Command that produces this log: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 ---------------------------------------------------------------------------------------------------- > trainable params: >>> pad_vector: torch.Size([1, 1, 1024]) >>> xlmr.embeddings.word_embeddings.weight: torch.Size([250002, 1024]) >>> xlmr.embeddings.position_embeddings.weight: torch.Size([514, 1024]) >>> xlmr.embeddings.token_type_embeddings.weight: torch.Size([1, 1024]) >>> xlmr.embeddings.LayerNorm.weight: torch.Size([1024]) >>> xlmr.embeddings.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.0.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.0.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.0.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.0.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.0.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.0.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.0.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.1.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.1.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.1.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.1.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.1.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.1.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.1.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.2.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.2.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.2.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.2.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.2.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.2.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.2.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.3.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.3.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.3.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.3.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.3.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.3.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.3.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.4.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.4.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.4.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.4.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.4.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.4.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.4.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.5.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.5.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.5.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.5.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.5.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.5.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.5.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.6.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.6.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.6.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.6.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.6.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.6.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.6.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.7.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.7.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.7.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.7.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.7.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.7.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.7.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.8.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.8.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.8.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.8.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.8.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.8.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.8.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.9.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.9.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.9.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.9.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.9.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.9.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.9.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.10.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.10.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.10.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.10.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.10.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.10.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.10.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.11.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.11.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.11.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.11.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.11.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.11.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.11.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.12.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.12.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.12.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.12.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.12.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.12.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.12.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.13.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.13.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.13.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.13.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.13.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.13.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.13.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.14.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.14.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.14.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.14.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.14.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.14.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.14.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.15.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.15.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.15.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.15.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.15.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.15.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.15.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.16.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.16.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.16.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.16.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.16.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.16.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.16.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.17.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.17.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.17.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.17.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.17.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.17.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.17.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.18.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.18.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.18.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.18.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.18.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.18.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.18.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.19.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.19.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.19.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.19.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.19.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.19.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.19.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.20.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.20.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.20.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.20.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.20.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.20.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.20.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.21.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.21.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.21.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.21.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.21.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.21.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.21.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.22.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.22.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.22.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.22.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.22.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.22.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.22.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.query.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.query.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.key.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.key.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.self.value.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.self.value.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.dense.weight: torch.Size([1024, 1024]) >>> xlmr.encoder.layer.23.attention.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.23.attention.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.intermediate.dense.weight: torch.Size([4096, 1024]) >>> xlmr.encoder.layer.23.intermediate.dense.bias: torch.Size([4096]) >>> xlmr.encoder.layer.23.output.dense.weight: torch.Size([1024, 4096]) >>> xlmr.encoder.layer.23.output.dense.bias: torch.Size([1024]) >>> xlmr.encoder.layer.23.output.LayerNorm.weight: torch.Size([1024]) >>> xlmr.encoder.layer.23.output.LayerNorm.bias: torch.Size([1024]) >>> xlmr.pooler.dense.weight: torch.Size([1024, 1024]) >>> xlmr.pooler.dense.bias: torch.Size([1024]) >>> upos_embedding.weight: torch.Size([19, 50]) >>> dist_embedding.weight: torch.Size([104, 50]) >>> deprel_embedding.weight: torch.Size([38, 50]) >>> gcn_layer.W.0.weight: torch.Size([450, 50]) >>> gcn_layer.W.0.bias: torch.Size([450]) >>> gcn_layer.W.1.weight: torch.Size([450, 450]) >>> gcn_layer.W.1.bias: torch.Size([450]) >>> trans_rep.weight: torch.Size([1024, 2924]) >>> trans_rep.bias: torch.Size([1024]) >>> mention_label_ffn.layers.0.weight: torch.Size([350, 1024]) >>> mention_label_ffn.layers.0.bias: torch.Size([350]) >>> mention_label_ffn.layers.1.weight: torch.Size([11, 350]) >>> mention_label_ffn.layers.1.bias: torch.Size([11]) >>> trigger_label_ffn.layers.0.weight: torch.Size([450, 1024]) >>> trigger_label_ffn.layers.0.bias: torch.Size([450]) >>> trigger_label_ffn.layers.1.weight: torch.Size([233, 450]) >>> trigger_label_ffn.layers.1.bias: torch.Size([233]) >>> event_type_ffn.layers.0.weight: torch.Size([450, 1024]) >>> event_type_ffn.layers.0.bias: torch.Size([450]) >>> event_type_ffn.layers.1.weight: torch.Size([117, 450]) >>> event_type_ffn.layers.1.bias: torch.Size([117]) >>> role_type_ffn.layers.0.weight: torch.Size([350, 1024]) >>> role_type_ffn.layers.0.bias: torch.Size([350]) >>> role_type_ffn.layers.1.weight: torch.Size([6, 350]) >>> role_type_ffn.layers.1.bias: torch.Size([6]) >>> trigger_crf.transition: torch.Size([235, 235]) >>> mention_crf.transition: torch.Size([13, 13]) n_trainable_params: 564979817, n_nontrainable_params: 0 ---------------------------------------------------------------------------------------------------- ****************************** Epoch: 0 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 05:04:55.201918: step: 2/1541, loss: 14.896011352539062 2023-01-23 05:04:56.165571: step: 4/1541, loss: 20.289886474609375 2023-01-23 05:04:57.061388: step: 6/1541, loss: 22.343137741088867 2023-01-23 05:04:58.066134: step: 8/1541, loss: 18.046424865722656 2023-01-23 05:04:58.990564: step: 10/1541, loss: 14.61170768737793 2023-01-23 05:04:59.887375: step: 12/1541, loss: 16.921581268310547 2023-01-23 05:05:00.802077: step: 14/1541, loss: 16.49817657470703 2023-01-23 05:05:01.701046: step: 16/1541, loss: 19.168132781982422 2023-01-23 05:05:02.608915: step: 18/1541, loss: 14.827041625976562 2023-01-23 05:05:03.506208: step: 20/1541, loss: 14.791550636291504 2023-01-23 05:05:04.397917: step: 22/1541, loss: 15.243518829345703 2023-01-23 05:05:05.281116: step: 24/1541, loss: 18.91895866394043 2023-01-23 05:05:06.146263: step: 26/1541, loss: 11.601038932800293 2023-01-23 05:05:07.077823: step: 28/1541, loss: 18.10414695739746 2023-01-23 05:05:07.962534: step: 30/1541, loss: 12.378674507141113 2023-01-23 05:05:08.842863: step: 32/1541, loss: 16.905330657958984 2023-01-23 05:05:09.722054: step: 34/1541, loss: 22.514575958251953 2023-01-23 05:05:10.609642: step: 36/1541, loss: 13.414445877075195 2023-01-23 05:05:11.502893: step: 38/1541, loss: 14.449551582336426 2023-01-23 05:05:12.384511: step: 40/1541, loss: 14.456972122192383 2023-01-23 05:05:13.293379: step: 42/1541, loss: 15.951105117797852 2023-01-23 05:05:14.281320: step: 44/1541, loss: 21.469106674194336 2023-01-23 05:05:15.165192: step: 46/1541, loss: 13.575908660888672 2023-01-23 05:05:16.072406: step: 48/1541, loss: 13.864509582519531 2023-01-23 05:05:16.965151: step: 50/1541, loss: 10.484349250793457 2023-01-23 05:05:17.872050: step: 52/1541, loss: 18.29685401916504 2023-01-23 05:05:18.760964: step: 54/1541, loss: 12.423348426818848 2023-01-23 05:05:19.617186: step: 56/1541, loss: 14.12872314453125 2023-01-23 05:05:20.479748: step: 58/1541, loss: 11.70406723022461 2023-01-23 05:05:21.493206: step: 60/1541, loss: 25.7273006439209 2023-01-23 05:05:22.420668: step: 62/1541, loss: 10.126834869384766 2023-01-23 05:05:23.317942: step: 64/1541, loss: 17.302453994750977 2023-01-23 05:05:24.266356: step: 66/1541, loss: 21.859588623046875 2023-01-23 05:05:25.174135: step: 68/1541, loss: 9.925400733947754 2023-01-23 05:05:26.035449: step: 70/1541, loss: 14.116806983947754 2023-01-23 05:05:26.966386: step: 72/1541, loss: 16.55558967590332 2023-01-23 05:05:27.883744: step: 74/1541, loss: 21.503005981445312 2023-01-23 05:05:28.844430: step: 76/1541, loss: 17.5455379486084 2023-01-23 05:05:29.746793: step: 78/1541, loss: 16.800954818725586 2023-01-23 05:05:30.586902: step: 80/1541, loss: 10.179220199584961 2023-01-23 05:05:31.485050: step: 82/1541, loss: 13.755714416503906 2023-01-23 05:05:32.373889: step: 84/1541, loss: 15.40347671508789 2023-01-23 05:05:33.253436: step: 86/1541, loss: 17.560955047607422 2023-01-23 05:05:34.135642: step: 88/1541, loss: 14.019800186157227 2023-01-23 05:05:35.015809: step: 90/1541, loss: 17.888471603393555 2023-01-23 05:05:35.911922: step: 92/1541, loss: 17.097434997558594 2023-01-23 05:05:36.893652: step: 94/1541, loss: 10.338234901428223 2023-01-23 05:05:37.764523: step: 96/1541, loss: 13.46949291229248 2023-01-23 05:05:38.617516: step: 98/1541, loss: 12.893721580505371 2023-01-23 05:05:39.534961: step: 100/1541, loss: 16.48578453063965 2023-01-23 05:05:40.457958: step: 102/1541, loss: 15.13374137878418 2023-01-23 05:05:41.371088: step: 104/1541, loss: 12.545004844665527 2023-01-23 05:05:42.249748: step: 106/1541, loss: 16.540802001953125 2023-01-23 05:05:43.146154: step: 108/1541, loss: 16.786109924316406 2023-01-23 05:05:44.057542: step: 110/1541, loss: 13.224740028381348 2023-01-23 05:05:44.954954: step: 112/1541, loss: 12.370198249816895 2023-01-23 05:05:45.861010: step: 114/1541, loss: 22.48811912536621 2023-01-23 05:05:46.757462: step: 116/1541, loss: 16.499298095703125 2023-01-23 05:05:47.630068: step: 118/1541, loss: 19.109081268310547 2023-01-23 05:05:48.496073: step: 120/1541, loss: 16.965959548950195 2023-01-23 05:05:49.370845: step: 122/1541, loss: 13.976045608520508 2023-01-23 05:05:50.288859: step: 124/1541, loss: 21.23472785949707 2023-01-23 05:05:51.145605: step: 126/1541, loss: 17.671363830566406 2023-01-23 05:05:52.022861: step: 128/1541, loss: 13.742109298706055 2023-01-23 05:05:52.874390: step: 130/1541, loss: 15.394196510314941 2023-01-23 05:05:53.826896: step: 132/1541, loss: 11.524449348449707 2023-01-23 05:05:54.707620: step: 134/1541, loss: 10.539427757263184 2023-01-23 05:05:55.561278: step: 136/1541, loss: 14.167620658874512 2023-01-23 05:05:56.485858: step: 138/1541, loss: 14.415192604064941 2023-01-23 05:05:57.395155: step: 140/1541, loss: 16.417461395263672 2023-01-23 05:05:58.303527: step: 142/1541, loss: 18.637100219726562 2023-01-23 05:05:59.234838: step: 144/1541, loss: 15.671875953674316 2023-01-23 05:06:00.110628: step: 146/1541, loss: 17.066654205322266 2023-01-23 05:06:01.052926: step: 148/1541, loss: 14.613673210144043 2023-01-23 05:06:01.945552: step: 150/1541, loss: 9.76639461517334 2023-01-23 05:06:02.884200: step: 152/1541, loss: 15.137272834777832 2023-01-23 05:06:03.776346: step: 154/1541, loss: 15.800795555114746 2023-01-23 05:06:04.681737: step: 156/1541, loss: 17.987262725830078 2023-01-23 05:06:05.713084: step: 158/1541, loss: 16.536911010742188 2023-01-23 05:06:06.628798: step: 160/1541, loss: 13.439132690429688 2023-01-23 05:06:07.578962: step: 162/1541, loss: 18.737594604492188 2023-01-23 05:06:08.451437: step: 164/1541, loss: 16.03371810913086 2023-01-23 05:06:09.336951: step: 166/1541, loss: 9.879634857177734 2023-01-23 05:06:10.198952: step: 168/1541, loss: 9.601217269897461 2023-01-23 05:06:11.108188: step: 170/1541, loss: 16.934783935546875 2023-01-23 05:06:12.013992: step: 172/1541, loss: 13.551576614379883 2023-01-23 05:06:12.955767: step: 174/1541, loss: 23.274749755859375 2023-01-23 05:06:13.840062: step: 176/1541, loss: 11.713321685791016 2023-01-23 05:06:14.699735: step: 178/1541, loss: 12.315461158752441 2023-01-23 05:06:15.572708: step: 180/1541, loss: 10.944551467895508 2023-01-23 05:06:16.445333: step: 182/1541, loss: 10.743326187133789 2023-01-23 05:06:17.339953: step: 184/1541, loss: 12.665847778320312 2023-01-23 05:06:18.269789: step: 186/1541, loss: 14.206562042236328 2023-01-23 05:06:19.143453: step: 188/1541, loss: 15.023327827453613 2023-01-23 05:06:20.053939: step: 190/1541, loss: 25.705564498901367 2023-01-23 05:06:20.943917: step: 192/1541, loss: 21.22574806213379 2023-01-23 05:06:21.815830: step: 194/1541, loss: 10.704414367675781 2023-01-23 05:06:22.687655: step: 196/1541, loss: 14.904123306274414 2023-01-23 05:06:23.546695: step: 198/1541, loss: 13.205425262451172 2023-01-23 05:06:24.457075: step: 200/1541, loss: 11.21593189239502 2023-01-23 05:06:25.313558: step: 202/1541, loss: 10.229848861694336 2023-01-23 05:06:26.263212: step: 204/1541, loss: 14.747854232788086 2023-01-23 05:06:27.185931: step: 206/1541, loss: 27.5120906829834 2023-01-23 05:06:28.041831: step: 208/1541, loss: 16.902420043945312 2023-01-23 05:06:28.903393: step: 210/1541, loss: 14.855912208557129 2023-01-23 05:06:29.798972: step: 212/1541, loss: 19.973215103149414 2023-01-23 05:06:30.730024: step: 214/1541, loss: 15.723851203918457 2023-01-23 05:06:31.629925: step: 216/1541, loss: 15.669995307922363 2023-01-23 05:06:32.634327: step: 218/1541, loss: 28.13666343688965 2023-01-23 05:06:33.508395: step: 220/1541, loss: 13.9130220413208 2023-01-23 05:06:34.391300: step: 222/1541, loss: 12.109015464782715 2023-01-23 05:06:35.329208: step: 224/1541, loss: 11.974514961242676 2023-01-23 05:06:36.221947: step: 226/1541, loss: 14.881797790527344 2023-01-23 05:06:37.060452: step: 228/1541, loss: 9.578873634338379 2023-01-23 05:06:37.954223: step: 230/1541, loss: 13.509004592895508 2023-01-23 05:06:38.805725: step: 232/1541, loss: 11.705095291137695 2023-01-23 05:06:39.730176: step: 234/1541, loss: 16.700428009033203 2023-01-23 05:06:40.710843: step: 236/1541, loss: 17.242578506469727 2023-01-23 05:06:41.658452: step: 238/1541, loss: 17.440521240234375 2023-01-23 05:06:42.589554: step: 240/1541, loss: 13.617818832397461 2023-01-23 05:06:43.426318: step: 242/1541, loss: 9.945706367492676 2023-01-23 05:06:44.300479: step: 244/1541, loss: 10.373117446899414 2023-01-23 05:06:45.279223: step: 246/1541, loss: 12.068395614624023 2023-01-23 05:06:46.202223: step: 248/1541, loss: 18.349424362182617 2023-01-23 05:06:47.189805: step: 250/1541, loss: 13.032878875732422 2023-01-23 05:06:48.120934: step: 252/1541, loss: 12.518865585327148 2023-01-23 05:06:48.984731: step: 254/1541, loss: 13.871589660644531 2023-01-23 05:06:49.948888: step: 256/1541, loss: 10.981792449951172 2023-01-23 05:06:50.856185: step: 258/1541, loss: 13.587631225585938 2023-01-23 05:06:51.790325: step: 260/1541, loss: 14.627349853515625 2023-01-23 05:06:52.672164: step: 262/1541, loss: 12.447607040405273 2023-01-23 05:06:53.541944: step: 264/1541, loss: 9.660717964172363 2023-01-23 05:06:54.425962: step: 266/1541, loss: 17.733016967773438 2023-01-23 05:06:55.341800: step: 268/1541, loss: 11.426300048828125 2023-01-23 05:06:56.214707: step: 270/1541, loss: 12.286795616149902 2023-01-23 05:06:57.094269: step: 272/1541, loss: 12.188199996948242 2023-01-23 05:06:57.988979: step: 274/1541, loss: 13.187979698181152 2023-01-23 05:06:58.890167: step: 276/1541, loss: 14.713739395141602 2023-01-23 05:06:59.750160: step: 278/1541, loss: 10.51850700378418 2023-01-23 05:07:00.612147: step: 280/1541, loss: 9.641315460205078 2023-01-23 05:07:01.517676: step: 282/1541, loss: 9.36971378326416 2023-01-23 05:07:02.399654: step: 284/1541, loss: 9.2005033493042 2023-01-23 05:07:03.334117: step: 286/1541, loss: 15.331141471862793 2023-01-23 05:07:04.336017: step: 288/1541, loss: 28.959259033203125 2023-01-23 05:07:05.244759: step: 290/1541, loss: 13.8943452835083 2023-01-23 05:07:06.132917: step: 292/1541, loss: 10.775311470031738 2023-01-23 05:07:06.994043: step: 294/1541, loss: 8.198742866516113 2023-01-23 05:07:07.936725: step: 296/1541, loss: 12.628128051757812 2023-01-23 05:07:08.852281: step: 298/1541, loss: 13.257475852966309 2023-01-23 05:07:09.824368: step: 300/1541, loss: 16.509267807006836 2023-01-23 05:07:10.710498: step: 302/1541, loss: 10.622275352478027 2023-01-23 05:07:11.621661: step: 304/1541, loss: 14.735602378845215 2023-01-23 05:07:12.606404: step: 306/1541, loss: 13.431617736816406 2023-01-23 05:07:13.522537: step: 308/1541, loss: 10.78051471710205 2023-01-23 05:07:14.417422: step: 310/1541, loss: 13.137725830078125 2023-01-23 05:07:15.342848: step: 312/1541, loss: 11.351325035095215 2023-01-23 05:07:16.263327: step: 314/1541, loss: 12.135729789733887 2023-01-23 05:07:17.141100: step: 316/1541, loss: 16.59418296813965 2023-01-23 05:07:18.075383: step: 318/1541, loss: 14.969941139221191 2023-01-23 05:07:18.955874: step: 320/1541, loss: 8.376886367797852 2023-01-23 05:07:19.848339: step: 322/1541, loss: 11.881765365600586 2023-01-23 05:07:20.698083: step: 324/1541, loss: 12.375406265258789 2023-01-23 05:07:21.552108: step: 326/1541, loss: 8.270042419433594 2023-01-23 05:07:22.440125: step: 328/1541, loss: 10.202186584472656 2023-01-23 05:07:23.326123: step: 330/1541, loss: 10.510866165161133 2023-01-23 05:07:24.278376: step: 332/1541, loss: 8.12625503540039 2023-01-23 05:07:25.216253: step: 334/1541, loss: 10.550634384155273 2023-01-23 05:07:26.068578: step: 336/1541, loss: 10.654748916625977 2023-01-23 05:07:26.945908: step: 338/1541, loss: 10.591633796691895 2023-01-23 05:07:27.807468: step: 340/1541, loss: 6.952168941497803 2023-01-23 05:07:28.683489: step: 342/1541, loss: 9.167327880859375 2023-01-23 05:07:29.600198: step: 344/1541, loss: 13.941611289978027 2023-01-23 05:07:30.496457: step: 346/1541, loss: 13.703326225280762 2023-01-23 05:07:31.390085: step: 348/1541, loss: 8.403375625610352 2023-01-23 05:07:32.320508: step: 350/1541, loss: 10.342571258544922 2023-01-23 05:07:33.224975: step: 352/1541, loss: 7.679482460021973 2023-01-23 05:07:34.127323: step: 354/1541, loss: 9.09946060180664 2023-01-23 05:07:35.039721: step: 356/1541, loss: 10.527013778686523 2023-01-23 05:07:35.905559: step: 358/1541, loss: 9.065887451171875 2023-01-23 05:07:36.958311: step: 360/1541, loss: 12.415372848510742 2023-01-23 05:07:37.902956: step: 362/1541, loss: 9.751036643981934 2023-01-23 05:07:38.781233: step: 364/1541, loss: 8.522175788879395 2023-01-23 05:07:39.662639: step: 366/1541, loss: 5.258638381958008 2023-01-23 05:07:40.559419: step: 368/1541, loss: 9.28451156616211 2023-01-23 05:07:41.450841: step: 370/1541, loss: 10.638537406921387 2023-01-23 05:07:42.334072: step: 372/1541, loss: 5.801848888397217 2023-01-23 05:07:43.236776: step: 374/1541, loss: 7.366043567657471 2023-01-23 05:07:44.122079: step: 376/1541, loss: 5.476063251495361 2023-01-23 05:07:44.972107: step: 378/1541, loss: 11.773749351501465 2023-01-23 05:07:45.858787: step: 380/1541, loss: 10.527091979980469 2023-01-23 05:07:46.737757: step: 382/1541, loss: 6.270551681518555 2023-01-23 05:07:47.609976: step: 384/1541, loss: 9.400766372680664 2023-01-23 05:07:48.530881: step: 386/1541, loss: 8.078178405761719 2023-01-23 05:07:49.504348: step: 388/1541, loss: 5.781654357910156 2023-01-23 05:07:50.446767: step: 390/1541, loss: 6.047903537750244 2023-01-23 05:07:51.378871: step: 392/1541, loss: 7.276648998260498 2023-01-23 05:07:52.242600: step: 394/1541, loss: 5.411816596984863 2023-01-23 05:07:53.149304: step: 396/1541, loss: 4.4493608474731445 2023-01-23 05:07:54.028409: step: 398/1541, loss: 8.202818870544434 2023-01-23 05:07:54.916271: step: 400/1541, loss: 5.503396987915039 2023-01-23 05:07:55.828384: step: 402/1541, loss: 13.372604370117188 2023-01-23 05:07:56.729353: step: 404/1541, loss: 8.162095069885254 2023-01-23 05:07:57.680480: step: 406/1541, loss: 6.971880912780762 2023-01-23 05:07:58.562277: step: 408/1541, loss: 7.058540344238281 2023-01-23 05:07:59.449405: step: 410/1541, loss: 7.383754730224609 2023-01-23 05:08:00.378287: step: 412/1541, loss: 8.020293235778809 2023-01-23 05:08:01.234400: step: 414/1541, loss: 5.338465690612793 2023-01-23 05:08:02.138797: step: 416/1541, loss: 5.134068965911865 2023-01-23 05:08:03.023626: step: 418/1541, loss: 5.084214687347412 2023-01-23 05:08:03.908027: step: 420/1541, loss: 7.4147844314575195 2023-01-23 05:08:04.757295: step: 422/1541, loss: 4.3640923500061035 2023-01-23 05:08:05.638308: step: 424/1541, loss: 5.133774280548096 2023-01-23 05:08:06.493588: step: 426/1541, loss: 4.934262275695801 2023-01-23 05:08:07.411788: step: 428/1541, loss: 5.111780166625977 2023-01-23 05:08:08.317278: step: 430/1541, loss: 5.796981334686279 2023-01-23 05:08:09.215624: step: 432/1541, loss: 6.568942546844482 2023-01-23 05:08:10.103926: step: 434/1541, loss: 3.9474167823791504 2023-01-23 05:08:11.023734: step: 436/1541, loss: 6.148954391479492 2023-01-23 05:08:11.914934: step: 438/1541, loss: 4.145214080810547 2023-01-23 05:08:12.788191: step: 440/1541, loss: 5.076517105102539 2023-01-23 05:08:13.676033: step: 442/1541, loss: 7.13367223739624 2023-01-23 05:08:14.705371: step: 444/1541, loss: 6.113094806671143 2023-01-23 05:08:15.594547: step: 446/1541, loss: 7.508191108703613 2023-01-23 05:08:16.458568: step: 448/1541, loss: 8.69920825958252 2023-01-23 05:08:17.321684: step: 450/1541, loss: 4.047685623168945 2023-01-23 05:08:18.214023: step: 452/1541, loss: 6.28642463684082 2023-01-23 05:08:19.142319: step: 454/1541, loss: 11.157655715942383 2023-01-23 05:08:20.014940: step: 456/1541, loss: 7.138740062713623 2023-01-23 05:08:20.913058: step: 458/1541, loss: 7.9998579025268555 2023-01-23 05:08:21.802494: step: 460/1541, loss: 7.284845352172852 2023-01-23 05:08:22.706066: step: 462/1541, loss: 6.214392185211182 2023-01-23 05:08:23.575299: step: 464/1541, loss: 7.157645225524902 2023-01-23 05:08:24.475560: step: 466/1541, loss: 4.599192142486572 2023-01-23 05:08:25.358564: step: 468/1541, loss: 4.021225929260254 2023-01-23 05:08:26.257410: step: 470/1541, loss: 7.079800128936768 2023-01-23 05:08:27.141498: step: 472/1541, loss: 4.057326316833496 2023-01-23 05:08:28.020595: step: 474/1541, loss: 4.23392391204834 2023-01-23 05:08:28.955400: step: 476/1541, loss: 2.676656723022461 2023-01-23 05:08:29.861667: step: 478/1541, loss: 7.009961128234863 2023-01-23 05:08:30.805140: step: 480/1541, loss: 6.53196382522583 2023-01-23 05:08:31.693088: step: 482/1541, loss: 8.03916072845459 2023-01-23 05:08:32.663192: step: 484/1541, loss: 6.76717472076416 2023-01-23 05:08:33.535987: step: 486/1541, loss: 3.829130172729492 2023-01-23 05:08:34.427908: step: 488/1541, loss: 3.9816298484802246 2023-01-23 05:08:35.343767: step: 490/1541, loss: 4.208441257476807 2023-01-23 05:08:36.183199: step: 492/1541, loss: 6.957939624786377 2023-01-23 05:08:37.033970: step: 494/1541, loss: 4.0806965827941895 2023-01-23 05:08:37.906134: step: 496/1541, loss: 6.3337507247924805 2023-01-23 05:08:38.778114: step: 498/1541, loss: 4.480905055999756 2023-01-23 05:08:39.644779: step: 500/1541, loss: 4.461435317993164 2023-01-23 05:08:40.573648: step: 502/1541, loss: 3.812537431716919 2023-01-23 05:08:41.474613: step: 504/1541, loss: 3.5521645545959473 2023-01-23 05:08:42.372081: step: 506/1541, loss: 8.348159790039062 2023-01-23 05:08:43.251730: step: 508/1541, loss: 4.540029048919678 2023-01-23 05:08:44.160174: step: 510/1541, loss: 6.26772403717041 2023-01-23 05:08:45.081538: step: 512/1541, loss: 6.450508117675781 2023-01-23 05:08:46.000764: step: 514/1541, loss: 5.327237606048584 2023-01-23 05:08:46.897670: step: 516/1541, loss: 7.005661964416504 2023-01-23 05:08:47.792291: step: 518/1541, loss: 3.375298500061035 2023-01-23 05:08:48.793891: step: 520/1541, loss: 6.48797607421875 2023-01-23 05:08:49.713431: step: 522/1541, loss: 3.3542630672454834 2023-01-23 05:08:50.614924: step: 524/1541, loss: 3.815680503845215 2023-01-23 05:08:51.507586: step: 526/1541, loss: 3.993610143661499 2023-01-23 05:08:52.409203: step: 528/1541, loss: 6.151822566986084 2023-01-23 05:08:53.318093: step: 530/1541, loss: 6.028957366943359 2023-01-23 05:08:54.169175: step: 532/1541, loss: 7.0997161865234375 2023-01-23 05:08:55.031512: step: 534/1541, loss: 3.6653332710266113 2023-01-23 05:08:55.975260: step: 536/1541, loss: 9.26030158996582 2023-01-23 05:08:56.854951: step: 538/1541, loss: 4.388888359069824 2023-01-23 05:08:57.866588: step: 540/1541, loss: 4.423489093780518 2023-01-23 05:08:58.761442: step: 542/1541, loss: 4.260842800140381 2023-01-23 05:08:59.650052: step: 544/1541, loss: 4.48546838760376 2023-01-23 05:09:00.568688: step: 546/1541, loss: 4.735332012176514 2023-01-23 05:09:01.437214: step: 548/1541, loss: 4.6871795654296875 2023-01-23 05:09:02.361836: step: 550/1541, loss: 4.846988201141357 2023-01-23 05:09:03.236353: step: 552/1541, loss: 6.819343090057373 2023-01-23 05:09:04.149660: step: 554/1541, loss: 3.1561124324798584 2023-01-23 05:09:05.053271: step: 556/1541, loss: 3.5109338760375977 2023-01-23 05:09:05.948929: step: 558/1541, loss: 4.016413688659668 2023-01-23 05:09:06.877250: step: 560/1541, loss: 3.304412364959717 2023-01-23 05:09:07.742306: step: 562/1541, loss: 9.458431243896484 2023-01-23 05:09:08.644339: step: 564/1541, loss: 5.721556186676025 2023-01-23 05:09:09.547023: step: 566/1541, loss: 5.96155309677124 2023-01-23 05:09:10.435508: step: 568/1541, loss: 4.460544586181641 2023-01-23 05:09:11.277197: step: 570/1541, loss: 3.0413389205932617 2023-01-23 05:09:12.146157: step: 572/1541, loss: 4.099221229553223 2023-01-23 05:09:13.015955: step: 574/1541, loss: 4.811769962310791 2023-01-23 05:09:13.918486: step: 576/1541, loss: 4.076995849609375 2023-01-23 05:09:14.811692: step: 578/1541, loss: 2.1908063888549805 2023-01-23 05:09:15.680935: step: 580/1541, loss: 5.072532653808594 2023-01-23 05:09:16.569853: step: 582/1541, loss: 5.656835079193115 2023-01-23 05:09:17.460381: step: 584/1541, loss: 2.226346969604492 2023-01-23 05:09:18.324974: step: 586/1541, loss: 4.9131951332092285 2023-01-23 05:09:19.265580: step: 588/1541, loss: 3.640947103500366 2023-01-23 05:09:20.193755: step: 590/1541, loss: 5.402192115783691 2023-01-23 05:09:21.086233: step: 592/1541, loss: 4.2272539138793945 2023-01-23 05:09:21.961916: step: 594/1541, loss: 3.259627103805542 2023-01-23 05:09:22.853229: step: 596/1541, loss: 5.759600639343262 2023-01-23 05:09:23.753284: step: 598/1541, loss: 2.967677354812622 2023-01-23 05:09:24.660704: step: 600/1541, loss: 7.251462936401367 2023-01-23 05:09:25.524268: step: 602/1541, loss: 2.300006866455078 2023-01-23 05:09:26.382240: step: 604/1541, loss: 3.466928005218506 2023-01-23 05:09:27.283715: step: 606/1541, loss: 4.464787483215332 2023-01-23 05:09:28.140501: step: 608/1541, loss: 3.4037656784057617 2023-01-23 05:09:29.146002: step: 610/1541, loss: 7.092461109161377 2023-01-23 05:09:30.051562: step: 612/1541, loss: 4.829914569854736 2023-01-23 05:09:30.970788: step: 614/1541, loss: 5.354001045227051 2023-01-23 05:09:31.818821: step: 616/1541, loss: 7.749959468841553 2023-01-23 05:09:32.690030: step: 618/1541, loss: 3.966111898422241 2023-01-23 05:09:33.639420: step: 620/1541, loss: 2.011718273162842 2023-01-23 05:09:34.509741: step: 622/1541, loss: 5.232485771179199 2023-01-23 05:09:35.373884: step: 624/1541, loss: 10.818878173828125 2023-01-23 05:09:36.317150: step: 626/1541, loss: 3.450985908508301 2023-01-23 05:09:37.187775: step: 628/1541, loss: 4.989720344543457 2023-01-23 05:09:38.103239: step: 630/1541, loss: 4.111686706542969 2023-01-23 05:09:38.994547: step: 632/1541, loss: 4.982763290405273 2023-01-23 05:09:39.911492: step: 634/1541, loss: 3.7170729637145996 2023-01-23 05:09:40.774157: step: 636/1541, loss: 3.519911766052246 2023-01-23 05:09:41.702253: step: 638/1541, loss: 2.0635600090026855 2023-01-23 05:09:42.593673: step: 640/1541, loss: 2.452291965484619 2023-01-23 05:09:43.463646: step: 642/1541, loss: 3.3125758171081543 2023-01-23 05:09:44.349648: step: 644/1541, loss: 4.572969436645508 2023-01-23 05:09:45.236693: step: 646/1541, loss: 5.80129861831665 2023-01-23 05:09:46.141045: step: 648/1541, loss: 4.772152423858643 2023-01-23 05:09:47.002605: step: 650/1541, loss: 6.751237869262695 2023-01-23 05:09:47.869920: step: 652/1541, loss: 4.197153091430664 2023-01-23 05:09:48.802637: step: 654/1541, loss: 3.144000768661499 2023-01-23 05:09:49.724325: step: 656/1541, loss: 3.7897844314575195 2023-01-23 05:09:50.592569: step: 658/1541, loss: 3.216465711593628 2023-01-23 05:09:51.480872: step: 660/1541, loss: 3.2540738582611084 2023-01-23 05:09:52.341963: step: 662/1541, loss: 2.5100858211517334 2023-01-23 05:09:53.244439: step: 664/1541, loss: 6.134953498840332 2023-01-23 05:09:54.129989: step: 666/1541, loss: 3.0558724403381348 2023-01-23 05:09:55.013485: step: 668/1541, loss: 10.272424697875977 2023-01-23 05:09:55.918125: step: 670/1541, loss: 4.6190314292907715 2023-01-23 05:09:56.798767: step: 672/1541, loss: 3.312051296234131 2023-01-23 05:09:57.675953: step: 674/1541, loss: 4.469375133514404 2023-01-23 05:09:58.554899: step: 676/1541, loss: 3.228795289993286 2023-01-23 05:09:59.426201: step: 678/1541, loss: 2.7653534412384033 2023-01-23 05:10:00.345746: step: 680/1541, loss: 4.442948341369629 2023-01-23 05:10:01.251344: step: 682/1541, loss: 4.926861763000488 2023-01-23 05:10:02.198945: step: 684/1541, loss: 4.0978684425354 2023-01-23 05:10:03.086772: step: 686/1541, loss: 4.060983657836914 2023-01-23 05:10:03.981188: step: 688/1541, loss: 3.680670976638794 2023-01-23 05:10:04.906648: step: 690/1541, loss: 5.661717414855957 2023-01-23 05:10:05.819385: step: 692/1541, loss: 3.3364341259002686 2023-01-23 05:10:06.736360: step: 694/1541, loss: 2.9225239753723145 2023-01-23 05:10:07.689042: step: 696/1541, loss: 2.724670886993408 2023-01-23 05:10:08.551740: step: 698/1541, loss: 3.1095707416534424 2023-01-23 05:10:09.506024: step: 700/1541, loss: 4.1214752197265625 2023-01-23 05:10:10.424055: step: 702/1541, loss: 2.578127145767212 2023-01-23 05:10:11.297115: step: 704/1541, loss: 5.951290130615234 2023-01-23 05:10:12.190646: step: 706/1541, loss: 3.2709858417510986 2023-01-23 05:10:13.089955: step: 708/1541, loss: 2.4436821937561035 2023-01-23 05:10:13.988429: step: 710/1541, loss: 3.4373068809509277 2023-01-23 05:10:14.953321: step: 712/1541, loss: 3.4814581871032715 2023-01-23 05:10:15.882572: step: 714/1541, loss: 3.875108480453491 2023-01-23 05:10:16.729266: step: 716/1541, loss: 2.2136471271514893 2023-01-23 05:10:17.634817: step: 718/1541, loss: 4.717573165893555 2023-01-23 05:10:18.538143: step: 720/1541, loss: 7.236995220184326 2023-01-23 05:10:19.527269: step: 722/1541, loss: 4.480843544006348 2023-01-23 05:10:20.421558: step: 724/1541, loss: 5.0453572273254395 2023-01-23 05:10:21.347070: step: 726/1541, loss: 3.3123040199279785 2023-01-23 05:10:22.349965: step: 728/1541, loss: 3.7874321937561035 2023-01-23 05:10:23.201430: step: 730/1541, loss: 5.709663391113281 2023-01-23 05:10:24.123034: step: 732/1541, loss: 6.379438400268555 2023-01-23 05:10:24.981450: step: 734/1541, loss: 3.075469732284546 2023-01-23 05:10:25.967267: step: 736/1541, loss: 3.9298791885375977 2023-01-23 05:10:26.850713: step: 738/1541, loss: 4.50181770324707 2023-01-23 05:10:27.753156: step: 740/1541, loss: 2.6362483501434326 2023-01-23 05:10:28.649424: step: 742/1541, loss: 3.846482276916504 2023-01-23 05:10:29.507904: step: 744/1541, loss: 5.908433437347412 2023-01-23 05:10:30.385665: step: 746/1541, loss: 3.9933972358703613 2023-01-23 05:10:31.262249: step: 748/1541, loss: 3.575603485107422 2023-01-23 05:10:32.208780: step: 750/1541, loss: 5.493069171905518 2023-01-23 05:10:33.093338: step: 752/1541, loss: 2.471210479736328 2023-01-23 05:10:33.966723: step: 754/1541, loss: 4.7410478591918945 2023-01-23 05:10:34.846088: step: 756/1541, loss: 4.400306701660156 2023-01-23 05:10:35.750393: step: 758/1541, loss: 2.9990196228027344 2023-01-23 05:10:36.640224: step: 760/1541, loss: 3.4297213554382324 2023-01-23 05:10:37.525589: step: 762/1541, loss: 2.470597982406616 2023-01-23 05:10:38.371463: step: 764/1541, loss: 2.350137710571289 2023-01-23 05:10:39.271140: step: 766/1541, loss: 2.0197553634643555 2023-01-23 05:10:40.126953: step: 768/1541, loss: 4.368199825286865 2023-01-23 05:10:41.033616: step: 770/1541, loss: 5.737679481506348 2023-01-23 05:10:41.903963: step: 772/1541, loss: 2.1960833072662354 2023-01-23 05:10:42.802193: step: 774/1541, loss: 5.675147533416748 2023-01-23 05:10:43.663366: step: 776/1541, loss: 3.5716280937194824 2023-01-23 05:10:44.620622: step: 778/1541, loss: 5.240971565246582 2023-01-23 05:10:45.526101: step: 780/1541, loss: 2.375906229019165 2023-01-23 05:10:46.408947: step: 782/1541, loss: 4.608270645141602 2023-01-23 05:10:47.303319: step: 784/1541, loss: 2.629014492034912 2023-01-23 05:10:48.191982: step: 786/1541, loss: 5.635354518890381 2023-01-23 05:10:49.074896: step: 788/1541, loss: 3.3931660652160645 2023-01-23 05:10:49.936507: step: 790/1541, loss: 3.401309013366699 2023-01-23 05:10:50.863574: step: 792/1541, loss: 4.143005847930908 2023-01-23 05:10:51.739028: step: 794/1541, loss: 4.247649669647217 2023-01-23 05:10:52.665755: step: 796/1541, loss: 4.214217185974121 2023-01-23 05:10:53.591602: step: 798/1541, loss: 3.3595128059387207 2023-01-23 05:10:54.525181: step: 800/1541, loss: 4.2044358253479 2023-01-23 05:10:55.408096: step: 802/1541, loss: 4.674581050872803 2023-01-23 05:10:56.267020: step: 804/1541, loss: 6.511136054992676 2023-01-23 05:10:57.173714: step: 806/1541, loss: 3.837322235107422 2023-01-23 05:10:58.091391: step: 808/1541, loss: 7.694106101989746 2023-01-23 05:10:59.068196: step: 810/1541, loss: 3.225917100906372 2023-01-23 05:10:59.947757: step: 812/1541, loss: 3.38158917427063 2023-01-23 05:11:00.844929: step: 814/1541, loss: 7.466368198394775 2023-01-23 05:11:01.779163: step: 816/1541, loss: 6.244002819061279 2023-01-23 05:11:02.717125: step: 818/1541, loss: 5.614564418792725 2023-01-23 05:11:03.580845: step: 820/1541, loss: 2.2344472408294678 2023-01-23 05:11:04.474588: step: 822/1541, loss: 5.437931060791016 2023-01-23 05:11:05.383195: step: 824/1541, loss: 6.1003241539001465 2023-01-23 05:11:06.308494: step: 826/1541, loss: 3.3683552742004395 2023-01-23 05:11:07.224296: step: 828/1541, loss: 2.8430023193359375 2023-01-23 05:11:08.102998: step: 830/1541, loss: 3.2997851371765137 2023-01-23 05:11:08.957421: step: 832/1541, loss: 2.769927978515625 2023-01-23 05:11:09.828538: step: 834/1541, loss: 3.977215528488159 2023-01-23 05:11:10.705845: step: 836/1541, loss: 5.0410284996032715 2023-01-23 05:11:11.593395: step: 838/1541, loss: 4.464787006378174 2023-01-23 05:11:12.455624: step: 840/1541, loss: 2.3246312141418457 2023-01-23 05:11:13.319070: step: 842/1541, loss: 3.2977421283721924 2023-01-23 05:11:14.256397: step: 844/1541, loss: 7.364193916320801 2023-01-23 05:11:15.138152: step: 846/1541, loss: 3.542060136795044 2023-01-23 05:11:16.026072: step: 848/1541, loss: 5.847805976867676 2023-01-23 05:11:16.899881: step: 850/1541, loss: 3.4626266956329346 2023-01-23 05:11:17.828367: step: 852/1541, loss: 5.154491901397705 2023-01-23 05:11:18.705992: step: 854/1541, loss: 5.1155853271484375 2023-01-23 05:11:19.558833: step: 856/1541, loss: 3.715864896774292 2023-01-23 05:11:20.429792: step: 858/1541, loss: 4.075436115264893 2023-01-23 05:11:21.365009: step: 860/1541, loss: 4.29785680770874 2023-01-23 05:11:22.296637: step: 862/1541, loss: 2.957780122756958 2023-01-23 05:11:23.261681: step: 864/1541, loss: 3.5591700077056885 2023-01-23 05:11:24.173096: step: 866/1541, loss: 3.106982946395874 2023-01-23 05:11:25.069596: step: 868/1541, loss: 3.92694354057312 2023-01-23 05:11:25.963853: step: 870/1541, loss: 4.607773780822754 2023-01-23 05:11:26.886631: step: 872/1541, loss: 6.763466835021973 2023-01-23 05:11:27.775304: step: 874/1541, loss: 4.374809265136719 2023-01-23 05:11:28.707866: step: 876/1541, loss: 3.649214029312134 2023-01-23 05:11:29.618823: step: 878/1541, loss: 6.533082962036133 2023-01-23 05:11:30.496405: step: 880/1541, loss: 3.472522735595703 2023-01-23 05:11:31.393199: step: 882/1541, loss: 3.343729257583618 2023-01-23 05:11:32.374145: step: 884/1541, loss: 2.095874547958374 2023-01-23 05:11:33.254049: step: 886/1541, loss: 5.5841450691223145 2023-01-23 05:11:34.090359: step: 888/1541, loss: 2.7591757774353027 2023-01-23 05:11:34.987641: step: 890/1541, loss: 5.90484094619751 2023-01-23 05:11:35.890062: step: 892/1541, loss: 2.1706507205963135 2023-01-23 05:11:36.791464: step: 894/1541, loss: 4.285267353057861 2023-01-23 05:11:37.660416: step: 896/1541, loss: 2.8478548526763916 2023-01-23 05:11:38.639970: step: 898/1541, loss: 4.170920372009277 2023-01-23 05:11:39.514191: step: 900/1541, loss: 5.2313151359558105 2023-01-23 05:11:40.408682: step: 902/1541, loss: 2.333449363708496 2023-01-23 05:11:41.324001: step: 904/1541, loss: 4.408980846405029 2023-01-23 05:11:42.243629: step: 906/1541, loss: 5.329829216003418 2023-01-23 05:11:43.164885: step: 908/1541, loss: 6.801540374755859 2023-01-23 05:11:44.101985: step: 910/1541, loss: 5.4480977058410645 2023-01-23 05:11:44.974537: step: 912/1541, loss: 3.4017796516418457 2023-01-23 05:11:45.865390: step: 914/1541, loss: 4.53005838394165 2023-01-23 05:11:46.809864: step: 916/1541, loss: 6.243847846984863 2023-01-23 05:11:47.699739: step: 918/1541, loss: 2.4218573570251465 2023-01-23 05:11:48.612379: step: 920/1541, loss: 4.277845859527588 2023-01-23 05:11:49.463089: step: 922/1541, loss: 2.6044058799743652 2023-01-23 05:11:50.336548: step: 924/1541, loss: 5.421505451202393 2023-01-23 05:11:51.283631: step: 926/1541, loss: 2.476356267929077 2023-01-23 05:11:52.172138: step: 928/1541, loss: 2.032877206802368 2023-01-23 05:11:53.093161: step: 930/1541, loss: 5.9327826499938965 2023-01-23 05:11:53.962005: step: 932/1541, loss: 2.4857335090637207 2023-01-23 05:11:54.902352: step: 934/1541, loss: 4.576376438140869 2023-01-23 05:11:55.767526: step: 936/1541, loss: 3.3368029594421387 2023-01-23 05:11:56.642164: step: 938/1541, loss: 5.318083763122559 2023-01-23 05:11:57.555589: step: 940/1541, loss: 3.1735570430755615 2023-01-23 05:11:58.457274: step: 942/1541, loss: 1.6922273635864258 2023-01-23 05:11:59.338033: step: 944/1541, loss: 2.6820497512817383 2023-01-23 05:12:00.196003: step: 946/1541, loss: 3.503749370574951 2023-01-23 05:12:01.121570: step: 948/1541, loss: 2.177628517150879 2023-01-23 05:12:02.040775: step: 950/1541, loss: 2.4885754585266113 2023-01-23 05:12:02.963161: step: 952/1541, loss: 3.9697680473327637 2023-01-23 05:12:03.845569: step: 954/1541, loss: 3.0796568393707275 2023-01-23 05:12:04.782753: step: 956/1541, loss: 5.622129440307617 2023-01-23 05:12:05.736851: step: 958/1541, loss: 8.584484100341797 2023-01-23 05:12:06.587528: step: 960/1541, loss: 2.4463112354278564 2023-01-23 05:12:07.512695: step: 962/1541, loss: 2.272512197494507 2023-01-23 05:12:08.485014: step: 964/1541, loss: 3.2077689170837402 2023-01-23 05:12:09.410932: step: 966/1541, loss: 5.46510124206543 2023-01-23 05:12:10.300040: step: 968/1541, loss: 2.9945437908172607 2023-01-23 05:12:11.192336: step: 970/1541, loss: 6.780055046081543 2023-01-23 05:12:12.069055: step: 972/1541, loss: 2.4195168018341064 2023-01-23 05:12:13.107456: step: 974/1541, loss: 1.9949817657470703 2023-01-23 05:12:13.974842: step: 976/1541, loss: 9.515571594238281 2023-01-23 05:12:14.858824: step: 978/1541, loss: 9.717366218566895 2023-01-23 05:12:15.743532: step: 980/1541, loss: 2.4512810707092285 2023-01-23 05:12:16.645463: step: 982/1541, loss: 2.940486192703247 2023-01-23 05:12:17.493240: step: 984/1541, loss: 4.923232078552246 2023-01-23 05:12:18.384463: step: 986/1541, loss: 4.831637859344482 2023-01-23 05:12:19.325325: step: 988/1541, loss: 3.184844732284546 2023-01-23 05:12:20.206897: step: 990/1541, loss: 3.939814567565918 2023-01-23 05:12:21.148135: step: 992/1541, loss: 2.5068154335021973 2023-01-23 05:12:22.036254: step: 994/1541, loss: 3.366342067718506 2023-01-23 05:12:22.928345: step: 996/1541, loss: 2.1876163482666016 2023-01-23 05:12:23.845388: step: 998/1541, loss: 3.1237003803253174 2023-01-23 05:12:24.772010: step: 1000/1541, loss: 1.8174586296081543 2023-01-23 05:12:25.658448: step: 1002/1541, loss: 4.658470630645752 2023-01-23 05:12:26.566137: step: 1004/1541, loss: 1.9084432125091553 2023-01-23 05:12:27.484695: step: 1006/1541, loss: 8.80937671661377 2023-01-23 05:12:28.350503: step: 1008/1541, loss: 4.6523637771606445 2023-01-23 05:12:29.234676: step: 1010/1541, loss: 4.013759613037109 2023-01-23 05:12:30.120235: step: 1012/1541, loss: 3.4185705184936523 2023-01-23 05:12:31.022825: step: 1014/1541, loss: 4.903395175933838 2023-01-23 05:12:31.946519: step: 1016/1541, loss: 2.5887839794158936 2023-01-23 05:12:32.869525: step: 1018/1541, loss: 4.489517688751221 2023-01-23 05:12:33.808242: step: 1020/1541, loss: 3.2452621459960938 2023-01-23 05:12:34.758127: step: 1022/1541, loss: 4.1305952072143555 2023-01-23 05:12:35.659474: step: 1024/1541, loss: 2.3782753944396973 2023-01-23 05:12:36.591226: step: 1026/1541, loss: 6.47962760925293 2023-01-23 05:12:37.452855: step: 1028/1541, loss: 4.9638471603393555 2023-01-23 05:12:38.397761: step: 1030/1541, loss: 1.6796746253967285 2023-01-23 05:12:39.342376: step: 1032/1541, loss: 3.390737533569336 2023-01-23 05:12:40.289542: step: 1034/1541, loss: 1.5792797803878784 2023-01-23 05:12:41.181470: step: 1036/1541, loss: 2.9843931198120117 2023-01-23 05:12:42.084336: step: 1038/1541, loss: 2.7397854328155518 2023-01-23 05:12:42.995253: step: 1040/1541, loss: 5.115817070007324 2023-01-23 05:12:43.903143: step: 1042/1541, loss: 2.6445698738098145 2023-01-23 05:12:44.794100: step: 1044/1541, loss: 1.619409203529358 2023-01-23 05:12:45.716776: step: 1046/1541, loss: 2.934901714324951 2023-01-23 05:12:46.590676: step: 1048/1541, loss: 1.803674340248108 2023-01-23 05:12:47.481671: step: 1050/1541, loss: 7.2734856605529785 2023-01-23 05:12:48.379464: step: 1052/1541, loss: 3.712313652038574 2023-01-23 05:12:49.331457: step: 1054/1541, loss: 5.055439472198486 2023-01-23 05:12:50.183869: step: 1056/1541, loss: 1.398912787437439 2023-01-23 05:12:51.038732: step: 1058/1541, loss: 2.5574910640716553 2023-01-23 05:12:51.975650: step: 1060/1541, loss: 4.2671284675598145 2023-01-23 05:12:52.899234: step: 1062/1541, loss: 2.7502095699310303 2023-01-23 05:12:53.849894: step: 1064/1541, loss: 3.187042236328125 2023-01-23 05:12:54.740579: step: 1066/1541, loss: 3.5256924629211426 2023-01-23 05:12:55.603764: step: 1068/1541, loss: 5.128590106964111 2023-01-23 05:12:56.483773: step: 1070/1541, loss: 4.537866115570068 2023-01-23 05:12:57.366816: step: 1072/1541, loss: 3.64961576461792 2023-01-23 05:12:58.268326: step: 1074/1541, loss: 1.6214724779129028 2023-01-23 05:12:59.162423: step: 1076/1541, loss: 5.227659225463867 2023-01-23 05:13:00.056148: step: 1078/1541, loss: 4.470484256744385 2023-01-23 05:13:00.988787: step: 1080/1541, loss: 2.519711494445801 2023-01-23 05:13:01.945936: step: 1082/1541, loss: 5.198164463043213 2023-01-23 05:13:02.804446: step: 1084/1541, loss: 2.3980495929718018 2023-01-23 05:13:03.748693: step: 1086/1541, loss: 2.649501085281372 2023-01-23 05:13:04.616641: step: 1088/1541, loss: 1.8109287023544312 2023-01-23 05:13:05.570906: step: 1090/1541, loss: 3.2621145248413086 2023-01-23 05:13:06.470330: step: 1092/1541, loss: 3.2648091316223145 2023-01-23 05:13:07.359229: step: 1094/1541, loss: 1.7164586782455444 2023-01-23 05:13:08.311636: step: 1096/1541, loss: 3.1959116458892822 2023-01-23 05:13:09.199542: step: 1098/1541, loss: 3.1182518005371094 2023-01-23 05:13:10.059112: step: 1100/1541, loss: 2.227625846862793 2023-01-23 05:13:10.980100: step: 1102/1541, loss: 2.0438344478607178 2023-01-23 05:13:11.877155: step: 1104/1541, loss: 2.334669589996338 2023-01-23 05:13:12.815443: step: 1106/1541, loss: 2.970388889312744 2023-01-23 05:13:13.704540: step: 1108/1541, loss: 2.2319881916046143 2023-01-23 05:13:14.641570: step: 1110/1541, loss: 3.5768697261810303 2023-01-23 05:13:15.591532: step: 1112/1541, loss: 1.1931216716766357 2023-01-23 05:13:16.485396: step: 1114/1541, loss: 3.0863451957702637 2023-01-23 05:13:17.345635: step: 1116/1541, loss: 2.7503433227539062 2023-01-23 05:13:18.237892: step: 1118/1541, loss: 4.187647819519043 2023-01-23 05:13:19.121410: step: 1120/1541, loss: 3.397799491882324 2023-01-23 05:13:20.086999: step: 1122/1541, loss: 2.905571460723877 2023-01-23 05:13:21.001752: step: 1124/1541, loss: 2.3993916511535645 2023-01-23 05:13:21.908631: step: 1126/1541, loss: 1.2772283554077148 2023-01-23 05:13:22.823139: step: 1128/1541, loss: 2.7718563079833984 2023-01-23 05:13:23.705615: step: 1130/1541, loss: 2.9497885704040527 2023-01-23 05:13:24.590313: step: 1132/1541, loss: 5.788618564605713 2023-01-23 05:13:25.509612: step: 1134/1541, loss: 2.3953065872192383 2023-01-23 05:13:26.427055: step: 1136/1541, loss: 2.9485535621643066 2023-01-23 05:13:27.374422: step: 1138/1541, loss: 2.956460475921631 2023-01-23 05:13:28.273457: step: 1140/1541, loss: 1.92601478099823 2023-01-23 05:13:29.126059: step: 1142/1541, loss: 2.2318456172943115 2023-01-23 05:13:30.019228: step: 1144/1541, loss: 1.1826106309890747 2023-01-23 05:13:30.935285: step: 1146/1541, loss: 3.082468032836914 2023-01-23 05:13:31.937501: step: 1148/1541, loss: 3.633897304534912 2023-01-23 05:13:32.827510: step: 1150/1541, loss: 1.7797327041625977 2023-01-23 05:13:33.734293: step: 1152/1541, loss: 8.065730094909668 2023-01-23 05:13:34.662479: step: 1154/1541, loss: 2.5585179328918457 2023-01-23 05:13:35.555099: step: 1156/1541, loss: 2.1464245319366455 2023-01-23 05:13:36.439829: step: 1158/1541, loss: 2.1504476070404053 2023-01-23 05:13:37.324710: step: 1160/1541, loss: 2.9259989261627197 2023-01-23 05:13:38.220097: step: 1162/1541, loss: 2.590718984603882 2023-01-23 05:13:39.128542: step: 1164/1541, loss: 1.5548310279846191 2023-01-23 05:13:40.029612: step: 1166/1541, loss: 2.1792242527008057 2023-01-23 05:13:40.927710: step: 1168/1541, loss: 1.991245985031128 2023-01-23 05:13:41.833105: step: 1170/1541, loss: 4.491457939147949 2023-01-23 05:13:42.725201: step: 1172/1541, loss: 7.264886856079102 2023-01-23 05:13:43.626927: step: 1174/1541, loss: 2.449263095855713 2023-01-23 05:13:44.530346: step: 1176/1541, loss: 2.3786797523498535 2023-01-23 05:13:45.449610: step: 1178/1541, loss: 1.776977777481079 2023-01-23 05:13:46.311070: step: 1180/1541, loss: 2.6962060928344727 2023-01-23 05:13:47.286399: step: 1182/1541, loss: 4.0183424949646 2023-01-23 05:13:48.144713: step: 1184/1541, loss: 2.4782698154449463 2023-01-23 05:13:49.015539: step: 1186/1541, loss: 5.091773509979248 2023-01-23 05:13:49.904857: step: 1188/1541, loss: 11.428810119628906 2023-01-23 05:13:50.812065: step: 1190/1541, loss: 2.476348876953125 2023-01-23 05:13:51.723861: step: 1192/1541, loss: 3.0601916313171387 2023-01-23 05:13:52.662019: step: 1194/1541, loss: 2.3158273696899414 2023-01-23 05:13:53.563719: step: 1196/1541, loss: 2.029017925262451 2023-01-23 05:13:54.436402: step: 1198/1541, loss: 2.884587049484253 2023-01-23 05:13:55.360874: step: 1200/1541, loss: 4.2901506423950195 2023-01-23 05:13:56.215567: step: 1202/1541, loss: 3.2197890281677246 2023-01-23 05:13:57.106549: step: 1204/1541, loss: 2.672617197036743 2023-01-23 05:13:58.016333: step: 1206/1541, loss: 6.09974479675293 2023-01-23 05:13:58.956658: step: 1208/1541, loss: 3.349771499633789 2023-01-23 05:13:59.875473: step: 1210/1541, loss: 2.1177408695220947 2023-01-23 05:14:00.822948: step: 1212/1541, loss: 2.278090238571167 2023-01-23 05:14:01.709229: step: 1214/1541, loss: 3.045269012451172 2023-01-23 05:14:02.641547: step: 1216/1541, loss: 2.15691876411438 2023-01-23 05:14:03.533287: step: 1218/1541, loss: 5.266175746917725 2023-01-23 05:14:04.476827: step: 1220/1541, loss: 4.545399188995361 2023-01-23 05:14:05.357804: step: 1222/1541, loss: 1.2207260131835938 2023-01-23 05:14:06.245383: step: 1224/1541, loss: 3.574289560317993 2023-01-23 05:14:07.165867: step: 1226/1541, loss: 10.766623497009277 2023-01-23 05:14:08.063354: step: 1228/1541, loss: 1.8134820461273193 2023-01-23 05:14:08.966213: step: 1230/1541, loss: 2.3146982192993164 2023-01-23 05:14:09.856053: step: 1232/1541, loss: 4.067654609680176 2023-01-23 05:14:10.739349: step: 1234/1541, loss: 1.92323637008667 2023-01-23 05:14:11.594821: step: 1236/1541, loss: 1.8503495454788208 2023-01-23 05:14:12.495425: step: 1238/1541, loss: 2.4183566570281982 2023-01-23 05:14:13.366990: step: 1240/1541, loss: 2.559445858001709 2023-01-23 05:14:14.220144: step: 1242/1541, loss: 1.3276734352111816 2023-01-23 05:14:15.087399: step: 1244/1541, loss: 1.5349628925323486 2023-01-23 05:14:15.991916: step: 1246/1541, loss: 2.1050829887390137 2023-01-23 05:14:16.904971: step: 1248/1541, loss: 5.284878730773926 2023-01-23 05:14:17.775095: step: 1250/1541, loss: 2.610255479812622 2023-01-23 05:14:18.680032: step: 1252/1541, loss: 3.5091264247894287 2023-01-23 05:14:19.567944: step: 1254/1541, loss: 6.008684158325195 2023-01-23 05:14:20.438799: step: 1256/1541, loss: 3.4649412631988525 2023-01-23 05:14:21.356922: step: 1258/1541, loss: 1.8123550415039062 2023-01-23 05:14:22.248411: step: 1260/1541, loss: 5.2001261711120605 2023-01-23 05:14:23.146358: step: 1262/1541, loss: 2.8271772861480713 2023-01-23 05:14:24.054884: step: 1264/1541, loss: 5.721043586730957 2023-01-23 05:14:24.958063: step: 1266/1541, loss: 2.284836530685425 2023-01-23 05:14:25.836475: step: 1268/1541, loss: 2.2219622135162354 2023-01-23 05:14:26.713616: step: 1270/1541, loss: 1.4908833503723145 2023-01-23 05:14:27.572601: step: 1272/1541, loss: 2.549482583999634 2023-01-23 05:14:28.423436: step: 1274/1541, loss: 1.6297540664672852 2023-01-23 05:14:29.282771: step: 1276/1541, loss: 2.7403693199157715 2023-01-23 05:14:30.164276: step: 1278/1541, loss: 11.338337898254395 2023-01-23 05:14:31.078303: step: 1280/1541, loss: 1.9585529565811157 2023-01-23 05:14:31.973754: step: 1282/1541, loss: 4.671322345733643 2023-01-23 05:14:32.879859: step: 1284/1541, loss: 3.0727484226226807 2023-01-23 05:14:33.763573: step: 1286/1541, loss: 3.8632988929748535 2023-01-23 05:14:34.679503: step: 1288/1541, loss: 1.533441185951233 2023-01-23 05:14:35.561716: step: 1290/1541, loss: 4.702691078186035 2023-01-23 05:14:36.464493: step: 1292/1541, loss: 1.9407955408096313 2023-01-23 05:14:37.360626: step: 1294/1541, loss: 2.6487507820129395 2023-01-23 05:14:38.237548: step: 1296/1541, loss: 2.0427303314208984 2023-01-23 05:14:39.135636: step: 1298/1541, loss: 3.0747666358947754 2023-01-23 05:14:40.029236: step: 1300/1541, loss: 1.5445908308029175 2023-01-23 05:14:40.957969: step: 1302/1541, loss: 4.65545129776001 2023-01-23 05:14:41.877146: step: 1304/1541, loss: 1.9940463304519653 2023-01-23 05:14:42.855632: step: 1306/1541, loss: 2.6490871906280518 2023-01-23 05:14:43.768536: step: 1308/1541, loss: 1.9622193574905396 2023-01-23 05:14:44.652933: step: 1310/1541, loss: 2.5327303409576416 2023-01-23 05:14:45.541022: step: 1312/1541, loss: 2.873953342437744 2023-01-23 05:14:46.408581: step: 1314/1541, loss: 2.0306336879730225 2023-01-23 05:14:47.307711: step: 1316/1541, loss: 2.603915214538574 2023-01-23 05:14:48.243465: step: 1318/1541, loss: 3.0166757106781006 2023-01-23 05:14:49.139028: step: 1320/1541, loss: 1.0980439186096191 2023-01-23 05:14:50.020947: step: 1322/1541, loss: 9.320212364196777 2023-01-23 05:14:50.895412: step: 1324/1541, loss: 2.0959091186523438 2023-01-23 05:14:51.752746: step: 1326/1541, loss: 1.4901636838912964 2023-01-23 05:14:52.664262: step: 1328/1541, loss: 9.192411422729492 2023-01-23 05:14:53.645352: step: 1330/1541, loss: 2.1900711059570312 2023-01-23 05:14:54.524793: step: 1332/1541, loss: 2.405038833618164 2023-01-23 05:14:55.391414: step: 1334/1541, loss: 2.2324180603027344 2023-01-23 05:14:56.295409: step: 1336/1541, loss: 1.9735215902328491 2023-01-23 05:14:57.177842: step: 1338/1541, loss: 1.7913317680358887 2023-01-23 05:14:58.031769: step: 1340/1541, loss: 1.5863088369369507 2023-01-23 05:14:58.926538: step: 1342/1541, loss: 1.6174112558364868 2023-01-23 05:14:59.803065: step: 1344/1541, loss: 2.5715749263763428 2023-01-23 05:15:00.693502: step: 1346/1541, loss: 2.3911454677581787 2023-01-23 05:15:01.599316: step: 1348/1541, loss: 2.308403491973877 2023-01-23 05:15:02.492618: step: 1350/1541, loss: 2.102966785430908 2023-01-23 05:15:03.380511: step: 1352/1541, loss: 3.908891439437866 2023-01-23 05:15:04.231875: step: 1354/1541, loss: 2.9518730640411377 2023-01-23 05:15:05.119168: step: 1356/1541, loss: 2.6161553859710693 2023-01-23 05:15:05.981295: step: 1358/1541, loss: 2.9736077785491943 2023-01-23 05:15:06.868279: step: 1360/1541, loss: 4.915577411651611 2023-01-23 05:15:07.767500: step: 1362/1541, loss: 2.8604040145874023 2023-01-23 05:15:08.613391: step: 1364/1541, loss: 1.261566162109375 2023-01-23 05:15:09.459219: step: 1366/1541, loss: 2.3064424991607666 2023-01-23 05:15:10.344351: step: 1368/1541, loss: 1.7926650047302246 2023-01-23 05:15:11.240997: step: 1370/1541, loss: 2.054575204849243 2023-01-23 05:15:12.107931: step: 1372/1541, loss: 1.849778175354004 2023-01-23 05:15:13.029492: step: 1374/1541, loss: 1.39596688747406 2023-01-23 05:15:14.002417: step: 1376/1541, loss: 2.00221848487854 2023-01-23 05:15:14.892386: step: 1378/1541, loss: 4.675865173339844 2023-01-23 05:15:15.778022: step: 1380/1541, loss: 1.4164812564849854 2023-01-23 05:15:16.699545: step: 1382/1541, loss: 5.748347282409668 2023-01-23 05:15:17.574078: step: 1384/1541, loss: 1.5301847457885742 2023-01-23 05:15:18.442015: step: 1386/1541, loss: 2.31630277633667 2023-01-23 05:15:19.322258: step: 1388/1541, loss: 1.8464980125427246 2023-01-23 05:15:20.192658: step: 1390/1541, loss: 2.5196242332458496 2023-01-23 05:15:21.086369: step: 1392/1541, loss: 2.9216949939727783 2023-01-23 05:15:21.945912: step: 1394/1541, loss: 3.7964606285095215 2023-01-23 05:15:22.800840: step: 1396/1541, loss: 3.6977591514587402 2023-01-23 05:15:23.684904: step: 1398/1541, loss: 2.633193254470825 2023-01-23 05:15:24.565987: step: 1400/1541, loss: 1.5678797960281372 2023-01-23 05:15:25.408422: step: 1402/1541, loss: 5.669009208679199 2023-01-23 05:15:26.288075: step: 1404/1541, loss: 2.648658514022827 2023-01-23 05:15:27.225424: step: 1406/1541, loss: 2.122039318084717 2023-01-23 05:15:28.106741: step: 1408/1541, loss: 1.841199517250061 2023-01-23 05:15:29.009298: step: 1410/1541, loss: 1.55251145362854 2023-01-23 05:15:29.905461: step: 1412/1541, loss: 0.8561643958091736 2023-01-23 05:15:30.821201: step: 1414/1541, loss: 2.264878749847412 2023-01-23 05:15:31.729864: step: 1416/1541, loss: 3.74137282371521 2023-01-23 05:15:32.658650: step: 1418/1541, loss: 5.1886887550354 2023-01-23 05:15:33.525035: step: 1420/1541, loss: 2.6090545654296875 2023-01-23 05:15:34.403115: step: 1422/1541, loss: 3.064929962158203 2023-01-23 05:15:35.332764: step: 1424/1541, loss: 2.4369800090789795 2023-01-23 05:15:36.219600: step: 1426/1541, loss: 2.2719383239746094 2023-01-23 05:15:37.147481: step: 1428/1541, loss: 2.7070813179016113 2023-01-23 05:15:38.034690: step: 1430/1541, loss: 2.390378713607788 2023-01-23 05:15:38.939091: step: 1432/1541, loss: 5.150574684143066 2023-01-23 05:15:39.820653: step: 1434/1541, loss: 3.1126508712768555 2023-01-23 05:15:40.673615: step: 1436/1541, loss: 3.903477430343628 2023-01-23 05:15:41.578972: step: 1438/1541, loss: 1.7687982320785522 2023-01-23 05:15:42.464992: step: 1440/1541, loss: 2.0287177562713623 2023-01-23 05:15:43.361564: step: 1442/1541, loss: 1.8782650232315063 2023-01-23 05:15:44.264547: step: 1444/1541, loss: 1.9137734174728394 2023-01-23 05:15:45.182909: step: 1446/1541, loss: 5.08827018737793 2023-01-23 05:15:46.045906: step: 1448/1541, loss: 3.957463026046753 2023-01-23 05:15:46.911096: step: 1450/1541, loss: 1.4149820804595947 2023-01-23 05:15:47.796725: step: 1452/1541, loss: 2.003631114959717 2023-01-23 05:15:48.724823: step: 1454/1541, loss: 4.629313945770264 2023-01-23 05:15:49.609010: step: 1456/1541, loss: 2.420494556427002 2023-01-23 05:15:50.493280: step: 1458/1541, loss: 9.544032096862793 2023-01-23 05:15:51.411934: step: 1460/1541, loss: 2.560499429702759 2023-01-23 05:15:52.385264: step: 1462/1541, loss: 2.9787518978118896 2023-01-23 05:15:53.285631: step: 1464/1541, loss: 1.8794710636138916 2023-01-23 05:15:54.153979: step: 1466/1541, loss: 2.1122448444366455 2023-01-23 05:15:55.115992: step: 1468/1541, loss: 2.694821834564209 2023-01-23 05:15:56.013602: step: 1470/1541, loss: 1.1753813028335571 2023-01-23 05:15:56.995040: step: 1472/1541, loss: 2.066025495529175 2023-01-23 05:15:57.926387: step: 1474/1541, loss: 2.0301175117492676 2023-01-23 05:15:58.821060: step: 1476/1541, loss: 1.4194300174713135 2023-01-23 05:15:59.735456: step: 1478/1541, loss: 1.7516411542892456 2023-01-23 05:16:00.595586: step: 1480/1541, loss: 2.2055788040161133 2023-01-23 05:16:01.606759: step: 1482/1541, loss: 1.803576111793518 2023-01-23 05:16:02.470297: step: 1484/1541, loss: 3.2596323490142822 2023-01-23 05:16:03.335703: step: 1486/1541, loss: 1.5952363014221191 2023-01-23 05:16:04.262223: step: 1488/1541, loss: 2.0296571254730225 2023-01-23 05:16:05.127653: step: 1490/1541, loss: 2.84836483001709 2023-01-23 05:16:06.022564: step: 1492/1541, loss: 3.018235921859741 2023-01-23 05:16:06.887950: step: 1494/1541, loss: 1.8847817182540894 2023-01-23 05:16:07.822105: step: 1496/1541, loss: 3.0650854110717773 2023-01-23 05:16:08.734302: step: 1498/1541, loss: 1.731103777885437 2023-01-23 05:16:09.605996: step: 1500/1541, loss: 4.799679756164551 2023-01-23 05:16:10.518370: step: 1502/1541, loss: 4.205718994140625 2023-01-23 05:16:11.439000: step: 1504/1541, loss: 3.369028329849243 2023-01-23 05:16:12.372065: step: 1506/1541, loss: 2.630521774291992 2023-01-23 05:16:13.272259: step: 1508/1541, loss: 2.327301502227783 2023-01-23 05:16:14.166984: step: 1510/1541, loss: 8.271418571472168 2023-01-23 05:16:15.114722: step: 1512/1541, loss: 2.3474457263946533 2023-01-23 05:16:16.021141: step: 1514/1541, loss: 1.472830891609192 2023-01-23 05:16:16.921617: step: 1516/1541, loss: 1.925399661064148 2023-01-23 05:16:17.785198: step: 1518/1541, loss: 1.3310664892196655 2023-01-23 05:16:18.746795: step: 1520/1541, loss: 8.94774055480957 2023-01-23 05:16:19.670007: step: 1522/1541, loss: 2.609194755554199 2023-01-23 05:16:20.531903: step: 1524/1541, loss: 3.6626527309417725 2023-01-23 05:16:21.437854: step: 1526/1541, loss: 2.1052472591400146 2023-01-23 05:16:22.315013: step: 1528/1541, loss: 2.5507609844207764 2023-01-23 05:16:23.216661: step: 1530/1541, loss: 1.3927693367004395 2023-01-23 05:16:24.143863: step: 1532/1541, loss: 1.5520473718643188 2023-01-23 05:16:25.046869: step: 1534/1541, loss: 3.9288277626037598 2023-01-23 05:16:25.887565: step: 1536/1541, loss: 1.9424042701721191 2023-01-23 05:16:26.855046: step: 1538/1541, loss: 1.7202082872390747 2023-01-23 05:16:27.750816: step: 1540/1541, loss: 2.195024013519287 2023-01-23 05:16:28.678577: step: 1542/1541, loss: 3.359971761703491 2023-01-23 05:16:29.559009: step: 1544/1541, loss: 1.0830938816070557 2023-01-23 05:16:30.442263: step: 1546/1541, loss: 2.346796751022339 2023-01-23 05:16:31.326255: step: 1548/1541, loss: 3.2162044048309326 2023-01-23 05:16:32.223413: step: 1550/1541, loss: 4.4385786056518555 2023-01-23 05:16:33.102459: step: 1552/1541, loss: 3.0257904529571533 2023-01-23 05:16:33.991835: step: 1554/1541, loss: 2.044212579727173 2023-01-23 05:16:34.904327: step: 1556/1541, loss: 3.441161632537842 2023-01-23 05:16:35.751160: step: 1558/1541, loss: 1.8406471014022827 2023-01-23 05:16:36.624413: step: 1560/1541, loss: 1.5935579538345337 2023-01-23 05:16:37.516287: step: 1562/1541, loss: 3.742830753326416 2023-01-23 05:16:38.423015: step: 1564/1541, loss: 4.485775470733643 2023-01-23 05:16:39.261229: step: 1566/1541, loss: 1.5484029054641724 2023-01-23 05:16:40.106756: step: 1568/1541, loss: 0.8182669878005981 2023-01-23 05:16:41.029092: step: 1570/1541, loss: 4.35479211807251 2023-01-23 05:16:41.955340: step: 1572/1541, loss: 2.080078363418579 2023-01-23 05:16:42.867033: step: 1574/1541, loss: 3.114391326904297 2023-01-23 05:16:43.720968: step: 1576/1541, loss: 4.810726642608643 2023-01-23 05:16:44.616736: step: 1578/1541, loss: 2.766826868057251 2023-01-23 05:16:45.499298: step: 1580/1541, loss: 2.9865684509277344 2023-01-23 05:16:46.377707: step: 1582/1541, loss: 1.4297053813934326 2023-01-23 05:16:47.277626: step: 1584/1541, loss: 2.0832090377807617 2023-01-23 05:16:48.150959: step: 1586/1541, loss: 4.819056034088135 2023-01-23 05:16:49.034001: step: 1588/1541, loss: 4.601319789886475 2023-01-23 05:16:49.970635: step: 1590/1541, loss: 1.2554559707641602 2023-01-23 05:16:50.908025: step: 1592/1541, loss: 4.708141326904297 2023-01-23 05:16:51.864111: step: 1594/1541, loss: 3.760302782058716 2023-01-23 05:16:52.769720: step: 1596/1541, loss: 3.6788742542266846 2023-01-23 05:16:53.681944: step: 1598/1541, loss: 2.1046950817108154 2023-01-23 05:16:54.581902: step: 1600/1541, loss: 1.8405556678771973 2023-01-23 05:16:55.458630: step: 1602/1541, loss: 1.959875226020813 2023-01-23 05:16:56.357843: step: 1604/1541, loss: 4.103623867034912 2023-01-23 05:16:57.302547: step: 1606/1541, loss: 1.4673289060592651 2023-01-23 05:16:58.177572: step: 1608/1541, loss: 2.0023860931396484 2023-01-23 05:16:59.028691: step: 1610/1541, loss: 1.1942498683929443 2023-01-23 05:16:59.880098: step: 1612/1541, loss: 1.7219032049179077 2023-01-23 05:17:00.754498: step: 1614/1541, loss: 2.9655771255493164 2023-01-23 05:17:01.677193: step: 1616/1541, loss: 4.83092737197876 2023-01-23 05:17:02.598695: step: 1618/1541, loss: 2.611741065979004 2023-01-23 05:17:03.517758: step: 1620/1541, loss: 2.470998525619507 2023-01-23 05:17:04.425239: step: 1622/1541, loss: 2.119483709335327 2023-01-23 05:17:05.293920: step: 1624/1541, loss: 1.9544509649276733 2023-01-23 05:17:06.235763: step: 1626/1541, loss: 4.982128143310547 2023-01-23 05:17:07.122745: step: 1628/1541, loss: 3.7605903148651123 2023-01-23 05:17:08.063583: step: 1630/1541, loss: 3.179110050201416 2023-01-23 05:17:08.942889: step: 1632/1541, loss: 1.633370041847229 2023-01-23 05:17:09.829976: step: 1634/1541, loss: 2.6311380863189697 2023-01-23 05:17:10.776592: step: 1636/1541, loss: 2.4847593307495117 2023-01-23 05:17:11.673401: step: 1638/1541, loss: 1.3703001737594604 2023-01-23 05:17:12.579564: step: 1640/1541, loss: 2.6462574005126953 2023-01-23 05:17:13.495867: step: 1642/1541, loss: 5.701605796813965 2023-01-23 05:17:14.409724: step: 1644/1541, loss: 1.644205093383789 2023-01-23 05:17:15.276162: step: 1646/1541, loss: 2.421766757965088 2023-01-23 05:17:16.172041: step: 1648/1541, loss: 2.023583173751831 2023-01-23 05:17:17.101371: step: 1650/1541, loss: 3.18644380569458 2023-01-23 05:17:17.967736: step: 1652/1541, loss: 2.698401927947998 2023-01-23 05:17:18.841649: step: 1654/1541, loss: 1.634724736213684 2023-01-23 05:17:19.716161: step: 1656/1541, loss: 3.7199199199676514 2023-01-23 05:17:20.608018: step: 1658/1541, loss: 3.6199307441711426 2023-01-23 05:17:21.481336: step: 1660/1541, loss: 1.2353140115737915 2023-01-23 05:17:22.367126: step: 1662/1541, loss: 2.3276619911193848 2023-01-23 05:17:23.308896: step: 1664/1541, loss: 6.194576263427734 2023-01-23 05:17:24.187876: step: 1666/1541, loss: 4.4523606300354 2023-01-23 05:17:25.069240: step: 1668/1541, loss: 4.204910755157471 2023-01-23 05:17:25.961995: step: 1670/1541, loss: 2.7184667587280273 2023-01-23 05:17:26.835729: step: 1672/1541, loss: 4.906307220458984 2023-01-23 05:17:27.727122: step: 1674/1541, loss: 4.747424125671387 2023-01-23 05:17:28.619327: step: 1676/1541, loss: 2.3073365688323975 2023-01-23 05:17:29.508214: step: 1678/1541, loss: 1.9376428127288818 2023-01-23 05:17:30.374157: step: 1680/1541, loss: 1.3919730186462402 2023-01-23 05:17:31.229686: step: 1682/1541, loss: 2.0704524517059326 2023-01-23 05:17:32.091037: step: 1684/1541, loss: 2.304966449737549 2023-01-23 05:17:32.985945: step: 1686/1541, loss: 2.562845468521118 2023-01-23 05:17:33.864635: step: 1688/1541, loss: 1.5487667322158813 2023-01-23 05:17:34.775014: step: 1690/1541, loss: 1.9541542530059814 2023-01-23 05:17:35.666294: step: 1692/1541, loss: 1.03623366355896 2023-01-23 05:17:36.547901: step: 1694/1541, loss: 1.4680819511413574 2023-01-23 05:17:37.435104: step: 1696/1541, loss: 1.1064772605895996 2023-01-23 05:17:38.299506: step: 1698/1541, loss: 2.0264017581939697 2023-01-23 05:17:39.171855: step: 1700/1541, loss: 2.661195755004883 2023-01-23 05:17:40.109291: step: 1702/1541, loss: 1.9155138731002808 2023-01-23 05:17:41.008291: step: 1704/1541, loss: 1.5695405006408691 2023-01-23 05:17:41.912583: step: 1706/1541, loss: 4.617031574249268 2023-01-23 05:17:42.798581: step: 1708/1541, loss: 2.7830417156219482 2023-01-23 05:17:43.705570: step: 1710/1541, loss: 1.6343655586242676 2023-01-23 05:17:44.565675: step: 1712/1541, loss: 2.678792715072632 2023-01-23 05:17:45.475323: step: 1714/1541, loss: 2.9210610389709473 2023-01-23 05:17:46.391266: step: 1716/1541, loss: 2.180471181869507 2023-01-23 05:17:47.257397: step: 1718/1541, loss: 3.7415356636047363 2023-01-23 05:17:48.121310: step: 1720/1541, loss: 1.700747013092041 2023-01-23 05:17:49.017505: step: 1722/1541, loss: 1.5951017141342163 2023-01-23 05:17:49.894868: step: 1724/1541, loss: 1.4442273378372192 2023-01-23 05:17:50.797750: step: 1726/1541, loss: 4.217314720153809 2023-01-23 05:17:51.662704: step: 1728/1541, loss: 1.3286865949630737 2023-01-23 05:17:52.555777: step: 1730/1541, loss: 4.380549430847168 2023-01-23 05:17:53.389102: step: 1732/1541, loss: 4.561209678649902 2023-01-23 05:17:54.236754: step: 1734/1541, loss: 2.8032755851745605 2023-01-23 05:17:55.104657: step: 1736/1541, loss: 1.5037816762924194 2023-01-23 05:17:55.987406: step: 1738/1541, loss: 1.8783172369003296 2023-01-23 05:17:56.889079: step: 1740/1541, loss: 2.155783176422119 2023-01-23 05:17:57.774660: step: 1742/1541, loss: 1.2596371173858643 2023-01-23 05:17:58.717638: step: 1744/1541, loss: 1.3074121475219727 2023-01-23 05:17:59.626069: step: 1746/1541, loss: 2.4874355792999268 2023-01-23 05:18:00.514389: step: 1748/1541, loss: 4.3300580978393555 2023-01-23 05:18:01.443462: step: 1750/1541, loss: 2.145622730255127 2023-01-23 05:18:02.344889: step: 1752/1541, loss: 3.464348793029785 2023-01-23 05:18:03.264009: step: 1754/1541, loss: 2.1928517818450928 2023-01-23 05:18:04.164169: step: 1756/1541, loss: 2.0173938274383545 2023-01-23 05:18:05.053612: step: 1758/1541, loss: 1.0779139995574951 2023-01-23 05:18:05.929001: step: 1760/1541, loss: 1.655928611755371 2023-01-23 05:18:06.826798: step: 1762/1541, loss: 1.9884828329086304 2023-01-23 05:18:07.703325: step: 1764/1541, loss: 4.064584732055664 2023-01-23 05:18:08.559597: step: 1766/1541, loss: 1.7070220708847046 2023-01-23 05:18:09.431013: step: 1768/1541, loss: 1.5534310340881348 2023-01-23 05:18:10.317563: step: 1770/1541, loss: 1.3992664813995361 2023-01-23 05:18:11.177303: step: 1772/1541, loss: 1.7737503051757812 2023-01-23 05:18:12.025653: step: 1774/1541, loss: 1.0296419858932495 2023-01-23 05:18:12.937565: step: 1776/1541, loss: 1.7880563735961914 2023-01-23 05:18:13.823717: step: 1778/1541, loss: 2.029548406600952 2023-01-23 05:18:14.809992: step: 1780/1541, loss: 6.533682346343994 2023-01-23 05:18:15.754931: step: 1782/1541, loss: 2.3393123149871826 2023-01-23 05:18:16.661473: step: 1784/1541, loss: 1.8221287727355957 2023-01-23 05:18:17.549623: step: 1786/1541, loss: 1.6811237335205078 2023-01-23 05:18:18.444956: step: 1788/1541, loss: 2.5717175006866455 2023-01-23 05:18:19.336459: step: 1790/1541, loss: 2.0062665939331055 2023-01-23 05:18:20.233743: step: 1792/1541, loss: 4.314602851867676 2023-01-23 05:18:21.142273: step: 1794/1541, loss: 1.6795583963394165 2023-01-23 05:18:22.007868: step: 1796/1541, loss: 2.230997085571289 2023-01-23 05:18:22.966509: step: 1798/1541, loss: 2.2451236248016357 2023-01-23 05:18:23.845667: step: 1800/1541, loss: 2.4006266593933105 2023-01-23 05:18:24.799839: step: 1802/1541, loss: 4.060626029968262 2023-01-23 05:18:25.666844: step: 1804/1541, loss: 3.7821407318115234 2023-01-23 05:18:26.556353: step: 1806/1541, loss: 2.5280745029449463 2023-01-23 05:18:27.451295: step: 1808/1541, loss: 4.150521278381348 2023-01-23 05:18:28.288310: step: 1810/1541, loss: 1.8781874179840088 2023-01-23 05:18:29.173741: step: 1812/1541, loss: 1.4667001962661743 2023-01-23 05:18:30.084155: step: 1814/1541, loss: 10.136836051940918 2023-01-23 05:18:30.954212: step: 1816/1541, loss: 4.275468826293945 2023-01-23 05:18:31.813160: step: 1818/1541, loss: 1.8006255626678467 2023-01-23 05:18:32.772795: step: 1820/1541, loss: 2.713373899459839 2023-01-23 05:18:33.626886: step: 1822/1541, loss: 1.853053331375122 2023-01-23 05:18:34.510138: step: 1824/1541, loss: 0.8215959668159485 2023-01-23 05:18:35.412325: step: 1826/1541, loss: 6.4982123374938965 2023-01-23 05:18:36.291567: step: 1828/1541, loss: 2.559427261352539 2023-01-23 05:18:37.164801: step: 1830/1541, loss: 1.902858853340149 2023-01-23 05:18:38.036255: step: 1832/1541, loss: 1.384061336517334 2023-01-23 05:18:38.909621: step: 1834/1541, loss: 1.9659199714660645 2023-01-23 05:18:39.792961: step: 1836/1541, loss: 4.72265625 2023-01-23 05:18:40.674158: step: 1838/1541, loss: 3.7287821769714355 2023-01-23 05:18:41.562811: step: 1840/1541, loss: 1.6320637464523315 2023-01-23 05:18:42.442131: step: 1842/1541, loss: 2.6251752376556396 2023-01-23 05:18:43.349042: step: 1844/1541, loss: 1.0906521081924438 2023-01-23 05:18:44.191923: step: 1846/1541, loss: 3.562819004058838 2023-01-23 05:18:45.121757: step: 1848/1541, loss: 1.761415958404541 2023-01-23 05:18:46.001910: step: 1850/1541, loss: 4.214339256286621 2023-01-23 05:18:46.891966: step: 1852/1541, loss: 2.5246267318725586 2023-01-23 05:18:47.777090: step: 1854/1541, loss: 2.766356945037842 2023-01-23 05:18:48.668402: step: 1856/1541, loss: 1.7583528757095337 2023-01-23 05:18:49.598461: step: 1858/1541, loss: 3.4718315601348877 2023-01-23 05:18:50.462680: step: 1860/1541, loss: 1.7508716583251953 2023-01-23 05:18:51.313186: step: 1862/1541, loss: 2.0509541034698486 2023-01-23 05:18:52.163933: step: 1864/1541, loss: 1.1424425840377808 2023-01-23 05:18:53.063054: step: 1866/1541, loss: 2.5271494388580322 2023-01-23 05:18:53.950679: step: 1868/1541, loss: 2.117082357406616 2023-01-23 05:18:54.851741: step: 1870/1541, loss: 1.570325255393982 2023-01-23 05:18:55.720851: step: 1872/1541, loss: 1.912516474723816 2023-01-23 05:18:56.657520: step: 1874/1541, loss: 1.9841355085372925 2023-01-23 05:18:57.555109: step: 1876/1541, loss: 2.1667189598083496 2023-01-23 05:18:58.459548: step: 1878/1541, loss: 1.9171863794326782 2023-01-23 05:18:59.358050: step: 1880/1541, loss: 3.454853057861328 2023-01-23 05:19:00.317046: step: 1882/1541, loss: 4.412927627563477 2023-01-23 05:19:01.208053: step: 1884/1541, loss: 4.0674967765808105 2023-01-23 05:19:02.093725: step: 1886/1541, loss: 2.2278952598571777 2023-01-23 05:19:02.983485: step: 1888/1541, loss: 1.4100065231323242 2023-01-23 05:19:03.901842: step: 1890/1541, loss: 1.0073977708816528 2023-01-23 05:19:04.845931: step: 1892/1541, loss: 2.0201308727264404 2023-01-23 05:19:05.765994: step: 1894/1541, loss: 0.9979107975959778 2023-01-23 05:19:06.637976: step: 1896/1541, loss: 1.8325252532958984 2023-01-23 05:19:07.528711: step: 1898/1541, loss: 1.9774835109710693 2023-01-23 05:19:08.431545: step: 1900/1541, loss: 1.2547006607055664 2023-01-23 05:19:09.344612: step: 1902/1541, loss: 1.5919681787490845 2023-01-23 05:19:10.236492: step: 1904/1541, loss: 2.8399834632873535 2023-01-23 05:19:11.118228: step: 1906/1541, loss: 3.387883424758911 2023-01-23 05:19:12.025015: step: 1908/1541, loss: 1.430403470993042 2023-01-23 05:19:12.895795: step: 1910/1541, loss: 3.9390780925750732 2023-01-23 05:19:13.792241: step: 1912/1541, loss: 2.5167601108551025 2023-01-23 05:19:14.683276: step: 1914/1541, loss: 1.110376238822937 2023-01-23 05:19:15.553736: step: 1916/1541, loss: 8.161396980285645 2023-01-23 05:19:16.461752: step: 1918/1541, loss: 1.2628825902938843 2023-01-23 05:19:17.341936: step: 1920/1541, loss: 1.138550043106079 2023-01-23 05:19:18.239167: step: 1922/1541, loss: 1.8771400451660156 2023-01-23 05:19:19.133871: step: 1924/1541, loss: 3.609248161315918 2023-01-23 05:19:20.009487: step: 1926/1541, loss: 1.1045013666152954 2023-01-23 05:19:20.885713: step: 1928/1541, loss: 2.812190532684326 2023-01-23 05:19:21.777668: step: 1930/1541, loss: 1.0642207860946655 2023-01-23 05:19:22.625011: step: 1932/1541, loss: 4.5081868171691895 2023-01-23 05:19:23.488329: step: 1934/1541, loss: 5.363675117492676 2023-01-23 05:19:24.397673: step: 1936/1541, loss: 1.9983315467834473 2023-01-23 05:19:25.317175: step: 1938/1541, loss: 1.9214266538619995 2023-01-23 05:19:26.234795: step: 1940/1541, loss: 4.118586540222168 2023-01-23 05:19:27.124942: step: 1942/1541, loss: 4.102306842803955 2023-01-23 05:19:28.022564: step: 1944/1541, loss: 1.7031679153442383 2023-01-23 05:19:28.932847: step: 1946/1541, loss: 2.9995357990264893 2023-01-23 05:19:29.819267: step: 1948/1541, loss: 5.133607864379883 2023-01-23 05:19:30.709111: step: 1950/1541, loss: 1.9347405433654785 2023-01-23 05:19:31.600774: step: 1952/1541, loss: 1.56049382686615 2023-01-23 05:19:32.473960: step: 1954/1541, loss: 2.3906803131103516 2023-01-23 05:19:33.378003: step: 1956/1541, loss: 1.8704876899719238 2023-01-23 05:19:34.288141: step: 1958/1541, loss: 4.832060813903809 2023-01-23 05:19:35.206177: step: 1960/1541, loss: 2.2777979373931885 2023-01-23 05:19:36.101682: step: 1962/1541, loss: 4.77734375 2023-01-23 05:19:37.053155: step: 1964/1541, loss: 1.715171217918396 2023-01-23 05:19:37.914136: step: 1966/1541, loss: 1.8105462789535522 2023-01-23 05:19:38.793849: step: 1968/1541, loss: 1.402416467666626 2023-01-23 05:19:39.648970: step: 1970/1541, loss: 1.5440303087234497 2023-01-23 05:19:40.533378: step: 1972/1541, loss: 1.7403090000152588 2023-01-23 05:19:41.445869: step: 1974/1541, loss: 5.242641448974609 2023-01-23 05:19:42.334995: step: 1976/1541, loss: 1.528476595878601 2023-01-23 05:19:43.213716: step: 1978/1541, loss: 2.632462501525879 2023-01-23 05:19:44.096252: step: 1980/1541, loss: 1.9003878831863403 2023-01-23 05:19:45.026949: step: 1982/1541, loss: 1.4275633096694946 2023-01-23 05:19:45.978396: step: 1984/1541, loss: 1.5773749351501465 2023-01-23 05:19:46.929608: step: 1986/1541, loss: 1.5818686485290527 2023-01-23 05:19:47.842740: step: 1988/1541, loss: 2.9167072772979736 2023-01-23 05:19:48.712257: step: 1990/1541, loss: 2.6630327701568604 2023-01-23 05:19:49.615968: step: 1992/1541, loss: 3.6701223850250244 2023-01-23 05:19:50.504290: step: 1994/1541, loss: 3.633632183074951 2023-01-23 05:19:51.407684: step: 1996/1541, loss: 2.4668941497802734 2023-01-23 05:19:52.259074: step: 1998/1541, loss: 1.1441385746002197 2023-01-23 05:19:53.114079: step: 2000/1541, loss: 1.7054327726364136 2023-01-23 05:19:53.992705: step: 2002/1541, loss: 2.234971523284912 2023-01-23 05:19:54.839536: step: 2004/1541, loss: 4.046039581298828 2023-01-23 05:19:55.748594: step: 2006/1541, loss: 1.2547180652618408 2023-01-23 05:19:56.652321: step: 2008/1541, loss: 1.5910327434539795 2023-01-23 05:19:57.534430: step: 2010/1541, loss: 1.5502417087554932 2023-01-23 05:19:58.425946: step: 2012/1541, loss: 1.554152250289917 2023-01-23 05:19:59.287801: step: 2014/1541, loss: 2.162843942642212 2023-01-23 05:20:00.213437: step: 2016/1541, loss: 1.7943614721298218 2023-01-23 05:20:01.136263: step: 2018/1541, loss: 5.753375053405762 2023-01-23 05:20:02.013485: step: 2020/1541, loss: 2.3293728828430176 2023-01-23 05:20:02.940228: step: 2022/1541, loss: 3.205216884613037 2023-01-23 05:20:03.836660: step: 2024/1541, loss: 1.8199824094772339 2023-01-23 05:20:04.756392: step: 2026/1541, loss: 4.436468601226807 2023-01-23 05:20:05.715023: step: 2028/1541, loss: 4.37652063369751 2023-01-23 05:20:06.570653: step: 2030/1541, loss: 1.4246588945388794 2023-01-23 05:20:07.497429: step: 2032/1541, loss: 1.5509793758392334 2023-01-23 05:20:08.371509: step: 2034/1541, loss: 2.1542270183563232 2023-01-23 05:20:09.323247: step: 2036/1541, loss: 3.6518800258636475 2023-01-23 05:20:10.200286: step: 2038/1541, loss: 2.189365863800049 2023-01-23 05:20:11.059938: step: 2040/1541, loss: 1.5207124948501587 2023-01-23 05:20:11.942456: step: 2042/1541, loss: 1.746291995048523 2023-01-23 05:20:12.812857: step: 2044/1541, loss: 0.8352614045143127 2023-01-23 05:20:13.701987: step: 2046/1541, loss: 4.210352897644043 2023-01-23 05:20:14.609330: step: 2048/1541, loss: 5.812264442443848 2023-01-23 05:20:15.451301: step: 2050/1541, loss: 1.2622138261795044 2023-01-23 05:20:16.287376: step: 2052/1541, loss: 1.8035099506378174 2023-01-23 05:20:17.141019: step: 2054/1541, loss: 0.9173163175582886 2023-01-23 05:20:18.048490: step: 2056/1541, loss: 1.6232742071151733 2023-01-23 05:20:18.921824: step: 2058/1541, loss: 2.6977477073669434 2023-01-23 05:20:19.788502: step: 2060/1541, loss: 0.9550980925559998 2023-01-23 05:20:20.686514: step: 2062/1541, loss: 1.190104365348816 2023-01-23 05:20:21.553377: step: 2064/1541, loss: 1.2633552551269531 2023-01-23 05:20:22.444081: step: 2066/1541, loss: 4.20432710647583 2023-01-23 05:20:23.349768: step: 2068/1541, loss: 0.7232124209403992 2023-01-23 05:20:24.255008: step: 2070/1541, loss: 1.9956836700439453 2023-01-23 05:20:25.136247: step: 2072/1541, loss: 1.7366946935653687 2023-01-23 05:20:26.020841: step: 2074/1541, loss: 2.068406820297241 2023-01-23 05:20:26.910251: step: 2076/1541, loss: 1.7923097610473633 2023-01-23 05:20:27.804467: step: 2078/1541, loss: 1.2345894575119019 2023-01-23 05:20:28.762599: step: 2080/1541, loss: 1.2350115776062012 2023-01-23 05:20:29.683274: step: 2082/1541, loss: 7.867457389831543 2023-01-23 05:20:30.563657: step: 2084/1541, loss: 3.7712457180023193 2023-01-23 05:20:31.461069: step: 2086/1541, loss: 1.5730760097503662 2023-01-23 05:20:32.336043: step: 2088/1541, loss: 2.5892980098724365 2023-01-23 05:20:33.205369: step: 2090/1541, loss: 1.902688980102539 2023-01-23 05:20:34.154871: step: 2092/1541, loss: 2.166581153869629 2023-01-23 05:20:35.044954: step: 2094/1541, loss: 0.6016511917114258 2023-01-23 05:20:35.924454: step: 2096/1541, loss: 2.03464412689209 2023-01-23 05:20:36.816278: step: 2098/1541, loss: 2.078531265258789 2023-01-23 05:20:37.700041: step: 2100/1541, loss: 5.823918342590332 2023-01-23 05:20:38.589122: step: 2102/1541, loss: 1.6873928308486938 2023-01-23 05:20:39.477585: step: 2104/1541, loss: 2.985450506210327 2023-01-23 05:20:40.377658: step: 2106/1541, loss: 1.647905945777893 2023-01-23 05:20:41.269377: step: 2108/1541, loss: 3.8957650661468506 2023-01-23 05:20:42.177483: step: 2110/1541, loss: 1.8744374513626099 2023-01-23 05:20:43.093054: step: 2112/1541, loss: 1.3908456563949585 2023-01-23 05:20:44.040789: step: 2114/1541, loss: 1.3252264261245728 2023-01-23 05:20:44.951499: step: 2116/1541, loss: 2.3204901218414307 2023-01-23 05:20:45.832299: step: 2118/1541, loss: 1.0710622072219849 2023-01-23 05:20:46.747172: step: 2120/1541, loss: 2.173370122909546 2023-01-23 05:20:47.688286: step: 2122/1541, loss: 1.0639480352401733 2023-01-23 05:20:48.605109: step: 2124/1541, loss: 2.209861993789673 2023-01-23 05:20:49.486811: step: 2126/1541, loss: 1.4012866020202637 2023-01-23 05:20:50.397622: step: 2128/1541, loss: 1.7055704593658447 2023-01-23 05:20:51.265822: step: 2130/1541, loss: 2.694674015045166 2023-01-23 05:20:52.173654: step: 2132/1541, loss: 3.4129421710968018 2023-01-23 05:20:53.069787: step: 2134/1541, loss: 2.65954327583313 2023-01-23 05:20:53.945309: step: 2136/1541, loss: 1.435075283050537 2023-01-23 05:20:54.878670: step: 2138/1541, loss: 2.438185214996338 2023-01-23 05:20:55.743985: step: 2140/1541, loss: 2.7057764530181885 2023-01-23 05:20:56.674321: step: 2142/1541, loss: 2.060103416442871 2023-01-23 05:20:57.587983: step: 2144/1541, loss: 1.8429057598114014 2023-01-23 05:20:58.479271: step: 2146/1541, loss: 1.8159714937210083 2023-01-23 05:20:59.352833: step: 2148/1541, loss: 1.91159188747406 2023-01-23 05:21:00.284627: step: 2150/1541, loss: 3.08048677444458 2023-01-23 05:21:01.166746: step: 2152/1541, loss: 1.6190927028656006 2023-01-23 05:21:02.062615: step: 2154/1541, loss: 2.088414430618286 2023-01-23 05:21:02.980682: step: 2156/1541, loss: 2.383549690246582 2023-01-23 05:21:03.853042: step: 2158/1541, loss: 1.1233373880386353 2023-01-23 05:21:04.787561: step: 2160/1541, loss: 1.8911073207855225 2023-01-23 05:21:05.680352: step: 2162/1541, loss: 1.3582717180252075 2023-01-23 05:21:06.625915: step: 2164/1541, loss: 1.1565086841583252 2023-01-23 05:21:07.531072: step: 2166/1541, loss: 8.311968803405762 2023-01-23 05:21:08.410986: step: 2168/1541, loss: 0.9185317158699036 2023-01-23 05:21:09.314627: step: 2170/1541, loss: 2.6065282821655273 2023-01-23 05:21:10.245207: step: 2172/1541, loss: 1.4939602613449097 2023-01-23 05:21:11.098521: step: 2174/1541, loss: 0.6820581555366516 2023-01-23 05:21:12.028539: step: 2176/1541, loss: 8.620575904846191 2023-01-23 05:21:12.932724: step: 2178/1541, loss: 1.8230619430541992 2023-01-23 05:21:13.835393: step: 2180/1541, loss: 2.0165834426879883 2023-01-23 05:21:14.840600: step: 2182/1541, loss: 1.9993414878845215 2023-01-23 05:21:15.759653: step: 2184/1541, loss: 1.7548502683639526 2023-01-23 05:21:16.652390: step: 2186/1541, loss: 1.57802414894104 2023-01-23 05:21:17.568922: step: 2188/1541, loss: 1.8088868856430054 2023-01-23 05:21:18.440141: step: 2190/1541, loss: 2.935232162475586 2023-01-23 05:21:19.351896: step: 2192/1541, loss: 3.7652502059936523 2023-01-23 05:21:20.224604: step: 2194/1541, loss: 3.6256906986236572 2023-01-23 05:21:21.083119: step: 2196/1541, loss: 4.111072063446045 2023-01-23 05:21:21.972269: step: 2198/1541, loss: 0.6989383101463318 2023-01-23 05:21:22.885221: step: 2200/1541, loss: 1.9531363248825073 2023-01-23 05:21:23.769491: step: 2202/1541, loss: 1.4090478420257568 2023-01-23 05:21:24.638919: step: 2204/1541, loss: 2.120286703109741 2023-01-23 05:21:25.539043: step: 2206/1541, loss: 1.9776567220687866 2023-01-23 05:21:26.394712: step: 2208/1541, loss: 1.8622372150421143 2023-01-23 05:21:27.318349: step: 2210/1541, loss: 1.1680629253387451 2023-01-23 05:21:28.189355: step: 2212/1541, loss: 4.180178642272949 2023-01-23 05:21:29.048127: step: 2214/1541, loss: 1.3466726541519165 2023-01-23 05:21:29.930936: step: 2216/1541, loss: 0.8340966701507568 2023-01-23 05:21:30.810391: step: 2218/1541, loss: 1.687229871749878 2023-01-23 05:21:31.678749: step: 2220/1541, loss: 1.514694333076477 2023-01-23 05:21:32.551178: step: 2222/1541, loss: 0.9175174236297607 2023-01-23 05:21:33.449966: step: 2224/1541, loss: 1.8969907760620117 2023-01-23 05:21:34.350396: step: 2226/1541, loss: 7.794639587402344 2023-01-23 05:21:35.215007: step: 2228/1541, loss: 1.1844249963760376 2023-01-23 05:21:36.069161: step: 2230/1541, loss: 0.8097560405731201 2023-01-23 05:21:36.988939: step: 2232/1541, loss: 0.9802626967430115 2023-01-23 05:21:37.904651: step: 2234/1541, loss: 1.6953321695327759 2023-01-23 05:21:38.799824: step: 2236/1541, loss: 1.6739609241485596 2023-01-23 05:21:39.652007: step: 2238/1541, loss: 2.456373929977417 2023-01-23 05:21:40.562153: step: 2240/1541, loss: 2.0780322551727295 2023-01-23 05:21:41.487088: step: 2242/1541, loss: 1.9386276006698608 2023-01-23 05:21:42.340574: step: 2244/1541, loss: 1.966787576675415 2023-01-23 05:21:43.249002: step: 2246/1541, loss: 1.743822693824768 2023-01-23 05:21:44.136171: step: 2248/1541, loss: 2.3492043018341064 2023-01-23 05:21:45.052314: step: 2250/1541, loss: 2.5515525341033936 2023-01-23 05:21:45.969223: step: 2252/1541, loss: 1.5591766834259033 2023-01-23 05:21:46.880973: step: 2254/1541, loss: 0.9824085831642151 2023-01-23 05:21:47.757922: step: 2256/1541, loss: 1.2093533277511597 2023-01-23 05:21:48.670232: step: 2258/1541, loss: 1.6143525838851929 2023-01-23 05:21:49.574947: step: 2260/1541, loss: 1.8345943689346313 2023-01-23 05:21:50.435741: step: 2262/1541, loss: 1.637512445449829 2023-01-23 05:21:51.345132: step: 2264/1541, loss: 2.055290699005127 2023-01-23 05:21:52.211495: step: 2266/1541, loss: 1.6151273250579834 2023-01-23 05:21:53.099678: step: 2268/1541, loss: 2.232996702194214 2023-01-23 05:21:54.010139: step: 2270/1541, loss: 1.0062097311019897 2023-01-23 05:21:54.905888: step: 2272/1541, loss: 1.5408074855804443 2023-01-23 05:21:55.783613: step: 2274/1541, loss: 1.1285287141799927 2023-01-23 05:21:56.713346: step: 2276/1541, loss: 4.820800304412842 2023-01-23 05:21:57.622131: step: 2278/1541, loss: 1.4802440404891968 2023-01-23 05:21:58.514636: step: 2280/1541, loss: 1.1879022121429443 2023-01-23 05:21:59.416734: step: 2282/1541, loss: 4.733030319213867 2023-01-23 05:22:00.253577: step: 2284/1541, loss: 2.3198421001434326 2023-01-23 05:22:01.143070: step: 2286/1541, loss: 2.5265469551086426 2023-01-23 05:22:02.030498: step: 2288/1541, loss: 1.296872615814209 2023-01-23 05:22:02.939683: step: 2290/1541, loss: 1.5709202289581299 2023-01-23 05:22:03.814189: step: 2292/1541, loss: 1.3067691326141357 2023-01-23 05:22:04.712185: step: 2294/1541, loss: 0.6139172315597534 2023-01-23 05:22:05.591798: step: 2296/1541, loss: 9.573974609375 2023-01-23 05:22:06.474742: step: 2298/1541, loss: 1.0723689794540405 2023-01-23 05:22:07.318018: step: 2300/1541, loss: 2.2330613136291504 2023-01-23 05:22:08.157340: step: 2302/1541, loss: 1.5683144330978394 2023-01-23 05:22:09.052044: step: 2304/1541, loss: 1.4856786727905273 2023-01-23 05:22:09.945763: step: 2306/1541, loss: 1.364242434501648 2023-01-23 05:22:10.820730: step: 2308/1541, loss: 0.7118258476257324 2023-01-23 05:22:11.692472: step: 2310/1541, loss: 1.9832863807678223 2023-01-23 05:22:12.586604: step: 2312/1541, loss: 2.244054079055786 2023-01-23 05:22:13.509913: step: 2314/1541, loss: 4.034794330596924 2023-01-23 05:22:14.446126: step: 2316/1541, loss: 0.9910688400268555 2023-01-23 05:22:15.309639: step: 2318/1541, loss: 0.9763568043708801 2023-01-23 05:22:16.207327: step: 2320/1541, loss: 1.0485481023788452 2023-01-23 05:22:17.101692: step: 2322/1541, loss: 1.4424996376037598 2023-01-23 05:22:17.998275: step: 2324/1541, loss: 1.9401463270187378 2023-01-23 05:22:18.830511: step: 2326/1541, loss: 0.9597500562667847 2023-01-23 05:22:19.688841: step: 2328/1541, loss: 2.346728563308716 2023-01-23 05:22:20.594617: step: 2330/1541, loss: 1.1869890689849854 2023-01-23 05:22:21.471303: step: 2332/1541, loss: 1.993728756904602 2023-01-23 05:22:22.383422: step: 2334/1541, loss: 2.365079164505005 2023-01-23 05:22:23.251654: step: 2336/1541, loss: 1.5295416116714478 2023-01-23 05:22:24.147885: step: 2338/1541, loss: 1.3254387378692627 2023-01-23 05:22:25.046873: step: 2340/1541, loss: 1.924581527709961 2023-01-23 05:22:25.916951: step: 2342/1541, loss: 0.8583758473396301 2023-01-23 05:22:26.848109: step: 2344/1541, loss: 1.2917628288269043 2023-01-23 05:22:27.721814: step: 2346/1541, loss: 2.6405439376831055 2023-01-23 05:22:28.629320: step: 2348/1541, loss: 1.6386146545410156 2023-01-23 05:22:29.499676: step: 2350/1541, loss: 7.608634948730469 2023-01-23 05:22:30.365149: step: 2352/1541, loss: 2.05985426902771 2023-01-23 05:22:31.291502: step: 2354/1541, loss: 3.594451665878296 2023-01-23 05:22:32.175278: step: 2356/1541, loss: 1.6200612783432007 2023-01-23 05:22:33.075342: step: 2358/1541, loss: 1.5062443017959595 2023-01-23 05:22:34.017313: step: 2360/1541, loss: 1.4029595851898193 2023-01-23 05:22:34.908965: step: 2362/1541, loss: 4.154743194580078 2023-01-23 05:22:35.821660: step: 2364/1541, loss: 1.2654221057891846 2023-01-23 05:22:36.712666: step: 2366/1541, loss: 1.8121272325515747 2023-01-23 05:22:37.600272: step: 2368/1541, loss: 1.3265725374221802 2023-01-23 05:22:38.528138: step: 2370/1541, loss: 1.0713659524917603 2023-01-23 05:22:39.454622: step: 2372/1541, loss: 3.6121253967285156 2023-01-23 05:22:40.330063: step: 2374/1541, loss: 1.1752221584320068 2023-01-23 05:22:41.230394: step: 2376/1541, loss: 3.3123281002044678 2023-01-23 05:22:42.112049: step: 2378/1541, loss: 1.1229305267333984 2023-01-23 05:22:42.956733: step: 2380/1541, loss: 1.1593762636184692 2023-01-23 05:22:43.818431: step: 2382/1541, loss: 1.1464898586273193 2023-01-23 05:22:44.673635: step: 2384/1541, loss: 1.4818578958511353 2023-01-23 05:22:45.560209: step: 2386/1541, loss: 1.469234585762024 2023-01-23 05:22:46.461947: step: 2388/1541, loss: 1.5345383882522583 2023-01-23 05:22:47.362505: step: 2390/1541, loss: 2.2739760875701904 2023-01-23 05:22:48.316519: step: 2392/1541, loss: 2.459646224975586 2023-01-23 05:22:49.212422: step: 2394/1541, loss: 0.5508168935775757 2023-01-23 05:22:50.110687: step: 2396/1541, loss: 0.7237075567245483 2023-01-23 05:22:51.017541: step: 2398/1541, loss: 2.2444710731506348 2023-01-23 05:22:51.903577: step: 2400/1541, loss: 2.1214120388031006 2023-01-23 05:22:52.764841: step: 2402/1541, loss: 1.7352217435836792 2023-01-23 05:22:53.723166: step: 2404/1541, loss: 1.5326379537582397 2023-01-23 05:22:54.618052: step: 2406/1541, loss: 1.0148358345031738 2023-01-23 05:22:55.516544: step: 2408/1541, loss: 1.7572298049926758 2023-01-23 05:22:56.440619: step: 2410/1541, loss: 1.5401924848556519 2023-01-23 05:22:57.299248: step: 2412/1541, loss: 9.041254997253418 2023-01-23 05:22:58.159616: step: 2414/1541, loss: 1.8502929210662842 2023-01-23 05:22:59.058224: step: 2416/1541, loss: 1.1940444707870483 2023-01-23 05:22:59.985909: step: 2418/1541, loss: 0.6922133564949036 2023-01-23 05:23:00.868932: step: 2420/1541, loss: 1.0775649547576904 2023-01-23 05:23:01.766377: step: 2422/1541, loss: 1.431461215019226 2023-01-23 05:23:02.776071: step: 2424/1541, loss: 2.253335952758789 2023-01-23 05:23:03.626140: step: 2426/1541, loss: 1.6017404794692993 2023-01-23 05:23:04.518370: step: 2428/1541, loss: 1.230046272277832 2023-01-23 05:23:05.408211: step: 2430/1541, loss: 1.2937284708023071 2023-01-23 05:23:06.283615: step: 2432/1541, loss: 1.32586669921875 2023-01-23 05:23:07.205631: step: 2434/1541, loss: 1.0053274631500244 2023-01-23 05:23:08.097675: step: 2436/1541, loss: 3.6503593921661377 2023-01-23 05:23:08.968092: step: 2438/1541, loss: 2.8455827236175537 2023-01-23 05:23:09.931840: step: 2440/1541, loss: 1.624405860900879 2023-01-23 05:23:10.791254: step: 2442/1541, loss: 1.5474679470062256 2023-01-23 05:23:11.676029: step: 2444/1541, loss: 1.4231942892074585 2023-01-23 05:23:12.569183: step: 2446/1541, loss: 7.993699550628662 2023-01-23 05:23:13.476735: step: 2448/1541, loss: 1.0958960056304932 2023-01-23 05:23:14.370670: step: 2450/1541, loss: 1.7797749042510986 2023-01-23 05:23:15.289357: step: 2452/1541, loss: 1.6096348762512207 2023-01-23 05:23:16.174022: step: 2454/1541, loss: 1.2487030029296875 2023-01-23 05:23:17.061527: step: 2456/1541, loss: 1.1893631219863892 2023-01-23 05:23:17.964271: step: 2458/1541, loss: 1.4915043115615845 2023-01-23 05:23:18.833506: step: 2460/1541, loss: 1.881641149520874 2023-01-23 05:23:19.722798: step: 2462/1541, loss: 1.2938014268875122 2023-01-23 05:23:20.609827: step: 2464/1541, loss: 0.758283257484436 2023-01-23 05:23:21.492222: step: 2466/1541, loss: 8.128525733947754 2023-01-23 05:23:22.371688: step: 2468/1541, loss: 1.8294073343276978 2023-01-23 05:23:23.246384: step: 2470/1541, loss: 0.7393746376037598 2023-01-23 05:23:24.101409: step: 2472/1541, loss: 3.745117664337158 2023-01-23 05:23:25.002066: step: 2474/1541, loss: 1.8777518272399902 2023-01-23 05:23:25.902471: step: 2476/1541, loss: 3.7648680210113525 2023-01-23 05:23:26.804029: step: 2478/1541, loss: 3.695951223373413 2023-01-23 05:23:27.716785: step: 2480/1541, loss: 1.5167803764343262 2023-01-23 05:23:28.631198: step: 2482/1541, loss: 1.20683753490448 2023-01-23 05:23:29.506084: step: 2484/1541, loss: 1.116026759147644 2023-01-23 05:23:30.364548: step: 2486/1541, loss: 1.1652849912643433 2023-01-23 05:23:31.254306: step: 2488/1541, loss: 2.256601572036743 2023-01-23 05:23:32.190458: step: 2490/1541, loss: 1.0373196601867676 2023-01-23 05:23:33.148948: step: 2492/1541, loss: 1.4822585582733154 2023-01-23 05:23:34.053347: step: 2494/1541, loss: 1.2324481010437012 2023-01-23 05:23:34.919613: step: 2496/1541, loss: 4.1517815589904785 2023-01-23 05:23:35.842833: step: 2498/1541, loss: 4.068368911743164 2023-01-23 05:23:36.735198: step: 2500/1541, loss: 0.8613705039024353 2023-01-23 05:23:37.669902: step: 2502/1541, loss: 1.438187599182129 2023-01-23 05:23:38.549759: step: 2504/1541, loss: 2.860494613647461 2023-01-23 05:23:39.425911: step: 2506/1541, loss: 4.615652084350586 2023-01-23 05:23:40.356801: step: 2508/1541, loss: 1.217963457107544 2023-01-23 05:23:41.228519: step: 2510/1541, loss: 2.638984441757202 2023-01-23 05:23:42.135842: step: 2512/1541, loss: 2.066831111907959 2023-01-23 05:23:43.002516: step: 2514/1541, loss: 0.9471815228462219 2023-01-23 05:23:43.909039: step: 2516/1541, loss: 2.41744065284729 2023-01-23 05:23:44.872617: step: 2518/1541, loss: 1.5019463300704956 2023-01-23 05:23:45.722938: step: 2520/1541, loss: 3.311530351638794 2023-01-23 05:23:46.641947: step: 2522/1541, loss: 2.5453946590423584 2023-01-23 05:23:47.505661: step: 2524/1541, loss: 0.759962797164917 2023-01-23 05:23:48.357222: step: 2526/1541, loss: 1.2217485904693604 2023-01-23 05:23:49.253069: step: 2528/1541, loss: 3.825263738632202 2023-01-23 05:23:50.126871: step: 2530/1541, loss: 1.2400273084640503 2023-01-23 05:23:51.003260: step: 2532/1541, loss: 2.1334831714630127 2023-01-23 05:23:51.894130: step: 2534/1541, loss: 0.9765654802322388 2023-01-23 05:23:52.835156: step: 2536/1541, loss: 2.335155487060547 2023-01-23 05:23:53.789451: step: 2538/1541, loss: 4.509556770324707 2023-01-23 05:23:54.729731: step: 2540/1541, loss: 2.3830654621124268 2023-01-23 05:23:55.705917: step: 2542/1541, loss: 1.9115793704986572 2023-01-23 05:23:56.587528: step: 2544/1541, loss: 1.3075093030929565 2023-01-23 05:23:57.588379: step: 2546/1541, loss: 2.197103261947632 2023-01-23 05:23:58.451032: step: 2548/1541, loss: 0.756209671497345 2023-01-23 05:23:59.387817: step: 2550/1541, loss: 1.244315266609192 2023-01-23 05:24:00.282042: step: 2552/1541, loss: 1.760947346687317 2023-01-23 05:24:01.190517: step: 2554/1541, loss: 2.7705864906311035 2023-01-23 05:24:02.176310: step: 2556/1541, loss: 6.9242329597473145 2023-01-23 05:24:03.061918: step: 2558/1541, loss: 2.224479913711548 2023-01-23 05:24:03.908734: step: 2560/1541, loss: 1.3424986600875854 2023-01-23 05:24:04.838008: step: 2562/1541, loss: 1.60308837890625 2023-01-23 05:24:05.750070: step: 2564/1541, loss: 1.6335557699203491 2023-01-23 05:24:06.663023: step: 2566/1541, loss: 0.9554516077041626 2023-01-23 05:24:07.594867: step: 2568/1541, loss: 1.74332857131958 2023-01-23 05:24:08.472339: step: 2570/1541, loss: 1.890310525894165 2023-01-23 05:24:09.426044: step: 2572/1541, loss: 2.0113353729248047 2023-01-23 05:24:10.390281: step: 2574/1541, loss: 2.6957316398620605 2023-01-23 05:24:11.301887: step: 2576/1541, loss: 1.228758454322815 2023-01-23 05:24:12.232539: step: 2578/1541, loss: 3.171997547149658 2023-01-23 05:24:13.120330: step: 2580/1541, loss: 1.9788039922714233 2023-01-23 05:24:14.020387: step: 2582/1541, loss: 2.0513975620269775 2023-01-23 05:24:14.893698: step: 2584/1541, loss: 7.868622303009033 2023-01-23 05:24:15.814319: step: 2586/1541, loss: 1.958996057510376 2023-01-23 05:24:16.698640: step: 2588/1541, loss: 0.8402938842773438 2023-01-23 05:24:17.568713: step: 2590/1541, loss: 1.3447397947311401 2023-01-23 05:24:18.473511: step: 2592/1541, loss: 1.5540891885757446 2023-01-23 05:24:19.370069: step: 2594/1541, loss: 0.8092209100723267 2023-01-23 05:24:20.262517: step: 2596/1541, loss: 1.3478142023086548 2023-01-23 05:24:21.189640: step: 2598/1541, loss: 1.3063747882843018 2023-01-23 05:24:22.035774: step: 2600/1541, loss: 1.1322283744812012 2023-01-23 05:24:22.948048: step: 2602/1541, loss: 1.2857098579406738 2023-01-23 05:24:23.907998: step: 2604/1541, loss: 1.8042312860488892 2023-01-23 05:24:24.811873: step: 2606/1541, loss: 1.3143510818481445 2023-01-23 05:24:25.699831: step: 2608/1541, loss: 0.8016505837440491 2023-01-23 05:24:26.603660: step: 2610/1541, loss: 1.767406940460205 2023-01-23 05:24:27.512139: step: 2612/1541, loss: 1.5216186046600342 2023-01-23 05:24:28.401814: step: 2614/1541, loss: 1.8250293731689453 2023-01-23 05:24:29.280357: step: 2616/1541, loss: 3.6240594387054443 2023-01-23 05:24:30.175666: step: 2618/1541, loss: 1.1899993419647217 2023-01-23 05:24:31.028022: step: 2620/1541, loss: 7.4469709396362305 2023-01-23 05:24:31.962033: step: 2622/1541, loss: 1.1883656978607178 2023-01-23 05:24:32.836750: step: 2624/1541, loss: 1.2525975704193115 2023-01-23 05:24:33.736633: step: 2626/1541, loss: 3.9045653343200684 2023-01-23 05:24:34.663529: step: 2628/1541, loss: 4.020846843719482 2023-01-23 05:24:35.572191: step: 2630/1541, loss: 1.6337851285934448 2023-01-23 05:24:36.483397: step: 2632/1541, loss: 4.7386627197265625 2023-01-23 05:24:37.358991: step: 2634/1541, loss: 1.6468091011047363 2023-01-23 05:24:38.236960: step: 2636/1541, loss: 3.262298107147217 2023-01-23 05:24:39.152221: step: 2638/1541, loss: 1.2887628078460693 2023-01-23 05:24:40.068695: step: 2640/1541, loss: 1.3166285753250122 2023-01-23 05:24:40.997184: step: 2642/1541, loss: 3.1762185096740723 2023-01-23 05:24:41.901872: step: 2644/1541, loss: 8.055333137512207 2023-01-23 05:24:42.827148: step: 2646/1541, loss: 0.8855505585670471 2023-01-23 05:24:43.757391: step: 2648/1541, loss: 4.214301586151123 2023-01-23 05:24:44.690240: step: 2650/1541, loss: 1.690877079963684 2023-01-23 05:24:45.553936: step: 2652/1541, loss: 1.8706430196762085 2023-01-23 05:24:46.513436: step: 2654/1541, loss: 0.754935622215271 2023-01-23 05:24:47.452056: step: 2656/1541, loss: 2.3139805793762207 2023-01-23 05:24:48.331420: step: 2658/1541, loss: 2.4892776012420654 2023-01-23 05:24:49.203951: step: 2660/1541, loss: 8.383882522583008 2023-01-23 05:24:50.098473: step: 2662/1541, loss: 1.9904476404190063 2023-01-23 05:24:50.972925: step: 2664/1541, loss: 1.0905537605285645 2023-01-23 05:24:51.820042: step: 2666/1541, loss: 1.9003878831863403 2023-01-23 05:24:52.689426: step: 2668/1541, loss: 4.305884838104248 2023-01-23 05:24:53.593667: step: 2670/1541, loss: 1.479646921157837 2023-01-23 05:24:54.500650: step: 2672/1541, loss: 0.8308634757995605 2023-01-23 05:24:55.384175: step: 2674/1541, loss: 4.503164291381836 2023-01-23 05:24:56.287771: step: 2676/1541, loss: 1.8275569677352905 2023-01-23 05:24:57.212176: step: 2678/1541, loss: 4.272184371948242 2023-01-23 05:24:58.122406: step: 2680/1541, loss: 2.118454933166504 2023-01-23 05:24:59.016710: step: 2682/1541, loss: 8.070246696472168 2023-01-23 05:24:59.879569: step: 2684/1541, loss: 1.9426676034927368 2023-01-23 05:25:00.739989: step: 2686/1541, loss: 1.9288140535354614 2023-01-23 05:25:01.636790: step: 2688/1541, loss: 1.1235263347625732 2023-01-23 05:25:02.529564: step: 2690/1541, loss: 1.0966674089431763 2023-01-23 05:25:03.461722: step: 2692/1541, loss: 2.2443227767944336 2023-01-23 05:25:04.384386: step: 2694/1541, loss: 0.9417904019355774 2023-01-23 05:25:05.271736: step: 2696/1541, loss: 2.16884708404541 2023-01-23 05:25:06.169350: step: 2698/1541, loss: 1.2371097803115845 2023-01-23 05:25:07.061226: step: 2700/1541, loss: 1.1461690664291382 2023-01-23 05:25:07.935747: step: 2702/1541, loss: 1.9201375246047974 2023-01-23 05:25:08.851030: step: 2704/1541, loss: 1.3605461120605469 2023-01-23 05:25:09.777848: step: 2706/1541, loss: 2.4821321964263916 2023-01-23 05:25:10.656403: step: 2708/1541, loss: 0.9109839797019958 2023-01-23 05:25:11.538719: step: 2710/1541, loss: 1.2288711071014404 2023-01-23 05:25:12.422846: step: 2712/1541, loss: 0.6462988257408142 2023-01-23 05:25:13.325474: step: 2714/1541, loss: 1.3379722833633423 2023-01-23 05:25:14.213156: step: 2716/1541, loss: 1.305911660194397 2023-01-23 05:25:15.096582: step: 2718/1541, loss: 1.9923208951950073 2023-01-23 05:25:16.024196: step: 2720/1541, loss: 1.5019173622131348 2023-01-23 05:25:16.921539: step: 2722/1541, loss: 1.3752142190933228 2023-01-23 05:25:17.783610: step: 2724/1541, loss: 1.1479017734527588 2023-01-23 05:25:18.655634: step: 2726/1541, loss: 1.864386796951294 2023-01-23 05:25:19.543718: step: 2728/1541, loss: 2.146570920944214 2023-01-23 05:25:20.465150: step: 2730/1541, loss: 1.3391013145446777 2023-01-23 05:25:21.387446: step: 2732/1541, loss: 9.476547241210938 2023-01-23 05:25:22.263286: step: 2734/1541, loss: 2.89176082611084 2023-01-23 05:25:23.154694: step: 2736/1541, loss: 3.593827724456787 2023-01-23 05:25:24.091953: step: 2738/1541, loss: 1.3417167663574219 2023-01-23 05:25:25.009496: step: 2740/1541, loss: 1.9887014627456665 2023-01-23 05:25:25.908169: step: 2742/1541, loss: 2.2106356620788574 2023-01-23 05:25:26.809924: step: 2744/1541, loss: 1.1578723192214966 2023-01-23 05:25:27.690657: step: 2746/1541, loss: 1.1174311637878418 2023-01-23 05:25:28.585544: step: 2748/1541, loss: 1.488038182258606 2023-01-23 05:25:29.460456: step: 2750/1541, loss: 1.5130860805511475 2023-01-23 05:25:30.346573: step: 2752/1541, loss: 1.17044198513031 2023-01-23 05:25:31.224351: step: 2754/1541, loss: 3.7526049613952637 2023-01-23 05:25:32.141104: step: 2756/1541, loss: 1.995558261871338 2023-01-23 05:25:33.073407: step: 2758/1541, loss: 1.1388853788375854 2023-01-23 05:25:33.992579: step: 2760/1541, loss: 1.6097601652145386 2023-01-23 05:25:34.919241: step: 2762/1541, loss: 2.7532711029052734 2023-01-23 05:25:35.803868: step: 2764/1541, loss: 0.7669805884361267 2023-01-23 05:25:36.666849: step: 2766/1541, loss: 4.043911933898926 2023-01-23 05:25:37.533688: step: 2768/1541, loss: 1.1402643918991089 2023-01-23 05:25:38.438172: step: 2770/1541, loss: 2.3839573860168457 2023-01-23 05:25:39.320297: step: 2772/1541, loss: 0.5764088034629822 2023-01-23 05:25:40.204826: step: 2774/1541, loss: 1.5327543020248413 2023-01-23 05:25:41.099809: step: 2776/1541, loss: 0.9932537078857422 2023-01-23 05:25:41.990586: step: 2778/1541, loss: 1.5342392921447754 2023-01-23 05:25:42.877909: step: 2780/1541, loss: 0.6883306503295898 2023-01-23 05:25:43.773892: step: 2782/1541, loss: 0.9877280592918396 2023-01-23 05:25:44.720238: step: 2784/1541, loss: 0.9690669178962708 2023-01-23 05:25:45.610690: step: 2786/1541, loss: 2.0216476917266846 2023-01-23 05:25:46.530882: step: 2788/1541, loss: 1.131047010421753 2023-01-23 05:25:47.448890: step: 2790/1541, loss: 3.187326192855835 2023-01-23 05:25:48.422322: step: 2792/1541, loss: 0.5892218351364136 2023-01-23 05:25:49.335576: step: 2794/1541, loss: 1.0124305486679077 2023-01-23 05:25:50.281208: step: 2796/1541, loss: 1.937996506690979 2023-01-23 05:25:51.265821: step: 2798/1541, loss: 2.0942492485046387 2023-01-23 05:25:52.172578: step: 2800/1541, loss: 0.8299248814582825 2023-01-23 05:25:53.067308: step: 2802/1541, loss: 1.404651165008545 2023-01-23 05:25:53.948928: step: 2804/1541, loss: 0.9778680801391602 2023-01-23 05:25:54.813477: step: 2806/1541, loss: 2.278322696685791 2023-01-23 05:25:55.703207: step: 2808/1541, loss: 1.2927545309066772 2023-01-23 05:25:56.609499: step: 2810/1541, loss: 8.806572914123535 2023-01-23 05:25:57.506274: step: 2812/1541, loss: 1.342309832572937 2023-01-23 05:25:58.363213: step: 2814/1541, loss: 0.5484181642532349 2023-01-23 05:25:59.289719: step: 2816/1541, loss: 2.182404041290283 2023-01-23 05:26:00.241231: step: 2818/1541, loss: 3.5628836154937744 2023-01-23 05:26:01.123157: step: 2820/1541, loss: 1.586952567100525 2023-01-23 05:26:01.995644: step: 2822/1541, loss: 1.163137435913086 2023-01-23 05:26:02.904587: step: 2824/1541, loss: 0.8995743989944458 2023-01-23 05:26:03.777421: step: 2826/1541, loss: 1.0046004056930542 2023-01-23 05:26:04.691249: step: 2828/1541, loss: 0.9475236535072327 2023-01-23 05:26:05.589967: step: 2830/1541, loss: 7.925792694091797 2023-01-23 05:26:06.506476: step: 2832/1541, loss: 3.6988048553466797 2023-01-23 05:26:07.373805: step: 2834/1541, loss: 3.604248046875 2023-01-23 05:26:08.266257: step: 2836/1541, loss: 1.860438585281372 2023-01-23 05:26:09.146202: step: 2838/1541, loss: 0.8575451374053955 2023-01-23 05:26:10.023396: step: 2840/1541, loss: 1.397740364074707 2023-01-23 05:26:10.911100: step: 2842/1541, loss: 1.3278436660766602 2023-01-23 05:26:11.837284: step: 2844/1541, loss: 1.970719814300537 2023-01-23 05:26:12.726926: step: 2846/1541, loss: 2.18874454498291 2023-01-23 05:26:13.619755: step: 2848/1541, loss: 1.8762884140014648 2023-01-23 05:26:14.540294: step: 2850/1541, loss: 1.6363840103149414 2023-01-23 05:26:15.429046: step: 2852/1541, loss: 1.9295823574066162 2023-01-23 05:26:16.289832: step: 2854/1541, loss: 0.8057435154914856 2023-01-23 05:26:17.191692: step: 2856/1541, loss: 1.3521955013275146 2023-01-23 05:26:18.046494: step: 2858/1541, loss: 1.7332265377044678 2023-01-23 05:26:18.873670: step: 2860/1541, loss: 1.2460838556289673 2023-01-23 05:26:19.714367: step: 2862/1541, loss: 0.6009072065353394 2023-01-23 05:26:20.582878: step: 2864/1541, loss: 0.944879412651062 2023-01-23 05:26:21.469887: step: 2866/1541, loss: 2.0712947845458984 2023-01-23 05:26:22.320106: step: 2868/1541, loss: 1.280534029006958 2023-01-23 05:26:23.243293: step: 2870/1541, loss: 1.3212311267852783 2023-01-23 05:26:24.145026: step: 2872/1541, loss: 3.660036325454712 2023-01-23 05:26:25.077978: step: 2874/1541, loss: 1.6792181730270386 2023-01-23 05:26:25.981496: step: 2876/1541, loss: 2.1273176670074463 2023-01-23 05:26:26.831255: step: 2878/1541, loss: 1.2508752346038818 2023-01-23 05:26:27.695031: step: 2880/1541, loss: 1.3221341371536255 2023-01-23 05:26:28.571341: step: 2882/1541, loss: 1.3174196481704712 2023-01-23 05:26:29.461334: step: 2884/1541, loss: 0.9677441120147705 2023-01-23 05:26:30.330006: step: 2886/1541, loss: 9.206088066101074 2023-01-23 05:26:31.205182: step: 2888/1541, loss: 0.8744150400161743 2023-01-23 05:26:32.135955: step: 2890/1541, loss: 1.0399476289749146 2023-01-23 05:26:33.014420: step: 2892/1541, loss: 4.603531837463379 2023-01-23 05:26:33.885277: step: 2894/1541, loss: 1.2203471660614014 2023-01-23 05:26:34.809661: step: 2896/1541, loss: 0.6451488733291626 2023-01-23 05:26:35.697166: step: 2898/1541, loss: 0.7109506726264954 2023-01-23 05:26:36.566509: step: 2900/1541, loss: 0.9148759841918945 2023-01-23 05:26:37.510157: step: 2902/1541, loss: 0.47687336802482605 2023-01-23 05:26:38.354732: step: 2904/1541, loss: 0.8504735231399536 2023-01-23 05:26:39.250590: step: 2906/1541, loss: 3.5485403537750244 2023-01-23 05:26:40.115371: step: 2908/1541, loss: 0.7315296530723572 2023-01-23 05:26:41.026359: step: 2910/1541, loss: 10.504484176635742 2023-01-23 05:26:41.957088: step: 2912/1541, loss: 3.46793270111084 2023-01-23 05:26:42.852824: step: 2914/1541, loss: 2.3945205211639404 2023-01-23 05:26:43.785387: step: 2916/1541, loss: 0.9774551391601562 2023-01-23 05:26:44.630352: step: 2918/1541, loss: 0.9577452540397644 2023-01-23 05:26:45.584048: step: 2920/1541, loss: 1.2078745365142822 2023-01-23 05:26:46.443807: step: 2922/1541, loss: 0.995376706123352 2023-01-23 05:26:47.338345: step: 2924/1541, loss: 1.1346396207809448 2023-01-23 05:26:48.246470: step: 2926/1541, loss: 1.2125309705734253 2023-01-23 05:26:49.143082: step: 2928/1541, loss: 0.9837161302566528 2023-01-23 05:26:50.117143: step: 2930/1541, loss: 1.980522632598877 2023-01-23 05:26:50.994488: step: 2932/1541, loss: 1.292456865310669 2023-01-23 05:26:51.932536: step: 2934/1541, loss: 1.146838903427124 2023-01-23 05:26:52.827387: step: 2936/1541, loss: 1.8499728441238403 2023-01-23 05:26:53.672203: step: 2938/1541, loss: 1.1705281734466553 2023-01-23 05:26:54.596039: step: 2940/1541, loss: 2.0617830753326416 2023-01-23 05:26:55.513562: step: 2942/1541, loss: 1.5195602178573608 2023-01-23 05:26:56.460349: step: 2944/1541, loss: 3.256021738052368 2023-01-23 05:26:57.347683: step: 2946/1541, loss: 1.7572238445281982 2023-01-23 05:26:58.208620: step: 2948/1541, loss: 1.5864205360412598 2023-01-23 05:26:59.088210: step: 2950/1541, loss: 1.615798830986023 2023-01-23 05:26:59.985098: step: 2952/1541, loss: 1.501753568649292 2023-01-23 05:27:00.857929: step: 2954/1541, loss: 1.1817737817764282 2023-01-23 05:27:01.764278: step: 2956/1541, loss: 2.127715826034546 2023-01-23 05:27:02.660897: step: 2958/1541, loss: 0.8060524463653564 2023-01-23 05:27:03.530982: step: 2960/1541, loss: 7.607016086578369 2023-01-23 05:27:04.416977: step: 2962/1541, loss: 1.7063175439834595 2023-01-23 05:27:05.298347: step: 2964/1541, loss: 2.598639726638794 2023-01-23 05:27:06.171873: step: 2966/1541, loss: 4.833226680755615 2023-01-23 05:27:07.044927: step: 2968/1541, loss: 5.756259918212891 2023-01-23 05:27:07.967852: step: 2970/1541, loss: 6.619884490966797 2023-01-23 05:27:08.824908: step: 2972/1541, loss: 7.63740348815918 2023-01-23 05:27:09.682330: step: 2974/1541, loss: 0.9742403030395508 2023-01-23 05:27:10.569034: step: 2976/1541, loss: 1.257499098777771 2023-01-23 05:27:11.489899: step: 2978/1541, loss: 2.161750078201294 2023-01-23 05:27:12.342340: step: 2980/1541, loss: 0.9605618715286255 2023-01-23 05:27:13.230288: step: 2982/1541, loss: 1.0655783414840698 2023-01-23 05:27:14.110876: step: 2984/1541, loss: 3.820685863494873 2023-01-23 05:27:14.969026: step: 2986/1541, loss: 1.4583468437194824 2023-01-23 05:27:15.835042: step: 2988/1541, loss: 1.5455797910690308 2023-01-23 05:27:16.742379: step: 2990/1541, loss: 1.1200236082077026 2023-01-23 05:27:17.714386: step: 2992/1541, loss: 1.776532769203186 2023-01-23 05:27:18.574971: step: 2994/1541, loss: 1.531255841255188 2023-01-23 05:27:19.505487: step: 2996/1541, loss: 2.639392375946045 2023-01-23 05:27:20.380698: step: 2998/1541, loss: 7.519433498382568 2023-01-23 05:27:21.339108: step: 3000/1541, loss: 1.8690398931503296 2023-01-23 05:27:22.232603: step: 3002/1541, loss: 0.7794058918952942 2023-01-23 05:27:23.124293: step: 3004/1541, loss: 1.3832523822784424 2023-01-23 05:27:24.047852: step: 3006/1541, loss: 8.40812873840332 2023-01-23 05:27:24.911117: step: 3008/1541, loss: 3.561389446258545 2023-01-23 05:27:25.812322: step: 3010/1541, loss: 1.840760588645935 2023-01-23 05:27:26.725127: step: 3012/1541, loss: 0.9485416412353516 2023-01-23 05:27:27.627016: step: 3014/1541, loss: 0.7415302395820618 2023-01-23 05:27:28.489881: step: 3016/1541, loss: 0.8764742016792297 2023-01-23 05:27:29.356011: step: 3018/1541, loss: 7.8784871101379395 2023-01-23 05:27:30.230607: step: 3020/1541, loss: 4.535917282104492 2023-01-23 05:27:31.177636: step: 3022/1541, loss: 1.5671112537384033 2023-01-23 05:27:32.070299: step: 3024/1541, loss: 1.195308804512024 2023-01-23 05:27:32.934857: step: 3026/1541, loss: 1.6653354167938232 2023-01-23 05:27:33.845343: step: 3028/1541, loss: 1.9767820835113525 2023-01-23 05:27:34.749084: step: 3030/1541, loss: 3.283111095428467 2023-01-23 05:27:35.634155: step: 3032/1541, loss: 0.8444740772247314 2023-01-23 05:27:36.536783: step: 3034/1541, loss: 0.9107363224029541 2023-01-23 05:27:37.407025: step: 3036/1541, loss: 1.5799704790115356 2023-01-23 05:27:38.333790: step: 3038/1541, loss: 1.94826340675354 2023-01-23 05:27:39.250760: step: 3040/1541, loss: 1.86981999874115 2023-01-23 05:27:40.119355: step: 3042/1541, loss: 0.585951030254364 2023-01-23 05:27:41.032155: step: 3044/1541, loss: 1.9084727764129639 2023-01-23 05:27:41.927532: step: 3046/1541, loss: 1.1765892505645752 2023-01-23 05:27:42.796173: step: 3048/1541, loss: 1.1659613847732544 2023-01-23 05:27:43.670869: step: 3050/1541, loss: 1.3762762546539307 2023-01-23 05:27:44.566222: step: 3052/1541, loss: 7.957155227661133 2023-01-23 05:27:45.442683: step: 3054/1541, loss: 4.736542224884033 2023-01-23 05:27:46.313751: step: 3056/1541, loss: 1.4350587129592896 2023-01-23 05:27:47.227531: step: 3058/1541, loss: 1.5286368131637573 2023-01-23 05:27:48.122434: step: 3060/1541, loss: 1.6953418254852295 2023-01-23 05:27:49.041061: step: 3062/1541, loss: 1.3497073650360107 2023-01-23 05:27:49.925595: step: 3064/1541, loss: 1.1381691694259644 2023-01-23 05:27:50.820012: step: 3066/1541, loss: 0.9064314961433411 2023-01-23 05:27:51.692084: step: 3068/1541, loss: 1.0035529136657715 2023-01-23 05:27:52.559921: step: 3070/1541, loss: 2.8471484184265137 2023-01-23 05:27:53.461839: step: 3072/1541, loss: 1.8429137468338013 2023-01-23 05:27:54.292522: step: 3074/1541, loss: 1.1606879234313965 2023-01-23 05:27:55.133200: step: 3076/1541, loss: 1.1053235530853271 2023-01-23 05:27:56.141489: step: 3078/1541, loss: 2.149735450744629 2023-01-23 05:27:56.998862: step: 3080/1541, loss: 1.9480302333831787 2023-01-23 05:27:57.836188: step: 3082/1541, loss: 1.3286948204040527 2023-01-23 05:27:58.559433: step: 3084/1541, loss: 2.06603741645813 ================================================== Loss: 4.368 -------------------- Dev Chinese: {'event': {'p': 0.6212664277180406, 'r': 0.6924101198402131, 'f1': 0.6549118387909321}, 'argument': {'p': 0.6323709190533013, 'r': 0.49513507557413056, 'f1': 0.5554010786431359}, 'combined': 0.36373874168064324, 'epoch': 0} Test Chinese: {'event': {'p': 0.6030494216614091, 'r': 0.6872378669862192, 'f1': 0.6423970876505182}, 'argument': {'p': 0.6443200173378743, 'r': 0.5357955277533802, 'f1': 0.5850677675886701}, 'combined': 0.3758458299771519, 'epoch': 0} Dev Korean: {'event': {'p': 0.5617647058823529, 'r': 0.762982689747004, 'f1': 0.6470920383963863}, 'argument': {'p': 0.6286803309975603, 'r': 0.49483226052711193, 'f1': 0.5537833962577786}, 'combined': 0.3583488267145197, 'epoch': 0} Test Korean: {'event': {'p': 0.5981887511916111, 'r': 0.7519472738166567, 'f1': 0.6663127156888771}, 'argument': {'p': 0.6443360636425584, 'r': 0.5473124864917484, 'f1': 0.5918744634712435}, 'combined': 0.3943734811024213, 'epoch': 0} Dev Russian: {'event': {'p': 0.6445497630331753, 'r': 0.7243675099866844, 'f1': 0.6821316614420063}, 'argument': {'p': 0.6206783521749645, 'r': 0.49187592721234025, 'f1': 0.5488212946245071}, 'combined': 0.3743683815369679, 'epoch': 0} Test Russian: {'event': {'p': 0.6431187859759289, 'r': 0.7363690832834032, 'f1': 0.6865921787709497}, 'argument': {'p': 0.6437771971949983, 'r': 0.5453389035822671, 'f1': 0.5904835543646312}, 'combined': 0.40542139011962663, 'epoch': 0} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6152375889217995, 'r': 0.4875467685795392, 'f1': 0.5439995523098016}, 'combined': 0.3998629187918199, 'epoch': 0} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.4080930487137384, 'r': 0.1378692732141008, 'f1': 0.20610760036047393}, 'combined': 0.1265857703001336, 'epoch': 0} Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.32857142857142857, 'r': 0.26744186046511625, 'f1': 0.2948717948717949}, 'combined': 0.16007326007326006, 'epoch': 0} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6212664277180406, 'r': 0.6924101198402131, 'f1': 0.6549118387909321}, 'argument': {'p': 0.6323709190533013, 'r': 0.49513507557413056, 'f1': 0.5554010786431359}, 'combined': 0.36373874168064324, 'epoch': 0} Test for Chinese: {'event': {'p': 0.6030494216614091, 'r': 0.6872378669862192, 'f1': 0.6423970876505182}, 'argument': {'p': 0.6443200173378743, 'r': 0.5357955277533802, 'f1': 0.5850677675886701}, 'combined': 0.3758458299771519, 'epoch': 0} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6152375889217995, 'r': 0.4875467685795392, 'f1': 0.5439995523098016}, 'combined': 0.3998629187918199, 'epoch': 0} -------------------- Dev for Korean: {'event': {'p': 0.5617647058823529, 'r': 0.762982689747004, 'f1': 0.6470920383963863}, 'argument': {'p': 0.6286803309975603, 'r': 0.49483226052711193, 'f1': 0.5537833962577786}, 'combined': 0.3583488267145197, 'epoch': 0} Test for Korean: {'event': {'p': 0.5981887511916111, 'r': 0.7519472738166567, 'f1': 0.6663127156888771}, 'argument': {'p': 0.6443360636425584, 'r': 0.5473124864917484, 'f1': 0.5918744634712435}, 'combined': 0.3943734811024213, 'epoch': 0} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.4080930487137384, 'r': 0.1378692732141008, 'f1': 0.20610760036047393}, 'combined': 0.1265857703001336, 'epoch': 0} -------------------- Dev for Russian: {'event': {'p': 0.6445497630331753, 'r': 0.7243675099866844, 'f1': 0.6821316614420063}, 'argument': {'p': 0.6206783521749645, 'r': 0.49187592721234025, 'f1': 0.5488212946245071}, 'combined': 0.3743683815369679, 'epoch': 0} Test for Russian: {'event': {'p': 0.6431187859759289, 'r': 0.7363690832834032, 'f1': 0.6865921787709497}, 'argument': {'p': 0.6437771971949983, 'r': 0.5453389035822671, 'f1': 0.5904835543646312}, 'combined': 0.40542139011962663, 'epoch': 0} Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.32857142857142857, 'r': 0.26744186046511625, 'f1': 0.2948717948717949}, 'combined': 0.16007326007326006, 'epoch': 0} ****************************** Epoch: 1 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 05:33:27.143708: step: 2/1541, loss: 4.1971025466918945 2023-01-23 05:33:28.094915: step: 4/1541, loss: 0.9349716305732727 2023-01-23 05:33:29.006678: step: 6/1541, loss: 2.345768451690674 2023-01-23 05:33:29.886511: step: 8/1541, loss: 0.6827350854873657 2023-01-23 05:33:30.756207: step: 10/1541, loss: 1.9772930145263672 2023-01-23 05:33:31.698622: step: 12/1541, loss: 1.0974581241607666 2023-01-23 05:33:32.637225: step: 14/1541, loss: 1.3544845581054688 2023-01-23 05:33:33.535070: step: 16/1541, loss: 1.0869818925857544 2023-01-23 05:33:34.385462: step: 18/1541, loss: 0.5491856336593628 2023-01-23 05:33:35.280531: step: 20/1541, loss: 0.9053882360458374 2023-01-23 05:33:36.131200: step: 22/1541, loss: 1.7346388101577759 2023-01-23 05:33:36.999080: step: 24/1541, loss: 0.8602963089942932 2023-01-23 05:33:37.892463: step: 26/1541, loss: 0.7133936882019043 2023-01-23 05:33:38.787403: step: 28/1541, loss: 1.0057522058486938 2023-01-23 05:33:39.770793: step: 30/1541, loss: 1.9250956773757935 2023-01-23 05:33:40.692998: step: 32/1541, loss: 1.5635870695114136 2023-01-23 05:33:41.609744: step: 34/1541, loss: 1.5110901594161987 2023-01-23 05:33:42.508917: step: 36/1541, loss: 2.374096632003784 2023-01-23 05:33:43.398199: step: 38/1541, loss: 0.7221804857254028 2023-01-23 05:33:44.300195: step: 40/1541, loss: 2.38568377494812 2023-01-23 05:33:45.226037: step: 42/1541, loss: 1.5934250354766846 2023-01-23 05:33:46.161384: step: 44/1541, loss: 2.3461995124816895 2023-01-23 05:33:47.080782: step: 46/1541, loss: 1.694205403327942 2023-01-23 05:33:47.953803: step: 48/1541, loss: 1.0758965015411377 2023-01-23 05:33:48.842084: step: 50/1541, loss: 0.8423731923103333 2023-01-23 05:33:49.687258: step: 52/1541, loss: 1.0651322603225708 2023-01-23 05:33:50.576146: step: 54/1541, loss: 1.2727490663528442 2023-01-23 05:33:51.446540: step: 56/1541, loss: 0.5279873609542847 2023-01-23 05:33:52.333257: step: 58/1541, loss: 3.9443159103393555 2023-01-23 05:33:53.220393: step: 60/1541, loss: 1.1150546073913574 2023-01-23 05:33:54.125967: step: 62/1541, loss: 1.9135013818740845 2023-01-23 05:33:54.997779: step: 64/1541, loss: 0.9610504508018494 2023-01-23 05:33:55.926333: step: 66/1541, loss: 1.6484239101409912 2023-01-23 05:33:56.795450: step: 68/1541, loss: 1.377852201461792 2023-01-23 05:33:57.668006: step: 70/1541, loss: 0.7091913819313049 2023-01-23 05:33:58.579224: step: 72/1541, loss: 4.281059741973877 2023-01-23 05:33:59.480015: step: 74/1541, loss: 1.6027878522872925 2023-01-23 05:34:00.370356: step: 76/1541, loss: 2.650484323501587 2023-01-23 05:34:01.264857: step: 78/1541, loss: 0.9443067312240601 2023-01-23 05:34:02.201486: step: 80/1541, loss: 2.0563485622406006 2023-01-23 05:34:03.092018: step: 82/1541, loss: 3.3584210872650146 2023-01-23 05:34:04.090678: step: 84/1541, loss: 1.356446623802185 2023-01-23 05:34:04.971614: step: 86/1541, loss: 1.8799824714660645 2023-01-23 05:34:05.858394: step: 88/1541, loss: 0.8507281541824341 2023-01-23 05:34:06.791896: step: 90/1541, loss: 1.6840351819992065 2023-01-23 05:34:07.675400: step: 92/1541, loss: 1.0465998649597168 2023-01-23 05:34:08.553419: step: 94/1541, loss: 0.8924545645713806 2023-01-23 05:34:09.486695: step: 96/1541, loss: 1.9401443004608154 2023-01-23 05:34:10.372893: step: 98/1541, loss: 0.996609628200531 2023-01-23 05:34:11.280365: step: 100/1541, loss: 1.3358983993530273 2023-01-23 05:34:12.155818: step: 102/1541, loss: 1.1484875679016113 2023-01-23 05:34:13.019232: step: 104/1541, loss: 3.757835865020752 2023-01-23 05:34:13.911513: step: 106/1541, loss: 2.111769199371338 2023-01-23 05:34:14.819100: step: 108/1541, loss: 1.2980635166168213 2023-01-23 05:34:15.756280: step: 110/1541, loss: 3.2786338329315186 2023-01-23 05:34:16.721820: step: 112/1541, loss: 0.9595214128494263 2023-01-23 05:34:17.639325: step: 114/1541, loss: 4.721959590911865 2023-01-23 05:34:18.575815: step: 116/1541, loss: 0.8978781700134277 2023-01-23 05:34:19.449257: step: 118/1541, loss: 1.837475061416626 2023-01-23 05:34:20.328665: step: 120/1541, loss: 1.5556107759475708 2023-01-23 05:34:21.209465: step: 122/1541, loss: 12.229978561401367 2023-01-23 05:34:22.110929: step: 124/1541, loss: 2.201319694519043 2023-01-23 05:34:23.007607: step: 126/1541, loss: 1.8417608737945557 2023-01-23 05:34:23.907009: step: 128/1541, loss: 1.9783761501312256 2023-01-23 05:34:24.755843: step: 130/1541, loss: 0.7910740971565247 2023-01-23 05:34:25.625572: step: 132/1541, loss: 0.7825881838798523 2023-01-23 05:34:26.554757: step: 134/1541, loss: 3.809465169906616 2023-01-23 05:34:27.410885: step: 136/1541, loss: 1.5446447134017944 2023-01-23 05:34:28.303715: step: 138/1541, loss: 6.890410900115967 2023-01-23 05:34:29.189589: step: 140/1541, loss: 1.6173641681671143 2023-01-23 05:34:30.069031: step: 142/1541, loss: 2.129877805709839 2023-01-23 05:34:30.989840: step: 144/1541, loss: 1.1788281202316284 2023-01-23 05:34:31.891891: step: 146/1541, loss: 3.7281715869903564 2023-01-23 05:34:32.829287: step: 148/1541, loss: 1.104671835899353 2023-01-23 05:34:33.751106: step: 150/1541, loss: 3.6942713260650635 2023-01-23 05:34:34.666926: step: 152/1541, loss: 3.7993955612182617 2023-01-23 05:34:35.535362: step: 154/1541, loss: 1.7970781326293945 2023-01-23 05:34:36.447096: step: 156/1541, loss: 1.4070806503295898 2023-01-23 05:34:37.306692: step: 158/1541, loss: 0.5030681490898132 2023-01-23 05:34:38.218584: step: 160/1541, loss: 1.8112101554870605 2023-01-23 05:34:39.083382: step: 162/1541, loss: 1.6272774934768677 2023-01-23 05:34:39.954234: step: 164/1541, loss: 1.5255776643753052 2023-01-23 05:34:40.900055: step: 166/1541, loss: 1.5772103071212769 2023-01-23 05:34:41.765776: step: 168/1541, loss: 1.4526962041854858 2023-01-23 05:34:42.653809: step: 170/1541, loss: 7.261528015136719 2023-01-23 05:34:43.535349: step: 172/1541, loss: 0.8781511783599854 2023-01-23 05:34:44.437759: step: 174/1541, loss: 1.4271398782730103 2023-01-23 05:34:45.363533: step: 176/1541, loss: 1.161170244216919 2023-01-23 05:34:46.276446: step: 178/1541, loss: 3.248987913131714 2023-01-23 05:34:47.186876: step: 180/1541, loss: 1.4803651571273804 2023-01-23 05:34:48.088639: step: 182/1541, loss: 1.9713718891143799 2023-01-23 05:34:48.977958: step: 184/1541, loss: 2.0489439964294434 2023-01-23 05:34:49.959621: step: 186/1541, loss: 1.5370756387710571 2023-01-23 05:34:50.820968: step: 188/1541, loss: 1.3874900341033936 2023-01-23 05:34:51.708484: step: 190/1541, loss: 0.5498137474060059 2023-01-23 05:34:52.590190: step: 192/1541, loss: 7.192009925842285 2023-01-23 05:34:53.536520: step: 194/1541, loss: 2.5171189308166504 2023-01-23 05:34:54.496941: step: 196/1541, loss: 1.3210700750350952 2023-01-23 05:34:55.398735: step: 198/1541, loss: 0.9558416604995728 2023-01-23 05:34:56.250827: step: 200/1541, loss: 0.854266881942749 2023-01-23 05:34:57.181784: step: 202/1541, loss: 1.9775217771530151 2023-01-23 05:34:58.132672: step: 204/1541, loss: 0.7576758861541748 2023-01-23 05:34:59.021453: step: 206/1541, loss: 4.4165215492248535 2023-01-23 05:34:59.926358: step: 208/1541, loss: 1.4501429796218872 2023-01-23 05:35:00.855312: step: 210/1541, loss: 1.7347021102905273 2023-01-23 05:35:01.736712: step: 212/1541, loss: 1.6366264820098877 2023-01-23 05:35:02.677406: step: 214/1541, loss: 1.0478545427322388 2023-01-23 05:35:03.550807: step: 216/1541, loss: 0.8181278109550476 2023-01-23 05:35:04.420630: step: 218/1541, loss: 0.8307151794433594 2023-01-23 05:35:05.344321: step: 220/1541, loss: 0.7109382152557373 2023-01-23 05:35:06.245903: step: 222/1541, loss: 4.034669399261475 2023-01-23 05:35:07.146977: step: 224/1541, loss: 0.4254451096057892 2023-01-23 05:35:08.005225: step: 226/1541, loss: 3.5372514724731445 2023-01-23 05:35:08.877802: step: 228/1541, loss: 1.4351738691329956 2023-01-23 05:35:09.799673: step: 230/1541, loss: 2.5277175903320312 2023-01-23 05:35:10.684616: step: 232/1541, loss: 1.5248980522155762 2023-01-23 05:35:11.626476: step: 234/1541, loss: 1.0597617626190186 2023-01-23 05:35:12.539239: step: 236/1541, loss: 0.9959869980812073 2023-01-23 05:35:13.432560: step: 238/1541, loss: 0.7124783992767334 2023-01-23 05:35:14.334709: step: 240/1541, loss: 0.9499538540840149 2023-01-23 05:35:15.309098: step: 242/1541, loss: 1.7836637496948242 2023-01-23 05:35:16.244081: step: 244/1541, loss: 0.7485816478729248 2023-01-23 05:35:17.147408: step: 246/1541, loss: 3.772355318069458 2023-01-23 05:35:18.014823: step: 248/1541, loss: 1.7367113828659058 2023-01-23 05:35:18.907873: step: 250/1541, loss: 0.7633818984031677 2023-01-23 05:35:19.780026: step: 252/1541, loss: 0.8297060132026672 2023-01-23 05:35:20.674505: step: 254/1541, loss: 10.962618827819824 2023-01-23 05:35:21.569625: step: 256/1541, loss: 0.9236879944801331 2023-01-23 05:35:22.451748: step: 258/1541, loss: 2.0822646617889404 2023-01-23 05:35:23.319087: step: 260/1541, loss: 0.9829673171043396 2023-01-23 05:35:24.233557: step: 262/1541, loss: 0.9189820885658264 2023-01-23 05:35:25.123409: step: 264/1541, loss: 1.2471133470535278 2023-01-23 05:35:26.069907: step: 266/1541, loss: 1.1279224157333374 2023-01-23 05:35:27.035929: step: 268/1541, loss: 1.517466425895691 2023-01-23 05:35:27.944314: step: 270/1541, loss: 0.6695058941841125 2023-01-23 05:35:28.923374: step: 272/1541, loss: 3.475691080093384 2023-01-23 05:35:29.809168: step: 274/1541, loss: 1.5141534805297852 2023-01-23 05:35:30.703366: step: 276/1541, loss: 1.7267464399337769 2023-01-23 05:35:31.586120: step: 278/1541, loss: 0.6793108582496643 2023-01-23 05:35:32.548035: step: 280/1541, loss: 3.4711036682128906 2023-01-23 05:35:33.475988: step: 282/1541, loss: 2.0617363452911377 2023-01-23 05:35:34.369773: step: 284/1541, loss: 4.001616954803467 2023-01-23 05:35:35.276658: step: 286/1541, loss: 1.8309670686721802 2023-01-23 05:35:36.148872: step: 288/1541, loss: 1.686366081237793 2023-01-23 05:35:37.004346: step: 290/1541, loss: 3.2278072834014893 2023-01-23 05:35:37.874792: step: 292/1541, loss: 3.477879762649536 2023-01-23 05:35:38.764181: step: 294/1541, loss: 0.9919607043266296 2023-01-23 05:35:39.690436: step: 296/1541, loss: 2.7843470573425293 2023-01-23 05:35:40.603069: step: 298/1541, loss: 1.2743535041809082 2023-01-23 05:35:41.524149: step: 300/1541, loss: 1.2321581840515137 2023-01-23 05:35:42.444170: step: 302/1541, loss: 1.5077533721923828 2023-01-23 05:35:43.300344: step: 304/1541, loss: 0.5443652868270874 2023-01-23 05:35:44.205455: step: 306/1541, loss: 0.5976079702377319 2023-01-23 05:35:45.132250: step: 308/1541, loss: 1.5918645858764648 2023-01-23 05:35:45.992662: step: 310/1541, loss: 1.2950342893600464 2023-01-23 05:35:46.892276: step: 312/1541, loss: 0.9637414813041687 2023-01-23 05:35:47.829302: step: 314/1541, loss: 1.7584478855133057 2023-01-23 05:35:48.741248: step: 316/1541, loss: 1.0759673118591309 2023-01-23 05:35:49.639718: step: 318/1541, loss: 0.626264214515686 2023-01-23 05:35:50.541358: step: 320/1541, loss: 1.0279558897018433 2023-01-23 05:35:51.468097: step: 322/1541, loss: 5.4127912521362305 2023-01-23 05:35:52.358884: step: 324/1541, loss: 1.199666142463684 2023-01-23 05:35:53.274744: step: 326/1541, loss: 1.455627202987671 2023-01-23 05:35:54.183199: step: 328/1541, loss: 1.945790410041809 2023-01-23 05:35:55.065136: step: 330/1541, loss: 1.827146291732788 2023-01-23 05:35:56.013083: step: 332/1541, loss: 2.6432833671569824 2023-01-23 05:35:56.929078: step: 334/1541, loss: 3.6004559993743896 2023-01-23 05:35:57.816594: step: 336/1541, loss: 1.2711782455444336 2023-01-23 05:35:58.655950: step: 338/1541, loss: 11.878430366516113 2023-01-23 05:35:59.668171: step: 340/1541, loss: 3.2601916790008545 2023-01-23 05:36:00.567453: step: 342/1541, loss: 1.1169662475585938 2023-01-23 05:36:01.463200: step: 344/1541, loss: 0.5049434900283813 2023-01-23 05:36:02.352763: step: 346/1541, loss: 1.474503993988037 2023-01-23 05:36:03.287322: step: 348/1541, loss: 1.5517735481262207 2023-01-23 05:36:04.143110: step: 350/1541, loss: 7.230674743652344 2023-01-23 05:36:05.062868: step: 352/1541, loss: 2.629558801651001 2023-01-23 05:36:05.918087: step: 354/1541, loss: 7.249820709228516 2023-01-23 05:36:06.834637: step: 356/1541, loss: 3.24574613571167 2023-01-23 05:36:07.727595: step: 358/1541, loss: 1.0299227237701416 2023-01-23 05:36:08.681064: step: 360/1541, loss: 0.9389128684997559 2023-01-23 05:36:09.624749: step: 362/1541, loss: 2.415729522705078 2023-01-23 05:36:10.504152: step: 364/1541, loss: 2.0728368759155273 2023-01-23 05:36:11.393291: step: 366/1541, loss: 3.400285005569458 2023-01-23 05:36:12.345597: step: 368/1541, loss: 1.0175992250442505 2023-01-23 05:36:13.226166: step: 370/1541, loss: 0.734133243560791 2023-01-23 05:36:14.120515: step: 372/1541, loss: 1.6643877029418945 2023-01-23 05:36:15.080119: step: 374/1541, loss: 1.459455966949463 2023-01-23 05:36:15.977021: step: 376/1541, loss: 3.912799596786499 2023-01-23 05:36:16.892806: step: 378/1541, loss: 1.2912272214889526 2023-01-23 05:36:17.790200: step: 380/1541, loss: 1.9407049417495728 2023-01-23 05:36:18.646942: step: 382/1541, loss: 0.6321556568145752 2023-01-23 05:36:19.535949: step: 384/1541, loss: 7.333801746368408 2023-01-23 05:36:20.419594: step: 386/1541, loss: 0.8363317251205444 2023-01-23 05:36:21.329385: step: 388/1541, loss: 1.7586272954940796 2023-01-23 05:36:22.192036: step: 390/1541, loss: 3.3862974643707275 2023-01-23 05:36:23.050880: step: 392/1541, loss: 1.2152516841888428 2023-01-23 05:36:23.919980: step: 394/1541, loss: 1.493565320968628 2023-01-23 05:36:24.861707: step: 396/1541, loss: 0.9373268485069275 2023-01-23 05:36:25.791255: step: 398/1541, loss: 1.1044968366622925 2023-01-23 05:36:26.674678: step: 400/1541, loss: 2.308537006378174 2023-01-23 05:36:27.544834: step: 402/1541, loss: 1.1341270208358765 2023-01-23 05:36:28.432597: step: 404/1541, loss: 0.44310462474823 2023-01-23 05:36:29.336095: step: 406/1541, loss: 5.898744106292725 2023-01-23 05:36:30.225164: step: 408/1541, loss: 0.5240208506584167 2023-01-23 05:36:31.134585: step: 410/1541, loss: 1.0712952613830566 2023-01-23 05:36:32.050052: step: 412/1541, loss: 1.1985784769058228 2023-01-23 05:36:32.986767: step: 414/1541, loss: 3.628972291946411 2023-01-23 05:36:33.896332: step: 416/1541, loss: 2.2719435691833496 2023-01-23 05:36:34.763764: step: 418/1541, loss: 1.1169573068618774 2023-01-23 05:36:35.653293: step: 420/1541, loss: 1.2777281999588013 2023-01-23 05:36:36.560712: step: 422/1541, loss: 1.5896871089935303 2023-01-23 05:36:37.451940: step: 424/1541, loss: 4.799841403961182 2023-01-23 05:36:38.347630: step: 426/1541, loss: 3.7213587760925293 2023-01-23 05:36:39.273414: step: 428/1541, loss: 1.3226982355117798 2023-01-23 05:36:40.207124: step: 430/1541, loss: 0.8105829954147339 2023-01-23 05:36:41.121325: step: 432/1541, loss: 2.0109047889709473 2023-01-23 05:36:42.004702: step: 434/1541, loss: 1.9641839265823364 2023-01-23 05:36:42.859346: step: 436/1541, loss: 1.751057505607605 2023-01-23 05:36:43.734317: step: 438/1541, loss: 1.3246973752975464 2023-01-23 05:36:44.595647: step: 440/1541, loss: 1.0939093828201294 2023-01-23 05:36:45.479504: step: 442/1541, loss: 0.905599057674408 2023-01-23 05:36:46.389274: step: 444/1541, loss: 1.3458037376403809 2023-01-23 05:36:47.237222: step: 446/1541, loss: 1.0083476305007935 2023-01-23 05:36:48.155031: step: 448/1541, loss: 1.21114981174469 2023-01-23 05:36:49.038025: step: 450/1541, loss: 0.849929928779602 2023-01-23 05:36:50.006248: step: 452/1541, loss: 0.6494117975234985 2023-01-23 05:36:50.851689: step: 454/1541, loss: 0.9672610759735107 2023-01-23 05:36:51.790956: step: 456/1541, loss: 1.500563144683838 2023-01-23 05:36:52.652266: step: 458/1541, loss: 1.9154329299926758 2023-01-23 05:36:53.549928: step: 460/1541, loss: 1.4190824031829834 2023-01-23 05:36:54.421384: step: 462/1541, loss: 0.8715049028396606 2023-01-23 05:36:55.306208: step: 464/1541, loss: 1.6953321695327759 2023-01-23 05:36:56.176036: step: 466/1541, loss: 0.7572718858718872 2023-01-23 05:36:57.075212: step: 468/1541, loss: 1.3892462253570557 2023-01-23 05:36:58.038594: step: 470/1541, loss: 1.120063304901123 2023-01-23 05:36:58.916779: step: 472/1541, loss: 6.897828578948975 2023-01-23 05:36:59.867024: step: 474/1541, loss: 1.394565224647522 2023-01-23 05:37:00.824541: step: 476/1541, loss: 0.7307003736495972 2023-01-23 05:37:01.748405: step: 478/1541, loss: 2.938091278076172 2023-01-23 05:37:02.649702: step: 480/1541, loss: 0.4497970938682556 2023-01-23 05:37:03.572446: step: 482/1541, loss: 3.9382855892181396 2023-01-23 05:37:04.439693: step: 484/1541, loss: 1.5181831121444702 2023-01-23 05:37:05.326210: step: 486/1541, loss: 0.9562671780586243 2023-01-23 05:37:06.224616: step: 488/1541, loss: 1.068907618522644 2023-01-23 05:37:07.103391: step: 490/1541, loss: 1.866097331047058 2023-01-23 05:37:08.042836: step: 492/1541, loss: 3.7070975303649902 2023-01-23 05:37:08.998965: step: 494/1541, loss: 3.0042266845703125 2023-01-23 05:37:09.860536: step: 496/1541, loss: 6.507486820220947 2023-01-23 05:37:10.763526: step: 498/1541, loss: 0.782077968120575 2023-01-23 05:37:11.681354: step: 500/1541, loss: 0.9860808849334717 2023-01-23 05:37:12.562002: step: 502/1541, loss: 2.2432453632354736 2023-01-23 05:37:13.498083: step: 504/1541, loss: 4.807198524475098 2023-01-23 05:37:14.346910: step: 506/1541, loss: 1.3125077486038208 2023-01-23 05:37:15.208999: step: 508/1541, loss: 7.235714912414551 2023-01-23 05:37:16.091433: step: 510/1541, loss: 1.1958669424057007 2023-01-23 05:37:17.039806: step: 512/1541, loss: 0.9133376479148865 2023-01-23 05:37:18.024833: step: 514/1541, loss: 1.7919678688049316 2023-01-23 05:37:18.952780: step: 516/1541, loss: 1.4577739238739014 2023-01-23 05:37:19.955738: step: 518/1541, loss: 3.6716578006744385 2023-01-23 05:37:20.844542: step: 520/1541, loss: 1.6406279802322388 2023-01-23 05:37:21.758902: step: 522/1541, loss: 0.8637295961380005 2023-01-23 05:37:22.663723: step: 524/1541, loss: 1.6228654384613037 2023-01-23 05:37:23.546386: step: 526/1541, loss: 0.6536268591880798 2023-01-23 05:37:24.455989: step: 528/1541, loss: 1.2812243700027466 2023-01-23 05:37:25.434778: step: 530/1541, loss: 8.11650562286377 2023-01-23 05:37:26.342240: step: 532/1541, loss: 0.917040228843689 2023-01-23 05:37:27.230906: step: 534/1541, loss: 0.6483702659606934 2023-01-23 05:37:28.087477: step: 536/1541, loss: 1.0460906028747559 2023-01-23 05:37:29.006312: step: 538/1541, loss: 2.04006028175354 2023-01-23 05:37:29.946441: step: 540/1541, loss: 1.992754578590393 2023-01-23 05:37:30.893281: step: 542/1541, loss: 3.540119171142578 2023-01-23 05:37:31.768829: step: 544/1541, loss: 3.4494149684906006 2023-01-23 05:37:32.684149: step: 546/1541, loss: 1.2552653551101685 2023-01-23 05:37:33.672198: step: 548/1541, loss: 1.2237857580184937 2023-01-23 05:37:34.554113: step: 550/1541, loss: 1.6333658695220947 2023-01-23 05:37:35.474403: step: 552/1541, loss: 2.0777862071990967 2023-01-23 05:37:36.409050: step: 554/1541, loss: 2.655524730682373 2023-01-23 05:37:37.394278: step: 556/1541, loss: 2.031102180480957 2023-01-23 05:37:38.269414: step: 558/1541, loss: 1.4719327688217163 2023-01-23 05:37:39.123276: step: 560/1541, loss: 0.6857996582984924 2023-01-23 05:37:40.076119: step: 562/1541, loss: 0.9241499900817871 2023-01-23 05:37:41.008702: step: 564/1541, loss: 3.2500498294830322 2023-01-23 05:37:41.883784: step: 566/1541, loss: 1.2777106761932373 2023-01-23 05:37:42.725125: step: 568/1541, loss: 1.1165506839752197 2023-01-23 05:37:43.689956: step: 570/1541, loss: 3.784125566482544 2023-01-23 05:37:44.557056: step: 572/1541, loss: 2.912179470062256 2023-01-23 05:37:45.477777: step: 574/1541, loss: 1.6378874778747559 2023-01-23 05:37:46.355496: step: 576/1541, loss: 0.8585452437400818 2023-01-23 05:37:47.257340: step: 578/1541, loss: 1.1992607116699219 2023-01-23 05:37:48.127190: step: 580/1541, loss: 3.447479248046875 2023-01-23 05:37:48.985842: step: 582/1541, loss: 0.5673176646232605 2023-01-23 05:37:49.928846: step: 584/1541, loss: 1.7320998907089233 2023-01-23 05:37:50.815457: step: 586/1541, loss: 3.86997652053833 2023-01-23 05:37:51.774670: step: 588/1541, loss: 1.1768850088119507 2023-01-23 05:37:52.658954: step: 590/1541, loss: 0.8497107625007629 2023-01-23 05:37:53.539931: step: 592/1541, loss: 1.0358024835586548 2023-01-23 05:37:54.393973: step: 594/1541, loss: 0.9843288064002991 2023-01-23 05:37:55.280246: step: 596/1541, loss: 1.3087875843048096 2023-01-23 05:37:56.201987: step: 598/1541, loss: 1.0556484460830688 2023-01-23 05:37:57.077884: step: 600/1541, loss: 0.8240888118743896 2023-01-23 05:37:57.966892: step: 602/1541, loss: 2.0508720874786377 2023-01-23 05:37:58.866300: step: 604/1541, loss: 1.164616346359253 2023-01-23 05:37:59.723425: step: 606/1541, loss: 0.6173651218414307 2023-01-23 05:38:00.614082: step: 608/1541, loss: 2.029489040374756 2023-01-23 05:38:01.489694: step: 610/1541, loss: 2.4544496536254883 2023-01-23 05:38:02.361409: step: 612/1541, loss: 1.6903481483459473 2023-01-23 05:38:03.208666: step: 614/1541, loss: 0.67680823802948 2023-01-23 05:38:04.123527: step: 616/1541, loss: 0.5806572437286377 2023-01-23 05:38:05.002735: step: 618/1541, loss: 3.2720494270324707 2023-01-23 05:38:05.895017: step: 620/1541, loss: 1.9786704778671265 2023-01-23 05:38:06.792912: step: 622/1541, loss: 1.6957507133483887 2023-01-23 05:38:07.680530: step: 624/1541, loss: 0.7497726082801819 2023-01-23 05:38:08.523112: step: 626/1541, loss: 1.1738263368606567 2023-01-23 05:38:09.373834: step: 628/1541, loss: 1.9140822887420654 2023-01-23 05:38:10.283117: step: 630/1541, loss: 0.9419687390327454 2023-01-23 05:38:11.142228: step: 632/1541, loss: 1.092646837234497 2023-01-23 05:38:12.051139: step: 634/1541, loss: 1.4527274370193481 2023-01-23 05:38:12.934206: step: 636/1541, loss: 1.1887577772140503 2023-01-23 05:38:13.832360: step: 638/1541, loss: 0.6949604153633118 2023-01-23 05:38:14.713048: step: 640/1541, loss: 1.2462588548660278 2023-01-23 05:38:15.575682: step: 642/1541, loss: 1.235133171081543 2023-01-23 05:38:16.473555: step: 644/1541, loss: 0.4895918369293213 2023-01-23 05:38:17.378268: step: 646/1541, loss: 4.999019622802734 2023-01-23 05:38:18.284237: step: 648/1541, loss: 1.1032328605651855 2023-01-23 05:38:19.221080: step: 650/1541, loss: 1.0122920274734497 2023-01-23 05:38:20.129199: step: 652/1541, loss: 3.4123764038085938 2023-01-23 05:38:20.997836: step: 654/1541, loss: 0.5303221344947815 2023-01-23 05:38:21.920354: step: 656/1541, loss: 1.6116917133331299 2023-01-23 05:38:22.805343: step: 658/1541, loss: 1.7867141962051392 2023-01-23 05:38:23.672039: step: 660/1541, loss: 0.6609826683998108 2023-01-23 05:38:24.582526: step: 662/1541, loss: 1.8862817287445068 2023-01-23 05:38:25.477395: step: 664/1541, loss: 1.5467135906219482 2023-01-23 05:38:26.367554: step: 666/1541, loss: 3.876251697540283 2023-01-23 05:38:27.382257: step: 668/1541, loss: 1.007912278175354 2023-01-23 05:38:28.248894: step: 670/1541, loss: 0.821125328540802 2023-01-23 05:38:29.137379: step: 672/1541, loss: 3.807948350906372 2023-01-23 05:38:30.058345: step: 674/1541, loss: 2.305638313293457 2023-01-23 05:38:30.975295: step: 676/1541, loss: 3.5812366008758545 2023-01-23 05:38:31.901619: step: 678/1541, loss: 1.2806730270385742 2023-01-23 05:38:32.780569: step: 680/1541, loss: 1.4951961040496826 2023-01-23 05:38:33.651075: step: 682/1541, loss: 3.6852235794067383 2023-01-23 05:38:34.589644: step: 684/1541, loss: 0.8421140313148499 2023-01-23 05:38:35.475141: step: 686/1541, loss: 6.339329242706299 2023-01-23 05:38:36.361504: step: 688/1541, loss: 2.3710787296295166 2023-01-23 05:38:37.247569: step: 690/1541, loss: 1.82338547706604 2023-01-23 05:38:38.231536: step: 692/1541, loss: 2.021684169769287 2023-01-23 05:38:39.111836: step: 694/1541, loss: 0.9735959768295288 2023-01-23 05:38:40.022096: step: 696/1541, loss: 1.074405550956726 2023-01-23 05:38:40.900521: step: 698/1541, loss: 1.5403125286102295 2023-01-23 05:38:41.778345: step: 700/1541, loss: 0.9360654354095459 2023-01-23 05:38:42.678746: step: 702/1541, loss: 2.0079541206359863 2023-01-23 05:38:43.589137: step: 704/1541, loss: 0.9981701970100403 2023-01-23 05:38:44.488865: step: 706/1541, loss: 0.556450366973877 2023-01-23 05:38:45.377303: step: 708/1541, loss: 2.1730449199676514 2023-01-23 05:38:46.309426: step: 710/1541, loss: 4.812526702880859 2023-01-23 05:38:47.189333: step: 712/1541, loss: 0.8779131770133972 2023-01-23 05:38:48.056184: step: 714/1541, loss: 1.1348655223846436 2023-01-23 05:38:48.920168: step: 716/1541, loss: 1.2197082042694092 2023-01-23 05:38:49.814772: step: 718/1541, loss: 0.927503228187561 2023-01-23 05:38:50.734832: step: 720/1541, loss: 4.382750988006592 2023-01-23 05:38:51.653724: step: 722/1541, loss: 1.206597089767456 2023-01-23 05:38:52.527088: step: 724/1541, loss: 0.8663541078567505 2023-01-23 05:38:53.420173: step: 726/1541, loss: 3.3012945652008057 2023-01-23 05:38:54.363527: step: 728/1541, loss: 9.899883270263672 2023-01-23 05:38:55.307281: step: 730/1541, loss: 2.6362767219543457 2023-01-23 05:38:56.201965: step: 732/1541, loss: 0.9460655450820923 2023-01-23 05:38:57.110469: step: 734/1541, loss: 1.486143946647644 2023-01-23 05:38:58.016118: step: 736/1541, loss: 3.2982351779937744 2023-01-23 05:38:58.895472: step: 738/1541, loss: 4.523658752441406 2023-01-23 05:38:59.824909: step: 740/1541, loss: 1.6599339246749878 2023-01-23 05:39:00.716936: step: 742/1541, loss: 3.534318447113037 2023-01-23 05:39:01.657370: step: 744/1541, loss: 1.6631746292114258 2023-01-23 05:39:02.571962: step: 746/1541, loss: 1.2761447429656982 2023-01-23 05:39:03.455435: step: 748/1541, loss: 0.9081157445907593 2023-01-23 05:39:04.341384: step: 750/1541, loss: 1.5911427736282349 2023-01-23 05:39:05.285322: step: 752/1541, loss: 1.4705891609191895 2023-01-23 05:39:06.203109: step: 754/1541, loss: 2.359771251678467 2023-01-23 05:39:07.076919: step: 756/1541, loss: 1.1116478443145752 2023-01-23 05:39:07.972061: step: 758/1541, loss: 0.6801003217697144 2023-01-23 05:39:08.835034: step: 760/1541, loss: 1.2552380561828613 2023-01-23 05:39:09.729887: step: 762/1541, loss: 1.701100468635559 2023-01-23 05:39:10.591699: step: 764/1541, loss: 0.5407575964927673 2023-01-23 05:39:11.483267: step: 766/1541, loss: 1.6372880935668945 2023-01-23 05:39:12.454541: step: 768/1541, loss: 1.5123751163482666 2023-01-23 05:39:13.328150: step: 770/1541, loss: 3.2079434394836426 2023-01-23 05:39:14.266050: step: 772/1541, loss: 3.7659363746643066 2023-01-23 05:39:15.191613: step: 774/1541, loss: 2.806173801422119 2023-01-23 05:39:16.085657: step: 776/1541, loss: 0.5104154348373413 2023-01-23 05:39:16.969742: step: 778/1541, loss: 1.0944037437438965 2023-01-23 05:39:17.891787: step: 780/1541, loss: 0.6311538219451904 2023-01-23 05:39:18.803512: step: 782/1541, loss: 1.0688350200653076 2023-01-23 05:39:19.660912: step: 784/1541, loss: 0.3722623884677887 2023-01-23 05:39:20.611544: step: 786/1541, loss: 1.3489611148834229 2023-01-23 05:39:21.543292: step: 788/1541, loss: 1.6529481410980225 2023-01-23 05:39:22.474250: step: 790/1541, loss: 1.0841896533966064 2023-01-23 05:39:23.370666: step: 792/1541, loss: 1.205864429473877 2023-01-23 05:39:24.293127: step: 794/1541, loss: 1.6919013261795044 2023-01-23 05:39:25.228988: step: 796/1541, loss: 8.166736602783203 2023-01-23 05:39:26.151115: step: 798/1541, loss: 3.954695701599121 2023-01-23 05:39:27.046370: step: 800/1541, loss: 3.6065731048583984 2023-01-23 05:39:28.031526: step: 802/1541, loss: 1.4020764827728271 2023-01-23 05:39:28.969833: step: 804/1541, loss: 0.6554213762283325 2023-01-23 05:39:29.910749: step: 806/1541, loss: 0.6691296100616455 2023-01-23 05:39:30.811991: step: 808/1541, loss: 1.2838385105133057 2023-01-23 05:39:31.696732: step: 810/1541, loss: 1.01707923412323 2023-01-23 05:39:32.604700: step: 812/1541, loss: 1.0517486333847046 2023-01-23 05:39:33.592911: step: 814/1541, loss: 0.8140899538993835 2023-01-23 05:39:34.515029: step: 816/1541, loss: 3.285930633544922 2023-01-23 05:39:35.384923: step: 818/1541, loss: 1.3024712800979614 2023-01-23 05:39:36.237131: step: 820/1541, loss: 0.5409526824951172 2023-01-23 05:39:37.171659: step: 822/1541, loss: 1.9046930074691772 2023-01-23 05:39:38.058891: step: 824/1541, loss: 1.0749166011810303 2023-01-23 05:39:38.977669: step: 826/1541, loss: 0.8627020716667175 2023-01-23 05:39:39.855913: step: 828/1541, loss: 0.9999043941497803 2023-01-23 05:39:40.795714: step: 830/1541, loss: 1.3739582300186157 2023-01-23 05:39:41.666019: step: 832/1541, loss: 0.9103407263755798 2023-01-23 05:39:42.564318: step: 834/1541, loss: 2.0455121994018555 2023-01-23 05:39:43.508838: step: 836/1541, loss: 1.0655159950256348 2023-01-23 05:39:44.428086: step: 838/1541, loss: 1.3451341390609741 2023-01-23 05:39:45.274711: step: 840/1541, loss: 0.9672605395317078 2023-01-23 05:39:46.176279: step: 842/1541, loss: 0.7030159831047058 2023-01-23 05:39:47.068918: step: 844/1541, loss: 3.6105761528015137 2023-01-23 05:39:47.961099: step: 846/1541, loss: 2.915496349334717 2023-01-23 05:39:48.845521: step: 848/1541, loss: 0.9145600199699402 2023-01-23 05:39:49.779721: step: 850/1541, loss: 2.4000179767608643 2023-01-23 05:39:50.684062: step: 852/1541, loss: 0.9903382658958435 2023-01-23 05:39:51.601434: step: 854/1541, loss: 2.4247987270355225 2023-01-23 05:39:52.490521: step: 856/1541, loss: 0.6311684846878052 2023-01-23 05:39:53.422367: step: 858/1541, loss: 1.1254591941833496 2023-01-23 05:39:54.322415: step: 860/1541, loss: 4.176446437835693 2023-01-23 05:39:55.310425: step: 862/1541, loss: 4.126867771148682 2023-01-23 05:39:56.196348: step: 864/1541, loss: 0.8681380152702332 2023-01-23 05:39:57.153058: step: 866/1541, loss: 0.9118281602859497 2023-01-23 05:39:58.097931: step: 868/1541, loss: 1.3308370113372803 2023-01-23 05:39:59.040328: step: 870/1541, loss: 3.740147352218628 2023-01-23 05:39:59.940930: step: 872/1541, loss: 1.1564916372299194 2023-01-23 05:40:00.847146: step: 874/1541, loss: 1.9042667150497437 2023-01-23 05:40:01.780621: step: 876/1541, loss: 1.2170295715332031 2023-01-23 05:40:02.650911: step: 878/1541, loss: 1.7399678230285645 2023-01-23 05:40:03.543489: step: 880/1541, loss: 0.7578601241111755 2023-01-23 05:40:04.423775: step: 882/1541, loss: 1.0148885250091553 2023-01-23 05:40:05.343006: step: 884/1541, loss: 1.263624906539917 2023-01-23 05:40:06.284752: step: 886/1541, loss: 1.2854769229888916 2023-01-23 05:40:07.260396: step: 888/1541, loss: 1.0669692754745483 2023-01-23 05:40:08.144616: step: 890/1541, loss: 0.6661832332611084 2023-01-23 05:40:09.008954: step: 892/1541, loss: 1.2997742891311646 2023-01-23 05:40:09.952589: step: 894/1541, loss: 1.61784827709198 2023-01-23 05:40:10.842177: step: 896/1541, loss: 0.8550770878791809 2023-01-23 05:40:11.763652: step: 898/1541, loss: 1.6829886436462402 2023-01-23 05:40:12.653577: step: 900/1541, loss: 0.7500547766685486 2023-01-23 05:40:13.568531: step: 902/1541, loss: 1.8062907457351685 2023-01-23 05:40:14.459165: step: 904/1541, loss: 0.5518333315849304 2023-01-23 05:40:15.340371: step: 906/1541, loss: 1.553147315979004 2023-01-23 05:40:16.208734: step: 908/1541, loss: 1.6763486862182617 2023-01-23 05:40:17.153265: step: 910/1541, loss: 3.720822811126709 2023-01-23 05:40:18.031857: step: 912/1541, loss: 0.9188764691352844 2023-01-23 05:40:18.915296: step: 914/1541, loss: 1.3281248807907104 2023-01-23 05:40:19.816258: step: 916/1541, loss: 0.8420493602752686 2023-01-23 05:40:20.678329: step: 918/1541, loss: 1.4265919923782349 2023-01-23 05:40:21.606506: step: 920/1541, loss: 1.0775556564331055 2023-01-23 05:40:22.463691: step: 922/1541, loss: 1.2726500034332275 2023-01-23 05:40:23.393281: step: 924/1541, loss: 1.2653234004974365 2023-01-23 05:40:24.318590: step: 926/1541, loss: 1.2147839069366455 2023-01-23 05:40:25.196716: step: 928/1541, loss: 0.6598694920539856 2023-01-23 05:40:26.056317: step: 930/1541, loss: 0.7534202337265015 2023-01-23 05:40:26.921542: step: 932/1541, loss: 1.0738219022750854 2023-01-23 05:40:27.788950: step: 934/1541, loss: 2.894136428833008 2023-01-23 05:40:28.684750: step: 936/1541, loss: 1.007870078086853 2023-01-23 05:40:29.561269: step: 938/1541, loss: 1.0105441808700562 2023-01-23 05:40:30.478120: step: 940/1541, loss: 3.0708179473876953 2023-01-23 05:40:31.378104: step: 942/1541, loss: 0.5720767378807068 2023-01-23 05:40:32.242624: step: 944/1541, loss: 2.7712483406066895 2023-01-23 05:40:33.122796: step: 946/1541, loss: 2.2608728408813477 2023-01-23 05:40:34.061751: step: 948/1541, loss: 1.042288064956665 2023-01-23 05:40:34.948891: step: 950/1541, loss: 0.7256917357444763 2023-01-23 05:40:35.855717: step: 952/1541, loss: 0.5664117336273193 2023-01-23 05:40:36.766456: step: 954/1541, loss: 2.045644998550415 2023-01-23 05:40:37.724845: step: 956/1541, loss: 0.9101589918136597 2023-01-23 05:40:38.600925: step: 958/1541, loss: 2.0411057472229004 2023-01-23 05:40:39.549189: step: 960/1541, loss: 1.4617764949798584 2023-01-23 05:40:40.386443: step: 962/1541, loss: 0.6183345913887024 2023-01-23 05:40:41.284805: step: 964/1541, loss: 0.8436709642410278 2023-01-23 05:40:42.159143: step: 966/1541, loss: 4.376078128814697 2023-01-23 05:40:43.119893: step: 968/1541, loss: 1.3750861883163452 2023-01-23 05:40:44.042378: step: 970/1541, loss: 2.1217939853668213 2023-01-23 05:40:45.078768: step: 972/1541, loss: 1.117215871810913 2023-01-23 05:40:45.941098: step: 974/1541, loss: 1.1326651573181152 2023-01-23 05:40:46.854213: step: 976/1541, loss: 1.112348198890686 2023-01-23 05:40:47.731466: step: 978/1541, loss: 1.1802499294281006 2023-01-23 05:40:48.638941: step: 980/1541, loss: 0.7806020379066467 2023-01-23 05:40:49.584390: step: 982/1541, loss: 1.4217561483383179 2023-01-23 05:40:50.502786: step: 984/1541, loss: 1.6498178243637085 2023-01-23 05:40:51.401752: step: 986/1541, loss: 3.2792139053344727 2023-01-23 05:40:52.347452: step: 988/1541, loss: 1.1087559461593628 2023-01-23 05:40:53.203822: step: 990/1541, loss: 0.7434332370758057 2023-01-23 05:40:54.109992: step: 992/1541, loss: 1.349881887435913 2023-01-23 05:40:54.951443: step: 994/1541, loss: 1.4467982053756714 2023-01-23 05:40:55.942245: step: 996/1541, loss: 1.3478795289993286 2023-01-23 05:40:56.879251: step: 998/1541, loss: 0.8214835524559021 2023-01-23 05:40:57.777085: step: 1000/1541, loss: 1.1331408023834229 2023-01-23 05:40:58.648069: step: 1002/1541, loss: 0.7606554627418518 2023-01-23 05:40:59.531145: step: 1004/1541, loss: 0.630927324295044 2023-01-23 05:41:00.420586: step: 1006/1541, loss: 1.554240107536316 2023-01-23 05:41:01.391892: step: 1008/1541, loss: 1.9428428411483765 2023-01-23 05:41:02.314696: step: 1010/1541, loss: 2.429514169692993 2023-01-23 05:41:03.169654: step: 1012/1541, loss: 2.521569013595581 2023-01-23 05:41:04.054141: step: 1014/1541, loss: 3.274343729019165 2023-01-23 05:41:04.978067: step: 1016/1541, loss: 1.0180275440216064 2023-01-23 05:41:05.863492: step: 1018/1541, loss: 1.4314817190170288 2023-01-23 05:41:06.752058: step: 1020/1541, loss: 1.7817782163619995 2023-01-23 05:41:07.685347: step: 1022/1541, loss: 1.262722373008728 2023-01-23 05:41:08.555163: step: 1024/1541, loss: 2.321058988571167 2023-01-23 05:41:09.492294: step: 1026/1541, loss: 1.0083696842193604 2023-01-23 05:41:10.400659: step: 1028/1541, loss: 0.9067942500114441 2023-01-23 05:41:11.341922: step: 1030/1541, loss: 1.606325626373291 2023-01-23 05:41:12.218517: step: 1032/1541, loss: 2.944315195083618 2023-01-23 05:41:13.077992: step: 1034/1541, loss: 0.8938825130462646 2023-01-23 05:41:13.949132: step: 1036/1541, loss: 1.1191421747207642 2023-01-23 05:41:14.970349: step: 1038/1541, loss: 1.2402135133743286 2023-01-23 05:41:15.895408: step: 1040/1541, loss: 0.9875919818878174 2023-01-23 05:41:16.796535: step: 1042/1541, loss: 0.9655954837799072 2023-01-23 05:41:17.728202: step: 1044/1541, loss: 2.089672565460205 2023-01-23 05:41:18.610992: step: 1046/1541, loss: 1.083481788635254 2023-01-23 05:41:19.565810: step: 1048/1541, loss: 7.223906993865967 2023-01-23 05:41:20.478547: step: 1050/1541, loss: 0.9653569459915161 2023-01-23 05:41:21.394982: step: 1052/1541, loss: 1.1590183973312378 2023-01-23 05:41:22.346674: step: 1054/1541, loss: 1.1445742845535278 2023-01-23 05:41:23.320523: step: 1056/1541, loss: 2.698007822036743 2023-01-23 05:41:24.257839: step: 1058/1541, loss: 1.3785991668701172 2023-01-23 05:41:25.175807: step: 1060/1541, loss: 1.4730342626571655 2023-01-23 05:41:26.054813: step: 1062/1541, loss: 1.0530670881271362 2023-01-23 05:41:26.955119: step: 1064/1541, loss: 0.5014369487762451 2023-01-23 05:41:27.846976: step: 1066/1541, loss: 0.4162033796310425 2023-01-23 05:41:28.759435: step: 1068/1541, loss: 1.719609260559082 2023-01-23 05:41:29.635401: step: 1070/1541, loss: 1.1483678817749023 2023-01-23 05:41:30.510531: step: 1072/1541, loss: 0.7773274779319763 2023-01-23 05:41:31.361085: step: 1074/1541, loss: 3.1375012397766113 2023-01-23 05:41:32.263538: step: 1076/1541, loss: 0.7670774459838867 2023-01-23 05:41:33.179307: step: 1078/1541, loss: 1.0666108131408691 2023-01-23 05:41:34.100976: step: 1080/1541, loss: 2.1556951999664307 2023-01-23 05:41:34.980682: step: 1082/1541, loss: 1.2310165166854858 2023-01-23 05:41:35.854314: step: 1084/1541, loss: 1.3084360361099243 2023-01-23 05:41:36.757408: step: 1086/1541, loss: 0.8338478207588196 2023-01-23 05:41:37.619624: step: 1088/1541, loss: 0.9082543849945068 2023-01-23 05:41:38.526893: step: 1090/1541, loss: 1.1349190473556519 2023-01-23 05:41:39.406864: step: 1092/1541, loss: 0.8678721189498901 2023-01-23 05:41:40.364290: step: 1094/1541, loss: 0.8703381419181824 2023-01-23 05:41:41.224629: step: 1096/1541, loss: 0.9448682069778442 2023-01-23 05:41:42.119577: step: 1098/1541, loss: 0.6700873970985413 2023-01-23 05:41:42.999061: step: 1100/1541, loss: 0.864135205745697 2023-01-23 05:41:43.848689: step: 1102/1541, loss: 1.0474739074707031 2023-01-23 05:41:44.704389: step: 1104/1541, loss: 1.1490641832351685 2023-01-23 05:41:45.588614: step: 1106/1541, loss: 1.9170598983764648 2023-01-23 05:41:46.463218: step: 1108/1541, loss: 0.497238427400589 2023-01-23 05:41:47.333875: step: 1110/1541, loss: 0.4793906807899475 2023-01-23 05:41:48.217227: step: 1112/1541, loss: 1.328864574432373 2023-01-23 05:41:49.100982: step: 1114/1541, loss: 0.9266353249549866 2023-01-23 05:41:50.004297: step: 1116/1541, loss: 3.5386886596679688 2023-01-23 05:41:50.920158: step: 1118/1541, loss: 1.6983413696289062 2023-01-23 05:41:51.793104: step: 1120/1541, loss: 1.1637567281723022 2023-01-23 05:41:52.675548: step: 1122/1541, loss: 1.0516726970672607 2023-01-23 05:41:53.549136: step: 1124/1541, loss: 0.7888895273208618 2023-01-23 05:41:54.495654: step: 1126/1541, loss: 1.3940727710723877 2023-01-23 05:41:55.377874: step: 1128/1541, loss: 1.299588918685913 2023-01-23 05:41:56.280572: step: 1130/1541, loss: 0.7229692339897156 2023-01-23 05:41:57.207710: step: 1132/1541, loss: 3.130647897720337 2023-01-23 05:41:58.100404: step: 1134/1541, loss: 7.6350483894348145 2023-01-23 05:41:58.976930: step: 1136/1541, loss: 0.8330357670783997 2023-01-23 05:41:59.914308: step: 1138/1541, loss: 3.9817891120910645 2023-01-23 05:42:00.797154: step: 1140/1541, loss: 1.1766108274459839 2023-01-23 05:42:01.744556: step: 1142/1541, loss: 0.8310659527778625 2023-01-23 05:42:02.628045: step: 1144/1541, loss: 0.7005575895309448 2023-01-23 05:42:03.515722: step: 1146/1541, loss: 1.0435000658035278 2023-01-23 05:42:04.358839: step: 1148/1541, loss: 0.5893216133117676 2023-01-23 05:42:05.269040: step: 1150/1541, loss: 0.8827902674674988 2023-01-23 05:42:06.190452: step: 1152/1541, loss: 0.7183274626731873 2023-01-23 05:42:07.075812: step: 1154/1541, loss: 1.7627038955688477 2023-01-23 05:42:07.927769: step: 1156/1541, loss: 0.9964470267295837 2023-01-23 05:42:08.823517: step: 1158/1541, loss: 0.35042139887809753 2023-01-23 05:42:09.702091: step: 1160/1541, loss: 1.1248302459716797 2023-01-23 05:42:10.596845: step: 1162/1541, loss: 1.5252195596694946 2023-01-23 05:42:11.537668: step: 1164/1541, loss: 1.0603771209716797 2023-01-23 05:42:12.492694: step: 1166/1541, loss: 1.9926527738571167 2023-01-23 05:42:13.349627: step: 1168/1541, loss: 0.8988853096961975 2023-01-23 05:42:14.233247: step: 1170/1541, loss: 0.9285564422607422 2023-01-23 05:42:15.128153: step: 1172/1541, loss: 1.4795451164245605 2023-01-23 05:42:16.040512: step: 1174/1541, loss: 1.3135836124420166 2023-01-23 05:42:16.939287: step: 1176/1541, loss: 2.4556634426116943 2023-01-23 05:42:17.806590: step: 1178/1541, loss: 1.204218864440918 2023-01-23 05:42:18.700425: step: 1180/1541, loss: 0.7517573833465576 2023-01-23 05:42:19.587773: step: 1182/1541, loss: 3.3576817512512207 2023-01-23 05:42:20.507004: step: 1184/1541, loss: 1.4383503198623657 2023-01-23 05:42:21.354866: step: 1186/1541, loss: 0.6701014637947083 2023-01-23 05:42:22.340118: step: 1188/1541, loss: 1.1260266304016113 2023-01-23 05:42:23.231893: step: 1190/1541, loss: 1.073509693145752 2023-01-23 05:42:24.115927: step: 1192/1541, loss: 1.654390811920166 2023-01-23 05:42:24.980610: step: 1194/1541, loss: 1.076635479927063 2023-01-23 05:42:25.918315: step: 1196/1541, loss: 0.9624966382980347 2023-01-23 05:42:26.808423: step: 1198/1541, loss: 1.194273591041565 2023-01-23 05:42:27.693974: step: 1200/1541, loss: 0.7406443953514099 2023-01-23 05:42:28.592879: step: 1202/1541, loss: 4.236730098724365 2023-01-23 05:42:29.491660: step: 1204/1541, loss: 1.4188071489334106 2023-01-23 05:42:30.506038: step: 1206/1541, loss: 1.527176022529602 2023-01-23 05:42:31.372963: step: 1208/1541, loss: 1.6665074825286865 2023-01-23 05:42:32.274753: step: 1210/1541, loss: 2.2908644676208496 2023-01-23 05:42:33.173373: step: 1212/1541, loss: 1.8637827634811401 2023-01-23 05:42:34.069644: step: 1214/1541, loss: 0.7605485320091248 2023-01-23 05:42:35.025048: step: 1216/1541, loss: 0.8999946117401123 2023-01-23 05:42:35.966216: step: 1218/1541, loss: 1.6918113231658936 2023-01-23 05:42:36.828804: step: 1220/1541, loss: 1.683457851409912 2023-01-23 05:42:37.726344: step: 1222/1541, loss: 1.329915165901184 2023-01-23 05:42:38.628731: step: 1224/1541, loss: 1.3772979974746704 2023-01-23 05:42:39.505522: step: 1226/1541, loss: 1.7025526762008667 2023-01-23 05:42:40.449950: step: 1228/1541, loss: 1.630307674407959 2023-01-23 05:42:41.378643: step: 1230/1541, loss: 1.07450532913208 2023-01-23 05:42:42.242542: step: 1232/1541, loss: 1.1046234369277954 2023-01-23 05:42:43.172299: step: 1234/1541, loss: 1.0370769500732422 2023-01-23 05:42:44.083547: step: 1236/1541, loss: 1.6527724266052246 2023-01-23 05:42:45.020760: step: 1238/1541, loss: 1.7760194540023804 2023-01-23 05:42:45.888976: step: 1240/1541, loss: 0.973755419254303 2023-01-23 05:42:46.760492: step: 1242/1541, loss: 0.7106781005859375 2023-01-23 05:42:47.746171: step: 1244/1541, loss: 1.4621316194534302 2023-01-23 05:42:48.676639: step: 1246/1541, loss: 0.9841216802597046 2023-01-23 05:42:49.548474: step: 1248/1541, loss: 0.7304646372795105 2023-01-23 05:42:50.446418: step: 1250/1541, loss: 1.5208619832992554 2023-01-23 05:42:51.339664: step: 1252/1541, loss: 3.3498411178588867 2023-01-23 05:42:52.219758: step: 1254/1541, loss: 1.1288200616836548 2023-01-23 05:42:53.078928: step: 1256/1541, loss: 1.0744966268539429 2023-01-23 05:42:54.006926: step: 1258/1541, loss: 1.100638508796692 2023-01-23 05:42:54.905890: step: 1260/1541, loss: 1.4606635570526123 2023-01-23 05:42:55.765422: step: 1262/1541, loss: 3.125903606414795 2023-01-23 05:42:56.645024: step: 1264/1541, loss: 1.3439881801605225 2023-01-23 05:42:57.587563: step: 1266/1541, loss: 2.252878427505493 2023-01-23 05:42:58.522025: step: 1268/1541, loss: 1.1532516479492188 2023-01-23 05:42:59.412255: step: 1270/1541, loss: 0.8205967545509338 2023-01-23 05:43:00.277616: step: 1272/1541, loss: 1.0379847288131714 2023-01-23 05:43:01.225887: step: 1274/1541, loss: 1.493708610534668 2023-01-23 05:43:02.159113: step: 1276/1541, loss: 1.0340602397918701 2023-01-23 05:43:03.087602: step: 1278/1541, loss: 1.8470746278762817 2023-01-23 05:43:04.010706: step: 1280/1541, loss: 1.199310302734375 2023-01-23 05:43:04.862627: step: 1282/1541, loss: 2.410456895828247 2023-01-23 05:43:05.838077: step: 1284/1541, loss: 0.6385321617126465 2023-01-23 05:43:06.713981: step: 1286/1541, loss: 1.605905532836914 2023-01-23 05:43:07.680908: step: 1288/1541, loss: 2.30562686920166 2023-01-23 05:43:08.616511: step: 1290/1541, loss: 1.0734028816223145 2023-01-23 05:43:09.513442: step: 1292/1541, loss: 0.7593689560890198 2023-01-23 05:43:10.429027: step: 1294/1541, loss: 0.4658966660499573 2023-01-23 05:43:11.380136: step: 1296/1541, loss: 1.8390272855758667 2023-01-23 05:43:12.296774: step: 1298/1541, loss: 0.6313835382461548 2023-01-23 05:43:13.257171: step: 1300/1541, loss: 1.809190273284912 2023-01-23 05:43:14.149123: step: 1302/1541, loss: 0.7321100831031799 2023-01-23 05:43:15.067394: step: 1304/1541, loss: 0.6296976804733276 2023-01-23 05:43:16.015546: step: 1306/1541, loss: 0.822694718837738 2023-01-23 05:43:16.908928: step: 1308/1541, loss: 1.207354187965393 2023-01-23 05:43:17.811209: step: 1310/1541, loss: 0.5077361464500427 2023-01-23 05:43:18.787364: step: 1312/1541, loss: 1.110336422920227 2023-01-23 05:43:19.674282: step: 1314/1541, loss: 0.7979339957237244 2023-01-23 05:43:20.567916: step: 1316/1541, loss: 1.072474718093872 2023-01-23 05:43:21.435078: step: 1318/1541, loss: 2.034578323364258 2023-01-23 05:43:22.326463: step: 1320/1541, loss: 1.735090732574463 2023-01-23 05:43:23.168045: step: 1322/1541, loss: 3.3113932609558105 2023-01-23 05:43:24.079796: step: 1324/1541, loss: 8.112021446228027 2023-01-23 05:43:24.959667: step: 1326/1541, loss: 1.1057132482528687 2023-01-23 05:43:25.844595: step: 1328/1541, loss: 0.6884925961494446 2023-01-23 05:43:26.748807: step: 1330/1541, loss: 0.6329652667045593 2023-01-23 05:43:27.648366: step: 1332/1541, loss: 0.6705090403556824 2023-01-23 05:43:28.542953: step: 1334/1541, loss: 0.8210510611534119 2023-01-23 05:43:29.452779: step: 1336/1541, loss: 1.5607315301895142 2023-01-23 05:43:30.365349: step: 1338/1541, loss: 1.3072490692138672 2023-01-23 05:43:31.242304: step: 1340/1541, loss: 1.1966631412506104 2023-01-23 05:43:32.284481: step: 1342/1541, loss: 1.0705561637878418 2023-01-23 05:43:33.157992: step: 1344/1541, loss: 0.5803431272506714 2023-01-23 05:43:34.039116: step: 1346/1541, loss: 1.065632939338684 2023-01-23 05:43:34.945487: step: 1348/1541, loss: 0.8458457589149475 2023-01-23 05:43:35.844857: step: 1350/1541, loss: 1.186862826347351 2023-01-23 05:43:36.749841: step: 1352/1541, loss: 2.2673633098602295 2023-01-23 05:43:37.642718: step: 1354/1541, loss: 1.2490134239196777 2023-01-23 05:43:38.563209: step: 1356/1541, loss: 1.108992576599121 2023-01-23 05:43:39.436081: step: 1358/1541, loss: 2.526517868041992 2023-01-23 05:43:40.318569: step: 1360/1541, loss: 1.2034488916397095 2023-01-23 05:43:41.194748: step: 1362/1541, loss: 1.57362699508667 2023-01-23 05:43:42.114655: step: 1364/1541, loss: 8.92070198059082 2023-01-23 05:43:43.057698: step: 1366/1541, loss: 0.7650074362754822 2023-01-23 05:43:43.960231: step: 1368/1541, loss: 1.2815308570861816 2023-01-23 05:43:44.862914: step: 1370/1541, loss: 1.3602535724639893 2023-01-23 05:43:45.767513: step: 1372/1541, loss: 0.8700757026672363 2023-01-23 05:43:46.722762: step: 1374/1541, loss: 3.250920534133911 2023-01-23 05:43:47.599645: step: 1376/1541, loss: 1.4738929271697998 2023-01-23 05:43:48.456608: step: 1378/1541, loss: 1.3059556484222412 2023-01-23 05:43:49.335625: step: 1380/1541, loss: 2.7736573219299316 2023-01-23 05:43:50.191546: step: 1382/1541, loss: 0.8150578737258911 2023-01-23 05:43:51.064635: step: 1384/1541, loss: 3.537484645843506 2023-01-23 05:43:51.967310: step: 1386/1541, loss: 3.0387768745422363 2023-01-23 05:43:52.826532: step: 1388/1541, loss: 0.5253520011901855 2023-01-23 05:43:53.784447: step: 1390/1541, loss: 1.2493298053741455 2023-01-23 05:43:54.698409: step: 1392/1541, loss: 0.8072249293327332 2023-01-23 05:43:55.618485: step: 1394/1541, loss: 1.505059838294983 2023-01-23 05:43:56.581145: step: 1396/1541, loss: 1.2521692514419556 2023-01-23 05:43:57.540812: step: 1398/1541, loss: 0.77716064453125 2023-01-23 05:43:58.403848: step: 1400/1541, loss: 0.780018150806427 2023-01-23 05:43:59.306081: step: 1402/1541, loss: 1.2337162494659424 2023-01-23 05:44:00.225145: step: 1404/1541, loss: 2.156062602996826 2023-01-23 05:44:01.177993: step: 1406/1541, loss: 0.8916817307472229 2023-01-23 05:44:02.057329: step: 1408/1541, loss: 1.019667387008667 2023-01-23 05:44:02.924796: step: 1410/1541, loss: 1.6657134294509888 2023-01-23 05:44:03.818462: step: 1412/1541, loss: 0.942428708076477 2023-01-23 05:44:04.744184: step: 1414/1541, loss: 1.7487175464630127 2023-01-23 05:44:05.681744: step: 1416/1541, loss: 1.586052417755127 2023-01-23 05:44:06.608960: step: 1418/1541, loss: 0.989467978477478 2023-01-23 05:44:07.553637: step: 1420/1541, loss: 3.72799015045166 2023-01-23 05:44:08.499738: step: 1422/1541, loss: 3.270643472671509 2023-01-23 05:44:09.447956: step: 1424/1541, loss: 1.3398218154907227 2023-01-23 05:44:10.317261: step: 1426/1541, loss: 1.967323660850525 2023-01-23 05:44:11.219991: step: 1428/1541, loss: 0.5129668712615967 2023-01-23 05:44:12.095902: step: 1430/1541, loss: 0.9193381071090698 2023-01-23 05:44:13.047462: step: 1432/1541, loss: 1.8118643760681152 2023-01-23 05:44:13.957252: step: 1434/1541, loss: 1.4005398750305176 2023-01-23 05:44:14.879839: step: 1436/1541, loss: 0.9203896522521973 2023-01-23 05:44:15.788896: step: 1438/1541, loss: 1.0631682872772217 2023-01-23 05:44:16.693609: step: 1440/1541, loss: 1.3773738145828247 2023-01-23 05:44:17.647608: step: 1442/1541, loss: 1.6785871982574463 2023-01-23 05:44:18.625403: step: 1444/1541, loss: 1.3237967491149902 2023-01-23 05:44:19.523276: step: 1446/1541, loss: 7.91560697555542 2023-01-23 05:44:20.410066: step: 1448/1541, loss: 1.1357182264328003 2023-01-23 05:44:21.314206: step: 1450/1541, loss: 0.7736692428588867 2023-01-23 05:44:22.182497: step: 1452/1541, loss: 0.47160667181015015 2023-01-23 05:44:23.118745: step: 1454/1541, loss: 1.5217994451522827 2023-01-23 05:44:23.988925: step: 1456/1541, loss: 0.6986419558525085 2023-01-23 05:44:24.833990: step: 1458/1541, loss: 0.7418634295463562 2023-01-23 05:44:25.744419: step: 1460/1541, loss: 3.1624040603637695 2023-01-23 05:44:26.660573: step: 1462/1541, loss: 1.0770245790481567 2023-01-23 05:44:27.587224: step: 1464/1541, loss: 0.940931499004364 2023-01-23 05:44:28.463733: step: 1466/1541, loss: 1.929619550704956 2023-01-23 05:44:29.344567: step: 1468/1541, loss: 1.7318669557571411 2023-01-23 05:44:30.219087: step: 1470/1541, loss: 5.904043197631836 2023-01-23 05:44:31.113319: step: 1472/1541, loss: 1.3642975091934204 2023-01-23 05:44:31.987339: step: 1474/1541, loss: 0.7200710773468018 2023-01-23 05:44:32.919842: step: 1476/1541, loss: 0.9424739480018616 2023-01-23 05:44:33.791144: step: 1478/1541, loss: 0.9995847344398499 2023-01-23 05:44:34.722472: step: 1480/1541, loss: 1.4477812051773071 2023-01-23 05:44:35.642017: step: 1482/1541, loss: 1.3524032831192017 2023-01-23 05:44:36.535223: step: 1484/1541, loss: 1.4630777835845947 2023-01-23 05:44:37.425743: step: 1486/1541, loss: 3.2109427452087402 2023-01-23 05:44:38.303200: step: 1488/1541, loss: 3.0914127826690674 2023-01-23 05:44:39.184640: step: 1490/1541, loss: 0.7041040062904358 2023-01-23 05:44:40.169372: step: 1492/1541, loss: 1.1098215579986572 2023-01-23 05:44:41.070499: step: 1494/1541, loss: 1.0646458864212036 2023-01-23 05:44:41.955069: step: 1496/1541, loss: 0.7781038284301758 2023-01-23 05:44:42.817059: step: 1498/1541, loss: 3.1600563526153564 2023-01-23 05:44:43.743385: step: 1500/1541, loss: 0.671109676361084 2023-01-23 05:44:44.666898: step: 1502/1541, loss: 1.0317310094833374 2023-01-23 05:44:45.551873: step: 1504/1541, loss: 0.943771243095398 2023-01-23 05:44:46.458006: step: 1506/1541, loss: 1.8224860429763794 2023-01-23 05:44:47.336804: step: 1508/1541, loss: 0.6239438056945801 2023-01-23 05:44:48.262466: step: 1510/1541, loss: 3.4383280277252197 2023-01-23 05:44:49.138240: step: 1512/1541, loss: 1.3556110858917236 2023-01-23 05:44:50.006995: step: 1514/1541, loss: 0.8630532026290894 2023-01-23 05:44:50.943785: step: 1516/1541, loss: 0.8935507535934448 2023-01-23 05:44:51.980299: step: 1518/1541, loss: 1.677878737449646 2023-01-23 05:44:52.851532: step: 1520/1541, loss: 4.769081115722656 2023-01-23 05:44:53.726500: step: 1522/1541, loss: 0.8705658912658691 2023-01-23 05:44:54.676913: step: 1524/1541, loss: 7.381287097930908 2023-01-23 05:44:55.565143: step: 1526/1541, loss: 0.6420593857765198 2023-01-23 05:44:56.443275: step: 1528/1541, loss: 3.837244987487793 2023-01-23 05:44:57.339117: step: 1530/1541, loss: 1.5833369493484497 2023-01-23 05:44:58.176402: step: 1532/1541, loss: 1.1883057355880737 2023-01-23 05:44:59.109677: step: 1534/1541, loss: 1.023776888847351 2023-01-23 05:45:00.002283: step: 1536/1541, loss: 0.7736681699752808 2023-01-23 05:45:00.950340: step: 1538/1541, loss: 0.5829282402992249 2023-01-23 05:45:01.904436: step: 1540/1541, loss: 0.912101686000824 2023-01-23 05:45:02.798184: step: 1542/1541, loss: 0.5790547132492065 2023-01-23 05:45:03.706340: step: 1544/1541, loss: 1.2643414735794067 2023-01-23 05:45:04.622244: step: 1546/1541, loss: 0.7917556762695312 2023-01-23 05:45:05.497691: step: 1548/1541, loss: 0.9288641810417175 2023-01-23 05:45:06.425201: step: 1550/1541, loss: 6.156045913696289 2023-01-23 05:45:07.388498: step: 1552/1541, loss: 1.8666776418685913 2023-01-23 05:45:08.250078: step: 1554/1541, loss: 1.9716951847076416 2023-01-23 05:45:09.140198: step: 1556/1541, loss: 0.38349801301956177 2023-01-23 05:45:10.042116: step: 1558/1541, loss: 0.935680627822876 2023-01-23 05:45:10.892819: step: 1560/1541, loss: 0.8560469150543213 2023-01-23 05:45:11.798198: step: 1562/1541, loss: 0.809477686882019 2023-01-23 05:45:12.674430: step: 1564/1541, loss: 1.4326037168502808 2023-01-23 05:45:13.561090: step: 1566/1541, loss: 0.7021955251693726 2023-01-23 05:45:14.435578: step: 1568/1541, loss: 0.9265598654747009 2023-01-23 05:45:15.355751: step: 1570/1541, loss: 3.105553388595581 2023-01-23 05:45:16.292240: step: 1572/1541, loss: 1.4246962070465088 2023-01-23 05:45:17.223317: step: 1574/1541, loss: 1.8809161186218262 2023-01-23 05:45:18.144156: step: 1576/1541, loss: 0.8886463642120361 2023-01-23 05:45:19.048093: step: 1578/1541, loss: 1.1214207410812378 2023-01-23 05:45:20.049332: step: 1580/1541, loss: 1.173776626586914 2023-01-23 05:45:20.956123: step: 1582/1541, loss: 0.9738593697547913 2023-01-23 05:45:21.948947: step: 1584/1541, loss: 3.109908103942871 2023-01-23 05:45:22.877240: step: 1586/1541, loss: 0.7699854969978333 2023-01-23 05:45:23.824917: step: 1588/1541, loss: 1.11375892162323 2023-01-23 05:45:24.666678: step: 1590/1541, loss: 0.6457172632217407 2023-01-23 05:45:25.609214: step: 1592/1541, loss: 0.965558648109436 2023-01-23 05:45:26.521722: step: 1594/1541, loss: 5.833235263824463 2023-01-23 05:45:27.424277: step: 1596/1541, loss: 0.7704886794090271 2023-01-23 05:45:28.326285: step: 1598/1541, loss: 0.8634025454521179 2023-01-23 05:45:29.196996: step: 1600/1541, loss: 0.7885506749153137 2023-01-23 05:45:30.225452: step: 1602/1541, loss: 0.4399392604827881 2023-01-23 05:45:31.151907: step: 1604/1541, loss: 0.6204991936683655 2023-01-23 05:45:32.074884: step: 1606/1541, loss: 1.3192076683044434 2023-01-23 05:45:32.991331: step: 1608/1541, loss: 1.956972599029541 2023-01-23 05:45:33.882422: step: 1610/1541, loss: 1.867971658706665 2023-01-23 05:45:34.767268: step: 1612/1541, loss: 0.42448103427886963 2023-01-23 05:45:35.645292: step: 1614/1541, loss: 0.9063072800636292 2023-01-23 05:45:36.557945: step: 1616/1541, loss: 1.143440842628479 2023-01-23 05:45:37.452519: step: 1618/1541, loss: 0.8120134472846985 2023-01-23 05:45:38.343296: step: 1620/1541, loss: 0.9403348565101624 2023-01-23 05:45:39.204202: step: 1622/1541, loss: 0.9480847716331482 2023-01-23 05:45:40.113924: step: 1624/1541, loss: 0.9993159174919128 2023-01-23 05:45:41.013108: step: 1626/1541, loss: 1.101432204246521 2023-01-23 05:45:41.915237: step: 1628/1541, loss: 1.4609699249267578 2023-01-23 05:45:42.896378: step: 1630/1541, loss: 2.274533748626709 2023-01-23 05:45:43.768405: step: 1632/1541, loss: 0.6743190288543701 2023-01-23 05:45:44.668300: step: 1634/1541, loss: 0.6621866226196289 2023-01-23 05:45:45.569571: step: 1636/1541, loss: 3.2905216217041016 2023-01-23 05:45:46.463038: step: 1638/1541, loss: 2.528634786605835 2023-01-23 05:45:47.344393: step: 1640/1541, loss: 2.013946771621704 2023-01-23 05:45:48.306851: step: 1642/1541, loss: 1.537509799003601 2023-01-23 05:45:49.190895: step: 1644/1541, loss: 0.9447765350341797 2023-01-23 05:45:50.074077: step: 1646/1541, loss: 0.8682153820991516 2023-01-23 05:45:50.971447: step: 1648/1541, loss: 0.9676815271377563 2023-01-23 05:45:51.867531: step: 1650/1541, loss: 0.9221408367156982 2023-01-23 05:45:52.768077: step: 1652/1541, loss: 1.8628968000411987 2023-01-23 05:45:53.688986: step: 1654/1541, loss: 0.993924617767334 2023-01-23 05:45:54.664080: step: 1656/1541, loss: 3.743267774581909 2023-01-23 05:45:55.600783: step: 1658/1541, loss: 0.9217202663421631 2023-01-23 05:45:56.526219: step: 1660/1541, loss: 0.9569315910339355 2023-01-23 05:45:57.461775: step: 1662/1541, loss: 1.6647839546203613 2023-01-23 05:45:58.371708: step: 1664/1541, loss: 0.9546614289283752 2023-01-23 05:45:59.267014: step: 1666/1541, loss: 0.7078774571418762 2023-01-23 05:46:00.224931: step: 1668/1541, loss: 0.8713483810424805 2023-01-23 05:46:01.120553: step: 1670/1541, loss: 0.5330361723899841 2023-01-23 05:46:02.019724: step: 1672/1541, loss: 1.087542176246643 2023-01-23 05:46:02.912034: step: 1674/1541, loss: 0.6625708341598511 2023-01-23 05:46:03.805783: step: 1676/1541, loss: 0.9301619529724121 2023-01-23 05:46:04.681048: step: 1678/1541, loss: 1.7340351343154907 2023-01-23 05:46:05.607275: step: 1680/1541, loss: 0.7971063256263733 2023-01-23 05:46:06.511533: step: 1682/1541, loss: 0.9444530606269836 2023-01-23 05:46:07.391682: step: 1684/1541, loss: 0.8164727687835693 2023-01-23 05:46:08.249707: step: 1686/1541, loss: 1.0145444869995117 2023-01-23 05:46:09.151339: step: 1688/1541, loss: 1.0963927507400513 2023-01-23 05:46:10.053574: step: 1690/1541, loss: 1.0452269315719604 2023-01-23 05:46:10.925257: step: 1692/1541, loss: 1.5847702026367188 2023-01-23 05:46:11.870966: step: 1694/1541, loss: 0.8736699819564819 2023-01-23 05:46:12.783093: step: 1696/1541, loss: 3.732642412185669 2023-01-23 05:46:13.678117: step: 1698/1541, loss: 1.0282161235809326 2023-01-23 05:46:14.553128: step: 1700/1541, loss: 0.6784660220146179 2023-01-23 05:46:15.455017: step: 1702/1541, loss: 1.651206374168396 2023-01-23 05:46:16.317098: step: 1704/1541, loss: 0.5707840919494629 2023-01-23 05:46:17.196581: step: 1706/1541, loss: 0.6646072864532471 2023-01-23 05:46:18.073562: step: 1708/1541, loss: 0.6712616682052612 2023-01-23 05:46:18.960935: step: 1710/1541, loss: 0.7892976999282837 2023-01-23 05:46:19.865653: step: 1712/1541, loss: 0.95856112241745 2023-01-23 05:46:20.734963: step: 1714/1541, loss: 1.5341455936431885 2023-01-23 05:46:21.620700: step: 1716/1541, loss: 1.2331271171569824 2023-01-23 05:46:22.483300: step: 1718/1541, loss: 0.9266798496246338 2023-01-23 05:46:23.481724: step: 1720/1541, loss: 0.9120493531227112 2023-01-23 05:46:24.349146: step: 1722/1541, loss: 1.176834225654602 2023-01-23 05:46:25.267007: step: 1724/1541, loss: 0.747674822807312 2023-01-23 05:46:26.158646: step: 1726/1541, loss: 1.5760486125946045 2023-01-23 05:46:27.042365: step: 1728/1541, loss: 1.181246280670166 2023-01-23 05:46:27.948932: step: 1730/1541, loss: 0.8975985050201416 2023-01-23 05:46:28.800095: step: 1732/1541, loss: 3.8191792964935303 2023-01-23 05:46:29.671471: step: 1734/1541, loss: 1.2317984104156494 2023-01-23 05:46:30.647290: step: 1736/1541, loss: 0.8093115091323853 2023-01-23 05:46:31.539756: step: 1738/1541, loss: 1.4378955364227295 2023-01-23 05:46:32.407993: step: 1740/1541, loss: 1.542055606842041 2023-01-23 05:46:33.302824: step: 1742/1541, loss: 0.7377315759658813 2023-01-23 05:46:34.186040: step: 1744/1541, loss: 0.4725762903690338 2023-01-23 05:46:35.126444: step: 1746/1541, loss: 1.8090299367904663 2023-01-23 05:46:35.987459: step: 1748/1541, loss: 0.6980903744697571 2023-01-23 05:46:36.923144: step: 1750/1541, loss: 3.4914562702178955 2023-01-23 05:46:37.812775: step: 1752/1541, loss: 1.0727226734161377 2023-01-23 05:46:38.700514: step: 1754/1541, loss: 1.0636626482009888 2023-01-23 05:46:39.617938: step: 1756/1541, loss: 8.829401016235352 2023-01-23 05:46:40.571095: step: 1758/1541, loss: 1.6832274198532104 2023-01-23 05:46:41.438892: step: 1760/1541, loss: 0.8351961374282837 2023-01-23 05:46:42.336738: step: 1762/1541, loss: 0.9122751355171204 2023-01-23 05:46:43.175588: step: 1764/1541, loss: 0.4436698853969574 2023-01-23 05:46:44.051467: step: 1766/1541, loss: 1.2489668130874634 2023-01-23 05:46:44.884849: step: 1768/1541, loss: 8.020298957824707 2023-01-23 05:46:45.747912: step: 1770/1541, loss: 0.8816038370132446 2023-01-23 05:46:46.689065: step: 1772/1541, loss: 2.2527637481689453 2023-01-23 05:46:47.584972: step: 1774/1541, loss: 1.1503829956054688 2023-01-23 05:46:48.476229: step: 1776/1541, loss: 0.7120333909988403 2023-01-23 05:46:49.377542: step: 1778/1541, loss: 1.0285701751708984 2023-01-23 05:46:50.269448: step: 1780/1541, loss: 1.4467754364013672 2023-01-23 05:46:51.155120: step: 1782/1541, loss: 2.581105947494507 2023-01-23 05:46:52.025178: step: 1784/1541, loss: 3.386091947555542 2023-01-23 05:46:52.919393: step: 1786/1541, loss: 0.8626254796981812 2023-01-23 05:46:53.828717: step: 1788/1541, loss: 0.8906155228614807 2023-01-23 05:46:54.736645: step: 1790/1541, loss: 1.7587776184082031 2023-01-23 05:46:55.657416: step: 1792/1541, loss: 0.8545590043067932 2023-01-23 05:46:56.597207: step: 1794/1541, loss: 1.5654702186584473 2023-01-23 05:46:57.479639: step: 1796/1541, loss: 0.7939989566802979 2023-01-23 05:46:58.391388: step: 1798/1541, loss: 0.9883958101272583 2023-01-23 05:46:59.344334: step: 1800/1541, loss: 1.3572686910629272 2023-01-23 05:47:00.224647: step: 1802/1541, loss: 1.7470496892929077 2023-01-23 05:47:01.118932: step: 1804/1541, loss: 1.1008243560791016 2023-01-23 05:47:02.011546: step: 1806/1541, loss: 1.1975288391113281 2023-01-23 05:47:02.879518: step: 1808/1541, loss: 1.91683030128479 2023-01-23 05:47:03.805611: step: 1810/1541, loss: 1.5102578401565552 2023-01-23 05:47:04.664671: step: 1812/1541, loss: 0.35128456354141235 2023-01-23 05:47:05.559486: step: 1814/1541, loss: 0.5740936994552612 2023-01-23 05:47:06.441619: step: 1816/1541, loss: 0.6952744722366333 2023-01-23 05:47:07.327664: step: 1818/1541, loss: 0.7140803337097168 2023-01-23 05:47:08.248634: step: 1820/1541, loss: 0.6144803166389465 2023-01-23 05:47:09.159706: step: 1822/1541, loss: 2.9791970252990723 2023-01-23 05:47:10.042985: step: 1824/1541, loss: 1.232001781463623 2023-01-23 05:47:10.986407: step: 1826/1541, loss: 1.1732232570648193 2023-01-23 05:47:11.897833: step: 1828/1541, loss: 0.9680188298225403 2023-01-23 05:47:12.839712: step: 1830/1541, loss: 4.435646057128906 2023-01-23 05:47:13.739461: step: 1832/1541, loss: 1.149672031402588 2023-01-23 05:47:14.654521: step: 1834/1541, loss: 0.9941973686218262 2023-01-23 05:47:15.524389: step: 1836/1541, loss: 0.6256107091903687 2023-01-23 05:47:16.411882: step: 1838/1541, loss: 0.7756385803222656 2023-01-23 05:47:17.332275: step: 1840/1541, loss: 1.167420744895935 2023-01-23 05:47:18.246324: step: 1842/1541, loss: 0.3874271512031555 2023-01-23 05:47:19.154007: step: 1844/1541, loss: 3.3144514560699463 2023-01-23 05:47:20.029599: step: 1846/1541, loss: 0.4213959872722626 2023-01-23 05:47:20.988738: step: 1848/1541, loss: 1.0107088088989258 2023-01-23 05:47:21.881484: step: 1850/1541, loss: 1.517749547958374 2023-01-23 05:47:22.757259: step: 1852/1541, loss: 3.2569100856781006 2023-01-23 05:47:23.653636: step: 1854/1541, loss: 0.8703327178955078 2023-01-23 05:47:24.564241: step: 1856/1541, loss: 0.5990935564041138 2023-01-23 05:47:25.528653: step: 1858/1541, loss: 0.8613108992576599 2023-01-23 05:47:26.502688: step: 1860/1541, loss: 2.047675132751465 2023-01-23 05:47:27.405791: step: 1862/1541, loss: 1.7994658946990967 2023-01-23 05:47:28.253717: step: 1864/1541, loss: 0.9962131977081299 2023-01-23 05:47:29.133899: step: 1866/1541, loss: 0.8398077487945557 2023-01-23 05:47:29.991806: step: 1868/1541, loss: 1.4674913883209229 2023-01-23 05:47:30.894397: step: 1870/1541, loss: 0.6214888691902161 2023-01-23 05:47:31.789527: step: 1872/1541, loss: 1.7064208984375 2023-01-23 05:47:32.670564: step: 1874/1541, loss: 1.6080268621444702 2023-01-23 05:47:33.556963: step: 1876/1541, loss: 1.532414197921753 2023-01-23 05:47:34.444698: step: 1878/1541, loss: 0.9709233045578003 2023-01-23 05:47:35.382224: step: 1880/1541, loss: 0.7757185101509094 2023-01-23 05:47:36.307425: step: 1882/1541, loss: 3.580662727355957 2023-01-23 05:47:37.216183: step: 1884/1541, loss: 0.8530250787734985 2023-01-23 05:47:38.116071: step: 1886/1541, loss: 1.1853258609771729 2023-01-23 05:47:38.993228: step: 1888/1541, loss: 3.591494560241699 2023-01-23 05:47:39.886578: step: 1890/1541, loss: 2.483961582183838 2023-01-23 05:47:40.807661: step: 1892/1541, loss: 3.643636703491211 2023-01-23 05:47:41.717080: step: 1894/1541, loss: 1.0048573017120361 2023-01-23 05:47:42.617380: step: 1896/1541, loss: 0.8018789291381836 2023-01-23 05:47:43.495198: step: 1898/1541, loss: 0.49042749404907227 2023-01-23 05:47:44.412990: step: 1900/1541, loss: 1.1174449920654297 2023-01-23 05:47:45.383247: step: 1902/1541, loss: 2.0552027225494385 2023-01-23 05:47:46.250136: step: 1904/1541, loss: 8.642248153686523 2023-01-23 05:47:47.141625: step: 1906/1541, loss: 0.850340723991394 2023-01-23 05:47:48.046311: step: 1908/1541, loss: 5.726912498474121 2023-01-23 05:47:48.940808: step: 1910/1541, loss: 7.659621715545654 2023-01-23 05:47:49.876847: step: 1912/1541, loss: 0.9433638453483582 2023-01-23 05:47:50.766032: step: 1914/1541, loss: 1.1159456968307495 2023-01-23 05:47:51.700495: step: 1916/1541, loss: 0.7652104496955872 2023-01-23 05:47:52.597108: step: 1918/1541, loss: 1.4333832263946533 2023-01-23 05:47:53.513672: step: 1920/1541, loss: 1.0600587129592896 2023-01-23 05:47:54.408040: step: 1922/1541, loss: 0.741326630115509 2023-01-23 05:47:55.263577: step: 1924/1541, loss: 0.4841635823249817 2023-01-23 05:47:56.139760: step: 1926/1541, loss: 0.7433091402053833 2023-01-23 05:47:57.003274: step: 1928/1541, loss: 0.4082028865814209 2023-01-23 05:47:57.864926: step: 1930/1541, loss: 1.302456259727478 2023-01-23 05:47:58.786645: step: 1932/1541, loss: 1.1340404748916626 2023-01-23 05:47:59.699090: step: 1934/1541, loss: 1.2239160537719727 2023-01-23 05:48:00.603740: step: 1936/1541, loss: 0.7807915210723877 2023-01-23 05:48:01.470845: step: 1938/1541, loss: 0.4954378008842468 2023-01-23 05:48:02.339013: step: 1940/1541, loss: 1.340192437171936 2023-01-23 05:48:03.230603: step: 1942/1541, loss: 0.7937080264091492 2023-01-23 05:48:04.131637: step: 1944/1541, loss: 1.2937145233154297 2023-01-23 05:48:05.012513: step: 1946/1541, loss: 0.9964765310287476 2023-01-23 05:48:05.873737: step: 1948/1541, loss: 0.9241880178451538 2023-01-23 05:48:06.782150: step: 1950/1541, loss: 1.2505896091461182 2023-01-23 05:48:07.679881: step: 1952/1541, loss: 1.1071304082870483 2023-01-23 05:48:08.561025: step: 1954/1541, loss: 0.7056686878204346 2023-01-23 05:48:09.435798: step: 1956/1541, loss: 0.6877257823944092 2023-01-23 05:48:10.342600: step: 1958/1541, loss: 1.0306971073150635 2023-01-23 05:48:11.226762: step: 1960/1541, loss: 0.5704895257949829 2023-01-23 05:48:12.084619: step: 1962/1541, loss: 3.0788040161132812 2023-01-23 05:48:12.976784: step: 1964/1541, loss: 3.3428027629852295 2023-01-23 05:48:13.868973: step: 1966/1541, loss: 1.1960439682006836 2023-01-23 05:48:14.735780: step: 1968/1541, loss: 0.7930858731269836 2023-01-23 05:48:15.628031: step: 1970/1541, loss: 0.821029782295227 2023-01-23 05:48:16.528925: step: 1972/1541, loss: 1.3927429914474487 2023-01-23 05:48:17.488693: step: 1974/1541, loss: 1.2244428396224976 2023-01-23 05:48:18.395192: step: 1976/1541, loss: 1.2754173278808594 2023-01-23 05:48:19.273741: step: 1978/1541, loss: 2.041835308074951 2023-01-23 05:48:20.167731: step: 1980/1541, loss: 7.641847133636475 2023-01-23 05:48:21.046442: step: 1982/1541, loss: 1.8437442779541016 2023-01-23 05:48:21.891235: step: 1984/1541, loss: 1.0289632081985474 2023-01-23 05:48:22.800061: step: 1986/1541, loss: 1.0766364336013794 2023-01-23 05:48:23.706956: step: 1988/1541, loss: 1.407832145690918 2023-01-23 05:48:24.621604: step: 1990/1541, loss: 0.8264808654785156 2023-01-23 05:48:25.494706: step: 1992/1541, loss: 0.6051866412162781 2023-01-23 05:48:26.370652: step: 1994/1541, loss: 10.001145362854004 2023-01-23 05:48:27.245249: step: 1996/1541, loss: 1.133946180343628 2023-01-23 05:48:28.264911: step: 1998/1541, loss: 1.041033148765564 2023-01-23 05:48:29.185625: step: 2000/1541, loss: 1.3185733556747437 2023-01-23 05:48:30.108449: step: 2002/1541, loss: 1.9865723848342896 2023-01-23 05:48:30.997036: step: 2004/1541, loss: 0.5105816125869751 2023-01-23 05:48:31.901360: step: 2006/1541, loss: 0.9503446817398071 2023-01-23 05:48:32.758654: step: 2008/1541, loss: 0.8186506628990173 2023-01-23 05:48:33.652285: step: 2010/1541, loss: 8.218330383300781 2023-01-23 05:48:34.629550: step: 2012/1541, loss: 1.0751675367355347 2023-01-23 05:48:35.564106: step: 2014/1541, loss: 1.476009488105774 2023-01-23 05:48:36.435264: step: 2016/1541, loss: 0.45391765236854553 2023-01-23 05:48:37.353821: step: 2018/1541, loss: 0.8897143006324768 2023-01-23 05:48:38.220774: step: 2020/1541, loss: 1.3989678621292114 2023-01-23 05:48:39.155130: step: 2022/1541, loss: 3.55574893951416 2023-01-23 05:48:40.056543: step: 2024/1541, loss: 7.277781963348389 2023-01-23 05:48:40.943690: step: 2026/1541, loss: 1.111338496208191 2023-01-23 05:48:41.844605: step: 2028/1541, loss: 0.954392671585083 2023-01-23 05:48:42.744427: step: 2030/1541, loss: 1.4226704835891724 2023-01-23 05:48:43.640277: step: 2032/1541, loss: 7.970160961151123 2023-01-23 05:48:44.576769: step: 2034/1541, loss: 0.7073707580566406 2023-01-23 05:48:45.518933: step: 2036/1541, loss: 0.8171622157096863 2023-01-23 05:48:46.422784: step: 2038/1541, loss: 3.310791254043579 2023-01-23 05:48:47.339198: step: 2040/1541, loss: 1.6403135061264038 2023-01-23 05:48:48.211348: step: 2042/1541, loss: 0.6574971079826355 2023-01-23 05:48:49.108879: step: 2044/1541, loss: 1.4409444332122803 2023-01-23 05:48:50.011325: step: 2046/1541, loss: 0.7131723761558533 2023-01-23 05:48:50.907364: step: 2048/1541, loss: 1.6765259504318237 2023-01-23 05:48:51.806125: step: 2050/1541, loss: 1.0203757286071777 2023-01-23 05:48:52.692704: step: 2052/1541, loss: 1.645006537437439 2023-01-23 05:48:53.582262: step: 2054/1541, loss: 3.309107780456543 2023-01-23 05:48:54.454950: step: 2056/1541, loss: 1.127404808998108 2023-01-23 05:48:55.355121: step: 2058/1541, loss: 1.244137167930603 2023-01-23 05:48:56.247511: step: 2060/1541, loss: 1.1213637590408325 2023-01-23 05:48:57.152085: step: 2062/1541, loss: 1.1153082847595215 2023-01-23 05:48:58.037878: step: 2064/1541, loss: 3.2341270446777344 2023-01-23 05:48:58.935315: step: 2066/1541, loss: 1.0664972066879272 2023-01-23 05:48:59.805729: step: 2068/1541, loss: 0.7879246473312378 2023-01-23 05:49:00.729455: step: 2070/1541, loss: 3.506587266921997 2023-01-23 05:49:01.607555: step: 2072/1541, loss: 2.3880653381347656 2023-01-23 05:49:02.526895: step: 2074/1541, loss: 1.5443556308746338 2023-01-23 05:49:03.462069: step: 2076/1541, loss: 1.0727609395980835 2023-01-23 05:49:04.325708: step: 2078/1541, loss: 1.4757349491119385 2023-01-23 05:49:05.292503: step: 2080/1541, loss: 3.711775302886963 2023-01-23 05:49:06.212983: step: 2082/1541, loss: 0.8660513758659363 2023-01-23 05:49:07.081837: step: 2084/1541, loss: 1.341848611831665 2023-01-23 05:49:08.029640: step: 2086/1541, loss: 1.9697248935699463 2023-01-23 05:49:08.962465: step: 2088/1541, loss: 0.36015474796295166 2023-01-23 05:49:09.852311: step: 2090/1541, loss: 0.5358319878578186 2023-01-23 05:49:10.726749: step: 2092/1541, loss: 1.3674163818359375 2023-01-23 05:49:11.597785: step: 2094/1541, loss: 0.5450463891029358 2023-01-23 05:49:12.506030: step: 2096/1541, loss: 1.3263216018676758 2023-01-23 05:49:13.396599: step: 2098/1541, loss: 0.3419762849807739 2023-01-23 05:49:14.294778: step: 2100/1541, loss: 0.6099209785461426 2023-01-23 05:49:15.190636: step: 2102/1541, loss: 1.0058131217956543 2023-01-23 05:49:16.094736: step: 2104/1541, loss: 3.2056543827056885 2023-01-23 05:49:16.974864: step: 2106/1541, loss: 7.072738170623779 2023-01-23 05:49:17.912933: step: 2108/1541, loss: 7.793522357940674 2023-01-23 05:49:18.803155: step: 2110/1541, loss: 1.2805925607681274 2023-01-23 05:49:19.697435: step: 2112/1541, loss: 0.6790681481361389 2023-01-23 05:49:20.612977: step: 2114/1541, loss: 1.0272376537322998 2023-01-23 05:49:21.533132: step: 2116/1541, loss: 0.7355534434318542 2023-01-23 05:49:22.388695: step: 2118/1541, loss: 6.966733455657959 2023-01-23 05:49:23.293894: step: 2120/1541, loss: 0.8822993636131287 2023-01-23 05:49:24.221469: step: 2122/1541, loss: 2.43953275680542 2023-01-23 05:49:25.109633: step: 2124/1541, loss: 0.9038236737251282 2023-01-23 05:49:26.000445: step: 2126/1541, loss: 0.6071888208389282 2023-01-23 05:49:26.999966: step: 2128/1541, loss: 2.3428077697753906 2023-01-23 05:49:28.037174: step: 2130/1541, loss: 0.7400037050247192 2023-01-23 05:49:28.962360: step: 2132/1541, loss: 3.3759143352508545 2023-01-23 05:49:29.957119: step: 2134/1541, loss: 1.029738187789917 2023-01-23 05:49:30.854561: step: 2136/1541, loss: 1.503726601600647 2023-01-23 05:49:31.739753: step: 2138/1541, loss: 6.494084358215332 2023-01-23 05:49:32.629486: step: 2140/1541, loss: 0.6076438426971436 2023-01-23 05:49:33.539466: step: 2142/1541, loss: 1.1404173374176025 2023-01-23 05:49:34.494298: step: 2144/1541, loss: 1.3232250213623047 2023-01-23 05:49:35.426850: step: 2146/1541, loss: 1.29173743724823 2023-01-23 05:49:36.356302: step: 2148/1541, loss: 0.7767961025238037 2023-01-23 05:49:37.241978: step: 2150/1541, loss: 0.9107699990272522 2023-01-23 05:49:38.140882: step: 2152/1541, loss: 2.9749560356140137 2023-01-23 05:49:39.044635: step: 2154/1541, loss: 3.5143625736236572 2023-01-23 05:49:39.909241: step: 2156/1541, loss: 0.7719957232475281 2023-01-23 05:49:40.842127: step: 2158/1541, loss: 0.7719888091087341 2023-01-23 05:49:41.722513: step: 2160/1541, loss: 3.7117466926574707 2023-01-23 05:49:42.599099: step: 2162/1541, loss: 0.9400259852409363 2023-01-23 05:49:43.559090: step: 2164/1541, loss: 0.5394126772880554 2023-01-23 05:49:44.477683: step: 2166/1541, loss: 3.3167078495025635 2023-01-23 05:49:45.387574: step: 2168/1541, loss: 3.5786094665527344 2023-01-23 05:49:46.303073: step: 2170/1541, loss: 0.7252414226531982 2023-01-23 05:49:47.211543: step: 2172/1541, loss: 0.792153000831604 2023-01-23 05:49:48.093203: step: 2174/1541, loss: 3.782865524291992 2023-01-23 05:49:48.944217: step: 2176/1541, loss: 1.2177002429962158 2023-01-23 05:49:49.852597: step: 2178/1541, loss: 0.9793555736541748 2023-01-23 05:49:50.792391: step: 2180/1541, loss: 0.6786515116691589 2023-01-23 05:49:51.736642: step: 2182/1541, loss: 1.214904546737671 2023-01-23 05:49:52.611875: step: 2184/1541, loss: 1.832457423210144 2023-01-23 05:49:53.491361: step: 2186/1541, loss: 1.5103331804275513 2023-01-23 05:49:54.385121: step: 2188/1541, loss: 0.9955049157142639 2023-01-23 05:49:55.273779: step: 2190/1541, loss: 1.5360244512557983 2023-01-23 05:49:56.170672: step: 2192/1541, loss: 1.052651286125183 2023-01-23 05:49:57.046666: step: 2194/1541, loss: 1.2041621208190918 2023-01-23 05:49:57.913972: step: 2196/1541, loss: 1.3396846055984497 2023-01-23 05:49:58.801518: step: 2198/1541, loss: 1.2132294178009033 2023-01-23 05:49:59.728009: step: 2200/1541, loss: 3.7162091732025146 2023-01-23 05:50:00.687042: step: 2202/1541, loss: 1.1964356899261475 2023-01-23 05:50:01.582122: step: 2204/1541, loss: 2.6095244884490967 2023-01-23 05:50:02.487666: step: 2206/1541, loss: 0.566669762134552 2023-01-23 05:50:03.354702: step: 2208/1541, loss: 1.4456372261047363 2023-01-23 05:50:04.312001: step: 2210/1541, loss: 0.8757597208023071 2023-01-23 05:50:05.195966: step: 2212/1541, loss: 0.9662368893623352 2023-01-23 05:50:06.081128: step: 2214/1541, loss: 2.9297900199890137 2023-01-23 05:50:06.958801: step: 2216/1541, loss: 3.801628828048706 2023-01-23 05:50:07.856404: step: 2218/1541, loss: 0.4866335988044739 2023-01-23 05:50:08.786037: step: 2220/1541, loss: 0.7142246961593628 2023-01-23 05:50:09.736401: step: 2222/1541, loss: 2.198418378829956 2023-01-23 05:50:10.688386: step: 2224/1541, loss: 3.535555124282837 2023-01-23 05:50:11.634574: step: 2226/1541, loss: 0.7807604074478149 2023-01-23 05:50:12.493015: step: 2228/1541, loss: 0.7101367712020874 2023-01-23 05:50:13.394290: step: 2230/1541, loss: 1.137657642364502 2023-01-23 05:50:14.300817: step: 2232/1541, loss: 1.8569183349609375 2023-01-23 05:50:15.170640: step: 2234/1541, loss: 0.9683172106742859 2023-01-23 05:50:16.084644: step: 2236/1541, loss: 1.2674094438552856 2023-01-23 05:50:16.975115: step: 2238/1541, loss: 0.9521849751472473 2023-01-23 05:50:17.876093: step: 2240/1541, loss: 1.0477728843688965 2023-01-23 05:50:18.769808: step: 2242/1541, loss: 0.9601643085479736 2023-01-23 05:50:19.657863: step: 2244/1541, loss: 1.0905344486236572 2023-01-23 05:50:20.545156: step: 2246/1541, loss: 0.9711612462997437 2023-01-23 05:50:21.390035: step: 2248/1541, loss: 0.7831916809082031 2023-01-23 05:50:22.301675: step: 2250/1541, loss: 1.2324614524841309 2023-01-23 05:50:23.193526: step: 2252/1541, loss: 1.205335259437561 2023-01-23 05:50:24.070269: step: 2254/1541, loss: 6.905212879180908 2023-01-23 05:50:24.973955: step: 2256/1541, loss: 1.2185367345809937 2023-01-23 05:50:25.865807: step: 2258/1541, loss: 1.029007077217102 2023-01-23 05:50:26.772575: step: 2260/1541, loss: 0.9538677930831909 2023-01-23 05:50:27.668498: step: 2262/1541, loss: 0.6260356307029724 2023-01-23 05:50:28.550769: step: 2264/1541, loss: 0.5462222695350647 2023-01-23 05:50:29.436873: step: 2266/1541, loss: 0.4899228811264038 2023-01-23 05:50:30.350724: step: 2268/1541, loss: 1.7291207313537598 2023-01-23 05:50:31.267477: step: 2270/1541, loss: 3.1561992168426514 2023-01-23 05:50:32.199146: step: 2272/1541, loss: 0.8036328554153442 2023-01-23 05:50:33.103640: step: 2274/1541, loss: 7.6643805503845215 2023-01-23 05:50:33.973489: step: 2276/1541, loss: 0.4368211030960083 2023-01-23 05:50:34.872783: step: 2278/1541, loss: 5.6685943603515625 2023-01-23 05:50:35.731692: step: 2280/1541, loss: 1.2691125869750977 2023-01-23 05:50:36.619673: step: 2282/1541, loss: 1.055359959602356 2023-01-23 05:50:37.460401: step: 2284/1541, loss: 1.0000576972961426 2023-01-23 05:50:38.384203: step: 2286/1541, loss: 1.2852064371109009 2023-01-23 05:50:39.428711: step: 2288/1541, loss: 5.763049125671387 2023-01-23 05:50:40.297484: step: 2290/1541, loss: 3.1561968326568604 2023-01-23 05:50:41.161357: step: 2292/1541, loss: 7.885923385620117 2023-01-23 05:50:42.027085: step: 2294/1541, loss: 3.6916732788085938 2023-01-23 05:50:42.981624: step: 2296/1541, loss: 0.9403122067451477 2023-01-23 05:50:43.862671: step: 2298/1541, loss: 1.5926812887191772 2023-01-23 05:50:44.776567: step: 2300/1541, loss: 11.760555267333984 2023-01-23 05:50:45.677337: step: 2302/1541, loss: 3.6168336868286133 2023-01-23 05:50:46.571344: step: 2304/1541, loss: 0.9602026343345642 2023-01-23 05:50:47.518409: step: 2306/1541, loss: 1.1944420337677002 2023-01-23 05:50:48.444174: step: 2308/1541, loss: 3.0937705039978027 2023-01-23 05:50:49.327077: step: 2310/1541, loss: 1.1159334182739258 2023-01-23 05:50:50.231805: step: 2312/1541, loss: 0.7335776686668396 2023-01-23 05:50:51.153679: step: 2314/1541, loss: 1.001627802848816 2023-01-23 05:50:52.091365: step: 2316/1541, loss: 1.1170276403427124 2023-01-23 05:50:53.022388: step: 2318/1541, loss: 1.496739149093628 2023-01-23 05:50:53.902942: step: 2320/1541, loss: 0.6170822381973267 2023-01-23 05:50:54.808950: step: 2322/1541, loss: 0.8064273595809937 2023-01-23 05:50:55.691927: step: 2324/1541, loss: 1.321675419807434 2023-01-23 05:50:56.568405: step: 2326/1541, loss: 1.4072458744049072 2023-01-23 05:50:57.430492: step: 2328/1541, loss: 1.982292890548706 2023-01-23 05:50:58.346535: step: 2330/1541, loss: 1.336192011833191 2023-01-23 05:50:59.238384: step: 2332/1541, loss: 1.0201631784439087 2023-01-23 05:51:00.203894: step: 2334/1541, loss: 0.9311142563819885 2023-01-23 05:51:01.086441: step: 2336/1541, loss: 3.1956591606140137 2023-01-23 05:51:01.991384: step: 2338/1541, loss: 1.1940898895263672 2023-01-23 05:51:02.851474: step: 2340/1541, loss: 0.5234074592590332 2023-01-23 05:51:03.739802: step: 2342/1541, loss: 1.3217651844024658 2023-01-23 05:51:04.645462: step: 2344/1541, loss: 1.0124012231826782 2023-01-23 05:51:05.516209: step: 2346/1541, loss: 3.0945444107055664 2023-01-23 05:51:06.418126: step: 2348/1541, loss: 1.1824884414672852 2023-01-23 05:51:07.385208: step: 2350/1541, loss: 3.285996198654175 2023-01-23 05:51:08.292583: step: 2352/1541, loss: 0.9567068815231323 2023-01-23 05:51:09.204243: step: 2354/1541, loss: 0.4666426479816437 2023-01-23 05:51:10.100463: step: 2356/1541, loss: 0.8866299390792847 2023-01-23 05:51:10.959589: step: 2358/1541, loss: 5.379018783569336 2023-01-23 05:51:11.851811: step: 2360/1541, loss: 2.926666736602783 2023-01-23 05:51:12.775916: step: 2362/1541, loss: 1.7926605939865112 2023-01-23 05:51:13.664254: step: 2364/1541, loss: 5.97123908996582 2023-01-23 05:51:14.544760: step: 2366/1541, loss: 1.093945026397705 2023-01-23 05:51:15.417111: step: 2368/1541, loss: 1.4239124059677124 2023-01-23 05:51:16.301262: step: 2370/1541, loss: 0.7410780787467957 2023-01-23 05:51:17.249773: step: 2372/1541, loss: 1.0420570373535156 2023-01-23 05:51:18.105283: step: 2374/1541, loss: 0.9316134452819824 2023-01-23 05:51:18.962645: step: 2376/1541, loss: 1.0389883518218994 2023-01-23 05:51:19.826071: step: 2378/1541, loss: 0.622592568397522 2023-01-23 05:51:20.708358: step: 2380/1541, loss: 0.9616953730583191 2023-01-23 05:51:21.576709: step: 2382/1541, loss: 0.9854007363319397 2023-01-23 05:51:22.472633: step: 2384/1541, loss: 0.5515356063842773 2023-01-23 05:51:23.408946: step: 2386/1541, loss: 1.0921467542648315 2023-01-23 05:51:24.285268: step: 2388/1541, loss: 0.9806354641914368 2023-01-23 05:51:25.220618: step: 2390/1541, loss: 1.3406041860580444 2023-01-23 05:51:26.124718: step: 2392/1541, loss: 1.1549359560012817 2023-01-23 05:51:27.051157: step: 2394/1541, loss: 1.6924018859863281 2023-01-23 05:51:27.969607: step: 2396/1541, loss: 0.8078910112380981 2023-01-23 05:51:28.870790: step: 2398/1541, loss: 0.6076231598854065 2023-01-23 05:51:29.813280: step: 2400/1541, loss: 1.5652137994766235 2023-01-23 05:51:30.751910: step: 2402/1541, loss: 4.078782081604004 2023-01-23 05:51:31.684003: step: 2404/1541, loss: 1.564069151878357 2023-01-23 05:51:32.629058: step: 2406/1541, loss: 1.2758206129074097 2023-01-23 05:51:33.546215: step: 2408/1541, loss: 0.8143766522407532 2023-01-23 05:51:34.386845: step: 2410/1541, loss: 1.0385676622390747 2023-01-23 05:51:35.288754: step: 2412/1541, loss: 0.7673496603965759 2023-01-23 05:51:36.170022: step: 2414/1541, loss: 1.7792487144470215 2023-01-23 05:51:37.055189: step: 2416/1541, loss: 9.656599044799805 2023-01-23 05:51:37.916987: step: 2418/1541, loss: 0.9536055326461792 2023-01-23 05:51:38.872210: step: 2420/1541, loss: 0.32052886486053467 2023-01-23 05:51:39.787350: step: 2422/1541, loss: 1.6169437170028687 2023-01-23 05:51:40.699977: step: 2424/1541, loss: 3.657209634780884 2023-01-23 05:51:41.562773: step: 2426/1541, loss: 3.115934133529663 2023-01-23 05:51:42.427491: step: 2428/1541, loss: 0.9971089959144592 2023-01-23 05:51:43.342434: step: 2430/1541, loss: 3.324580192565918 2023-01-23 05:51:44.260135: step: 2432/1541, loss: 0.9571393728256226 2023-01-23 05:51:45.154387: step: 2434/1541, loss: 3.351165294647217 2023-01-23 05:51:46.069776: step: 2436/1541, loss: 0.8674553036689758 2023-01-23 05:51:47.040175: step: 2438/1541, loss: 1.3168283700942993 2023-01-23 05:51:47.966053: step: 2440/1541, loss: 1.5864427089691162 2023-01-23 05:51:48.870329: step: 2442/1541, loss: 2.2269933223724365 2023-01-23 05:51:49.724301: step: 2444/1541, loss: 3.395852565765381 2023-01-23 05:51:50.579755: step: 2446/1541, loss: 3.412198066711426 2023-01-23 05:51:51.488094: step: 2448/1541, loss: 0.5948735475540161 2023-01-23 05:51:52.435331: step: 2450/1541, loss: 1.5271275043487549 2023-01-23 05:51:53.313699: step: 2452/1541, loss: 1.1373271942138672 2023-01-23 05:51:54.218209: step: 2454/1541, loss: 0.5719513297080994 2023-01-23 05:51:55.108165: step: 2456/1541, loss: 0.7726990580558777 2023-01-23 05:51:56.109949: step: 2458/1541, loss: 1.7209863662719727 2023-01-23 05:51:57.068927: step: 2460/1541, loss: 0.8046477437019348 2023-01-23 05:51:57.950258: step: 2462/1541, loss: 0.5500941276550293 2023-01-23 05:51:58.810037: step: 2464/1541, loss: 1.2812957763671875 2023-01-23 05:51:59.632570: step: 2466/1541, loss: 2.964468479156494 2023-01-23 05:52:00.517346: step: 2468/1541, loss: 1.2177072763442993 2023-01-23 05:52:01.425120: step: 2470/1541, loss: 0.6526669859886169 2023-01-23 05:52:02.364227: step: 2472/1541, loss: 0.7909387946128845 2023-01-23 05:52:03.260217: step: 2474/1541, loss: 1.821987509727478 2023-01-23 05:52:04.206293: step: 2476/1541, loss: 0.5480170249938965 2023-01-23 05:52:05.172944: step: 2478/1541, loss: 0.8409291505813599 2023-01-23 05:52:06.077618: step: 2480/1541, loss: 1.017047643661499 2023-01-23 05:52:07.003491: step: 2482/1541, loss: 1.8949432373046875 2023-01-23 05:52:07.877345: step: 2484/1541, loss: 9.894153594970703 2023-01-23 05:52:08.780983: step: 2486/1541, loss: 0.6824736595153809 2023-01-23 05:52:09.652769: step: 2488/1541, loss: 3.1362175941467285 2023-01-23 05:52:10.523274: step: 2490/1541, loss: 1.5228359699249268 2023-01-23 05:52:11.439005: step: 2492/1541, loss: 1.468172311782837 2023-01-23 05:52:12.334621: step: 2494/1541, loss: 6.756662368774414 2023-01-23 05:52:13.228373: step: 2496/1541, loss: 1.1496511697769165 2023-01-23 05:52:14.141150: step: 2498/1541, loss: 0.5223819017410278 2023-01-23 05:52:15.060108: step: 2500/1541, loss: 1.449004888534546 2023-01-23 05:52:15.954084: step: 2502/1541, loss: 6.9199652671813965 2023-01-23 05:52:16.895584: step: 2504/1541, loss: 1.670634150505066 2023-01-23 05:52:17.806429: step: 2506/1541, loss: 0.5944319367408752 2023-01-23 05:52:18.682964: step: 2508/1541, loss: 0.9307965040206909 2023-01-23 05:52:19.608332: step: 2510/1541, loss: 1.6904001235961914 2023-01-23 05:52:20.499021: step: 2512/1541, loss: 1.0307583808898926 2023-01-23 05:52:21.429997: step: 2514/1541, loss: 1.6685043573379517 2023-01-23 05:52:22.332955: step: 2516/1541, loss: 1.955463171005249 2023-01-23 05:52:23.219483: step: 2518/1541, loss: 1.4658690690994263 2023-01-23 05:52:24.107138: step: 2520/1541, loss: 2.878953695297241 2023-01-23 05:52:25.017647: step: 2522/1541, loss: 0.9503520727157593 2023-01-23 05:52:25.920948: step: 2524/1541, loss: 0.9976903796195984 2023-01-23 05:52:26.818697: step: 2526/1541, loss: 0.9021797180175781 2023-01-23 05:52:27.713248: step: 2528/1541, loss: 1.1462548971176147 2023-01-23 05:52:28.599321: step: 2530/1541, loss: 2.629164695739746 2023-01-23 05:52:29.500818: step: 2532/1541, loss: 0.6515032052993774 2023-01-23 05:52:30.384338: step: 2534/1541, loss: 0.42563655972480774 2023-01-23 05:52:31.259930: step: 2536/1541, loss: 0.7744763493537903 2023-01-23 05:52:32.198039: step: 2538/1541, loss: 0.8021766543388367 2023-01-23 05:52:33.126681: step: 2540/1541, loss: 1.9946657419204712 2023-01-23 05:52:33.995615: step: 2542/1541, loss: 0.3912610411643982 2023-01-23 05:52:34.959443: step: 2544/1541, loss: 0.6284394860267639 2023-01-23 05:52:35.836450: step: 2546/1541, loss: 0.5373064875602722 2023-01-23 05:52:36.755053: step: 2548/1541, loss: 3.0081772804260254 2023-01-23 05:52:37.650964: step: 2550/1541, loss: 0.8419257998466492 2023-01-23 05:52:38.535979: step: 2552/1541, loss: 1.5725195407867432 2023-01-23 05:52:39.410551: step: 2554/1541, loss: 3.967608690261841 2023-01-23 05:52:40.334577: step: 2556/1541, loss: 1.0345458984375 2023-01-23 05:52:41.212327: step: 2558/1541, loss: 1.0793428421020508 2023-01-23 05:52:42.103936: step: 2560/1541, loss: 7.218868732452393 2023-01-23 05:52:43.023940: step: 2562/1541, loss: 1.705614447593689 2023-01-23 05:52:43.906093: step: 2564/1541, loss: 1.254467248916626 2023-01-23 05:52:44.866711: step: 2566/1541, loss: 0.763964831829071 2023-01-23 05:52:45.745085: step: 2568/1541, loss: 1.0359454154968262 2023-01-23 05:52:46.582836: step: 2570/1541, loss: 0.9973042011260986 2023-01-23 05:52:47.484881: step: 2572/1541, loss: 0.6900523900985718 2023-01-23 05:52:48.392338: step: 2574/1541, loss: 3.6191301345825195 2023-01-23 05:52:49.318045: step: 2576/1541, loss: 3.040308952331543 2023-01-23 05:52:50.233273: step: 2578/1541, loss: 4.334196090698242 2023-01-23 05:52:51.129167: step: 2580/1541, loss: 1.1059584617614746 2023-01-23 05:52:52.013019: step: 2582/1541, loss: 2.0937790870666504 2023-01-23 05:52:52.898690: step: 2584/1541, loss: 1.3143349885940552 2023-01-23 05:52:53.802987: step: 2586/1541, loss: 1.0947142839431763 2023-01-23 05:52:54.753277: step: 2588/1541, loss: 1.9500670433044434 2023-01-23 05:52:55.649157: step: 2590/1541, loss: 0.9860620498657227 2023-01-23 05:52:56.521538: step: 2592/1541, loss: 1.0785123109817505 2023-01-23 05:52:57.379853: step: 2594/1541, loss: 3.6604971885681152 2023-01-23 05:52:58.221063: step: 2596/1541, loss: 1.4685149192810059 2023-01-23 05:52:59.154706: step: 2598/1541, loss: 0.591059148311615 2023-01-23 05:53:00.013658: step: 2600/1541, loss: 1.6717040538787842 2023-01-23 05:53:00.898065: step: 2602/1541, loss: 1.2886375188827515 2023-01-23 05:53:01.812533: step: 2604/1541, loss: 3.6002562046051025 2023-01-23 05:53:02.697371: step: 2606/1541, loss: 0.620428740978241 2023-01-23 05:53:03.602057: step: 2608/1541, loss: 1.3031609058380127 2023-01-23 05:53:04.469206: step: 2610/1541, loss: 0.6476550698280334 2023-01-23 05:53:05.337738: step: 2612/1541, loss: 0.9179323315620422 2023-01-23 05:53:06.219019: step: 2614/1541, loss: 1.1353003978729248 2023-01-23 05:53:07.097826: step: 2616/1541, loss: 1.674009919166565 2023-01-23 05:53:07.997196: step: 2618/1541, loss: 3.122767448425293 2023-01-23 05:53:08.884189: step: 2620/1541, loss: 1.0489646196365356 2023-01-23 05:53:09.772327: step: 2622/1541, loss: 0.9177893400192261 2023-01-23 05:53:10.651934: step: 2624/1541, loss: 0.4740184247493744 2023-01-23 05:53:11.550025: step: 2626/1541, loss: 1.7753924131393433 2023-01-23 05:53:12.472177: step: 2628/1541, loss: 1.7715998888015747 2023-01-23 05:53:13.368327: step: 2630/1541, loss: 3.911055564880371 2023-01-23 05:53:14.262740: step: 2632/1541, loss: 0.8775210976600647 2023-01-23 05:53:15.142680: step: 2634/1541, loss: 0.8247302770614624 2023-01-23 05:53:16.121501: step: 2636/1541, loss: 1.387371301651001 2023-01-23 05:53:17.019012: step: 2638/1541, loss: 3.7697019577026367 2023-01-23 05:53:18.044530: step: 2640/1541, loss: 1.7951878309249878 2023-01-23 05:53:18.986676: step: 2642/1541, loss: 0.5283887982368469 2023-01-23 05:53:19.888427: step: 2644/1541, loss: 1.4480810165405273 2023-01-23 05:53:20.792666: step: 2646/1541, loss: 0.6134321093559265 2023-01-23 05:53:21.700682: step: 2648/1541, loss: 4.698853015899658 2023-01-23 05:53:22.621178: step: 2650/1541, loss: 0.5680108070373535 2023-01-23 05:53:23.510445: step: 2652/1541, loss: 3.1801557540893555 2023-01-23 05:53:24.403636: step: 2654/1541, loss: 0.4215753376483917 2023-01-23 05:53:25.331856: step: 2656/1541, loss: 0.6948131918907166 2023-01-23 05:53:26.205756: step: 2658/1541, loss: 0.8548398613929749 2023-01-23 05:53:27.089915: step: 2660/1541, loss: 1.6667693853378296 2023-01-23 05:53:27.949026: step: 2662/1541, loss: 0.8200783133506775 2023-01-23 05:53:28.847700: step: 2664/1541, loss: 2.587698459625244 2023-01-23 05:53:29.720596: step: 2666/1541, loss: 0.5590710043907166 2023-01-23 05:53:30.637011: step: 2668/1541, loss: 0.8013485670089722 2023-01-23 05:53:31.523741: step: 2670/1541, loss: 0.4861069917678833 2023-01-23 05:53:32.454439: step: 2672/1541, loss: 1.4778811931610107 2023-01-23 05:53:33.372669: step: 2674/1541, loss: 4.088742256164551 2023-01-23 05:53:34.239436: step: 2676/1541, loss: 0.4219130277633667 2023-01-23 05:53:35.158488: step: 2678/1541, loss: 0.6566819548606873 2023-01-23 05:53:36.047281: step: 2680/1541, loss: 2.3819613456726074 2023-01-23 05:53:36.925305: step: 2682/1541, loss: 0.5947490930557251 2023-01-23 05:53:37.794740: step: 2684/1541, loss: 1.0137662887573242 2023-01-23 05:53:38.789928: step: 2686/1541, loss: 1.7063837051391602 2023-01-23 05:53:39.691254: step: 2688/1541, loss: 0.9401820302009583 2023-01-23 05:53:40.629428: step: 2690/1541, loss: 1.3533486127853394 2023-01-23 05:53:41.536984: step: 2692/1541, loss: 0.7116901278495789 2023-01-23 05:53:42.517969: step: 2694/1541, loss: 1.4601348638534546 2023-01-23 05:53:43.397145: step: 2696/1541, loss: 3.260981559753418 2023-01-23 05:53:44.371557: step: 2698/1541, loss: 0.6775116324424744 2023-01-23 05:53:45.288679: step: 2700/1541, loss: 0.6575390100479126 2023-01-23 05:53:46.189724: step: 2702/1541, loss: 1.9802268743515015 2023-01-23 05:53:47.095829: step: 2704/1541, loss: 1.8617838621139526 2023-01-23 05:53:48.059508: step: 2706/1541, loss: 1.2904982566833496 2023-01-23 05:53:48.963235: step: 2708/1541, loss: 1.183772087097168 2023-01-23 05:53:49.823418: step: 2710/1541, loss: 1.9653394222259521 2023-01-23 05:53:50.695923: step: 2712/1541, loss: 0.529231607913971 2023-01-23 05:53:51.594030: step: 2714/1541, loss: 0.37133723497390747 2023-01-23 05:53:52.494718: step: 2716/1541, loss: 8.1231107711792 2023-01-23 05:53:53.427993: step: 2718/1541, loss: 1.0324889421463013 2023-01-23 05:53:54.348879: step: 2720/1541, loss: 1.0208712816238403 2023-01-23 05:53:55.247907: step: 2722/1541, loss: 0.5865964889526367 2023-01-23 05:53:56.130572: step: 2724/1541, loss: 1.1299488544464111 2023-01-23 05:53:57.058553: step: 2726/1541, loss: 1.6602710485458374 2023-01-23 05:53:58.008356: step: 2728/1541, loss: 1.37510347366333 2023-01-23 05:53:58.922253: step: 2730/1541, loss: 1.1917293071746826 2023-01-23 05:53:59.822142: step: 2732/1541, loss: 1.2649472951889038 2023-01-23 05:54:00.735644: step: 2734/1541, loss: 0.9712379574775696 2023-01-23 05:54:01.606332: step: 2736/1541, loss: 0.6598154902458191 2023-01-23 05:54:02.595808: step: 2738/1541, loss: 1.3655952215194702 2023-01-23 05:54:03.485297: step: 2740/1541, loss: 1.3952833414077759 2023-01-23 05:54:04.353693: step: 2742/1541, loss: 1.3816155195236206 2023-01-23 05:54:05.233583: step: 2744/1541, loss: 1.0463604927062988 2023-01-23 05:54:06.130541: step: 2746/1541, loss: 1.0053391456604004 2023-01-23 05:54:07.058522: step: 2748/1541, loss: 0.8339454531669617 2023-01-23 05:54:07.965827: step: 2750/1541, loss: 1.1431487798690796 2023-01-23 05:54:08.907241: step: 2752/1541, loss: 1.2655584812164307 2023-01-23 05:54:09.867537: step: 2754/1541, loss: 1.386808156967163 2023-01-23 05:54:10.754191: step: 2756/1541, loss: 0.8235559463500977 2023-01-23 05:54:11.611593: step: 2758/1541, loss: 1.7699248790740967 2023-01-23 05:54:12.535047: step: 2760/1541, loss: 1.2542880773544312 2023-01-23 05:54:13.403517: step: 2762/1541, loss: 3.013338327407837 2023-01-23 05:54:14.308998: step: 2764/1541, loss: 1.1692737340927124 2023-01-23 05:54:15.210222: step: 2766/1541, loss: 0.9697920083999634 2023-01-23 05:54:16.167657: step: 2768/1541, loss: 1.0608932971954346 2023-01-23 05:54:17.077519: step: 2770/1541, loss: 0.6283561587333679 2023-01-23 05:54:17.932637: step: 2772/1541, loss: 1.2860479354858398 2023-01-23 05:54:18.834625: step: 2774/1541, loss: 1.2369771003723145 2023-01-23 05:54:19.789904: step: 2776/1541, loss: 0.5986993908882141 2023-01-23 05:54:20.704234: step: 2778/1541, loss: 1.4107699394226074 2023-01-23 05:54:21.604903: step: 2780/1541, loss: 1.8674606084823608 2023-01-23 05:54:22.518950: step: 2782/1541, loss: 1.8065152168273926 2023-01-23 05:54:23.460049: step: 2784/1541, loss: 0.2866623103618622 2023-01-23 05:54:24.356407: step: 2786/1541, loss: 2.886091470718384 2023-01-23 05:54:25.222313: step: 2788/1541, loss: 1.4707976579666138 2023-01-23 05:54:26.144127: step: 2790/1541, loss: 1.3614892959594727 2023-01-23 05:54:27.048516: step: 2792/1541, loss: 3.1028366088867188 2023-01-23 05:54:27.915119: step: 2794/1541, loss: 0.7401425838470459 2023-01-23 05:54:28.786372: step: 2796/1541, loss: 0.6173005104064941 2023-01-23 05:54:29.648511: step: 2798/1541, loss: 0.5956574082374573 2023-01-23 05:54:30.511535: step: 2800/1541, loss: 0.5095840692520142 2023-01-23 05:54:31.407881: step: 2802/1541, loss: 3.1303298473358154 2023-01-23 05:54:32.343516: step: 2804/1541, loss: 1.2293022871017456 2023-01-23 05:54:33.207420: step: 2806/1541, loss: 4.184839725494385 2023-01-23 05:54:34.073562: step: 2808/1541, loss: 0.6239885091781616 2023-01-23 05:54:34.980178: step: 2810/1541, loss: 1.0356664657592773 2023-01-23 05:54:35.885682: step: 2812/1541, loss: 0.639864981174469 2023-01-23 05:54:36.856907: step: 2814/1541, loss: 0.5730032920837402 2023-01-23 05:54:37.717328: step: 2816/1541, loss: 0.4939374625682831 2023-01-23 05:54:38.617441: step: 2818/1541, loss: 0.7995100617408752 2023-01-23 05:54:39.488889: step: 2820/1541, loss: 2.459681272506714 2023-01-23 05:54:40.437662: step: 2822/1541, loss: 0.6026960015296936 2023-01-23 05:54:41.340653: step: 2824/1541, loss: 9.450944900512695 2023-01-23 05:54:42.264652: step: 2826/1541, loss: 1.2911726236343384 2023-01-23 05:54:43.122631: step: 2828/1541, loss: 1.3795766830444336 2023-01-23 05:54:44.009414: step: 2830/1541, loss: 1.0361974239349365 2023-01-23 05:54:44.917306: step: 2832/1541, loss: 0.8555634617805481 2023-01-23 05:54:45.804589: step: 2834/1541, loss: 0.5535539388656616 2023-01-23 05:54:46.699378: step: 2836/1541, loss: 0.6048506498336792 2023-01-23 05:54:47.558724: step: 2838/1541, loss: 0.7358758449554443 2023-01-23 05:54:48.483332: step: 2840/1541, loss: 4.1723480224609375 2023-01-23 05:54:49.445190: step: 2842/1541, loss: 1.4279345273971558 2023-01-23 05:54:50.415299: step: 2844/1541, loss: 0.8967524766921997 2023-01-23 05:54:51.315640: step: 2846/1541, loss: 1.2867815494537354 2023-01-23 05:54:52.279839: step: 2848/1541, loss: 0.8487733006477356 2023-01-23 05:54:53.139464: step: 2850/1541, loss: 0.617648184299469 2023-01-23 05:54:54.016665: step: 2852/1541, loss: 1.057720422744751 2023-01-23 05:54:54.907752: step: 2854/1541, loss: 4.48367977142334 2023-01-23 05:54:55.832451: step: 2856/1541, loss: 1.8307714462280273 2023-01-23 05:54:56.715430: step: 2858/1541, loss: 0.5416795611381531 2023-01-23 05:54:57.613528: step: 2860/1541, loss: 1.6180833578109741 2023-01-23 05:54:58.524606: step: 2862/1541, loss: 3.0260753631591797 2023-01-23 05:54:59.435905: step: 2864/1541, loss: 0.9470680356025696 2023-01-23 05:55:00.326495: step: 2866/1541, loss: 0.7081654667854309 2023-01-23 05:55:01.226360: step: 2868/1541, loss: 1.1401175260543823 2023-01-23 05:55:02.211646: step: 2870/1541, loss: 1.9309234619140625 2023-01-23 05:55:03.097914: step: 2872/1541, loss: 1.6955448389053345 2023-01-23 05:55:03.970375: step: 2874/1541, loss: 0.49176186323165894 2023-01-23 05:55:04.863694: step: 2876/1541, loss: 1.2024197578430176 2023-01-23 05:55:05.726137: step: 2878/1541, loss: 4.228167533874512 2023-01-23 05:55:06.608630: step: 2880/1541, loss: 0.8027759194374084 2023-01-23 05:55:07.505460: step: 2882/1541, loss: 1.192202091217041 2023-01-23 05:55:08.402766: step: 2884/1541, loss: 2.7816412448883057 2023-01-23 05:55:09.302551: step: 2886/1541, loss: 0.49498671293258667 2023-01-23 05:55:10.236340: step: 2888/1541, loss: 0.7390719056129456 2023-01-23 05:55:11.122039: step: 2890/1541, loss: 1.0062644481658936 2023-01-23 05:55:12.045370: step: 2892/1541, loss: 1.4403761625289917 2023-01-23 05:55:12.981085: step: 2894/1541, loss: 1.013120412826538 2023-01-23 05:55:13.863271: step: 2896/1541, loss: 0.7099908590316772 2023-01-23 05:55:14.809543: step: 2898/1541, loss: 1.060322880744934 2023-01-23 05:55:15.682527: step: 2900/1541, loss: 1.252247929573059 2023-01-23 05:55:16.557163: step: 2902/1541, loss: 0.7433171272277832 2023-01-23 05:55:17.457342: step: 2904/1541, loss: 0.7981131076812744 2023-01-23 05:55:18.357086: step: 2906/1541, loss: 3.155137777328491 2023-01-23 05:55:19.255494: step: 2908/1541, loss: 1.677430272102356 2023-01-23 05:55:20.094734: step: 2910/1541, loss: 0.9523622393608093 2023-01-23 05:55:20.977766: step: 2912/1541, loss: 0.6009687185287476 2023-01-23 05:55:21.865058: step: 2914/1541, loss: 1.0872831344604492 2023-01-23 05:55:22.726646: step: 2916/1541, loss: 1.213489055633545 2023-01-23 05:55:23.692634: step: 2918/1541, loss: 0.5860543251037598 2023-01-23 05:55:24.578409: step: 2920/1541, loss: 0.7795441746711731 2023-01-23 05:55:25.457615: step: 2922/1541, loss: 1.0175490379333496 2023-01-23 05:55:26.374949: step: 2924/1541, loss: 3.5811736583709717 2023-01-23 05:55:27.220606: step: 2926/1541, loss: 3.561856508255005 2023-01-23 05:55:28.103833: step: 2928/1541, loss: 1.1721906661987305 2023-01-23 05:55:28.976071: step: 2930/1541, loss: 0.5314958691596985 2023-01-23 05:55:29.848513: step: 2932/1541, loss: 6.452524185180664 2023-01-23 05:55:30.788938: step: 2934/1541, loss: 2.204942226409912 2023-01-23 05:55:31.667959: step: 2936/1541, loss: 0.6651293635368347 2023-01-23 05:55:32.640080: step: 2938/1541, loss: 3.8215198516845703 2023-01-23 05:55:33.491384: step: 2940/1541, loss: 1.2380255460739136 2023-01-23 05:55:34.422624: step: 2942/1541, loss: 1.106656789779663 2023-01-23 05:55:35.351601: step: 2944/1541, loss: 1.9135332107543945 2023-01-23 05:55:36.215551: step: 2946/1541, loss: 0.7205308675765991 2023-01-23 05:55:37.134528: step: 2948/1541, loss: 0.3448394536972046 2023-01-23 05:55:38.002898: step: 2950/1541, loss: 7.505201816558838 2023-01-23 05:55:38.893721: step: 2952/1541, loss: 1.2060288190841675 2023-01-23 05:55:39.812814: step: 2954/1541, loss: 1.3034595251083374 2023-01-23 05:55:40.764054: step: 2956/1541, loss: 7.72320032119751 2023-01-23 05:55:41.663939: step: 2958/1541, loss: 0.8546711802482605 2023-01-23 05:55:42.527122: step: 2960/1541, loss: 0.7932264804840088 2023-01-23 05:55:43.421717: step: 2962/1541, loss: 1.524991512298584 2023-01-23 05:55:44.349472: step: 2964/1541, loss: 1.7368240356445312 2023-01-23 05:55:45.201154: step: 2966/1541, loss: 1.0784385204315186 2023-01-23 05:55:46.066759: step: 2968/1541, loss: 1.0289911031723022 2023-01-23 05:55:46.928710: step: 2970/1541, loss: 0.8647348880767822 2023-01-23 05:55:47.875715: step: 2972/1541, loss: 1.6700230836868286 2023-01-23 05:55:48.772608: step: 2974/1541, loss: 1.9552762508392334 2023-01-23 05:55:49.634003: step: 2976/1541, loss: 1.5472861528396606 2023-01-23 05:55:50.493571: step: 2978/1541, loss: 0.889133632183075 2023-01-23 05:55:51.399551: step: 2980/1541, loss: 0.6971845030784607 2023-01-23 05:55:52.319830: step: 2982/1541, loss: 1.5470569133758545 2023-01-23 05:55:53.226550: step: 2984/1541, loss: 1.2500483989715576 2023-01-23 05:55:54.094364: step: 2986/1541, loss: 0.8706246018409729 2023-01-23 05:55:54.979308: step: 2988/1541, loss: 0.8173624873161316 2023-01-23 05:55:55.859108: step: 2990/1541, loss: 0.5378542542457581 2023-01-23 05:55:56.774429: step: 2992/1541, loss: 0.6644731163978577 2023-01-23 05:55:57.703788: step: 2994/1541, loss: 0.7617189884185791 2023-01-23 05:55:58.589489: step: 2996/1541, loss: 0.976319432258606 2023-01-23 05:55:59.477818: step: 2998/1541, loss: 1.0191468000411987 2023-01-23 05:56:00.398171: step: 3000/1541, loss: 3.972334623336792 2023-01-23 05:56:01.301094: step: 3002/1541, loss: 3.2437448501586914 2023-01-23 05:56:02.233623: step: 3004/1541, loss: 7.640218734741211 2023-01-23 05:56:03.180945: step: 3006/1541, loss: 0.6462056040763855 2023-01-23 05:56:04.057208: step: 3008/1541, loss: 1.488471508026123 2023-01-23 05:56:04.955997: step: 3010/1541, loss: 0.6738666892051697 2023-01-23 05:56:05.852926: step: 3012/1541, loss: 0.9970131516456604 2023-01-23 05:56:06.734360: step: 3014/1541, loss: 1.5537387132644653 2023-01-23 05:56:07.627228: step: 3016/1541, loss: 0.7742883563041687 2023-01-23 05:56:08.575900: step: 3018/1541, loss: 1.333538293838501 2023-01-23 05:56:09.568312: step: 3020/1541, loss: 0.9483049511909485 2023-01-23 05:56:10.424015: step: 3022/1541, loss: 0.944429874420166 2023-01-23 05:56:11.365062: step: 3024/1541, loss: 1.939847707748413 2023-01-23 05:56:12.249335: step: 3026/1541, loss: 1.4853533506393433 2023-01-23 05:56:13.172414: step: 3028/1541, loss: 4.701409816741943 2023-01-23 05:56:14.067150: step: 3030/1541, loss: 6.606198310852051 2023-01-23 05:56:15.091561: step: 3032/1541, loss: 0.9353384971618652 2023-01-23 05:56:15.990277: step: 3034/1541, loss: 0.6245204210281372 2023-01-23 05:56:16.868140: step: 3036/1541, loss: 1.4014081954956055 2023-01-23 05:56:17.756155: step: 3038/1541, loss: 1.6824719905853271 2023-01-23 05:56:18.624119: step: 3040/1541, loss: 1.5566657781600952 2023-01-23 05:56:19.521837: step: 3042/1541, loss: 1.3054996728897095 2023-01-23 05:56:20.385815: step: 3044/1541, loss: 0.7400035858154297 2023-01-23 05:56:21.272482: step: 3046/1541, loss: 0.7538264989852905 2023-01-23 05:56:22.187916: step: 3048/1541, loss: 6.938909530639648 2023-01-23 05:56:23.057499: step: 3050/1541, loss: 1.5990957021713257 2023-01-23 05:56:23.914388: step: 3052/1541, loss: 0.3570311963558197 2023-01-23 05:56:24.815642: step: 3054/1541, loss: 1.3065482378005981 2023-01-23 05:56:25.712099: step: 3056/1541, loss: 3.3225841522216797 2023-01-23 05:56:26.626224: step: 3058/1541, loss: 1.0684494972229004 2023-01-23 05:56:27.517301: step: 3060/1541, loss: 1.5028562545776367 2023-01-23 05:56:28.411597: step: 3062/1541, loss: 1.3359549045562744 2023-01-23 05:56:29.310331: step: 3064/1541, loss: 0.9455095529556274 2023-01-23 05:56:30.223225: step: 3066/1541, loss: 3.82375168800354 2023-01-23 05:56:31.123245: step: 3068/1541, loss: 1.165527582168579 2023-01-23 05:56:31.975693: step: 3070/1541, loss: 1.422320008277893 2023-01-23 05:56:32.844400: step: 3072/1541, loss: 0.7161415815353394 2023-01-23 05:56:33.776877: step: 3074/1541, loss: 0.8383992910385132 2023-01-23 05:56:34.641371: step: 3076/1541, loss: 0.8483666181564331 2023-01-23 05:56:35.588687: step: 3078/1541, loss: 0.912675142288208 2023-01-23 05:56:36.497259: step: 3080/1541, loss: 1.1674151420593262 2023-01-23 05:56:37.402098: step: 3082/1541, loss: 1.6990808248519897 2023-01-23 05:56:38.060717: step: 3084/1541, loss: 0.5273277163505554 ================================================== Loss: 1.719 -------------------- Dev Chinese: {'event': {'p': 0.6367980884109916, 'r': 0.7097203728362184, 'f1': 0.6712846347607053}, 'argument': {'p': 0.6729578904388359, 'r': 0.5990583090121413, 'f1': 0.6338614493379393}, 'combined': 0.4255014515077099, 'epoch': 1} Test Chinese: {'event': {'p': 0.6151419558359621, 'r': 0.701018573996405, 'f1': 0.6552786334360123}, 'argument': {'p': 0.7112215570847993, 'r': 0.6529247081434223, 'f1': 0.6808274734486968}, 'combined': 0.446131696407155, 'epoch': 1} Dev Korean: {'event': {'p': 0.5794117647058824, 'r': 0.7869507323568575, 'f1': 0.6674195369847544}, 'argument': {'p': 0.6709033183918488, 'r': 0.5991927054503958, 'f1': 0.633023593329179}, 'combined': 0.42249231356018613, 'epoch': 1} Test Korean: {'event': {'p': 0.6101048617731173, 'r': 0.7669263031755542, 'f1': 0.6795858773559862}, 'argument': {'p': 0.7136545672623013, 'r': 0.6632701036094307, 'f1': 0.6875405006284541}, 'combined': 0.4672428143373619, 'epoch': 1} Dev Russian: {'event': {'p': 0.659952606635071, 'r': 0.7416777629826897, 'f1': 0.6984326018808777}, 'argument': {'p': 0.6741587330290469, 'r': 0.5985758502439995, 'f1': 0.6341230011751896}, 'combined': 0.4428921776232986, 'epoch': 1} Test Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7127746635632035, 'r': 0.6583194973727579, 'f1': 0.684465694079899}, 'combined': 0.476831687440019, 'epoch': 1} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6567493112947658, 'r': 0.5753821399839099, 'f1': 0.6133790737564322}, 'combined': 0.4403747196200026, 'epoch': 1} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.45908936301793446, 'r': 0.2171368608868609, 'f1': 0.2948280312959212}, 'combined': 0.1810754837880461, 'epoch': 1} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.4982400932400932, 'r': 0.5793489456280154, 'f1': 0.5357420357420357}, 'combined': 0.2602175602175602, 'epoch': 1} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6367980884109916, 'r': 0.7097203728362184, 'f1': 0.6712846347607053}, 'argument': {'p': 0.6729578904388359, 'r': 0.5990583090121413, 'f1': 0.6338614493379393}, 'combined': 0.4255014515077099, 'epoch': 1} Test for Chinese: {'event': {'p': 0.6151419558359621, 'r': 0.701018573996405, 'f1': 0.6552786334360123}, 'argument': {'p': 0.7112215570847993, 'r': 0.6529247081434223, 'f1': 0.6808274734486968}, 'combined': 0.446131696407155, 'epoch': 1} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6567493112947658, 'r': 0.5753821399839099, 'f1': 0.6133790737564322}, 'combined': 0.4403747196200026, 'epoch': 1} -------------------- Dev for Korean: {'event': {'p': 0.5794117647058824, 'r': 0.7869507323568575, 'f1': 0.6674195369847544}, 'argument': {'p': 0.6709033183918488, 'r': 0.5991927054503958, 'f1': 0.633023593329179}, 'combined': 0.42249231356018613, 'epoch': 1} Test for Korean: {'event': {'p': 0.6101048617731173, 'r': 0.7669263031755542, 'f1': 0.6795858773559862}, 'argument': {'p': 0.7136545672623013, 'r': 0.6632701036094307, 'f1': 0.6875405006284541}, 'combined': 0.4672428143373619, 'epoch': 1} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.45908936301793446, 'r': 0.2171368608868609, 'f1': 0.2948280312959212}, 'combined': 0.1810754837880461, 'epoch': 1} -------------------- Dev for Russian: {'event': {'p': 0.659952606635071, 'r': 0.7416777629826897, 'f1': 0.6984326018808777}, 'argument': {'p': 0.6741587330290469, 'r': 0.5985758502439995, 'f1': 0.6341230011751896}, 'combined': 0.4428921776232986, 'epoch': 1} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7127746635632035, 'r': 0.6583194973727579, 'f1': 0.684465694079899}, 'combined': 0.476831687440019, 'epoch': 1} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.4982400932400932, 'r': 0.5793489456280154, 'f1': 0.5357420357420357}, 'combined': 0.2602175602175602, 'epoch': 1} ****************************** Epoch: 2 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 06:02:20.517374: step: 2/1541, loss: 1.0842299461364746 2023-01-23 06:02:21.372505: step: 4/1541, loss: 0.4085042178630829 2023-01-23 06:02:22.287613: step: 6/1541, loss: 0.40158525109291077 2023-01-23 06:02:23.175210: step: 8/1541, loss: 0.5233862996101379 2023-01-23 06:02:24.072768: step: 10/1541, loss: 0.8812931776046753 2023-01-23 06:02:24.960545: step: 12/1541, loss: 1.0712823867797852 2023-01-23 06:02:25.864215: step: 14/1541, loss: 0.7556844353675842 2023-01-23 06:02:26.871487: step: 16/1541, loss: 2.235907793045044 2023-01-23 06:02:27.848939: step: 18/1541, loss: 0.8915612101554871 2023-01-23 06:02:28.733706: step: 20/1541, loss: 0.4643523395061493 2023-01-23 06:02:29.579720: step: 22/1541, loss: 1.6858181953430176 2023-01-23 06:02:30.448761: step: 24/1541, loss: 0.6810336709022522 2023-01-23 06:02:31.375320: step: 26/1541, loss: 0.7807514667510986 2023-01-23 06:02:32.263303: step: 28/1541, loss: 1.0987297296524048 2023-01-23 06:02:33.137899: step: 30/1541, loss: 1.559050440788269 2023-01-23 06:02:34.050651: step: 32/1541, loss: 2.717447519302368 2023-01-23 06:02:34.932864: step: 34/1541, loss: 3.006192684173584 2023-01-23 06:02:35.793303: step: 36/1541, loss: 1.0470757484436035 2023-01-23 06:02:36.737817: step: 38/1541, loss: 3.665649175643921 2023-01-23 06:02:37.606273: step: 40/1541, loss: 0.27472180128097534 2023-01-23 06:02:38.483249: step: 42/1541, loss: 3.3684372901916504 2023-01-23 06:02:39.461416: step: 44/1541, loss: 1.1889629364013672 2023-01-23 06:02:40.316559: step: 46/1541, loss: 0.9115402698516846 2023-01-23 06:02:41.176052: step: 48/1541, loss: 1.3958491086959839 2023-01-23 06:02:42.104103: step: 50/1541, loss: 0.6139430403709412 2023-01-23 06:02:43.001104: step: 52/1541, loss: 3.0746190547943115 2023-01-23 06:02:43.888314: step: 54/1541, loss: 1.3168458938598633 2023-01-23 06:02:44.814584: step: 56/1541, loss: 0.7888268232345581 2023-01-23 06:02:45.689577: step: 58/1541, loss: 0.7597337961196899 2023-01-23 06:02:46.629434: step: 60/1541, loss: 1.7224498987197876 2023-01-23 06:02:47.508328: step: 62/1541, loss: 0.6131554841995239 2023-01-23 06:02:48.425364: step: 64/1541, loss: 1.0219615697860718 2023-01-23 06:02:49.282850: step: 66/1541, loss: 0.5064013004302979 2023-01-23 06:02:50.182882: step: 68/1541, loss: 1.2145657539367676 2023-01-23 06:02:51.035587: step: 70/1541, loss: 0.8255782127380371 2023-01-23 06:02:51.930073: step: 72/1541, loss: 1.156599760055542 2023-01-23 06:02:52.848272: step: 74/1541, loss: 0.6015948057174683 2023-01-23 06:02:53.721986: step: 76/1541, loss: 0.816048264503479 2023-01-23 06:02:54.614203: step: 78/1541, loss: 1.0028977394104004 2023-01-23 06:02:55.522843: step: 80/1541, loss: 0.5207762718200684 2023-01-23 06:02:56.518616: step: 82/1541, loss: 0.6063483357429504 2023-01-23 06:02:57.420841: step: 84/1541, loss: 0.4492877423763275 2023-01-23 06:02:58.335883: step: 86/1541, loss: 0.8254169225692749 2023-01-23 06:02:59.213774: step: 88/1541, loss: 0.3743269145488739 2023-01-23 06:03:00.093777: step: 90/1541, loss: 2.9706361293792725 2023-01-23 06:03:00.971234: step: 92/1541, loss: 3.336660385131836 2023-01-23 06:03:01.900606: step: 94/1541, loss: 1.4257338047027588 2023-01-23 06:03:02.813588: step: 96/1541, loss: 0.5605168342590332 2023-01-23 06:03:03.715800: step: 98/1541, loss: 1.5185729265213013 2023-01-23 06:03:04.630704: step: 100/1541, loss: 1.3954075574874878 2023-01-23 06:03:05.470432: step: 102/1541, loss: 0.9971560835838318 2023-01-23 06:03:06.394425: step: 104/1541, loss: 7.128517150878906 2023-01-23 06:03:07.400511: step: 106/1541, loss: 0.9896306395530701 2023-01-23 06:03:08.279439: step: 108/1541, loss: 0.493122935295105 2023-01-23 06:03:09.190309: step: 110/1541, loss: 0.710349440574646 2023-01-23 06:03:10.067521: step: 112/1541, loss: 1.195454478263855 2023-01-23 06:03:11.014199: step: 114/1541, loss: 3.3632349967956543 2023-01-23 06:03:11.975330: step: 116/1541, loss: 1.0761960744857788 2023-01-23 06:03:12.886760: step: 118/1541, loss: 0.4781879186630249 2023-01-23 06:03:13.793242: step: 120/1541, loss: 3.9244301319122314 2023-01-23 06:03:14.721414: step: 122/1541, loss: 0.4743001461029053 2023-01-23 06:03:15.604389: step: 124/1541, loss: 3.4500441551208496 2023-01-23 06:03:16.529306: step: 126/1541, loss: 0.5705998539924622 2023-01-23 06:03:17.418795: step: 128/1541, loss: 2.7005112171173096 2023-01-23 06:03:18.322485: step: 130/1541, loss: 5.467536449432373 2023-01-23 06:03:19.261663: step: 132/1541, loss: 5.094424247741699 2023-01-23 06:03:20.160249: step: 134/1541, loss: 0.5013097524642944 2023-01-23 06:03:21.053859: step: 136/1541, loss: 0.6839438080787659 2023-01-23 06:03:21.956210: step: 138/1541, loss: 0.6833668351173401 2023-01-23 06:03:22.835800: step: 140/1541, loss: 0.9116135239601135 2023-01-23 06:03:23.725909: step: 142/1541, loss: 0.583612322807312 2023-01-23 06:03:24.630656: step: 144/1541, loss: 0.6172298789024353 2023-01-23 06:03:25.542187: step: 146/1541, loss: 0.8440280556678772 2023-01-23 06:03:26.401802: step: 148/1541, loss: 0.49058014154434204 2023-01-23 06:03:27.303635: step: 150/1541, loss: 1.14420485496521 2023-01-23 06:03:28.216315: step: 152/1541, loss: 0.668762743473053 2023-01-23 06:03:29.152896: step: 154/1541, loss: 0.9455240964889526 2023-01-23 06:03:30.048005: step: 156/1541, loss: 0.8566352128982544 2023-01-23 06:03:31.024118: step: 158/1541, loss: 1.2392785549163818 2023-01-23 06:03:31.897191: step: 160/1541, loss: 1.6482411623001099 2023-01-23 06:03:32.834138: step: 162/1541, loss: 0.9709704518318176 2023-01-23 06:03:33.750106: step: 164/1541, loss: 1.9008769989013672 2023-01-23 06:03:34.688863: step: 166/1541, loss: 0.6517098546028137 2023-01-23 06:03:35.683551: step: 168/1541, loss: 1.118730902671814 2023-01-23 06:03:36.570336: step: 170/1541, loss: 0.6410865187644958 2023-01-23 06:03:37.519650: step: 172/1541, loss: 3.895099401473999 2023-01-23 06:03:38.428657: step: 174/1541, loss: 0.6229751110076904 2023-01-23 06:03:39.323484: step: 176/1541, loss: 1.1146655082702637 2023-01-23 06:03:40.225763: step: 178/1541, loss: 1.0398555994033813 2023-01-23 06:03:41.106048: step: 180/1541, loss: 1.2145922183990479 2023-01-23 06:03:42.003315: step: 182/1541, loss: 0.750381350517273 2023-01-23 06:03:42.856789: step: 184/1541, loss: 4.256346225738525 2023-01-23 06:03:43.738220: step: 186/1541, loss: 0.9606316089630127 2023-01-23 06:03:44.598900: step: 188/1541, loss: 0.8480685353279114 2023-01-23 06:03:45.481672: step: 190/1541, loss: 0.6121640801429749 2023-01-23 06:03:46.329014: step: 192/1541, loss: 2.887930154800415 2023-01-23 06:03:47.168294: step: 194/1541, loss: 0.3467995524406433 2023-01-23 06:03:48.052924: step: 196/1541, loss: 0.8326122760772705 2023-01-23 06:03:48.956523: step: 198/1541, loss: 0.9562017917633057 2023-01-23 06:03:49.886029: step: 200/1541, loss: 0.38190191984176636 2023-01-23 06:03:50.774014: step: 202/1541, loss: 1.7034168243408203 2023-01-23 06:03:51.632047: step: 204/1541, loss: 1.284226655960083 2023-01-23 06:03:52.647660: step: 206/1541, loss: 0.4573616087436676 2023-01-23 06:03:53.528889: step: 208/1541, loss: 3.3526227474212646 2023-01-23 06:03:54.396747: step: 210/1541, loss: 0.719551146030426 2023-01-23 06:03:55.260103: step: 212/1541, loss: 0.8888340592384338 2023-01-23 06:03:56.211226: step: 214/1541, loss: 0.7337324619293213 2023-01-23 06:03:57.100732: step: 216/1541, loss: 0.815495491027832 2023-01-23 06:03:58.003415: step: 218/1541, loss: 0.4488155245780945 2023-01-23 06:03:58.871968: step: 220/1541, loss: 0.753823459148407 2023-01-23 06:03:59.783592: step: 222/1541, loss: 0.832984209060669 2023-01-23 06:04:00.758263: step: 224/1541, loss: 0.9534823298454285 2023-01-23 06:04:01.708752: step: 226/1541, loss: 1.2361153364181519 2023-01-23 06:04:02.605105: step: 228/1541, loss: 0.41915059089660645 2023-01-23 06:04:03.537997: step: 230/1541, loss: 1.7847784757614136 2023-01-23 06:04:04.404648: step: 232/1541, loss: 0.6170253753662109 2023-01-23 06:04:05.328662: step: 234/1541, loss: 3.0003535747528076 2023-01-23 06:04:06.245078: step: 236/1541, loss: 1.188470721244812 2023-01-23 06:04:07.171381: step: 238/1541, loss: 0.9473892450332642 2023-01-23 06:04:08.055985: step: 240/1541, loss: 1.0082483291625977 2023-01-23 06:04:09.031697: step: 242/1541, loss: 0.6911032795906067 2023-01-23 06:04:10.024993: step: 244/1541, loss: 0.5860386490821838 2023-01-23 06:04:11.002779: step: 246/1541, loss: 1.628117322921753 2023-01-23 06:04:11.928834: step: 248/1541, loss: 0.8031197786331177 2023-01-23 06:04:12.797703: step: 250/1541, loss: 1.0843873023986816 2023-01-23 06:04:13.805538: step: 252/1541, loss: 0.669636607170105 2023-01-23 06:04:14.753395: step: 254/1541, loss: 1.2326221466064453 2023-01-23 06:04:15.713841: step: 256/1541, loss: 1.0721523761749268 2023-01-23 06:04:16.656957: step: 258/1541, loss: 0.9723489284515381 2023-01-23 06:04:17.640892: step: 260/1541, loss: 0.4092591404914856 2023-01-23 06:04:18.478304: step: 262/1541, loss: 0.7832691669464111 2023-01-23 06:04:19.368023: step: 264/1541, loss: 0.5158959031105042 2023-01-23 06:04:20.251246: step: 266/1541, loss: 0.6312225461006165 2023-01-23 06:04:21.118504: step: 268/1541, loss: 1.1452438831329346 2023-01-23 06:04:22.033507: step: 270/1541, loss: 2.339473009109497 2023-01-23 06:04:22.919010: step: 272/1541, loss: 0.6732713580131531 2023-01-23 06:04:23.800934: step: 274/1541, loss: 0.40130847692489624 2023-01-23 06:04:24.717199: step: 276/1541, loss: 2.6330788135528564 2023-01-23 06:04:25.603897: step: 278/1541, loss: 1.1531634330749512 2023-01-23 06:04:26.489321: step: 280/1541, loss: 10.167716026306152 2023-01-23 06:04:27.360754: step: 282/1541, loss: 0.8071631789207458 2023-01-23 06:04:28.244152: step: 284/1541, loss: 1.145243763923645 2023-01-23 06:04:29.128800: step: 286/1541, loss: 0.36756592988967896 2023-01-23 06:04:30.019087: step: 288/1541, loss: 0.6387955546379089 2023-01-23 06:04:30.945642: step: 290/1541, loss: 0.6016181707382202 2023-01-23 06:04:31.881307: step: 292/1541, loss: 0.4676283299922943 2023-01-23 06:04:32.808845: step: 294/1541, loss: 0.29346537590026855 2023-01-23 06:04:33.701894: step: 296/1541, loss: 1.034338116645813 2023-01-23 06:04:34.621890: step: 298/1541, loss: 0.5770531296730042 2023-01-23 06:04:35.478046: step: 300/1541, loss: 0.7996317744255066 2023-01-23 06:04:36.390051: step: 302/1541, loss: 0.7917903661727905 2023-01-23 06:04:37.280586: step: 304/1541, loss: 0.4415876269340515 2023-01-23 06:04:38.141700: step: 306/1541, loss: 0.7234194874763489 2023-01-23 06:04:39.036817: step: 308/1541, loss: 0.31453004479408264 2023-01-23 06:04:39.944432: step: 310/1541, loss: 2.9319393634796143 2023-01-23 06:04:40.829710: step: 312/1541, loss: 0.8129768371582031 2023-01-23 06:04:41.700767: step: 314/1541, loss: 0.5721260905265808 2023-01-23 06:04:42.579707: step: 316/1541, loss: 0.5558392405509949 2023-01-23 06:04:43.485606: step: 318/1541, loss: 1.5373923778533936 2023-01-23 06:04:44.359574: step: 320/1541, loss: 0.83409583568573 2023-01-23 06:04:45.237045: step: 322/1541, loss: 0.7775567770004272 2023-01-23 06:04:46.151428: step: 324/1541, loss: 1.2965229749679565 2023-01-23 06:04:47.029464: step: 326/1541, loss: 1.988840103149414 2023-01-23 06:04:47.903505: step: 328/1541, loss: 3.196136713027954 2023-01-23 06:04:48.870118: step: 330/1541, loss: 1.1212564706802368 2023-01-23 06:04:49.710136: step: 332/1541, loss: 3.230910301208496 2023-01-23 06:04:50.591524: step: 334/1541, loss: 0.8902889490127563 2023-01-23 06:04:51.464186: step: 336/1541, loss: 3.3413455486297607 2023-01-23 06:04:52.346692: step: 338/1541, loss: 0.5266730785369873 2023-01-23 06:04:53.218652: step: 340/1541, loss: 0.6320213079452515 2023-01-23 06:04:54.101697: step: 342/1541, loss: 0.8804971575737 2023-01-23 06:04:55.093600: step: 344/1541, loss: 0.5225503444671631 2023-01-23 06:04:55.977091: step: 346/1541, loss: 1.2921485900878906 2023-01-23 06:04:56.934889: step: 348/1541, loss: 7.378821849822998 2023-01-23 06:04:57.814978: step: 350/1541, loss: 0.7965288162231445 2023-01-23 06:04:58.687214: step: 352/1541, loss: 0.7731444239616394 2023-01-23 06:04:59.627309: step: 354/1541, loss: 1.34514582157135 2023-01-23 06:05:00.624552: step: 356/1541, loss: 0.35918861627578735 2023-01-23 06:05:01.533602: step: 358/1541, loss: 1.5605276823043823 2023-01-23 06:05:02.454117: step: 360/1541, loss: 1.3137246370315552 2023-01-23 06:05:03.380091: step: 362/1541, loss: 0.6959186792373657 2023-01-23 06:05:04.266644: step: 364/1541, loss: 7.395930290222168 2023-01-23 06:05:05.133561: step: 366/1541, loss: 1.4199934005737305 2023-01-23 06:05:06.083296: step: 368/1541, loss: 0.9886400103569031 2023-01-23 06:05:06.997813: step: 370/1541, loss: 3.274172067642212 2023-01-23 06:05:07.941747: step: 372/1541, loss: 0.3677622377872467 2023-01-23 06:05:08.833358: step: 374/1541, loss: 0.5831061005592346 2023-01-23 06:05:09.777154: step: 376/1541, loss: 3.2060046195983887 2023-01-23 06:05:10.646546: step: 378/1541, loss: 0.881058931350708 2023-01-23 06:05:11.519534: step: 380/1541, loss: 2.8867228031158447 2023-01-23 06:05:12.411199: step: 382/1541, loss: 0.6921019554138184 2023-01-23 06:05:13.349012: step: 384/1541, loss: 0.6749218106269836 2023-01-23 06:05:14.228414: step: 386/1541, loss: 0.38867467641830444 2023-01-23 06:05:15.136424: step: 388/1541, loss: 9.852973937988281 2023-01-23 06:05:16.028654: step: 390/1541, loss: 3.189366340637207 2023-01-23 06:05:16.937619: step: 392/1541, loss: 1.69962477684021 2023-01-23 06:05:17.788306: step: 394/1541, loss: 3.158879518508911 2023-01-23 06:05:18.659995: step: 396/1541, loss: 0.9666619896888733 2023-01-23 06:05:19.608996: step: 398/1541, loss: 0.621080756187439 2023-01-23 06:05:20.486041: step: 400/1541, loss: 1.4902883768081665 2023-01-23 06:05:21.414200: step: 402/1541, loss: 0.44777488708496094 2023-01-23 06:05:22.394817: step: 404/1541, loss: 1.1564463376998901 2023-01-23 06:05:23.333343: step: 406/1541, loss: 0.8491976261138916 2023-01-23 06:05:24.244037: step: 408/1541, loss: 0.22592273354530334 2023-01-23 06:05:25.183690: step: 410/1541, loss: 1.1440554857254028 2023-01-23 06:05:26.094747: step: 412/1541, loss: 1.2330561876296997 2023-01-23 06:05:26.947630: step: 414/1541, loss: 0.6575852036476135 2023-01-23 06:05:27.828814: step: 416/1541, loss: 0.5263446569442749 2023-01-23 06:05:28.783685: step: 418/1541, loss: 1.8171651363372803 2023-01-23 06:05:29.631673: step: 420/1541, loss: 0.29677629470825195 2023-01-23 06:05:30.530541: step: 422/1541, loss: 0.44833117723464966 2023-01-23 06:05:31.407479: step: 424/1541, loss: 0.8974947929382324 2023-01-23 06:05:32.283883: step: 426/1541, loss: 0.7019454836845398 2023-01-23 06:05:33.177248: step: 428/1541, loss: 1.0000931024551392 2023-01-23 06:05:34.091254: step: 430/1541, loss: 1.0915241241455078 2023-01-23 06:05:34.945395: step: 432/1541, loss: 0.6146532893180847 2023-01-23 06:05:35.822645: step: 434/1541, loss: 0.31519776582717896 2023-01-23 06:05:36.681755: step: 436/1541, loss: 0.5491338968276978 2023-01-23 06:05:37.556559: step: 438/1541, loss: 0.5660479664802551 2023-01-23 06:05:38.495159: step: 440/1541, loss: 1.0836881399154663 2023-01-23 06:05:39.422795: step: 442/1541, loss: 0.669329047203064 2023-01-23 06:05:40.346783: step: 444/1541, loss: 0.8762286901473999 2023-01-23 06:05:41.329759: step: 446/1541, loss: 1.8657819032669067 2023-01-23 06:05:42.206356: step: 448/1541, loss: 3.2871670722961426 2023-01-23 06:05:43.128567: step: 450/1541, loss: 0.8604196310043335 2023-01-23 06:05:44.069801: step: 452/1541, loss: 1.2834731340408325 2023-01-23 06:05:44.981897: step: 454/1541, loss: 0.9515957236289978 2023-01-23 06:05:45.862583: step: 456/1541, loss: 1.1208947896957397 2023-01-23 06:05:46.732086: step: 458/1541, loss: 0.49263617396354675 2023-01-23 06:05:47.651291: step: 460/1541, loss: 1.1245378255844116 2023-01-23 06:05:48.603867: step: 462/1541, loss: 0.9075873494148254 2023-01-23 06:05:49.518965: step: 464/1541, loss: 1.4586338996887207 2023-01-23 06:05:50.412732: step: 466/1541, loss: 0.6318588256835938 2023-01-23 06:05:51.317584: step: 468/1541, loss: 0.5884584188461304 2023-01-23 06:05:52.274317: step: 470/1541, loss: 0.6032633185386658 2023-01-23 06:05:53.209389: step: 472/1541, loss: 1.171263575553894 2023-01-23 06:05:54.146205: step: 474/1541, loss: 0.7670339345932007 2023-01-23 06:05:55.087048: step: 476/1541, loss: 1.1253306865692139 2023-01-23 06:05:55.967257: step: 478/1541, loss: 0.440674364566803 2023-01-23 06:05:56.870679: step: 480/1541, loss: 0.5366206169128418 2023-01-23 06:05:57.760923: step: 482/1541, loss: 0.3966054618358612 2023-01-23 06:05:58.628883: step: 484/1541, loss: 1.5532509088516235 2023-01-23 06:05:59.479906: step: 486/1541, loss: 0.25129178166389465 2023-01-23 06:06:00.362685: step: 488/1541, loss: 1.0228286981582642 2023-01-23 06:06:01.234821: step: 490/1541, loss: 1.2282928228378296 2023-01-23 06:06:02.203702: step: 492/1541, loss: 0.8199703693389893 2023-01-23 06:06:03.130080: step: 494/1541, loss: 0.8553823828697205 2023-01-23 06:06:04.028125: step: 496/1541, loss: 1.508705973625183 2023-01-23 06:06:04.936226: step: 498/1541, loss: 0.8304944038391113 2023-01-23 06:06:05.882437: step: 500/1541, loss: 2.96541690826416 2023-01-23 06:06:06.769524: step: 502/1541, loss: 1.0358978509902954 2023-01-23 06:06:07.703899: step: 504/1541, loss: 0.574733316898346 2023-01-23 06:06:08.569780: step: 506/1541, loss: 0.6076946258544922 2023-01-23 06:06:09.431124: step: 508/1541, loss: 0.4767799377441406 2023-01-23 06:06:10.312561: step: 510/1541, loss: 1.3718150854110718 2023-01-23 06:06:11.289977: step: 512/1541, loss: 1.281416893005371 2023-01-23 06:06:12.189395: step: 514/1541, loss: 1.2923064231872559 2023-01-23 06:06:13.065226: step: 516/1541, loss: 0.9965929985046387 2023-01-23 06:06:13.988159: step: 518/1541, loss: 1.4238591194152832 2023-01-23 06:06:14.960126: step: 520/1541, loss: 0.8377792239189148 2023-01-23 06:06:15.844321: step: 522/1541, loss: 0.5717670917510986 2023-01-23 06:06:16.723764: step: 524/1541, loss: 0.727001965045929 2023-01-23 06:06:17.594391: step: 526/1541, loss: 0.80942302942276 2023-01-23 06:06:18.520042: step: 528/1541, loss: 2.1042628288269043 2023-01-23 06:06:19.406335: step: 530/1541, loss: 0.7178446054458618 2023-01-23 06:06:20.291720: step: 532/1541, loss: 1.6249668598175049 2023-01-23 06:06:21.205228: step: 534/1541, loss: 0.755119800567627 2023-01-23 06:06:22.063826: step: 536/1541, loss: 1.0135992765426636 2023-01-23 06:06:22.989060: step: 538/1541, loss: 0.928926408290863 2023-01-23 06:06:23.960849: step: 540/1541, loss: 3.24985933303833 2023-01-23 06:06:24.883067: step: 542/1541, loss: 1.0843132734298706 2023-01-23 06:06:25.744705: step: 544/1541, loss: 1.1081384420394897 2023-01-23 06:06:26.627940: step: 546/1541, loss: 1.028221845626831 2023-01-23 06:06:27.507942: step: 548/1541, loss: 0.579384982585907 2023-01-23 06:06:28.427538: step: 550/1541, loss: 1.1210236549377441 2023-01-23 06:06:29.344529: step: 552/1541, loss: 3.1292455196380615 2023-01-23 06:06:30.224608: step: 554/1541, loss: 0.6648995280265808 2023-01-23 06:06:31.175270: step: 556/1541, loss: 1.0940513610839844 2023-01-23 06:06:32.061912: step: 558/1541, loss: 0.4036954343318939 2023-01-23 06:06:32.953344: step: 560/1541, loss: 0.7115387916564941 2023-01-23 06:06:33.811402: step: 562/1541, loss: 0.5358917713165283 2023-01-23 06:06:34.705598: step: 564/1541, loss: 0.7160725593566895 2023-01-23 06:06:35.680842: step: 566/1541, loss: 0.7778767347335815 2023-01-23 06:06:36.582883: step: 568/1541, loss: 1.1403563022613525 2023-01-23 06:06:37.504884: step: 570/1541, loss: 0.8934710621833801 2023-01-23 06:06:38.402876: step: 572/1541, loss: 0.5261983275413513 2023-01-23 06:06:39.298900: step: 574/1541, loss: 1.0724828243255615 2023-01-23 06:06:40.179616: step: 576/1541, loss: 1.383747935295105 2023-01-23 06:06:41.118544: step: 578/1541, loss: 1.8787741661071777 2023-01-23 06:06:41.996333: step: 580/1541, loss: 0.4258998930454254 2023-01-23 06:06:42.883547: step: 582/1541, loss: 1.4158307313919067 2023-01-23 06:06:43.846505: step: 584/1541, loss: 0.8460870385169983 2023-01-23 06:06:44.717495: step: 586/1541, loss: 1.4279632568359375 2023-01-23 06:06:45.686611: step: 588/1541, loss: 2.05594539642334 2023-01-23 06:06:46.609213: step: 590/1541, loss: 0.6734881401062012 2023-01-23 06:06:47.509875: step: 592/1541, loss: 0.8030192255973816 2023-01-23 06:06:48.352575: step: 594/1541, loss: 3.2004830837249756 2023-01-23 06:06:49.256831: step: 596/1541, loss: 0.5537482500076294 2023-01-23 06:06:50.140307: step: 598/1541, loss: 3.3633697032928467 2023-01-23 06:06:51.053764: step: 600/1541, loss: 1.108134150505066 2023-01-23 06:06:51.968473: step: 602/1541, loss: 0.6017315983772278 2023-01-23 06:06:52.865651: step: 604/1541, loss: 1.3505676984786987 2023-01-23 06:06:53.733625: step: 606/1541, loss: 0.6303764581680298 2023-01-23 06:06:54.602123: step: 608/1541, loss: 0.6234962344169617 2023-01-23 06:06:55.473146: step: 610/1541, loss: 0.8091936111450195 2023-01-23 06:06:56.403912: step: 612/1541, loss: 0.4352775514125824 2023-01-23 06:06:57.296315: step: 614/1541, loss: 0.6571961641311646 2023-01-23 06:06:58.174014: step: 616/1541, loss: 0.7115850448608398 2023-01-23 06:06:59.070210: step: 618/1541, loss: 7.94683837890625 2023-01-23 06:06:59.973392: step: 620/1541, loss: 1.1406170129776 2023-01-23 06:07:00.846042: step: 622/1541, loss: 3.4070475101470947 2023-01-23 06:07:01.703210: step: 624/1541, loss: 0.7648618817329407 2023-01-23 06:07:02.682347: step: 626/1541, loss: 3.2074170112609863 2023-01-23 06:07:03.564186: step: 628/1541, loss: 0.9382791519165039 2023-01-23 06:07:04.454556: step: 630/1541, loss: 1.0099016427993774 2023-01-23 06:07:05.372664: step: 632/1541, loss: 0.7877002358436584 2023-01-23 06:07:06.269895: step: 634/1541, loss: 1.8355894088745117 2023-01-23 06:07:07.181377: step: 636/1541, loss: 0.5981012582778931 2023-01-23 06:07:08.174447: step: 638/1541, loss: 1.3067864179611206 2023-01-23 06:07:09.041186: step: 640/1541, loss: 0.4795820415019989 2023-01-23 06:07:09.980401: step: 642/1541, loss: 2.234280824661255 2023-01-23 06:07:10.898582: step: 644/1541, loss: 1.092376947402954 2023-01-23 06:07:11.753081: step: 646/1541, loss: 0.6129226684570312 2023-01-23 06:07:12.644215: step: 648/1541, loss: 1.7347650527954102 2023-01-23 06:07:13.518235: step: 650/1541, loss: 0.7383831739425659 2023-01-23 06:07:14.389433: step: 652/1541, loss: 0.7733483910560608 2023-01-23 06:07:15.301440: step: 654/1541, loss: 1.022044062614441 2023-01-23 06:07:16.273427: step: 656/1541, loss: 1.1616767644882202 2023-01-23 06:07:17.200070: step: 658/1541, loss: 3.282824993133545 2023-01-23 06:07:18.066300: step: 660/1541, loss: 0.7928851246833801 2023-01-23 06:07:18.952534: step: 662/1541, loss: 6.874966144561768 2023-01-23 06:07:19.868297: step: 664/1541, loss: 0.4492018520832062 2023-01-23 06:07:20.755450: step: 666/1541, loss: 1.6138306856155396 2023-01-23 06:07:21.666579: step: 668/1541, loss: 3.4058597087860107 2023-01-23 06:07:22.571340: step: 670/1541, loss: 1.2189241647720337 2023-01-23 06:07:23.474763: step: 672/1541, loss: 0.6342324614524841 2023-01-23 06:07:24.418137: step: 674/1541, loss: 3.4791083335876465 2023-01-23 06:07:25.319551: step: 676/1541, loss: 0.4483400583267212 2023-01-23 06:07:26.236585: step: 678/1541, loss: 0.42605748772621155 2023-01-23 06:07:27.141596: step: 680/1541, loss: 0.9109981060028076 2023-01-23 06:07:28.085240: step: 682/1541, loss: 0.8723540902137756 2023-01-23 06:07:28.986133: step: 684/1541, loss: 2.8451366424560547 2023-01-23 06:07:29.895253: step: 686/1541, loss: 1.2119343280792236 2023-01-23 06:07:30.843357: step: 688/1541, loss: 1.1118212938308716 2023-01-23 06:07:31.699859: step: 690/1541, loss: 0.7275924682617188 2023-01-23 06:07:32.689867: step: 692/1541, loss: 0.7389818429946899 2023-01-23 06:07:33.673462: step: 694/1541, loss: 0.879856526851654 2023-01-23 06:07:34.588016: step: 696/1541, loss: 3.513880729675293 2023-01-23 06:07:35.469776: step: 698/1541, loss: 0.6762356162071228 2023-01-23 06:07:36.393871: step: 700/1541, loss: 3.2241785526275635 2023-01-23 06:07:37.285122: step: 702/1541, loss: 0.28021806478500366 2023-01-23 06:07:38.166038: step: 704/1541, loss: 0.8350697755813599 2023-01-23 06:07:39.062940: step: 706/1541, loss: 1.521878719329834 2023-01-23 06:07:39.921941: step: 708/1541, loss: 3.148465633392334 2023-01-23 06:07:40.866810: step: 710/1541, loss: 0.9293065667152405 2023-01-23 06:07:41.803519: step: 712/1541, loss: 3.252933979034424 2023-01-23 06:07:42.769587: step: 714/1541, loss: 1.463368535041809 2023-01-23 06:07:43.661645: step: 716/1541, loss: 0.4527479112148285 2023-01-23 06:07:44.572519: step: 718/1541, loss: 0.5378947257995605 2023-01-23 06:07:45.512692: step: 720/1541, loss: 0.6226381063461304 2023-01-23 06:07:46.421699: step: 722/1541, loss: 0.6113121509552002 2023-01-23 06:07:47.321183: step: 724/1541, loss: 1.0874873399734497 2023-01-23 06:07:48.243236: step: 726/1541, loss: 0.6373812556266785 2023-01-23 06:07:49.123829: step: 728/1541, loss: 0.5625945925712585 2023-01-23 06:07:50.037333: step: 730/1541, loss: 1.1573193073272705 2023-01-23 06:07:50.948945: step: 732/1541, loss: 2.010258197784424 2023-01-23 06:07:51.839592: step: 734/1541, loss: 1.3202099800109863 2023-01-23 06:07:52.754488: step: 736/1541, loss: 0.7844889760017395 2023-01-23 06:07:53.654578: step: 738/1541, loss: 0.7085586190223694 2023-01-23 06:07:54.540800: step: 740/1541, loss: 1.4283255338668823 2023-01-23 06:07:55.453393: step: 742/1541, loss: 0.8599850535392761 2023-01-23 06:07:56.353645: step: 744/1541, loss: 6.8123955726623535 2023-01-23 06:07:57.293675: step: 746/1541, loss: 1.4584054946899414 2023-01-23 06:07:58.214305: step: 748/1541, loss: 1.2781236171722412 2023-01-23 06:07:59.109369: step: 750/1541, loss: 0.9907675981521606 2023-01-23 06:08:00.026991: step: 752/1541, loss: 0.9241988658905029 2023-01-23 06:08:00.970751: step: 754/1541, loss: 1.3339763879776 2023-01-23 06:08:01.883562: step: 756/1541, loss: 0.24890492856502533 2023-01-23 06:08:02.810705: step: 758/1541, loss: 0.8197736144065857 2023-01-23 06:08:03.682371: step: 760/1541, loss: 3.676710605621338 2023-01-23 06:08:04.558478: step: 762/1541, loss: 1.145090937614441 2023-01-23 06:08:05.450649: step: 764/1541, loss: 1.0399566888809204 2023-01-23 06:08:06.338062: step: 766/1541, loss: 5.3690338134765625 2023-01-23 06:08:07.206311: step: 768/1541, loss: 1.0298552513122559 2023-01-23 06:08:08.115268: step: 770/1541, loss: 0.7898194789886475 2023-01-23 06:08:09.177885: step: 772/1541, loss: 1.2268116474151611 2023-01-23 06:08:10.091451: step: 774/1541, loss: 0.5242905616760254 2023-01-23 06:08:11.017614: step: 776/1541, loss: 0.6066920161247253 2023-01-23 06:08:11.877803: step: 778/1541, loss: 0.8209066390991211 2023-01-23 06:08:12.800306: step: 780/1541, loss: 0.9852568507194519 2023-01-23 06:08:13.745025: step: 782/1541, loss: 1.1889169216156006 2023-01-23 06:08:14.713442: step: 784/1541, loss: 0.7630932331085205 2023-01-23 06:08:15.642384: step: 786/1541, loss: 3.22802996635437 2023-01-23 06:08:16.536408: step: 788/1541, loss: 0.5967560410499573 2023-01-23 06:08:17.459372: step: 790/1541, loss: 1.021149754524231 2023-01-23 06:08:18.360896: step: 792/1541, loss: 0.608543336391449 2023-01-23 06:08:19.302582: step: 794/1541, loss: 0.9219532608985901 2023-01-23 06:08:20.231053: step: 796/1541, loss: 0.6067726612091064 2023-01-23 06:08:21.125571: step: 798/1541, loss: 1.391600251197815 2023-01-23 06:08:22.023151: step: 800/1541, loss: 3.4503941535949707 2023-01-23 06:08:22.942318: step: 802/1541, loss: 7.208365440368652 2023-01-23 06:08:23.824578: step: 804/1541, loss: 1.0656275749206543 2023-01-23 06:08:24.783409: step: 806/1541, loss: 0.8153055906295776 2023-01-23 06:08:25.695902: step: 808/1541, loss: 0.7353812456130981 2023-01-23 06:08:26.634179: step: 810/1541, loss: 1.0545639991760254 2023-01-23 06:08:27.541030: step: 812/1541, loss: 1.6295037269592285 2023-01-23 06:08:28.455720: step: 814/1541, loss: 0.6074888110160828 2023-01-23 06:08:29.412970: step: 816/1541, loss: 0.8943935632705688 2023-01-23 06:08:30.349382: step: 818/1541, loss: 3.3490078449249268 2023-01-23 06:08:31.248097: step: 820/1541, loss: 0.5947072505950928 2023-01-23 06:08:32.172143: step: 822/1541, loss: 1.0879915952682495 2023-01-23 06:08:33.057245: step: 824/1541, loss: 0.5735174417495728 2023-01-23 06:08:33.952504: step: 826/1541, loss: 1.5425647497177124 2023-01-23 06:08:34.905841: step: 828/1541, loss: 0.9069414734840393 2023-01-23 06:08:35.809133: step: 830/1541, loss: 0.6240565776824951 2023-01-23 06:08:36.658203: step: 832/1541, loss: 0.5991878509521484 2023-01-23 06:08:37.564347: step: 834/1541, loss: 1.0509235858917236 2023-01-23 06:08:38.486986: step: 836/1541, loss: 0.5619840621948242 2023-01-23 06:08:39.407738: step: 838/1541, loss: 0.5879969000816345 2023-01-23 06:08:40.326953: step: 840/1541, loss: 1.0613270998001099 2023-01-23 06:08:41.245174: step: 842/1541, loss: 0.6534615159034729 2023-01-23 06:08:42.185413: step: 844/1541, loss: 0.5973045229911804 2023-01-23 06:08:43.073378: step: 846/1541, loss: 0.3775921165943146 2023-01-23 06:08:43.928682: step: 848/1541, loss: 1.1164231300354004 2023-01-23 06:08:44.840872: step: 850/1541, loss: 0.7531360983848572 2023-01-23 06:08:45.695532: step: 852/1541, loss: 0.5433734655380249 2023-01-23 06:08:46.581919: step: 854/1541, loss: 1.00020170211792 2023-01-23 06:08:47.483458: step: 856/1541, loss: 1.4554187059402466 2023-01-23 06:08:48.433043: step: 858/1541, loss: 4.186690330505371 2023-01-23 06:08:49.322241: step: 860/1541, loss: 0.682915985584259 2023-01-23 06:08:50.165750: step: 862/1541, loss: 0.5580244660377502 2023-01-23 06:08:51.099107: step: 864/1541, loss: 3.3291826248168945 2023-01-23 06:08:51.999918: step: 866/1541, loss: 0.8231256604194641 2023-01-23 06:08:52.919027: step: 868/1541, loss: 1.0565810203552246 2023-01-23 06:08:53.784472: step: 870/1541, loss: 0.5444850921630859 2023-01-23 06:08:54.712772: step: 872/1541, loss: 0.7172694206237793 2023-01-23 06:08:55.637315: step: 874/1541, loss: 0.6971880197525024 2023-01-23 06:08:56.514418: step: 876/1541, loss: 0.41331183910369873 2023-01-23 06:08:57.396763: step: 878/1541, loss: 0.47428297996520996 2023-01-23 06:08:58.326950: step: 880/1541, loss: 3.1860158443450928 2023-01-23 06:08:59.217065: step: 882/1541, loss: 0.7522020936012268 2023-01-23 06:09:00.095269: step: 884/1541, loss: 0.9582337737083435 2023-01-23 06:09:01.022892: step: 886/1541, loss: 0.9607124328613281 2023-01-23 06:09:01.947133: step: 888/1541, loss: 0.9434884786605835 2023-01-23 06:09:02.848847: step: 890/1541, loss: 1.0435131788253784 2023-01-23 06:09:03.780984: step: 892/1541, loss: 0.49362367391586304 2023-01-23 06:09:04.686332: step: 894/1541, loss: 1.1769993305206299 2023-01-23 06:09:05.569255: step: 896/1541, loss: 0.9365693926811218 2023-01-23 06:09:06.482785: step: 898/1541, loss: 1.1771310567855835 2023-01-23 06:09:07.359574: step: 900/1541, loss: 1.215015172958374 2023-01-23 06:09:08.287872: step: 902/1541, loss: 1.341173529624939 2023-01-23 06:09:09.180043: step: 904/1541, loss: 0.7792098522186279 2023-01-23 06:09:10.085953: step: 906/1541, loss: 0.7705392837524414 2023-01-23 06:09:11.002337: step: 908/1541, loss: 3.5973024368286133 2023-01-23 06:09:11.906471: step: 910/1541, loss: 0.6181037425994873 2023-01-23 06:09:12.826773: step: 912/1541, loss: 0.9687687158584595 2023-01-23 06:09:13.779647: step: 914/1541, loss: 6.8633599281311035 2023-01-23 06:09:14.653914: step: 916/1541, loss: 5.574495792388916 2023-01-23 06:09:15.522774: step: 918/1541, loss: 0.6820458173751831 2023-01-23 06:09:16.398770: step: 920/1541, loss: 1.4027281999588013 2023-01-23 06:09:17.280673: step: 922/1541, loss: 1.092954397201538 2023-01-23 06:09:18.168640: step: 924/1541, loss: 0.6134316325187683 2023-01-23 06:09:19.086215: step: 926/1541, loss: 0.9263119101524353 2023-01-23 06:09:19.947412: step: 928/1541, loss: 0.6413142681121826 2023-01-23 06:09:20.839980: step: 930/1541, loss: 1.1787152290344238 2023-01-23 06:09:21.747879: step: 932/1541, loss: 0.921108603477478 2023-01-23 06:09:22.694392: step: 934/1541, loss: 3.7241005897521973 2023-01-23 06:09:23.555661: step: 936/1541, loss: 0.6855366230010986 2023-01-23 06:09:24.449574: step: 938/1541, loss: 0.6406561136245728 2023-01-23 06:09:25.343493: step: 940/1541, loss: 1.2049546241760254 2023-01-23 06:09:26.245549: step: 942/1541, loss: 0.5515554547309875 2023-01-23 06:09:27.162084: step: 944/1541, loss: 1.1346579790115356 2023-01-23 06:09:28.038934: step: 946/1541, loss: 1.232867956161499 2023-01-23 06:09:28.909711: step: 948/1541, loss: 1.1641216278076172 2023-01-23 06:09:29.868081: step: 950/1541, loss: 0.9378539323806763 2023-01-23 06:09:30.771847: step: 952/1541, loss: 0.7047277092933655 2023-01-23 06:09:31.655082: step: 954/1541, loss: 7.1338067054748535 2023-01-23 06:09:32.598702: step: 956/1541, loss: 1.4046657085418701 2023-01-23 06:09:33.484149: step: 958/1541, loss: 0.6439711451530457 2023-01-23 06:09:34.459685: step: 960/1541, loss: 1.3414767980575562 2023-01-23 06:09:35.372703: step: 962/1541, loss: 9.149271965026855 2023-01-23 06:09:36.273762: step: 964/1541, loss: 0.8240286111831665 2023-01-23 06:09:37.180571: step: 966/1541, loss: 0.4458608031272888 2023-01-23 06:09:38.089169: step: 968/1541, loss: 0.7106008529663086 2023-01-23 06:09:38.973604: step: 970/1541, loss: 1.9227155447006226 2023-01-23 06:09:39.844328: step: 972/1541, loss: 0.8514583110809326 2023-01-23 06:09:40.757058: step: 974/1541, loss: 3.0872485637664795 2023-01-23 06:09:41.693296: step: 976/1541, loss: 2.904339551925659 2023-01-23 06:09:42.575812: step: 978/1541, loss: 0.5224287509918213 2023-01-23 06:09:43.521420: step: 980/1541, loss: 1.2437260150909424 2023-01-23 06:09:44.477125: step: 982/1541, loss: 0.5629892349243164 2023-01-23 06:09:45.391970: step: 984/1541, loss: 4.919036865234375 2023-01-23 06:09:46.299882: step: 986/1541, loss: 1.0493333339691162 2023-01-23 06:09:47.208233: step: 988/1541, loss: 0.5600607991218567 2023-01-23 06:09:48.145178: step: 990/1541, loss: 0.9219413995742798 2023-01-23 06:09:49.097258: step: 992/1541, loss: 0.48091426491737366 2023-01-23 06:09:49.970409: step: 994/1541, loss: 0.47707661986351013 2023-01-23 06:09:50.870667: step: 996/1541, loss: 1.5602446794509888 2023-01-23 06:09:51.800395: step: 998/1541, loss: 2.854649305343628 2023-01-23 06:09:52.714194: step: 1000/1541, loss: 1.1715010404586792 2023-01-23 06:09:53.634636: step: 1002/1541, loss: 0.7337331771850586 2023-01-23 06:09:54.529161: step: 1004/1541, loss: 0.9235069155693054 2023-01-23 06:09:55.410476: step: 1006/1541, loss: 1.170503854751587 2023-01-23 06:09:56.282253: step: 1008/1541, loss: 1.559706211090088 2023-01-23 06:09:57.144732: step: 1010/1541, loss: 0.8772782683372498 2023-01-23 06:09:58.057147: step: 1012/1541, loss: 0.6029947996139526 2023-01-23 06:09:58.952219: step: 1014/1541, loss: 1.118882656097412 2023-01-23 06:09:59.826653: step: 1016/1541, loss: 0.24913446605205536 2023-01-23 06:10:00.763424: step: 1018/1541, loss: 0.7699289321899414 2023-01-23 06:10:01.664929: step: 1020/1541, loss: 2.197374105453491 2023-01-23 06:10:02.519610: step: 1022/1541, loss: 0.4967978000640869 2023-01-23 06:10:03.385587: step: 1024/1541, loss: 1.108178734779358 2023-01-23 06:10:04.264947: step: 1026/1541, loss: 0.7110145092010498 2023-01-23 06:10:05.161300: step: 1028/1541, loss: 1.0849682092666626 2023-01-23 06:10:06.094106: step: 1030/1541, loss: 0.8573702573776245 2023-01-23 06:10:06.984983: step: 1032/1541, loss: 0.8483251333236694 2023-01-23 06:10:07.892904: step: 1034/1541, loss: 2.190706491470337 2023-01-23 06:10:08.774575: step: 1036/1541, loss: 0.9976227283477783 2023-01-23 06:10:09.702385: step: 1038/1541, loss: 1.0819528102874756 2023-01-23 06:10:10.615829: step: 1040/1541, loss: 3.558183431625366 2023-01-23 06:10:11.593250: step: 1042/1541, loss: 2.641399621963501 2023-01-23 06:10:12.475662: step: 1044/1541, loss: 0.46754032373428345 2023-01-23 06:10:13.379223: step: 1046/1541, loss: 2.3733456134796143 2023-01-23 06:10:14.340145: step: 1048/1541, loss: 0.37057796120643616 2023-01-23 06:10:15.259896: step: 1050/1541, loss: 0.6896893382072449 2023-01-23 06:10:16.181584: step: 1052/1541, loss: 1.115694522857666 2023-01-23 06:10:17.090442: step: 1054/1541, loss: 2.9349560737609863 2023-01-23 06:10:17.964927: step: 1056/1541, loss: 0.6680170297622681 2023-01-23 06:10:18.918464: step: 1058/1541, loss: 0.401197224855423 2023-01-23 06:10:19.798800: step: 1060/1541, loss: 0.8025256991386414 2023-01-23 06:10:20.736680: step: 1062/1541, loss: 0.6682007312774658 2023-01-23 06:10:21.612144: step: 1064/1541, loss: 0.6554346084594727 2023-01-23 06:10:22.534166: step: 1066/1541, loss: 9.171821594238281 2023-01-23 06:10:23.434896: step: 1068/1541, loss: 1.0463252067565918 2023-01-23 06:10:24.344759: step: 1070/1541, loss: 1.2036232948303223 2023-01-23 06:10:25.287775: step: 1072/1541, loss: 0.6416072845458984 2023-01-23 06:10:26.167034: step: 1074/1541, loss: 1.1839526891708374 2023-01-23 06:10:27.125585: step: 1076/1541, loss: 1.447993278503418 2023-01-23 06:10:28.046646: step: 1078/1541, loss: 0.6926154494285583 2023-01-23 06:10:29.050800: step: 1080/1541, loss: 0.7798793911933899 2023-01-23 06:10:29.979369: step: 1082/1541, loss: 0.991130530834198 2023-01-23 06:10:30.887723: step: 1084/1541, loss: 0.7321574687957764 2023-01-23 06:10:31.775732: step: 1086/1541, loss: 1.7539706230163574 2023-01-23 06:10:32.686190: step: 1088/1541, loss: 0.48982635140419006 2023-01-23 06:10:33.646745: step: 1090/1541, loss: 0.7913321852684021 2023-01-23 06:10:34.518020: step: 1092/1541, loss: 0.5210430026054382 2023-01-23 06:10:35.449787: step: 1094/1541, loss: 1.6258831024169922 2023-01-23 06:10:36.333894: step: 1096/1541, loss: 0.8300791382789612 2023-01-23 06:10:37.217688: step: 1098/1541, loss: 0.5212824940681458 2023-01-23 06:10:38.141255: step: 1100/1541, loss: 0.8813067078590393 2023-01-23 06:10:39.044583: step: 1102/1541, loss: 4.968353748321533 2023-01-23 06:10:39.976454: step: 1104/1541, loss: 1.9997648000717163 2023-01-23 06:10:40.989951: step: 1106/1541, loss: 1.6117804050445557 2023-01-23 06:10:41.891926: step: 1108/1541, loss: 0.7587181329727173 2023-01-23 06:10:42.823978: step: 1110/1541, loss: 3.7702414989471436 2023-01-23 06:10:43.688207: step: 1112/1541, loss: 0.9968839883804321 2023-01-23 06:10:44.603679: step: 1114/1541, loss: 0.8675002455711365 2023-01-23 06:10:45.470194: step: 1116/1541, loss: 1.290806770324707 2023-01-23 06:10:46.322839: step: 1118/1541, loss: 1.2626841068267822 2023-01-23 06:10:47.213889: step: 1120/1541, loss: 3.4838132858276367 2023-01-23 06:10:48.144873: step: 1122/1541, loss: 0.8629285097122192 2023-01-23 06:10:49.045347: step: 1124/1541, loss: 0.8146528005599976 2023-01-23 06:10:49.908454: step: 1126/1541, loss: 1.100506067276001 2023-01-23 06:10:50.801591: step: 1128/1541, loss: 0.9663857817649841 2023-01-23 06:10:51.723571: step: 1130/1541, loss: 0.8190878629684448 2023-01-23 06:10:52.645131: step: 1132/1541, loss: 1.8697410821914673 2023-01-23 06:10:53.571416: step: 1134/1541, loss: 0.9229376316070557 2023-01-23 06:10:54.461985: step: 1136/1541, loss: 0.8602582812309265 2023-01-23 06:10:55.357202: step: 1138/1541, loss: 0.9091413021087646 2023-01-23 06:10:56.217637: step: 1140/1541, loss: 0.6372926831245422 2023-01-23 06:10:57.129059: step: 1142/1541, loss: 0.9218734502792358 2023-01-23 06:10:58.048376: step: 1144/1541, loss: 0.47278091311454773 2023-01-23 06:10:58.919087: step: 1146/1541, loss: 1.4644460678100586 2023-01-23 06:10:59.874240: step: 1148/1541, loss: 0.7990533709526062 2023-01-23 06:11:00.752352: step: 1150/1541, loss: 0.3862702250480652 2023-01-23 06:11:01.661942: step: 1152/1541, loss: 3.008620262145996 2023-01-23 06:11:02.583141: step: 1154/1541, loss: 3.0851080417633057 2023-01-23 06:11:03.434762: step: 1156/1541, loss: 0.5873055458068848 2023-01-23 06:11:04.336106: step: 1158/1541, loss: 0.9877141118049622 2023-01-23 06:11:05.275284: step: 1160/1541, loss: 0.8488134741783142 2023-01-23 06:11:06.191855: step: 1162/1541, loss: 8.174722671508789 2023-01-23 06:11:07.134027: step: 1164/1541, loss: 0.7818748950958252 2023-01-23 06:11:08.114441: step: 1166/1541, loss: 1.4896891117095947 2023-01-23 06:11:09.043193: step: 1168/1541, loss: 0.8792564272880554 2023-01-23 06:11:09.938654: step: 1170/1541, loss: 1.1739306449890137 2023-01-23 06:11:10.800505: step: 1172/1541, loss: 0.6882426738739014 2023-01-23 06:11:11.718594: step: 1174/1541, loss: 1.0947675704956055 2023-01-23 06:11:12.634456: step: 1176/1541, loss: 0.7447579503059387 2023-01-23 06:11:13.502267: step: 1178/1541, loss: 1.0291908979415894 2023-01-23 06:11:14.382421: step: 1180/1541, loss: 0.41751518845558167 2023-01-23 06:11:15.288823: step: 1182/1541, loss: 0.7706080079078674 2023-01-23 06:11:16.145005: step: 1184/1541, loss: 0.7668555378913879 2023-01-23 06:11:17.002983: step: 1186/1541, loss: 1.8627022504806519 2023-01-23 06:11:17.864773: step: 1188/1541, loss: 0.7419856786727905 2023-01-23 06:11:18.762684: step: 1190/1541, loss: 3.0121512413024902 2023-01-23 06:11:19.633097: step: 1192/1541, loss: 0.6258103847503662 2023-01-23 06:11:20.543927: step: 1194/1541, loss: 0.7897453904151917 2023-01-23 06:11:21.506461: step: 1196/1541, loss: 2.0695807933807373 2023-01-23 06:11:22.408152: step: 1198/1541, loss: 0.47542399168014526 2023-01-23 06:11:23.322828: step: 1200/1541, loss: 3.1418824195861816 2023-01-23 06:11:24.198530: step: 1202/1541, loss: 0.4858379364013672 2023-01-23 06:11:25.145657: step: 1204/1541, loss: 1.9393789768218994 2023-01-23 06:11:26.018991: step: 1206/1541, loss: 1.1152236461639404 2023-01-23 06:11:26.953299: step: 1208/1541, loss: 0.49129122495651245 2023-01-23 06:11:27.867706: step: 1210/1541, loss: 0.590606153011322 2023-01-23 06:11:28.722899: step: 1212/1541, loss: 1.3564203977584839 2023-01-23 06:11:29.633184: step: 1214/1541, loss: 0.4764903783798218 2023-01-23 06:11:30.496644: step: 1216/1541, loss: 0.6238796710968018 2023-01-23 06:11:31.398247: step: 1218/1541, loss: 1.6751925945281982 2023-01-23 06:11:32.338477: step: 1220/1541, loss: 0.599242091178894 2023-01-23 06:11:33.200868: step: 1222/1541, loss: 1.117151141166687 2023-01-23 06:11:34.105149: step: 1224/1541, loss: 0.8679577112197876 2023-01-23 06:11:34.987264: step: 1226/1541, loss: 1.3149783611297607 2023-01-23 06:11:35.862330: step: 1228/1541, loss: 0.4735216796398163 2023-01-23 06:11:36.699864: step: 1230/1541, loss: 3.0148696899414062 2023-01-23 06:11:37.702679: step: 1232/1541, loss: 0.5906912684440613 2023-01-23 06:11:38.577893: step: 1234/1541, loss: 0.6312682628631592 2023-01-23 06:11:39.547795: step: 1236/1541, loss: 0.6938125491142273 2023-01-23 06:11:40.421958: step: 1238/1541, loss: 0.6046555638313293 2023-01-23 06:11:41.390446: step: 1240/1541, loss: 1.3758453130722046 2023-01-23 06:11:42.305062: step: 1242/1541, loss: 1.1054795980453491 2023-01-23 06:11:43.245587: step: 1244/1541, loss: 7.556565761566162 2023-01-23 06:11:44.179442: step: 1246/1541, loss: 5.234997749328613 2023-01-23 06:11:45.125780: step: 1248/1541, loss: 1.038024663925171 2023-01-23 06:11:45.993355: step: 1250/1541, loss: 0.8917043805122375 2023-01-23 06:11:46.858374: step: 1252/1541, loss: 0.8787680864334106 2023-01-23 06:11:47.727841: step: 1254/1541, loss: 1.34928560256958 2023-01-23 06:11:48.612949: step: 1256/1541, loss: 3.317173719406128 2023-01-23 06:11:49.527252: step: 1258/1541, loss: 1.0242866277694702 2023-01-23 06:11:50.490220: step: 1260/1541, loss: 3.262087345123291 2023-01-23 06:11:51.424639: step: 1262/1541, loss: 0.5540900826454163 2023-01-23 06:11:52.306331: step: 1264/1541, loss: 0.5816890001296997 2023-01-23 06:11:53.216026: step: 1266/1541, loss: 1.0150407552719116 2023-01-23 06:11:54.108903: step: 1268/1541, loss: 0.8962600827217102 2023-01-23 06:11:55.014569: step: 1270/1541, loss: 0.8153046369552612 2023-01-23 06:11:55.875829: step: 1272/1541, loss: 1.192114233970642 2023-01-23 06:11:56.794769: step: 1274/1541, loss: 1.2348363399505615 2023-01-23 06:11:57.787790: step: 1276/1541, loss: 1.4182263612747192 2023-01-23 06:11:58.652789: step: 1278/1541, loss: 3.22055983543396 2023-01-23 06:11:59.582645: step: 1280/1541, loss: 0.7487128973007202 2023-01-23 06:12:00.496093: step: 1282/1541, loss: 1.51981782913208 2023-01-23 06:12:01.401054: step: 1284/1541, loss: 1.1758719682693481 2023-01-23 06:12:02.272722: step: 1286/1541, loss: 0.690365195274353 2023-01-23 06:12:03.192733: step: 1288/1541, loss: 14.346052169799805 2023-01-23 06:12:04.055435: step: 1290/1541, loss: 1.3855336904525757 2023-01-23 06:12:04.971690: step: 1292/1541, loss: 2.2566003799438477 2023-01-23 06:12:05.883453: step: 1294/1541, loss: 0.8951109051704407 2023-01-23 06:12:06.808576: step: 1296/1541, loss: 1.289095163345337 2023-01-23 06:12:07.692000: step: 1298/1541, loss: 0.4299917221069336 2023-01-23 06:12:08.584069: step: 1300/1541, loss: 1.1136445999145508 2023-01-23 06:12:09.557766: step: 1302/1541, loss: 1.4252532720565796 2023-01-23 06:12:10.438197: step: 1304/1541, loss: 0.863063633441925 2023-01-23 06:12:11.323862: step: 1306/1541, loss: 2.9676878452301025 2023-01-23 06:12:12.237012: step: 1308/1541, loss: 0.7803908586502075 2023-01-23 06:12:13.109928: step: 1310/1541, loss: 3.2497124671936035 2023-01-23 06:12:14.001389: step: 1312/1541, loss: 0.8864841461181641 2023-01-23 06:12:14.887914: step: 1314/1541, loss: 0.8976959586143494 2023-01-23 06:12:15.799722: step: 1316/1541, loss: 0.6039854884147644 2023-01-23 06:12:16.754012: step: 1318/1541, loss: 0.6039818525314331 2023-01-23 06:12:17.680262: step: 1320/1541, loss: 0.6553876996040344 2023-01-23 06:12:18.588479: step: 1322/1541, loss: 0.6739190816879272 2023-01-23 06:12:19.464122: step: 1324/1541, loss: 2.830360174179077 2023-01-23 06:12:20.459064: step: 1326/1541, loss: 0.6210525631904602 2023-01-23 06:12:21.335836: step: 1328/1541, loss: 0.6763706207275391 2023-01-23 06:12:22.240986: step: 1330/1541, loss: 1.1437785625457764 2023-01-23 06:12:23.125319: step: 1332/1541, loss: 0.9675253033638 2023-01-23 06:12:24.081964: step: 1334/1541, loss: 1.293861746788025 2023-01-23 06:12:24.971147: step: 1336/1541, loss: 1.0802596807479858 2023-01-23 06:12:25.839606: step: 1338/1541, loss: 0.5853816270828247 2023-01-23 06:12:26.714462: step: 1340/1541, loss: 1.082392930984497 2023-01-23 06:12:27.606577: step: 1342/1541, loss: 2.405494451522827 2023-01-23 06:12:28.525229: step: 1344/1541, loss: 4.021509647369385 2023-01-23 06:12:29.470902: step: 1346/1541, loss: 1.300338625907898 2023-01-23 06:12:30.351348: step: 1348/1541, loss: 7.6860480308532715 2023-01-23 06:12:31.264601: step: 1350/1541, loss: 1.9442931413650513 2023-01-23 06:12:32.256610: step: 1352/1541, loss: 1.380598545074463 2023-01-23 06:12:33.269283: step: 1354/1541, loss: 3.915172576904297 2023-01-23 06:12:34.163098: step: 1356/1541, loss: 3.173275947570801 2023-01-23 06:12:35.037171: step: 1358/1541, loss: 1.0074448585510254 2023-01-23 06:12:35.896228: step: 1360/1541, loss: 1.7189011573791504 2023-01-23 06:12:36.762669: step: 1362/1541, loss: 0.6120641231536865 2023-01-23 06:12:37.674503: step: 1364/1541, loss: 1.5192146301269531 2023-01-23 06:12:38.557613: step: 1366/1541, loss: 0.7642854452133179 2023-01-23 06:12:39.517095: step: 1368/1541, loss: 1.2906478643417358 2023-01-23 06:12:40.418503: step: 1370/1541, loss: 1.060886025428772 2023-01-23 06:12:41.382206: step: 1372/1541, loss: 1.2862917184829712 2023-01-23 06:12:42.245816: step: 1374/1541, loss: 1.248691439628601 2023-01-23 06:12:43.227668: step: 1376/1541, loss: 0.6748846769332886 2023-01-23 06:12:44.124271: step: 1378/1541, loss: 0.7523787617683411 2023-01-23 06:12:45.007183: step: 1380/1541, loss: 0.6007325649261475 2023-01-23 06:12:45.872247: step: 1382/1541, loss: 0.6456117630004883 2023-01-23 06:12:46.838760: step: 1384/1541, loss: 0.8631650805473328 2023-01-23 06:12:47.754230: step: 1386/1541, loss: 1.9893641471862793 2023-01-23 06:12:48.605973: step: 1388/1541, loss: 0.8679466843605042 2023-01-23 06:12:49.513746: step: 1390/1541, loss: 0.6742953658103943 2023-01-23 06:12:50.401850: step: 1392/1541, loss: 0.48400795459747314 2023-01-23 06:12:51.328898: step: 1394/1541, loss: 3.123619318008423 2023-01-23 06:12:52.209229: step: 1396/1541, loss: 2.9791572093963623 2023-01-23 06:12:53.102944: step: 1398/1541, loss: 0.8635810017585754 2023-01-23 06:12:54.031640: step: 1400/1541, loss: 0.9626864194869995 2023-01-23 06:12:54.912876: step: 1402/1541, loss: 0.5839535593986511 2023-01-23 06:12:55.797676: step: 1404/1541, loss: 0.5599238872528076 2023-01-23 06:12:56.715283: step: 1406/1541, loss: 1.0835529565811157 2023-01-23 06:12:57.618516: step: 1408/1541, loss: 1.0644500255584717 2023-01-23 06:12:58.669498: step: 1410/1541, loss: 3.1401000022888184 2023-01-23 06:12:59.554542: step: 1412/1541, loss: 0.3836117088794708 2023-01-23 06:13:00.446090: step: 1414/1541, loss: 0.6977033615112305 2023-01-23 06:13:01.380620: step: 1416/1541, loss: 1.412029504776001 2023-01-23 06:13:02.250221: step: 1418/1541, loss: 0.5503904819488525 2023-01-23 06:13:03.196573: step: 1420/1541, loss: 0.362935334444046 2023-01-23 06:13:04.060462: step: 1422/1541, loss: 5.734084129333496 2023-01-23 06:13:04.980813: step: 1424/1541, loss: 0.8867378830909729 2023-01-23 06:13:05.857581: step: 1426/1541, loss: 1.0602189302444458 2023-01-23 06:13:06.745453: step: 1428/1541, loss: 1.544142246246338 2023-01-23 06:13:07.631129: step: 1430/1541, loss: 3.274991273880005 2023-01-23 06:13:08.538091: step: 1432/1541, loss: 3.7779293060302734 2023-01-23 06:13:09.402740: step: 1434/1541, loss: 1.2960580587387085 2023-01-23 06:13:10.313108: step: 1436/1541, loss: 0.8041568994522095 2023-01-23 06:13:11.209764: step: 1438/1541, loss: 0.8214539885520935 2023-01-23 06:13:12.057729: step: 1440/1541, loss: 0.6081314086914062 2023-01-23 06:13:13.008209: step: 1442/1541, loss: 2.1610233783721924 2023-01-23 06:13:13.890914: step: 1444/1541, loss: 1.2803657054901123 2023-01-23 06:13:14.740513: step: 1446/1541, loss: 0.4119725823402405 2023-01-23 06:13:15.654045: step: 1448/1541, loss: 1.6944278478622437 2023-01-23 06:13:16.511677: step: 1450/1541, loss: 0.6437435150146484 2023-01-23 06:13:17.431549: step: 1452/1541, loss: 3.0868561267852783 2023-01-23 06:13:18.316343: step: 1454/1541, loss: 0.5927602052688599 2023-01-23 06:13:19.269238: step: 1456/1541, loss: 1.0824534893035889 2023-01-23 06:13:20.163351: step: 1458/1541, loss: 1.779234766960144 2023-01-23 06:13:21.046508: step: 1460/1541, loss: 0.8162235021591187 2023-01-23 06:13:21.940985: step: 1462/1541, loss: 0.9963820576667786 2023-01-23 06:13:22.900566: step: 1464/1541, loss: 0.9694298505783081 2023-01-23 06:13:23.781934: step: 1466/1541, loss: 1.1648634672164917 2023-01-23 06:13:24.688913: step: 1468/1541, loss: 3.602691411972046 2023-01-23 06:13:25.567233: step: 1470/1541, loss: 5.666265964508057 2023-01-23 06:13:26.452586: step: 1472/1541, loss: 1.2495273351669312 2023-01-23 06:13:27.296852: step: 1474/1541, loss: 0.5939812660217285 2023-01-23 06:13:28.161079: step: 1476/1541, loss: 2.360938549041748 2023-01-23 06:13:29.057570: step: 1478/1541, loss: 0.9161884188652039 2023-01-23 06:13:29.965864: step: 1480/1541, loss: 0.9282068610191345 2023-01-23 06:13:30.900123: step: 1482/1541, loss: 0.435289204120636 2023-01-23 06:13:31.761901: step: 1484/1541, loss: 0.8980650901794434 2023-01-23 06:13:32.641706: step: 1486/1541, loss: 7.211140155792236 2023-01-23 06:13:33.570715: step: 1488/1541, loss: 1.0111594200134277 2023-01-23 06:13:34.437867: step: 1490/1541, loss: 0.942772388458252 2023-01-23 06:13:35.402486: step: 1492/1541, loss: 1.5500329732894897 2023-01-23 06:13:36.330892: step: 1494/1541, loss: 0.8187576532363892 2023-01-23 06:13:37.284094: step: 1496/1541, loss: 0.6806201338768005 2023-01-23 06:13:38.174080: step: 1498/1541, loss: 2.9949121475219727 2023-01-23 06:13:39.103744: step: 1500/1541, loss: 0.5140451788902283 2023-01-23 06:13:40.045903: step: 1502/1541, loss: 0.27253150939941406 2023-01-23 06:13:40.986465: step: 1504/1541, loss: 0.46596601605415344 2023-01-23 06:13:41.839163: step: 1506/1541, loss: 1.230682611465454 2023-01-23 06:13:42.795052: step: 1508/1541, loss: 0.5805516839027405 2023-01-23 06:13:43.671125: step: 1510/1541, loss: 0.6395880579948425 2023-01-23 06:13:44.539354: step: 1512/1541, loss: 0.5983230471611023 2023-01-23 06:13:45.428820: step: 1514/1541, loss: 0.6514688730239868 2023-01-23 06:13:46.328285: step: 1516/1541, loss: 1.3322347402572632 2023-01-23 06:13:47.206808: step: 1518/1541, loss: 0.7709673643112183 2023-01-23 06:13:48.103611: step: 1520/1541, loss: 9.047223091125488 2023-01-23 06:13:48.984425: step: 1522/1541, loss: 0.31272679567337036 2023-01-23 06:13:49.920146: step: 1524/1541, loss: 1.3886469602584839 2023-01-23 06:13:50.825292: step: 1526/1541, loss: 0.7729814052581787 2023-01-23 06:13:51.703385: step: 1528/1541, loss: 0.6815182566642761 2023-01-23 06:13:52.586082: step: 1530/1541, loss: 7.187314987182617 2023-01-23 06:13:53.474753: step: 1532/1541, loss: 1.168449878692627 2023-01-23 06:13:54.372050: step: 1534/1541, loss: 2.9429514408111572 2023-01-23 06:13:55.314720: step: 1536/1541, loss: 0.448452889919281 2023-01-23 06:13:56.224390: step: 1538/1541, loss: 0.7494457960128784 2023-01-23 06:13:57.122765: step: 1540/1541, loss: 0.8144969344139099 2023-01-23 06:13:57.993525: step: 1542/1541, loss: 0.659684419631958 2023-01-23 06:13:58.895326: step: 1544/1541, loss: 0.8729724287986755 2023-01-23 06:13:59.758021: step: 1546/1541, loss: 0.8085091710090637 2023-01-23 06:14:00.689477: step: 1548/1541, loss: 0.6519178748130798 2023-01-23 06:14:01.581506: step: 1550/1541, loss: 0.7470054626464844 2023-01-23 06:14:02.524790: step: 1552/1541, loss: 1.9450386762619019 2023-01-23 06:14:03.411570: step: 1554/1541, loss: 0.8136687278747559 2023-01-23 06:14:04.284647: step: 1556/1541, loss: 1.091008186340332 2023-01-23 06:14:05.170148: step: 1558/1541, loss: 2.97301983833313 2023-01-23 06:14:06.093204: step: 1560/1541, loss: 0.9613404273986816 2023-01-23 06:14:07.097246: step: 1562/1541, loss: 3.755885124206543 2023-01-23 06:14:08.007415: step: 1564/1541, loss: 0.6392372250556946 2023-01-23 06:14:08.906317: step: 1566/1541, loss: 0.6175170540809631 2023-01-23 06:14:09.812938: step: 1568/1541, loss: 0.7201215028762817 2023-01-23 06:14:10.720195: step: 1570/1541, loss: 7.281790733337402 2023-01-23 06:14:11.601072: step: 1572/1541, loss: 0.4122826159000397 2023-01-23 06:14:12.492884: step: 1574/1541, loss: 0.7176757454872131 2023-01-23 06:14:13.364682: step: 1576/1541, loss: 5.09929084777832 2023-01-23 06:14:14.276890: step: 1578/1541, loss: 1.16304612159729 2023-01-23 06:14:15.151402: step: 1580/1541, loss: 0.5335566401481628 2023-01-23 06:14:16.061811: step: 1582/1541, loss: 0.5986900925636292 2023-01-23 06:14:16.947015: step: 1584/1541, loss: 1.5647543668746948 2023-01-23 06:14:17.944888: step: 1586/1541, loss: 1.3765051364898682 2023-01-23 06:14:18.840806: step: 1588/1541, loss: 0.8396426439285278 2023-01-23 06:14:19.772122: step: 1590/1541, loss: 0.8885896801948547 2023-01-23 06:14:20.659037: step: 1592/1541, loss: 1.0683884620666504 2023-01-23 06:14:21.545443: step: 1594/1541, loss: 3.210847854614258 2023-01-23 06:14:22.432570: step: 1596/1541, loss: 0.6198535561561584 2023-01-23 06:14:23.383340: step: 1598/1541, loss: 1.180993914604187 2023-01-23 06:14:24.332265: step: 1600/1541, loss: 0.33340123295783997 2023-01-23 06:14:25.210072: step: 1602/1541, loss: 3.003649950027466 2023-01-23 06:14:26.126255: step: 1604/1541, loss: 0.2149377018213272 2023-01-23 06:14:27.028135: step: 1606/1541, loss: 0.6532079577445984 2023-01-23 06:14:27.909642: step: 1608/1541, loss: 7.59411096572876 2023-01-23 06:14:28.832916: step: 1610/1541, loss: 0.9962292909622192 2023-01-23 06:14:29.777838: step: 1612/1541, loss: 5.205307960510254 2023-01-23 06:14:30.672998: step: 1614/1541, loss: 3.735288381576538 2023-01-23 06:14:31.623689: step: 1616/1541, loss: 0.4613346457481384 2023-01-23 06:14:32.473533: step: 1618/1541, loss: 1.5957118272781372 2023-01-23 06:14:33.381494: step: 1620/1541, loss: 1.7394417524337769 2023-01-23 06:14:34.265271: step: 1622/1541, loss: 0.8810384273529053 2023-01-23 06:14:35.200756: step: 1624/1541, loss: 0.896928608417511 2023-01-23 06:14:36.121936: step: 1626/1541, loss: 1.2674888372421265 2023-01-23 06:14:37.094335: step: 1628/1541, loss: 0.8771577477455139 2023-01-23 06:14:37.981776: step: 1630/1541, loss: 0.5855873823165894 2023-01-23 06:14:38.911913: step: 1632/1541, loss: 0.6544532179832458 2023-01-23 06:14:39.807806: step: 1634/1541, loss: 0.6620293259620667 2023-01-23 06:14:40.688239: step: 1636/1541, loss: 1.0076371431350708 2023-01-23 06:14:41.555419: step: 1638/1541, loss: 0.6409502625465393 2023-01-23 06:14:42.432267: step: 1640/1541, loss: 3.0894057750701904 2023-01-23 06:14:43.299221: step: 1642/1541, loss: 0.46290937066078186 2023-01-23 06:14:44.236828: step: 1644/1541, loss: 1.0313587188720703 2023-01-23 06:14:45.136695: step: 1646/1541, loss: 9.269983291625977 2023-01-23 06:14:46.022731: step: 1648/1541, loss: 0.8975704908370972 2023-01-23 06:14:46.952098: step: 1650/1541, loss: 1.2981914281845093 2023-01-23 06:14:47.846776: step: 1652/1541, loss: 1.0794116258621216 2023-01-23 06:14:48.748155: step: 1654/1541, loss: 1.0266178846359253 2023-01-23 06:14:49.695157: step: 1656/1541, loss: 1.5780915021896362 2023-01-23 06:14:50.628035: step: 1658/1541, loss: 0.5186257362365723 2023-01-23 06:14:51.551067: step: 1660/1541, loss: 0.7121044993400574 2023-01-23 06:14:52.426951: step: 1662/1541, loss: 3.1549630165100098 2023-01-23 06:14:53.308086: step: 1664/1541, loss: 1.6386936902999878 2023-01-23 06:14:54.188317: step: 1666/1541, loss: 0.5217248201370239 2023-01-23 06:14:55.062587: step: 1668/1541, loss: 0.9214282631874084 2023-01-23 06:14:55.992352: step: 1670/1541, loss: 0.4984312057495117 2023-01-23 06:14:56.883747: step: 1672/1541, loss: 0.5527347922325134 2023-01-23 06:14:57.772444: step: 1674/1541, loss: 0.4257009029388428 2023-01-23 06:14:58.662943: step: 1676/1541, loss: 0.8329899311065674 2023-01-23 06:14:59.539584: step: 1678/1541, loss: 0.6854647397994995 2023-01-23 06:15:00.410265: step: 1680/1541, loss: 1.5190309286117554 2023-01-23 06:15:01.306190: step: 1682/1541, loss: 0.6530845761299133 2023-01-23 06:15:02.232767: step: 1684/1541, loss: 0.7268908619880676 2023-01-23 06:15:03.151167: step: 1686/1541, loss: 1.3738822937011719 2023-01-23 06:15:04.045215: step: 1688/1541, loss: 0.9305717945098877 2023-01-23 06:15:04.944933: step: 1690/1541, loss: 0.7508413791656494 2023-01-23 06:15:05.841236: step: 1692/1541, loss: 0.6232801079750061 2023-01-23 06:15:06.785978: step: 1694/1541, loss: 0.5209993124008179 2023-01-23 06:15:07.737141: step: 1696/1541, loss: 3.4478933811187744 2023-01-23 06:15:08.624706: step: 1698/1541, loss: 0.8642138838768005 2023-01-23 06:15:09.502566: step: 1700/1541, loss: 6.6029229164123535 2023-01-23 06:15:10.409326: step: 1702/1541, loss: 0.754277229309082 2023-01-23 06:15:11.359369: step: 1704/1541, loss: 1.066371202468872 2023-01-23 06:15:12.258337: step: 1706/1541, loss: 1.053605079650879 2023-01-23 06:15:13.158691: step: 1708/1541, loss: 2.822681427001953 2023-01-23 06:15:14.172240: step: 1710/1541, loss: 1.3750109672546387 2023-01-23 06:15:15.107059: step: 1712/1541, loss: 0.7877618670463562 2023-01-23 06:15:16.014195: step: 1714/1541, loss: 1.127634882926941 2023-01-23 06:15:16.877459: step: 1716/1541, loss: 0.9881335496902466 2023-01-23 06:15:17.779373: step: 1718/1541, loss: 0.5292832255363464 2023-01-23 06:15:18.693030: step: 1720/1541, loss: 1.9245182275772095 2023-01-23 06:15:19.626775: step: 1722/1541, loss: 0.4839951992034912 2023-01-23 06:15:20.508347: step: 1724/1541, loss: 0.8021033406257629 2023-01-23 06:15:21.410983: step: 1726/1541, loss: 1.5288969278335571 2023-01-23 06:15:22.342343: step: 1728/1541, loss: 2.118170976638794 2023-01-23 06:15:23.279585: step: 1730/1541, loss: 0.3775020241737366 2023-01-23 06:15:24.153915: step: 1732/1541, loss: 1.1277945041656494 2023-01-23 06:15:25.055198: step: 1734/1541, loss: 0.5241987109184265 2023-01-23 06:15:25.940401: step: 1736/1541, loss: 1.0300315618515015 2023-01-23 06:15:26.797211: step: 1738/1541, loss: 3.1917335987091064 2023-01-23 06:15:27.659809: step: 1740/1541, loss: 0.9023053050041199 2023-01-23 06:15:28.594654: step: 1742/1541, loss: 1.314857840538025 2023-01-23 06:15:29.512480: step: 1744/1541, loss: 1.2829909324645996 2023-01-23 06:15:30.394371: step: 1746/1541, loss: 3.6986734867095947 2023-01-23 06:15:31.334097: step: 1748/1541, loss: 0.7723771333694458 2023-01-23 06:15:32.242987: step: 1750/1541, loss: 6.877243995666504 2023-01-23 06:15:33.156391: step: 1752/1541, loss: 1.7521421909332275 2023-01-23 06:15:34.012967: step: 1754/1541, loss: 0.9824827313423157 2023-01-23 06:15:34.898416: step: 1756/1541, loss: 0.8734093308448792 2023-01-23 06:15:35.837273: step: 1758/1541, loss: 1.228522539138794 2023-01-23 06:15:36.755904: step: 1760/1541, loss: 0.45291244983673096 2023-01-23 06:15:37.632876: step: 1762/1541, loss: 0.5094897747039795 2023-01-23 06:15:38.558625: step: 1764/1541, loss: 0.5580121278762817 2023-01-23 06:15:39.427799: step: 1766/1541, loss: 0.47305774688720703 2023-01-23 06:15:40.268153: step: 1768/1541, loss: 0.42592519521713257 2023-01-23 06:15:41.217950: step: 1770/1541, loss: 0.8224814534187317 2023-01-23 06:15:42.211322: step: 1772/1541, loss: 5.865386962890625 2023-01-23 06:15:43.107720: step: 1774/1541, loss: 3.2363834381103516 2023-01-23 06:15:44.027784: step: 1776/1541, loss: 1.9721297025680542 2023-01-23 06:15:44.933926: step: 1778/1541, loss: 1.4852701425552368 2023-01-23 06:15:45.818111: step: 1780/1541, loss: 1.0170552730560303 2023-01-23 06:15:46.715290: step: 1782/1541, loss: 0.6750995516777039 2023-01-23 06:15:47.640952: step: 1784/1541, loss: 1.4547079801559448 2023-01-23 06:15:48.528599: step: 1786/1541, loss: 0.32594218850135803 2023-01-23 06:15:49.434854: step: 1788/1541, loss: 2.1291356086730957 2023-01-23 06:15:50.327623: step: 1790/1541, loss: 0.5509874224662781 2023-01-23 06:15:51.241976: step: 1792/1541, loss: 1.3738166093826294 2023-01-23 06:15:52.179624: step: 1794/1541, loss: 0.48137176036834717 2023-01-23 06:15:53.060121: step: 1796/1541, loss: 0.9927275776863098 2023-01-23 06:15:53.946153: step: 1798/1541, loss: 0.6558511257171631 2023-01-23 06:15:54.808299: step: 1800/1541, loss: 1.577419400215149 2023-01-23 06:15:55.717536: step: 1802/1541, loss: 0.5149428248405457 2023-01-23 06:15:56.654745: step: 1804/1541, loss: 0.3644672632217407 2023-01-23 06:15:57.537896: step: 1806/1541, loss: 0.5022827386856079 2023-01-23 06:15:58.443025: step: 1808/1541, loss: 0.7800127267837524 2023-01-23 06:15:59.358783: step: 1810/1541, loss: 0.7936766147613525 2023-01-23 06:16:00.307637: step: 1812/1541, loss: 2.715399742126465 2023-01-23 06:16:01.212519: step: 1814/1541, loss: 0.5094972252845764 2023-01-23 06:16:02.079934: step: 1816/1541, loss: 0.8606261014938354 2023-01-23 06:16:02.945131: step: 1818/1541, loss: 1.3666244745254517 2023-01-23 06:16:03.810315: step: 1820/1541, loss: 1.490639090538025 2023-01-23 06:16:04.721846: step: 1822/1541, loss: 0.6241985559463501 2023-01-23 06:16:05.609304: step: 1824/1541, loss: 1.1701509952545166 2023-01-23 06:16:06.501760: step: 1826/1541, loss: 0.6400866508483887 2023-01-23 06:16:07.424027: step: 1828/1541, loss: 3.1889302730560303 2023-01-23 06:16:08.309978: step: 1830/1541, loss: 0.6355395913124084 2023-01-23 06:16:09.176479: step: 1832/1541, loss: 0.5675042867660522 2023-01-23 06:16:10.111946: step: 1834/1541, loss: 0.5969852209091187 2023-01-23 06:16:10.960225: step: 1836/1541, loss: 0.8585351705551147 2023-01-23 06:16:11.863987: step: 1838/1541, loss: 3.372614622116089 2023-01-23 06:16:12.728663: step: 1840/1541, loss: 3.069138288497925 2023-01-23 06:16:13.613732: step: 1842/1541, loss: 0.7614673972129822 2023-01-23 06:16:14.484128: step: 1844/1541, loss: 1.8831942081451416 2023-01-23 06:16:15.340154: step: 1846/1541, loss: 0.3231796324253082 2023-01-23 06:16:16.250828: step: 1848/1541, loss: 3.0233237743377686 2023-01-23 06:16:17.140492: step: 1850/1541, loss: 0.8084602355957031 2023-01-23 06:16:18.037787: step: 1852/1541, loss: 0.6058071851730347 2023-01-23 06:16:19.011745: step: 1854/1541, loss: 0.4291985034942627 2023-01-23 06:16:19.927609: step: 1856/1541, loss: 1.4880887269973755 2023-01-23 06:16:20.798087: step: 1858/1541, loss: 1.2882657051086426 2023-01-23 06:16:21.732022: step: 1860/1541, loss: 3.7441580295562744 2023-01-23 06:16:22.608566: step: 1862/1541, loss: 0.7468857169151306 2023-01-23 06:16:23.544351: step: 1864/1541, loss: 3.4009666442871094 2023-01-23 06:16:24.461900: step: 1866/1541, loss: 1.2112784385681152 2023-01-23 06:16:25.343152: step: 1868/1541, loss: 0.8301485776901245 2023-01-23 06:16:26.232596: step: 1870/1541, loss: 0.604171872138977 2023-01-23 06:16:27.115138: step: 1872/1541, loss: 0.2577320337295532 2023-01-23 06:16:28.017946: step: 1874/1541, loss: 0.6666606664657593 2023-01-23 06:16:28.912891: step: 1876/1541, loss: 2.3519535064697266 2023-01-23 06:16:29.807225: step: 1878/1541, loss: 0.7719713449478149 2023-01-23 06:16:30.663079: step: 1880/1541, loss: 2.8918726444244385 2023-01-23 06:16:31.558570: step: 1882/1541, loss: 0.8578299283981323 2023-01-23 06:16:32.469659: step: 1884/1541, loss: 1.0338979959487915 2023-01-23 06:16:33.403406: step: 1886/1541, loss: 0.8705443143844604 2023-01-23 06:16:34.384324: step: 1888/1541, loss: 1.360536813735962 2023-01-23 06:16:35.304260: step: 1890/1541, loss: 0.75165855884552 2023-01-23 06:16:36.201960: step: 1892/1541, loss: 0.6143703460693359 2023-01-23 06:16:37.072615: step: 1894/1541, loss: 3.6602978706359863 2023-01-23 06:16:37.968397: step: 1896/1541, loss: 0.5216072797775269 2023-01-23 06:16:38.852567: step: 1898/1541, loss: 3.3784754276275635 2023-01-23 06:16:39.725309: step: 1900/1541, loss: 0.5829849243164062 2023-01-23 06:16:40.651489: step: 1902/1541, loss: 2.778698682785034 2023-01-23 06:16:41.629873: step: 1904/1541, loss: 1.8020814657211304 2023-01-23 06:16:42.569563: step: 1906/1541, loss: 0.7111164927482605 2023-01-23 06:16:43.478627: step: 1908/1541, loss: 2.985992670059204 2023-01-23 06:16:44.345036: step: 1910/1541, loss: 1.1331230401992798 2023-01-23 06:16:45.290649: step: 1912/1541, loss: 1.42844557762146 2023-01-23 06:16:46.196749: step: 1914/1541, loss: 0.8505947589874268 2023-01-23 06:16:47.123323: step: 1916/1541, loss: 3.152423620223999 2023-01-23 06:16:47.974841: step: 1918/1541, loss: 0.5289524793624878 2023-01-23 06:16:48.878822: step: 1920/1541, loss: 0.30513352155685425 2023-01-23 06:16:49.803623: step: 1922/1541, loss: 0.9715765118598938 2023-01-23 06:16:50.669501: step: 1924/1541, loss: 0.7390958666801453 2023-01-23 06:16:51.579866: step: 1926/1541, loss: 1.0647401809692383 2023-01-23 06:16:52.447515: step: 1928/1541, loss: 0.7196833491325378 2023-01-23 06:16:53.358108: step: 1930/1541, loss: 3.4528636932373047 2023-01-23 06:16:54.247893: step: 1932/1541, loss: 0.8254364728927612 2023-01-23 06:16:55.148793: step: 1934/1541, loss: 1.0255893468856812 2023-01-23 06:16:56.019517: step: 1936/1541, loss: 0.7544925808906555 2023-01-23 06:16:56.911422: step: 1938/1541, loss: 0.9242674112319946 2023-01-23 06:16:57.816935: step: 1940/1541, loss: 3.000290632247925 2023-01-23 06:16:58.690577: step: 1942/1541, loss: 0.4949204623699188 2023-01-23 06:16:59.585845: step: 1944/1541, loss: 0.7313095331192017 2023-01-23 06:17:00.478982: step: 1946/1541, loss: 0.7328236103057861 2023-01-23 06:17:01.369999: step: 1948/1541, loss: 3.7993485927581787 2023-01-23 06:17:02.243968: step: 1950/1541, loss: 0.5847676396369934 2023-01-23 06:17:03.163089: step: 1952/1541, loss: 0.7119752168655396 2023-01-23 06:17:04.103777: step: 1954/1541, loss: 0.9368599057197571 2023-01-23 06:17:05.095285: step: 1956/1541, loss: 0.24538902938365936 2023-01-23 06:17:06.006307: step: 1958/1541, loss: 0.6376075148582458 2023-01-23 06:17:06.902058: step: 1960/1541, loss: 0.3509662449359894 2023-01-23 06:17:07.835847: step: 1962/1541, loss: 2.767871618270874 2023-01-23 06:17:08.850045: step: 1964/1541, loss: 0.6118468046188354 2023-01-23 06:17:09.772558: step: 1966/1541, loss: 2.920836925506592 2023-01-23 06:17:10.656968: step: 1968/1541, loss: 0.6454060077667236 2023-01-23 06:17:11.587026: step: 1970/1541, loss: 2.786980628967285 2023-01-23 06:17:12.467958: step: 1972/1541, loss: 1.3588521480560303 2023-01-23 06:17:13.340435: step: 1974/1541, loss: 0.3664551377296448 2023-01-23 06:17:14.216249: step: 1976/1541, loss: 1.7982879877090454 2023-01-23 06:17:15.148975: step: 1978/1541, loss: 2.8358726501464844 2023-01-23 06:17:16.026698: step: 1980/1541, loss: 0.9735183715820312 2023-01-23 06:17:16.905420: step: 1982/1541, loss: 0.6613637804985046 2023-01-23 06:17:17.845437: step: 1984/1541, loss: 1.6411877870559692 2023-01-23 06:17:18.713879: step: 1986/1541, loss: 0.8061266541481018 2023-01-23 06:17:19.687806: step: 1988/1541, loss: 0.9178816676139832 2023-01-23 06:17:20.563541: step: 1990/1541, loss: 0.3969566226005554 2023-01-23 06:17:21.474125: step: 1992/1541, loss: 1.0270793437957764 2023-01-23 06:17:22.372135: step: 1994/1541, loss: 0.7578715682029724 2023-01-23 06:17:23.239138: step: 1996/1541, loss: 6.656989574432373 2023-01-23 06:17:24.158956: step: 1998/1541, loss: 0.9444878697395325 2023-01-23 06:17:25.058674: step: 2000/1541, loss: 5.674219608306885 2023-01-23 06:17:25.929563: step: 2002/1541, loss: 1.0059316158294678 2023-01-23 06:17:26.851535: step: 2004/1541, loss: 1.4627747535705566 2023-01-23 06:17:27.705177: step: 2006/1541, loss: 1.3392294645309448 2023-01-23 06:17:28.650720: step: 2008/1541, loss: 1.1639821529388428 2023-01-23 06:17:29.518704: step: 2010/1541, loss: 1.3365882635116577 2023-01-23 06:17:30.412061: step: 2012/1541, loss: 0.9313067197799683 2023-01-23 06:17:31.313486: step: 2014/1541, loss: 6.571728229522705 2023-01-23 06:17:32.213395: step: 2016/1541, loss: 0.45108750462532043 2023-01-23 06:17:33.151532: step: 2018/1541, loss: 0.7935319542884827 2023-01-23 06:17:34.013693: step: 2020/1541, loss: 0.6969478726387024 2023-01-23 06:17:34.914438: step: 2022/1541, loss: 0.5584802627563477 2023-01-23 06:17:35.820016: step: 2024/1541, loss: 0.3569866716861725 2023-01-23 06:17:36.707331: step: 2026/1541, loss: 3.542849063873291 2023-01-23 06:17:37.606045: step: 2028/1541, loss: 1.2561486959457397 2023-01-23 06:17:38.491169: step: 2030/1541, loss: 1.364909291267395 2023-01-23 06:17:39.398593: step: 2032/1541, loss: 0.8116141557693481 2023-01-23 06:17:40.308041: step: 2034/1541, loss: 0.6361746191978455 2023-01-23 06:17:41.201378: step: 2036/1541, loss: 3.475182056427002 2023-01-23 06:17:42.094494: step: 2038/1541, loss: 0.797814667224884 2023-01-23 06:17:42.973264: step: 2040/1541, loss: 0.532953679561615 2023-01-23 06:17:43.876055: step: 2042/1541, loss: 0.16963006556034088 2023-01-23 06:17:44.782863: step: 2044/1541, loss: 0.5394595265388489 2023-01-23 06:17:45.646311: step: 2046/1541, loss: 0.891075074672699 2023-01-23 06:17:46.526906: step: 2048/1541, loss: 0.2592358887195587 2023-01-23 06:17:47.400490: step: 2050/1541, loss: 1.2335379123687744 2023-01-23 06:17:48.325920: step: 2052/1541, loss: 0.7309386730194092 2023-01-23 06:17:49.225292: step: 2054/1541, loss: 0.6026654243469238 2023-01-23 06:17:50.085486: step: 2056/1541, loss: 0.42177534103393555 2023-01-23 06:17:51.000867: step: 2058/1541, loss: 3.0217068195343018 2023-01-23 06:17:51.908958: step: 2060/1541, loss: 0.7411055564880371 2023-01-23 06:17:52.782197: step: 2062/1541, loss: 1.3609685897827148 2023-01-23 06:17:53.673196: step: 2064/1541, loss: 0.49907368421554565 2023-01-23 06:17:54.576194: step: 2066/1541, loss: 0.4236309826374054 2023-01-23 06:17:55.495162: step: 2068/1541, loss: 0.6542349457740784 2023-01-23 06:17:56.355885: step: 2070/1541, loss: 0.8788590431213379 2023-01-23 06:17:57.257327: step: 2072/1541, loss: 6.566308975219727 2023-01-23 06:17:58.156342: step: 2074/1541, loss: 5.825507640838623 2023-01-23 06:17:59.031226: step: 2076/1541, loss: 0.31090328097343445 2023-01-23 06:17:59.894335: step: 2078/1541, loss: 0.7113879323005676 2023-01-23 06:18:00.770993: step: 2080/1541, loss: 3.2432918548583984 2023-01-23 06:18:01.653547: step: 2082/1541, loss: 1.037372350692749 2023-01-23 06:18:02.593235: step: 2084/1541, loss: 3.4522910118103027 2023-01-23 06:18:03.504774: step: 2086/1541, loss: 3.0760862827301025 2023-01-23 06:18:04.410983: step: 2088/1541, loss: 0.27422645688056946 2023-01-23 06:18:05.320650: step: 2090/1541, loss: 0.735831081867218 2023-01-23 06:18:06.177378: step: 2092/1541, loss: 1.4952483177185059 2023-01-23 06:18:07.040773: step: 2094/1541, loss: 0.5907489061355591 2023-01-23 06:18:07.898060: step: 2096/1541, loss: 0.7328497767448425 2023-01-23 06:18:08.791475: step: 2098/1541, loss: 1.0623048543930054 2023-01-23 06:18:09.689993: step: 2100/1541, loss: 0.6335358619689941 2023-01-23 06:18:10.632260: step: 2102/1541, loss: 0.6029292941093445 2023-01-23 06:18:11.477239: step: 2104/1541, loss: 0.5837475657463074 2023-01-23 06:18:12.352464: step: 2106/1541, loss: 0.6255063414573669 2023-01-23 06:18:13.244272: step: 2108/1541, loss: 0.7058619856834412 2023-01-23 06:18:14.130439: step: 2110/1541, loss: 6.705358505249023 2023-01-23 06:18:15.020314: step: 2112/1541, loss: 0.8114763498306274 2023-01-23 06:18:15.921822: step: 2114/1541, loss: 1.0006277561187744 2023-01-23 06:18:16.822575: step: 2116/1541, loss: 1.1135385036468506 2023-01-23 06:18:17.699473: step: 2118/1541, loss: 0.6821252703666687 2023-01-23 06:18:18.682211: step: 2120/1541, loss: 0.950761616230011 2023-01-23 06:18:19.592215: step: 2122/1541, loss: 0.7364336252212524 2023-01-23 06:18:20.506359: step: 2124/1541, loss: 1.3690991401672363 2023-01-23 06:18:21.425648: step: 2126/1541, loss: 0.7482011914253235 2023-01-23 06:18:22.364491: step: 2128/1541, loss: 0.9864398241043091 2023-01-23 06:18:23.230176: step: 2130/1541, loss: 1.9938374757766724 2023-01-23 06:18:24.134205: step: 2132/1541, loss: 0.8848446607589722 2023-01-23 06:18:25.030419: step: 2134/1541, loss: 7.645805835723877 2023-01-23 06:18:25.910911: step: 2136/1541, loss: 0.9412444829940796 2023-01-23 06:18:26.771528: step: 2138/1541, loss: 2.964768886566162 2023-01-23 06:18:27.682756: step: 2140/1541, loss: 1.0336576700210571 2023-01-23 06:18:28.599036: step: 2142/1541, loss: 0.9611425399780273 2023-01-23 06:18:29.510190: step: 2144/1541, loss: 3.604565143585205 2023-01-23 06:18:30.381948: step: 2146/1541, loss: 0.9795684218406677 2023-01-23 06:18:31.287621: step: 2148/1541, loss: 3.828861713409424 2023-01-23 06:18:32.252555: step: 2150/1541, loss: 1.211870551109314 2023-01-23 06:18:33.156024: step: 2152/1541, loss: 3.399214029312134 2023-01-23 06:18:34.016351: step: 2154/1541, loss: 0.5332638621330261 2023-01-23 06:18:34.907437: step: 2156/1541, loss: 3.763561487197876 2023-01-23 06:18:35.805390: step: 2158/1541, loss: 1.1018357276916504 2023-01-23 06:18:36.795957: step: 2160/1541, loss: 0.37316280603408813 2023-01-23 06:18:37.691617: step: 2162/1541, loss: 1.1314542293548584 2023-01-23 06:18:38.608685: step: 2164/1541, loss: 0.7833516001701355 2023-01-23 06:18:39.551227: step: 2166/1541, loss: 0.5560997128486633 2023-01-23 06:18:40.430785: step: 2168/1541, loss: 0.6542417407035828 2023-01-23 06:18:41.431914: step: 2170/1541, loss: 1.5721008777618408 2023-01-23 06:18:42.339838: step: 2172/1541, loss: 0.45316627621650696 2023-01-23 06:18:43.209421: step: 2174/1541, loss: 2.7794339656829834 2023-01-23 06:18:44.066292: step: 2176/1541, loss: 0.8355571031570435 2023-01-23 06:18:45.002942: step: 2178/1541, loss: 1.8232636451721191 2023-01-23 06:18:45.872579: step: 2180/1541, loss: 0.4342576563358307 2023-01-23 06:18:46.796240: step: 2182/1541, loss: 0.5821227431297302 2023-01-23 06:18:47.692537: step: 2184/1541, loss: 7.989980697631836 2023-01-23 06:18:48.576558: step: 2186/1541, loss: 1.22769296169281 2023-01-23 06:18:49.480806: step: 2188/1541, loss: 0.7591037750244141 2023-01-23 06:18:50.357526: step: 2190/1541, loss: 0.35472387075424194 2023-01-23 06:18:51.261171: step: 2192/1541, loss: 1.3592760562896729 2023-01-23 06:18:52.148753: step: 2194/1541, loss: 3.147512435913086 2023-01-23 06:18:53.032444: step: 2196/1541, loss: 0.23799742758274078 2023-01-23 06:18:53.941645: step: 2198/1541, loss: 0.47226080298423767 2023-01-23 06:18:54.824517: step: 2200/1541, loss: 0.7289937138557434 2023-01-23 06:18:55.697633: step: 2202/1541, loss: 1.3127846717834473 2023-01-23 06:18:56.597155: step: 2204/1541, loss: 1.2055658102035522 2023-01-23 06:18:57.484233: step: 2206/1541, loss: 0.8708028793334961 2023-01-23 06:18:58.380963: step: 2208/1541, loss: 0.7494773268699646 2023-01-23 06:18:59.268459: step: 2210/1541, loss: 1.4827178716659546 2023-01-23 06:19:00.182284: step: 2212/1541, loss: 3.218194007873535 2023-01-23 06:19:01.037733: step: 2214/1541, loss: 0.7736766338348389 2023-01-23 06:19:01.951039: step: 2216/1541, loss: 0.9202073812484741 2023-01-23 06:19:02.856061: step: 2218/1541, loss: 0.8346894979476929 2023-01-23 06:19:03.758954: step: 2220/1541, loss: 3.017698049545288 2023-01-23 06:19:04.658990: step: 2222/1541, loss: 2.94488525390625 2023-01-23 06:19:05.544499: step: 2224/1541, loss: 2.932439088821411 2023-01-23 06:19:06.464495: step: 2226/1541, loss: 3.3285720348358154 2023-01-23 06:19:07.370286: step: 2228/1541, loss: 3.3137097358703613 2023-01-23 06:19:08.298911: step: 2230/1541, loss: 1.06268310546875 2023-01-23 06:19:09.250620: step: 2232/1541, loss: 0.44528329372406006 2023-01-23 06:19:10.142007: step: 2234/1541, loss: 0.4357832372188568 2023-01-23 06:19:11.029387: step: 2236/1541, loss: 0.5704194903373718 2023-01-23 06:19:11.903113: step: 2238/1541, loss: 0.9187765717506409 2023-01-23 06:19:12.758816: step: 2240/1541, loss: 0.7287315130233765 2023-01-23 06:19:13.625588: step: 2242/1541, loss: 0.4157564342021942 2023-01-23 06:19:14.521076: step: 2244/1541, loss: 0.6580114364624023 2023-01-23 06:19:15.368273: step: 2246/1541, loss: 0.5813758373260498 2023-01-23 06:19:16.226974: step: 2248/1541, loss: 0.5654714703559875 2023-01-23 06:19:17.164035: step: 2250/1541, loss: 1.0169563293457031 2023-01-23 06:19:18.049536: step: 2252/1541, loss: 0.6278765797615051 2023-01-23 06:19:19.013133: step: 2254/1541, loss: 0.5425857901573181 2023-01-23 06:19:19.921738: step: 2256/1541, loss: 1.0609327554702759 2023-01-23 06:19:20.790164: step: 2258/1541, loss: 0.4730667471885681 2023-01-23 06:19:21.651989: step: 2260/1541, loss: 0.859410285949707 2023-01-23 06:19:22.584959: step: 2262/1541, loss: 1.5787659883499146 2023-01-23 06:19:23.462432: step: 2264/1541, loss: 0.4175575077533722 2023-01-23 06:19:24.375906: step: 2266/1541, loss: 3.7374072074890137 2023-01-23 06:19:25.248186: step: 2268/1541, loss: 0.7078477740287781 2023-01-23 06:19:26.174573: step: 2270/1541, loss: 0.6912528872489929 2023-01-23 06:19:27.091639: step: 2272/1541, loss: 0.4899320602416992 2023-01-23 06:19:28.004523: step: 2274/1541, loss: 1.1446365118026733 2023-01-23 06:19:28.872538: step: 2276/1541, loss: 2.322678565979004 2023-01-23 06:19:29.761192: step: 2278/1541, loss: 1.011906385421753 2023-01-23 06:19:30.661500: step: 2280/1541, loss: 1.928571343421936 2023-01-23 06:19:31.571041: step: 2282/1541, loss: 1.0504958629608154 2023-01-23 06:19:32.438968: step: 2284/1541, loss: 1.2029213905334473 2023-01-23 06:19:33.316006: step: 2286/1541, loss: 1.0397330522537231 2023-01-23 06:19:34.192024: step: 2288/1541, loss: 0.5987926125526428 2023-01-23 06:19:35.078648: step: 2290/1541, loss: 1.3282678127288818 2023-01-23 06:19:36.037924: step: 2292/1541, loss: 0.3941587209701538 2023-01-23 06:19:37.044840: step: 2294/1541, loss: 3.565288782119751 2023-01-23 06:19:37.936765: step: 2296/1541, loss: 0.8130354285240173 2023-01-23 06:19:38.821393: step: 2298/1541, loss: 0.3923815190792084 2023-01-23 06:19:39.712517: step: 2300/1541, loss: 1.11227548122406 2023-01-23 06:19:40.654101: step: 2302/1541, loss: 1.312785029411316 2023-01-23 06:19:41.565863: step: 2304/1541, loss: 1.4252727031707764 2023-01-23 06:19:42.471878: step: 2306/1541, loss: 1.6598831415176392 2023-01-23 06:19:43.364397: step: 2308/1541, loss: 0.7396066784858704 2023-01-23 06:19:44.259188: step: 2310/1541, loss: 0.33214014768600464 2023-01-23 06:19:45.164625: step: 2312/1541, loss: 3.0494801998138428 2023-01-23 06:19:46.085970: step: 2314/1541, loss: 3.481748580932617 2023-01-23 06:19:47.000189: step: 2316/1541, loss: 0.672597348690033 2023-01-23 06:19:47.856520: step: 2318/1541, loss: 0.6775630116462708 2023-01-23 06:19:48.813858: step: 2320/1541, loss: 0.5968133211135864 2023-01-23 06:19:49.671918: step: 2322/1541, loss: 1.301484227180481 2023-01-23 06:19:50.514369: step: 2324/1541, loss: 2.8881759643554688 2023-01-23 06:19:51.406890: step: 2326/1541, loss: 0.8298876285552979 2023-01-23 06:19:52.255454: step: 2328/1541, loss: 3.4362237453460693 2023-01-23 06:19:53.172640: step: 2330/1541, loss: 3.1882569789886475 2023-01-23 06:19:54.059708: step: 2332/1541, loss: 0.9369710087776184 2023-01-23 06:19:54.940008: step: 2334/1541, loss: 0.7516934871673584 2023-01-23 06:19:55.807208: step: 2336/1541, loss: 1.1861857175827026 2023-01-23 06:19:56.732410: step: 2338/1541, loss: 1.2015355825424194 2023-01-23 06:19:57.592153: step: 2340/1541, loss: 0.6536665558815002 2023-01-23 06:19:58.501064: step: 2342/1541, loss: 3.1607279777526855 2023-01-23 06:19:59.417319: step: 2344/1541, loss: 0.8922918438911438 2023-01-23 06:20:00.335628: step: 2346/1541, loss: 0.6214417219161987 2023-01-23 06:20:01.276773: step: 2348/1541, loss: 0.8091288805007935 2023-01-23 06:20:02.197526: step: 2350/1541, loss: 0.8153366446495056 2023-01-23 06:20:03.099807: step: 2352/1541, loss: 1.0431721210479736 2023-01-23 06:20:04.022053: step: 2354/1541, loss: 1.0682713985443115 2023-01-23 06:20:04.918946: step: 2356/1541, loss: 1.0617870092391968 2023-01-23 06:20:05.805980: step: 2358/1541, loss: 3.405257225036621 2023-01-23 06:20:06.702375: step: 2360/1541, loss: 1.1875232458114624 2023-01-23 06:20:07.603549: step: 2362/1541, loss: 0.7296475768089294 2023-01-23 06:20:08.472992: step: 2364/1541, loss: 8.907679557800293 2023-01-23 06:20:09.361362: step: 2366/1541, loss: 0.6222957968711853 2023-01-23 06:20:10.261708: step: 2368/1541, loss: 1.4717810153961182 2023-01-23 06:20:11.135301: step: 2370/1541, loss: 0.5906047224998474 2023-01-23 06:20:12.025534: step: 2372/1541, loss: 3.704151153564453 2023-01-23 06:20:12.909723: step: 2374/1541, loss: 1.270249366760254 2023-01-23 06:20:13.819658: step: 2376/1541, loss: 0.9675750732421875 2023-01-23 06:20:14.702942: step: 2378/1541, loss: 2.9877171516418457 2023-01-23 06:20:15.532264: step: 2380/1541, loss: 1.2302380800247192 2023-01-23 06:20:16.480626: step: 2382/1541, loss: 1.3435094356536865 2023-01-23 06:20:17.371535: step: 2384/1541, loss: 0.8973102569580078 2023-01-23 06:20:18.294296: step: 2386/1541, loss: 1.2805026769638062 2023-01-23 06:20:19.196609: step: 2388/1541, loss: 0.4238364100456238 2023-01-23 06:20:20.101858: step: 2390/1541, loss: 1.3897582292556763 2023-01-23 06:20:20.975301: step: 2392/1541, loss: 0.34063705801963806 2023-01-23 06:20:21.824025: step: 2394/1541, loss: 0.9026727080345154 2023-01-23 06:20:22.692820: step: 2396/1541, loss: 0.9821183681488037 2023-01-23 06:20:23.633941: step: 2398/1541, loss: 0.3596597909927368 2023-01-23 06:20:24.549397: step: 2400/1541, loss: 0.8154051303863525 2023-01-23 06:20:25.415424: step: 2402/1541, loss: 0.5546463131904602 2023-01-23 06:20:26.369302: step: 2404/1541, loss: 5.192855358123779 2023-01-23 06:20:27.308310: step: 2406/1541, loss: 1.2835406064987183 2023-01-23 06:20:28.253794: step: 2408/1541, loss: 0.5583070516586304 2023-01-23 06:20:29.150947: step: 2410/1541, loss: 1.0301942825317383 2023-01-23 06:20:30.033896: step: 2412/1541, loss: 0.6671984195709229 2023-01-23 06:20:30.883038: step: 2414/1541, loss: 0.5416455864906311 2023-01-23 06:20:31.748908: step: 2416/1541, loss: 0.6869326233863831 2023-01-23 06:20:32.624137: step: 2418/1541, loss: 0.48976343870162964 2023-01-23 06:20:33.558819: step: 2420/1541, loss: 1.218613862991333 2023-01-23 06:20:34.416358: step: 2422/1541, loss: 2.7360024452209473 2023-01-23 06:20:35.294278: step: 2424/1541, loss: 0.6742629408836365 2023-01-23 06:20:36.169025: step: 2426/1541, loss: 0.4626135528087616 2023-01-23 06:20:37.119815: step: 2428/1541, loss: 1.1467591524124146 2023-01-23 06:20:38.070060: step: 2430/1541, loss: 3.3012332916259766 2023-01-23 06:20:38.934749: step: 2432/1541, loss: 2.034870147705078 2023-01-23 06:20:39.830262: step: 2434/1541, loss: 0.5430470705032349 2023-01-23 06:20:40.724084: step: 2436/1541, loss: 0.40850546956062317 2023-01-23 06:20:41.604807: step: 2438/1541, loss: 0.929557204246521 2023-01-23 06:20:42.509082: step: 2440/1541, loss: 2.391979694366455 2023-01-23 06:20:43.380935: step: 2442/1541, loss: 0.33908361196517944 2023-01-23 06:20:44.272964: step: 2444/1541, loss: 1.1976252794265747 2023-01-23 06:20:45.179594: step: 2446/1541, loss: 1.509323239326477 2023-01-23 06:20:46.111097: step: 2448/1541, loss: 0.6189027428627014 2023-01-23 06:20:47.034586: step: 2450/1541, loss: 0.8877214193344116 2023-01-23 06:20:47.920697: step: 2452/1541, loss: 0.5736804604530334 2023-01-23 06:20:48.796453: step: 2454/1541, loss: 5.978498935699463 2023-01-23 06:20:49.668062: step: 2456/1541, loss: 0.9590590596199036 2023-01-23 06:20:50.540493: step: 2458/1541, loss: 2.403799533843994 2023-01-23 06:20:51.393507: step: 2460/1541, loss: 0.8852792978286743 2023-01-23 06:20:52.280330: step: 2462/1541, loss: 0.7191036343574524 2023-01-23 06:20:53.169581: step: 2464/1541, loss: 0.8593957424163818 2023-01-23 06:20:54.044386: step: 2466/1541, loss: 0.5741857290267944 2023-01-23 06:20:54.945326: step: 2468/1541, loss: 0.45947685837745667 2023-01-23 06:20:55.848990: step: 2470/1541, loss: 0.4560586214065552 2023-01-23 06:20:56.710842: step: 2472/1541, loss: 5.7472124099731445 2023-01-23 06:20:57.592946: step: 2474/1541, loss: 1.2050213813781738 2023-01-23 06:20:58.496658: step: 2476/1541, loss: 0.49819812178611755 2023-01-23 06:20:59.369437: step: 2478/1541, loss: 1.3557488918304443 2023-01-23 06:21:00.259305: step: 2480/1541, loss: 0.8130670189857483 2023-01-23 06:21:01.187151: step: 2482/1541, loss: 0.646308422088623 2023-01-23 06:21:02.135539: step: 2484/1541, loss: 3.238051414489746 2023-01-23 06:21:03.033522: step: 2486/1541, loss: 0.9584316611289978 2023-01-23 06:21:03.899295: step: 2488/1541, loss: 0.8008666038513184 2023-01-23 06:21:04.822803: step: 2490/1541, loss: 3.8155336380004883 2023-01-23 06:21:05.721202: step: 2492/1541, loss: 1.3203152418136597 2023-01-23 06:21:06.593689: step: 2494/1541, loss: 7.40425968170166 2023-01-23 06:21:07.477589: step: 2496/1541, loss: 0.4635293483734131 2023-01-23 06:21:08.372883: step: 2498/1541, loss: 1.2104694843292236 2023-01-23 06:21:09.289483: step: 2500/1541, loss: 1.037290096282959 2023-01-23 06:21:10.183289: step: 2502/1541, loss: 0.5361167788505554 2023-01-23 06:21:11.079601: step: 2504/1541, loss: 0.5224415063858032 2023-01-23 06:21:11.961565: step: 2506/1541, loss: 4.025001049041748 2023-01-23 06:21:12.904187: step: 2508/1541, loss: 3.1660003662109375 2023-01-23 06:21:13.773292: step: 2510/1541, loss: 0.5896605253219604 2023-01-23 06:21:14.737984: step: 2512/1541, loss: 0.8368892073631287 2023-01-23 06:21:15.614430: step: 2514/1541, loss: 6.59021520614624 2023-01-23 06:21:16.532241: step: 2516/1541, loss: 0.5818345546722412 2023-01-23 06:21:17.461451: step: 2518/1541, loss: 0.7071280479431152 2023-01-23 06:21:18.344564: step: 2520/1541, loss: 0.8133126497268677 2023-01-23 06:21:19.272573: step: 2522/1541, loss: 0.5452579259872437 2023-01-23 06:21:20.188195: step: 2524/1541, loss: 0.6929330825805664 2023-01-23 06:21:21.100715: step: 2526/1541, loss: 7.243762969970703 2023-01-23 06:21:21.988249: step: 2528/1541, loss: 0.3259367048740387 2023-01-23 06:21:22.919338: step: 2530/1541, loss: 2.7755165100097656 2023-01-23 06:21:23.825854: step: 2532/1541, loss: 2.8040719032287598 2023-01-23 06:21:24.750074: step: 2534/1541, loss: 3.219625234603882 2023-01-23 06:21:25.673172: step: 2536/1541, loss: 0.4053783714771271 2023-01-23 06:21:26.626176: step: 2538/1541, loss: 2.963953733444214 2023-01-23 06:21:27.527920: step: 2540/1541, loss: 0.7653759121894836 2023-01-23 06:21:28.404719: step: 2542/1541, loss: 2.8141210079193115 2023-01-23 06:21:29.318763: step: 2544/1541, loss: 3.2002079486846924 2023-01-23 06:21:30.201619: step: 2546/1541, loss: 1.4161577224731445 2023-01-23 06:21:31.165120: step: 2548/1541, loss: 1.221207857131958 2023-01-23 06:21:32.061040: step: 2550/1541, loss: 2.4256749153137207 2023-01-23 06:21:32.946746: step: 2552/1541, loss: 2.828275203704834 2023-01-23 06:21:33.808980: step: 2554/1541, loss: 0.8852969408035278 2023-01-23 06:21:34.694647: step: 2556/1541, loss: 5.100340843200684 2023-01-23 06:21:35.636138: step: 2558/1541, loss: 1.5477641820907593 2023-01-23 06:21:36.501947: step: 2560/1541, loss: 2.883704662322998 2023-01-23 06:21:37.377935: step: 2562/1541, loss: 1.5399507284164429 2023-01-23 06:21:38.275100: step: 2564/1541, loss: 1.3764921426773071 2023-01-23 06:21:39.324158: step: 2566/1541, loss: 1.6761316061019897 2023-01-23 06:21:40.233837: step: 2568/1541, loss: 1.2944729328155518 2023-01-23 06:21:41.125185: step: 2570/1541, loss: 0.6787909269332886 2023-01-23 06:21:42.038173: step: 2572/1541, loss: 1.6799583435058594 2023-01-23 06:21:42.935900: step: 2574/1541, loss: 1.5131546258926392 2023-01-23 06:21:43.822124: step: 2576/1541, loss: 0.9151994585990906 2023-01-23 06:21:44.683851: step: 2578/1541, loss: 1.6490201950073242 2023-01-23 06:21:45.540528: step: 2580/1541, loss: 3.0091867446899414 2023-01-23 06:21:46.443250: step: 2582/1541, loss: 0.3913668394088745 2023-01-23 06:21:47.380521: step: 2584/1541, loss: 0.4942114055156708 2023-01-23 06:21:48.266869: step: 2586/1541, loss: 1.1791224479675293 2023-01-23 06:21:49.179802: step: 2588/1541, loss: 0.7575547695159912 2023-01-23 06:21:50.066579: step: 2590/1541, loss: 1.3033249378204346 2023-01-23 06:21:50.973038: step: 2592/1541, loss: 3.6032137870788574 2023-01-23 06:21:51.825651: step: 2594/1541, loss: 3.021115779876709 2023-01-23 06:21:52.721113: step: 2596/1541, loss: 2.3462531566619873 2023-01-23 06:21:53.634367: step: 2598/1541, loss: 1.4647926092147827 2023-01-23 06:21:54.534103: step: 2600/1541, loss: 1.3912956714630127 2023-01-23 06:21:55.435536: step: 2602/1541, loss: 1.6466398239135742 2023-01-23 06:21:56.313541: step: 2604/1541, loss: 4.1922502517700195 2023-01-23 06:21:57.216196: step: 2606/1541, loss: 3.1423139572143555 2023-01-23 06:21:58.080099: step: 2608/1541, loss: 0.21759973466396332 2023-01-23 06:21:58.977704: step: 2610/1541, loss: 0.952065646648407 2023-01-23 06:21:59.866333: step: 2612/1541, loss: 1.048408031463623 2023-01-23 06:22:00.769882: step: 2614/1541, loss: 4.9450459480285645 2023-01-23 06:22:01.650096: step: 2616/1541, loss: 0.6574743986129761 2023-01-23 06:22:02.556481: step: 2618/1541, loss: 6.835864067077637 2023-01-23 06:22:03.497091: step: 2620/1541, loss: 3.3347461223602295 2023-01-23 06:22:04.392688: step: 2622/1541, loss: 1.2286297082901 2023-01-23 06:22:05.316742: step: 2624/1541, loss: 0.5550464391708374 2023-01-23 06:22:06.207546: step: 2626/1541, loss: 0.9635210633277893 2023-01-23 06:22:07.178052: step: 2628/1541, loss: 3.114511013031006 2023-01-23 06:22:08.074059: step: 2630/1541, loss: 3.3179564476013184 2023-01-23 06:22:09.005420: step: 2632/1541, loss: 2.9214203357696533 2023-01-23 06:22:09.901050: step: 2634/1541, loss: 0.6297857165336609 2023-01-23 06:22:10.794016: step: 2636/1541, loss: 0.5946261286735535 2023-01-23 06:22:11.664426: step: 2638/1541, loss: 0.30643659830093384 2023-01-23 06:22:12.540390: step: 2640/1541, loss: 0.4095361530780792 2023-01-23 06:22:13.474891: step: 2642/1541, loss: 3.051405668258667 2023-01-23 06:22:14.379140: step: 2644/1541, loss: 2.635145902633667 2023-01-23 06:22:15.277034: step: 2646/1541, loss: 0.49581849575042725 2023-01-23 06:22:16.151485: step: 2648/1541, loss: 0.5026257038116455 2023-01-23 06:22:17.077610: step: 2650/1541, loss: 3.443415641784668 2023-01-23 06:22:17.966850: step: 2652/1541, loss: 0.46487486362457275 2023-01-23 06:22:18.858476: step: 2654/1541, loss: 0.8319246172904968 2023-01-23 06:22:19.756876: step: 2656/1541, loss: 1.2548755407333374 2023-01-23 06:22:20.680832: step: 2658/1541, loss: 0.5918166637420654 2023-01-23 06:22:21.586610: step: 2660/1541, loss: 0.4026414155960083 2023-01-23 06:22:22.488599: step: 2662/1541, loss: 1.1513690948486328 2023-01-23 06:22:23.365960: step: 2664/1541, loss: 3.3341543674468994 2023-01-23 06:22:24.377157: step: 2666/1541, loss: 0.9312742948532104 2023-01-23 06:22:25.282237: step: 2668/1541, loss: 2.485084056854248 2023-01-23 06:22:26.202924: step: 2670/1541, loss: 4.303116798400879 2023-01-23 06:22:27.092318: step: 2672/1541, loss: 1.787066102027893 2023-01-23 06:22:27.985352: step: 2674/1541, loss: 0.8948470950126648 2023-01-23 06:22:28.847862: step: 2676/1541, loss: 3.553537130355835 2023-01-23 06:22:29.745716: step: 2678/1541, loss: 1.2312273979187012 2023-01-23 06:22:30.654397: step: 2680/1541, loss: 0.4230033755302429 2023-01-23 06:22:31.580043: step: 2682/1541, loss: 0.957822322845459 2023-01-23 06:22:32.504353: step: 2684/1541, loss: 1.0764386653900146 2023-01-23 06:22:33.408304: step: 2686/1541, loss: 0.4250023365020752 2023-01-23 06:22:34.332436: step: 2688/1541, loss: 0.7737668752670288 2023-01-23 06:22:35.225267: step: 2690/1541, loss: 0.9483950138092041 2023-01-23 06:22:36.154399: step: 2692/1541, loss: 3.612102508544922 2023-01-23 06:22:37.114441: step: 2694/1541, loss: 0.9378568530082703 2023-01-23 06:22:38.016162: step: 2696/1541, loss: 1.4416707754135132 2023-01-23 06:22:38.916678: step: 2698/1541, loss: 0.7652645707130432 2023-01-23 06:22:39.805281: step: 2700/1541, loss: 0.5252288579940796 2023-01-23 06:22:40.687430: step: 2702/1541, loss: 5.442856788635254 2023-01-23 06:22:41.569646: step: 2704/1541, loss: 0.9169837832450867 2023-01-23 06:22:42.508119: step: 2706/1541, loss: 0.9359731674194336 2023-01-23 06:22:43.420431: step: 2708/1541, loss: 0.8765154480934143 2023-01-23 06:22:44.307308: step: 2710/1541, loss: 0.9862391948699951 2023-01-23 06:22:45.180246: step: 2712/1541, loss: 0.4589812457561493 2023-01-23 06:22:46.158761: step: 2714/1541, loss: 0.7694450616836548 2023-01-23 06:22:47.103782: step: 2716/1541, loss: 0.7939561605453491 2023-01-23 06:22:47.995696: step: 2718/1541, loss: 0.668418824672699 2023-01-23 06:22:48.881242: step: 2720/1541, loss: 0.6471141576766968 2023-01-23 06:22:49.748402: step: 2722/1541, loss: 0.7552224397659302 2023-01-23 06:22:50.665131: step: 2724/1541, loss: 0.6657864451408386 2023-01-23 06:22:51.569911: step: 2726/1541, loss: 0.568876326084137 2023-01-23 06:22:52.457562: step: 2728/1541, loss: 0.951387882232666 2023-01-23 06:22:53.374110: step: 2730/1541, loss: 0.6467517614364624 2023-01-23 06:22:54.321237: step: 2732/1541, loss: 6.866145610809326 2023-01-23 06:22:55.214148: step: 2734/1541, loss: 1.5233852863311768 2023-01-23 06:22:56.126563: step: 2736/1541, loss: 0.7181416749954224 2023-01-23 06:22:57.052386: step: 2738/1541, loss: 0.501171886920929 2023-01-23 06:22:58.002702: step: 2740/1541, loss: 6.336933612823486 2023-01-23 06:22:58.916266: step: 2742/1541, loss: 0.8141660094261169 2023-01-23 06:22:59.785913: step: 2744/1541, loss: 0.7015358805656433 2023-01-23 06:23:00.645854: step: 2746/1541, loss: 0.501120388507843 2023-01-23 06:23:01.582397: step: 2748/1541, loss: 1.6932491064071655 2023-01-23 06:23:02.474621: step: 2750/1541, loss: 0.7814431190490723 2023-01-23 06:23:03.378757: step: 2752/1541, loss: 0.2924814224243164 2023-01-23 06:23:04.286097: step: 2754/1541, loss: 0.5145712494850159 2023-01-23 06:23:05.169148: step: 2756/1541, loss: 0.5965408682823181 2023-01-23 06:23:06.061099: step: 2758/1541, loss: 3.2810165882110596 2023-01-23 06:23:06.947631: step: 2760/1541, loss: 2.93320631980896 2023-01-23 06:23:07.853061: step: 2762/1541, loss: 0.5309401154518127 2023-01-23 06:23:08.741692: step: 2764/1541, loss: 1.0938820838928223 2023-01-23 06:23:09.622449: step: 2766/1541, loss: 0.6665436029434204 2023-01-23 06:23:10.561363: step: 2768/1541, loss: 0.7830813527107239 2023-01-23 06:23:11.465844: step: 2770/1541, loss: 0.4169347584247589 2023-01-23 06:23:12.415390: step: 2772/1541, loss: 0.46904197335243225 2023-01-23 06:23:13.307543: step: 2774/1541, loss: 0.8702046275138855 2023-01-23 06:23:14.210385: step: 2776/1541, loss: 1.5025231838226318 2023-01-23 06:23:15.083661: step: 2778/1541, loss: 0.5016480088233948 2023-01-23 06:23:15.922005: step: 2780/1541, loss: 0.8035240173339844 2023-01-23 06:23:16.827946: step: 2782/1541, loss: 0.3668767511844635 2023-01-23 06:23:17.696855: step: 2784/1541, loss: 1.0856398344039917 2023-01-23 06:23:18.608776: step: 2786/1541, loss: 0.9791877269744873 2023-01-23 06:23:19.498331: step: 2788/1541, loss: 0.46159619092941284 2023-01-23 06:23:20.419391: step: 2790/1541, loss: 1.604039192199707 2023-01-23 06:23:21.283346: step: 2792/1541, loss: 1.0450676679611206 2023-01-23 06:23:22.192459: step: 2794/1541, loss: 3.1253395080566406 2023-01-23 06:23:23.086343: step: 2796/1541, loss: 0.8358378410339355 2023-01-23 06:23:24.019204: step: 2798/1541, loss: 0.3587515950202942 2023-01-23 06:23:24.935199: step: 2800/1541, loss: 0.639725923538208 2023-01-23 06:23:25.842901: step: 2802/1541, loss: 0.7596039175987244 2023-01-23 06:23:26.755739: step: 2804/1541, loss: 0.4590551257133484 2023-01-23 06:23:27.656637: step: 2806/1541, loss: 1.1243935823440552 2023-01-23 06:23:28.568242: step: 2808/1541, loss: 0.9951379299163818 2023-01-23 06:23:29.492739: step: 2810/1541, loss: 1.2835438251495361 2023-01-23 06:23:30.396911: step: 2812/1541, loss: 3.0371367931365967 2023-01-23 06:23:31.322620: step: 2814/1541, loss: 0.5902575254440308 2023-01-23 06:23:32.271078: step: 2816/1541, loss: 0.4885922968387604 2023-01-23 06:23:33.111998: step: 2818/1541, loss: 0.46939051151275635 2023-01-23 06:23:33.969193: step: 2820/1541, loss: 1.0590600967407227 2023-01-23 06:23:34.896820: step: 2822/1541, loss: 3.8691749572753906 2023-01-23 06:23:35.791723: step: 2824/1541, loss: 0.9561951160430908 2023-01-23 06:23:36.694997: step: 2826/1541, loss: 0.6825037002563477 2023-01-23 06:23:37.701794: step: 2828/1541, loss: 1.2982150316238403 2023-01-23 06:23:38.595940: step: 2830/1541, loss: 0.795693576335907 2023-01-23 06:23:39.530730: step: 2832/1541, loss: 0.7316094040870667 2023-01-23 06:23:40.413911: step: 2834/1541, loss: 0.7674456834793091 2023-01-23 06:23:41.329372: step: 2836/1541, loss: 0.7852445840835571 2023-01-23 06:23:42.281373: step: 2838/1541, loss: 0.4664333760738373 2023-01-23 06:23:43.166468: step: 2840/1541, loss: 0.7423755526542664 2023-01-23 06:23:44.056619: step: 2842/1541, loss: 0.7701488137245178 2023-01-23 06:23:44.947494: step: 2844/1541, loss: 0.698620080947876 2023-01-23 06:23:45.886014: step: 2846/1541, loss: 0.6458317637443542 2023-01-23 06:23:46.805780: step: 2848/1541, loss: 1.2344783544540405 2023-01-23 06:23:47.727614: step: 2850/1541, loss: 1.250192642211914 2023-01-23 06:23:48.644167: step: 2852/1541, loss: 1.1919580698013306 2023-01-23 06:23:49.564264: step: 2854/1541, loss: 0.7202158570289612 2023-01-23 06:23:50.484507: step: 2856/1541, loss: 1.423413872718811 2023-01-23 06:23:51.345991: step: 2858/1541, loss: 0.6106268167495728 2023-01-23 06:23:52.252846: step: 2860/1541, loss: 0.6685186624526978 2023-01-23 06:23:53.164586: step: 2862/1541, loss: 0.9114799499511719 2023-01-23 06:23:54.062212: step: 2864/1541, loss: 0.2854057252407074 2023-01-23 06:23:54.948717: step: 2866/1541, loss: 5.312473773956299 2023-01-23 06:23:55.888608: step: 2868/1541, loss: 1.475170612335205 2023-01-23 06:23:56.781775: step: 2870/1541, loss: 3.094062566757202 2023-01-23 06:23:57.674978: step: 2872/1541, loss: 3.155470848083496 2023-01-23 06:23:58.560088: step: 2874/1541, loss: 0.5950455069541931 2023-01-23 06:23:59.428763: step: 2876/1541, loss: 1.55536687374115 2023-01-23 06:24:00.384241: step: 2878/1541, loss: 0.40149986743927 2023-01-23 06:24:01.264289: step: 2880/1541, loss: 0.5593255758285522 2023-01-23 06:24:02.130040: step: 2882/1541, loss: 0.6632104516029358 2023-01-23 06:24:03.080799: step: 2884/1541, loss: 0.8845832347869873 2023-01-23 06:24:03.944954: step: 2886/1541, loss: 0.7242321372032166 2023-01-23 06:24:04.809821: step: 2888/1541, loss: 1.0690860748291016 2023-01-23 06:24:05.709101: step: 2890/1541, loss: 0.40824636816978455 2023-01-23 06:24:06.596633: step: 2892/1541, loss: 1.1798344850540161 2023-01-23 06:24:07.481824: step: 2894/1541, loss: 0.6429632902145386 2023-01-23 06:24:08.389521: step: 2896/1541, loss: 0.8386955857276917 2023-01-23 06:24:09.372087: step: 2898/1541, loss: 0.6766651272773743 2023-01-23 06:24:10.259270: step: 2900/1541, loss: 0.7537204027175903 2023-01-23 06:24:11.286871: step: 2902/1541, loss: 1.4898053407669067 2023-01-23 06:24:12.199615: step: 2904/1541, loss: 0.8477781414985657 2023-01-23 06:24:13.049867: step: 2906/1541, loss: 5.550699234008789 2023-01-23 06:24:13.912566: step: 2908/1541, loss: 3.1247410774230957 2023-01-23 06:24:14.873056: step: 2910/1541, loss: 0.6098529100418091 2023-01-23 06:24:15.776878: step: 2912/1541, loss: 1.0809458494186401 2023-01-23 06:24:16.658782: step: 2914/1541, loss: 0.6584835052490234 2023-01-23 06:24:17.545489: step: 2916/1541, loss: 1.0203074216842651 2023-01-23 06:24:18.473086: step: 2918/1541, loss: 1.284130334854126 2023-01-23 06:24:19.388556: step: 2920/1541, loss: 0.8174715638160706 2023-01-23 06:24:20.277561: step: 2922/1541, loss: 2.658342123031616 2023-01-23 06:24:21.137429: step: 2924/1541, loss: 3.1425492763519287 2023-01-23 06:24:22.090714: step: 2926/1541, loss: 2.3825581073760986 2023-01-23 06:24:22.962674: step: 2928/1541, loss: 1.5714914798736572 2023-01-23 06:24:23.859895: step: 2930/1541, loss: 1.3429350852966309 2023-01-23 06:24:24.743132: step: 2932/1541, loss: 0.3567076325416565 2023-01-23 06:24:25.602724: step: 2934/1541, loss: 0.968883752822876 2023-01-23 06:24:26.483203: step: 2936/1541, loss: 6.923331260681152 2023-01-23 06:24:27.426085: step: 2938/1541, loss: 0.9016653895378113 2023-01-23 06:24:28.348648: step: 2940/1541, loss: 1.1819475889205933 2023-01-23 06:24:29.246265: step: 2942/1541, loss: 0.6947866082191467 2023-01-23 06:24:30.137300: step: 2944/1541, loss: 1.640610933303833 2023-01-23 06:24:31.075614: step: 2946/1541, loss: 0.8456208109855652 2023-01-23 06:24:31.962107: step: 2948/1541, loss: 0.7910149097442627 2023-01-23 06:24:32.852537: step: 2950/1541, loss: 1.1603639125823975 2023-01-23 06:24:33.778143: step: 2952/1541, loss: 1.240493655204773 2023-01-23 06:24:34.668877: step: 2954/1541, loss: 0.60820072889328 2023-01-23 06:24:35.533100: step: 2956/1541, loss: 0.522866427898407 2023-01-23 06:24:36.428239: step: 2958/1541, loss: 2.814815044403076 2023-01-23 06:24:37.348596: step: 2960/1541, loss: 1.3221328258514404 2023-01-23 06:24:38.252132: step: 2962/1541, loss: 0.8742246627807617 2023-01-23 06:24:39.135230: step: 2964/1541, loss: 0.5027730464935303 2023-01-23 06:24:40.046203: step: 2966/1541, loss: 1.1612470149993896 2023-01-23 06:24:40.953972: step: 2968/1541, loss: 0.49461787939071655 2023-01-23 06:24:41.832460: step: 2970/1541, loss: 0.5078362822532654 2023-01-23 06:24:42.734611: step: 2972/1541, loss: 0.4511033892631531 2023-01-23 06:24:43.605342: step: 2974/1541, loss: 0.7152271866798401 2023-01-23 06:24:44.466248: step: 2976/1541, loss: 0.4789341390132904 2023-01-23 06:24:45.432024: step: 2978/1541, loss: 0.4870545268058777 2023-01-23 06:24:46.277850: step: 2980/1541, loss: 0.4993591904640198 2023-01-23 06:24:47.173230: step: 2982/1541, loss: 1.8381478786468506 2023-01-23 06:24:48.063211: step: 2984/1541, loss: 6.003362655639648 2023-01-23 06:24:48.969726: step: 2986/1541, loss: 0.6532694697380066 2023-01-23 06:24:49.872369: step: 2988/1541, loss: 0.5077416896820068 2023-01-23 06:24:50.774665: step: 2990/1541, loss: 0.9230139255523682 2023-01-23 06:24:51.655597: step: 2992/1541, loss: 0.40327349305152893 2023-01-23 06:24:52.562635: step: 2994/1541, loss: 0.9990313053131104 2023-01-23 06:24:53.487988: step: 2996/1541, loss: 0.5308389663696289 2023-01-23 06:24:54.389917: step: 2998/1541, loss: 1.4679962396621704 2023-01-23 06:24:55.279553: step: 3000/1541, loss: 4.019498825073242 2023-01-23 06:24:56.183189: step: 3002/1541, loss: 0.8078593015670776 2023-01-23 06:24:57.064216: step: 3004/1541, loss: 2.1271767616271973 2023-01-23 06:24:57.992306: step: 3006/1541, loss: 0.3156262934207916 2023-01-23 06:24:58.862305: step: 3008/1541, loss: 0.553762674331665 2023-01-23 06:24:59.891605: step: 3010/1541, loss: 3.689039945602417 2023-01-23 06:25:00.767836: step: 3012/1541, loss: 3.1958441734313965 2023-01-23 06:25:01.645039: step: 3014/1541, loss: 0.6760873794555664 2023-01-23 06:25:02.590630: step: 3016/1541, loss: 0.34504324197769165 2023-01-23 06:25:03.433996: step: 3018/1541, loss: 0.6179415583610535 2023-01-23 06:25:04.347573: step: 3020/1541, loss: 1.3013800382614136 2023-01-23 06:25:05.234116: step: 3022/1541, loss: 0.6224789023399353 2023-01-23 06:25:06.166109: step: 3024/1541, loss: 1.2319406270980835 2023-01-23 06:25:07.028663: step: 3026/1541, loss: 0.5365291833877563 2023-01-23 06:25:07.946135: step: 3028/1541, loss: 0.9298970699310303 2023-01-23 06:25:08.881213: step: 3030/1541, loss: 1.3497583866119385 2023-01-23 06:25:09.839395: step: 3032/1541, loss: 0.5014330744743347 2023-01-23 06:25:10.743208: step: 3034/1541, loss: 0.7593774795532227 2023-01-23 06:25:11.610456: step: 3036/1541, loss: 1.0449283123016357 2023-01-23 06:25:12.548360: step: 3038/1541, loss: 0.6863257884979248 2023-01-23 06:25:13.450249: step: 3040/1541, loss: 1.075376033782959 2023-01-23 06:25:14.329574: step: 3042/1541, loss: 3.0526299476623535 2023-01-23 06:25:15.165942: step: 3044/1541, loss: 2.7499918937683105 2023-01-23 06:25:16.090175: step: 3046/1541, loss: 0.8554897904396057 2023-01-23 06:25:16.985131: step: 3048/1541, loss: 0.755613386631012 2023-01-23 06:25:17.876354: step: 3050/1541, loss: 0.741803765296936 2023-01-23 06:25:18.795160: step: 3052/1541, loss: 0.8438074588775635 2023-01-23 06:25:19.675373: step: 3054/1541, loss: 0.9863602519035339 2023-01-23 06:25:20.572174: step: 3056/1541, loss: 0.46278253197669983 2023-01-23 06:25:21.460162: step: 3058/1541, loss: 1.2240314483642578 2023-01-23 06:25:22.362316: step: 3060/1541, loss: 0.7158665060997009 2023-01-23 06:25:23.291177: step: 3062/1541, loss: 0.8298373222351074 2023-01-23 06:25:24.166949: step: 3064/1541, loss: 5.261936187744141 2023-01-23 06:25:25.118646: step: 3066/1541, loss: 1.0574442148208618 2023-01-23 06:25:25.986587: step: 3068/1541, loss: 0.30305156111717224 2023-01-23 06:25:26.885235: step: 3070/1541, loss: 1.186387300491333 2023-01-23 06:25:27.795132: step: 3072/1541, loss: 0.6206837892532349 2023-01-23 06:25:28.775061: step: 3074/1541, loss: 3.9644570350646973 2023-01-23 06:25:29.657975: step: 3076/1541, loss: 1.397201657295227 2023-01-23 06:25:30.554087: step: 3078/1541, loss: 0.43224313855171204 2023-01-23 06:25:31.467021: step: 3080/1541, loss: 0.6397441625595093 2023-01-23 06:25:32.376952: step: 3082/1541, loss: 0.47613269090652466 2023-01-23 06:25:33.012166: step: 3084/1541, loss: 2.1677894592285156 ================================================== Loss: 1.439 -------------------- Dev Chinese: {'event': {'p': 0.6212664277180406, 'r': 0.6924101198402131, 'f1': 0.6549118387909321}, 'argument': {'p': 0.6893128369893616, 'r': 0.624505647186943, 'f1': 0.6553108584831598}, 'combined': 0.42917083930887046, 'epoch': 2} Test Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.7288561511943926, 'r': 0.6674983629560891, 'f1': 0.6968291831658773}, 'combined': 0.45310483430724363, 'epoch': 2} Dev Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.6879318328246672, 'r': 0.6346725941543704, 'f1': 0.6602298798250162}, 'combined': 0.4339399097212416, 'epoch': 2} Test Korean: {'event': {'p': 0.607721639656816, 'r': 0.7639304973037747, 'f1': 0.6769312450225644}, 'argument': {'p': 0.7251491588582952, 'r': 0.6793663287655232, 'f1': 0.701511554983885}, 'combined': 0.4748750903129564, 'epoch': 2} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6840848719584073, 'r': 0.6320573327507272, 'f1': 0.6570427693878401}, 'combined': 0.4547807005668812, 'epoch': 2} Test Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7226548990349742, 'r': 0.676241111117048, 'f1': 0.6986780265739436}, 'combined': 0.48673268108251827, 'epoch': 2} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6497835497835497, 'r': 0.6037811745776347, 'f1': 0.6259382819015846}, 'combined': 0.4600913866968912, 'epoch': 2} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.5048776455026455, 'r': 0.24561615186615185, 'f1': 0.33046536796536796}, 'combined': 0.2029629818999898, 'epoch': 2} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.49764818514818515, 'r': 0.5092213987562825, 'f1': 0.5033682792303481}, 'combined': 0.25887511503275046, 'epoch': 2} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6212664277180406, 'r': 0.6924101198402131, 'f1': 0.6549118387909321}, 'argument': {'p': 0.6893128369893616, 'r': 0.624505647186943, 'f1': 0.6553108584831598}, 'combined': 0.42917083930887046, 'epoch': 2} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.7288561511943926, 'r': 0.6674983629560891, 'f1': 0.6968291831658773}, 'combined': 0.45310483430724363, 'epoch': 2} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6497835497835497, 'r': 0.6037811745776347, 'f1': 0.6259382819015846}, 'combined': 0.4600913866968912, 'epoch': 2} -------------------- Dev for Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.6879318328246672, 'r': 0.6346725941543704, 'f1': 0.6602298798250162}, 'combined': 0.4339399097212416, 'epoch': 2} Test for Korean: {'event': {'p': 0.607721639656816, 'r': 0.7639304973037747, 'f1': 0.6769312450225644}, 'argument': {'p': 0.7251491588582952, 'r': 0.6793663287655232, 'f1': 0.701511554983885}, 'combined': 0.4748750903129564, 'epoch': 2} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.5048776455026455, 'r': 0.24561615186615185, 'f1': 0.33046536796536796}, 'combined': 0.2029629818999898, 'epoch': 2} -------------------- Dev for Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6840848719584073, 'r': 0.6320573327507272, 'f1': 0.6570427693878401}, 'combined': 0.4547807005668812, 'epoch': 2} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7226548990349742, 'r': 0.676241111117048, 'f1': 0.6986780265739436}, 'combined': 0.48673268108251827, 'epoch': 2} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.49764818514818515, 'r': 0.5092213987562825, 'f1': 0.5033682792303481}, 'combined': 0.25887511503275046, 'epoch': 2} ****************************** Epoch: 3 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 06:31:15.003191: step: 2/1541, loss: 0.9028211236000061 2023-01-23 06:31:15.878525: step: 4/1541, loss: 1.573695421218872 2023-01-23 06:31:16.763157: step: 6/1541, loss: 4.94209623336792 2023-01-23 06:31:17.644255: step: 8/1541, loss: 3.4177794456481934 2023-01-23 06:31:18.584362: step: 10/1541, loss: 0.8049171566963196 2023-01-23 06:31:19.440175: step: 12/1541, loss: 2.7964096069335938 2023-01-23 06:31:20.368707: step: 14/1541, loss: 0.5578429698944092 2023-01-23 06:31:21.354583: step: 16/1541, loss: 0.9336281418800354 2023-01-23 06:31:22.212868: step: 18/1541, loss: 1.2688517570495605 2023-01-23 06:31:23.094668: step: 20/1541, loss: 0.8133953809738159 2023-01-23 06:31:24.042877: step: 22/1541, loss: 1.8076751232147217 2023-01-23 06:31:24.932049: step: 24/1541, loss: 0.5293958187103271 2023-01-23 06:31:25.811306: step: 26/1541, loss: 0.5649165511131287 2023-01-23 06:31:26.752333: step: 28/1541, loss: 1.288214921951294 2023-01-23 06:31:27.622836: step: 30/1541, loss: 0.10888871550559998 2023-01-23 06:31:28.596059: step: 32/1541, loss: 0.5151811838150024 2023-01-23 06:31:29.486375: step: 34/1541, loss: 1.0579068660736084 2023-01-23 06:31:30.353205: step: 36/1541, loss: 0.3068389594554901 2023-01-23 06:31:31.241880: step: 38/1541, loss: 0.306343138217926 2023-01-23 06:31:32.196819: step: 40/1541, loss: 0.12281239032745361 2023-01-23 06:31:33.096586: step: 42/1541, loss: 0.4973805546760559 2023-01-23 06:31:33.964686: step: 44/1541, loss: 0.49360182881355286 2023-01-23 06:31:34.902929: step: 46/1541, loss: 0.42247799038887024 2023-01-23 06:31:35.839680: step: 48/1541, loss: 0.9461613297462463 2023-01-23 06:31:36.741050: step: 50/1541, loss: 0.6517974138259888 2023-01-23 06:31:37.597663: step: 52/1541, loss: 1.2692545652389526 2023-01-23 06:31:38.488440: step: 54/1541, loss: 0.2896275520324707 2023-01-23 06:31:39.408953: step: 56/1541, loss: 2.504589319229126 2023-01-23 06:31:40.274125: step: 58/1541, loss: 0.4513208866119385 2023-01-23 06:31:41.161318: step: 60/1541, loss: 0.5389995574951172 2023-01-23 06:31:42.043987: step: 62/1541, loss: 0.3331499397754669 2023-01-23 06:31:42.979935: step: 64/1541, loss: 0.5400927662849426 2023-01-23 06:31:43.879734: step: 66/1541, loss: 0.7706707119941711 2023-01-23 06:31:44.760311: step: 68/1541, loss: 2.303234577178955 2023-01-23 06:31:45.662596: step: 70/1541, loss: 0.7249647378921509 2023-01-23 06:31:46.605967: step: 72/1541, loss: 1.2495440244674683 2023-01-23 06:31:47.487124: step: 74/1541, loss: 0.5452345609664917 2023-01-23 06:31:48.412889: step: 76/1541, loss: 0.30676695704460144 2023-01-23 06:31:49.289491: step: 78/1541, loss: 1.1242094039916992 2023-01-23 06:31:50.157413: step: 80/1541, loss: 0.7531300187110901 2023-01-23 06:31:51.083830: step: 82/1541, loss: 0.7952859997749329 2023-01-23 06:31:52.002854: step: 84/1541, loss: 2.693742513656616 2023-01-23 06:31:52.995841: step: 86/1541, loss: 0.6214902400970459 2023-01-23 06:31:53.918413: step: 88/1541, loss: 0.5432215929031372 2023-01-23 06:31:54.871114: step: 90/1541, loss: 0.42302998900413513 2023-01-23 06:31:55.773192: step: 92/1541, loss: 0.5191811323165894 2023-01-23 06:31:56.617489: step: 94/1541, loss: 0.5174882411956787 2023-01-23 06:31:57.487338: step: 96/1541, loss: 0.5904658436775208 2023-01-23 06:31:58.415414: step: 98/1541, loss: 0.8721208572387695 2023-01-23 06:31:59.324590: step: 100/1541, loss: 1.5348371267318726 2023-01-23 06:32:00.219870: step: 102/1541, loss: 0.6607627272605896 2023-01-23 06:32:01.121767: step: 104/1541, loss: 2.715390682220459 2023-01-23 06:32:02.037616: step: 106/1541, loss: 0.6595248579978943 2023-01-23 06:32:02.944248: step: 108/1541, loss: 0.46389564871788025 2023-01-23 06:32:03.819045: step: 110/1541, loss: 0.8597419261932373 2023-01-23 06:32:04.734590: step: 112/1541, loss: 1.022499680519104 2023-01-23 06:32:05.648004: step: 114/1541, loss: 0.7661647796630859 2023-01-23 06:32:06.557157: step: 116/1541, loss: 0.8163541555404663 2023-01-23 06:32:07.474756: step: 118/1541, loss: 0.7742744088172913 2023-01-23 06:32:08.373620: step: 120/1541, loss: 1.1064059734344482 2023-01-23 06:32:09.240999: step: 122/1541, loss: 2.04740571975708 2023-01-23 06:32:10.142051: step: 124/1541, loss: 0.6688390374183655 2023-01-23 06:32:11.006428: step: 126/1541, loss: 1.0306615829467773 2023-01-23 06:32:11.887765: step: 128/1541, loss: 1.8714745044708252 2023-01-23 06:32:12.796343: step: 130/1541, loss: 0.4698355793952942 2023-01-23 06:32:13.706322: step: 132/1541, loss: 1.2492139339447021 2023-01-23 06:32:14.610281: step: 134/1541, loss: 1.2078793048858643 2023-01-23 06:32:15.497339: step: 136/1541, loss: 0.5591988563537598 2023-01-23 06:32:16.437095: step: 138/1541, loss: 1.1457442045211792 2023-01-23 06:32:17.312056: step: 140/1541, loss: 0.8387815356254578 2023-01-23 06:32:18.234993: step: 142/1541, loss: 0.5739314556121826 2023-01-23 06:32:19.154342: step: 144/1541, loss: 1.049291968345642 2023-01-23 06:32:20.044296: step: 146/1541, loss: 0.6167237162590027 2023-01-23 06:32:20.934673: step: 148/1541, loss: 0.9591178297996521 2023-01-23 06:32:21.830493: step: 150/1541, loss: 0.4428805112838745 2023-01-23 06:32:22.789345: step: 152/1541, loss: 0.463135302066803 2023-01-23 06:32:23.684617: step: 154/1541, loss: 0.6479989886283875 2023-01-23 06:32:24.626821: step: 156/1541, loss: 0.543422281742096 2023-01-23 06:32:25.525988: step: 158/1541, loss: 0.3552815914154053 2023-01-23 06:32:26.425453: step: 160/1541, loss: 0.6570674180984497 2023-01-23 06:32:27.381338: step: 162/1541, loss: 0.7955978512763977 2023-01-23 06:32:28.273865: step: 164/1541, loss: 3.022063732147217 2023-01-23 06:32:29.171838: step: 166/1541, loss: 0.4363641142845154 2023-01-23 06:32:30.079181: step: 168/1541, loss: 2.425072193145752 2023-01-23 06:32:30.958547: step: 170/1541, loss: 0.4196503460407257 2023-01-23 06:32:31.893734: step: 172/1541, loss: 0.6275849342346191 2023-01-23 06:32:32.828619: step: 174/1541, loss: 1.1615703105926514 2023-01-23 06:32:33.713284: step: 176/1541, loss: 1.930449366569519 2023-01-23 06:32:34.583576: step: 178/1541, loss: 0.5078964829444885 2023-01-23 06:32:35.446285: step: 180/1541, loss: 0.524239182472229 2023-01-23 06:32:36.318076: step: 182/1541, loss: 0.3575831949710846 2023-01-23 06:32:37.179763: step: 184/1541, loss: 0.46595433354377747 2023-01-23 06:32:38.073224: step: 186/1541, loss: 0.6157121062278748 2023-01-23 06:32:38.985651: step: 188/1541, loss: 0.8833168745040894 2023-01-23 06:32:39.863953: step: 190/1541, loss: 0.9409406781196594 2023-01-23 06:32:40.771887: step: 192/1541, loss: 0.5576081275939941 2023-01-23 06:32:41.631495: step: 194/1541, loss: 2.639446496963501 2023-01-23 06:32:42.544310: step: 196/1541, loss: 0.5468037128448486 2023-01-23 06:32:43.433223: step: 198/1541, loss: 0.5476449728012085 2023-01-23 06:32:44.340804: step: 200/1541, loss: 1.909948468208313 2023-01-23 06:32:45.218125: step: 202/1541, loss: 0.946615993976593 2023-01-23 06:32:46.193826: step: 204/1541, loss: 6.871638774871826 2023-01-23 06:32:47.117871: step: 206/1541, loss: 1.8735215663909912 2023-01-23 06:32:47.977795: step: 208/1541, loss: 1.5237479209899902 2023-01-23 06:32:48.841415: step: 210/1541, loss: 0.17500092089176178 2023-01-23 06:32:49.793077: step: 212/1541, loss: 0.7589967250823975 2023-01-23 06:32:50.700072: step: 214/1541, loss: 1.1566452980041504 2023-01-23 06:32:51.603129: step: 216/1541, loss: 1.5513778924942017 2023-01-23 06:32:52.562913: step: 218/1541, loss: 0.6024172902107239 2023-01-23 06:32:53.463941: step: 220/1541, loss: 1.0718498229980469 2023-01-23 06:32:54.397713: step: 222/1541, loss: 1.0902801752090454 2023-01-23 06:32:55.297251: step: 224/1541, loss: 0.658811092376709 2023-01-23 06:32:56.174197: step: 226/1541, loss: 0.655961275100708 2023-01-23 06:32:57.068759: step: 228/1541, loss: 0.9338037967681885 2023-01-23 06:32:57.928365: step: 230/1541, loss: 0.22555625438690186 2023-01-23 06:32:58.821606: step: 232/1541, loss: 0.6320006847381592 2023-01-23 06:32:59.748630: step: 234/1541, loss: 0.6545066833496094 2023-01-23 06:33:00.625873: step: 236/1541, loss: 1.2921063899993896 2023-01-23 06:33:01.525467: step: 238/1541, loss: 0.4833991825580597 2023-01-23 06:33:02.424150: step: 240/1541, loss: 1.087501883506775 2023-01-23 06:33:03.340412: step: 242/1541, loss: 0.9910438656806946 2023-01-23 06:33:04.212992: step: 244/1541, loss: 0.8318524956703186 2023-01-23 06:33:05.128687: step: 246/1541, loss: 1.0871226787567139 2023-01-23 06:33:06.011050: step: 248/1541, loss: 0.5712933540344238 2023-01-23 06:33:06.917831: step: 250/1541, loss: 0.8070554733276367 2023-01-23 06:33:07.812510: step: 252/1541, loss: 0.7788692712783813 2023-01-23 06:33:08.726172: step: 254/1541, loss: 0.9448153972625732 2023-01-23 06:33:09.599942: step: 256/1541, loss: 0.21705317497253418 2023-01-23 06:33:10.585290: step: 258/1541, loss: 0.5230154991149902 2023-01-23 06:33:11.549340: step: 260/1541, loss: 1.1073598861694336 2023-01-23 06:33:12.449406: step: 262/1541, loss: 0.5660895705223083 2023-01-23 06:33:13.346532: step: 264/1541, loss: 0.39230877161026 2023-01-23 06:33:14.226468: step: 266/1541, loss: 1.3534950017929077 2023-01-23 06:33:15.155274: step: 268/1541, loss: 0.4205110967159271 2023-01-23 06:33:16.105935: step: 270/1541, loss: 0.8090956211090088 2023-01-23 06:33:17.026075: step: 272/1541, loss: 0.40315648913383484 2023-01-23 06:33:18.010351: step: 274/1541, loss: 2.3188915252685547 2023-01-23 06:33:18.911864: step: 276/1541, loss: 1.0315361022949219 2023-01-23 06:33:19.854220: step: 278/1541, loss: 0.6599130630493164 2023-01-23 06:33:20.715465: step: 280/1541, loss: 0.839260458946228 2023-01-23 06:33:21.645102: step: 282/1541, loss: 0.5747161507606506 2023-01-23 06:33:22.496725: step: 284/1541, loss: 0.27409955859184265 2023-01-23 06:33:23.409652: step: 286/1541, loss: 0.6091787219047546 2023-01-23 06:33:24.352393: step: 288/1541, loss: 2.442810535430908 2023-01-23 06:33:25.272218: step: 290/1541, loss: 0.32332926988601685 2023-01-23 06:33:26.281718: step: 292/1541, loss: 0.7365201711654663 2023-01-23 06:33:27.184388: step: 294/1541, loss: 0.5395795702934265 2023-01-23 06:33:28.077083: step: 296/1541, loss: 1.5858051776885986 2023-01-23 06:33:28.997132: step: 298/1541, loss: 0.5582224130630493 2023-01-23 06:33:29.853606: step: 300/1541, loss: 0.3539136052131653 2023-01-23 06:33:30.755030: step: 302/1541, loss: 0.5730241537094116 2023-01-23 06:33:31.623403: step: 304/1541, loss: 0.48161545395851135 2023-01-23 06:33:32.582742: step: 306/1541, loss: 0.9887155890464783 2023-01-23 06:33:33.454287: step: 308/1541, loss: 0.6949543356895447 2023-01-23 06:33:34.342786: step: 310/1541, loss: 0.8361755013465881 2023-01-23 06:33:35.283624: step: 312/1541, loss: 0.9178257584571838 2023-01-23 06:33:36.150002: step: 314/1541, loss: 0.5726152062416077 2023-01-23 06:33:37.042421: step: 316/1541, loss: 3.276439905166626 2023-01-23 06:33:38.043564: step: 318/1541, loss: 2.187657117843628 2023-01-23 06:33:38.963871: step: 320/1541, loss: 1.021615743637085 2023-01-23 06:33:39.864623: step: 322/1541, loss: 1.1666561365127563 2023-01-23 06:33:40.796926: step: 324/1541, loss: 0.5319563746452332 2023-01-23 06:33:41.702928: step: 326/1541, loss: 0.9381054639816284 2023-01-23 06:33:42.562932: step: 328/1541, loss: 4.920280456542969 2023-01-23 06:33:43.444503: step: 330/1541, loss: 0.349736750125885 2023-01-23 06:33:44.320848: step: 332/1541, loss: 0.6262247562408447 2023-01-23 06:33:45.230261: step: 334/1541, loss: 0.8106998801231384 2023-01-23 06:33:46.124645: step: 336/1541, loss: 1.3328921794891357 2023-01-23 06:33:47.054231: step: 338/1541, loss: 0.3957418203353882 2023-01-23 06:33:47.909161: step: 340/1541, loss: 2.1689870357513428 2023-01-23 06:33:48.807934: step: 342/1541, loss: 1.8436009883880615 2023-01-23 06:33:49.674394: step: 344/1541, loss: 0.9293293952941895 2023-01-23 06:33:50.595281: step: 346/1541, loss: 0.44351986050605774 2023-01-23 06:33:51.539891: step: 348/1541, loss: 1.1043894290924072 2023-01-23 06:33:52.431615: step: 350/1541, loss: 0.38091719150543213 2023-01-23 06:33:53.307035: step: 352/1541, loss: 0.7172551155090332 2023-01-23 06:33:54.228510: step: 354/1541, loss: 0.9803442358970642 2023-01-23 06:33:55.121762: step: 356/1541, loss: 0.7226430177688599 2023-01-23 06:33:56.030615: step: 358/1541, loss: 1.9771754741668701 2023-01-23 06:33:56.899572: step: 360/1541, loss: 0.8134808540344238 2023-01-23 06:33:57.804470: step: 362/1541, loss: 0.38614946603775024 2023-01-23 06:33:58.704199: step: 364/1541, loss: 3.090630531311035 2023-01-23 06:33:59.567047: step: 366/1541, loss: 0.08888125419616699 2023-01-23 06:34:00.436546: step: 368/1541, loss: 0.8925016522407532 2023-01-23 06:34:01.349858: step: 370/1541, loss: 1.202490210533142 2023-01-23 06:34:02.297337: step: 372/1541, loss: 0.8578289151191711 2023-01-23 06:34:03.182474: step: 374/1541, loss: 0.9074280858039856 2023-01-23 06:34:04.052655: step: 376/1541, loss: 1.880523681640625 2023-01-23 06:34:04.950389: step: 378/1541, loss: 0.720535933971405 2023-01-23 06:34:05.840230: step: 380/1541, loss: 0.3791486918926239 2023-01-23 06:34:06.733476: step: 382/1541, loss: 1.4996592998504639 2023-01-23 06:34:07.618266: step: 384/1541, loss: 0.4353596270084381 2023-01-23 06:34:08.504457: step: 386/1541, loss: 0.784839928150177 2023-01-23 06:34:09.388245: step: 388/1541, loss: 0.4689538776874542 2023-01-23 06:34:10.398553: step: 390/1541, loss: 0.8412795066833496 2023-01-23 06:34:11.428856: step: 392/1541, loss: 0.701643168926239 2023-01-23 06:34:12.344790: step: 394/1541, loss: 1.3216997385025024 2023-01-23 06:34:13.276520: step: 396/1541, loss: 4.005177974700928 2023-01-23 06:34:14.171780: step: 398/1541, loss: 0.6172772645950317 2023-01-23 06:34:15.051109: step: 400/1541, loss: 1.2522869110107422 2023-01-23 06:34:15.960445: step: 402/1541, loss: 0.36475852131843567 2023-01-23 06:34:16.809066: step: 404/1541, loss: 0.5917754173278809 2023-01-23 06:34:17.670634: step: 406/1541, loss: 0.7162378430366516 2023-01-23 06:34:18.551112: step: 408/1541, loss: 1.1949933767318726 2023-01-23 06:34:19.454830: step: 410/1541, loss: 0.7708856463432312 2023-01-23 06:34:20.306043: step: 412/1541, loss: 0.6216400861740112 2023-01-23 06:34:21.258120: step: 414/1541, loss: 0.45840492844581604 2023-01-23 06:34:22.146886: step: 416/1541, loss: 1.6394200325012207 2023-01-23 06:34:23.072395: step: 418/1541, loss: 1.0380512475967407 2023-01-23 06:34:23.958870: step: 420/1541, loss: 0.5850008726119995 2023-01-23 06:34:24.948368: step: 422/1541, loss: 1.5947394371032715 2023-01-23 06:34:25.876279: step: 424/1541, loss: 0.6470819711685181 2023-01-23 06:34:26.763645: step: 426/1541, loss: 0.9483911395072937 2023-01-23 06:34:27.651305: step: 428/1541, loss: 0.6316141486167908 2023-01-23 06:34:28.544247: step: 430/1541, loss: 1.012607216835022 2023-01-23 06:34:29.397995: step: 432/1541, loss: 0.6566970944404602 2023-01-23 06:34:30.263693: step: 434/1541, loss: 0.47875624895095825 2023-01-23 06:34:31.167540: step: 436/1541, loss: 2.0920639038085938 2023-01-23 06:34:32.061044: step: 438/1541, loss: 0.7471731305122375 2023-01-23 06:34:32.935639: step: 440/1541, loss: 0.697247326374054 2023-01-23 06:34:33.821066: step: 442/1541, loss: 0.7982158660888672 2023-01-23 06:34:34.707278: step: 444/1541, loss: 1.1100620031356812 2023-01-23 06:34:35.637094: step: 446/1541, loss: 1.5197736024856567 2023-01-23 06:34:36.522004: step: 448/1541, loss: 0.5840264558792114 2023-01-23 06:34:37.409342: step: 450/1541, loss: 0.4913291931152344 2023-01-23 06:34:38.326312: step: 452/1541, loss: 0.7512368559837341 2023-01-23 06:34:39.278887: step: 454/1541, loss: 0.7964184284210205 2023-01-23 06:34:40.154688: step: 456/1541, loss: 0.6492730975151062 2023-01-23 06:34:41.076443: step: 458/1541, loss: 0.5272347927093506 2023-01-23 06:34:41.997396: step: 460/1541, loss: 0.5266190767288208 2023-01-23 06:34:42.934004: step: 462/1541, loss: 0.9193801283836365 2023-01-23 06:34:43.867475: step: 464/1541, loss: 0.440056711435318 2023-01-23 06:34:44.828761: step: 466/1541, loss: 0.4729878902435303 2023-01-23 06:34:45.731714: step: 468/1541, loss: 1.3414725065231323 2023-01-23 06:34:46.650669: step: 470/1541, loss: 0.8989822268486023 2023-01-23 06:34:47.554965: step: 472/1541, loss: 2.8532488346099854 2023-01-23 06:34:48.432698: step: 474/1541, loss: 0.4537593722343445 2023-01-23 06:34:49.341318: step: 476/1541, loss: 0.8947107195854187 2023-01-23 06:34:50.271684: step: 478/1541, loss: 0.7705023884773254 2023-01-23 06:34:51.161027: step: 480/1541, loss: 0.4835634231567383 2023-01-23 06:34:52.048537: step: 482/1541, loss: 3.1738924980163574 2023-01-23 06:34:52.917016: step: 484/1541, loss: 1.4043534994125366 2023-01-23 06:34:53.829329: step: 486/1541, loss: 0.996874988079071 2023-01-23 06:34:54.707254: step: 488/1541, loss: 1.4699829816818237 2023-01-23 06:34:55.745695: step: 490/1541, loss: 0.8923912644386292 2023-01-23 06:34:56.610980: step: 492/1541, loss: 3.4237964153289795 2023-01-23 06:34:57.507442: step: 494/1541, loss: 0.44394996762275696 2023-01-23 06:34:58.381805: step: 496/1541, loss: 1.7398362159729004 2023-01-23 06:34:59.294229: step: 498/1541, loss: 1.0133665800094604 2023-01-23 06:35:00.168532: step: 500/1541, loss: 0.7163920402526855 2023-01-23 06:35:01.115149: step: 502/1541, loss: 0.7791969180107117 2023-01-23 06:35:02.005474: step: 504/1541, loss: 1.0614194869995117 2023-01-23 06:35:02.891592: step: 506/1541, loss: 2.6786699295043945 2023-01-23 06:35:03.788278: step: 508/1541, loss: 1.3708758354187012 2023-01-23 06:35:04.693194: step: 510/1541, loss: 0.5468690395355225 2023-01-23 06:35:05.564335: step: 512/1541, loss: 2.6236274242401123 2023-01-23 06:35:06.449533: step: 514/1541, loss: 0.9049848318099976 2023-01-23 06:35:07.355965: step: 516/1541, loss: 3.5157294273376465 2023-01-23 06:35:08.287868: step: 518/1541, loss: 0.5305950045585632 2023-01-23 06:35:09.215911: step: 520/1541, loss: 0.7530981302261353 2023-01-23 06:35:10.136788: step: 522/1541, loss: 0.5152182579040527 2023-01-23 06:35:11.048183: step: 524/1541, loss: 0.798346221446991 2023-01-23 06:35:11.960033: step: 526/1541, loss: 0.8648821115493774 2023-01-23 06:35:12.947230: step: 528/1541, loss: 0.9538264274597168 2023-01-23 06:35:13.833659: step: 530/1541, loss: 4.006590366363525 2023-01-23 06:35:14.745167: step: 532/1541, loss: 2.175889730453491 2023-01-23 06:35:15.654115: step: 534/1541, loss: 0.7774574756622314 2023-01-23 06:35:16.546830: step: 536/1541, loss: 0.2763115465641022 2023-01-23 06:35:17.477878: step: 538/1541, loss: 1.8338302373886108 2023-01-23 06:35:18.416760: step: 540/1541, loss: 0.3154491186141968 2023-01-23 06:35:19.335587: step: 542/1541, loss: 0.5840924382209778 2023-01-23 06:35:20.281019: step: 544/1541, loss: 3.078226089477539 2023-01-23 06:35:21.252905: step: 546/1541, loss: 0.5541974306106567 2023-01-23 06:35:22.130375: step: 548/1541, loss: 0.6786800026893616 2023-01-23 06:35:23.007540: step: 550/1541, loss: 0.754055380821228 2023-01-23 06:35:23.873855: step: 552/1541, loss: 3.416487455368042 2023-01-23 06:35:24.769840: step: 554/1541, loss: 0.36533838510513306 2023-01-23 06:35:25.742371: step: 556/1541, loss: 1.3422983884811401 2023-01-23 06:35:26.639520: step: 558/1541, loss: 0.637301504611969 2023-01-23 06:35:27.521863: step: 560/1541, loss: 0.6624619960784912 2023-01-23 06:35:28.438061: step: 562/1541, loss: 0.35710668563842773 2023-01-23 06:35:29.405926: step: 564/1541, loss: 0.5137689709663391 2023-01-23 06:35:30.285988: step: 566/1541, loss: 0.7968524098396301 2023-01-23 06:35:31.264213: step: 568/1541, loss: 1.4085010290145874 2023-01-23 06:35:32.232444: step: 570/1541, loss: 0.631669819355011 2023-01-23 06:35:33.120930: step: 572/1541, loss: 0.31435513496398926 2023-01-23 06:35:33.993959: step: 574/1541, loss: 2.1013975143432617 2023-01-23 06:35:34.867474: step: 576/1541, loss: 0.9474978446960449 2023-01-23 06:35:35.739632: step: 578/1541, loss: 0.44868558645248413 2023-01-23 06:35:36.728217: step: 580/1541, loss: 1.1124540567398071 2023-01-23 06:35:37.645115: step: 582/1541, loss: 0.8725740909576416 2023-01-23 06:35:38.602831: step: 584/1541, loss: 0.8870869874954224 2023-01-23 06:35:39.522173: step: 586/1541, loss: 0.7554352283477783 2023-01-23 06:35:40.411446: step: 588/1541, loss: 0.899314284324646 2023-01-23 06:35:41.249183: step: 590/1541, loss: 0.6401288509368896 2023-01-23 06:35:42.161648: step: 592/1541, loss: 0.8358885049819946 2023-01-23 06:35:43.077924: step: 594/1541, loss: 0.8909014463424683 2023-01-23 06:35:43.945415: step: 596/1541, loss: 0.408298522233963 2023-01-23 06:35:44.855546: step: 598/1541, loss: 0.9714298248291016 2023-01-23 06:35:45.692108: step: 600/1541, loss: 0.46236011385917664 2023-01-23 06:35:46.570774: step: 602/1541, loss: 0.7215027809143066 2023-01-23 06:35:47.467406: step: 604/1541, loss: 1.0369234085083008 2023-01-23 06:35:48.398116: step: 606/1541, loss: 0.7465509176254272 2023-01-23 06:35:49.271161: step: 608/1541, loss: 1.3381640911102295 2023-01-23 06:35:50.160095: step: 610/1541, loss: 0.98729407787323 2023-01-23 06:35:51.064677: step: 612/1541, loss: 0.7461050152778625 2023-01-23 06:35:51.986366: step: 614/1541, loss: 1.1920825242996216 2023-01-23 06:35:52.881552: step: 616/1541, loss: 1.1754859685897827 2023-01-23 06:35:53.769413: step: 618/1541, loss: 1.7083642482757568 2023-01-23 06:35:54.671377: step: 620/1541, loss: 0.7617034912109375 2023-01-23 06:35:55.607540: step: 622/1541, loss: 0.5889648199081421 2023-01-23 06:35:56.518418: step: 624/1541, loss: 0.13948610424995422 2023-01-23 06:35:57.445002: step: 626/1541, loss: 0.8557418584823608 2023-01-23 06:35:58.355483: step: 628/1541, loss: 0.48669129610061646 2023-01-23 06:35:59.246136: step: 630/1541, loss: 1.4227725267410278 2023-01-23 06:36:00.135921: step: 632/1541, loss: 0.6098557114601135 2023-01-23 06:36:01.022226: step: 634/1541, loss: 0.6447705030441284 2023-01-23 06:36:01.952935: step: 636/1541, loss: 0.8373202681541443 2023-01-23 06:36:02.824586: step: 638/1541, loss: 0.22098033130168915 2023-01-23 06:36:03.722281: step: 640/1541, loss: 0.3181632161140442 2023-01-23 06:36:04.657445: step: 642/1541, loss: 0.66607666015625 2023-01-23 06:36:05.568253: step: 644/1541, loss: 0.3276757001876831 2023-01-23 06:36:06.473205: step: 646/1541, loss: 1.4142816066741943 2023-01-23 06:36:07.320246: step: 648/1541, loss: 0.8085877895355225 2023-01-23 06:36:08.203937: step: 650/1541, loss: 1.0758000612258911 2023-01-23 06:36:09.125005: step: 652/1541, loss: 0.6916043162345886 2023-01-23 06:36:10.075151: step: 654/1541, loss: 0.9933962821960449 2023-01-23 06:36:10.930814: step: 656/1541, loss: 0.23200353980064392 2023-01-23 06:36:11.825779: step: 658/1541, loss: 1.361348271369934 2023-01-23 06:36:12.690558: step: 660/1541, loss: 0.2623150050640106 2023-01-23 06:36:13.585404: step: 662/1541, loss: 0.5078107714653015 2023-01-23 06:36:14.475146: step: 664/1541, loss: 1.2015784978866577 2023-01-23 06:36:15.403159: step: 666/1541, loss: 1.3190709352493286 2023-01-23 06:36:16.309397: step: 668/1541, loss: 0.554481029510498 2023-01-23 06:36:17.222244: step: 670/1541, loss: 0.6798027157783508 2023-01-23 06:36:18.104061: step: 672/1541, loss: 0.6627827286720276 2023-01-23 06:36:19.042069: step: 674/1541, loss: 1.1342592239379883 2023-01-23 06:36:19.926667: step: 676/1541, loss: 0.5056343674659729 2023-01-23 06:36:20.843618: step: 678/1541, loss: 0.8044824004173279 2023-01-23 06:36:21.775438: step: 680/1541, loss: 0.31065765023231506 2023-01-23 06:36:22.675796: step: 682/1541, loss: 0.41616711020469666 2023-01-23 06:36:23.554542: step: 684/1541, loss: 3.4564671516418457 2023-01-23 06:36:24.451657: step: 686/1541, loss: 0.7293639779090881 2023-01-23 06:36:25.377594: step: 688/1541, loss: 0.2908788025379181 2023-01-23 06:36:26.262000: step: 690/1541, loss: 1.0309430360794067 2023-01-23 06:36:27.165530: step: 692/1541, loss: 0.9942004084587097 2023-01-23 06:36:28.005495: step: 694/1541, loss: 0.3396347165107727 2023-01-23 06:36:28.881657: step: 696/1541, loss: 0.5631922483444214 2023-01-23 06:36:29.806249: step: 698/1541, loss: 1.3379002809524536 2023-01-23 06:36:30.704038: step: 700/1541, loss: 0.7414276599884033 2023-01-23 06:36:31.620414: step: 702/1541, loss: 0.7604549527168274 2023-01-23 06:36:32.453920: step: 704/1541, loss: 0.6501567363739014 2023-01-23 06:36:33.392616: step: 706/1541, loss: 0.6348217725753784 2023-01-23 06:36:34.332135: step: 708/1541, loss: 0.8372791409492493 2023-01-23 06:36:35.268091: step: 710/1541, loss: 0.3770727813243866 2023-01-23 06:36:36.168731: step: 712/1541, loss: 0.9591401219367981 2023-01-23 06:36:37.029797: step: 714/1541, loss: 0.632231593132019 2023-01-23 06:36:37.912768: step: 716/1541, loss: 0.6612465977668762 2023-01-23 06:36:38.808733: step: 718/1541, loss: 0.6960633993148804 2023-01-23 06:36:39.709460: step: 720/1541, loss: 0.9851177930831909 2023-01-23 06:36:40.585600: step: 722/1541, loss: 0.703617513179779 2023-01-23 06:36:41.488815: step: 724/1541, loss: 0.6601245999336243 2023-01-23 06:36:42.368605: step: 726/1541, loss: 0.8263582587242126 2023-01-23 06:36:43.285616: step: 728/1541, loss: 1.3387439250946045 2023-01-23 06:36:44.133523: step: 730/1541, loss: 0.7323122024536133 2023-01-23 06:36:45.049254: step: 732/1541, loss: 0.6870481967926025 2023-01-23 06:36:46.097671: step: 734/1541, loss: 0.8773958683013916 2023-01-23 06:36:47.011437: step: 736/1541, loss: 6.633556365966797 2023-01-23 06:36:47.874301: step: 738/1541, loss: 0.5397921800613403 2023-01-23 06:36:48.759129: step: 740/1541, loss: 1.535725712776184 2023-01-23 06:36:49.654875: step: 742/1541, loss: 0.3350277543067932 2023-01-23 06:36:50.548302: step: 744/1541, loss: 0.4680337607860565 2023-01-23 06:36:51.512343: step: 746/1541, loss: 0.673588752746582 2023-01-23 06:36:52.424098: step: 748/1541, loss: 0.7009245753288269 2023-01-23 06:36:53.326799: step: 750/1541, loss: 0.48030394315719604 2023-01-23 06:36:54.215487: step: 752/1541, loss: 1.7848973274230957 2023-01-23 06:36:55.070804: step: 754/1541, loss: 1.9428610801696777 2023-01-23 06:36:55.975242: step: 756/1541, loss: 3.347470760345459 2023-01-23 06:36:56.893215: step: 758/1541, loss: 0.8397004008293152 2023-01-23 06:36:57.814575: step: 760/1541, loss: 0.9368717670440674 2023-01-23 06:36:58.712523: step: 762/1541, loss: 1.3807293176651 2023-01-23 06:36:59.619076: step: 764/1541, loss: 0.9552587270736694 2023-01-23 06:37:00.503555: step: 766/1541, loss: 0.4495600163936615 2023-01-23 06:37:01.414814: step: 768/1541, loss: 0.8947204351425171 2023-01-23 06:37:02.350503: step: 770/1541, loss: 3.0233583450317383 2023-01-23 06:37:03.241832: step: 772/1541, loss: 2.7202672958374023 2023-01-23 06:37:04.128757: step: 774/1541, loss: 0.8332232236862183 2023-01-23 06:37:05.041723: step: 776/1541, loss: 0.6402210593223572 2023-01-23 06:37:06.011404: step: 778/1541, loss: 2.260681390762329 2023-01-23 06:37:06.944030: step: 780/1541, loss: 1.2467519044876099 2023-01-23 06:37:07.884086: step: 782/1541, loss: 0.7434388399124146 2023-01-23 06:37:08.800392: step: 784/1541, loss: 0.4986606538295746 2023-01-23 06:37:09.771952: step: 786/1541, loss: 1.1933550834655762 2023-01-23 06:37:10.667851: step: 788/1541, loss: 0.18682093918323517 2023-01-23 06:37:11.658104: step: 790/1541, loss: 0.3955092132091522 2023-01-23 06:37:12.582319: step: 792/1541, loss: 1.3172367811203003 2023-01-23 06:37:13.453464: step: 794/1541, loss: 0.30565303564071655 2023-01-23 06:37:14.351350: step: 796/1541, loss: 1.2619524002075195 2023-01-23 06:37:15.214611: step: 798/1541, loss: 0.4544592499732971 2023-01-23 06:37:16.149451: step: 800/1541, loss: 0.9383355379104614 2023-01-23 06:37:17.030597: step: 802/1541, loss: 0.8604469299316406 2023-01-23 06:37:17.945544: step: 804/1541, loss: 0.6515401601791382 2023-01-23 06:37:18.846825: step: 806/1541, loss: 2.038900136947632 2023-01-23 06:37:19.765298: step: 808/1541, loss: 0.3797001540660858 2023-01-23 06:37:20.668271: step: 810/1541, loss: 1.5281001329421997 2023-01-23 06:37:21.521845: step: 812/1541, loss: 0.4955388903617859 2023-01-23 06:37:22.410231: step: 814/1541, loss: 1.2224317789077759 2023-01-23 06:37:23.331610: step: 816/1541, loss: 0.7806413173675537 2023-01-23 06:37:24.240719: step: 818/1541, loss: 2.5399723052978516 2023-01-23 06:37:25.124879: step: 820/1541, loss: 1.013603687286377 2023-01-23 06:37:26.026722: step: 822/1541, loss: 0.5019419193267822 2023-01-23 06:37:26.978681: step: 824/1541, loss: 0.38051313161849976 2023-01-23 06:37:27.872218: step: 826/1541, loss: 0.7569939494132996 2023-01-23 06:37:28.753571: step: 828/1541, loss: 0.7205168008804321 2023-01-23 06:37:29.631550: step: 830/1541, loss: 1.6676883697509766 2023-01-23 06:37:30.534304: step: 832/1541, loss: 1.087617039680481 2023-01-23 06:37:31.436584: step: 834/1541, loss: 1.2500394582748413 2023-01-23 06:37:32.405392: step: 836/1541, loss: 1.1546094417572021 2023-01-23 06:37:33.280314: step: 838/1541, loss: 0.6801774501800537 2023-01-23 06:37:34.157542: step: 840/1541, loss: 0.7541422843933105 2023-01-23 06:37:35.087215: step: 842/1541, loss: 3.7423346042633057 2023-01-23 06:37:36.031180: step: 844/1541, loss: 0.39928463101387024 2023-01-23 06:37:36.869988: step: 846/1541, loss: 1.0472617149353027 2023-01-23 06:37:37.773005: step: 848/1541, loss: 0.6477592587471008 2023-01-23 06:37:38.657906: step: 850/1541, loss: 0.6414051651954651 2023-01-23 06:37:39.563725: step: 852/1541, loss: 0.7836608290672302 2023-01-23 06:37:40.461621: step: 854/1541, loss: 0.7534359693527222 2023-01-23 06:37:41.359142: step: 856/1541, loss: 0.18540631234645844 2023-01-23 06:37:42.231567: step: 858/1541, loss: 1.3248279094696045 2023-01-23 06:37:43.149029: step: 860/1541, loss: 0.8851649761199951 2023-01-23 06:37:44.051515: step: 862/1541, loss: 1.660304069519043 2023-01-23 06:37:44.962213: step: 864/1541, loss: 2.5207369327545166 2023-01-23 06:37:45.880763: step: 866/1541, loss: 0.534072756767273 2023-01-23 06:37:46.808554: step: 868/1541, loss: 2.380902051925659 2023-01-23 06:37:47.692363: step: 870/1541, loss: 0.686244785785675 2023-01-23 06:37:48.575130: step: 872/1541, loss: 1.0367966890335083 2023-01-23 06:37:49.519186: step: 874/1541, loss: 6.5264997482299805 2023-01-23 06:37:50.515440: step: 876/1541, loss: 1.1959048509597778 2023-01-23 06:37:51.428574: step: 878/1541, loss: 0.7215486168861389 2023-01-23 06:37:52.330380: step: 880/1541, loss: 1.3001707792282104 2023-01-23 06:37:53.203794: step: 882/1541, loss: 1.3308008909225464 2023-01-23 06:37:54.254599: step: 884/1541, loss: 0.6887264847755432 2023-01-23 06:37:55.168946: step: 886/1541, loss: 0.5065585374832153 2023-01-23 06:37:56.058165: step: 888/1541, loss: 0.2962415814399719 2023-01-23 06:37:56.951106: step: 890/1541, loss: 0.7430912852287292 2023-01-23 06:37:57.857875: step: 892/1541, loss: 0.6843736171722412 2023-01-23 06:37:58.747021: step: 894/1541, loss: 0.2542130649089813 2023-01-23 06:37:59.629318: step: 896/1541, loss: 1.2111656665802002 2023-01-23 06:38:00.548636: step: 898/1541, loss: 0.3245658874511719 2023-01-23 06:38:01.509043: step: 900/1541, loss: 0.8724348545074463 2023-01-23 06:38:02.390168: step: 902/1541, loss: 0.6499044895172119 2023-01-23 06:38:03.270001: step: 904/1541, loss: 1.0960031747817993 2023-01-23 06:38:04.132915: step: 906/1541, loss: 1.3300424814224243 2023-01-23 06:38:05.009824: step: 908/1541, loss: 0.40916121006011963 2023-01-23 06:38:05.915768: step: 910/1541, loss: 0.4001888632774353 2023-01-23 06:38:06.794112: step: 912/1541, loss: 0.9027525186538696 2023-01-23 06:38:07.688825: step: 914/1541, loss: 1.621448278427124 2023-01-23 06:38:08.575088: step: 916/1541, loss: 1.2934238910675049 2023-01-23 06:38:09.440786: step: 918/1541, loss: 0.690418541431427 2023-01-23 06:38:10.359773: step: 920/1541, loss: 0.5539952516555786 2023-01-23 06:38:11.233235: step: 922/1541, loss: 0.9075661897659302 2023-01-23 06:38:12.126451: step: 924/1541, loss: 0.3920205235481262 2023-01-23 06:38:13.016093: step: 926/1541, loss: 1.5430269241333008 2023-01-23 06:38:13.895391: step: 928/1541, loss: 0.8175615668296814 2023-01-23 06:38:14.850423: step: 930/1541, loss: 0.5922805070877075 2023-01-23 06:38:15.746097: step: 932/1541, loss: 0.7768229842185974 2023-01-23 06:38:16.603792: step: 934/1541, loss: 0.8548039197921753 2023-01-23 06:38:17.475819: step: 936/1541, loss: 0.26087629795074463 2023-01-23 06:38:18.373441: step: 938/1541, loss: 1.0018784999847412 2023-01-23 06:38:19.285160: step: 940/1541, loss: 0.5503091812133789 2023-01-23 06:38:20.238819: step: 942/1541, loss: 0.9606398344039917 2023-01-23 06:38:21.146941: step: 944/1541, loss: 0.68174809217453 2023-01-23 06:38:22.018382: step: 946/1541, loss: 0.38726723194122314 2023-01-23 06:38:22.911398: step: 948/1541, loss: 3.329343557357788 2023-01-23 06:38:23.902120: step: 950/1541, loss: 0.8625974655151367 2023-01-23 06:38:24.814915: step: 952/1541, loss: 1.0649571418762207 2023-01-23 06:38:25.702913: step: 954/1541, loss: 0.8920539617538452 2023-01-23 06:38:26.616586: step: 956/1541, loss: 0.6460154056549072 2023-01-23 06:38:27.516914: step: 958/1541, loss: 0.5250446200370789 2023-01-23 06:38:28.394980: step: 960/1541, loss: 0.8206812739372253 2023-01-23 06:38:29.305472: step: 962/1541, loss: 0.34842193126678467 2023-01-23 06:38:30.216456: step: 964/1541, loss: 0.40264326333999634 2023-01-23 06:38:31.088937: step: 966/1541, loss: 2.971804618835449 2023-01-23 06:38:32.002167: step: 968/1541, loss: 1.6197431087493896 2023-01-23 06:38:32.961370: step: 970/1541, loss: 0.5494863986968994 2023-01-23 06:38:33.888399: step: 972/1541, loss: 0.32241955399513245 2023-01-23 06:38:34.774721: step: 974/1541, loss: 1.186905860900879 2023-01-23 06:38:35.657884: step: 976/1541, loss: 0.7387056946754456 2023-01-23 06:38:36.658829: step: 978/1541, loss: 8.150591850280762 2023-01-23 06:38:37.560854: step: 980/1541, loss: 0.7029252648353577 2023-01-23 06:38:38.479823: step: 982/1541, loss: 0.47561296820640564 2023-01-23 06:38:39.362934: step: 984/1541, loss: 0.4368738830089569 2023-01-23 06:38:40.313864: step: 986/1541, loss: 0.8651410937309265 2023-01-23 06:38:41.234182: step: 988/1541, loss: 0.7926746606826782 2023-01-23 06:38:42.175262: step: 990/1541, loss: 1.1993060111999512 2023-01-23 06:38:43.070389: step: 992/1541, loss: 0.7819361090660095 2023-01-23 06:38:43.934154: step: 994/1541, loss: 0.6310235857963562 2023-01-23 06:38:44.800112: step: 996/1541, loss: 0.5255594849586487 2023-01-23 06:38:45.719166: step: 998/1541, loss: 1.083694577217102 2023-01-23 06:38:46.581205: step: 1000/1541, loss: 0.7084569931030273 2023-01-23 06:38:47.514367: step: 1002/1541, loss: 0.8388201594352722 2023-01-23 06:38:48.411560: step: 1004/1541, loss: 1.5757157802581787 2023-01-23 06:38:49.344892: step: 1006/1541, loss: 0.9095553159713745 2023-01-23 06:38:50.226620: step: 1008/1541, loss: 0.5020133256912231 2023-01-23 06:38:51.119814: step: 1010/1541, loss: 0.4844200611114502 2023-01-23 06:38:52.042390: step: 1012/1541, loss: 0.4789103865623474 2023-01-23 06:38:52.965608: step: 1014/1541, loss: 1.112331748008728 2023-01-23 06:38:53.929950: step: 1016/1541, loss: 6.550941467285156 2023-01-23 06:38:54.807716: step: 1018/1541, loss: 0.4833506941795349 2023-01-23 06:38:55.759311: step: 1020/1541, loss: 1.1959383487701416 2023-01-23 06:38:56.655719: step: 1022/1541, loss: 0.5884564518928528 2023-01-23 06:38:57.537787: step: 1024/1541, loss: 0.496226042509079 2023-01-23 06:38:58.453335: step: 1026/1541, loss: 0.41074153780937195 2023-01-23 06:38:59.315773: step: 1028/1541, loss: 0.6166683435440063 2023-01-23 06:39:00.212940: step: 1030/1541, loss: 0.45258471369743347 2023-01-23 06:39:01.090538: step: 1032/1541, loss: 1.4061435461044312 2023-01-23 06:39:02.031037: step: 1034/1541, loss: 1.1362452507019043 2023-01-23 06:39:02.966685: step: 1036/1541, loss: 0.5654826760292053 2023-01-23 06:39:03.871847: step: 1038/1541, loss: 0.8129411935806274 2023-01-23 06:39:04.812811: step: 1040/1541, loss: 0.44373252987861633 2023-01-23 06:39:05.712232: step: 1042/1541, loss: 7.480105400085449 2023-01-23 06:39:06.665694: step: 1044/1541, loss: 0.910639762878418 2023-01-23 06:39:07.589080: step: 1046/1541, loss: 0.6867595911026001 2023-01-23 06:39:08.494533: step: 1048/1541, loss: 0.3622855842113495 2023-01-23 06:39:09.366233: step: 1050/1541, loss: 1.0858131647109985 2023-01-23 06:39:10.322876: step: 1052/1541, loss: 2.563728094100952 2023-01-23 06:39:11.263809: step: 1054/1541, loss: 0.6461468935012817 2023-01-23 06:39:12.150006: step: 1056/1541, loss: 0.6977890729904175 2023-01-23 06:39:13.129439: step: 1058/1541, loss: 0.9903779029846191 2023-01-23 06:39:14.056776: step: 1060/1541, loss: 6.668257713317871 2023-01-23 06:39:14.955869: step: 1062/1541, loss: 0.36390164494514465 2023-01-23 06:39:15.856866: step: 1064/1541, loss: 1.0682647228240967 2023-01-23 06:39:16.734139: step: 1066/1541, loss: 0.9783334732055664 2023-01-23 06:39:17.667760: step: 1068/1541, loss: 2.527163505554199 2023-01-23 06:39:18.545256: step: 1070/1541, loss: 1.3171495199203491 2023-01-23 06:39:19.443800: step: 1072/1541, loss: 1.4351794719696045 2023-01-23 06:39:20.286025: step: 1074/1541, loss: 0.8191138505935669 2023-01-23 06:39:21.157240: step: 1076/1541, loss: 0.5290677547454834 2023-01-23 06:39:22.049977: step: 1078/1541, loss: 1.2164334058761597 2023-01-23 06:39:22.972966: step: 1080/1541, loss: 1.1370126008987427 2023-01-23 06:39:23.824108: step: 1082/1541, loss: 0.2627529501914978 2023-01-23 06:39:24.760472: step: 1084/1541, loss: 0.5583773255348206 2023-01-23 06:39:25.640576: step: 1086/1541, loss: 0.7035115957260132 2023-01-23 06:39:26.524205: step: 1088/1541, loss: 0.36526253819465637 2023-01-23 06:39:27.426004: step: 1090/1541, loss: 0.9213729500770569 2023-01-23 06:39:28.357699: step: 1092/1541, loss: 1.3191567659378052 2023-01-23 06:39:29.220665: step: 1094/1541, loss: 1.1162793636322021 2023-01-23 06:39:30.113187: step: 1096/1541, loss: 0.41165226697921753 2023-01-23 06:39:31.081989: step: 1098/1541, loss: 0.9362925887107849 2023-01-23 06:39:31.985053: step: 1100/1541, loss: 0.7798605561256409 2023-01-23 06:39:32.864181: step: 1102/1541, loss: 0.16487076878547668 2023-01-23 06:39:33.760509: step: 1104/1541, loss: 0.5935863256454468 2023-01-23 06:39:34.685146: step: 1106/1541, loss: 0.2811445891857147 2023-01-23 06:39:35.651930: step: 1108/1541, loss: 1.336443543434143 2023-01-23 06:39:36.583171: step: 1110/1541, loss: 0.6151405572891235 2023-01-23 06:39:37.489420: step: 1112/1541, loss: 0.3040907680988312 2023-01-23 06:39:38.371400: step: 1114/1541, loss: 0.37990301847457886 2023-01-23 06:39:39.249425: step: 1116/1541, loss: 0.4590398669242859 2023-01-23 06:39:40.125420: step: 1118/1541, loss: 0.2823704183101654 2023-01-23 06:39:41.009861: step: 1120/1541, loss: 0.239175945520401 2023-01-23 06:39:41.948233: step: 1122/1541, loss: 0.33194348216056824 2023-01-23 06:39:42.869321: step: 1124/1541, loss: 1.478451132774353 2023-01-23 06:39:43.845049: step: 1126/1541, loss: 0.4766128957271576 2023-01-23 06:39:44.727327: step: 1128/1541, loss: 0.5811964869499207 2023-01-23 06:39:45.711141: step: 1130/1541, loss: 1.25765860080719 2023-01-23 06:39:46.622603: step: 1132/1541, loss: 1.7227647304534912 2023-01-23 06:39:47.536783: step: 1134/1541, loss: 0.6634745001792908 2023-01-23 06:39:48.423207: step: 1136/1541, loss: 1.130824089050293 2023-01-23 06:39:49.324168: step: 1138/1541, loss: 0.42473703622817993 2023-01-23 06:39:50.202179: step: 1140/1541, loss: 0.5680400133132935 2023-01-23 06:39:51.082403: step: 1142/1541, loss: 1.0668765306472778 2023-01-23 06:39:52.021040: step: 1144/1541, loss: 0.6717916131019592 2023-01-23 06:39:52.890178: step: 1146/1541, loss: 2.8778998851776123 2023-01-23 06:39:53.810764: step: 1148/1541, loss: 0.3859432637691498 2023-01-23 06:39:54.690829: step: 1150/1541, loss: 0.5325082540512085 2023-01-23 06:39:55.564806: step: 1152/1541, loss: 0.4790201187133789 2023-01-23 06:39:56.436288: step: 1154/1541, loss: 4.024336338043213 2023-01-23 06:39:57.375300: step: 1156/1541, loss: 0.5917052030563354 2023-01-23 06:39:58.229004: step: 1158/1541, loss: 0.6210066080093384 2023-01-23 06:39:59.131915: step: 1160/1541, loss: 0.7825270295143127 2023-01-23 06:40:00.012077: step: 1162/1541, loss: 0.33168718218803406 2023-01-23 06:40:00.930159: step: 1164/1541, loss: 0.6984900832176208 2023-01-23 06:40:01.804273: step: 1166/1541, loss: 0.7672134637832642 2023-01-23 06:40:02.686685: step: 1168/1541, loss: 0.38762012124061584 2023-01-23 06:40:03.567819: step: 1170/1541, loss: 2.8251702785491943 2023-01-23 06:40:04.418567: step: 1172/1541, loss: 0.2193288803100586 2023-01-23 06:40:05.307865: step: 1174/1541, loss: 0.3036029040813446 2023-01-23 06:40:06.228317: step: 1176/1541, loss: 0.5794126391410828 2023-01-23 06:40:07.089773: step: 1178/1541, loss: 0.862725019454956 2023-01-23 06:40:07.960022: step: 1180/1541, loss: 0.9223181009292603 2023-01-23 06:40:08.875546: step: 1182/1541, loss: 0.4587200880050659 2023-01-23 06:40:09.839878: step: 1184/1541, loss: 0.7653606534004211 2023-01-23 06:40:10.733116: step: 1186/1541, loss: 0.4762817919254303 2023-01-23 06:40:11.630886: step: 1188/1541, loss: 0.7203959226608276 2023-01-23 06:40:12.545793: step: 1190/1541, loss: 1.0476033687591553 2023-01-23 06:40:13.442628: step: 1192/1541, loss: 0.3538758158683777 2023-01-23 06:40:14.305041: step: 1194/1541, loss: 1.6779236793518066 2023-01-23 06:40:15.161094: step: 1196/1541, loss: 0.8207522630691528 2023-01-23 06:40:16.032899: step: 1198/1541, loss: 0.4813753664493561 2023-01-23 06:40:16.922135: step: 1200/1541, loss: 0.6306105852127075 2023-01-23 06:40:17.846998: step: 1202/1541, loss: 0.5476863384246826 2023-01-23 06:40:18.724897: step: 1204/1541, loss: 0.6436626315116882 2023-01-23 06:40:19.601427: step: 1206/1541, loss: 0.6326565146446228 2023-01-23 06:40:20.504882: step: 1208/1541, loss: 0.4091580808162689 2023-01-23 06:40:21.448515: step: 1210/1541, loss: 0.6093223094940186 2023-01-23 06:40:22.303563: step: 1212/1541, loss: 1.7774516344070435 2023-01-23 06:40:23.170763: step: 1214/1541, loss: 0.8603309988975525 2023-01-23 06:40:24.043081: step: 1216/1541, loss: 1.3372783660888672 2023-01-23 06:40:25.057373: step: 1218/1541, loss: 0.8134014010429382 2023-01-23 06:40:25.970196: step: 1220/1541, loss: 1.1163023710250854 2023-01-23 06:40:26.945578: step: 1222/1541, loss: 2.958195209503174 2023-01-23 06:40:27.835114: step: 1224/1541, loss: 0.6675653457641602 2023-01-23 06:40:28.722474: step: 1226/1541, loss: 1.6563724279403687 2023-01-23 06:40:29.584509: step: 1228/1541, loss: 0.47434550523757935 2023-01-23 06:40:30.502427: step: 1230/1541, loss: 0.456096887588501 2023-01-23 06:40:31.403972: step: 1232/1541, loss: 0.7794057130813599 2023-01-23 06:40:32.353620: step: 1234/1541, loss: 0.8409135341644287 2023-01-23 06:40:33.261920: step: 1236/1541, loss: 0.8541877865791321 2023-01-23 06:40:34.150687: step: 1238/1541, loss: 1.323466181755066 2023-01-23 06:40:35.002238: step: 1240/1541, loss: 0.8012590408325195 2023-01-23 06:40:35.899238: step: 1242/1541, loss: 0.21444857120513916 2023-01-23 06:40:36.820140: step: 1244/1541, loss: 0.37619104981422424 2023-01-23 06:40:37.716557: step: 1246/1541, loss: 0.6623082160949707 2023-01-23 06:40:38.583198: step: 1248/1541, loss: 2.5941669940948486 2023-01-23 06:40:39.471510: step: 1250/1541, loss: 0.8530385494232178 2023-01-23 06:40:40.387408: step: 1252/1541, loss: 0.906473696231842 2023-01-23 06:40:41.275688: step: 1254/1541, loss: 3.1011195182800293 2023-01-23 06:40:42.238466: step: 1256/1541, loss: 1.3416824340820312 2023-01-23 06:40:43.166803: step: 1258/1541, loss: 1.5423755645751953 2023-01-23 06:40:44.051478: step: 1260/1541, loss: 1.0062613487243652 2023-01-23 06:40:44.972006: step: 1262/1541, loss: 0.6470316052436829 2023-01-23 06:40:45.838006: step: 1264/1541, loss: 0.7800775170326233 2023-01-23 06:40:46.744691: step: 1266/1541, loss: 0.7202046513557434 2023-01-23 06:40:47.627367: step: 1268/1541, loss: 1.559973120689392 2023-01-23 06:40:48.509570: step: 1270/1541, loss: 0.5851792693138123 2023-01-23 06:40:49.411062: step: 1272/1541, loss: 1.2402113676071167 2023-01-23 06:40:50.358276: step: 1274/1541, loss: 0.5063768029212952 2023-01-23 06:40:51.293298: step: 1276/1541, loss: 1.3856195211410522 2023-01-23 06:40:52.165377: step: 1278/1541, loss: 0.9234365224838257 2023-01-23 06:40:53.136661: step: 1280/1541, loss: 1.0044183731079102 2023-01-23 06:40:54.030419: step: 1282/1541, loss: 0.7752442359924316 2023-01-23 06:40:54.916297: step: 1284/1541, loss: 0.5995250344276428 2023-01-23 06:40:55.790236: step: 1286/1541, loss: 1.4523900747299194 2023-01-23 06:40:56.698825: step: 1288/1541, loss: 0.8580552935600281 2023-01-23 06:40:57.587023: step: 1290/1541, loss: 0.821552038192749 2023-01-23 06:40:58.468084: step: 1292/1541, loss: 0.4600581228733063 2023-01-23 06:40:59.329236: step: 1294/1541, loss: 1.086768627166748 2023-01-23 06:41:00.196857: step: 1296/1541, loss: 0.8653886318206787 2023-01-23 06:41:01.125590: step: 1298/1541, loss: 1.1257551908493042 2023-01-23 06:41:02.018863: step: 1300/1541, loss: 1.0901049375534058 2023-01-23 06:41:02.903708: step: 1302/1541, loss: 0.6320667862892151 2023-01-23 06:41:03.787688: step: 1304/1541, loss: 0.3242650032043457 2023-01-23 06:41:04.718611: step: 1306/1541, loss: 0.5490827560424805 2023-01-23 06:41:05.637032: step: 1308/1541, loss: 0.6563199162483215 2023-01-23 06:41:06.543973: step: 1310/1541, loss: 0.640777587890625 2023-01-23 06:41:07.450586: step: 1312/1541, loss: 1.1810389757156372 2023-01-23 06:41:08.420851: step: 1314/1541, loss: 1.3944340944290161 2023-01-23 06:41:09.299739: step: 1316/1541, loss: 0.526054859161377 2023-01-23 06:41:10.194579: step: 1318/1541, loss: 1.0973824262619019 2023-01-23 06:41:11.124518: step: 1320/1541, loss: 1.9307359457015991 2023-01-23 06:41:12.065692: step: 1322/1541, loss: 0.9011867046356201 2023-01-23 06:41:12.951660: step: 1324/1541, loss: 1.1226177215576172 2023-01-23 06:41:13.821760: step: 1326/1541, loss: 1.0631659030914307 2023-01-23 06:41:14.725451: step: 1328/1541, loss: 0.6454088091850281 2023-01-23 06:41:15.602679: step: 1330/1541, loss: 0.4392593502998352 2023-01-23 06:41:16.459938: step: 1332/1541, loss: 0.44268617033958435 2023-01-23 06:41:17.372396: step: 1334/1541, loss: 0.6177557110786438 2023-01-23 06:41:18.296977: step: 1336/1541, loss: 0.7014282941818237 2023-01-23 06:41:19.236295: step: 1338/1541, loss: 0.7027501463890076 2023-01-23 06:41:20.136185: step: 1340/1541, loss: 0.9388504028320312 2023-01-23 06:41:21.044856: step: 1342/1541, loss: 0.7662605047225952 2023-01-23 06:41:21.921852: step: 1344/1541, loss: 0.9016537070274353 2023-01-23 06:41:22.832064: step: 1346/1541, loss: 6.344604969024658 2023-01-23 06:41:23.750465: step: 1348/1541, loss: 0.4067622721195221 2023-01-23 06:41:24.616315: step: 1350/1541, loss: 0.33939245343208313 2023-01-23 06:41:25.538852: step: 1352/1541, loss: 0.30250486731529236 2023-01-23 06:41:26.423517: step: 1354/1541, loss: 0.8566956520080566 2023-01-23 06:41:27.331340: step: 1356/1541, loss: 0.7533009052276611 2023-01-23 06:41:28.284612: step: 1358/1541, loss: 1.5644490718841553 2023-01-23 06:41:29.263097: step: 1360/1541, loss: 1.0696879625320435 2023-01-23 06:41:30.151219: step: 1362/1541, loss: 2.1194145679473877 2023-01-23 06:41:31.109947: step: 1364/1541, loss: 0.3223094046115875 2023-01-23 06:41:32.025172: step: 1366/1541, loss: 0.8427702784538269 2023-01-23 06:41:32.918133: step: 1368/1541, loss: 0.8315945267677307 2023-01-23 06:41:33.827269: step: 1370/1541, loss: 0.31320831179618835 2023-01-23 06:41:34.698548: step: 1372/1541, loss: 0.4666074216365814 2023-01-23 06:41:35.628856: step: 1374/1541, loss: 1.0670888423919678 2023-01-23 06:41:36.511660: step: 1376/1541, loss: 3.4896719455718994 2023-01-23 06:41:37.401697: step: 1378/1541, loss: 0.4869098365306854 2023-01-23 06:41:38.268777: step: 1380/1541, loss: 2.070019245147705 2023-01-23 06:41:39.168453: step: 1382/1541, loss: 0.6784193515777588 2023-01-23 06:41:40.063291: step: 1384/1541, loss: 1.0659816265106201 2023-01-23 06:41:40.959377: step: 1386/1541, loss: 0.5443854331970215 2023-01-23 06:41:41.855806: step: 1388/1541, loss: 1.1356784105300903 2023-01-23 06:41:42.756610: step: 1390/1541, loss: 0.9178000092506409 2023-01-23 06:41:43.611481: step: 1392/1541, loss: 0.7222245931625366 2023-01-23 06:41:44.496867: step: 1394/1541, loss: 0.8255337476730347 2023-01-23 06:41:45.415570: step: 1396/1541, loss: 0.5143912434577942 2023-01-23 06:41:46.306699: step: 1398/1541, loss: 1.0076344013214111 2023-01-23 06:41:47.182592: step: 1400/1541, loss: 0.7720345258712769 2023-01-23 06:41:48.051668: step: 1402/1541, loss: 0.7272462844848633 2023-01-23 06:41:48.951244: step: 1404/1541, loss: 0.48593413829803467 2023-01-23 06:41:49.862468: step: 1406/1541, loss: 0.6469850540161133 2023-01-23 06:41:50.742031: step: 1408/1541, loss: 2.3056561946868896 2023-01-23 06:41:51.679753: step: 1410/1541, loss: 0.8910068273544312 2023-01-23 06:41:52.600916: step: 1412/1541, loss: 0.8780710697174072 2023-01-23 06:41:53.497576: step: 1414/1541, loss: 0.9808399677276611 2023-01-23 06:41:54.396809: step: 1416/1541, loss: 0.7661013603210449 2023-01-23 06:41:55.274132: step: 1418/1541, loss: 1.0284671783447266 2023-01-23 06:41:56.167204: step: 1420/1541, loss: 0.961997926235199 2023-01-23 06:41:57.063779: step: 1422/1541, loss: 0.17873947322368622 2023-01-23 06:41:58.033908: step: 1424/1541, loss: 0.7594978213310242 2023-01-23 06:41:58.961975: step: 1426/1541, loss: 0.6697920560836792 2023-01-23 06:41:59.872601: step: 1428/1541, loss: 0.5663468241691589 2023-01-23 06:42:00.806532: step: 1430/1541, loss: 1.130133867263794 2023-01-23 06:42:01.703231: step: 1432/1541, loss: 1.073093056678772 2023-01-23 06:42:02.594932: step: 1434/1541, loss: 0.4705008566379547 2023-01-23 06:42:03.449878: step: 1436/1541, loss: 0.8039664030075073 2023-01-23 06:42:04.322287: step: 1438/1541, loss: 1.5985350608825684 2023-01-23 06:42:05.177018: step: 1440/1541, loss: 1.5808451175689697 2023-01-23 06:42:06.049797: step: 1442/1541, loss: 0.5115436315536499 2023-01-23 06:42:06.994760: step: 1444/1541, loss: 3.5646426677703857 2023-01-23 06:42:07.916790: step: 1446/1541, loss: 0.7089202404022217 2023-01-23 06:42:08.829408: step: 1448/1541, loss: 0.740655779838562 2023-01-23 06:42:09.746866: step: 1450/1541, loss: 2.082674980163574 2023-01-23 06:42:10.711208: step: 1452/1541, loss: 0.5573066473007202 2023-01-23 06:42:11.594926: step: 1454/1541, loss: 0.634887158870697 2023-01-23 06:42:12.526482: step: 1456/1541, loss: 0.8172979950904846 2023-01-23 06:42:13.451324: step: 1458/1541, loss: 0.8032956719398499 2023-01-23 06:42:14.360968: step: 1460/1541, loss: 1.371992588043213 2023-01-23 06:42:15.286381: step: 1462/1541, loss: 0.6333311200141907 2023-01-23 06:42:16.165741: step: 1464/1541, loss: 1.964092493057251 2023-01-23 06:42:17.065772: step: 1466/1541, loss: 0.9582867622375488 2023-01-23 06:42:17.950732: step: 1468/1541, loss: 1.2367054224014282 2023-01-23 06:42:18.840228: step: 1470/1541, loss: 0.9340256452560425 2023-01-23 06:42:19.717243: step: 1472/1541, loss: 1.2223869562149048 2023-01-23 06:42:20.603810: step: 1474/1541, loss: 0.9550120830535889 2023-01-23 06:42:21.485206: step: 1476/1541, loss: 1.3080902099609375 2023-01-23 06:42:22.380242: step: 1478/1541, loss: 1.1334044933319092 2023-01-23 06:42:23.308690: step: 1480/1541, loss: 7.313815593719482 2023-01-23 06:42:24.151771: step: 1482/1541, loss: 0.4466223418712616 2023-01-23 06:42:25.035046: step: 1484/1541, loss: 0.6631645560264587 2023-01-23 06:42:25.911350: step: 1486/1541, loss: 1.0109248161315918 2023-01-23 06:42:26.773988: step: 1488/1541, loss: 0.422256737947464 2023-01-23 06:42:27.671095: step: 1490/1541, loss: 0.9566593766212463 2023-01-23 06:42:28.604441: step: 1492/1541, loss: 0.5436001420021057 2023-01-23 06:42:29.555999: step: 1494/1541, loss: 0.8141472935676575 2023-01-23 06:42:30.411982: step: 1496/1541, loss: 1.6539688110351562 2023-01-23 06:42:31.314487: step: 1498/1541, loss: 7.08626651763916 2023-01-23 06:42:32.215055: step: 1500/1541, loss: 1.2333731651306152 2023-01-23 06:42:33.086711: step: 1502/1541, loss: 0.3888709843158722 2023-01-23 06:42:33.988587: step: 1504/1541, loss: 0.48379969596862793 2023-01-23 06:42:34.910429: step: 1506/1541, loss: 0.7449113130569458 2023-01-23 06:42:35.798883: step: 1508/1541, loss: 2.9246153831481934 2023-01-23 06:42:36.648863: step: 1510/1541, loss: 1.8804465532302856 2023-01-23 06:42:37.569851: step: 1512/1541, loss: 0.7856930494308472 2023-01-23 06:42:38.514240: step: 1514/1541, loss: 0.717444896697998 2023-01-23 06:42:39.403563: step: 1516/1541, loss: 0.633033037185669 2023-01-23 06:42:40.260045: step: 1518/1541, loss: 0.40021830797195435 2023-01-23 06:42:41.140735: step: 1520/1541, loss: 0.9992275238037109 2023-01-23 06:42:42.103521: step: 1522/1541, loss: 0.9144898653030396 2023-01-23 06:42:42.973967: step: 1524/1541, loss: 2.057093620300293 2023-01-23 06:42:43.903444: step: 1526/1541, loss: 0.30468204617500305 2023-01-23 06:42:44.792877: step: 1528/1541, loss: 0.5112572908401489 2023-01-23 06:42:45.712221: step: 1530/1541, loss: 0.8991495370864868 2023-01-23 06:42:46.601062: step: 1532/1541, loss: 0.7687198519706726 2023-01-23 06:42:47.489685: step: 1534/1541, loss: 0.29980751872062683 2023-01-23 06:42:48.402020: step: 1536/1541, loss: 0.37098294496536255 2023-01-23 06:42:49.301807: step: 1538/1541, loss: 0.8133745193481445 2023-01-23 06:42:50.208009: step: 1540/1541, loss: 1.0281974077224731 2023-01-23 06:42:51.103208: step: 1542/1541, loss: 0.40977731347084045 2023-01-23 06:42:51.997035: step: 1544/1541, loss: 0.9924387335777283 2023-01-23 06:42:52.863633: step: 1546/1541, loss: 0.5139762759208679 2023-01-23 06:42:53.713485: step: 1548/1541, loss: 0.4511182904243469 2023-01-23 06:42:54.603454: step: 1550/1541, loss: 1.1202762126922607 2023-01-23 06:42:55.529831: step: 1552/1541, loss: 0.4969879388809204 2023-01-23 06:42:56.529194: step: 1554/1541, loss: 0.9411928057670593 2023-01-23 06:42:57.372435: step: 1556/1541, loss: 1.4240410327911377 2023-01-23 06:42:58.249628: step: 1558/1541, loss: 1.1257107257843018 2023-01-23 06:42:59.193873: step: 1560/1541, loss: 0.5979064106941223 2023-01-23 06:43:00.110291: step: 1562/1541, loss: 3.470090627670288 2023-01-23 06:43:00.983409: step: 1564/1541, loss: 0.4941713809967041 2023-01-23 06:43:01.916442: step: 1566/1541, loss: 0.9626144766807556 2023-01-23 06:43:02.770745: step: 1568/1541, loss: 1.1983435153961182 2023-01-23 06:43:03.711464: step: 1570/1541, loss: 0.5340893864631653 2023-01-23 06:43:04.601860: step: 1572/1541, loss: 1.266279697418213 2023-01-23 06:43:05.508549: step: 1574/1541, loss: 1.2241837978363037 2023-01-23 06:43:06.432556: step: 1576/1541, loss: 0.7842676043510437 2023-01-23 06:43:07.325201: step: 1578/1541, loss: 0.7088889479637146 2023-01-23 06:43:08.254460: step: 1580/1541, loss: 1.135498285293579 2023-01-23 06:43:09.145098: step: 1582/1541, loss: 0.4505389630794525 2023-01-23 06:43:10.032409: step: 1584/1541, loss: 0.723930299282074 2023-01-23 06:43:10.914108: step: 1586/1541, loss: 0.6556326150894165 2023-01-23 06:43:11.825626: step: 1588/1541, loss: 0.8788834810256958 2023-01-23 06:43:12.718225: step: 1590/1541, loss: 0.9694266319274902 2023-01-23 06:43:13.578855: step: 1592/1541, loss: 0.7478014230728149 2023-01-23 06:43:14.481238: step: 1594/1541, loss: 0.6551346182823181 2023-01-23 06:43:15.380322: step: 1596/1541, loss: 0.8724607229232788 2023-01-23 06:43:16.377745: step: 1598/1541, loss: 0.31827813386917114 2023-01-23 06:43:17.224398: step: 1600/1541, loss: 0.8635590672492981 2023-01-23 06:43:18.133911: step: 1602/1541, loss: 1.1757875680923462 2023-01-23 06:43:19.029617: step: 1604/1541, loss: 0.580451250076294 2023-01-23 06:43:19.963730: step: 1606/1541, loss: 0.34818533062934875 2023-01-23 06:43:20.940871: step: 1608/1541, loss: 0.32073915004730225 2023-01-23 06:43:21.848882: step: 1610/1541, loss: 0.41857674717903137 2023-01-23 06:43:22.720499: step: 1612/1541, loss: 0.782903254032135 2023-01-23 06:43:23.615690: step: 1614/1541, loss: 0.4035111963748932 2023-01-23 06:43:24.478655: step: 1616/1541, loss: 0.6869609355926514 2023-01-23 06:43:25.356780: step: 1618/1541, loss: 0.6783681511878967 2023-01-23 06:43:26.236281: step: 1620/1541, loss: 1.2410794496536255 2023-01-23 06:43:27.111398: step: 1622/1541, loss: 0.788941502571106 2023-01-23 06:43:27.996785: step: 1624/1541, loss: 0.6541446447372437 2023-01-23 06:43:28.935253: step: 1626/1541, loss: 0.35994499921798706 2023-01-23 06:43:29.808767: step: 1628/1541, loss: 0.48510295152664185 2023-01-23 06:43:30.652560: step: 1630/1541, loss: 2.4645349979400635 2023-01-23 06:43:31.558550: step: 1632/1541, loss: 0.43908342719078064 2023-01-23 06:43:32.469003: step: 1634/1541, loss: 0.41881099343299866 2023-01-23 06:43:33.376749: step: 1636/1541, loss: 1.0283589363098145 2023-01-23 06:43:34.271725: step: 1638/1541, loss: 0.40620994567871094 2023-01-23 06:43:35.155487: step: 1640/1541, loss: 0.570451021194458 2023-01-23 06:43:36.079433: step: 1642/1541, loss: 0.5948958992958069 2023-01-23 06:43:36.981419: step: 1644/1541, loss: 0.7888221144676208 2023-01-23 06:43:37.885456: step: 1646/1541, loss: 0.6132601499557495 2023-01-23 06:43:38.796158: step: 1648/1541, loss: 0.9577312469482422 2023-01-23 06:43:39.690187: step: 1650/1541, loss: 3.140507936477661 2023-01-23 06:43:40.581766: step: 1652/1541, loss: 0.7092705368995667 2023-01-23 06:43:41.544048: step: 1654/1541, loss: 0.7335548400878906 2023-01-23 06:43:42.473655: step: 1656/1541, loss: 1.1796858310699463 2023-01-23 06:43:43.371586: step: 1658/1541, loss: 0.38818299770355225 2023-01-23 06:43:44.242491: step: 1660/1541, loss: 0.7191767692565918 2023-01-23 06:43:45.150683: step: 1662/1541, loss: 1.162920355796814 2023-01-23 06:43:46.073631: step: 1664/1541, loss: 1.4865009784698486 2023-01-23 06:43:46.955043: step: 1666/1541, loss: 0.2001688927412033 2023-01-23 06:43:47.857846: step: 1668/1541, loss: 0.7983331680297852 2023-01-23 06:43:48.745603: step: 1670/1541, loss: 0.359075665473938 2023-01-23 06:43:49.611262: step: 1672/1541, loss: 0.6455905437469482 2023-01-23 06:43:50.470819: step: 1674/1541, loss: 0.5644928216934204 2023-01-23 06:43:51.332649: step: 1676/1541, loss: 0.7242259979248047 2023-01-23 06:43:52.214923: step: 1678/1541, loss: 0.3522080183029175 2023-01-23 06:43:53.129429: step: 1680/1541, loss: 0.46616804599761963 2023-01-23 06:43:54.029638: step: 1682/1541, loss: 1.052606225013733 2023-01-23 06:43:54.927122: step: 1684/1541, loss: 0.7505136132240295 2023-01-23 06:43:55.866394: step: 1686/1541, loss: 0.5082679986953735 2023-01-23 06:43:56.758705: step: 1688/1541, loss: 0.6031419634819031 2023-01-23 06:43:57.651513: step: 1690/1541, loss: 1.2580857276916504 2023-01-23 06:43:58.570059: step: 1692/1541, loss: 1.0642876625061035 2023-01-23 06:43:59.470754: step: 1694/1541, loss: 1.382574439048767 2023-01-23 06:44:00.377589: step: 1696/1541, loss: 0.36951741576194763 2023-01-23 06:44:01.228086: step: 1698/1541, loss: 0.4546344578266144 2023-01-23 06:44:02.183922: step: 1700/1541, loss: 1.0959447622299194 2023-01-23 06:44:03.098258: step: 1702/1541, loss: 0.7482584118843079 2023-01-23 06:44:03.997940: step: 1704/1541, loss: 1.002400279045105 2023-01-23 06:44:04.847753: step: 1706/1541, loss: 6.573192119598389 2023-01-23 06:44:05.788779: step: 1708/1541, loss: 0.3953841030597687 2023-01-23 06:44:06.704835: step: 1710/1541, loss: 0.2781665623188019 2023-01-23 06:44:07.554220: step: 1712/1541, loss: 0.35669955611228943 2023-01-23 06:44:08.456043: step: 1714/1541, loss: 0.8959910869598389 2023-01-23 06:44:09.368724: step: 1716/1541, loss: 0.5731189846992493 2023-01-23 06:44:10.261798: step: 1718/1541, loss: 0.8185200691223145 2023-01-23 06:44:11.143888: step: 1720/1541, loss: 0.4717133641242981 2023-01-23 06:44:12.027436: step: 1722/1541, loss: 0.4113573431968689 2023-01-23 06:44:12.918038: step: 1724/1541, loss: 0.9999281167984009 2023-01-23 06:44:13.849590: step: 1726/1541, loss: 0.8184273838996887 2023-01-23 06:44:14.710940: step: 1728/1541, loss: 0.4095836877822876 2023-01-23 06:44:15.662098: step: 1730/1541, loss: 1.3962668180465698 2023-01-23 06:44:16.563093: step: 1732/1541, loss: 0.47461432218551636 2023-01-23 06:44:17.495792: step: 1734/1541, loss: 1.6024272441864014 2023-01-23 06:44:18.384671: step: 1736/1541, loss: 1.3881564140319824 2023-01-23 06:44:19.271192: step: 1738/1541, loss: 0.21081581711769104 2023-01-23 06:44:20.168944: step: 1740/1541, loss: 0.8084633350372314 2023-01-23 06:44:21.053291: step: 1742/1541, loss: 0.7064900398254395 2023-01-23 06:44:21.976059: step: 1744/1541, loss: 1.317008137702942 2023-01-23 06:44:22.865652: step: 1746/1541, loss: 0.5309041142463684 2023-01-23 06:44:23.775520: step: 1748/1541, loss: 0.9540004730224609 2023-01-23 06:44:24.664007: step: 1750/1541, loss: 0.7069873809814453 2023-01-23 06:44:25.556857: step: 1752/1541, loss: 0.39982616901397705 2023-01-23 06:44:26.440573: step: 1754/1541, loss: 0.24366582930088043 2023-01-23 06:44:27.300024: step: 1756/1541, loss: 0.8532106876373291 2023-01-23 06:44:28.226918: step: 1758/1541, loss: 0.579615592956543 2023-01-23 06:44:29.218025: step: 1760/1541, loss: 1.247733235359192 2023-01-23 06:44:30.113521: step: 1762/1541, loss: 0.32083913683891296 2023-01-23 06:44:30.996495: step: 1764/1541, loss: 0.8600441217422485 2023-01-23 06:44:31.947778: step: 1766/1541, loss: 1.619818091392517 2023-01-23 06:44:32.802280: step: 1768/1541, loss: 0.6975657343864441 2023-01-23 06:44:33.666317: step: 1770/1541, loss: 1.2287274599075317 2023-01-23 06:44:34.571825: step: 1772/1541, loss: 0.8304265141487122 2023-01-23 06:44:35.478696: step: 1774/1541, loss: 0.8491084575653076 2023-01-23 06:44:36.378014: step: 1776/1541, loss: 2.594869375228882 2023-01-23 06:44:37.256989: step: 1778/1541, loss: 1.317665457725525 2023-01-23 06:44:38.153700: step: 1780/1541, loss: 0.9733105897903442 2023-01-23 06:44:39.082329: step: 1782/1541, loss: 0.47661468386650085 2023-01-23 06:44:39.950315: step: 1784/1541, loss: 0.5158620476722717 2023-01-23 06:44:40.857633: step: 1786/1541, loss: 1.0517373085021973 2023-01-23 06:44:41.776083: step: 1788/1541, loss: 1.099124789237976 2023-01-23 06:44:42.741860: step: 1790/1541, loss: 5.7982587814331055 2023-01-23 06:44:43.663399: step: 1792/1541, loss: 1.634935736656189 2023-01-23 06:44:44.551588: step: 1794/1541, loss: 0.4170394837856293 2023-01-23 06:44:45.487990: step: 1796/1541, loss: 2.3697502613067627 2023-01-23 06:44:46.417324: step: 1798/1541, loss: 1.0093731880187988 2023-01-23 06:44:47.345050: step: 1800/1541, loss: 7.092848777770996 2023-01-23 06:44:48.214511: step: 1802/1541, loss: 0.7251479625701904 2023-01-23 06:44:49.127086: step: 1804/1541, loss: 0.8825317025184631 2023-01-23 06:44:50.029312: step: 1806/1541, loss: 1.3414050340652466 2023-01-23 06:44:50.956118: step: 1808/1541, loss: 0.5975500345230103 2023-01-23 06:44:51.843646: step: 1810/1541, loss: 0.37770241498947144 2023-01-23 06:44:52.729684: step: 1812/1541, loss: 0.9841466546058655 2023-01-23 06:44:53.616726: step: 1814/1541, loss: 0.5182368159294128 2023-01-23 06:44:54.480568: step: 1816/1541, loss: 0.3320621848106384 2023-01-23 06:44:55.371562: step: 1818/1541, loss: 0.8915314674377441 2023-01-23 06:44:56.268063: step: 1820/1541, loss: 1.9552762508392334 2023-01-23 06:44:57.143923: step: 1822/1541, loss: 0.6504198312759399 2023-01-23 06:44:58.108802: step: 1824/1541, loss: 0.8742539882659912 2023-01-23 06:44:59.037048: step: 1826/1541, loss: 0.7448981404304504 2023-01-23 06:44:59.934503: step: 1828/1541, loss: 0.5565341711044312 2023-01-23 06:45:00.858543: step: 1830/1541, loss: 0.7557823061943054 2023-01-23 06:45:01.784467: step: 1832/1541, loss: 1.1458436250686646 2023-01-23 06:45:02.644443: step: 1834/1541, loss: 1.6013305187225342 2023-01-23 06:45:03.535900: step: 1836/1541, loss: 0.4829016923904419 2023-01-23 06:45:04.437868: step: 1838/1541, loss: 0.34596577286720276 2023-01-23 06:45:05.339229: step: 1840/1541, loss: 0.5238755345344543 2023-01-23 06:45:06.201146: step: 1842/1541, loss: 1.7837423086166382 2023-01-23 06:45:07.086701: step: 1844/1541, loss: 1.2454261779785156 2023-01-23 06:45:07.957403: step: 1846/1541, loss: 0.5019118189811707 2023-01-23 06:45:08.844733: step: 1848/1541, loss: 0.34921783208847046 2023-01-23 06:45:09.718719: step: 1850/1541, loss: 0.6981067061424255 2023-01-23 06:45:10.665654: step: 1852/1541, loss: 0.7654427289962769 2023-01-23 06:45:11.559527: step: 1854/1541, loss: 1.5890506505966187 2023-01-23 06:45:12.458757: step: 1856/1541, loss: 0.5461728572845459 2023-01-23 06:45:13.302898: step: 1858/1541, loss: 0.33652207255363464 2023-01-23 06:45:14.194141: step: 1860/1541, loss: 1.6316591501235962 2023-01-23 06:45:15.204051: step: 1862/1541, loss: 0.4873630106449127 2023-01-23 06:45:16.096656: step: 1864/1541, loss: 0.6730571985244751 2023-01-23 06:45:16.996609: step: 1866/1541, loss: 0.8691725730895996 2023-01-23 06:45:17.912725: step: 1868/1541, loss: 0.817023515701294 2023-01-23 06:45:18.804940: step: 1870/1541, loss: 0.7800642848014832 2023-01-23 06:45:19.683703: step: 1872/1541, loss: 0.5964586734771729 2023-01-23 06:45:20.592781: step: 1874/1541, loss: 1.0744729042053223 2023-01-23 06:45:21.481391: step: 1876/1541, loss: 0.8041605353355408 2023-01-23 06:45:22.333654: step: 1878/1541, loss: 0.29998984932899475 2023-01-23 06:45:23.245751: step: 1880/1541, loss: 0.7351741194725037 2023-01-23 06:45:24.128792: step: 1882/1541, loss: 1.0429216623306274 2023-01-23 06:45:25.022869: step: 1884/1541, loss: 0.44225335121154785 2023-01-23 06:45:25.897315: step: 1886/1541, loss: 0.16282057762145996 2023-01-23 06:45:26.759347: step: 1888/1541, loss: 2.3095247745513916 2023-01-23 06:45:27.690007: step: 1890/1541, loss: 1.3083707094192505 2023-01-23 06:45:28.558738: step: 1892/1541, loss: 0.4557702839374542 2023-01-23 06:45:29.465675: step: 1894/1541, loss: 0.33642464876174927 2023-01-23 06:45:30.345609: step: 1896/1541, loss: 1.3052973747253418 2023-01-23 06:45:31.233670: step: 1898/1541, loss: 0.5561493039131165 2023-01-23 06:45:32.242923: step: 1900/1541, loss: 0.654228687286377 2023-01-23 06:45:33.206827: step: 1902/1541, loss: 0.4576205015182495 2023-01-23 06:45:34.048703: step: 1904/1541, loss: 0.0960405170917511 2023-01-23 06:45:34.927286: step: 1906/1541, loss: 6.189648628234863 2023-01-23 06:45:35.822339: step: 1908/1541, loss: 1.3671081066131592 2023-01-23 06:45:36.703838: step: 1910/1541, loss: 0.5706384778022766 2023-01-23 06:45:37.598076: step: 1912/1541, loss: 0.5418373346328735 2023-01-23 06:45:38.535120: step: 1914/1541, loss: 0.6985363364219666 2023-01-23 06:45:39.420752: step: 1916/1541, loss: 0.2150004357099533 2023-01-23 06:45:40.303862: step: 1918/1541, loss: 0.33040741086006165 2023-01-23 06:45:41.243840: step: 1920/1541, loss: 1.4101399183273315 2023-01-23 06:45:42.161994: step: 1922/1541, loss: 0.17906847596168518 2023-01-23 06:45:43.037055: step: 1924/1541, loss: 0.4484371244907379 2023-01-23 06:45:43.942665: step: 1926/1541, loss: 0.42108088731765747 2023-01-23 06:45:44.813015: step: 1928/1541, loss: 0.8664386868476868 2023-01-23 06:45:45.723700: step: 1930/1541, loss: 0.34574997425079346 2023-01-23 06:45:46.665047: step: 1932/1541, loss: 0.51533442735672 2023-01-23 06:45:47.540184: step: 1934/1541, loss: 0.7694912552833557 2023-01-23 06:45:48.473532: step: 1936/1541, loss: 0.8365513682365417 2023-01-23 06:45:49.342294: step: 1938/1541, loss: 0.40376168489456177 2023-01-23 06:45:50.251040: step: 1940/1541, loss: 0.6502347588539124 2023-01-23 06:45:51.129396: step: 1942/1541, loss: 0.6952436566352844 2023-01-23 06:45:51.998076: step: 1944/1541, loss: 0.4855495095252991 2023-01-23 06:45:52.925284: step: 1946/1541, loss: 1.8278247117996216 2023-01-23 06:45:53.793231: step: 1948/1541, loss: 0.6041792035102844 2023-01-23 06:45:54.692768: step: 1950/1541, loss: 1.101405143737793 2023-01-23 06:45:55.577678: step: 1952/1541, loss: 1.141900658607483 2023-01-23 06:45:56.475183: step: 1954/1541, loss: 0.7933424115180969 2023-01-23 06:45:57.323752: step: 1956/1541, loss: 0.4659547805786133 2023-01-23 06:45:58.223266: step: 1958/1541, loss: 0.5512209534645081 2023-01-23 06:45:59.147437: step: 1960/1541, loss: 0.6409626603126526 2023-01-23 06:46:00.023619: step: 1962/1541, loss: 0.7904356122016907 2023-01-23 06:46:00.903107: step: 1964/1541, loss: 0.7811813950538635 2023-01-23 06:46:01.804247: step: 1966/1541, loss: 0.5294086337089539 2023-01-23 06:46:02.689497: step: 1968/1541, loss: 0.7754720449447632 2023-01-23 06:46:03.593392: step: 1970/1541, loss: 0.719467282295227 2023-01-23 06:46:04.449750: step: 1972/1541, loss: 0.7607446312904358 2023-01-23 06:46:05.335038: step: 1974/1541, loss: 0.2168852537870407 2023-01-23 06:46:06.196447: step: 1976/1541, loss: 0.7915745973587036 2023-01-23 06:46:07.090662: step: 1978/1541, loss: 0.813129186630249 2023-01-23 06:46:08.015741: step: 1980/1541, loss: 0.3096815347671509 2023-01-23 06:46:08.931682: step: 1982/1541, loss: 0.6333261728286743 2023-01-23 06:46:09.867673: step: 1984/1541, loss: 0.7561307549476624 2023-01-23 06:46:10.790043: step: 1986/1541, loss: 0.7025243043899536 2023-01-23 06:46:11.687466: step: 1988/1541, loss: 1.0701054334640503 2023-01-23 06:46:12.567271: step: 1990/1541, loss: 0.577343761920929 2023-01-23 06:46:13.572070: step: 1992/1541, loss: 0.7164175510406494 2023-01-23 06:46:14.488622: step: 1994/1541, loss: 1.1360938549041748 2023-01-23 06:46:15.427182: step: 1996/1541, loss: 0.26391446590423584 2023-01-23 06:46:16.337539: step: 1998/1541, loss: 1.2744789123535156 2023-01-23 06:46:17.221010: step: 2000/1541, loss: 0.295468270778656 2023-01-23 06:46:18.112291: step: 2002/1541, loss: 0.6309043169021606 2023-01-23 06:46:19.010481: step: 2004/1541, loss: 1.3001328706741333 2023-01-23 06:46:19.885319: step: 2006/1541, loss: 0.41920721530914307 2023-01-23 06:46:20.764794: step: 2008/1541, loss: 0.2663883864879608 2023-01-23 06:46:21.697790: step: 2010/1541, loss: 0.42955198884010315 2023-01-23 06:46:22.557964: step: 2012/1541, loss: 2.7160418033599854 2023-01-23 06:46:23.484237: step: 2014/1541, loss: 3.310499668121338 2023-01-23 06:46:24.397703: step: 2016/1541, loss: 0.8261672258377075 2023-01-23 06:46:25.292120: step: 2018/1541, loss: 1.429916262626648 2023-01-23 06:46:26.173550: step: 2020/1541, loss: 1.1077749729156494 2023-01-23 06:46:27.138422: step: 2022/1541, loss: 1.1991369724273682 2023-01-23 06:46:28.001317: step: 2024/1541, loss: 0.9557051658630371 2023-01-23 06:46:28.899860: step: 2026/1541, loss: 0.8093137741088867 2023-01-23 06:46:29.781546: step: 2028/1541, loss: 1.1060272455215454 2023-01-23 06:46:30.738582: step: 2030/1541, loss: 1.009509563446045 2023-01-23 06:46:31.721126: step: 2032/1541, loss: 1.4744184017181396 2023-01-23 06:46:32.686489: step: 2034/1541, loss: 0.6003499031066895 2023-01-23 06:46:33.550133: step: 2036/1541, loss: 0.3755190372467041 2023-01-23 06:46:34.458712: step: 2038/1541, loss: 0.9294828176498413 2023-01-23 06:46:35.398654: step: 2040/1541, loss: 1.145634412765503 2023-01-23 06:46:36.327379: step: 2042/1541, loss: 0.7142961025238037 2023-01-23 06:46:37.212326: step: 2044/1541, loss: 1.344887137413025 2023-01-23 06:46:38.145981: step: 2046/1541, loss: 1.3894096612930298 2023-01-23 06:46:39.062483: step: 2048/1541, loss: 0.8021796941757202 2023-01-23 06:46:39.922006: step: 2050/1541, loss: 0.7147513031959534 2023-01-23 06:46:40.872859: step: 2052/1541, loss: 0.661957323551178 2023-01-23 06:46:41.838947: step: 2054/1541, loss: 0.3601118326187134 2023-01-23 06:46:42.724693: step: 2056/1541, loss: 0.7905454039573669 2023-01-23 06:46:43.612532: step: 2058/1541, loss: 0.403897762298584 2023-01-23 06:46:44.560334: step: 2060/1541, loss: 3.404750347137451 2023-01-23 06:46:45.474960: step: 2062/1541, loss: 1.1059650182724 2023-01-23 06:46:46.323203: step: 2064/1541, loss: 0.5784094333648682 2023-01-23 06:46:47.169095: step: 2066/1541, loss: 1.0413938760757446 2023-01-23 06:46:48.074325: step: 2068/1541, loss: 0.5024186372756958 2023-01-23 06:46:48.982831: step: 2070/1541, loss: 1.594160556793213 2023-01-23 06:46:49.835246: step: 2072/1541, loss: 0.617767333984375 2023-01-23 06:46:50.730189: step: 2074/1541, loss: 0.41172555088996887 2023-01-23 06:46:51.630390: step: 2076/1541, loss: 0.7586117386817932 2023-01-23 06:46:52.559331: step: 2078/1541, loss: 2.8168632984161377 2023-01-23 06:46:53.400947: step: 2080/1541, loss: 0.48010382056236267 2023-01-23 06:46:54.308625: step: 2082/1541, loss: 0.6693029403686523 2023-01-23 06:46:55.221723: step: 2084/1541, loss: 3.1300134658813477 2023-01-23 06:46:56.109310: step: 2086/1541, loss: 0.3899635076522827 2023-01-23 06:46:57.020443: step: 2088/1541, loss: 0.7208084464073181 2023-01-23 06:46:57.986934: step: 2090/1541, loss: 0.7373349070549011 2023-01-23 06:46:58.881814: step: 2092/1541, loss: 0.6634597778320312 2023-01-23 06:46:59.765465: step: 2094/1541, loss: 0.8615911602973938 2023-01-23 06:47:00.691779: step: 2096/1541, loss: 0.6388450264930725 2023-01-23 06:47:01.681982: step: 2098/1541, loss: 1.964943528175354 2023-01-23 06:47:02.613177: step: 2100/1541, loss: 0.7074921131134033 2023-01-23 06:47:03.496024: step: 2102/1541, loss: 0.7674378156661987 2023-01-23 06:47:04.357475: step: 2104/1541, loss: 0.3850720524787903 2023-01-23 06:47:05.284887: step: 2106/1541, loss: 0.8928897380828857 2023-01-23 06:47:06.181522: step: 2108/1541, loss: 0.6871625781059265 2023-01-23 06:47:07.136265: step: 2110/1541, loss: 1.1400734186172485 2023-01-23 06:47:08.063143: step: 2112/1541, loss: 0.42630892992019653 2023-01-23 06:47:08.940795: step: 2114/1541, loss: 0.718319296836853 2023-01-23 06:47:09.823527: step: 2116/1541, loss: 0.5910254120826721 2023-01-23 06:47:10.719448: step: 2118/1541, loss: 0.7382182478904724 2023-01-23 06:47:11.627483: step: 2120/1541, loss: 1.6141372919082642 2023-01-23 06:47:12.587135: step: 2122/1541, loss: 1.195786952972412 2023-01-23 06:47:13.494991: step: 2124/1541, loss: 0.6456395983695984 2023-01-23 06:47:14.370896: step: 2126/1541, loss: 1.4975507259368896 2023-01-23 06:47:15.237682: step: 2128/1541, loss: 0.6074206233024597 2023-01-23 06:47:16.123118: step: 2130/1541, loss: 0.9634552001953125 2023-01-23 06:47:17.065023: step: 2132/1541, loss: 0.977846086025238 2023-01-23 06:47:18.009559: step: 2134/1541, loss: 2.2388906478881836 2023-01-23 06:47:18.942243: step: 2136/1541, loss: 0.3556741178035736 2023-01-23 06:47:19.841950: step: 2138/1541, loss: 0.5068196654319763 2023-01-23 06:47:20.824634: step: 2140/1541, loss: 0.9873697757720947 2023-01-23 06:47:21.730910: step: 2142/1541, loss: 0.5732144713401794 2023-01-23 06:47:22.582572: step: 2144/1541, loss: 0.27571380138397217 2023-01-23 06:47:23.477293: step: 2146/1541, loss: 1.4703731536865234 2023-01-23 06:47:24.416537: step: 2148/1541, loss: 0.6788676381111145 2023-01-23 06:47:25.345791: step: 2150/1541, loss: 1.484250783920288 2023-01-23 06:47:26.247805: step: 2152/1541, loss: 0.8077383041381836 2023-01-23 06:47:27.131939: step: 2154/1541, loss: 3.955021381378174 2023-01-23 06:47:28.139035: step: 2156/1541, loss: 0.7304299473762512 2023-01-23 06:47:29.130442: step: 2158/1541, loss: 0.706314742565155 2023-01-23 06:47:30.094932: step: 2160/1541, loss: 7.180294036865234 2023-01-23 06:47:30.984588: step: 2162/1541, loss: 0.32839348912239075 2023-01-23 06:47:31.875533: step: 2164/1541, loss: 0.7828777432441711 2023-01-23 06:47:32.795764: step: 2166/1541, loss: 0.747093677520752 2023-01-23 06:47:33.657227: step: 2168/1541, loss: 1.3475550413131714 2023-01-23 06:47:34.589414: step: 2170/1541, loss: 0.4960767328739166 2023-01-23 06:47:35.491968: step: 2172/1541, loss: 0.2756556272506714 2023-01-23 06:47:36.397346: step: 2174/1541, loss: 0.8334110975265503 2023-01-23 06:47:37.306707: step: 2176/1541, loss: 0.3485627770423889 2023-01-23 06:47:38.263322: step: 2178/1541, loss: 0.361382395029068 2023-01-23 06:47:39.163581: step: 2180/1541, loss: 1.2683204412460327 2023-01-23 06:47:40.161630: step: 2182/1541, loss: 0.41138771176338196 2023-01-23 06:47:41.042687: step: 2184/1541, loss: 0.6197523474693298 2023-01-23 06:47:41.915417: step: 2186/1541, loss: 0.7227849960327148 2023-01-23 06:47:42.806059: step: 2188/1541, loss: 1.6981453895568848 2023-01-23 06:47:43.702896: step: 2190/1541, loss: 0.38624852895736694 2023-01-23 06:47:44.642046: step: 2192/1541, loss: 1.2912828922271729 2023-01-23 06:47:45.564011: step: 2194/1541, loss: 1.3072032928466797 2023-01-23 06:47:46.531144: step: 2196/1541, loss: 0.9904956221580505 2023-01-23 06:47:47.458984: step: 2198/1541, loss: 1.5974173545837402 2023-01-23 06:47:48.359983: step: 2200/1541, loss: 0.9502768516540527 2023-01-23 06:47:49.282711: step: 2202/1541, loss: 0.653713583946228 2023-01-23 06:47:50.186701: step: 2204/1541, loss: 0.6110556125640869 2023-01-23 06:47:51.133311: step: 2206/1541, loss: 1.4189484119415283 2023-01-23 06:47:52.062669: step: 2208/1541, loss: 0.6562969088554382 2023-01-23 06:47:52.967623: step: 2210/1541, loss: 0.26135319471359253 2023-01-23 06:47:53.828639: step: 2212/1541, loss: 1.0444345474243164 2023-01-23 06:47:54.791578: step: 2214/1541, loss: 0.7713236212730408 2023-01-23 06:47:55.673182: step: 2216/1541, loss: 0.43206608295440674 2023-01-23 06:47:56.559852: step: 2218/1541, loss: 0.6467539668083191 2023-01-23 06:47:57.444062: step: 2220/1541, loss: 0.5409233570098877 2023-01-23 06:47:58.416689: step: 2222/1541, loss: 0.8525407910346985 2023-01-23 06:47:59.304791: step: 2224/1541, loss: 0.4863995909690857 2023-01-23 06:48:00.181371: step: 2226/1541, loss: 0.7053370475769043 2023-01-23 06:48:01.099955: step: 2228/1541, loss: 1.0763587951660156 2023-01-23 06:48:01.962298: step: 2230/1541, loss: 0.8343582153320312 2023-01-23 06:48:02.876896: step: 2232/1541, loss: 1.105286717414856 2023-01-23 06:48:03.787460: step: 2234/1541, loss: 0.3825916349887848 2023-01-23 06:48:04.697744: step: 2236/1541, loss: 1.6528966426849365 2023-01-23 06:48:05.665330: step: 2238/1541, loss: 0.3701210618019104 2023-01-23 06:48:06.567015: step: 2240/1541, loss: 2.23374605178833 2023-01-23 06:48:07.477181: step: 2242/1541, loss: 2.0480916500091553 2023-01-23 06:48:08.366679: step: 2244/1541, loss: 0.8383574485778809 2023-01-23 06:48:09.237101: step: 2246/1541, loss: 1.2132318019866943 2023-01-23 06:48:10.132876: step: 2248/1541, loss: 0.5360615849494934 2023-01-23 06:48:11.023450: step: 2250/1541, loss: 1.1722114086151123 2023-01-23 06:48:11.907287: step: 2252/1541, loss: 0.904130756855011 2023-01-23 06:48:12.828526: step: 2254/1541, loss: 0.9186338186264038 2023-01-23 06:48:13.696377: step: 2256/1541, loss: 0.2685052454471588 2023-01-23 06:48:14.692577: step: 2258/1541, loss: 0.38748395442962646 2023-01-23 06:48:15.622724: step: 2260/1541, loss: 0.5476024150848389 2023-01-23 06:48:16.538067: step: 2262/1541, loss: 0.3833293616771698 2023-01-23 06:48:17.430328: step: 2264/1541, loss: 0.5030086636543274 2023-01-23 06:48:18.343087: step: 2266/1541, loss: 0.6720563769340515 2023-01-23 06:48:19.257469: step: 2268/1541, loss: 0.8800429701805115 2023-01-23 06:48:20.153964: step: 2270/1541, loss: 0.9969038367271423 2023-01-23 06:48:21.070334: step: 2272/1541, loss: 0.5195922255516052 2023-01-23 06:48:22.073335: step: 2274/1541, loss: 0.9625077247619629 2023-01-23 06:48:22.946859: step: 2276/1541, loss: 0.7369780540466309 2023-01-23 06:48:23.834645: step: 2278/1541, loss: 7.2213053703308105 2023-01-23 06:48:24.728901: step: 2280/1541, loss: 0.9860828518867493 2023-01-23 06:48:25.661271: step: 2282/1541, loss: 1.3054791688919067 2023-01-23 06:48:26.564938: step: 2284/1541, loss: 0.9645805954933167 2023-01-23 06:48:27.420835: step: 2286/1541, loss: 0.2769094407558441 2023-01-23 06:48:28.412281: step: 2288/1541, loss: 0.6319923996925354 2023-01-23 06:48:29.298856: step: 2290/1541, loss: 0.9443451166152954 2023-01-23 06:48:30.180046: step: 2292/1541, loss: 3.4578206539154053 2023-01-23 06:48:31.084813: step: 2294/1541, loss: 1.1043415069580078 2023-01-23 06:48:31.960030: step: 2296/1541, loss: 2.1255486011505127 2023-01-23 06:48:32.861819: step: 2298/1541, loss: 0.5311383605003357 2023-01-23 06:48:33.800840: step: 2300/1541, loss: 1.5421456098556519 2023-01-23 06:48:34.728448: step: 2302/1541, loss: 0.515013575553894 2023-01-23 06:48:35.592493: step: 2304/1541, loss: 0.7344585061073303 2023-01-23 06:48:36.467676: step: 2306/1541, loss: 10.514425277709961 2023-01-23 06:48:37.376624: step: 2308/1541, loss: 0.6702826619148254 2023-01-23 06:48:38.218391: step: 2310/1541, loss: 0.1995503455400467 2023-01-23 06:48:39.219690: step: 2312/1541, loss: 0.8221060633659363 2023-01-23 06:48:40.121802: step: 2314/1541, loss: 0.7845015525817871 2023-01-23 06:48:41.024446: step: 2316/1541, loss: 0.9387956857681274 2023-01-23 06:48:41.942024: step: 2318/1541, loss: 3.123007297515869 2023-01-23 06:48:42.838044: step: 2320/1541, loss: 0.5456344485282898 2023-01-23 06:48:43.809385: step: 2322/1541, loss: 1.476929783821106 2023-01-23 06:48:44.712780: step: 2324/1541, loss: 0.6001620292663574 2023-01-23 06:48:45.648400: step: 2326/1541, loss: 1.1269776821136475 2023-01-23 06:48:46.535182: step: 2328/1541, loss: 0.9226977229118347 2023-01-23 06:48:47.411603: step: 2330/1541, loss: 0.4558781087398529 2023-01-23 06:48:48.277353: step: 2332/1541, loss: 0.5255179405212402 2023-01-23 06:48:49.153419: step: 2334/1541, loss: 0.1410297453403473 2023-01-23 06:48:50.064773: step: 2336/1541, loss: 0.5549169182777405 2023-01-23 06:48:50.954661: step: 2338/1541, loss: 1.0958492755889893 2023-01-23 06:48:51.882705: step: 2340/1541, loss: 1.055955410003662 2023-01-23 06:48:52.740979: step: 2342/1541, loss: 0.5798937082290649 2023-01-23 06:48:53.666406: step: 2344/1541, loss: 0.8050045967102051 2023-01-23 06:48:54.544833: step: 2346/1541, loss: 0.7817032933235168 2023-01-23 06:48:55.428191: step: 2348/1541, loss: 0.2228207290172577 2023-01-23 06:48:56.286775: step: 2350/1541, loss: 0.6391920447349548 2023-01-23 06:48:57.137150: step: 2352/1541, loss: 0.2221401184797287 2023-01-23 06:48:58.089655: step: 2354/1541, loss: 0.6865585446357727 2023-01-23 06:48:58.984955: step: 2356/1541, loss: 0.8028790354728699 2023-01-23 06:48:59.871108: step: 2358/1541, loss: 0.9837208390235901 2023-01-23 06:49:00.807487: step: 2360/1541, loss: 0.8199555277824402 2023-01-23 06:49:01.751545: step: 2362/1541, loss: 0.4669540524482727 2023-01-23 06:49:02.645380: step: 2364/1541, loss: 0.43838271498680115 2023-01-23 06:49:03.548902: step: 2366/1541, loss: 0.8382354974746704 2023-01-23 06:49:04.446487: step: 2368/1541, loss: 0.32458382844924927 2023-01-23 06:49:05.338379: step: 2370/1541, loss: 1.6547236442565918 2023-01-23 06:49:06.216908: step: 2372/1541, loss: 0.8494725823402405 2023-01-23 06:49:07.076820: step: 2374/1541, loss: 3.430144786834717 2023-01-23 06:49:07.965715: step: 2376/1541, loss: 0.7241089940071106 2023-01-23 06:49:08.865375: step: 2378/1541, loss: 0.49305200576782227 2023-01-23 06:49:09.756180: step: 2380/1541, loss: 0.8892881274223328 2023-01-23 06:49:10.627613: step: 2382/1541, loss: 0.913061797618866 2023-01-23 06:49:11.491844: step: 2384/1541, loss: 0.8562459945678711 2023-01-23 06:49:12.409105: step: 2386/1541, loss: 0.7059752941131592 2023-01-23 06:49:13.410188: step: 2388/1541, loss: 0.8547857403755188 2023-01-23 06:49:14.303523: step: 2390/1541, loss: 3.1336841583251953 2023-01-23 06:49:15.196866: step: 2392/1541, loss: 0.41992053389549255 2023-01-23 06:49:16.144681: step: 2394/1541, loss: 1.1703267097473145 2023-01-23 06:49:17.026293: step: 2396/1541, loss: 0.6547970175743103 2023-01-23 06:49:17.893441: step: 2398/1541, loss: 1.0892293453216553 2023-01-23 06:49:18.783922: step: 2400/1541, loss: 3.1275923252105713 2023-01-23 06:49:19.696442: step: 2402/1541, loss: 0.3262634575366974 2023-01-23 06:49:20.586496: step: 2404/1541, loss: 0.35555732250213623 2023-01-23 06:49:21.532717: step: 2406/1541, loss: 0.5838974118232727 2023-01-23 06:49:22.398280: step: 2408/1541, loss: 0.6600779294967651 2023-01-23 06:49:23.272705: step: 2410/1541, loss: 0.5039718747138977 2023-01-23 06:49:24.193749: step: 2412/1541, loss: 0.4329797029495239 2023-01-23 06:49:25.072693: step: 2414/1541, loss: 0.4329453110694885 2023-01-23 06:49:25.953016: step: 2416/1541, loss: 0.41654127836227417 2023-01-23 06:49:26.855684: step: 2418/1541, loss: 1.2218579053878784 2023-01-23 06:49:27.766376: step: 2420/1541, loss: 0.724591851234436 2023-01-23 06:49:28.686614: step: 2422/1541, loss: 0.7520134449005127 2023-01-23 06:49:29.614572: step: 2424/1541, loss: 0.9466223120689392 2023-01-23 06:49:30.542876: step: 2426/1541, loss: 1.2705596685409546 2023-01-23 06:49:31.466744: step: 2428/1541, loss: 0.8427746295928955 2023-01-23 06:49:32.367983: step: 2430/1541, loss: 0.6605857610702515 2023-01-23 06:49:33.259476: step: 2432/1541, loss: 0.4048011302947998 2023-01-23 06:49:34.116824: step: 2434/1541, loss: 2.9850618839263916 2023-01-23 06:49:35.015976: step: 2436/1541, loss: 1.2179224491119385 2023-01-23 06:49:35.937002: step: 2438/1541, loss: 1.633911371231079 2023-01-23 06:49:36.828014: step: 2440/1541, loss: 0.549184262752533 2023-01-23 06:49:37.708133: step: 2442/1541, loss: 1.375182867050171 2023-01-23 06:49:38.619760: step: 2444/1541, loss: 0.5450568199157715 2023-01-23 06:49:39.503329: step: 2446/1541, loss: 0.6098089218139648 2023-01-23 06:49:40.411193: step: 2448/1541, loss: 1.474771499633789 2023-01-23 06:49:41.295326: step: 2450/1541, loss: 0.5317738056182861 2023-01-23 06:49:42.200689: step: 2452/1541, loss: 0.30113738775253296 2023-01-23 06:49:43.090175: step: 2454/1541, loss: 0.6066685318946838 2023-01-23 06:49:43.966776: step: 2456/1541, loss: 0.9552927017211914 2023-01-23 06:49:44.889443: step: 2458/1541, loss: 0.7288678884506226 2023-01-23 06:49:45.782092: step: 2460/1541, loss: 0.5818673372268677 2023-01-23 06:49:46.668718: step: 2462/1541, loss: 0.645623505115509 2023-01-23 06:49:47.589009: step: 2464/1541, loss: 0.6443738341331482 2023-01-23 06:49:48.486117: step: 2466/1541, loss: 0.4354906976222992 2023-01-23 06:49:49.421231: step: 2468/1541, loss: 0.8837670683860779 2023-01-23 06:49:50.337088: step: 2470/1541, loss: 2.0497992038726807 2023-01-23 06:49:51.208165: step: 2472/1541, loss: 0.2904987037181854 2023-01-23 06:49:52.088465: step: 2474/1541, loss: 0.4677445888519287 2023-01-23 06:49:53.012680: step: 2476/1541, loss: 0.4795485734939575 2023-01-23 06:49:53.894263: step: 2478/1541, loss: 1.3735769987106323 2023-01-23 06:49:54.771409: step: 2480/1541, loss: 0.8323858976364136 2023-01-23 06:49:55.656643: step: 2482/1541, loss: 0.5852367281913757 2023-01-23 06:49:56.562781: step: 2484/1541, loss: 1.2440295219421387 2023-01-23 06:49:57.441987: step: 2486/1541, loss: 0.15737667679786682 2023-01-23 06:49:58.377923: step: 2488/1541, loss: 1.738625407218933 2023-01-23 06:49:59.318869: step: 2490/1541, loss: 0.7460893988609314 2023-01-23 06:50:00.198719: step: 2492/1541, loss: 0.7434651255607605 2023-01-23 06:50:01.080490: step: 2494/1541, loss: 0.5969826579093933 2023-01-23 06:50:01.960658: step: 2496/1541, loss: 0.6520935297012329 2023-01-23 06:50:02.829062: step: 2498/1541, loss: 0.5086385607719421 2023-01-23 06:50:03.720659: step: 2500/1541, loss: 0.29394540190696716 2023-01-23 06:50:04.613008: step: 2502/1541, loss: 1.6932350397109985 2023-01-23 06:50:05.527755: step: 2504/1541, loss: 1.3116681575775146 2023-01-23 06:50:06.475087: step: 2506/1541, loss: 0.5861674547195435 2023-01-23 06:50:07.415529: step: 2508/1541, loss: 2.0051591396331787 2023-01-23 06:50:08.322081: step: 2510/1541, loss: 0.794586718082428 2023-01-23 06:50:09.225006: step: 2512/1541, loss: 1.328714370727539 2023-01-23 06:50:10.214357: step: 2514/1541, loss: 6.950294017791748 2023-01-23 06:50:11.112555: step: 2516/1541, loss: 0.8477199673652649 2023-01-23 06:50:12.018298: step: 2518/1541, loss: 0.9465923309326172 2023-01-23 06:50:12.882428: step: 2520/1541, loss: 0.755248486995697 2023-01-23 06:50:13.793630: step: 2522/1541, loss: 0.7449169158935547 2023-01-23 06:50:14.666429: step: 2524/1541, loss: 0.7235947847366333 2023-01-23 06:50:15.559851: step: 2526/1541, loss: 0.5238451361656189 2023-01-23 06:50:16.471805: step: 2528/1541, loss: 2.7537553310394287 2023-01-23 06:50:17.362289: step: 2530/1541, loss: 0.7705903053283691 2023-01-23 06:50:18.308053: step: 2532/1541, loss: 0.5359914302825928 2023-01-23 06:50:19.171916: step: 2534/1541, loss: 0.9879105091094971 2023-01-23 06:50:20.098847: step: 2536/1541, loss: 0.5153762698173523 2023-01-23 06:50:20.985139: step: 2538/1541, loss: 0.6992459297180176 2023-01-23 06:50:21.902737: step: 2540/1541, loss: 0.590569794178009 2023-01-23 06:50:22.829308: step: 2542/1541, loss: 3.000459671020508 2023-01-23 06:50:23.723398: step: 2544/1541, loss: 1.4330140352249146 2023-01-23 06:50:24.602128: step: 2546/1541, loss: 0.42886024713516235 2023-01-23 06:50:25.540686: step: 2548/1541, loss: 0.295803427696228 2023-01-23 06:50:26.439478: step: 2550/1541, loss: 0.2633872330188751 2023-01-23 06:50:27.308488: step: 2552/1541, loss: 0.6122350692749023 2023-01-23 06:50:28.223019: step: 2554/1541, loss: 0.49354270100593567 2023-01-23 06:50:29.140224: step: 2556/1541, loss: 0.4269046187400818 2023-01-23 06:50:30.060200: step: 2558/1541, loss: 3.096219539642334 2023-01-23 06:50:30.959915: step: 2560/1541, loss: 0.6148738265037537 2023-01-23 06:50:31.884176: step: 2562/1541, loss: 0.8375611901283264 2023-01-23 06:50:32.799591: step: 2564/1541, loss: 0.4141958951950073 2023-01-23 06:50:33.693055: step: 2566/1541, loss: 4.607567310333252 2023-01-23 06:50:34.590876: step: 2568/1541, loss: 0.38369524478912354 2023-01-23 06:50:35.515363: step: 2570/1541, loss: 0.6758881211280823 2023-01-23 06:50:36.415087: step: 2572/1541, loss: 0.7749309539794922 2023-01-23 06:50:37.339401: step: 2574/1541, loss: 0.4599250555038452 2023-01-23 06:50:38.230655: step: 2576/1541, loss: 2.8828089237213135 2023-01-23 06:50:39.149009: step: 2578/1541, loss: 3.3388986587524414 2023-01-23 06:50:40.044632: step: 2580/1541, loss: 0.6843503713607788 2023-01-23 06:50:40.992845: step: 2582/1541, loss: 0.642180860042572 2023-01-23 06:50:41.874356: step: 2584/1541, loss: 3.2269065380096436 2023-01-23 06:50:42.752071: step: 2586/1541, loss: 1.0480670928955078 2023-01-23 06:50:43.698986: step: 2588/1541, loss: 0.6322999596595764 2023-01-23 06:50:44.594817: step: 2590/1541, loss: 0.5632036924362183 2023-01-23 06:50:45.472921: step: 2592/1541, loss: 1.4971349239349365 2023-01-23 06:50:46.366489: step: 2594/1541, loss: 0.4308764636516571 2023-01-23 06:50:47.274750: step: 2596/1541, loss: 1.8763874769210815 2023-01-23 06:50:48.166845: step: 2598/1541, loss: 0.6467455625534058 2023-01-23 06:50:49.097074: step: 2600/1541, loss: 0.6139785647392273 2023-01-23 06:50:49.954470: step: 2602/1541, loss: 1.6762239933013916 2023-01-23 06:50:50.821999: step: 2604/1541, loss: 0.4821636974811554 2023-01-23 06:50:51.752698: step: 2606/1541, loss: 1.0220320224761963 2023-01-23 06:50:52.613274: step: 2608/1541, loss: 0.8432393074035645 2023-01-23 06:50:53.480393: step: 2610/1541, loss: 0.5761191844940186 2023-01-23 06:50:54.345802: step: 2612/1541, loss: 0.9979059100151062 2023-01-23 06:50:55.286776: step: 2614/1541, loss: 1.163330078125 2023-01-23 06:50:56.199988: step: 2616/1541, loss: 0.8648534417152405 2023-01-23 06:50:57.102220: step: 2618/1541, loss: 0.9025871157646179 2023-01-23 06:50:57.990731: step: 2620/1541, loss: 0.38662630319595337 2023-01-23 06:50:58.909824: step: 2622/1541, loss: 3.349167585372925 2023-01-23 06:50:59.791309: step: 2624/1541, loss: 0.8872421383857727 2023-01-23 06:51:00.676636: step: 2626/1541, loss: 0.6642637252807617 2023-01-23 06:51:01.614858: step: 2628/1541, loss: 0.4420936703681946 2023-01-23 06:51:02.543335: step: 2630/1541, loss: 0.889924168586731 2023-01-23 06:51:03.475329: step: 2632/1541, loss: 0.6690396666526794 2023-01-23 06:51:04.381065: step: 2634/1541, loss: 2.7502684593200684 2023-01-23 06:51:05.247840: step: 2636/1541, loss: 0.6712068319320679 2023-01-23 06:51:06.093961: step: 2638/1541, loss: 0.3080849349498749 2023-01-23 06:51:07.015258: step: 2640/1541, loss: 3.2183446884155273 2023-01-23 06:51:07.937992: step: 2642/1541, loss: 0.4785444140434265 2023-01-23 06:51:08.851979: step: 2644/1541, loss: 0.866632878780365 2023-01-23 06:51:09.737225: step: 2646/1541, loss: 0.9429449439048767 2023-01-23 06:51:10.662565: step: 2648/1541, loss: 3.810587167739868 2023-01-23 06:51:11.566359: step: 2650/1541, loss: 0.6649128794670105 2023-01-23 06:51:12.442387: step: 2652/1541, loss: 0.4290529191493988 2023-01-23 06:51:13.341701: step: 2654/1541, loss: 3.046032190322876 2023-01-23 06:51:14.269087: step: 2656/1541, loss: 0.8276188373565674 2023-01-23 06:51:15.146311: step: 2658/1541, loss: 1.6022956371307373 2023-01-23 06:51:16.042207: step: 2660/1541, loss: 0.45610862970352173 2023-01-23 06:51:16.938647: step: 2662/1541, loss: 0.8693602681159973 2023-01-23 06:51:17.883244: step: 2664/1541, loss: 0.35055646300315857 2023-01-23 06:51:18.772877: step: 2666/1541, loss: 0.7148089408874512 2023-01-23 06:51:19.655359: step: 2668/1541, loss: 0.8137701749801636 2023-01-23 06:51:20.630821: step: 2670/1541, loss: 0.43315479159355164 2023-01-23 06:51:21.526161: step: 2672/1541, loss: 1.4240559339523315 2023-01-23 06:51:22.425015: step: 2674/1541, loss: 0.5706973671913147 2023-01-23 06:51:23.311236: step: 2676/1541, loss: 1.216079592704773 2023-01-23 06:51:24.242212: step: 2678/1541, loss: 0.6342830061912537 2023-01-23 06:51:25.134149: step: 2680/1541, loss: 0.9856889247894287 2023-01-23 06:51:26.064001: step: 2682/1541, loss: 0.46922338008880615 2023-01-23 06:51:26.947247: step: 2684/1541, loss: 0.17606720328330994 2023-01-23 06:51:27.859393: step: 2686/1541, loss: 2.0702414512634277 2023-01-23 06:51:28.737578: step: 2688/1541, loss: 0.5337644219398499 2023-01-23 06:51:29.642033: step: 2690/1541, loss: 0.4967379570007324 2023-01-23 06:51:30.570835: step: 2692/1541, loss: 1.0560805797576904 2023-01-23 06:51:31.472829: step: 2694/1541, loss: 0.5877314805984497 2023-01-23 06:51:32.392631: step: 2696/1541, loss: 0.19150878489017487 2023-01-23 06:51:33.284057: step: 2698/1541, loss: 0.7799651622772217 2023-01-23 06:51:34.175241: step: 2700/1541, loss: 0.878520131111145 2023-01-23 06:51:35.051997: step: 2702/1541, loss: 0.9824098944664001 2023-01-23 06:51:35.964834: step: 2704/1541, loss: 0.7950394153594971 2023-01-23 06:51:36.888887: step: 2706/1541, loss: 2.0457417964935303 2023-01-23 06:51:37.816045: step: 2708/1541, loss: 0.7579077482223511 2023-01-23 06:51:38.716489: step: 2710/1541, loss: 0.27793416380882263 2023-01-23 06:51:39.622493: step: 2712/1541, loss: 0.8088700175285339 2023-01-23 06:51:40.490871: step: 2714/1541, loss: 0.8178289532661438 2023-01-23 06:51:41.422099: step: 2716/1541, loss: 0.842404305934906 2023-01-23 06:51:42.319634: step: 2718/1541, loss: 0.4403385818004608 2023-01-23 06:51:43.247719: step: 2720/1541, loss: 0.41237717866897583 2023-01-23 06:51:44.131401: step: 2722/1541, loss: 0.3788757622241974 2023-01-23 06:51:45.041858: step: 2724/1541, loss: 1.080094337463379 2023-01-23 06:51:45.935977: step: 2726/1541, loss: 0.6967707276344299 2023-01-23 06:51:46.853786: step: 2728/1541, loss: 0.20934036374092102 2023-01-23 06:51:47.719089: step: 2730/1541, loss: 1.6414629220962524 2023-01-23 06:51:48.593002: step: 2732/1541, loss: 0.4846853017807007 2023-01-23 06:51:49.516878: step: 2734/1541, loss: 0.5198751091957092 2023-01-23 06:51:50.396921: step: 2736/1541, loss: 0.5948711633682251 2023-01-23 06:51:51.329298: step: 2738/1541, loss: 1.2462186813354492 2023-01-23 06:51:52.199489: step: 2740/1541, loss: 0.43496111035346985 2023-01-23 06:51:53.082674: step: 2742/1541, loss: 0.5218947529792786 2023-01-23 06:51:53.958531: step: 2744/1541, loss: 0.4882807433605194 2023-01-23 06:51:54.894065: step: 2746/1541, loss: 0.6481642127037048 2023-01-23 06:51:55.807368: step: 2748/1541, loss: 1.1067698001861572 2023-01-23 06:51:56.727709: step: 2750/1541, loss: 0.6556779146194458 2023-01-23 06:51:57.603833: step: 2752/1541, loss: 0.4252396821975708 2023-01-23 06:51:58.525178: step: 2754/1541, loss: 1.2455329895019531 2023-01-23 06:51:59.469427: step: 2756/1541, loss: 0.654960572719574 2023-01-23 06:52:00.393345: step: 2758/1541, loss: 0.4640372693538666 2023-01-23 06:52:01.297575: step: 2760/1541, loss: 0.4654383361339569 2023-01-23 06:52:02.203294: step: 2762/1541, loss: 0.4567195475101471 2023-01-23 06:52:03.072177: step: 2764/1541, loss: 0.5366681218147278 2023-01-23 06:52:03.977299: step: 2766/1541, loss: 0.7204498052597046 2023-01-23 06:52:04.861393: step: 2768/1541, loss: 0.4622713327407837 2023-01-23 06:52:05.743238: step: 2770/1541, loss: 0.8868849873542786 2023-01-23 06:52:06.686784: step: 2772/1541, loss: 1.6879857778549194 2023-01-23 06:52:07.582838: step: 2774/1541, loss: 0.5513251423835754 2023-01-23 06:52:08.461710: step: 2776/1541, loss: 0.3488087058067322 2023-01-23 06:52:09.337637: step: 2778/1541, loss: 0.7439311742782593 2023-01-23 06:52:10.286050: step: 2780/1541, loss: 0.7540821433067322 2023-01-23 06:52:11.167355: step: 2782/1541, loss: 1.5715727806091309 2023-01-23 06:52:12.022124: step: 2784/1541, loss: 0.5931330323219299 2023-01-23 06:52:12.926613: step: 2786/1541, loss: 0.7840108275413513 2023-01-23 06:52:13.892576: step: 2788/1541, loss: 0.40015238523483276 2023-01-23 06:52:14.783747: step: 2790/1541, loss: 1.4232161045074463 2023-01-23 06:52:15.667025: step: 2792/1541, loss: 0.35715174674987793 2023-01-23 06:52:16.578229: step: 2794/1541, loss: 0.8894797563552856 2023-01-23 06:52:17.463873: step: 2796/1541, loss: 0.7369832396507263 2023-01-23 06:52:18.345948: step: 2798/1541, loss: 0.46147188544273376 2023-01-23 06:52:19.236369: step: 2800/1541, loss: 2.998225450515747 2023-01-23 06:52:20.102830: step: 2802/1541, loss: 0.8565020561218262 2023-01-23 06:52:20.995166: step: 2804/1541, loss: 3.232938051223755 2023-01-23 06:52:21.878795: step: 2806/1541, loss: 1.0100436210632324 2023-01-23 06:52:22.819750: step: 2808/1541, loss: 0.34319815039634705 2023-01-23 06:52:23.738528: step: 2810/1541, loss: 0.7877587676048279 2023-01-23 06:52:24.684150: step: 2812/1541, loss: 0.70042484998703 2023-01-23 06:52:25.605024: step: 2814/1541, loss: 0.5593509078025818 2023-01-23 06:52:26.502453: step: 2816/1541, loss: 0.5842348337173462 2023-01-23 06:52:27.431136: step: 2818/1541, loss: 1.892763376235962 2023-01-23 06:52:28.344808: step: 2820/1541, loss: 0.30918315052986145 2023-01-23 06:52:29.196173: step: 2822/1541, loss: 0.9348947405815125 2023-01-23 06:52:30.109763: step: 2824/1541, loss: 0.7976300716400146 2023-01-23 06:52:31.017389: step: 2826/1541, loss: 3.0586822032928467 2023-01-23 06:52:31.947127: step: 2828/1541, loss: 0.9904352426528931 2023-01-23 06:52:32.851289: step: 2830/1541, loss: 0.7892906665802002 2023-01-23 06:52:33.722828: step: 2832/1541, loss: 2.38131046295166 2023-01-23 06:52:34.616532: step: 2834/1541, loss: 0.8378028869628906 2023-01-23 06:52:35.501864: step: 2836/1541, loss: 1.0515649318695068 2023-01-23 06:52:36.454619: step: 2838/1541, loss: 1.742667555809021 2023-01-23 06:52:37.329000: step: 2840/1541, loss: 3.040266513824463 2023-01-23 06:52:38.198162: step: 2842/1541, loss: 0.738449215888977 2023-01-23 06:52:39.060175: step: 2844/1541, loss: 0.6985875964164734 2023-01-23 06:52:39.988353: step: 2846/1541, loss: 0.9253210425376892 2023-01-23 06:52:40.891230: step: 2848/1541, loss: 0.6545392274856567 2023-01-23 06:52:41.764069: step: 2850/1541, loss: 0.8485846519470215 2023-01-23 06:52:42.648007: step: 2852/1541, loss: 0.511599063873291 2023-01-23 06:52:43.661750: step: 2854/1541, loss: 2.010369300842285 2023-01-23 06:52:44.553660: step: 2856/1541, loss: 1.386923909187317 2023-01-23 06:52:45.452142: step: 2858/1541, loss: 0.5873668193817139 2023-01-23 06:52:46.367680: step: 2860/1541, loss: 0.39485180377960205 2023-01-23 06:52:47.265246: step: 2862/1541, loss: 1.0366625785827637 2023-01-23 06:52:48.153707: step: 2864/1541, loss: 0.569727897644043 2023-01-23 06:52:49.016015: step: 2866/1541, loss: 0.3537065088748932 2023-01-23 06:52:49.876162: step: 2868/1541, loss: 0.34461942315101624 2023-01-23 06:52:50.771080: step: 2870/1541, loss: 0.27726370096206665 2023-01-23 06:52:51.674197: step: 2872/1541, loss: 0.5902780294418335 2023-01-23 06:52:52.550884: step: 2874/1541, loss: 0.9238290190696716 2023-01-23 06:52:53.500720: step: 2876/1541, loss: 0.9691436290740967 2023-01-23 06:52:54.423556: step: 2878/1541, loss: 1.1488378047943115 2023-01-23 06:52:55.325587: step: 2880/1541, loss: 0.9448673725128174 2023-01-23 06:52:56.242854: step: 2882/1541, loss: 0.4089185893535614 2023-01-23 06:52:57.144683: step: 2884/1541, loss: 0.7142809629440308 2023-01-23 06:52:58.012159: step: 2886/1541, loss: 0.8896318078041077 2023-01-23 06:52:58.924023: step: 2888/1541, loss: 0.6487597227096558 2023-01-23 06:52:59.813677: step: 2890/1541, loss: 0.36323797702789307 2023-01-23 06:53:00.700815: step: 2892/1541, loss: 0.7220245003700256 2023-01-23 06:53:01.583450: step: 2894/1541, loss: 0.3093395531177521 2023-01-23 06:53:02.436722: step: 2896/1541, loss: 1.679253339767456 2023-01-23 06:53:03.310093: step: 2898/1541, loss: 1.7424025535583496 2023-01-23 06:53:04.223400: step: 2900/1541, loss: 0.6510205268859863 2023-01-23 06:53:05.155241: step: 2902/1541, loss: 0.6571189165115356 2023-01-23 06:53:06.087643: step: 2904/1541, loss: 1.7615669965744019 2023-01-23 06:53:06.994717: step: 2906/1541, loss: 0.8838924765586853 2023-01-23 06:53:07.842310: step: 2908/1541, loss: 0.8308296203613281 2023-01-23 06:53:08.748393: step: 2910/1541, loss: 0.34375542402267456 2023-01-23 06:53:09.623548: step: 2912/1541, loss: 1.213295578956604 2023-01-23 06:53:10.497565: step: 2914/1541, loss: 0.3382451832294464 2023-01-23 06:53:11.465012: step: 2916/1541, loss: 1.1634070873260498 2023-01-23 06:53:12.342289: step: 2918/1541, loss: 0.9108424186706543 2023-01-23 06:53:13.239110: step: 2920/1541, loss: 0.8060278296470642 2023-01-23 06:53:14.139450: step: 2922/1541, loss: 0.8946234583854675 2023-01-23 06:53:14.994794: step: 2924/1541, loss: 0.8488937020301819 2023-01-23 06:53:15.880200: step: 2926/1541, loss: 1.1606537103652954 2023-01-23 06:53:16.760136: step: 2928/1541, loss: 0.7988627552986145 2023-01-23 06:53:17.643497: step: 2930/1541, loss: 1.1261663436889648 2023-01-23 06:53:18.578875: step: 2932/1541, loss: 0.61208176612854 2023-01-23 06:53:19.503533: step: 2934/1541, loss: 0.38271552324295044 2023-01-23 06:53:20.482571: step: 2936/1541, loss: 0.5503991842269897 2023-01-23 06:53:21.375323: step: 2938/1541, loss: 0.3268078863620758 2023-01-23 06:53:22.301899: step: 2940/1541, loss: 1.1954219341278076 2023-01-23 06:53:23.212776: step: 2942/1541, loss: 0.31680750846862793 2023-01-23 06:53:24.175954: step: 2944/1541, loss: 1.505192756652832 2023-01-23 06:53:25.033882: step: 2946/1541, loss: 0.2952319383621216 2023-01-23 06:53:25.969851: step: 2948/1541, loss: 0.7960109114646912 2023-01-23 06:53:26.850381: step: 2950/1541, loss: 0.40666663646698 2023-01-23 06:53:27.745076: step: 2952/1541, loss: 0.8037883639335632 2023-01-23 06:53:28.660204: step: 2954/1541, loss: 0.6687396168708801 2023-01-23 06:53:29.516005: step: 2956/1541, loss: 0.7309952974319458 2023-01-23 06:53:30.372242: step: 2958/1541, loss: 0.939723551273346 2023-01-23 06:53:31.296465: step: 2960/1541, loss: 3.058190107345581 2023-01-23 06:53:32.204512: step: 2962/1541, loss: 0.5859863758087158 2023-01-23 06:53:33.096423: step: 2964/1541, loss: 0.4704339802265167 2023-01-23 06:53:33.980861: step: 2966/1541, loss: 0.5199567079544067 2023-01-23 06:53:34.981877: step: 2968/1541, loss: 0.4536302983760834 2023-01-23 06:53:35.860220: step: 2970/1541, loss: 0.2271750420331955 2023-01-23 06:53:36.754115: step: 2972/1541, loss: 0.5785194635391235 2023-01-23 06:53:37.655043: step: 2974/1541, loss: 3.301595449447632 2023-01-23 06:53:38.545388: step: 2976/1541, loss: 0.8071733713150024 2023-01-23 06:53:39.451160: step: 2978/1541, loss: 0.831230878829956 2023-01-23 06:53:40.396070: step: 2980/1541, loss: 1.1557142734527588 2023-01-23 06:53:41.251949: step: 2982/1541, loss: 0.40494590997695923 2023-01-23 06:53:42.207476: step: 2984/1541, loss: 1.0965934991836548 2023-01-23 06:53:43.082156: step: 2986/1541, loss: 0.5673129558563232 2023-01-23 06:53:43.971750: step: 2988/1541, loss: 0.47751790285110474 2023-01-23 06:53:44.853270: step: 2990/1541, loss: 1.8412965536117554 2023-01-23 06:53:45.696362: step: 2992/1541, loss: 0.5943275094032288 2023-01-23 06:53:46.580541: step: 2994/1541, loss: 0.5887250304222107 2023-01-23 06:53:47.559594: step: 2996/1541, loss: 1.0254136323928833 2023-01-23 06:53:48.452924: step: 2998/1541, loss: 0.5016969442367554 2023-01-23 06:53:49.331927: step: 3000/1541, loss: 0.7349323034286499 2023-01-23 06:53:50.193430: step: 3002/1541, loss: 1.026248574256897 2023-01-23 06:53:51.130698: step: 3004/1541, loss: 2.058320999145508 2023-01-23 06:53:52.100702: step: 3006/1541, loss: 1.0224571228027344 2023-01-23 06:53:52.988309: step: 3008/1541, loss: 0.7366580367088318 2023-01-23 06:53:53.860697: step: 3010/1541, loss: 0.9326617121696472 2023-01-23 06:53:54.748016: step: 3012/1541, loss: 0.7042102813720703 2023-01-23 06:53:55.642215: step: 3014/1541, loss: 1.9828485250473022 2023-01-23 06:53:56.537120: step: 3016/1541, loss: 0.6872323155403137 2023-01-23 06:53:57.458440: step: 3018/1541, loss: 0.7101231813430786 2023-01-23 06:53:58.414379: step: 3020/1541, loss: 0.566102921962738 2023-01-23 06:53:59.299694: step: 3022/1541, loss: 0.9892632961273193 2023-01-23 06:54:00.177207: step: 3024/1541, loss: 0.7636057734489441 2023-01-23 06:54:01.079885: step: 3026/1541, loss: 0.7082196474075317 2023-01-23 06:54:02.002959: step: 3028/1541, loss: 1.1732956171035767 2023-01-23 06:54:02.915886: step: 3030/1541, loss: 3.5514650344848633 2023-01-23 06:54:03.843491: step: 3032/1541, loss: 1.1558783054351807 2023-01-23 06:54:04.733050: step: 3034/1541, loss: 0.8644376993179321 2023-01-23 06:54:05.577211: step: 3036/1541, loss: 0.79531329870224 2023-01-23 06:54:06.481143: step: 3038/1541, loss: 0.6502455472946167 2023-01-23 06:54:07.362549: step: 3040/1541, loss: 1.1039066314697266 2023-01-23 06:54:08.258403: step: 3042/1541, loss: 0.7068555355072021 2023-01-23 06:54:09.181419: step: 3044/1541, loss: 0.674129068851471 2023-01-23 06:54:10.096438: step: 3046/1541, loss: 0.4715316891670227 2023-01-23 06:54:11.010297: step: 3048/1541, loss: 0.8583561778068542 2023-01-23 06:54:11.902779: step: 3050/1541, loss: 1.084531545639038 2023-01-23 06:54:12.772183: step: 3052/1541, loss: 0.766613781452179 2023-01-23 06:54:13.693221: step: 3054/1541, loss: 0.97287517786026 2023-01-23 06:54:14.660919: step: 3056/1541, loss: 1.2534469366073608 2023-01-23 06:54:15.548897: step: 3058/1541, loss: 3.5144453048706055 2023-01-23 06:54:16.458992: step: 3060/1541, loss: 0.2762492895126343 2023-01-23 06:54:17.311793: step: 3062/1541, loss: 1.0721877813339233 2023-01-23 06:54:18.216469: step: 3064/1541, loss: 2.866797685623169 2023-01-23 06:54:19.134228: step: 3066/1541, loss: 0.7746002674102783 2023-01-23 06:54:20.042724: step: 3068/1541, loss: 2.6702122688293457 2023-01-23 06:54:20.903042: step: 3070/1541, loss: 0.47900819778442383 2023-01-23 06:54:21.874989: step: 3072/1541, loss: 0.4965197443962097 2023-01-23 06:54:22.773254: step: 3074/1541, loss: 0.584229588508606 2023-01-23 06:54:23.714779: step: 3076/1541, loss: 0.9533994793891907 2023-01-23 06:54:24.614021: step: 3078/1541, loss: 0.8174363970756531 2023-01-23 06:54:25.479196: step: 3080/1541, loss: 0.4970117509365082 2023-01-23 06:54:26.445079: step: 3082/1541, loss: 0.6110126972198486 2023-01-23 06:54:27.090284: step: 3084/1541, loss: 1.4200888872146606 ================================================== Loss: 1.008 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6785420951989045, 'r': 0.6342893498598455, 'f1': 0.6556698897427617}, 'combined': 0.43353487671908053, 'epoch': 3} Test Chinese: {'event': {'p': 0.6146161934805467, 'r': 0.7004194128220491, 'f1': 0.6547185662279473}, 'argument': {'p': 0.7152752499932414, 'r': 0.683166178551939, 'f1': 0.6988520923025249}, 'combined': 0.4575514398777102, 'epoch': 3} Dev Korean: {'event': {'p': 0.5794117647058824, 'r': 0.7869507323568575, 'f1': 0.6674195369847544}, 'argument': {'p': 0.6730302718272186, 'r': 0.6402230614442431, 'f1': 0.6562168778212484}, 'combined': 0.43797196475703876, 'epoch': 3} Test Korean: {'event': {'p': 0.6158245948522403, 'r': 0.7741162372678251, 'f1': 0.6859569949561986}, 'argument': {'p': 0.710349703959979, 'r': 0.6981195087758472, 'f1': 0.7041815069913384}, 'combined': 0.48303823043950583, 'epoch': 3} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6726304715079724, 'r': 0.6407450541227186, 'f1': 0.6563007143962181}, 'combined': 0.45509002515499514, 'epoch': 3} Test Russian: {'event': {'p': 0.6546310832025117, 'r': 0.7495506291192331, 'f1': 0.6988826815642458}, 'argument': {'p': 0.7113660651562723, 'r': 0.6932372099423113, 'f1': 0.7021846453005923}, 'combined': 0.4907446878609168, 'epoch': 3} Chinese: {'event': {'p': 0.6507936507936508, 'r': 0.7592592592592593, 'f1': 0.7008547008547008}, 'argument': {'p': 0.5846397862926789, 'r': 0.6260302136408331, 'f1': 0.604627471294138}, 'combined': 0.42375600552238724, 'epoch': 3} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5277777777777778, 'r': 0.2368233618233618, 'f1': 0.3269419862340216}, 'combined': 0.1905016297741543, 'epoch': 3} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5364055059457359, 'r': 0.5613545992455375, 'f1': 0.5485965401717753}, 'combined': 0.28213536351691304, 'epoch': 3} New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6212664277180406, 'r': 0.6924101198402131, 'f1': 0.6549118387909321}, 'argument': {'p': 0.6893128369893616, 'r': 0.624505647186943, 'f1': 0.6553108584831598}, 'combined': 0.42917083930887046, 'epoch': 2} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.7288561511943926, 'r': 0.6674983629560891, 'f1': 0.6968291831658773}, 'combined': 0.45310483430724363, 'epoch': 2} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6497835497835497, 'r': 0.6037811745776347, 'f1': 0.6259382819015846}, 'combined': 0.4600913866968912, 'epoch': 2} -------------------- Dev for Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.6879318328246672, 'r': 0.6346725941543704, 'f1': 0.6602298798250162}, 'combined': 0.4339399097212416, 'epoch': 2} Test for Korean: {'event': {'p': 0.607721639656816, 'r': 0.7639304973037747, 'f1': 0.6769312450225644}, 'argument': {'p': 0.7251491588582952, 'r': 0.6793663287655232, 'f1': 0.701511554983885}, 'combined': 0.4748750903129564, 'epoch': 2} Korean: {'event': {'p': 0.609375, 'r': 0.6190476190476191, 'f1': 0.6141732283464567}, 'argument': {'p': 0.5048776455026455, 'r': 0.24561615186615185, 'f1': 0.33046536796536796}, 'combined': 0.2029629818999898, 'epoch': 2} -------------------- Dev for Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6726304715079724, 'r': 0.6407450541227186, 'f1': 0.6563007143962181}, 'combined': 0.45509002515499514, 'epoch': 3} Test for Russian: {'event': {'p': 0.6546310832025117, 'r': 0.7495506291192331, 'f1': 0.6988826815642458}, 'argument': {'p': 0.7113660651562723, 'r': 0.6932372099423113, 'f1': 0.7021846453005923}, 'combined': 0.4907446878609168, 'epoch': 3} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5364055059457359, 'r': 0.5613545992455375, 'f1': 0.5485965401717753}, 'combined': 0.28213536351691304, 'epoch': 3} ****************************** Epoch: 4 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 06:59:53.995185: step: 2/1541, loss: 0.35639065504074097 2023-01-23 06:59:54.891614: step: 4/1541, loss: 0.41663384437561035 2023-01-23 06:59:55.781023: step: 6/1541, loss: 0.8641396164894104 2023-01-23 06:59:56.630708: step: 8/1541, loss: 0.48891592025756836 2023-01-23 06:59:57.502583: step: 10/1541, loss: 0.20470018684864044 2023-01-23 06:59:58.449460: step: 12/1541, loss: 0.8402842879295349 2023-01-23 06:59:59.351922: step: 14/1541, loss: 0.6338161826133728 2023-01-23 07:00:00.294268: step: 16/1541, loss: 0.9755557179450989 2023-01-23 07:00:01.204795: step: 18/1541, loss: 0.5065597891807556 2023-01-23 07:00:02.043516: step: 20/1541, loss: 0.5316221714019775 2023-01-23 07:00:02.949155: step: 22/1541, loss: 0.25896939635276794 2023-01-23 07:00:03.933936: step: 24/1541, loss: 0.7775360941886902 2023-01-23 07:00:04.875369: step: 26/1541, loss: 0.646548867225647 2023-01-23 07:00:05.745959: step: 28/1541, loss: 0.7487905025482178 2023-01-23 07:00:06.634776: step: 30/1541, loss: 0.9475518465042114 2023-01-23 07:00:07.541675: step: 32/1541, loss: 0.7794806361198425 2023-01-23 07:00:08.488256: step: 34/1541, loss: 0.45696237683296204 2023-01-23 07:00:09.412940: step: 36/1541, loss: 0.8135567903518677 2023-01-23 07:00:10.286155: step: 38/1541, loss: 6.27531623840332 2023-01-23 07:00:11.201720: step: 40/1541, loss: 0.6271959543228149 2023-01-23 07:00:12.088421: step: 42/1541, loss: 0.6120291352272034 2023-01-23 07:00:12.967275: step: 44/1541, loss: 0.640003502368927 2023-01-23 07:00:13.854667: step: 46/1541, loss: 0.5082609057426453 2023-01-23 07:00:14.814660: step: 48/1541, loss: 0.5339298844337463 2023-01-23 07:00:15.737710: step: 50/1541, loss: 0.5510515570640564 2023-01-23 07:00:16.634076: step: 52/1541, loss: 1.056194543838501 2023-01-23 07:00:17.573841: step: 54/1541, loss: 0.5506774187088013 2023-01-23 07:00:18.557519: step: 56/1541, loss: 0.3631300926208496 2023-01-23 07:00:19.466825: step: 58/1541, loss: 0.9682166576385498 2023-01-23 07:00:20.364989: step: 60/1541, loss: 1.035837173461914 2023-01-23 07:00:21.240843: step: 62/1541, loss: 0.5179961919784546 2023-01-23 07:00:22.160124: step: 64/1541, loss: 0.23268994688987732 2023-01-23 07:00:23.053524: step: 66/1541, loss: 0.3806628882884979 2023-01-23 07:00:23.934682: step: 68/1541, loss: 0.9479464888572693 2023-01-23 07:00:24.845107: step: 70/1541, loss: 1.29695725440979 2023-01-23 07:00:25.773604: step: 72/1541, loss: 0.44676828384399414 2023-01-23 07:00:26.691534: step: 74/1541, loss: 0.4384376108646393 2023-01-23 07:00:27.574603: step: 76/1541, loss: 0.8668836951255798 2023-01-23 07:00:28.518937: step: 78/1541, loss: 0.2368919998407364 2023-01-23 07:00:29.439980: step: 80/1541, loss: 0.3334803581237793 2023-01-23 07:00:30.342475: step: 82/1541, loss: 0.43404775857925415 2023-01-23 07:00:31.233549: step: 84/1541, loss: 1.1454832553863525 2023-01-23 07:00:32.263632: step: 86/1541, loss: 0.5476458072662354 2023-01-23 07:00:33.187608: step: 88/1541, loss: 0.30376991629600525 2023-01-23 07:00:34.098068: step: 90/1541, loss: 1.4936141967773438 2023-01-23 07:00:34.968584: step: 92/1541, loss: 0.58907550573349 2023-01-23 07:00:35.821255: step: 94/1541, loss: 0.759922981262207 2023-01-23 07:00:36.674333: step: 96/1541, loss: 0.07023736834526062 2023-01-23 07:00:37.563181: step: 98/1541, loss: 0.5967851877212524 2023-01-23 07:00:38.470498: step: 100/1541, loss: 1.3800114393234253 2023-01-23 07:00:39.401339: step: 102/1541, loss: 0.37848854064941406 2023-01-23 07:00:40.329951: step: 104/1541, loss: 0.38708144426345825 2023-01-23 07:00:41.196031: step: 106/1541, loss: 0.4211001396179199 2023-01-23 07:00:42.118586: step: 108/1541, loss: 2.885019302368164 2023-01-23 07:00:43.045657: step: 110/1541, loss: 0.9167048931121826 2023-01-23 07:00:43.939788: step: 112/1541, loss: 0.7329390048980713 2023-01-23 07:00:44.795581: step: 114/1541, loss: 0.6040591597557068 2023-01-23 07:00:45.718142: step: 116/1541, loss: 0.8076047897338867 2023-01-23 07:00:46.622319: step: 118/1541, loss: 0.5490811467170715 2023-01-23 07:00:47.521715: step: 120/1541, loss: 0.5647960305213928 2023-01-23 07:00:48.447878: step: 122/1541, loss: 0.639639675617218 2023-01-23 07:00:49.310520: step: 124/1541, loss: 0.8193246126174927 2023-01-23 07:00:50.316552: step: 126/1541, loss: 0.7182811498641968 2023-01-23 07:00:51.211700: step: 128/1541, loss: 0.7598232626914978 2023-01-23 07:00:52.106674: step: 130/1541, loss: 0.4220511317253113 2023-01-23 07:00:53.003636: step: 132/1541, loss: 0.31029176712036133 2023-01-23 07:00:53.945770: step: 134/1541, loss: 0.4558878540992737 2023-01-23 07:00:54.816654: step: 136/1541, loss: 0.4931383728981018 2023-01-23 07:00:55.714022: step: 138/1541, loss: 0.47441738843917847 2023-01-23 07:00:56.653532: step: 140/1541, loss: 0.21571220457553864 2023-01-23 07:00:57.551295: step: 142/1541, loss: 0.9402888417243958 2023-01-23 07:00:58.437474: step: 144/1541, loss: 1.0152523517608643 2023-01-23 07:00:59.404925: step: 146/1541, loss: 0.7338749170303345 2023-01-23 07:01:00.306642: step: 148/1541, loss: 0.2885528802871704 2023-01-23 07:01:01.189179: step: 150/1541, loss: 0.46343478560447693 2023-01-23 07:01:02.150713: step: 152/1541, loss: 0.4147680103778839 2023-01-23 07:01:03.063193: step: 154/1541, loss: 0.4177910089492798 2023-01-23 07:01:03.970417: step: 156/1541, loss: 0.6275709867477417 2023-01-23 07:01:04.912201: step: 158/1541, loss: 0.40687090158462524 2023-01-23 07:01:05.810301: step: 160/1541, loss: 0.34343868494033813 2023-01-23 07:01:06.752147: step: 162/1541, loss: 0.4578886032104492 2023-01-23 07:01:07.655415: step: 164/1541, loss: 0.4287934899330139 2023-01-23 07:01:08.547233: step: 166/1541, loss: 1.2963716983795166 2023-01-23 07:01:09.391685: step: 168/1541, loss: 0.31687435507774353 2023-01-23 07:01:10.277932: step: 170/1541, loss: 0.6447929739952087 2023-01-23 07:01:11.166386: step: 172/1541, loss: 0.5489696264266968 2023-01-23 07:01:12.106280: step: 174/1541, loss: 0.31195229291915894 2023-01-23 07:01:13.001558: step: 176/1541, loss: 0.6146755218505859 2023-01-23 07:01:13.864062: step: 178/1541, loss: 0.7109798192977905 2023-01-23 07:01:14.818320: step: 180/1541, loss: 1.2230693101882935 2023-01-23 07:01:15.703876: step: 182/1541, loss: 0.5717586278915405 2023-01-23 07:01:16.586280: step: 184/1541, loss: 0.36476728320121765 2023-01-23 07:01:17.577896: step: 186/1541, loss: 1.203009009361267 2023-01-23 07:01:18.483010: step: 188/1541, loss: 0.31434354186058044 2023-01-23 07:01:19.405905: step: 190/1541, loss: 0.5349948406219482 2023-01-23 07:01:20.327235: step: 192/1541, loss: 0.8300599455833435 2023-01-23 07:01:21.236123: step: 194/1541, loss: 0.6334413290023804 2023-01-23 07:01:22.166800: step: 196/1541, loss: 6.505847930908203 2023-01-23 07:01:23.047285: step: 198/1541, loss: 0.5938680768013 2023-01-23 07:01:24.027080: step: 200/1541, loss: 0.5339093208312988 2023-01-23 07:01:24.902901: step: 202/1541, loss: 0.548084557056427 2023-01-23 07:01:25.768238: step: 204/1541, loss: 0.5031399130821228 2023-01-23 07:01:26.686222: step: 206/1541, loss: 0.6960732340812683 2023-01-23 07:01:27.573107: step: 208/1541, loss: 0.5777720212936401 2023-01-23 07:01:28.491810: step: 210/1541, loss: 2.885460615158081 2023-01-23 07:01:29.360747: step: 212/1541, loss: 0.6099592447280884 2023-01-23 07:01:30.294020: step: 214/1541, loss: 0.8610377907752991 2023-01-23 07:01:31.183536: step: 216/1541, loss: 0.2950378656387329 2023-01-23 07:01:32.162036: step: 218/1541, loss: 0.6504219174385071 2023-01-23 07:01:33.078874: step: 220/1541, loss: 1.173505187034607 2023-01-23 07:01:34.004477: step: 222/1541, loss: 1.5751672983169556 2023-01-23 07:01:34.907770: step: 224/1541, loss: 0.2435256838798523 2023-01-23 07:01:35.778706: step: 226/1541, loss: 0.4774916172027588 2023-01-23 07:01:36.680215: step: 228/1541, loss: 0.7695338726043701 2023-01-23 07:01:37.582044: step: 230/1541, loss: 0.36186274886131287 2023-01-23 07:01:38.459037: step: 232/1541, loss: 1.1956123113632202 2023-01-23 07:01:39.359419: step: 234/1541, loss: 0.5484254360198975 2023-01-23 07:01:40.257933: step: 236/1541, loss: 0.9056467413902283 2023-01-23 07:01:41.202698: step: 238/1541, loss: 0.865379273891449 2023-01-23 07:01:42.120951: step: 240/1541, loss: 0.5281542539596558 2023-01-23 07:01:43.081175: step: 242/1541, loss: 1.2576274871826172 2023-01-23 07:01:44.035285: step: 244/1541, loss: 0.5406676530838013 2023-01-23 07:01:44.897521: step: 246/1541, loss: 3.0097439289093018 2023-01-23 07:01:45.812861: step: 248/1541, loss: 0.34368592500686646 2023-01-23 07:01:46.719826: step: 250/1541, loss: 0.9659327864646912 2023-01-23 07:01:47.600113: step: 252/1541, loss: 0.3980919420719147 2023-01-23 07:01:48.532553: step: 254/1541, loss: 0.7041852474212646 2023-01-23 07:01:49.482059: step: 256/1541, loss: 0.5693432092666626 2023-01-23 07:01:50.350711: step: 258/1541, loss: 0.6082345247268677 2023-01-23 07:01:51.188018: step: 260/1541, loss: 0.22372999787330627 2023-01-23 07:01:52.102478: step: 262/1541, loss: 2.799172878265381 2023-01-23 07:01:53.084215: step: 264/1541, loss: 1.2157448530197144 2023-01-23 07:01:53.973161: step: 266/1541, loss: 0.3975808620452881 2023-01-23 07:01:54.924888: step: 268/1541, loss: 0.3142869174480438 2023-01-23 07:01:55.812007: step: 270/1541, loss: 0.423505961894989 2023-01-23 07:01:56.739686: step: 272/1541, loss: 1.8022465705871582 2023-01-23 07:01:57.592709: step: 274/1541, loss: 0.4474077820777893 2023-01-23 07:01:58.491902: step: 276/1541, loss: 0.7493696212768555 2023-01-23 07:01:59.412079: step: 278/1541, loss: 0.4819171726703644 2023-01-23 07:02:00.321974: step: 280/1541, loss: 0.321350634098053 2023-01-23 07:02:01.245278: step: 282/1541, loss: 0.7748561501502991 2023-01-23 07:02:02.217853: step: 284/1541, loss: 1.161037564277649 2023-01-23 07:02:03.122615: step: 286/1541, loss: 0.6595345735549927 2023-01-23 07:02:03.983820: step: 288/1541, loss: 0.4061495363712311 2023-01-23 07:02:04.914823: step: 290/1541, loss: 0.44273555278778076 2023-01-23 07:02:05.831732: step: 292/1541, loss: 0.27594050765037537 2023-01-23 07:02:06.716925: step: 294/1541, loss: 0.4930768609046936 2023-01-23 07:02:07.651927: step: 296/1541, loss: 0.6789840459823608 2023-01-23 07:02:08.544148: step: 298/1541, loss: 0.4404834806919098 2023-01-23 07:02:09.455685: step: 300/1541, loss: 0.752193808555603 2023-01-23 07:02:10.395107: step: 302/1541, loss: 0.6546227931976318 2023-01-23 07:02:11.273802: step: 304/1541, loss: 0.3260587155818939 2023-01-23 07:02:12.148893: step: 306/1541, loss: 1.652951717376709 2023-01-23 07:02:13.056125: step: 308/1541, loss: 0.35569918155670166 2023-01-23 07:02:13.917627: step: 310/1541, loss: 2.5215604305267334 2023-01-23 07:02:14.810773: step: 312/1541, loss: 1.1000125408172607 2023-01-23 07:02:15.688796: step: 314/1541, loss: 0.32277679443359375 2023-01-23 07:02:16.648328: step: 316/1541, loss: 0.6335071921348572 2023-01-23 07:02:17.585891: step: 318/1541, loss: 0.5416952967643738 2023-01-23 07:02:18.488390: step: 320/1541, loss: 1.4804561138153076 2023-01-23 07:02:19.403235: step: 322/1541, loss: 0.654325544834137 2023-01-23 07:02:20.270538: step: 324/1541, loss: 0.3848007023334503 2023-01-23 07:02:21.241103: step: 326/1541, loss: 0.6482894420623779 2023-01-23 07:02:22.116883: step: 328/1541, loss: 0.6123363971710205 2023-01-23 07:02:23.048679: step: 330/1541, loss: 1.0746396780014038 2023-01-23 07:02:23.994075: step: 332/1541, loss: 0.24523551762104034 2023-01-23 07:02:24.953754: step: 334/1541, loss: 0.5975744724273682 2023-01-23 07:02:25.892668: step: 336/1541, loss: 2.930387258529663 2023-01-23 07:02:26.763804: step: 338/1541, loss: 0.4643702805042267 2023-01-23 07:02:27.656448: step: 340/1541, loss: 0.7656936049461365 2023-01-23 07:02:28.553962: step: 342/1541, loss: 0.8431143164634705 2023-01-23 07:02:29.420774: step: 344/1541, loss: 0.8967902660369873 2023-01-23 07:02:30.293852: step: 346/1541, loss: 0.36316367983818054 2023-01-23 07:02:31.172425: step: 348/1541, loss: 0.6305859684944153 2023-01-23 07:02:32.081180: step: 350/1541, loss: 0.5818061232566833 2023-01-23 07:02:32.981852: step: 352/1541, loss: 0.4873957335948944 2023-01-23 07:02:33.877999: step: 354/1541, loss: 0.21681463718414307 2023-01-23 07:02:34.809299: step: 356/1541, loss: 0.5409770607948303 2023-01-23 07:02:35.690620: step: 358/1541, loss: 0.4304410517215729 2023-01-23 07:02:36.593348: step: 360/1541, loss: 1.8312546014785767 2023-01-23 07:02:37.501651: step: 362/1541, loss: 0.4920400381088257 2023-01-23 07:02:38.432750: step: 364/1541, loss: 0.9511473774909973 2023-01-23 07:02:39.340018: step: 366/1541, loss: 0.7562235593795776 2023-01-23 07:02:40.231148: step: 368/1541, loss: 0.5775697231292725 2023-01-23 07:02:41.108723: step: 370/1541, loss: 0.3200948238372803 2023-01-23 07:02:41.981749: step: 372/1541, loss: 0.27658888697624207 2023-01-23 07:02:42.898254: step: 374/1541, loss: 0.41858723759651184 2023-01-23 07:02:43.794177: step: 376/1541, loss: 0.7032800912857056 2023-01-23 07:02:44.711531: step: 378/1541, loss: 0.4384008049964905 2023-01-23 07:02:45.580080: step: 380/1541, loss: 1.0318403244018555 2023-01-23 07:02:46.477647: step: 382/1541, loss: 1.153670310974121 2023-01-23 07:02:47.359589: step: 384/1541, loss: 0.25036323070526123 2023-01-23 07:02:48.249322: step: 386/1541, loss: 0.7645015120506287 2023-01-23 07:02:49.212017: step: 388/1541, loss: 0.5571752190589905 2023-01-23 07:02:50.114260: step: 390/1541, loss: 0.8432124257087708 2023-01-23 07:02:51.067658: step: 392/1541, loss: 0.38512730598449707 2023-01-23 07:02:51.978412: step: 394/1541, loss: 1.067600965499878 2023-01-23 07:02:52.933789: step: 396/1541, loss: 0.7823648452758789 2023-01-23 07:02:53.825501: step: 398/1541, loss: 0.5591596961021423 2023-01-23 07:02:54.771236: step: 400/1541, loss: 0.4153532087802887 2023-01-23 07:02:55.657334: step: 402/1541, loss: 0.4889265298843384 2023-01-23 07:02:56.512491: step: 404/1541, loss: 0.5772254467010498 2023-01-23 07:02:57.406784: step: 406/1541, loss: 0.5782445669174194 2023-01-23 07:02:58.326099: step: 408/1541, loss: 0.9257734417915344 2023-01-23 07:02:59.204297: step: 410/1541, loss: 0.7348018288612366 2023-01-23 07:03:00.078051: step: 412/1541, loss: 0.5993162393569946 2023-01-23 07:03:00.999944: step: 414/1541, loss: 0.14453643560409546 2023-01-23 07:03:01.870406: step: 416/1541, loss: 0.8573649525642395 2023-01-23 07:03:02.765085: step: 418/1541, loss: 0.44033318758010864 2023-01-23 07:03:03.750426: step: 420/1541, loss: 0.8942620158195496 2023-01-23 07:03:04.694571: step: 422/1541, loss: 0.5965848565101624 2023-01-23 07:03:05.587281: step: 424/1541, loss: 0.3566063642501831 2023-01-23 07:03:06.464209: step: 426/1541, loss: 2.8668324947357178 2023-01-23 07:03:07.417162: step: 428/1541, loss: 2.687628984451294 2023-01-23 07:03:08.310549: step: 430/1541, loss: 0.5975289940834045 2023-01-23 07:03:09.219467: step: 432/1541, loss: 6.9565043449401855 2023-01-23 07:03:10.112061: step: 434/1541, loss: 0.7325390577316284 2023-01-23 07:03:11.016334: step: 436/1541, loss: 0.7506673336029053 2023-01-23 07:03:11.924421: step: 438/1541, loss: 0.3527127802371979 2023-01-23 07:03:12.785235: step: 440/1541, loss: 0.5559250116348267 2023-01-23 07:03:13.757039: step: 442/1541, loss: 0.7633219361305237 2023-01-23 07:03:14.607691: step: 444/1541, loss: 0.5333383083343506 2023-01-23 07:03:15.529122: step: 446/1541, loss: 1.0069669485092163 2023-01-23 07:03:16.448115: step: 448/1541, loss: 0.30800893902778625 2023-01-23 07:03:17.345097: step: 450/1541, loss: 0.6845381259918213 2023-01-23 07:03:18.199049: step: 452/1541, loss: 1.0263276100158691 2023-01-23 07:03:19.120004: step: 454/1541, loss: 0.3477324843406677 2023-01-23 07:03:20.021584: step: 456/1541, loss: 0.7921503186225891 2023-01-23 07:03:20.917169: step: 458/1541, loss: 0.787117600440979 2023-01-23 07:03:21.834672: step: 460/1541, loss: 0.2600216269493103 2023-01-23 07:03:22.769485: step: 462/1541, loss: 0.2581906318664551 2023-01-23 07:03:23.713351: step: 464/1541, loss: 0.2496635615825653 2023-01-23 07:03:24.590543: step: 466/1541, loss: 1.2410564422607422 2023-01-23 07:03:25.508039: step: 468/1541, loss: 0.4324294626712799 2023-01-23 07:03:26.440746: step: 470/1541, loss: 3.1685116291046143 2023-01-23 07:03:27.333672: step: 472/1541, loss: 1.0415068864822388 2023-01-23 07:03:28.225814: step: 474/1541, loss: 0.15704667568206787 2023-01-23 07:03:29.095899: step: 476/1541, loss: 0.33538952469825745 2023-01-23 07:03:29.992247: step: 478/1541, loss: 0.4855436682701111 2023-01-23 07:03:30.904225: step: 480/1541, loss: 0.8201401233673096 2023-01-23 07:03:31.891882: step: 482/1541, loss: 0.7034803628921509 2023-01-23 07:03:32.789798: step: 484/1541, loss: 0.5882121324539185 2023-01-23 07:03:33.757076: step: 486/1541, loss: 0.2049350142478943 2023-01-23 07:03:34.631328: step: 488/1541, loss: 0.3127138912677765 2023-01-23 07:03:35.606787: step: 490/1541, loss: 0.5594770908355713 2023-01-23 07:03:36.555044: step: 492/1541, loss: 0.8255093693733215 2023-01-23 07:03:37.442593: step: 494/1541, loss: 0.33069908618927 2023-01-23 07:03:38.323325: step: 496/1541, loss: 0.2817462384700775 2023-01-23 07:03:39.220963: step: 498/1541, loss: 0.3544328808784485 2023-01-23 07:03:40.134836: step: 500/1541, loss: 0.7436527609825134 2023-01-23 07:03:41.022431: step: 502/1541, loss: 0.5650792121887207 2023-01-23 07:03:41.875418: step: 504/1541, loss: 0.31052321195602417 2023-01-23 07:03:42.760650: step: 506/1541, loss: 0.2003660351037979 2023-01-23 07:03:43.695052: step: 508/1541, loss: 0.7977397441864014 2023-01-23 07:03:44.586737: step: 510/1541, loss: 0.5743427276611328 2023-01-23 07:03:45.490685: step: 512/1541, loss: 3.0924148559570312 2023-01-23 07:03:46.370004: step: 514/1541, loss: 0.47699975967407227 2023-01-23 07:03:47.300313: step: 516/1541, loss: 0.6128028631210327 2023-01-23 07:03:48.171162: step: 518/1541, loss: 0.4299311935901642 2023-01-23 07:03:49.078416: step: 520/1541, loss: 0.41679003834724426 2023-01-23 07:03:49.942903: step: 522/1541, loss: 0.49946966767311096 2023-01-23 07:03:50.852015: step: 524/1541, loss: 0.6300909519195557 2023-01-23 07:03:51.704765: step: 526/1541, loss: 0.8552273511886597 2023-01-23 07:03:52.588180: step: 528/1541, loss: 0.6142439842224121 2023-01-23 07:03:53.492844: step: 530/1541, loss: 0.3844347894191742 2023-01-23 07:03:54.386172: step: 532/1541, loss: 1.1744905710220337 2023-01-23 07:03:55.285837: step: 534/1541, loss: 1.087348222732544 2023-01-23 07:03:56.229286: step: 536/1541, loss: 1.0540391206741333 2023-01-23 07:03:57.126066: step: 538/1541, loss: 0.38007062673568726 2023-01-23 07:03:58.057014: step: 540/1541, loss: 0.6080347895622253 2023-01-23 07:03:58.946889: step: 542/1541, loss: 0.7435769438743591 2023-01-23 07:03:59.841843: step: 544/1541, loss: 0.2898463010787964 2023-01-23 07:04:00.716365: step: 546/1541, loss: 0.6714811325073242 2023-01-23 07:04:01.634899: step: 548/1541, loss: 0.7749018669128418 2023-01-23 07:04:02.539983: step: 550/1541, loss: 1.0352516174316406 2023-01-23 07:04:03.402719: step: 552/1541, loss: 0.6061456203460693 2023-01-23 07:04:04.269274: step: 554/1541, loss: 1.245184302330017 2023-01-23 07:04:05.183949: step: 556/1541, loss: 2.0449492931365967 2023-01-23 07:04:06.070489: step: 558/1541, loss: 1.2873525619506836 2023-01-23 07:04:06.945212: step: 560/1541, loss: 0.39493033289909363 2023-01-23 07:04:07.847275: step: 562/1541, loss: 0.7271957397460938 2023-01-23 07:04:08.737863: step: 564/1541, loss: 0.7720930576324463 2023-01-23 07:04:09.652190: step: 566/1541, loss: 0.1936616152524948 2023-01-23 07:04:10.501379: step: 568/1541, loss: 0.4131607413291931 2023-01-23 07:04:11.447389: step: 570/1541, loss: 0.6005046963691711 2023-01-23 07:04:12.343401: step: 572/1541, loss: 0.17009519040584564 2023-01-23 07:04:13.255638: step: 574/1541, loss: 1.4498467445373535 2023-01-23 07:04:14.109196: step: 576/1541, loss: 0.5299227237701416 2023-01-23 07:04:15.076837: step: 578/1541, loss: 0.69524085521698 2023-01-23 07:04:15.965664: step: 580/1541, loss: 0.5396324396133423 2023-01-23 07:04:16.906437: step: 582/1541, loss: 0.4543581008911133 2023-01-23 07:04:17.850699: step: 584/1541, loss: 0.5685120820999146 2023-01-23 07:04:18.749529: step: 586/1541, loss: 0.5892479419708252 2023-01-23 07:04:19.691421: step: 588/1541, loss: 2.930568218231201 2023-01-23 07:04:20.625893: step: 590/1541, loss: 0.6764990091323853 2023-01-23 07:04:21.486039: step: 592/1541, loss: 0.45327967405319214 2023-01-23 07:04:22.492043: step: 594/1541, loss: 0.842613935470581 2023-01-23 07:04:23.454539: step: 596/1541, loss: 0.4460006356239319 2023-01-23 07:04:24.352993: step: 598/1541, loss: 0.6604982018470764 2023-01-23 07:04:25.379938: step: 600/1541, loss: 0.45744115114212036 2023-01-23 07:04:26.244703: step: 602/1541, loss: 0.2564593255519867 2023-01-23 07:04:27.113676: step: 604/1541, loss: 0.9678825736045837 2023-01-23 07:04:28.014231: step: 606/1541, loss: 0.26225483417510986 2023-01-23 07:04:28.916569: step: 608/1541, loss: 1.1925156116485596 2023-01-23 07:04:29.841895: step: 610/1541, loss: 0.3425655663013458 2023-01-23 07:04:30.828586: step: 612/1541, loss: 1.5098025798797607 2023-01-23 07:04:31.748836: step: 614/1541, loss: 0.6285039186477661 2023-01-23 07:04:32.632011: step: 616/1541, loss: 0.9817387461662292 2023-01-23 07:04:33.565430: step: 618/1541, loss: 1.2222719192504883 2023-01-23 07:04:34.482141: step: 620/1541, loss: 2.674994468688965 2023-01-23 07:04:35.402324: step: 622/1541, loss: 0.6592932343482971 2023-01-23 07:04:36.267931: step: 624/1541, loss: 1.0900862216949463 2023-01-23 07:04:37.191423: step: 626/1541, loss: 0.46936169266700745 2023-01-23 07:04:38.040199: step: 628/1541, loss: 3.2454421520233154 2023-01-23 07:04:38.964904: step: 630/1541, loss: 0.38147038221359253 2023-01-23 07:04:39.907645: step: 632/1541, loss: 1.0242407321929932 2023-01-23 07:04:40.815990: step: 634/1541, loss: 0.1319034993648529 2023-01-23 07:04:41.714454: step: 636/1541, loss: 0.5940477848052979 2023-01-23 07:04:42.671850: step: 638/1541, loss: 0.5438511967658997 2023-01-23 07:04:43.582964: step: 640/1541, loss: 0.37284666299819946 2023-01-23 07:04:44.466868: step: 642/1541, loss: 0.684889554977417 2023-01-23 07:04:45.368796: step: 644/1541, loss: 0.9852976202964783 2023-01-23 07:04:46.228244: step: 646/1541, loss: 0.5146953463554382 2023-01-23 07:04:47.095999: step: 648/1541, loss: 0.3850066363811493 2023-01-23 07:04:48.050318: step: 650/1541, loss: 1.1937193870544434 2023-01-23 07:04:48.944803: step: 652/1541, loss: 1.1049550771713257 2023-01-23 07:04:49.843301: step: 654/1541, loss: 0.4454454779624939 2023-01-23 07:04:50.757004: step: 656/1541, loss: 0.8131265044212341 2023-01-23 07:04:51.623284: step: 658/1541, loss: 3.0441482067108154 2023-01-23 07:04:52.498324: step: 660/1541, loss: 0.3486039936542511 2023-01-23 07:04:53.370617: step: 662/1541, loss: 1.1379317045211792 2023-01-23 07:04:54.339415: step: 664/1541, loss: 0.35487788915634155 2023-01-23 07:04:55.214627: step: 666/1541, loss: 0.8556997179985046 2023-01-23 07:04:56.123816: step: 668/1541, loss: 0.4424804747104645 2023-01-23 07:04:57.009631: step: 670/1541, loss: 2.8349945545196533 2023-01-23 07:04:57.891242: step: 672/1541, loss: 0.49113354086875916 2023-01-23 07:04:58.762170: step: 674/1541, loss: 0.4158802926540375 2023-01-23 07:04:59.705812: step: 676/1541, loss: 0.6941125988960266 2023-01-23 07:05:00.606397: step: 678/1541, loss: 0.3964768946170807 2023-01-23 07:05:01.491669: step: 680/1541, loss: 0.6592745184898376 2023-01-23 07:05:02.349031: step: 682/1541, loss: 0.6572495102882385 2023-01-23 07:05:03.247162: step: 684/1541, loss: 0.316462904214859 2023-01-23 07:05:04.120254: step: 686/1541, loss: 0.770595371723175 2023-01-23 07:05:05.022335: step: 688/1541, loss: 0.5102452635765076 2023-01-23 07:05:05.897839: step: 690/1541, loss: 0.3972678780555725 2023-01-23 07:05:06.791015: step: 692/1541, loss: 0.7766573429107666 2023-01-23 07:05:07.662841: step: 694/1541, loss: 0.7234928607940674 2023-01-23 07:05:08.594316: step: 696/1541, loss: 6.780306816101074 2023-01-23 07:05:09.524395: step: 698/1541, loss: 0.3362690806388855 2023-01-23 07:05:10.433613: step: 700/1541, loss: 0.4615066647529602 2023-01-23 07:05:11.333324: step: 702/1541, loss: 0.5828859806060791 2023-01-23 07:05:12.200638: step: 704/1541, loss: 0.2744019627571106 2023-01-23 07:05:13.077828: step: 706/1541, loss: 0.5750451683998108 2023-01-23 07:05:13.969730: step: 708/1541, loss: 0.4278254210948944 2023-01-23 07:05:14.872072: step: 710/1541, loss: 0.4940546154975891 2023-01-23 07:05:15.736421: step: 712/1541, loss: 3.10323166847229 2023-01-23 07:05:16.599457: step: 714/1541, loss: 0.8295289278030396 2023-01-23 07:05:17.505885: step: 716/1541, loss: 0.5703974962234497 2023-01-23 07:05:18.369271: step: 718/1541, loss: 0.3127555549144745 2023-01-23 07:05:19.295487: step: 720/1541, loss: 0.6884352564811707 2023-01-23 07:05:20.211507: step: 722/1541, loss: 0.36161723732948303 2023-01-23 07:05:21.076867: step: 724/1541, loss: 0.4813966453075409 2023-01-23 07:05:21.994741: step: 726/1541, loss: 0.6804596781730652 2023-01-23 07:05:22.892611: step: 728/1541, loss: 1.274431586265564 2023-01-23 07:05:23.842725: step: 730/1541, loss: 2.895120143890381 2023-01-23 07:05:24.757249: step: 732/1541, loss: 1.252785325050354 2023-01-23 07:05:25.619235: step: 734/1541, loss: 0.4156200885772705 2023-01-23 07:05:26.502382: step: 736/1541, loss: 0.3636172115802765 2023-01-23 07:05:27.443343: step: 738/1541, loss: 0.8372044563293457 2023-01-23 07:05:28.347137: step: 740/1541, loss: 0.39562323689460754 2023-01-23 07:05:29.224605: step: 742/1541, loss: 0.5323712229728699 2023-01-23 07:05:30.107180: step: 744/1541, loss: 0.2836706340312958 2023-01-23 07:05:31.034579: step: 746/1541, loss: 0.47956833243370056 2023-01-23 07:05:31.964371: step: 748/1541, loss: 0.9522403478622437 2023-01-23 07:05:32.874031: step: 750/1541, loss: 1.2293517589569092 2023-01-23 07:05:33.777495: step: 752/1541, loss: 1.627590298652649 2023-01-23 07:05:34.659545: step: 754/1541, loss: 0.9683065414428711 2023-01-23 07:05:35.584140: step: 756/1541, loss: 0.5769877433776855 2023-01-23 07:05:36.469653: step: 758/1541, loss: 0.23092924058437347 2023-01-23 07:05:37.354562: step: 760/1541, loss: 0.5258589386940002 2023-01-23 07:05:38.245042: step: 762/1541, loss: 0.46789807081222534 2023-01-23 07:05:39.107201: step: 764/1541, loss: 0.5191705226898193 2023-01-23 07:05:39.988112: step: 766/1541, loss: 0.6154792904853821 2023-01-23 07:05:40.889530: step: 768/1541, loss: 0.9362555742263794 2023-01-23 07:05:41.778188: step: 770/1541, loss: 0.5595161318778992 2023-01-23 07:05:42.666033: step: 772/1541, loss: 0.27462485432624817 2023-01-23 07:05:43.576596: step: 774/1541, loss: 0.5608519315719604 2023-01-23 07:05:44.494558: step: 776/1541, loss: 1.2486770153045654 2023-01-23 07:05:45.436452: step: 778/1541, loss: 1.1526451110839844 2023-01-23 07:05:46.287375: step: 780/1541, loss: 0.6231037974357605 2023-01-23 07:05:47.208010: step: 782/1541, loss: 0.8937888145446777 2023-01-23 07:05:48.103199: step: 784/1541, loss: 0.2546788156032562 2023-01-23 07:05:48.974848: step: 786/1541, loss: 0.7430486083030701 2023-01-23 07:05:49.835096: step: 788/1541, loss: 0.5005574226379395 2023-01-23 07:05:50.703416: step: 790/1541, loss: 0.5705952644348145 2023-01-23 07:05:51.608702: step: 792/1541, loss: 0.41837117075920105 2023-01-23 07:05:52.490111: step: 794/1541, loss: 0.48230981826782227 2023-01-23 07:05:53.512851: step: 796/1541, loss: 0.6232258081436157 2023-01-23 07:05:54.412336: step: 798/1541, loss: 0.8015016913414001 2023-01-23 07:05:55.311583: step: 800/1541, loss: 0.7856835722923279 2023-01-23 07:05:56.235368: step: 802/1541, loss: 0.4889981150627136 2023-01-23 07:05:57.126888: step: 804/1541, loss: 0.3079782724380493 2023-01-23 07:05:58.053411: step: 806/1541, loss: 0.8131071329116821 2023-01-23 07:05:58.960036: step: 808/1541, loss: 0.40844324231147766 2023-01-23 07:05:59.821734: step: 810/1541, loss: 0.6499037146568298 2023-01-23 07:06:00.687261: step: 812/1541, loss: 0.795292854309082 2023-01-23 07:06:01.569602: step: 814/1541, loss: 0.613229513168335 2023-01-23 07:06:02.495848: step: 816/1541, loss: 0.5675347447395325 2023-01-23 07:06:03.378021: step: 818/1541, loss: 0.8992251753807068 2023-01-23 07:06:04.309158: step: 820/1541, loss: 0.3591192960739136 2023-01-23 07:06:05.308659: step: 822/1541, loss: 0.791569173336029 2023-01-23 07:06:06.216297: step: 824/1541, loss: 0.6150590181350708 2023-01-23 07:06:07.131019: step: 826/1541, loss: 0.2279568314552307 2023-01-23 07:06:08.044002: step: 828/1541, loss: 1.0448927879333496 2023-01-23 07:06:08.926658: step: 830/1541, loss: 0.1501631885766983 2023-01-23 07:06:09.853252: step: 832/1541, loss: 0.5951665043830872 2023-01-23 07:06:10.802772: step: 834/1541, loss: 0.7977077960968018 2023-01-23 07:06:11.668950: step: 836/1541, loss: 0.3463238477706909 2023-01-23 07:06:12.544956: step: 838/1541, loss: 1.390219807624817 2023-01-23 07:06:13.439302: step: 840/1541, loss: 0.8123387694358826 2023-01-23 07:06:14.331323: step: 842/1541, loss: 1.4998546838760376 2023-01-23 07:06:15.235805: step: 844/1541, loss: 0.67537522315979 2023-01-23 07:06:16.159205: step: 846/1541, loss: 0.7906548976898193 2023-01-23 07:06:17.017353: step: 848/1541, loss: 0.5402775406837463 2023-01-23 07:06:17.906062: step: 850/1541, loss: 0.34467965364456177 2023-01-23 07:06:18.841617: step: 852/1541, loss: 0.47802734375 2023-01-23 07:06:19.775885: step: 854/1541, loss: 0.7053552269935608 2023-01-23 07:06:20.683256: step: 856/1541, loss: 1.749342679977417 2023-01-23 07:06:21.600709: step: 858/1541, loss: 0.3648015558719635 2023-01-23 07:06:22.486684: step: 860/1541, loss: 0.7793250679969788 2023-01-23 07:06:23.417015: step: 862/1541, loss: 0.34726375341415405 2023-01-23 07:06:24.316952: step: 864/1541, loss: 0.3423007130622864 2023-01-23 07:06:25.191549: step: 866/1541, loss: 0.6155301332473755 2023-01-23 07:06:26.154359: step: 868/1541, loss: 1.0867915153503418 2023-01-23 07:06:27.073644: step: 870/1541, loss: 0.825012743473053 2023-01-23 07:06:28.041995: step: 872/1541, loss: 0.7349725365638733 2023-01-23 07:06:28.964302: step: 874/1541, loss: 0.4590951204299927 2023-01-23 07:06:29.881517: step: 876/1541, loss: 0.825746476650238 2023-01-23 07:06:30.810566: step: 878/1541, loss: 0.8326521515846252 2023-01-23 07:06:31.701683: step: 880/1541, loss: 0.36335524916648865 2023-01-23 07:06:32.671477: step: 882/1541, loss: 0.8209876418113708 2023-01-23 07:06:33.596823: step: 884/1541, loss: 0.4304547607898712 2023-01-23 07:06:34.534489: step: 886/1541, loss: 0.35783979296684265 2023-01-23 07:06:35.473074: step: 888/1541, loss: 0.8431710004806519 2023-01-23 07:06:36.344230: step: 890/1541, loss: 0.5288675427436829 2023-01-23 07:06:37.239346: step: 892/1541, loss: 0.31126025319099426 2023-01-23 07:06:38.128121: step: 894/1541, loss: 2.922851085662842 2023-01-23 07:06:39.006198: step: 896/1541, loss: 0.454779714345932 2023-01-23 07:06:39.945454: step: 898/1541, loss: 1.030845046043396 2023-01-23 07:06:40.820799: step: 900/1541, loss: 0.32891497015953064 2023-01-23 07:06:41.676537: step: 902/1541, loss: 2.694091796875 2023-01-23 07:06:42.583595: step: 904/1541, loss: 0.4757557809352875 2023-01-23 07:06:43.508197: step: 906/1541, loss: 0.5579558610916138 2023-01-23 07:06:44.363334: step: 908/1541, loss: 0.2255115807056427 2023-01-23 07:06:45.308719: step: 910/1541, loss: 3.023312568664551 2023-01-23 07:06:46.234992: step: 912/1541, loss: 1.0448017120361328 2023-01-23 07:06:47.216951: step: 914/1541, loss: 1.121856451034546 2023-01-23 07:06:48.107353: step: 916/1541, loss: 0.37933632731437683 2023-01-23 07:06:49.004723: step: 918/1541, loss: 0.24830792844295502 2023-01-23 07:06:49.968976: step: 920/1541, loss: 1.4730433225631714 2023-01-23 07:06:50.853799: step: 922/1541, loss: 0.28042158484458923 2023-01-23 07:06:51.719254: step: 924/1541, loss: 3.0454301834106445 2023-01-23 07:06:52.647886: step: 926/1541, loss: 0.4278239905834198 2023-01-23 07:06:53.512877: step: 928/1541, loss: 3.1468427181243896 2023-01-23 07:06:54.394865: step: 930/1541, loss: 0.32756906747817993 2023-01-23 07:06:55.308445: step: 932/1541, loss: 3.172213315963745 2023-01-23 07:06:56.172332: step: 934/1541, loss: 0.5511741042137146 2023-01-23 07:06:57.028312: step: 936/1541, loss: 3.142753839492798 2023-01-23 07:06:57.930912: step: 938/1541, loss: 1.205102801322937 2023-01-23 07:06:58.847774: step: 940/1541, loss: 1.631994366645813 2023-01-23 07:06:59.726609: step: 942/1541, loss: 0.5952964425086975 2023-01-23 07:07:00.675360: step: 944/1541, loss: 0.7590208053588867 2023-01-23 07:07:01.560430: step: 946/1541, loss: 0.7585088610649109 2023-01-23 07:07:02.426042: step: 948/1541, loss: 0.7346926331520081 2023-01-23 07:07:03.287156: step: 950/1541, loss: 0.33677947521209717 2023-01-23 07:07:04.275073: step: 952/1541, loss: 0.6682883501052856 2023-01-23 07:07:05.231851: step: 954/1541, loss: 0.5269983410835266 2023-01-23 07:07:06.196773: step: 956/1541, loss: 0.43157222867012024 2023-01-23 07:07:07.069193: step: 958/1541, loss: 0.8706896901130676 2023-01-23 07:07:08.059363: step: 960/1541, loss: 0.33963802456855774 2023-01-23 07:07:08.934734: step: 962/1541, loss: 1.1741546392440796 2023-01-23 07:07:09.852116: step: 964/1541, loss: 0.7508506178855896 2023-01-23 07:07:10.748301: step: 966/1541, loss: 0.2990075349807739 2023-01-23 07:07:11.692664: step: 968/1541, loss: 0.7949540019035339 2023-01-23 07:07:12.604578: step: 970/1541, loss: 0.8880837559700012 2023-01-23 07:07:13.501085: step: 972/1541, loss: 0.7244945168495178 2023-01-23 07:07:14.408883: step: 974/1541, loss: 0.7798898816108704 2023-01-23 07:07:15.322320: step: 976/1541, loss: 0.25348007678985596 2023-01-23 07:07:16.170316: step: 978/1541, loss: 0.5044070482254028 2023-01-23 07:07:17.076987: step: 980/1541, loss: 0.53548663854599 2023-01-23 07:07:17.958650: step: 982/1541, loss: 0.7569319009780884 2023-01-23 07:07:18.854254: step: 984/1541, loss: 0.5907227396965027 2023-01-23 07:07:19.744546: step: 986/1541, loss: 0.9991332292556763 2023-01-23 07:07:20.687884: step: 988/1541, loss: 0.6004302501678467 2023-01-23 07:07:21.539706: step: 990/1541, loss: 6.332016944885254 2023-01-23 07:07:22.400135: step: 992/1541, loss: 0.37137675285339355 2023-01-23 07:07:23.348211: step: 994/1541, loss: 0.7140220403671265 2023-01-23 07:07:24.261452: step: 996/1541, loss: 0.3823375701904297 2023-01-23 07:07:25.195685: step: 998/1541, loss: 3.0068347454071045 2023-01-23 07:07:26.154224: step: 1000/1541, loss: 0.8352782130241394 2023-01-23 07:07:27.018382: step: 1002/1541, loss: 0.37771040201187134 2023-01-23 07:07:27.931125: step: 1004/1541, loss: 1.711039423942566 2023-01-23 07:07:28.838095: step: 1006/1541, loss: 0.5462630987167358 2023-01-23 07:07:29.725486: step: 1008/1541, loss: 0.5317186713218689 2023-01-23 07:07:30.620046: step: 1010/1541, loss: 3.2197000980377197 2023-01-23 07:07:31.520696: step: 1012/1541, loss: 0.4043225944042206 2023-01-23 07:07:32.453451: step: 1014/1541, loss: 0.37581613659858704 2023-01-23 07:07:33.450930: step: 1016/1541, loss: 0.6520373821258545 2023-01-23 07:07:34.340084: step: 1018/1541, loss: 0.6268119812011719 2023-01-23 07:07:35.242206: step: 1020/1541, loss: 0.5820302963256836 2023-01-23 07:07:36.132397: step: 1022/1541, loss: 3.3393659591674805 2023-01-23 07:07:37.065016: step: 1024/1541, loss: 1.5802457332611084 2023-01-23 07:07:37.982627: step: 1026/1541, loss: 0.4873693883419037 2023-01-23 07:07:38.853899: step: 1028/1541, loss: 0.3527529537677765 2023-01-23 07:07:39.840272: step: 1030/1541, loss: 0.7292750477790833 2023-01-23 07:07:40.744385: step: 1032/1541, loss: 0.7755604386329651 2023-01-23 07:07:41.677185: step: 1034/1541, loss: 0.7627291083335876 2023-01-23 07:07:42.568951: step: 1036/1541, loss: 1.225442886352539 2023-01-23 07:07:43.451603: step: 1038/1541, loss: 0.5916361212730408 2023-01-23 07:07:44.400331: step: 1040/1541, loss: 0.7715797424316406 2023-01-23 07:07:45.256687: step: 1042/1541, loss: 0.24263207614421844 2023-01-23 07:07:46.139449: step: 1044/1541, loss: 0.4330085217952728 2023-01-23 07:07:47.060167: step: 1046/1541, loss: 0.44434258341789246 2023-01-23 07:07:47.944521: step: 1048/1541, loss: 1.4022741317749023 2023-01-23 07:07:48.951412: step: 1050/1541, loss: 1.326078176498413 2023-01-23 07:07:49.812941: step: 1052/1541, loss: 1.3286893367767334 2023-01-23 07:07:50.743840: step: 1054/1541, loss: 0.6286951899528503 2023-01-23 07:07:51.659834: step: 1056/1541, loss: 1.0339365005493164 2023-01-23 07:07:52.576271: step: 1058/1541, loss: 1.3705418109893799 2023-01-23 07:07:53.481669: step: 1060/1541, loss: 0.24314476549625397 2023-01-23 07:07:54.366369: step: 1062/1541, loss: 0.6693075299263 2023-01-23 07:07:55.258609: step: 1064/1541, loss: 0.3092575967311859 2023-01-23 07:07:56.154764: step: 1066/1541, loss: 0.2128313183784485 2023-01-23 07:07:57.057917: step: 1068/1541, loss: 0.22673338651657104 2023-01-23 07:07:57.928949: step: 1070/1541, loss: 0.3109179139137268 2023-01-23 07:07:58.832635: step: 1072/1541, loss: 1.430079460144043 2023-01-23 07:07:59.754852: step: 1074/1541, loss: 6.176510810852051 2023-01-23 07:08:00.686505: step: 1076/1541, loss: 1.5496623516082764 2023-01-23 07:08:01.582342: step: 1078/1541, loss: 1.6438047885894775 2023-01-23 07:08:02.560790: step: 1080/1541, loss: 0.5028643608093262 2023-01-23 07:08:03.501433: step: 1082/1541, loss: 1.2785100936889648 2023-01-23 07:08:04.397547: step: 1084/1541, loss: 0.506293773651123 2023-01-23 07:08:05.357879: step: 1086/1541, loss: 0.3883727490901947 2023-01-23 07:08:06.253052: step: 1088/1541, loss: 0.5895574688911438 2023-01-23 07:08:07.187704: step: 1090/1541, loss: 0.3378399610519409 2023-01-23 07:08:08.072686: step: 1092/1541, loss: 3.4543299674987793 2023-01-23 07:08:09.009852: step: 1094/1541, loss: 0.7211518883705139 2023-01-23 07:08:09.971953: step: 1096/1541, loss: 0.5298656821250916 2023-01-23 07:08:10.879796: step: 1098/1541, loss: 0.5185885429382324 2023-01-23 07:08:11.772199: step: 1100/1541, loss: 0.6890161633491516 2023-01-23 07:08:12.669305: step: 1102/1541, loss: 0.36733439564704895 2023-01-23 07:08:13.581389: step: 1104/1541, loss: 0.4536495506763458 2023-01-23 07:08:14.442926: step: 1106/1541, loss: 0.19767042994499207 2023-01-23 07:08:15.341128: step: 1108/1541, loss: 0.9169580936431885 2023-01-23 07:08:16.240450: step: 1110/1541, loss: 0.2816316485404968 2023-01-23 07:08:17.166619: step: 1112/1541, loss: 0.5626366138458252 2023-01-23 07:08:18.100471: step: 1114/1541, loss: 0.533022403717041 2023-01-23 07:08:18.929974: step: 1116/1541, loss: 0.36979547142982483 2023-01-23 07:08:19.864105: step: 1118/1541, loss: 0.5923240184783936 2023-01-23 07:08:20.734983: step: 1120/1541, loss: 0.30498191714286804 2023-01-23 07:08:21.607056: step: 1122/1541, loss: 1.1080201864242554 2023-01-23 07:08:22.508879: step: 1124/1541, loss: 0.3523523509502411 2023-01-23 07:08:23.436656: step: 1126/1541, loss: 0.48706936836242676 2023-01-23 07:08:24.339271: step: 1128/1541, loss: 0.8170691132545471 2023-01-23 07:08:25.241119: step: 1130/1541, loss: 1.4274065494537354 2023-01-23 07:08:26.127540: step: 1132/1541, loss: 0.4357204735279083 2023-01-23 07:08:27.029765: step: 1134/1541, loss: 0.7135902643203735 2023-01-23 07:08:28.016972: step: 1136/1541, loss: 0.9623558521270752 2023-01-23 07:08:28.924732: step: 1138/1541, loss: 0.8582772016525269 2023-01-23 07:08:29.813551: step: 1140/1541, loss: 0.7980756163597107 2023-01-23 07:08:30.691827: step: 1142/1541, loss: 0.46641167998313904 2023-01-23 07:08:31.584506: step: 1144/1541, loss: 0.561493456363678 2023-01-23 07:08:32.454568: step: 1146/1541, loss: 0.7946579456329346 2023-01-23 07:08:33.348955: step: 1148/1541, loss: 0.27544671297073364 2023-01-23 07:08:34.247572: step: 1150/1541, loss: 0.4142775237560272 2023-01-23 07:08:35.129878: step: 1152/1541, loss: 0.44001662731170654 2023-01-23 07:08:36.013146: step: 1154/1541, loss: 0.5027897357940674 2023-01-23 07:08:36.949129: step: 1156/1541, loss: 0.9551256895065308 2023-01-23 07:08:37.809252: step: 1158/1541, loss: 0.5945876836776733 2023-01-23 07:08:38.744230: step: 1160/1541, loss: 0.6265982389450073 2023-01-23 07:08:39.704790: step: 1162/1541, loss: 0.22276557981967926 2023-01-23 07:08:40.586351: step: 1164/1541, loss: 3.1982123851776123 2023-01-23 07:08:41.460279: step: 1166/1541, loss: 0.16922219097614288 2023-01-23 07:08:42.323618: step: 1168/1541, loss: 0.3075885772705078 2023-01-23 07:08:43.201191: step: 1170/1541, loss: 0.5465770363807678 2023-01-23 07:08:44.069059: step: 1172/1541, loss: 0.7360473275184631 2023-01-23 07:08:45.066226: step: 1174/1541, loss: 0.6573851704597473 2023-01-23 07:08:45.971983: step: 1176/1541, loss: 1.1040817499160767 2023-01-23 07:08:46.864847: step: 1178/1541, loss: 0.4297889173030853 2023-01-23 07:08:47.760941: step: 1180/1541, loss: 0.6516289710998535 2023-01-23 07:08:48.651721: step: 1182/1541, loss: 0.42247459292411804 2023-01-23 07:08:49.557197: step: 1184/1541, loss: 0.6799066066741943 2023-01-23 07:08:50.489249: step: 1186/1541, loss: 0.5161479115486145 2023-01-23 07:08:51.360987: step: 1188/1541, loss: 0.9538294076919556 2023-01-23 07:08:52.251659: step: 1190/1541, loss: 0.6444182991981506 2023-01-23 07:08:53.167066: step: 1192/1541, loss: 0.41797831654548645 2023-01-23 07:08:54.081493: step: 1194/1541, loss: 0.15568481385707855 2023-01-23 07:08:54.972552: step: 1196/1541, loss: 1.2385635375976562 2023-01-23 07:08:55.899647: step: 1198/1541, loss: 0.3466644287109375 2023-01-23 07:08:56.784186: step: 1200/1541, loss: 3.011399745941162 2023-01-23 07:08:57.668083: step: 1202/1541, loss: 0.9216405749320984 2023-01-23 07:08:58.607694: step: 1204/1541, loss: 1.381104588508606 2023-01-23 07:08:59.508521: step: 1206/1541, loss: 1.0265529155731201 2023-01-23 07:09:00.440493: step: 1208/1541, loss: 0.720730185508728 2023-01-23 07:09:01.348497: step: 1210/1541, loss: 0.8293285965919495 2023-01-23 07:09:02.289617: step: 1212/1541, loss: 1.499221682548523 2023-01-23 07:09:03.164810: step: 1214/1541, loss: 0.5413955450057983 2023-01-23 07:09:04.055641: step: 1216/1541, loss: 0.6923129558563232 2023-01-23 07:09:04.947567: step: 1218/1541, loss: 0.7168142795562744 2023-01-23 07:09:05.825049: step: 1220/1541, loss: 0.6273834705352783 2023-01-23 07:09:06.702872: step: 1222/1541, loss: 0.8828218579292297 2023-01-23 07:09:07.606366: step: 1224/1541, loss: 0.3112000823020935 2023-01-23 07:09:08.494549: step: 1226/1541, loss: 0.6963648200035095 2023-01-23 07:09:09.401009: step: 1228/1541, loss: 0.8705751895904541 2023-01-23 07:09:10.345288: step: 1230/1541, loss: 0.13895103335380554 2023-01-23 07:09:11.290462: step: 1232/1541, loss: 0.48411813378334045 2023-01-23 07:09:12.201207: step: 1234/1541, loss: 0.3604472875595093 2023-01-23 07:09:13.084083: step: 1236/1541, loss: 0.42455726861953735 2023-01-23 07:09:13.973765: step: 1238/1541, loss: 0.9551372528076172 2023-01-23 07:09:14.848345: step: 1240/1541, loss: 0.3742256164550781 2023-01-23 07:09:15.761418: step: 1242/1541, loss: 0.7142903804779053 2023-01-23 07:09:16.613016: step: 1244/1541, loss: 0.7813340425491333 2023-01-23 07:09:17.465037: step: 1246/1541, loss: 0.36689135432243347 2023-01-23 07:09:18.424817: step: 1248/1541, loss: 0.39394116401672363 2023-01-23 07:09:19.306613: step: 1250/1541, loss: 0.5839459896087646 2023-01-23 07:09:20.222764: step: 1252/1541, loss: 0.9225120544433594 2023-01-23 07:09:21.130281: step: 1254/1541, loss: 0.43414995074272156 2023-01-23 07:09:21.991464: step: 1256/1541, loss: 0.5975357890129089 2023-01-23 07:09:22.855272: step: 1258/1541, loss: 0.527336597442627 2023-01-23 07:09:23.770214: step: 1260/1541, loss: 1.5055923461914062 2023-01-23 07:09:24.696011: step: 1262/1541, loss: 3.231679677963257 2023-01-23 07:09:25.568104: step: 1264/1541, loss: 0.23533594608306885 2023-01-23 07:09:26.443488: step: 1266/1541, loss: 0.631679117679596 2023-01-23 07:09:27.344832: step: 1268/1541, loss: 0.5495873093605042 2023-01-23 07:09:28.236686: step: 1270/1541, loss: 0.7880931496620178 2023-01-23 07:09:29.127681: step: 1272/1541, loss: 1.115821361541748 2023-01-23 07:09:30.029466: step: 1274/1541, loss: 0.5413616299629211 2023-01-23 07:09:30.922396: step: 1276/1541, loss: 0.798545241355896 2023-01-23 07:09:31.846890: step: 1278/1541, loss: 0.6361815929412842 2023-01-23 07:09:32.756403: step: 1280/1541, loss: 0.37416473031044006 2023-01-23 07:09:33.658356: step: 1282/1541, loss: 0.5268056988716125 2023-01-23 07:09:34.523095: step: 1284/1541, loss: 0.49400752782821655 2023-01-23 07:09:35.444240: step: 1286/1541, loss: 1.4103567600250244 2023-01-23 07:09:36.421743: step: 1288/1541, loss: 0.7088167667388916 2023-01-23 07:09:37.349139: step: 1290/1541, loss: 2.6009528636932373 2023-01-23 07:09:38.302280: step: 1292/1541, loss: 0.4703707695007324 2023-01-23 07:09:39.234632: step: 1294/1541, loss: 0.7189325094223022 2023-01-23 07:09:40.122852: step: 1296/1541, loss: 0.6548221111297607 2023-01-23 07:09:40.996910: step: 1298/1541, loss: 0.6005114912986755 2023-01-23 07:09:41.890331: step: 1300/1541, loss: 0.31962981820106506 2023-01-23 07:09:42.766215: step: 1302/1541, loss: 0.9636273980140686 2023-01-23 07:09:43.688349: step: 1304/1541, loss: 7.468176364898682 2023-01-23 07:09:44.608338: step: 1306/1541, loss: 0.13319076597690582 2023-01-23 07:09:45.535814: step: 1308/1541, loss: 0.393419086933136 2023-01-23 07:09:46.413342: step: 1310/1541, loss: 0.34769245982170105 2023-01-23 07:09:47.314947: step: 1312/1541, loss: 0.5853679180145264 2023-01-23 07:09:48.234900: step: 1314/1541, loss: 0.912283182144165 2023-01-23 07:09:49.089668: step: 1316/1541, loss: 0.7436869144439697 2023-01-23 07:09:49.954000: step: 1318/1541, loss: 0.3926776647567749 2023-01-23 07:09:50.886468: step: 1320/1541, loss: 0.36410272121429443 2023-01-23 07:09:51.807918: step: 1322/1541, loss: 1.4057155847549438 2023-01-23 07:09:52.694071: step: 1324/1541, loss: 0.29958614706993103 2023-01-23 07:09:53.634526: step: 1326/1541, loss: 3.1735713481903076 2023-01-23 07:09:54.542588: step: 1328/1541, loss: 0.748624324798584 2023-01-23 07:09:55.438054: step: 1330/1541, loss: 0.5877135396003723 2023-01-23 07:09:56.313921: step: 1332/1541, loss: 0.39007872343063354 2023-01-23 07:09:57.246032: step: 1334/1541, loss: 6.784582614898682 2023-01-23 07:09:58.090589: step: 1336/1541, loss: 0.746543824672699 2023-01-23 07:09:58.938713: step: 1338/1541, loss: 0.24515214562416077 2023-01-23 07:09:59.866710: step: 1340/1541, loss: 6.733861446380615 2023-01-23 07:10:00.800604: step: 1342/1541, loss: 0.3436315059661865 2023-01-23 07:10:01.700257: step: 1344/1541, loss: 0.27686986327171326 2023-01-23 07:10:02.629806: step: 1346/1541, loss: 0.36159181594848633 2023-01-23 07:10:03.474394: step: 1348/1541, loss: 0.7487221956253052 2023-01-23 07:10:04.441620: step: 1350/1541, loss: 0.4117122292518616 2023-01-23 07:10:05.289022: step: 1352/1541, loss: 0.6206552982330322 2023-01-23 07:10:06.173172: step: 1354/1541, loss: 1.412663221359253 2023-01-23 07:10:07.066710: step: 1356/1541, loss: 0.4577077031135559 2023-01-23 07:10:07.963172: step: 1358/1541, loss: 0.33086177706718445 2023-01-23 07:10:08.877331: step: 1360/1541, loss: 0.6822864413261414 2023-01-23 07:10:09.800747: step: 1362/1541, loss: 0.9946550130844116 2023-01-23 07:10:10.693873: step: 1364/1541, loss: 6.044124126434326 2023-01-23 07:10:11.599090: step: 1366/1541, loss: 0.3239259719848633 2023-01-23 07:10:12.465410: step: 1368/1541, loss: 1.3310999870300293 2023-01-23 07:10:13.378288: step: 1370/1541, loss: 0.7912086844444275 2023-01-23 07:10:14.234207: step: 1372/1541, loss: 1.5469372272491455 2023-01-23 07:10:15.123645: step: 1374/1541, loss: 0.4334814250469208 2023-01-23 07:10:16.078507: step: 1376/1541, loss: 0.4963574707508087 2023-01-23 07:10:16.989144: step: 1378/1541, loss: 0.2679417133331299 2023-01-23 07:10:17.860340: step: 1380/1541, loss: 0.7041981220245361 2023-01-23 07:10:18.768730: step: 1382/1541, loss: 0.3239556849002838 2023-01-23 07:10:19.660982: step: 1384/1541, loss: 0.5806496739387512 2023-01-23 07:10:20.611110: step: 1386/1541, loss: 0.7596318125724792 2023-01-23 07:10:21.558669: step: 1388/1541, loss: 0.5190393924713135 2023-01-23 07:10:22.425874: step: 1390/1541, loss: 0.3505755066871643 2023-01-23 07:10:23.288508: step: 1392/1541, loss: 0.9166096448898315 2023-01-23 07:10:24.206030: step: 1394/1541, loss: 0.7298465967178345 2023-01-23 07:10:25.103318: step: 1396/1541, loss: 0.43578335642814636 2023-01-23 07:10:26.024749: step: 1398/1541, loss: 0.39258697628974915 2023-01-23 07:10:26.906736: step: 1400/1541, loss: 0.7265077233314514 2023-01-23 07:10:27.800477: step: 1402/1541, loss: 0.8392215967178345 2023-01-23 07:10:28.681256: step: 1404/1541, loss: 0.8964471220970154 2023-01-23 07:10:29.643353: step: 1406/1541, loss: 1.0078076124191284 2023-01-23 07:10:30.587991: step: 1408/1541, loss: 0.3053354322910309 2023-01-23 07:10:31.477389: step: 1410/1541, loss: 0.851036548614502 2023-01-23 07:10:32.438004: step: 1412/1541, loss: 0.5462062358856201 2023-01-23 07:10:33.330673: step: 1414/1541, loss: 0.9311255216598511 2023-01-23 07:10:34.198260: step: 1416/1541, loss: 0.9923354983329773 2023-01-23 07:10:35.080774: step: 1418/1541, loss: 1.5937767028808594 2023-01-23 07:10:35.959679: step: 1420/1541, loss: 1.977914571762085 2023-01-23 07:10:36.886981: step: 1422/1541, loss: 2.852006196975708 2023-01-23 07:10:37.783123: step: 1424/1541, loss: 0.4280204474925995 2023-01-23 07:10:38.711824: step: 1426/1541, loss: 0.744331419467926 2023-01-23 07:10:39.575616: step: 1428/1541, loss: 0.6444521546363831 2023-01-23 07:10:40.477723: step: 1430/1541, loss: 0.814409077167511 2023-01-23 07:10:41.402851: step: 1432/1541, loss: 1.1700477600097656 2023-01-23 07:10:42.311157: step: 1434/1541, loss: 0.78737872838974 2023-01-23 07:10:43.194746: step: 1436/1541, loss: 0.38766640424728394 2023-01-23 07:10:44.144184: step: 1438/1541, loss: 0.6575326919555664 2023-01-23 07:10:45.024488: step: 1440/1541, loss: 0.7336962223052979 2023-01-23 07:10:45.931389: step: 1442/1541, loss: 0.4976554214954376 2023-01-23 07:10:46.819220: step: 1444/1541, loss: 0.5149909853935242 2023-01-23 07:10:47.675944: step: 1446/1541, loss: 0.7516056895256042 2023-01-23 07:10:48.575184: step: 1448/1541, loss: 0.8908998966217041 2023-01-23 07:10:49.465156: step: 1450/1541, loss: 6.59464693069458 2023-01-23 07:10:50.379666: step: 1452/1541, loss: 3.1178171634674072 2023-01-23 07:10:51.268270: step: 1454/1541, loss: 0.3980099856853485 2023-01-23 07:10:52.216002: step: 1456/1541, loss: 0.37106001377105713 2023-01-23 07:10:53.120979: step: 1458/1541, loss: 0.820273756980896 2023-01-23 07:10:53.999128: step: 1460/1541, loss: 0.2751840054988861 2023-01-23 07:10:54.879194: step: 1462/1541, loss: 2.8544869422912598 2023-01-23 07:10:55.768221: step: 1464/1541, loss: 0.5324707627296448 2023-01-23 07:10:56.631196: step: 1466/1541, loss: 0.6259556412696838 2023-01-23 07:10:57.522757: step: 1468/1541, loss: 0.5420024394989014 2023-01-23 07:10:58.432627: step: 1470/1541, loss: 1.423000454902649 2023-01-23 07:10:59.344202: step: 1472/1541, loss: 0.4375300705432892 2023-01-23 07:11:00.224923: step: 1474/1541, loss: 2.4378983974456787 2023-01-23 07:11:01.087026: step: 1476/1541, loss: 0.415197491645813 2023-01-23 07:11:01.973136: step: 1478/1541, loss: 0.5927807092666626 2023-01-23 07:11:02.842020: step: 1480/1541, loss: 2.018341064453125 2023-01-23 07:11:03.704303: step: 1482/1541, loss: 4.123777866363525 2023-01-23 07:11:04.554200: step: 1484/1541, loss: 0.4469830095767975 2023-01-23 07:11:05.482405: step: 1486/1541, loss: 0.8481855392456055 2023-01-23 07:11:06.367284: step: 1488/1541, loss: 0.3410075306892395 2023-01-23 07:11:07.273889: step: 1490/1541, loss: 0.7632651925086975 2023-01-23 07:11:08.158155: step: 1492/1541, loss: 0.6970687508583069 2023-01-23 07:11:09.024734: step: 1494/1541, loss: 0.2455764263868332 2023-01-23 07:11:09.889404: step: 1496/1541, loss: 0.3506312072277069 2023-01-23 07:11:10.763971: step: 1498/1541, loss: 0.3761334717273712 2023-01-23 07:11:11.624988: step: 1500/1541, loss: 0.6780223846435547 2023-01-23 07:11:12.548659: step: 1502/1541, loss: 0.3659854233264923 2023-01-23 07:11:13.462145: step: 1504/1541, loss: 1.9130419492721558 2023-01-23 07:11:14.404530: step: 1506/1541, loss: 0.3091930150985718 2023-01-23 07:11:15.298194: step: 1508/1541, loss: 0.42545104026794434 2023-01-23 07:11:16.168154: step: 1510/1541, loss: 0.28110063076019287 2023-01-23 07:11:17.092301: step: 1512/1541, loss: 0.5087623596191406 2023-01-23 07:11:18.005661: step: 1514/1541, loss: 0.49900177121162415 2023-01-23 07:11:18.956156: step: 1516/1541, loss: 1.1835311651229858 2023-01-23 07:11:19.855688: step: 1518/1541, loss: 0.8421283960342407 2023-01-23 07:11:20.749712: step: 1520/1541, loss: 0.6373394727706909 2023-01-23 07:11:21.638549: step: 1522/1541, loss: 0.5297667980194092 2023-01-23 07:11:22.501578: step: 1524/1541, loss: 0.3584357798099518 2023-01-23 07:11:23.376750: step: 1526/1541, loss: 0.28850412368774414 2023-01-23 07:11:24.256476: step: 1528/1541, loss: 0.48592671751976013 2023-01-23 07:11:25.143225: step: 1530/1541, loss: 0.29682838916778564 2023-01-23 07:11:26.004408: step: 1532/1541, loss: 0.7604261636734009 2023-01-23 07:11:26.903206: step: 1534/1541, loss: 0.6481348276138306 2023-01-23 07:11:27.829519: step: 1536/1541, loss: 1.6676026582717896 2023-01-23 07:11:28.708025: step: 1538/1541, loss: 1.3601646423339844 2023-01-23 07:11:29.670235: step: 1540/1541, loss: 1.345478892326355 2023-01-23 07:11:30.600991: step: 1542/1541, loss: 1.2412645816802979 2023-01-23 07:11:31.507012: step: 1544/1541, loss: 0.726763129234314 2023-01-23 07:11:32.416457: step: 1546/1541, loss: 0.5773124694824219 2023-01-23 07:11:33.350455: step: 1548/1541, loss: 0.6168079376220703 2023-01-23 07:11:34.252230: step: 1550/1541, loss: 0.8383142948150635 2023-01-23 07:11:35.206627: step: 1552/1541, loss: 1.895760416984558 2023-01-23 07:11:36.155896: step: 1554/1541, loss: 0.4813665449619293 2023-01-23 07:11:37.052413: step: 1556/1541, loss: 0.7282122373580933 2023-01-23 07:11:37.951443: step: 1558/1541, loss: 1.5978951454162598 2023-01-23 07:11:38.855012: step: 1560/1541, loss: 0.3043636977672577 2023-01-23 07:11:39.814598: step: 1562/1541, loss: 0.40857163071632385 2023-01-23 07:11:40.710946: step: 1564/1541, loss: 0.8261014223098755 2023-01-23 07:11:41.596086: step: 1566/1541, loss: 0.5961694717407227 2023-01-23 07:11:42.495884: step: 1568/1541, loss: 1.7395622730255127 2023-01-23 07:11:43.385755: step: 1570/1541, loss: 0.45979973673820496 2023-01-23 07:11:44.279582: step: 1572/1541, loss: 0.5412535071372986 2023-01-23 07:11:45.219341: step: 1574/1541, loss: 1.7325899600982666 2023-01-23 07:11:46.182756: step: 1576/1541, loss: 1.6974241733551025 2023-01-23 07:11:47.065873: step: 1578/1541, loss: 0.5053521394729614 2023-01-23 07:11:47.954827: step: 1580/1541, loss: 1.551857829093933 2023-01-23 07:11:48.900631: step: 1582/1541, loss: 0.330157071352005 2023-01-23 07:11:49.797983: step: 1584/1541, loss: 0.4693409204483032 2023-01-23 07:11:50.682159: step: 1586/1541, loss: 0.3117128908634186 2023-01-23 07:11:51.615449: step: 1588/1541, loss: 1.8743761777877808 2023-01-23 07:11:52.529875: step: 1590/1541, loss: 0.5033422708511353 2023-01-23 07:11:53.397771: step: 1592/1541, loss: 0.3998693823814392 2023-01-23 07:11:54.281415: step: 1594/1541, loss: 0.24276110529899597 2023-01-23 07:11:55.142232: step: 1596/1541, loss: 0.6005135178565979 2023-01-23 07:11:55.995510: step: 1598/1541, loss: 0.27324655652046204 2023-01-23 07:11:56.951825: step: 1600/1541, loss: 0.5682473182678223 2023-01-23 07:11:57.842245: step: 1602/1541, loss: 0.7007032632827759 2023-01-23 07:11:58.889757: step: 1604/1541, loss: 4.04241943359375 2023-01-23 07:11:59.809607: step: 1606/1541, loss: 1.5418357849121094 2023-01-23 07:12:00.675378: step: 1608/1541, loss: 0.5847423076629639 2023-01-23 07:12:01.590474: step: 1610/1541, loss: 0.4671374559402466 2023-01-23 07:12:02.579982: step: 1612/1541, loss: 1.4064775705337524 2023-01-23 07:12:03.511489: step: 1614/1541, loss: 0.9222134947776794 2023-01-23 07:12:04.436338: step: 1616/1541, loss: 0.4858875274658203 2023-01-23 07:12:05.310422: step: 1618/1541, loss: 0.44146060943603516 2023-01-23 07:12:06.265113: step: 1620/1541, loss: 0.37794923782348633 2023-01-23 07:12:07.153837: step: 1622/1541, loss: 0.524360716342926 2023-01-23 07:12:08.039356: step: 1624/1541, loss: 0.5461680889129639 2023-01-23 07:12:08.910497: step: 1626/1541, loss: 0.3593406081199646 2023-01-23 07:12:09.805531: step: 1628/1541, loss: 0.6372485160827637 2023-01-23 07:12:10.679040: step: 1630/1541, loss: 0.7825139164924622 2023-01-23 07:12:11.623348: step: 1632/1541, loss: 0.8225283622741699 2023-01-23 07:12:12.515672: step: 1634/1541, loss: 0.237411767244339 2023-01-23 07:12:13.423581: step: 1636/1541, loss: 2.921783447265625 2023-01-23 07:12:14.299226: step: 1638/1541, loss: 0.6218536496162415 2023-01-23 07:12:15.166723: step: 1640/1541, loss: 0.5319405198097229 2023-01-23 07:12:16.064483: step: 1642/1541, loss: 4.377277374267578 2023-01-23 07:12:16.945820: step: 1644/1541, loss: 0.4046899378299713 2023-01-23 07:12:17.875794: step: 1646/1541, loss: 0.8604508638381958 2023-01-23 07:12:18.768103: step: 1648/1541, loss: 0.4711565673351288 2023-01-23 07:12:19.685837: step: 1650/1541, loss: 0.7416113018989563 2023-01-23 07:12:20.552142: step: 1652/1541, loss: 0.3560759127140045 2023-01-23 07:12:21.430541: step: 1654/1541, loss: 0.7791041135787964 2023-01-23 07:12:22.385566: step: 1656/1541, loss: 0.6601952314376831 2023-01-23 07:12:23.346176: step: 1658/1541, loss: 0.4796013832092285 2023-01-23 07:12:24.263427: step: 1660/1541, loss: 0.3238867521286011 2023-01-23 07:12:25.178071: step: 1662/1541, loss: 0.9560847282409668 2023-01-23 07:12:26.070251: step: 1664/1541, loss: 0.6033249497413635 2023-01-23 07:12:26.943877: step: 1666/1541, loss: 3.4829227924346924 2023-01-23 07:12:27.806700: step: 1668/1541, loss: 0.39191752672195435 2023-01-23 07:12:28.718731: step: 1670/1541, loss: 0.8254663348197937 2023-01-23 07:12:29.581053: step: 1672/1541, loss: 0.4350399076938629 2023-01-23 07:12:30.529264: step: 1674/1541, loss: 0.7588301301002502 2023-01-23 07:12:31.447571: step: 1676/1541, loss: 0.7878837585449219 2023-01-23 07:12:32.341324: step: 1678/1541, loss: 0.7410085797309875 2023-01-23 07:12:33.257268: step: 1680/1541, loss: 0.5985530614852905 2023-01-23 07:12:34.118709: step: 1682/1541, loss: 0.249372199177742 2023-01-23 07:12:35.015949: step: 1684/1541, loss: 0.9049299359321594 2023-01-23 07:12:35.926015: step: 1686/1541, loss: 0.7786528468132019 2023-01-23 07:12:36.797314: step: 1688/1541, loss: 1.206069827079773 2023-01-23 07:12:37.688717: step: 1690/1541, loss: 0.8576339483261108 2023-01-23 07:12:38.598096: step: 1692/1541, loss: 0.43716081976890564 2023-01-23 07:12:39.470487: step: 1694/1541, loss: 0.8808499574661255 2023-01-23 07:12:40.348654: step: 1696/1541, loss: 0.5345579385757446 2023-01-23 07:12:41.254545: step: 1698/1541, loss: 0.5438557863235474 2023-01-23 07:12:42.162225: step: 1700/1541, loss: 2.6615419387817383 2023-01-23 07:12:43.013419: step: 1702/1541, loss: 0.5211591124534607 2023-01-23 07:12:43.958939: step: 1704/1541, loss: 0.5010892152786255 2023-01-23 07:12:44.863859: step: 1706/1541, loss: 0.518625020980835 2023-01-23 07:12:45.822752: step: 1708/1541, loss: 0.24930566549301147 2023-01-23 07:12:46.710837: step: 1710/1541, loss: 0.6617375016212463 2023-01-23 07:12:47.564539: step: 1712/1541, loss: 0.8040343523025513 2023-01-23 07:12:48.495743: step: 1714/1541, loss: 1.9063901901245117 2023-01-23 07:12:49.379271: step: 1716/1541, loss: 0.745684802532196 2023-01-23 07:12:50.250291: step: 1718/1541, loss: 0.5673014521598816 2023-01-23 07:12:51.093247: step: 1720/1541, loss: 0.2605707049369812 2023-01-23 07:12:52.043132: step: 1722/1541, loss: 0.5950279235839844 2023-01-23 07:12:52.960803: step: 1724/1541, loss: 0.9718061089515686 2023-01-23 07:12:53.819305: step: 1726/1541, loss: 0.15923085808753967 2023-01-23 07:12:54.734630: step: 1728/1541, loss: 1.9030591249465942 2023-01-23 07:12:55.690827: step: 1730/1541, loss: 0.4040074646472931 2023-01-23 07:12:56.565644: step: 1732/1541, loss: 0.2816023528575897 2023-01-23 07:12:57.478181: step: 1734/1541, loss: 0.4042586386203766 2023-01-23 07:12:58.391205: step: 1736/1541, loss: 0.6893431544303894 2023-01-23 07:12:59.318855: step: 1738/1541, loss: 6.392664432525635 2023-01-23 07:13:00.222760: step: 1740/1541, loss: 0.388266921043396 2023-01-23 07:13:01.123607: step: 1742/1541, loss: 0.5214716792106628 2023-01-23 07:13:02.057013: step: 1744/1541, loss: 3.7225303649902344 2023-01-23 07:13:02.934097: step: 1746/1541, loss: 0.2758927345275879 2023-01-23 07:13:03.797207: step: 1748/1541, loss: 0.8663108348846436 2023-01-23 07:13:04.692661: step: 1750/1541, loss: 0.2063886821269989 2023-01-23 07:13:05.594538: step: 1752/1541, loss: 0.8208218812942505 2023-01-23 07:13:06.563261: step: 1754/1541, loss: 0.8172824382781982 2023-01-23 07:13:07.459760: step: 1756/1541, loss: 0.6100462675094604 2023-01-23 07:13:08.367112: step: 1758/1541, loss: 0.5066608786582947 2023-01-23 07:13:09.248774: step: 1760/1541, loss: 0.7537645697593689 2023-01-23 07:13:10.146493: step: 1762/1541, loss: 1.0091873407363892 2023-01-23 07:13:11.012519: step: 1764/1541, loss: 0.8778258562088013 2023-01-23 07:13:11.879440: step: 1766/1541, loss: 0.8796136975288391 2023-01-23 07:13:12.785252: step: 1768/1541, loss: 0.5704635977745056 2023-01-23 07:13:13.677885: step: 1770/1541, loss: 0.5928155183792114 2023-01-23 07:13:14.584026: step: 1772/1541, loss: 3.2192587852478027 2023-01-23 07:13:15.562608: step: 1774/1541, loss: 1.259356141090393 2023-01-23 07:13:16.540015: step: 1776/1541, loss: 0.5086947679519653 2023-01-23 07:13:17.450676: step: 1778/1541, loss: 0.5654317736625671 2023-01-23 07:13:18.340948: step: 1780/1541, loss: 1.028462290763855 2023-01-23 07:13:19.239757: step: 1782/1541, loss: 1.8443365097045898 2023-01-23 07:13:20.118900: step: 1784/1541, loss: 0.45674562454223633 2023-01-23 07:13:20.990062: step: 1786/1541, loss: 0.4199320375919342 2023-01-23 07:13:21.845214: step: 1788/1541, loss: 0.42111748456954956 2023-01-23 07:13:22.740168: step: 1790/1541, loss: 0.7534780502319336 2023-01-23 07:13:23.629527: step: 1792/1541, loss: 0.4507625997066498 2023-01-23 07:13:24.472564: step: 1794/1541, loss: 0.4904283285140991 2023-01-23 07:13:25.351197: step: 1796/1541, loss: 1.2002320289611816 2023-01-23 07:13:26.231372: step: 1798/1541, loss: 1.296310305595398 2023-01-23 07:13:27.107534: step: 1800/1541, loss: 1.1300678253173828 2023-01-23 07:13:27.993389: step: 1802/1541, loss: 0.3668097257614136 2023-01-23 07:13:28.881047: step: 1804/1541, loss: 0.6649410128593445 2023-01-23 07:13:29.785151: step: 1806/1541, loss: 0.5357317924499512 2023-01-23 07:13:30.727099: step: 1808/1541, loss: 1.1359283924102783 2023-01-23 07:13:31.630522: step: 1810/1541, loss: 0.23431912064552307 2023-01-23 07:13:32.554400: step: 1812/1541, loss: 0.590781569480896 2023-01-23 07:13:33.448917: step: 1814/1541, loss: 1.1400692462921143 2023-01-23 07:13:34.394702: step: 1816/1541, loss: 0.9807102084159851 2023-01-23 07:13:35.316042: step: 1818/1541, loss: 0.44451987743377686 2023-01-23 07:13:36.238152: step: 1820/1541, loss: 0.3916141986846924 2023-01-23 07:13:37.109430: step: 1822/1541, loss: 0.2126760184764862 2023-01-23 07:13:37.996050: step: 1824/1541, loss: 0.5824399590492249 2023-01-23 07:13:38.926171: step: 1826/1541, loss: 0.6717338562011719 2023-01-23 07:13:39.808983: step: 1828/1541, loss: 0.3491969406604767 2023-01-23 07:13:40.663379: step: 1830/1541, loss: 1.216694712638855 2023-01-23 07:13:41.525614: step: 1832/1541, loss: 0.45061686635017395 2023-01-23 07:13:42.425354: step: 1834/1541, loss: 0.3807448148727417 2023-01-23 07:13:43.273043: step: 1836/1541, loss: 0.31982484459877014 2023-01-23 07:13:44.178683: step: 1838/1541, loss: 0.35539180040359497 2023-01-23 07:13:45.076074: step: 1840/1541, loss: 0.3822993338108063 2023-01-23 07:13:46.041159: step: 1842/1541, loss: 0.7382948398590088 2023-01-23 07:13:47.028099: step: 1844/1541, loss: 0.7506478428840637 2023-01-23 07:13:47.965509: step: 1846/1541, loss: 0.34387677907943726 2023-01-23 07:13:48.838027: step: 1848/1541, loss: 0.4124142527580261 2023-01-23 07:13:49.734452: step: 1850/1541, loss: 0.4477013349533081 2023-01-23 07:13:50.628503: step: 1852/1541, loss: 0.5836302042007446 2023-01-23 07:13:51.584221: step: 1854/1541, loss: 0.5601509809494019 2023-01-23 07:13:52.520885: step: 1856/1541, loss: 0.5992144346237183 2023-01-23 07:13:53.416180: step: 1858/1541, loss: 0.4598747789859772 2023-01-23 07:13:54.288439: step: 1860/1541, loss: 0.4343724250793457 2023-01-23 07:13:55.222624: step: 1862/1541, loss: 0.6783872842788696 2023-01-23 07:13:56.096802: step: 1864/1541, loss: 0.5513607859611511 2023-01-23 07:13:57.025145: step: 1866/1541, loss: 0.509638249874115 2023-01-23 07:13:57.896447: step: 1868/1541, loss: 0.6495741605758667 2023-01-23 07:13:58.825701: step: 1870/1541, loss: 0.9945797920227051 2023-01-23 07:13:59.710196: step: 1872/1541, loss: 0.40146979689598083 2023-01-23 07:14:00.634445: step: 1874/1541, loss: 0.4479098618030548 2023-01-23 07:14:01.531922: step: 1876/1541, loss: 0.3504413962364197 2023-01-23 07:14:02.439575: step: 1878/1541, loss: 1.064084768295288 2023-01-23 07:14:03.378288: step: 1880/1541, loss: 0.6788337826728821 2023-01-23 07:14:04.284868: step: 1882/1541, loss: 0.5391358733177185 2023-01-23 07:14:05.252269: step: 1884/1541, loss: 0.5986460447311401 2023-01-23 07:14:06.169846: step: 1886/1541, loss: 1.0507512092590332 2023-01-23 07:14:07.061935: step: 1888/1541, loss: 0.28623372316360474 2023-01-23 07:14:07.990364: step: 1890/1541, loss: 0.750596821308136 2023-01-23 07:14:08.890458: step: 1892/1541, loss: 0.5575764179229736 2023-01-23 07:14:09.799866: step: 1894/1541, loss: 0.3993466794490814 2023-01-23 07:14:10.681547: step: 1896/1541, loss: 0.5182352066040039 2023-01-23 07:14:11.549594: step: 1898/1541, loss: 0.5117563009262085 2023-01-23 07:14:12.453448: step: 1900/1541, loss: 0.9480992555618286 2023-01-23 07:14:13.378042: step: 1902/1541, loss: 0.32087767124176025 2023-01-23 07:14:14.304517: step: 1904/1541, loss: 0.8266975283622742 2023-01-23 07:14:15.170607: step: 1906/1541, loss: 0.4017546474933624 2023-01-23 07:14:16.121429: step: 1908/1541, loss: 0.9960534572601318 2023-01-23 07:14:16.993109: step: 1910/1541, loss: 0.2293759286403656 2023-01-23 07:14:17.885215: step: 1912/1541, loss: 2.8708815574645996 2023-01-23 07:14:18.807166: step: 1914/1541, loss: 1.1036851406097412 2023-01-23 07:14:19.693427: step: 1916/1541, loss: 0.370820015668869 2023-01-23 07:14:20.584047: step: 1918/1541, loss: 0.8225513100624084 2023-01-23 07:14:21.471017: step: 1920/1541, loss: 0.7372161149978638 2023-01-23 07:14:22.355975: step: 1922/1541, loss: 0.3701385259628296 2023-01-23 07:14:23.287669: step: 1924/1541, loss: 1.7622582912445068 2023-01-23 07:14:24.279342: step: 1926/1541, loss: 0.48004212975502014 2023-01-23 07:14:25.190111: step: 1928/1541, loss: 1.6409085988998413 2023-01-23 07:14:26.106616: step: 1930/1541, loss: 0.27518585324287415 2023-01-23 07:14:26.981843: step: 1932/1541, loss: 0.9147339463233948 2023-01-23 07:14:27.877599: step: 1934/1541, loss: 0.2503392994403839 2023-01-23 07:14:28.810180: step: 1936/1541, loss: 1.1247316598892212 2023-01-23 07:14:29.726348: step: 1938/1541, loss: 1.6020492315292358 2023-01-23 07:14:30.628037: step: 1940/1541, loss: 2.3915293216705322 2023-01-23 07:14:31.498096: step: 1942/1541, loss: 0.49353229999542236 2023-01-23 07:14:32.421417: step: 1944/1541, loss: 2.250258684158325 2023-01-23 07:14:33.297456: step: 1946/1541, loss: 0.8497875332832336 2023-01-23 07:14:34.157236: step: 1948/1541, loss: 0.9345779418945312 2023-01-23 07:14:35.030690: step: 1950/1541, loss: 0.7539977431297302 2023-01-23 07:14:35.941980: step: 1952/1541, loss: 0.35122257471084595 2023-01-23 07:14:36.844439: step: 1954/1541, loss: 0.2326013296842575 2023-01-23 07:14:37.748779: step: 1956/1541, loss: 0.8697162866592407 2023-01-23 07:14:38.670044: step: 1958/1541, loss: 1.2743438482284546 2023-01-23 07:14:39.546472: step: 1960/1541, loss: 0.4109419584274292 2023-01-23 07:14:40.460870: step: 1962/1541, loss: 6.23045015335083 2023-01-23 07:14:41.335587: step: 1964/1541, loss: 0.6304009556770325 2023-01-23 07:14:42.208664: step: 1966/1541, loss: 0.4579157531261444 2023-01-23 07:14:43.096664: step: 1968/1541, loss: 0.6711437702178955 2023-01-23 07:14:43.985723: step: 1970/1541, loss: 0.4779266119003296 2023-01-23 07:14:44.912811: step: 1972/1541, loss: 0.4288347661495209 2023-01-23 07:14:45.815053: step: 1974/1541, loss: 1.1983976364135742 2023-01-23 07:14:46.728898: step: 1976/1541, loss: 2.992487907409668 2023-01-23 07:14:47.605096: step: 1978/1541, loss: 0.5515707731246948 2023-01-23 07:14:48.517806: step: 1980/1541, loss: 1.301308274269104 2023-01-23 07:14:49.391057: step: 1982/1541, loss: 0.5848730206489563 2023-01-23 07:14:50.271880: step: 1984/1541, loss: 0.8791427612304688 2023-01-23 07:14:51.144677: step: 1986/1541, loss: 0.4238378405570984 2023-01-23 07:14:52.117822: step: 1988/1541, loss: 1.5932902097702026 2023-01-23 07:14:53.043486: step: 1990/1541, loss: 1.106336236000061 2023-01-23 07:14:54.006416: step: 1992/1541, loss: 0.7627699971199036 2023-01-23 07:14:54.962359: step: 1994/1541, loss: 0.7650936245918274 2023-01-23 07:14:55.912216: step: 1996/1541, loss: 1.2915705442428589 2023-01-23 07:14:56.871185: step: 1998/1541, loss: 0.6465831398963928 2023-01-23 07:14:57.764119: step: 2000/1541, loss: 0.9116138815879822 2023-01-23 07:14:58.645723: step: 2002/1541, loss: 0.5836620330810547 2023-01-23 07:14:59.563978: step: 2004/1541, loss: 0.24719980359077454 2023-01-23 07:15:00.432213: step: 2006/1541, loss: 1.1516579389572144 2023-01-23 07:15:01.370488: step: 2008/1541, loss: 0.7630872130393982 2023-01-23 07:15:02.249224: step: 2010/1541, loss: 0.421832412481308 2023-01-23 07:15:03.185785: step: 2012/1541, loss: 3.107499361038208 2023-01-23 07:15:04.074846: step: 2014/1541, loss: 1.0662018060684204 2023-01-23 07:15:04.994985: step: 2016/1541, loss: 1.0698105096817017 2023-01-23 07:15:05.862894: step: 2018/1541, loss: 0.6510067582130432 2023-01-23 07:15:06.754642: step: 2020/1541, loss: 1.277667760848999 2023-01-23 07:15:07.649942: step: 2022/1541, loss: 0.7353147864341736 2023-01-23 07:15:08.566799: step: 2024/1541, loss: 0.7980136871337891 2023-01-23 07:15:09.457516: step: 2026/1541, loss: 0.6167649030685425 2023-01-23 07:15:10.370764: step: 2028/1541, loss: 0.9510681629180908 2023-01-23 07:15:11.294503: step: 2030/1541, loss: 0.7470505833625793 2023-01-23 07:15:12.146430: step: 2032/1541, loss: 0.24015139043331146 2023-01-23 07:15:13.025198: step: 2034/1541, loss: 0.6034011840820312 2023-01-23 07:15:13.963815: step: 2036/1541, loss: 0.49236762523651123 2023-01-23 07:15:14.843217: step: 2038/1541, loss: 0.6386058330535889 2023-01-23 07:15:15.716637: step: 2040/1541, loss: 0.703448474407196 2023-01-23 07:15:16.677145: step: 2042/1541, loss: 1.1634674072265625 2023-01-23 07:15:17.589684: step: 2044/1541, loss: 0.3256000578403473 2023-01-23 07:15:18.489081: step: 2046/1541, loss: 0.6635715365409851 2023-01-23 07:15:19.424522: step: 2048/1541, loss: 0.9901002645492554 2023-01-23 07:15:20.352085: step: 2050/1541, loss: 0.8227041363716125 2023-01-23 07:15:21.229003: step: 2052/1541, loss: 0.5537164807319641 2023-01-23 07:15:22.169469: step: 2054/1541, loss: 1.1429747343063354 2023-01-23 07:15:23.058929: step: 2056/1541, loss: 0.35180795192718506 2023-01-23 07:15:23.931950: step: 2058/1541, loss: 0.4150044918060303 2023-01-23 07:15:24.857304: step: 2060/1541, loss: 0.34383878111839294 2023-01-23 07:15:25.715720: step: 2062/1541, loss: 0.5689491033554077 2023-01-23 07:15:26.610138: step: 2064/1541, loss: 0.2737715542316437 2023-01-23 07:15:27.484022: step: 2066/1541, loss: 0.9809677600860596 2023-01-23 07:15:28.416772: step: 2068/1541, loss: 0.5107704997062683 2023-01-23 07:15:29.296521: step: 2070/1541, loss: 0.31426629424095154 2023-01-23 07:15:30.233866: step: 2072/1541, loss: 1.3735378980636597 2023-01-23 07:15:31.146438: step: 2074/1541, loss: 0.5535274744033813 2023-01-23 07:15:32.025587: step: 2076/1541, loss: 1.060181975364685 2023-01-23 07:15:32.958730: step: 2078/1541, loss: 0.5396811962127686 2023-01-23 07:15:33.843656: step: 2080/1541, loss: 0.5751852989196777 2023-01-23 07:15:34.743371: step: 2082/1541, loss: 0.9969896674156189 2023-01-23 07:15:35.656231: step: 2084/1541, loss: 0.765409529209137 2023-01-23 07:15:36.561187: step: 2086/1541, loss: 0.5760477185249329 2023-01-23 07:15:37.445292: step: 2088/1541, loss: 0.9968861937522888 2023-01-23 07:15:38.371379: step: 2090/1541, loss: 0.29680144786834717 2023-01-23 07:15:39.211306: step: 2092/1541, loss: 2.6680071353912354 2023-01-23 07:15:40.092113: step: 2094/1541, loss: 0.6043668389320374 2023-01-23 07:15:41.016852: step: 2096/1541, loss: 0.5349074006080627 2023-01-23 07:15:41.936323: step: 2098/1541, loss: 0.7518582344055176 2023-01-23 07:15:42.798316: step: 2100/1541, loss: 0.8822633624076843 2023-01-23 07:15:43.681406: step: 2102/1541, loss: 0.681294322013855 2023-01-23 07:15:44.592891: step: 2104/1541, loss: 0.3540164828300476 2023-01-23 07:15:45.483141: step: 2106/1541, loss: 0.2729845643043518 2023-01-23 07:15:46.333182: step: 2108/1541, loss: 0.5652635097503662 2023-01-23 07:15:47.220351: step: 2110/1541, loss: 0.6011605262756348 2023-01-23 07:15:48.208976: step: 2112/1541, loss: 0.7067053914070129 2023-01-23 07:15:49.193116: step: 2114/1541, loss: 0.519205629825592 2023-01-23 07:15:50.115447: step: 2116/1541, loss: 0.9279382824897766 2023-01-23 07:15:51.031827: step: 2118/1541, loss: 1.1799254417419434 2023-01-23 07:15:51.917018: step: 2120/1541, loss: 3.172380208969116 2023-01-23 07:15:52.857636: step: 2122/1541, loss: 0.4973227381706238 2023-01-23 07:15:53.736719: step: 2124/1541, loss: 0.620032787322998 2023-01-23 07:15:54.635270: step: 2126/1541, loss: 0.4614831805229187 2023-01-23 07:15:55.519618: step: 2128/1541, loss: 1.167452096939087 2023-01-23 07:15:56.387352: step: 2130/1541, loss: 0.28953680396080017 2023-01-23 07:15:57.279051: step: 2132/1541, loss: 0.49015092849731445 2023-01-23 07:15:58.151412: step: 2134/1541, loss: 0.5605309009552002 2023-01-23 07:15:59.030654: step: 2136/1541, loss: 0.42236971855163574 2023-01-23 07:15:59.892948: step: 2138/1541, loss: 0.2354937195777893 2023-01-23 07:16:00.848175: step: 2140/1541, loss: 0.5503669381141663 2023-01-23 07:16:01.713023: step: 2142/1541, loss: 0.7248419523239136 2023-01-23 07:16:02.644516: step: 2144/1541, loss: 0.941286563873291 2023-01-23 07:16:03.522502: step: 2146/1541, loss: 0.4829491972923279 2023-01-23 07:16:04.433613: step: 2148/1541, loss: 0.3410942852497101 2023-01-23 07:16:05.422495: step: 2150/1541, loss: 0.7025178074836731 2023-01-23 07:16:06.260870: step: 2152/1541, loss: 0.4511028528213501 2023-01-23 07:16:07.237447: step: 2154/1541, loss: 0.6079168319702148 2023-01-23 07:16:08.110805: step: 2156/1541, loss: 0.34219375252723694 2023-01-23 07:16:09.002646: step: 2158/1541, loss: 1.2100647687911987 2023-01-23 07:16:09.897991: step: 2160/1541, loss: 0.3786611557006836 2023-01-23 07:16:10.792389: step: 2162/1541, loss: 0.7104100584983826 2023-01-23 07:16:11.690655: step: 2164/1541, loss: 0.43934836983680725 2023-01-23 07:16:12.559028: step: 2166/1541, loss: 0.6811844706535339 2023-01-23 07:16:13.555416: step: 2168/1541, loss: 1.5092085599899292 2023-01-23 07:16:14.442832: step: 2170/1541, loss: 0.963554859161377 2023-01-23 07:16:15.325721: step: 2172/1541, loss: 0.2602635324001312 2023-01-23 07:16:16.241681: step: 2174/1541, loss: 0.7036210298538208 2023-01-23 07:16:17.115691: step: 2176/1541, loss: 0.873715341091156 2023-01-23 07:16:18.003102: step: 2178/1541, loss: 0.6282404065132141 2023-01-23 07:16:18.868519: step: 2180/1541, loss: 0.922182559967041 2023-01-23 07:16:19.741823: step: 2182/1541, loss: 0.5874648094177246 2023-01-23 07:16:20.619199: step: 2184/1541, loss: 0.5108764171600342 2023-01-23 07:16:21.510214: step: 2186/1541, loss: 1.03078031539917 2023-01-23 07:16:22.426500: step: 2188/1541, loss: 0.9313204288482666 2023-01-23 07:16:23.352902: step: 2190/1541, loss: 0.16730783879756927 2023-01-23 07:16:24.335035: step: 2192/1541, loss: 0.8756262063980103 2023-01-23 07:16:25.267444: step: 2194/1541, loss: 0.4968935251235962 2023-01-23 07:16:26.176391: step: 2196/1541, loss: 0.41921573877334595 2023-01-23 07:16:27.045815: step: 2198/1541, loss: 0.6754696369171143 2023-01-23 07:16:27.928813: step: 2200/1541, loss: 0.2914191484451294 2023-01-23 07:16:28.831373: step: 2202/1541, loss: 0.8192645311355591 2023-01-23 07:16:29.709369: step: 2204/1541, loss: 0.5609957575798035 2023-01-23 07:16:30.653438: step: 2206/1541, loss: 0.3729604184627533 2023-01-23 07:16:31.562205: step: 2208/1541, loss: 0.8917196393013 2023-01-23 07:16:32.479845: step: 2210/1541, loss: 0.9163622856140137 2023-01-23 07:16:33.383570: step: 2212/1541, loss: 0.722424328327179 2023-01-23 07:16:34.278214: step: 2214/1541, loss: 0.6305776834487915 2023-01-23 07:16:35.220957: step: 2216/1541, loss: 0.6287232041358948 2023-01-23 07:16:36.139157: step: 2218/1541, loss: 0.6542650461196899 2023-01-23 07:16:37.017635: step: 2220/1541, loss: 0.6680896878242493 2023-01-23 07:16:37.916285: step: 2222/1541, loss: 0.6784453988075256 2023-01-23 07:16:38.790160: step: 2224/1541, loss: 0.3303799331188202 2023-01-23 07:16:39.693814: step: 2226/1541, loss: 0.4983578324317932 2023-01-23 07:16:40.591203: step: 2228/1541, loss: 0.35692569613456726 2023-01-23 07:16:41.487500: step: 2230/1541, loss: 0.28957393765449524 2023-01-23 07:16:42.374510: step: 2232/1541, loss: 0.22048312425613403 2023-01-23 07:16:43.291930: step: 2234/1541, loss: 6.603104591369629 2023-01-23 07:16:44.181418: step: 2236/1541, loss: 0.47285759449005127 2023-01-23 07:16:45.093786: step: 2238/1541, loss: 1.2077679634094238 2023-01-23 07:16:45.977279: step: 2240/1541, loss: 0.4671773910522461 2023-01-23 07:16:46.868706: step: 2242/1541, loss: 0.6694093942642212 2023-01-23 07:16:47.834312: step: 2244/1541, loss: 0.179234579205513 2023-01-23 07:16:48.721655: step: 2246/1541, loss: 2.423741340637207 2023-01-23 07:16:49.643598: step: 2248/1541, loss: 1.3042333126068115 2023-01-23 07:16:50.513883: step: 2250/1541, loss: 0.3568224310874939 2023-01-23 07:16:51.424799: step: 2252/1541, loss: 0.8437696695327759 2023-01-23 07:16:52.372029: step: 2254/1541, loss: 0.9420576691627502 2023-01-23 07:16:53.250013: step: 2256/1541, loss: 0.658880352973938 2023-01-23 07:16:54.119600: step: 2258/1541, loss: 1.0848506689071655 2023-01-23 07:16:55.011959: step: 2260/1541, loss: 0.4516589641571045 2023-01-23 07:16:55.906547: step: 2262/1541, loss: 0.6111169457435608 2023-01-23 07:16:56.830273: step: 2264/1541, loss: 0.3461894392967224 2023-01-23 07:16:57.743969: step: 2266/1541, loss: 0.5406985282897949 2023-01-23 07:16:58.630447: step: 2268/1541, loss: 1.521793246269226 2023-01-23 07:16:59.521336: step: 2270/1541, loss: 0.5028892159461975 2023-01-23 07:17:00.434620: step: 2272/1541, loss: 0.8034173846244812 2023-01-23 07:17:01.283232: step: 2274/1541, loss: 0.41552141308784485 2023-01-23 07:17:02.237992: step: 2276/1541, loss: 0.22813817858695984 2023-01-23 07:17:03.171786: step: 2278/1541, loss: 0.9833031296730042 2023-01-23 07:17:04.035467: step: 2280/1541, loss: 0.34420397877693176 2023-01-23 07:17:04.980821: step: 2282/1541, loss: 0.39402705430984497 2023-01-23 07:17:05.888763: step: 2284/1541, loss: 0.39424461126327515 2023-01-23 07:17:06.770837: step: 2286/1541, loss: 0.43677839636802673 2023-01-23 07:17:07.678687: step: 2288/1541, loss: 0.2438240945339203 2023-01-23 07:17:08.575584: step: 2290/1541, loss: 2.7714943885803223 2023-01-23 07:17:09.461300: step: 2292/1541, loss: 0.41683852672576904 2023-01-23 07:17:10.366946: step: 2294/1541, loss: 0.5124780535697937 2023-01-23 07:17:11.227068: step: 2296/1541, loss: 0.665320634841919 2023-01-23 07:17:12.135698: step: 2298/1541, loss: 1.2101519107818604 2023-01-23 07:17:13.089138: step: 2300/1541, loss: 0.5142046213150024 2023-01-23 07:17:14.000311: step: 2302/1541, loss: 1.1562988758087158 2023-01-23 07:17:14.874849: step: 2304/1541, loss: 2.225954532623291 2023-01-23 07:17:15.821517: step: 2306/1541, loss: 0.6742997169494629 2023-01-23 07:17:16.700753: step: 2308/1541, loss: 0.24211157858371735 2023-01-23 07:17:17.680181: step: 2310/1541, loss: 1.030198574066162 2023-01-23 07:17:18.575072: step: 2312/1541, loss: 1.034774899482727 2023-01-23 07:17:19.466332: step: 2314/1541, loss: 0.8612681031227112 2023-01-23 07:17:20.330568: step: 2316/1541, loss: 0.4426204562187195 2023-01-23 07:17:21.238642: step: 2318/1541, loss: 3.7215704917907715 2023-01-23 07:17:22.095737: step: 2320/1541, loss: 0.3916597366333008 2023-01-23 07:17:23.021953: step: 2322/1541, loss: 0.5731411576271057 2023-01-23 07:17:23.891192: step: 2324/1541, loss: 1.137941837310791 2023-01-23 07:17:24.761286: step: 2326/1541, loss: 0.35440245270729065 2023-01-23 07:17:25.634145: step: 2328/1541, loss: 0.4396423399448395 2023-01-23 07:17:26.529885: step: 2330/1541, loss: 0.8140539526939392 2023-01-23 07:17:27.439063: step: 2332/1541, loss: 0.391892671585083 2023-01-23 07:17:28.318296: step: 2334/1541, loss: 0.37687721848487854 2023-01-23 07:17:29.224501: step: 2336/1541, loss: 0.28278249502182007 2023-01-23 07:17:30.122507: step: 2338/1541, loss: 1.125381588935852 2023-01-23 07:17:31.015392: step: 2340/1541, loss: 0.8060166239738464 2023-01-23 07:17:31.930524: step: 2342/1541, loss: 0.4149913489818573 2023-01-23 07:17:32.854369: step: 2344/1541, loss: 0.5160351991653442 2023-01-23 07:17:33.739108: step: 2346/1541, loss: 0.4336674213409424 2023-01-23 07:17:34.663825: step: 2348/1541, loss: 0.9636411070823669 2023-01-23 07:17:35.577018: step: 2350/1541, loss: 1.4764814376831055 2023-01-23 07:17:36.509736: step: 2352/1541, loss: 0.4611832797527313 2023-01-23 07:17:37.403036: step: 2354/1541, loss: 0.7047207951545715 2023-01-23 07:17:38.285782: step: 2356/1541, loss: 0.5265806317329407 2023-01-23 07:17:39.215733: step: 2358/1541, loss: 0.5202939510345459 2023-01-23 07:17:40.067423: step: 2360/1541, loss: 0.3765532374382019 2023-01-23 07:17:41.030345: step: 2362/1541, loss: 0.24486835300922394 2023-01-23 07:17:41.930395: step: 2364/1541, loss: 0.5140266418457031 2023-01-23 07:17:42.798118: step: 2366/1541, loss: 0.5198190212249756 2023-01-23 07:17:43.684865: step: 2368/1541, loss: 0.6194607615470886 2023-01-23 07:17:44.640672: step: 2370/1541, loss: 1.0483684539794922 2023-01-23 07:17:45.513483: step: 2372/1541, loss: 0.7178084850311279 2023-01-23 07:17:46.386444: step: 2374/1541, loss: 1.078718662261963 2023-01-23 07:17:47.328631: step: 2376/1541, loss: 1.1152591705322266 2023-01-23 07:17:48.197580: step: 2378/1541, loss: 0.752684473991394 2023-01-23 07:17:49.089683: step: 2380/1541, loss: 0.48379093408584595 2023-01-23 07:17:50.001967: step: 2382/1541, loss: 0.32187771797180176 2023-01-23 07:17:50.880610: step: 2384/1541, loss: 1.2325178384780884 2023-01-23 07:17:51.894530: step: 2386/1541, loss: 0.4681425392627716 2023-01-23 07:17:52.793251: step: 2388/1541, loss: 2.900757074356079 2023-01-23 07:17:53.654133: step: 2390/1541, loss: 0.38167160749435425 2023-01-23 07:17:54.539089: step: 2392/1541, loss: 0.30892303586006165 2023-01-23 07:17:55.448167: step: 2394/1541, loss: 1.020729660987854 2023-01-23 07:17:56.318360: step: 2396/1541, loss: 0.6562190651893616 2023-01-23 07:17:57.220084: step: 2398/1541, loss: 0.8769307136535645 2023-01-23 07:17:58.182932: step: 2400/1541, loss: 1.3794182538986206 2023-01-23 07:17:59.049247: step: 2402/1541, loss: 0.5977663397789001 2023-01-23 07:17:59.945599: step: 2404/1541, loss: 0.26315367221832275 2023-01-23 07:18:00.906307: step: 2406/1541, loss: 0.5546493530273438 2023-01-23 07:18:01.864354: step: 2408/1541, loss: 0.33775800466537476 2023-01-23 07:18:02.760221: step: 2410/1541, loss: 0.5247077345848083 2023-01-23 07:18:03.661761: step: 2412/1541, loss: 0.32465389370918274 2023-01-23 07:18:04.548484: step: 2414/1541, loss: 0.633709192276001 2023-01-23 07:18:05.439134: step: 2416/1541, loss: 0.2966197729110718 2023-01-23 07:18:06.326410: step: 2418/1541, loss: 0.6635856628417969 2023-01-23 07:18:07.212675: step: 2420/1541, loss: 0.5532323718070984 2023-01-23 07:18:08.232829: step: 2422/1541, loss: 0.5302297472953796 2023-01-23 07:18:09.140855: step: 2424/1541, loss: 0.4371577501296997 2023-01-23 07:18:10.051563: step: 2426/1541, loss: 0.3448833227157593 2023-01-23 07:18:10.953413: step: 2428/1541, loss: 0.5099800825119019 2023-01-23 07:18:11.826655: step: 2430/1541, loss: 2.1319234371185303 2023-01-23 07:18:12.695637: step: 2432/1541, loss: 0.6068223714828491 2023-01-23 07:18:13.534168: step: 2434/1541, loss: 1.4680486917495728 2023-01-23 07:18:14.448576: step: 2436/1541, loss: 2.2968356609344482 2023-01-23 07:18:15.389674: step: 2438/1541, loss: 2.589118003845215 2023-01-23 07:18:16.355291: step: 2440/1541, loss: 0.7587569355964661 2023-01-23 07:18:17.253764: step: 2442/1541, loss: 2.8511953353881836 2023-01-23 07:18:18.216273: step: 2444/1541, loss: 0.908310055732727 2023-01-23 07:18:19.135487: step: 2446/1541, loss: 0.2868632674217224 2023-01-23 07:18:20.020897: step: 2448/1541, loss: 0.18545836210250854 2023-01-23 07:18:20.967954: step: 2450/1541, loss: 0.28506332635879517 2023-01-23 07:18:21.840642: step: 2452/1541, loss: 0.3236418664455414 2023-01-23 07:18:22.754851: step: 2454/1541, loss: 1.5837562084197998 2023-01-23 07:18:23.647932: step: 2456/1541, loss: 0.7482810020446777 2023-01-23 07:18:24.558569: step: 2458/1541, loss: 1.9932167530059814 2023-01-23 07:18:25.485068: step: 2460/1541, loss: 0.4933367371559143 2023-01-23 07:18:26.415358: step: 2462/1541, loss: 0.4432217478752136 2023-01-23 07:18:27.303703: step: 2464/1541, loss: 0.5050555467605591 2023-01-23 07:18:28.254443: step: 2466/1541, loss: 0.9965957999229431 2023-01-23 07:18:29.181246: step: 2468/1541, loss: 0.5072495937347412 2023-01-23 07:18:30.094632: step: 2470/1541, loss: 0.6645073294639587 2023-01-23 07:18:30.967196: step: 2472/1541, loss: 0.4126676023006439 2023-01-23 07:18:31.948552: step: 2474/1541, loss: 0.5013630390167236 2023-01-23 07:18:32.852224: step: 2476/1541, loss: 1.4877753257751465 2023-01-23 07:18:33.748043: step: 2478/1541, loss: 0.2759493887424469 2023-01-23 07:18:34.605104: step: 2480/1541, loss: 0.4148620367050171 2023-01-23 07:18:35.489908: step: 2482/1541, loss: 0.8101373314857483 2023-01-23 07:18:36.357640: step: 2484/1541, loss: 0.39137062430381775 2023-01-23 07:18:37.220237: step: 2486/1541, loss: 0.638875424861908 2023-01-23 07:18:38.157393: step: 2488/1541, loss: 0.26264098286628723 2023-01-23 07:18:39.041510: step: 2490/1541, loss: 1.7871953248977661 2023-01-23 07:18:39.959368: step: 2492/1541, loss: 0.9743613600730896 2023-01-23 07:18:40.829961: step: 2494/1541, loss: 0.3818552792072296 2023-01-23 07:18:41.687139: step: 2496/1541, loss: 0.4915851652622223 2023-01-23 07:18:42.551717: step: 2498/1541, loss: 1.2495105266571045 2023-01-23 07:18:43.420412: step: 2500/1541, loss: 1.3675143718719482 2023-01-23 07:18:44.309381: step: 2502/1541, loss: 2.491264581680298 2023-01-23 07:18:45.169466: step: 2504/1541, loss: 0.3008931577205658 2023-01-23 07:18:46.062618: step: 2506/1541, loss: 0.7330483198165894 2023-01-23 07:18:47.017769: step: 2508/1541, loss: 2.074639320373535 2023-01-23 07:18:47.900139: step: 2510/1541, loss: 0.4366958439350128 2023-01-23 07:18:48.804525: step: 2512/1541, loss: 1.0131572484970093 2023-01-23 07:18:49.713723: step: 2514/1541, loss: 0.599662184715271 2023-01-23 07:18:50.605497: step: 2516/1541, loss: 0.4607948362827301 2023-01-23 07:18:51.473105: step: 2518/1541, loss: 2.9410479068756104 2023-01-23 07:18:52.341148: step: 2520/1541, loss: 0.8931824564933777 2023-01-23 07:18:53.200058: step: 2522/1541, loss: 2.498070478439331 2023-01-23 07:18:54.095397: step: 2524/1541, loss: 0.6885505318641663 2023-01-23 07:18:54.969952: step: 2526/1541, loss: 0.7279673218727112 2023-01-23 07:18:55.840409: step: 2528/1541, loss: 1.0098971128463745 2023-01-23 07:18:56.737598: step: 2530/1541, loss: 0.9701506495475769 2023-01-23 07:18:57.660903: step: 2532/1541, loss: 0.6906026601791382 2023-01-23 07:18:58.572259: step: 2534/1541, loss: 0.6084616780281067 2023-01-23 07:18:59.563803: step: 2536/1541, loss: 0.46532103419303894 2023-01-23 07:19:00.432264: step: 2538/1541, loss: 0.7914435267448425 2023-01-23 07:19:01.330536: step: 2540/1541, loss: 0.4780843257904053 2023-01-23 07:19:02.241809: step: 2542/1541, loss: 0.3173428177833557 2023-01-23 07:19:03.134472: step: 2544/1541, loss: 0.5414087772369385 2023-01-23 07:19:04.032860: step: 2546/1541, loss: 0.36078470945358276 2023-01-23 07:19:04.987776: step: 2548/1541, loss: 0.2539256811141968 2023-01-23 07:19:05.940389: step: 2550/1541, loss: 0.5829547047615051 2023-01-23 07:19:06.889080: step: 2552/1541, loss: 1.1484577655792236 2023-01-23 07:19:07.794848: step: 2554/1541, loss: 1.480112910270691 2023-01-23 07:19:08.677807: step: 2556/1541, loss: 0.40347936749458313 2023-01-23 07:19:09.551021: step: 2558/1541, loss: 1.1892870664596558 2023-01-23 07:19:10.499056: step: 2560/1541, loss: 0.7107985615730286 2023-01-23 07:19:11.364530: step: 2562/1541, loss: 0.4401226341724396 2023-01-23 07:19:12.321031: step: 2564/1541, loss: 0.8963729739189148 2023-01-23 07:19:13.183671: step: 2566/1541, loss: 0.3928122818470001 2023-01-23 07:19:14.064698: step: 2568/1541, loss: 0.40643271803855896 2023-01-23 07:19:14.946185: step: 2570/1541, loss: 0.7775219082832336 2023-01-23 07:19:15.841397: step: 2572/1541, loss: 0.651188850402832 2023-01-23 07:19:16.773392: step: 2574/1541, loss: 0.3563900589942932 2023-01-23 07:19:17.643350: step: 2576/1541, loss: 1.0825241804122925 2023-01-23 07:19:18.517419: step: 2578/1541, loss: 0.8618040084838867 2023-01-23 07:19:19.517577: step: 2580/1541, loss: 0.2098834216594696 2023-01-23 07:19:20.397816: step: 2582/1541, loss: 0.4352027475833893 2023-01-23 07:19:21.314931: step: 2584/1541, loss: 1.0893901586532593 2023-01-23 07:19:22.191608: step: 2586/1541, loss: 1.1210989952087402 2023-01-23 07:19:23.076873: step: 2588/1541, loss: 0.4020994007587433 2023-01-23 07:19:23.958991: step: 2590/1541, loss: 1.2864526510238647 2023-01-23 07:19:24.917187: step: 2592/1541, loss: 0.7348308563232422 2023-01-23 07:19:25.799831: step: 2594/1541, loss: 0.5219045877456665 2023-01-23 07:19:26.704589: step: 2596/1541, loss: 0.8255284428596497 2023-01-23 07:19:27.641508: step: 2598/1541, loss: 0.7234364748001099 2023-01-23 07:19:28.517332: step: 2600/1541, loss: 0.3280467092990875 2023-01-23 07:19:29.481428: step: 2602/1541, loss: 0.6687191128730774 2023-01-23 07:19:30.405364: step: 2604/1541, loss: 0.6326919198036194 2023-01-23 07:19:31.301013: step: 2606/1541, loss: 0.8390543460845947 2023-01-23 07:19:32.245869: step: 2608/1541, loss: 0.4657142758369446 2023-01-23 07:19:33.120607: step: 2610/1541, loss: 0.6110463738441467 2023-01-23 07:19:34.026866: step: 2612/1541, loss: 0.5550912618637085 2023-01-23 07:19:35.015946: step: 2614/1541, loss: 0.6102555990219116 2023-01-23 07:19:35.916590: step: 2616/1541, loss: 1.4807122945785522 2023-01-23 07:19:36.770746: step: 2618/1541, loss: 0.8026694655418396 2023-01-23 07:19:37.661134: step: 2620/1541, loss: 0.5977328419685364 2023-01-23 07:19:38.521069: step: 2622/1541, loss: 0.4359835982322693 2023-01-23 07:19:39.407511: step: 2624/1541, loss: 0.6213133335113525 2023-01-23 07:19:40.319383: step: 2626/1541, loss: 0.5593006014823914 2023-01-23 07:19:41.208605: step: 2628/1541, loss: 0.6549128293991089 2023-01-23 07:19:42.105495: step: 2630/1541, loss: 0.31414639949798584 2023-01-23 07:19:43.003413: step: 2632/1541, loss: 0.7816166281700134 2023-01-23 07:19:43.900119: step: 2634/1541, loss: 0.6091349720954895 2023-01-23 07:19:44.834647: step: 2636/1541, loss: 0.317308247089386 2023-01-23 07:19:45.797754: step: 2638/1541, loss: 7.318789005279541 2023-01-23 07:19:46.695420: step: 2640/1541, loss: 0.5807124376296997 2023-01-23 07:19:47.587867: step: 2642/1541, loss: 0.30474427342414856 2023-01-23 07:19:48.466714: step: 2644/1541, loss: 0.5841049551963806 2023-01-23 07:19:49.339080: step: 2646/1541, loss: 0.6605613827705383 2023-01-23 07:19:50.253062: step: 2648/1541, loss: 1.620143175125122 2023-01-23 07:19:51.163443: step: 2650/1541, loss: 0.4145105481147766 2023-01-23 07:19:52.052902: step: 2652/1541, loss: 1.1480250358581543 2023-01-23 07:19:52.923102: step: 2654/1541, loss: 0.40950649976730347 2023-01-23 07:19:53.806538: step: 2656/1541, loss: 2.3688266277313232 2023-01-23 07:19:54.688689: step: 2658/1541, loss: 0.7358992695808411 2023-01-23 07:19:55.585872: step: 2660/1541, loss: 0.6321792602539062 2023-01-23 07:19:56.468915: step: 2662/1541, loss: 0.35827890038490295 2023-01-23 07:19:57.349203: step: 2664/1541, loss: 0.3360077142715454 2023-01-23 07:19:58.314143: step: 2666/1541, loss: 0.8043849468231201 2023-01-23 07:19:59.239113: step: 2668/1541, loss: 3.4184725284576416 2023-01-23 07:20:00.205096: step: 2670/1541, loss: 0.47818121314048767 2023-01-23 07:20:01.071224: step: 2672/1541, loss: 0.43150782585144043 2023-01-23 07:20:01.953900: step: 2674/1541, loss: 0.5141602158546448 2023-01-23 07:20:02.822879: step: 2676/1541, loss: 0.7855662107467651 2023-01-23 07:20:03.726731: step: 2678/1541, loss: 0.4847332239151001 2023-01-23 07:20:04.631223: step: 2680/1541, loss: 0.6024501919746399 2023-01-23 07:20:05.478847: step: 2682/1541, loss: 0.6095185875892639 2023-01-23 07:20:06.335060: step: 2684/1541, loss: 2.6643924713134766 2023-01-23 07:20:07.202634: step: 2686/1541, loss: 0.7706067562103271 2023-01-23 07:20:08.088065: step: 2688/1541, loss: 0.5239839553833008 2023-01-23 07:20:08.977546: step: 2690/1541, loss: 0.7318999171257019 2023-01-23 07:20:09.852892: step: 2692/1541, loss: 0.5999041795730591 2023-01-23 07:20:10.752612: step: 2694/1541, loss: 0.5271798372268677 2023-01-23 07:20:11.645486: step: 2696/1541, loss: 0.7936012744903564 2023-01-23 07:20:12.516815: step: 2698/1541, loss: 0.14475570619106293 2023-01-23 07:20:13.390272: step: 2700/1541, loss: 0.6611546874046326 2023-01-23 07:20:14.308813: step: 2702/1541, loss: 1.1677154302597046 2023-01-23 07:20:15.342983: step: 2704/1541, loss: 0.7312822341918945 2023-01-23 07:20:16.212529: step: 2706/1541, loss: 0.5607249736785889 2023-01-23 07:20:17.179316: step: 2708/1541, loss: 0.2867429554462433 2023-01-23 07:20:18.080873: step: 2710/1541, loss: 0.8998821973800659 2023-01-23 07:20:19.028982: step: 2712/1541, loss: 0.34581074118614197 2023-01-23 07:20:19.950187: step: 2714/1541, loss: 0.6619461178779602 2023-01-23 07:20:20.816779: step: 2716/1541, loss: 0.46105802059173584 2023-01-23 07:20:21.701342: step: 2718/1541, loss: 0.637837290763855 2023-01-23 07:20:22.578846: step: 2720/1541, loss: 0.448752760887146 2023-01-23 07:20:23.473651: step: 2722/1541, loss: 0.5644210577011108 2023-01-23 07:20:24.395831: step: 2724/1541, loss: 0.692566454410553 2023-01-23 07:20:25.326651: step: 2726/1541, loss: 0.6320770382881165 2023-01-23 07:20:26.257042: step: 2728/1541, loss: 0.4813035726547241 2023-01-23 07:20:27.223891: step: 2730/1541, loss: 0.20865629613399506 2023-01-23 07:20:28.130671: step: 2732/1541, loss: 0.5526893734931946 2023-01-23 07:20:29.004471: step: 2734/1541, loss: 0.9769406318664551 2023-01-23 07:20:29.870725: step: 2736/1541, loss: 0.5156936645507812 2023-01-23 07:20:30.727597: step: 2738/1541, loss: 0.6387303471565247 2023-01-23 07:20:31.604808: step: 2740/1541, loss: 1.1331111192703247 2023-01-23 07:20:32.529507: step: 2742/1541, loss: 1.2576611042022705 2023-01-23 07:20:33.444351: step: 2744/1541, loss: 0.23936107754707336 2023-01-23 07:20:34.338575: step: 2746/1541, loss: 0.4797271192073822 2023-01-23 07:20:35.202667: step: 2748/1541, loss: 1.6017906665802002 2023-01-23 07:20:36.081808: step: 2750/1541, loss: 0.5504989624023438 2023-01-23 07:20:36.960618: step: 2752/1541, loss: 0.6104388236999512 2023-01-23 07:20:37.855887: step: 2754/1541, loss: 0.5625404119491577 2023-01-23 07:20:38.763571: step: 2756/1541, loss: 0.5933113098144531 2023-01-23 07:20:39.639509: step: 2758/1541, loss: 1.5667510032653809 2023-01-23 07:20:40.586118: step: 2760/1541, loss: 0.45760872960090637 2023-01-23 07:20:41.502748: step: 2762/1541, loss: 3.717636823654175 2023-01-23 07:20:42.410511: step: 2764/1541, loss: 0.5357465147972107 2023-01-23 07:20:43.326261: step: 2766/1541, loss: 0.43814176321029663 2023-01-23 07:20:44.223876: step: 2768/1541, loss: 3.0401806831359863 2023-01-23 07:20:45.128494: step: 2770/1541, loss: 1.003105878829956 2023-01-23 07:20:46.014803: step: 2772/1541, loss: 0.34919607639312744 2023-01-23 07:20:46.914150: step: 2774/1541, loss: 0.13687683641910553 2023-01-23 07:20:47.857263: step: 2776/1541, loss: 0.7906961441040039 2023-01-23 07:20:48.742129: step: 2778/1541, loss: 1.7188154458999634 2023-01-23 07:20:49.698146: step: 2780/1541, loss: 0.321750283241272 2023-01-23 07:20:50.626751: step: 2782/1541, loss: 0.7941024303436279 2023-01-23 07:20:51.592898: step: 2784/1541, loss: 0.929248571395874 2023-01-23 07:20:52.539748: step: 2786/1541, loss: 0.5189612507820129 2023-01-23 07:20:53.466928: step: 2788/1541, loss: 1.229341745376587 2023-01-23 07:20:54.394866: step: 2790/1541, loss: 0.12112805247306824 2023-01-23 07:20:55.280910: step: 2792/1541, loss: 0.48677244782447815 2023-01-23 07:20:56.158934: step: 2794/1541, loss: 0.4520006775856018 2023-01-23 07:20:57.104869: step: 2796/1541, loss: 0.86112380027771 2023-01-23 07:20:58.031684: step: 2798/1541, loss: 0.45364299416542053 2023-01-23 07:20:58.937979: step: 2800/1541, loss: 1.128076195716858 2023-01-23 07:20:59.788741: step: 2802/1541, loss: 0.3218053877353668 2023-01-23 07:21:00.645871: step: 2804/1541, loss: 1.196285367012024 2023-01-23 07:21:01.527873: step: 2806/1541, loss: 0.5432888269424438 2023-01-23 07:21:02.454028: step: 2808/1541, loss: 0.2938864231109619 2023-01-23 07:21:03.353114: step: 2810/1541, loss: 0.2924173176288605 2023-01-23 07:21:04.206525: step: 2812/1541, loss: 0.42567765712738037 2023-01-23 07:21:05.123982: step: 2814/1541, loss: 1.191068172454834 2023-01-23 07:21:06.001054: step: 2816/1541, loss: 0.7439656853675842 2023-01-23 07:21:06.925589: step: 2818/1541, loss: 0.5187950134277344 2023-01-23 07:21:07.854146: step: 2820/1541, loss: 0.8760961890220642 2023-01-23 07:21:08.804520: step: 2822/1541, loss: 0.25197410583496094 2023-01-23 07:21:09.700593: step: 2824/1541, loss: 0.554099440574646 2023-01-23 07:21:10.605319: step: 2826/1541, loss: 0.8334951996803284 2023-01-23 07:21:11.524969: step: 2828/1541, loss: 0.7166549563407898 2023-01-23 07:21:12.406132: step: 2830/1541, loss: 0.8019245862960815 2023-01-23 07:21:13.327711: step: 2832/1541, loss: 0.8612616062164307 2023-01-23 07:21:14.215743: step: 2834/1541, loss: 1.186253547668457 2023-01-23 07:21:15.233531: step: 2836/1541, loss: 0.6003003120422363 2023-01-23 07:21:16.156493: step: 2838/1541, loss: 0.36336132884025574 2023-01-23 07:21:17.036909: step: 2840/1541, loss: 1.1144295930862427 2023-01-23 07:21:17.950875: step: 2842/1541, loss: 1.7977415323257446 2023-01-23 07:21:18.842489: step: 2844/1541, loss: 1.5731899738311768 2023-01-23 07:21:19.721252: step: 2846/1541, loss: 0.39072054624557495 2023-01-23 07:21:20.641972: step: 2848/1541, loss: 0.34857720136642456 2023-01-23 07:21:21.543110: step: 2850/1541, loss: 3.050723075866699 2023-01-23 07:21:22.469264: step: 2852/1541, loss: 0.8455158472061157 2023-01-23 07:21:23.382378: step: 2854/1541, loss: 0.6067084670066833 2023-01-23 07:21:24.293904: step: 2856/1541, loss: 2.990532875061035 2023-01-23 07:21:25.223150: step: 2858/1541, loss: 0.5567166805267334 2023-01-23 07:21:26.109652: step: 2860/1541, loss: 0.5212400555610657 2023-01-23 07:21:26.981229: step: 2862/1541, loss: 0.601778507232666 2023-01-23 07:21:27.871475: step: 2864/1541, loss: 0.7560119032859802 2023-01-23 07:21:28.776961: step: 2866/1541, loss: 0.6445177793502808 2023-01-23 07:21:29.662885: step: 2868/1541, loss: 0.6859535574913025 2023-01-23 07:21:30.558452: step: 2870/1541, loss: 0.5815810561180115 2023-01-23 07:21:31.458105: step: 2872/1541, loss: 0.9703372716903687 2023-01-23 07:21:32.373645: step: 2874/1541, loss: 0.29194480180740356 2023-01-23 07:21:33.240930: step: 2876/1541, loss: 0.6611154079437256 2023-01-23 07:21:34.164803: step: 2878/1541, loss: 0.4071407616138458 2023-01-23 07:21:35.079115: step: 2880/1541, loss: 0.48312902450561523 2023-01-23 07:21:35.977348: step: 2882/1541, loss: 7.285450458526611 2023-01-23 07:21:36.922084: step: 2884/1541, loss: 0.3932226300239563 2023-01-23 07:21:37.850666: step: 2886/1541, loss: 0.9785760045051575 2023-01-23 07:21:38.743218: step: 2888/1541, loss: 1.0493805408477783 2023-01-23 07:21:39.680757: step: 2890/1541, loss: 0.6901190876960754 2023-01-23 07:21:40.561369: step: 2892/1541, loss: 1.0365842580795288 2023-01-23 07:21:41.474678: step: 2894/1541, loss: 0.5619301795959473 2023-01-23 07:21:42.365037: step: 2896/1541, loss: 0.5293359756469727 2023-01-23 07:21:43.221602: step: 2898/1541, loss: 0.7648407220840454 2023-01-23 07:21:44.121418: step: 2900/1541, loss: 0.7207226157188416 2023-01-23 07:21:45.006368: step: 2902/1541, loss: 0.46028727293014526 2023-01-23 07:21:45.928449: step: 2904/1541, loss: 0.6867094039916992 2023-01-23 07:21:46.812541: step: 2906/1541, loss: 0.48052623867988586 2023-01-23 07:21:47.696779: step: 2908/1541, loss: 0.42539525032043457 2023-01-23 07:21:48.606377: step: 2910/1541, loss: 1.5610014200210571 2023-01-23 07:21:49.465777: step: 2912/1541, loss: 1.0206916332244873 2023-01-23 07:21:50.386175: step: 2914/1541, loss: 0.16045910120010376 2023-01-23 07:21:51.286054: step: 2916/1541, loss: 0.8095287680625916 2023-01-23 07:21:52.244570: step: 2918/1541, loss: 1.594890832901001 2023-01-23 07:21:53.125508: step: 2920/1541, loss: 0.323334276676178 2023-01-23 07:21:54.040320: step: 2922/1541, loss: 0.6151981949806213 2023-01-23 07:21:54.916673: step: 2924/1541, loss: 0.8343360424041748 2023-01-23 07:21:55.860087: step: 2926/1541, loss: 1.1129964590072632 2023-01-23 07:21:56.779714: step: 2928/1541, loss: 0.9287270307540894 2023-01-23 07:21:57.658568: step: 2930/1541, loss: 5.969832420349121 2023-01-23 07:21:58.509043: step: 2932/1541, loss: 0.4852042496204376 2023-01-23 07:21:59.387642: step: 2934/1541, loss: 0.5911339521408081 2023-01-23 07:22:00.296779: step: 2936/1541, loss: 0.4138118624687195 2023-01-23 07:22:01.220833: step: 2938/1541, loss: 0.550220251083374 2023-01-23 07:22:02.150525: step: 2940/1541, loss: 0.5657113790512085 2023-01-23 07:22:03.051928: step: 2942/1541, loss: 1.0834767818450928 2023-01-23 07:22:03.969641: step: 2944/1541, loss: 0.6172381639480591 2023-01-23 07:22:04.872426: step: 2946/1541, loss: 0.8271017670631409 2023-01-23 07:22:05.848735: step: 2948/1541, loss: 0.554015040397644 2023-01-23 07:22:06.783758: step: 2950/1541, loss: 0.8456590175628662 2023-01-23 07:22:07.715072: step: 2952/1541, loss: 3.390636920928955 2023-01-23 07:22:08.594458: step: 2954/1541, loss: 0.49129268527030945 2023-01-23 07:22:09.472559: step: 2956/1541, loss: 0.1143307238817215 2023-01-23 07:22:10.341516: step: 2958/1541, loss: 0.839060366153717 2023-01-23 07:22:11.251017: step: 2960/1541, loss: 0.42945778369903564 2023-01-23 07:22:12.136018: step: 2962/1541, loss: 0.8026401996612549 2023-01-23 07:22:13.036091: step: 2964/1541, loss: 0.734004557132721 2023-01-23 07:22:13.946794: step: 2966/1541, loss: 0.40136203169822693 2023-01-23 07:22:14.823948: step: 2968/1541, loss: 2.3334197998046875 2023-01-23 07:22:15.787000: step: 2970/1541, loss: 2.2736337184906006 2023-01-23 07:22:16.719331: step: 2972/1541, loss: 2.105029582977295 2023-01-23 07:22:17.655682: step: 2974/1541, loss: 0.860198438167572 2023-01-23 07:22:18.520993: step: 2976/1541, loss: 0.37803056836128235 2023-01-23 07:22:19.476276: step: 2978/1541, loss: 1.3462575674057007 2023-01-23 07:22:20.390255: step: 2980/1541, loss: 0.5611346960067749 2023-01-23 07:22:21.310321: step: 2982/1541, loss: 0.5140706896781921 2023-01-23 07:22:22.215607: step: 2984/1541, loss: 0.3714025914669037 2023-01-23 07:22:23.086497: step: 2986/1541, loss: 1.0126116275787354 2023-01-23 07:22:24.005663: step: 2988/1541, loss: 0.39184901118278503 2023-01-23 07:22:24.940285: step: 2990/1541, loss: 0.3448980152606964 2023-01-23 07:22:25.809684: step: 2992/1541, loss: 1.1019608974456787 2023-01-23 07:22:26.701282: step: 2994/1541, loss: 0.3279367685317993 2023-01-23 07:22:27.589047: step: 2996/1541, loss: 0.940751314163208 2023-01-23 07:22:28.431853: step: 2998/1541, loss: 0.7180474400520325 2023-01-23 07:22:29.371086: step: 3000/1541, loss: 1.098190188407898 2023-01-23 07:22:30.248073: step: 3002/1541, loss: 0.4294660985469818 2023-01-23 07:22:31.169958: step: 3004/1541, loss: 0.738852858543396 2023-01-23 07:22:32.135083: step: 3006/1541, loss: 1.4313899278640747 2023-01-23 07:22:33.007541: step: 3008/1541, loss: 1.1039172410964966 2023-01-23 07:22:33.967873: step: 3010/1541, loss: 0.5292916297912598 2023-01-23 07:22:34.879410: step: 3012/1541, loss: 1.0139203071594238 2023-01-23 07:22:35.781178: step: 3014/1541, loss: 0.9426945447921753 2023-01-23 07:22:36.665108: step: 3016/1541, loss: 0.600770890712738 2023-01-23 07:22:37.600615: step: 3018/1541, loss: 0.7186826467514038 2023-01-23 07:22:38.502584: step: 3020/1541, loss: 0.5201526880264282 2023-01-23 07:22:39.429336: step: 3022/1541, loss: 0.519402801990509 2023-01-23 07:22:40.317619: step: 3024/1541, loss: 0.8061184883117676 2023-01-23 07:22:41.240959: step: 3026/1541, loss: 0.7103779911994934 2023-01-23 07:22:42.104867: step: 3028/1541, loss: 0.502190351486206 2023-01-23 07:22:43.019623: step: 3030/1541, loss: 0.5261127352714539 2023-01-23 07:22:43.957526: step: 3032/1541, loss: 0.46657246351242065 2023-01-23 07:22:44.863639: step: 3034/1541, loss: 0.6639288067817688 2023-01-23 07:22:45.751924: step: 3036/1541, loss: 1.0732192993164062 2023-01-23 07:22:46.661141: step: 3038/1541, loss: 0.42965584993362427 2023-01-23 07:22:47.557571: step: 3040/1541, loss: 0.5735238194465637 2023-01-23 07:22:48.436090: step: 3042/1541, loss: 0.7766016721725464 2023-01-23 07:22:49.325011: step: 3044/1541, loss: 0.45504066348075867 2023-01-23 07:22:50.227705: step: 3046/1541, loss: 1.6906509399414062 2023-01-23 07:22:51.132348: step: 3048/1541, loss: 1.1933882236480713 2023-01-23 07:22:52.013080: step: 3050/1541, loss: 0.25626271963119507 2023-01-23 07:22:52.915791: step: 3052/1541, loss: 1.1568474769592285 2023-01-23 07:22:53.789610: step: 3054/1541, loss: 0.6077896356582642 2023-01-23 07:22:54.658696: step: 3056/1541, loss: 0.4112831950187683 2023-01-23 07:22:55.568800: step: 3058/1541, loss: 0.4415893256664276 2023-01-23 07:22:56.479486: step: 3060/1541, loss: 0.4842418134212494 2023-01-23 07:22:57.410922: step: 3062/1541, loss: 0.6525493264198303 2023-01-23 07:22:58.281014: step: 3064/1541, loss: 0.5743505954742432 2023-01-23 07:22:59.145551: step: 3066/1541, loss: 0.23948977887630463 2023-01-23 07:23:00.029200: step: 3068/1541, loss: 2.889634609222412 2023-01-23 07:23:00.969034: step: 3070/1541, loss: 0.5499588251113892 2023-01-23 07:23:01.878237: step: 3072/1541, loss: 0.5783229470252991 2023-01-23 07:23:02.757893: step: 3074/1541, loss: 0.49486786127090454 2023-01-23 07:23:03.615669: step: 3076/1541, loss: 3.3617122173309326 2023-01-23 07:23:04.488058: step: 3078/1541, loss: 0.35378456115722656 2023-01-23 07:23:05.377891: step: 3080/1541, loss: 0.647625207901001 2023-01-23 07:23:06.249651: step: 3082/1541, loss: 0.7972532510757446 2023-01-23 07:23:06.887143: step: 3084/1541, loss: 0.5334597229957581 ================================================== Loss: 0.848 -------------------- Dev Chinese: {'event': {'p': 0.6344086021505376, 'r': 0.7070572569906791, 'f1': 0.6687657430730478}, 'argument': {'p': 0.6514595708798564, 'r': 0.6327452819648294, 'f1': 0.6419660678776685}, 'combined': 0.42932491441189163, 'epoch': 4} Test Chinese: {'event': {'p': 0.6140904311251314, 'r': 0.6998202516476932, 'f1': 0.6541584990198823}, 'argument': {'p': 0.694574043100388, 'r': 0.6872361023900377, 'f1': 0.6908855891084613}, 'combined': 0.45194867996565824, 'epoch': 4} Dev Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Dev Russian: {'event': {'p': 0.6575829383886256, 'r': 0.7390146471371505, 'f1': 0.695924764890282}, 'argument': {'p': 0.6561752698674627, 'r': 0.6438411482534127, 'f1': 0.6499496980470882}, 'combined': 0.45231609080392965, 'epoch': 4} Test Russian: {'event': {'p': 0.6530612244897959, 'r': 0.7477531455961653, 'f1': 0.6972067039106145}, 'argument': {'p': 0.6933810842434333, 'r': 0.694128262135937, 'f1': 0.6937544720109311}, 'combined': 0.4836902687539899, 'epoch': 4} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6281347962382445, 'r': 0.6448109412711183, 'f1': 0.6363636363636364}, 'combined': 0.4786324786324786, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5928679962013296, 'r': 0.6204432518386007, 'f1': 0.6063422688422688}, 'combined': 0.31183316683316686, 'epoch': 4} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6344086021505376, 'r': 0.7070572569906791, 'f1': 0.6687657430730478}, 'argument': {'p': 0.6514595708798564, 'r': 0.6327452819648294, 'f1': 0.6419660678776685}, 'combined': 0.42932491441189163, 'epoch': 4} Test for Chinese: {'event': {'p': 0.6140904311251314, 'r': 0.6998202516476932, 'f1': 0.6541584990198823}, 'argument': {'p': 0.694574043100388, 'r': 0.6872361023900377, 'f1': 0.6908855891084613}, 'combined': 0.45194867996565824, 'epoch': 4} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6281347962382445, 'r': 0.6448109412711183, 'f1': 0.6363636363636364}, 'combined': 0.4786324786324786, 'epoch': 4} -------------------- Dev for Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test for Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} -------------------- Dev for Russian: {'event': {'p': 0.6575829383886256, 'r': 0.7390146471371505, 'f1': 0.695924764890282}, 'argument': {'p': 0.6561752698674627, 'r': 0.6438411482534127, 'f1': 0.6499496980470882}, 'combined': 0.45231609080392965, 'epoch': 4} Test for Russian: {'event': {'p': 0.6530612244897959, 'r': 0.7477531455961653, 'f1': 0.6972067039106145}, 'argument': {'p': 0.6933810842434333, 'r': 0.694128262135937, 'f1': 0.6937544720109311}, 'combined': 0.4836902687539899, 'epoch': 4} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5928679962013296, 'r': 0.6204432518386007, 'f1': 0.6063422688422688}, 'combined': 0.31183316683316686, 'epoch': 4} ****************************** Epoch: 5 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 07:28:53.562883: step: 2/1541, loss: 3.340786933898926 2023-01-23 07:28:54.464738: step: 4/1541, loss: 0.18974259495735168 2023-01-23 07:28:55.311707: step: 6/1541, loss: 0.4740162193775177 2023-01-23 07:28:56.258884: step: 8/1541, loss: 0.9768191576004028 2023-01-23 07:28:57.215167: step: 10/1541, loss: 0.3041735887527466 2023-01-23 07:28:58.154307: step: 12/1541, loss: 0.543038010597229 2023-01-23 07:28:59.057467: step: 14/1541, loss: 0.5015403032302856 2023-01-23 07:28:59.958078: step: 16/1541, loss: 0.23296962678432465 2023-01-23 07:29:00.957008: step: 18/1541, loss: 0.58981853723526 2023-01-23 07:29:01.880709: step: 20/1541, loss: 0.4342830777168274 2023-01-23 07:29:02.771691: step: 22/1541, loss: 0.9760361909866333 2023-01-23 07:29:03.649160: step: 24/1541, loss: 0.26460081338882446 2023-01-23 07:29:04.577266: step: 26/1541, loss: 0.3089454174041748 2023-01-23 07:29:05.528538: step: 28/1541, loss: 0.5956353545188904 2023-01-23 07:29:06.430594: step: 30/1541, loss: 0.5855239629745483 2023-01-23 07:29:07.370725: step: 32/1541, loss: 0.28766006231307983 2023-01-23 07:29:08.238107: step: 34/1541, loss: 0.18036086857318878 2023-01-23 07:29:09.128198: step: 36/1541, loss: 2.8494300842285156 2023-01-23 07:29:10.068456: step: 38/1541, loss: 0.360843300819397 2023-01-23 07:29:10.992345: step: 40/1541, loss: 0.7617335915565491 2023-01-23 07:29:11.914273: step: 42/1541, loss: 0.4497760236263275 2023-01-23 07:29:12.830831: step: 44/1541, loss: 0.3363882303237915 2023-01-23 07:29:13.677221: step: 46/1541, loss: 0.2600462734699249 2023-01-23 07:29:14.570421: step: 48/1541, loss: 0.2229522317647934 2023-01-23 07:29:15.483719: step: 50/1541, loss: 0.29231083393096924 2023-01-23 07:29:16.376850: step: 52/1541, loss: 0.3618854582309723 2023-01-23 07:29:17.262385: step: 54/1541, loss: 0.24935217201709747 2023-01-23 07:29:18.244153: step: 56/1541, loss: 0.3914056420326233 2023-01-23 07:29:19.114811: step: 58/1541, loss: 0.9179509282112122 2023-01-23 07:29:20.029438: step: 60/1541, loss: 0.6110088229179382 2023-01-23 07:29:20.898753: step: 62/1541, loss: 0.25649285316467285 2023-01-23 07:29:21.780869: step: 64/1541, loss: 0.3216625154018402 2023-01-23 07:29:22.704080: step: 66/1541, loss: 0.2920592427253723 2023-01-23 07:29:23.570134: step: 68/1541, loss: 0.8416105508804321 2023-01-23 07:29:24.480806: step: 70/1541, loss: 0.30252954363822937 2023-01-23 07:29:25.387651: step: 72/1541, loss: 0.35192325711250305 2023-01-23 07:29:26.277951: step: 74/1541, loss: 0.4579179286956787 2023-01-23 07:29:27.150836: step: 76/1541, loss: 0.3737460672855377 2023-01-23 07:29:28.056955: step: 78/1541, loss: 0.23161157965660095 2023-01-23 07:29:28.918360: step: 80/1541, loss: 0.3845050036907196 2023-01-23 07:29:29.817023: step: 82/1541, loss: 0.6874634027481079 2023-01-23 07:29:30.774122: step: 84/1541, loss: 0.835837721824646 2023-01-23 07:29:31.755710: step: 86/1541, loss: 0.3452907204627991 2023-01-23 07:29:32.649895: step: 88/1541, loss: 0.6959424614906311 2023-01-23 07:29:33.525238: step: 90/1541, loss: 0.3063378632068634 2023-01-23 07:29:34.408034: step: 92/1541, loss: 0.29835161566734314 2023-01-23 07:29:35.258892: step: 94/1541, loss: 0.42356041073799133 2023-01-23 07:29:36.186613: step: 96/1541, loss: 5.405776500701904 2023-01-23 07:29:37.087302: step: 98/1541, loss: 0.9592600464820862 2023-01-23 07:29:37.985782: step: 100/1541, loss: 0.2970738410949707 2023-01-23 07:29:38.937239: step: 102/1541, loss: 0.5868827700614929 2023-01-23 07:29:39.811040: step: 104/1541, loss: 0.6285496950149536 2023-01-23 07:29:40.700751: step: 106/1541, loss: 0.1349743902683258 2023-01-23 07:29:41.679108: step: 108/1541, loss: 0.7402307391166687 2023-01-23 07:29:42.551832: step: 110/1541, loss: 0.3237200081348419 2023-01-23 07:29:43.565502: step: 112/1541, loss: 0.4505916237831116 2023-01-23 07:29:44.489701: step: 114/1541, loss: 0.549311101436615 2023-01-23 07:29:45.376263: step: 116/1541, loss: 0.5162381529808044 2023-01-23 07:29:46.277844: step: 118/1541, loss: 0.6139713525772095 2023-01-23 07:29:47.192127: step: 120/1541, loss: 3.2704286575317383 2023-01-23 07:29:48.075658: step: 122/1541, loss: 0.19298984110355377 2023-01-23 07:29:48.981422: step: 124/1541, loss: 0.3660401701927185 2023-01-23 07:29:49.872216: step: 126/1541, loss: 0.47584429383277893 2023-01-23 07:29:50.764779: step: 128/1541, loss: 0.4319215714931488 2023-01-23 07:29:51.634509: step: 130/1541, loss: 1.787013292312622 2023-01-23 07:29:52.498299: step: 132/1541, loss: 0.5435146689414978 2023-01-23 07:29:53.377845: step: 134/1541, loss: 0.48304420709609985 2023-01-23 07:29:54.267977: step: 136/1541, loss: 0.6499898433685303 2023-01-23 07:29:55.179533: step: 138/1541, loss: 0.250311940908432 2023-01-23 07:29:56.028995: step: 140/1541, loss: 0.3874608278274536 2023-01-23 07:29:56.895684: step: 142/1541, loss: 0.9633357524871826 2023-01-23 07:29:57.805703: step: 144/1541, loss: 0.5565085411071777 2023-01-23 07:29:58.679471: step: 146/1541, loss: 0.5075632929801941 2023-01-23 07:29:59.542869: step: 148/1541, loss: 0.2730993628501892 2023-01-23 07:30:00.440990: step: 150/1541, loss: 0.7580682039260864 2023-01-23 07:30:01.336845: step: 152/1541, loss: 1.0070931911468506 2023-01-23 07:30:02.256649: step: 154/1541, loss: 0.2726517617702484 2023-01-23 07:30:03.148911: step: 156/1541, loss: 0.542694091796875 2023-01-23 07:30:04.081026: step: 158/1541, loss: 0.8920679688453674 2023-01-23 07:30:04.954533: step: 160/1541, loss: 0.3868750333786011 2023-01-23 07:30:05.885251: step: 162/1541, loss: 0.5752679109573364 2023-01-23 07:30:06.779250: step: 164/1541, loss: 0.3969084322452545 2023-01-23 07:30:07.642844: step: 166/1541, loss: 0.22421382367610931 2023-01-23 07:30:08.549905: step: 168/1541, loss: 0.5533144474029541 2023-01-23 07:30:09.497875: step: 170/1541, loss: 0.7438063025474548 2023-01-23 07:30:10.381736: step: 172/1541, loss: 0.5885154008865356 2023-01-23 07:30:11.359874: step: 174/1541, loss: 0.46732425689697266 2023-01-23 07:30:12.296357: step: 176/1541, loss: 0.3754041790962219 2023-01-23 07:30:13.189261: step: 178/1541, loss: 0.1850953996181488 2023-01-23 07:30:14.120669: step: 180/1541, loss: 0.456657350063324 2023-01-23 07:30:15.009424: step: 182/1541, loss: 0.31054338812828064 2023-01-23 07:30:15.967679: step: 184/1541, loss: 0.2742825150489807 2023-01-23 07:30:16.814723: step: 186/1541, loss: 0.963435173034668 2023-01-23 07:30:17.762255: step: 188/1541, loss: 0.5109720826148987 2023-01-23 07:30:18.658595: step: 190/1541, loss: 0.8853522539138794 2023-01-23 07:30:19.535290: step: 192/1541, loss: 1.003728985786438 2023-01-23 07:30:20.443633: step: 194/1541, loss: 0.39271965622901917 2023-01-23 07:30:21.309005: step: 196/1541, loss: 0.21669727563858032 2023-01-23 07:30:22.176503: step: 198/1541, loss: 0.7230046987533569 2023-01-23 07:30:23.072902: step: 200/1541, loss: 0.3405720889568329 2023-01-23 07:30:23.972922: step: 202/1541, loss: 6.708118438720703 2023-01-23 07:30:24.903781: step: 204/1541, loss: 1.031044840812683 2023-01-23 07:30:25.784769: step: 206/1541, loss: 0.19585196673870087 2023-01-23 07:30:26.677176: step: 208/1541, loss: 0.4834668040275574 2023-01-23 07:30:27.679708: step: 210/1541, loss: 0.36991608142852783 2023-01-23 07:30:28.618610: step: 212/1541, loss: 0.3592241108417511 2023-01-23 07:30:29.487593: step: 214/1541, loss: 0.3711354434490204 2023-01-23 07:30:30.359307: step: 216/1541, loss: 1.163939118385315 2023-01-23 07:30:31.202603: step: 218/1541, loss: 0.7048569917678833 2023-01-23 07:30:32.116004: step: 220/1541, loss: 0.854511022567749 2023-01-23 07:30:32.987417: step: 222/1541, loss: 0.3896898329257965 2023-01-23 07:30:33.854774: step: 224/1541, loss: 0.5091009140014648 2023-01-23 07:30:34.707477: step: 226/1541, loss: 0.37220290303230286 2023-01-23 07:30:35.612787: step: 228/1541, loss: 0.3810894787311554 2023-01-23 07:30:36.547310: step: 230/1541, loss: 0.5748885869979858 2023-01-23 07:30:37.479718: step: 232/1541, loss: 0.47703030705451965 2023-01-23 07:30:38.353849: step: 234/1541, loss: 1.3495638370513916 2023-01-23 07:30:39.236248: step: 236/1541, loss: 0.5613813400268555 2023-01-23 07:30:40.109847: step: 238/1541, loss: 1.1554926633834839 2023-01-23 07:30:41.034542: step: 240/1541, loss: 0.7105746865272522 2023-01-23 07:30:41.934512: step: 242/1541, loss: 0.36991220712661743 2023-01-23 07:30:42.787934: step: 244/1541, loss: 0.6586021184921265 2023-01-23 07:30:43.754160: step: 246/1541, loss: 0.39255276322364807 2023-01-23 07:30:44.627477: step: 248/1541, loss: 0.17064975202083588 2023-01-23 07:30:45.542360: step: 250/1541, loss: 0.5080112814903259 2023-01-23 07:30:46.439878: step: 252/1541, loss: 0.34109199047088623 2023-01-23 07:30:47.304908: step: 254/1541, loss: 0.5904426574707031 2023-01-23 07:30:48.241507: step: 256/1541, loss: 0.27383458614349365 2023-01-23 07:30:49.134793: step: 258/1541, loss: 0.4321245849132538 2023-01-23 07:30:50.019635: step: 260/1541, loss: 0.09973005205392838 2023-01-23 07:30:51.042090: step: 262/1541, loss: 1.1730554103851318 2023-01-23 07:30:51.958193: step: 264/1541, loss: 0.4840889871120453 2023-01-23 07:30:52.874388: step: 266/1541, loss: 0.3996778130531311 2023-01-23 07:30:53.747302: step: 268/1541, loss: 1.0177876949310303 2023-01-23 07:30:54.634792: step: 270/1541, loss: 1.2633534669876099 2023-01-23 07:30:55.514048: step: 272/1541, loss: 0.6725727319717407 2023-01-23 07:30:56.456489: step: 274/1541, loss: 0.38797956705093384 2023-01-23 07:30:57.394990: step: 276/1541, loss: 1.5680862665176392 2023-01-23 07:30:58.281848: step: 278/1541, loss: 0.7277836203575134 2023-01-23 07:30:59.169070: step: 280/1541, loss: 0.6084587574005127 2023-01-23 07:31:00.054363: step: 282/1541, loss: 0.7521091103553772 2023-01-23 07:31:00.900343: step: 284/1541, loss: 0.3429635763168335 2023-01-23 07:31:01.871814: step: 286/1541, loss: 1.5702524185180664 2023-01-23 07:31:02.742060: step: 288/1541, loss: 0.6026183366775513 2023-01-23 07:31:03.663096: step: 290/1541, loss: 0.43015119433403015 2023-01-23 07:31:04.587401: step: 292/1541, loss: 0.9968951940536499 2023-01-23 07:31:05.506583: step: 294/1541, loss: 0.43939411640167236 2023-01-23 07:31:06.391686: step: 296/1541, loss: 0.10011263936758041 2023-01-23 07:31:07.329531: step: 298/1541, loss: 0.39266708493232727 2023-01-23 07:31:08.238009: step: 300/1541, loss: 0.26594725251197815 2023-01-23 07:31:09.100095: step: 302/1541, loss: 0.23221789300441742 2023-01-23 07:31:10.068887: step: 304/1541, loss: 0.3143714368343353 2023-01-23 07:31:10.969322: step: 306/1541, loss: 0.19470638036727905 2023-01-23 07:31:11.938977: step: 308/1541, loss: 0.8941380381584167 2023-01-23 07:31:12.841718: step: 310/1541, loss: 0.329299658536911 2023-01-23 07:31:13.777576: step: 312/1541, loss: 1.1554244756698608 2023-01-23 07:31:14.678002: step: 314/1541, loss: 0.5263737440109253 2023-01-23 07:31:15.550428: step: 316/1541, loss: 1.769208312034607 2023-01-23 07:31:16.459659: step: 318/1541, loss: 0.7324662804603577 2023-01-23 07:31:17.361665: step: 320/1541, loss: 0.23996306955814362 2023-01-23 07:31:18.282649: step: 322/1541, loss: 0.2995913028717041 2023-01-23 07:31:19.196043: step: 324/1541, loss: 0.3320801556110382 2023-01-23 07:31:20.066406: step: 326/1541, loss: 0.2108972668647766 2023-01-23 07:31:20.998820: step: 328/1541, loss: 0.32755741477012634 2023-01-23 07:31:21.886753: step: 330/1541, loss: 0.18323586881160736 2023-01-23 07:31:22.761388: step: 332/1541, loss: 0.35416868329048157 2023-01-23 07:31:23.670730: step: 334/1541, loss: 0.8011043667793274 2023-01-23 07:31:24.553238: step: 336/1541, loss: 0.15974347293376923 2023-01-23 07:31:25.413160: step: 338/1541, loss: 0.37025344371795654 2023-01-23 07:31:26.286971: step: 340/1541, loss: 0.3593824505805969 2023-01-23 07:31:27.166644: step: 342/1541, loss: 0.8434457778930664 2023-01-23 07:31:28.060109: step: 344/1541, loss: 0.7477440237998962 2023-01-23 07:31:28.949230: step: 346/1541, loss: 0.4183737337589264 2023-01-23 07:31:29.831604: step: 348/1541, loss: 0.49020594358444214 2023-01-23 07:31:30.772577: step: 350/1541, loss: 0.5506107807159424 2023-01-23 07:31:31.692692: step: 352/1541, loss: 0.4139193892478943 2023-01-23 07:31:32.674676: step: 354/1541, loss: 0.46524807810783386 2023-01-23 07:31:33.575277: step: 356/1541, loss: 0.4433940052986145 2023-01-23 07:31:34.452843: step: 358/1541, loss: 0.2012275755405426 2023-01-23 07:31:35.366057: step: 360/1541, loss: 0.6655287742614746 2023-01-23 07:31:36.280819: step: 362/1541, loss: 1.1649240255355835 2023-01-23 07:31:37.152023: step: 364/1541, loss: 0.1966528296470642 2023-01-23 07:31:38.081713: step: 366/1541, loss: 0.7436972856521606 2023-01-23 07:31:38.992022: step: 368/1541, loss: 0.3361406624317169 2023-01-23 07:31:39.866687: step: 370/1541, loss: 0.3439837694168091 2023-01-23 07:31:40.778516: step: 372/1541, loss: 0.5149533152580261 2023-01-23 07:31:41.745007: step: 374/1541, loss: 0.36256295442581177 2023-01-23 07:31:42.677286: step: 376/1541, loss: 0.2640995681285858 2023-01-23 07:31:43.600737: step: 378/1541, loss: 0.858375608921051 2023-01-23 07:31:44.445041: step: 380/1541, loss: 0.43845391273498535 2023-01-23 07:31:45.386021: step: 382/1541, loss: 0.22359691560268402 2023-01-23 07:31:46.260353: step: 384/1541, loss: 0.1508568823337555 2023-01-23 07:31:47.143832: step: 386/1541, loss: 0.14099439978599548 2023-01-23 07:31:48.077618: step: 388/1541, loss: 0.40093758702278137 2023-01-23 07:31:49.028124: step: 390/1541, loss: 0.849007785320282 2023-01-23 07:31:49.962433: step: 392/1541, loss: 0.9507867097854614 2023-01-23 07:31:50.907175: step: 394/1541, loss: 0.6385188698768616 2023-01-23 07:31:51.909816: step: 396/1541, loss: 0.4859314560890198 2023-01-23 07:31:52.795565: step: 398/1541, loss: 0.21177546679973602 2023-01-23 07:31:53.664006: step: 400/1541, loss: 0.5492739677429199 2023-01-23 07:31:54.537732: step: 402/1541, loss: 0.4857728183269501 2023-01-23 07:31:55.423938: step: 404/1541, loss: 0.5463724732398987 2023-01-23 07:31:56.361391: step: 406/1541, loss: 0.5547749996185303 2023-01-23 07:31:57.220262: step: 408/1541, loss: 0.45553505420684814 2023-01-23 07:31:58.108184: step: 410/1541, loss: 0.31507211923599243 2023-01-23 07:31:59.020593: step: 412/1541, loss: 0.16394783556461334 2023-01-23 07:31:59.898948: step: 414/1541, loss: 0.47014185786247253 2023-01-23 07:32:00.804574: step: 416/1541, loss: 0.19766241312026978 2023-01-23 07:32:01.811427: step: 418/1541, loss: 1.21162748336792 2023-01-23 07:32:02.701968: step: 420/1541, loss: 0.2125987410545349 2023-01-23 07:32:03.602984: step: 422/1541, loss: 0.5918274521827698 2023-01-23 07:32:04.471687: step: 424/1541, loss: 0.5659220218658447 2023-01-23 07:32:05.356019: step: 426/1541, loss: 3.3345935344696045 2023-01-23 07:32:06.234059: step: 428/1541, loss: 0.8914892077445984 2023-01-23 07:32:07.100585: step: 430/1541, loss: 0.43340760469436646 2023-01-23 07:32:08.001433: step: 432/1541, loss: 0.5456626415252686 2023-01-23 07:32:08.898948: step: 434/1541, loss: 0.8526597619056702 2023-01-23 07:32:09.752565: step: 436/1541, loss: 0.6797639727592468 2023-01-23 07:32:10.669825: step: 438/1541, loss: 0.389166921377182 2023-01-23 07:32:11.558717: step: 440/1541, loss: 0.23092736303806305 2023-01-23 07:32:12.424933: step: 442/1541, loss: 0.3422655761241913 2023-01-23 07:32:13.318608: step: 444/1541, loss: 0.3198194205760956 2023-01-23 07:32:14.221112: step: 446/1541, loss: 1.2809630632400513 2023-01-23 07:32:15.097550: step: 448/1541, loss: 0.23112282156944275 2023-01-23 07:32:15.975227: step: 450/1541, loss: 0.4557863175868988 2023-01-23 07:32:16.862565: step: 452/1541, loss: 0.6060137152671814 2023-01-23 07:32:17.744875: step: 454/1541, loss: 0.36244815587997437 2023-01-23 07:32:18.640839: step: 456/1541, loss: 1.4984952211380005 2023-01-23 07:32:19.556566: step: 458/1541, loss: 0.3738723695278168 2023-01-23 07:32:20.433900: step: 460/1541, loss: 0.1569189578294754 2023-01-23 07:32:21.337229: step: 462/1541, loss: 0.8380958437919617 2023-01-23 07:32:22.317831: step: 464/1541, loss: 0.5738366842269897 2023-01-23 07:32:23.269372: step: 466/1541, loss: 1.04380464553833 2023-01-23 07:32:24.123287: step: 468/1541, loss: 0.4320545494556427 2023-01-23 07:32:25.029555: step: 470/1541, loss: 0.4804771840572357 2023-01-23 07:32:25.908184: step: 472/1541, loss: 0.9600541591644287 2023-01-23 07:32:26.809286: step: 474/1541, loss: 0.7426703572273254 2023-01-23 07:32:27.683643: step: 476/1541, loss: 0.6337661147117615 2023-01-23 07:32:28.581565: step: 478/1541, loss: 0.6608461737632751 2023-01-23 07:32:29.488229: step: 480/1541, loss: 0.6174027323722839 2023-01-23 07:32:30.375096: step: 482/1541, loss: 0.16672086715698242 2023-01-23 07:32:31.268970: step: 484/1541, loss: 1.139941930770874 2023-01-23 07:32:32.200353: step: 486/1541, loss: 0.588100790977478 2023-01-23 07:32:33.093237: step: 488/1541, loss: 1.681828498840332 2023-01-23 07:32:33.984533: step: 490/1541, loss: 2.7927322387695312 2023-01-23 07:32:34.888320: step: 492/1541, loss: 0.5248489379882812 2023-01-23 07:32:35.777618: step: 494/1541, loss: 0.657597005367279 2023-01-23 07:32:36.661625: step: 496/1541, loss: 0.2688380181789398 2023-01-23 07:32:37.537919: step: 498/1541, loss: 2.7171196937561035 2023-01-23 07:32:38.432882: step: 500/1541, loss: 0.2321934849023819 2023-01-23 07:32:39.327708: step: 502/1541, loss: 0.42984455823898315 2023-01-23 07:32:40.206052: step: 504/1541, loss: 0.7538703680038452 2023-01-23 07:32:41.139170: step: 506/1541, loss: 0.5624375939369202 2023-01-23 07:32:42.080030: step: 508/1541, loss: 0.8440147638320923 2023-01-23 07:32:42.996858: step: 510/1541, loss: 0.2517772316932678 2023-01-23 07:32:43.874711: step: 512/1541, loss: 1.3779798746109009 2023-01-23 07:32:44.770221: step: 514/1541, loss: 0.6356848478317261 2023-01-23 07:32:45.666992: step: 516/1541, loss: 0.2545225918292999 2023-01-23 07:32:46.547386: step: 518/1541, loss: 0.36875107884407043 2023-01-23 07:32:47.446361: step: 520/1541, loss: 0.6301188468933105 2023-01-23 07:32:48.325940: step: 522/1541, loss: 0.35737326741218567 2023-01-23 07:32:49.218496: step: 524/1541, loss: 0.44064655900001526 2023-01-23 07:32:50.133001: step: 526/1541, loss: 0.4795300364494324 2023-01-23 07:32:51.051698: step: 528/1541, loss: 0.6056938171386719 2023-01-23 07:32:51.932669: step: 530/1541, loss: 0.24175024032592773 2023-01-23 07:32:52.810321: step: 532/1541, loss: 0.36962297558784485 2023-01-23 07:32:53.692125: step: 534/1541, loss: 0.1409192532300949 2023-01-23 07:32:54.614009: step: 536/1541, loss: 0.8468283414840698 2023-01-23 07:32:55.531649: step: 538/1541, loss: 0.7895552515983582 2023-01-23 07:32:56.515586: step: 540/1541, loss: 0.705319881439209 2023-01-23 07:32:57.412122: step: 542/1541, loss: 0.6148459315299988 2023-01-23 07:32:58.327568: step: 544/1541, loss: 0.5350528955459595 2023-01-23 07:32:59.219489: step: 546/1541, loss: 0.2070532590150833 2023-01-23 07:33:00.105274: step: 548/1541, loss: 0.2605850398540497 2023-01-23 07:33:00.973419: step: 550/1541, loss: 0.4183087646961212 2023-01-23 07:33:01.899265: step: 552/1541, loss: 0.5023930072784424 2023-01-23 07:33:02.780180: step: 554/1541, loss: 0.6313613653182983 2023-01-23 07:33:03.667653: step: 556/1541, loss: 1.1585301160812378 2023-01-23 07:33:04.627929: step: 558/1541, loss: 0.6942268013954163 2023-01-23 07:33:05.528672: step: 560/1541, loss: 0.34604161977767944 2023-01-23 07:33:06.450623: step: 562/1541, loss: 0.3180631995201111 2023-01-23 07:33:07.396197: step: 564/1541, loss: 0.9543481469154358 2023-01-23 07:33:08.301663: step: 566/1541, loss: 0.2621641457080841 2023-01-23 07:33:09.187084: step: 568/1541, loss: 0.4458746016025543 2023-01-23 07:33:10.124789: step: 570/1541, loss: 0.37923187017440796 2023-01-23 07:33:11.039893: step: 572/1541, loss: 0.32854950428009033 2023-01-23 07:33:11.924207: step: 574/1541, loss: 0.3391023874282837 2023-01-23 07:33:12.866819: step: 576/1541, loss: 0.7070028781890869 2023-01-23 07:33:13.813920: step: 578/1541, loss: 0.7241475582122803 2023-01-23 07:33:14.765317: step: 580/1541, loss: 0.8056244850158691 2023-01-23 07:33:15.678600: step: 582/1541, loss: 0.6771398782730103 2023-01-23 07:33:16.548303: step: 584/1541, loss: 0.5165730118751526 2023-01-23 07:33:17.451537: step: 586/1541, loss: 0.48470431566238403 2023-01-23 07:33:18.366175: step: 588/1541, loss: 1.2861186265945435 2023-01-23 07:33:19.261860: step: 590/1541, loss: 0.3269585967063904 2023-01-23 07:33:20.171351: step: 592/1541, loss: 0.6900144815444946 2023-01-23 07:33:21.068957: step: 594/1541, loss: 1.1078028678894043 2023-01-23 07:33:21.935688: step: 596/1541, loss: 0.37502291798591614 2023-01-23 07:33:22.826850: step: 598/1541, loss: 0.4093250334262848 2023-01-23 07:33:23.699027: step: 600/1541, loss: 0.5671740174293518 2023-01-23 07:33:24.603981: step: 602/1541, loss: 0.27573534846305847 2023-01-23 07:33:25.503130: step: 604/1541, loss: 0.5224339962005615 2023-01-23 07:33:26.457335: step: 606/1541, loss: 0.2899583876132965 2023-01-23 07:33:27.302543: step: 608/1541, loss: 0.3647334575653076 2023-01-23 07:33:28.192700: step: 610/1541, loss: 2.815751075744629 2023-01-23 07:33:29.149200: step: 612/1541, loss: 1.6500641107559204 2023-01-23 07:33:30.076325: step: 614/1541, loss: 0.612697958946228 2023-01-23 07:33:31.006170: step: 616/1541, loss: 0.09897247701883316 2023-01-23 07:33:31.937120: step: 618/1541, loss: 1.4806981086730957 2023-01-23 07:33:32.862945: step: 620/1541, loss: 1.1984652280807495 2023-01-23 07:33:33.771004: step: 622/1541, loss: 0.7037032246589661 2023-01-23 07:33:34.651443: step: 624/1541, loss: 0.46315866708755493 2023-01-23 07:33:35.582036: step: 626/1541, loss: 0.7122559547424316 2023-01-23 07:33:36.517443: step: 628/1541, loss: 2.847468852996826 2023-01-23 07:33:37.398190: step: 630/1541, loss: 0.6590356230735779 2023-01-23 07:33:38.275262: step: 632/1541, loss: 0.35500895977020264 2023-01-23 07:33:39.222225: step: 634/1541, loss: 0.3747750520706177 2023-01-23 07:33:40.169239: step: 636/1541, loss: 0.3968716263771057 2023-01-23 07:33:41.091196: step: 638/1541, loss: 1.1799367666244507 2023-01-23 07:33:41.977032: step: 640/1541, loss: 0.21660448610782623 2023-01-23 07:33:42.888134: step: 642/1541, loss: 2.558652400970459 2023-01-23 07:33:43.788740: step: 644/1541, loss: 0.4585115909576416 2023-01-23 07:33:44.659125: step: 646/1541, loss: 0.24658718705177307 2023-01-23 07:33:45.533392: step: 648/1541, loss: 0.3374466300010681 2023-01-23 07:33:46.463451: step: 650/1541, loss: 0.5788304209709167 2023-01-23 07:33:47.382765: step: 652/1541, loss: 0.08700685948133469 2023-01-23 07:33:48.233377: step: 654/1541, loss: 0.2513788044452667 2023-01-23 07:33:49.164478: step: 656/1541, loss: 0.2467605620622635 2023-01-23 07:33:50.093905: step: 658/1541, loss: 0.29263314604759216 2023-01-23 07:33:50.964373: step: 660/1541, loss: 1.8820878267288208 2023-01-23 07:33:51.820448: step: 662/1541, loss: 0.28988537192344666 2023-01-23 07:33:52.675322: step: 664/1541, loss: 1.1962512731552124 2023-01-23 07:33:53.583357: step: 666/1541, loss: 0.5317884087562561 2023-01-23 07:33:54.490127: step: 668/1541, loss: 0.731965959072113 2023-01-23 07:33:55.368381: step: 670/1541, loss: 0.40487736463546753 2023-01-23 07:33:56.232893: step: 672/1541, loss: 0.22328367829322815 2023-01-23 07:33:57.109298: step: 674/1541, loss: 0.519987940788269 2023-01-23 07:33:57.981635: step: 676/1541, loss: 0.43755385279655457 2023-01-23 07:33:58.887014: step: 678/1541, loss: 0.32992368936538696 2023-01-23 07:33:59.785745: step: 680/1541, loss: 0.295981228351593 2023-01-23 07:34:00.675717: step: 682/1541, loss: 0.5732437968254089 2023-01-23 07:34:01.633639: step: 684/1541, loss: 0.3983825147151947 2023-01-23 07:34:02.563487: step: 686/1541, loss: 3.9708406925201416 2023-01-23 07:34:03.527289: step: 688/1541, loss: 2.397359848022461 2023-01-23 07:34:04.424521: step: 690/1541, loss: 1.073156476020813 2023-01-23 07:34:05.314334: step: 692/1541, loss: 0.39596104621887207 2023-01-23 07:34:06.205119: step: 694/1541, loss: 3.1622438430786133 2023-01-23 07:34:07.081696: step: 696/1541, loss: 0.44007083773612976 2023-01-23 07:34:07.999764: step: 698/1541, loss: 0.7964169979095459 2023-01-23 07:34:08.881031: step: 700/1541, loss: 0.7529511451721191 2023-01-23 07:34:09.739857: step: 702/1541, loss: 0.6801660060882568 2023-01-23 07:34:10.593206: step: 704/1541, loss: 0.6035998463630676 2023-01-23 07:34:11.456716: step: 706/1541, loss: 1.0845617055892944 2023-01-23 07:34:12.329319: step: 708/1541, loss: 0.4772772490978241 2023-01-23 07:34:13.230455: step: 710/1541, loss: 0.5693469643592834 2023-01-23 07:34:14.109725: step: 712/1541, loss: 1.348981499671936 2023-01-23 07:34:14.995458: step: 714/1541, loss: 0.26434361934661865 2023-01-23 07:34:15.902967: step: 716/1541, loss: 0.7102653384208679 2023-01-23 07:34:16.790758: step: 718/1541, loss: 0.8062350749969482 2023-01-23 07:34:17.722336: step: 720/1541, loss: 0.9568884968757629 2023-01-23 07:34:18.619365: step: 722/1541, loss: 0.3691597282886505 2023-01-23 07:34:19.500232: step: 724/1541, loss: 0.6466783881187439 2023-01-23 07:34:20.407166: step: 726/1541, loss: 0.5456416010856628 2023-01-23 07:34:21.283019: step: 728/1541, loss: 0.490933895111084 2023-01-23 07:34:22.147001: step: 730/1541, loss: 0.47307679057121277 2023-01-23 07:34:23.038421: step: 732/1541, loss: 0.6571069359779358 2023-01-23 07:34:23.958014: step: 734/1541, loss: 0.6549094319343567 2023-01-23 07:34:24.859624: step: 736/1541, loss: 0.24127857387065887 2023-01-23 07:34:25.721475: step: 738/1541, loss: 0.6160316467285156 2023-01-23 07:34:26.621682: step: 740/1541, loss: 0.5908976197242737 2023-01-23 07:34:27.503694: step: 742/1541, loss: 0.7992885708808899 2023-01-23 07:34:28.443113: step: 744/1541, loss: 0.23754680156707764 2023-01-23 07:34:29.381148: step: 746/1541, loss: 4.044122695922852 2023-01-23 07:34:30.322233: step: 748/1541, loss: 0.492037832736969 2023-01-23 07:34:31.217771: step: 750/1541, loss: 0.3546375036239624 2023-01-23 07:34:32.149992: step: 752/1541, loss: 0.31478169560432434 2023-01-23 07:34:33.056241: step: 754/1541, loss: 0.3632951080799103 2023-01-23 07:34:33.940185: step: 756/1541, loss: 0.3735029697418213 2023-01-23 07:34:34.833316: step: 758/1541, loss: 0.37639716267585754 2023-01-23 07:34:35.728837: step: 760/1541, loss: 0.5683416724205017 2023-01-23 07:34:36.607902: step: 762/1541, loss: 2.6569113731384277 2023-01-23 07:34:37.488043: step: 764/1541, loss: 0.4164068102836609 2023-01-23 07:34:38.377634: step: 766/1541, loss: 0.5426962375640869 2023-01-23 07:34:39.247811: step: 768/1541, loss: 2.3600807189941406 2023-01-23 07:34:40.128494: step: 770/1541, loss: 0.7029240727424622 2023-01-23 07:34:41.093876: step: 772/1541, loss: 5.350169658660889 2023-01-23 07:34:41.994578: step: 774/1541, loss: 1.1363372802734375 2023-01-23 07:34:42.894137: step: 776/1541, loss: 0.2542482614517212 2023-01-23 07:34:43.898002: step: 778/1541, loss: 0.6072421073913574 2023-01-23 07:34:44.810483: step: 780/1541, loss: 0.5105410218238831 2023-01-23 07:34:45.688993: step: 782/1541, loss: 0.7634667158126831 2023-01-23 07:34:46.596825: step: 784/1541, loss: 0.3805483281612396 2023-01-23 07:34:47.526400: step: 786/1541, loss: 0.47462958097457886 2023-01-23 07:34:48.395366: step: 788/1541, loss: 0.9740381836891174 2023-01-23 07:34:49.315926: step: 790/1541, loss: 0.2404646873474121 2023-01-23 07:34:50.267633: step: 792/1541, loss: 0.3648764193058014 2023-01-23 07:34:51.178129: step: 794/1541, loss: 0.3686525821685791 2023-01-23 07:34:52.049961: step: 796/1541, loss: 0.28640174865722656 2023-01-23 07:34:52.995116: step: 798/1541, loss: 0.5526148676872253 2023-01-23 07:34:53.878648: step: 800/1541, loss: 0.5249645113945007 2023-01-23 07:34:54.808533: step: 802/1541, loss: 0.8303833603858948 2023-01-23 07:34:55.774063: step: 804/1541, loss: 0.664072573184967 2023-01-23 07:34:56.715157: step: 806/1541, loss: 0.5301144123077393 2023-01-23 07:34:57.610976: step: 808/1541, loss: 0.4630092978477478 2023-01-23 07:34:58.464100: step: 810/1541, loss: 0.11695808172225952 2023-01-23 07:34:59.387073: step: 812/1541, loss: 0.16067437827587128 2023-01-23 07:35:00.274021: step: 814/1541, loss: 0.2643561065196991 2023-01-23 07:35:01.196555: step: 816/1541, loss: 0.9450242519378662 2023-01-23 07:35:02.145438: step: 818/1541, loss: 0.7243616580963135 2023-01-23 07:35:03.087545: step: 820/1541, loss: 1.8225269317626953 2023-01-23 07:35:03.985663: step: 822/1541, loss: 0.7130281925201416 2023-01-23 07:35:04.859974: step: 824/1541, loss: 0.7605944275856018 2023-01-23 07:35:05.782175: step: 826/1541, loss: 0.2852214574813843 2023-01-23 07:35:06.696066: step: 828/1541, loss: 1.7955381870269775 2023-01-23 07:35:07.532467: step: 830/1541, loss: 0.6251822710037231 2023-01-23 07:35:08.427545: step: 832/1541, loss: 0.5454850196838379 2023-01-23 07:35:09.326071: step: 834/1541, loss: 0.22695212066173553 2023-01-23 07:35:10.236101: step: 836/1541, loss: 0.44255468249320984 2023-01-23 07:35:11.182514: step: 838/1541, loss: 0.235067218542099 2023-01-23 07:35:12.048661: step: 840/1541, loss: 0.9511951208114624 2023-01-23 07:35:13.013556: step: 842/1541, loss: 0.10420887917280197 2023-01-23 07:35:13.899218: step: 844/1541, loss: 2.8483359813690186 2023-01-23 07:35:14.772847: step: 846/1541, loss: 0.7827538251876831 2023-01-23 07:35:15.692075: step: 848/1541, loss: 0.5751728415489197 2023-01-23 07:35:16.643930: step: 850/1541, loss: 0.5445276498794556 2023-01-23 07:35:17.587748: step: 852/1541, loss: 0.49207645654678345 2023-01-23 07:35:18.493379: step: 854/1541, loss: 0.42468690872192383 2023-01-23 07:35:19.470817: step: 856/1541, loss: 0.676232099533081 2023-01-23 07:35:20.403035: step: 858/1541, loss: 0.7223939299583435 2023-01-23 07:35:21.301841: step: 860/1541, loss: 6.260109901428223 2023-01-23 07:35:22.167325: step: 862/1541, loss: 1.269473910331726 2023-01-23 07:35:23.047457: step: 864/1541, loss: 0.3684690594673157 2023-01-23 07:35:23.947292: step: 866/1541, loss: 0.2759038507938385 2023-01-23 07:35:24.809962: step: 868/1541, loss: 0.33237022161483765 2023-01-23 07:35:25.704992: step: 870/1541, loss: 0.4041779041290283 2023-01-23 07:35:26.581136: step: 872/1541, loss: 0.3030291199684143 2023-01-23 07:35:27.484670: step: 874/1541, loss: 0.5017563104629517 2023-01-23 07:35:28.370910: step: 876/1541, loss: 0.6335048675537109 2023-01-23 07:35:29.251386: step: 878/1541, loss: 2.6817948818206787 2023-01-23 07:35:30.156286: step: 880/1541, loss: 1.20499849319458 2023-01-23 07:35:31.017649: step: 882/1541, loss: 0.5371527075767517 2023-01-23 07:35:31.906827: step: 884/1541, loss: 0.36983442306518555 2023-01-23 07:35:32.814179: step: 886/1541, loss: 0.24775850772857666 2023-01-23 07:35:33.740135: step: 888/1541, loss: 1.6209282875061035 2023-01-23 07:35:34.663303: step: 890/1541, loss: 3.017122745513916 2023-01-23 07:35:35.563520: step: 892/1541, loss: 2.949517250061035 2023-01-23 07:35:36.445878: step: 894/1541, loss: 0.43164950609207153 2023-01-23 07:35:37.345022: step: 896/1541, loss: 0.6253476738929749 2023-01-23 07:35:38.266018: step: 898/1541, loss: 0.7035107016563416 2023-01-23 07:35:39.181147: step: 900/1541, loss: 0.3448043763637543 2023-01-23 07:35:40.073306: step: 902/1541, loss: 0.5305758714675903 2023-01-23 07:35:40.943972: step: 904/1541, loss: 0.6606314778327942 2023-01-23 07:35:41.865536: step: 906/1541, loss: 0.6259119510650635 2023-01-23 07:35:42.736918: step: 908/1541, loss: 0.47445422410964966 2023-01-23 07:35:43.626241: step: 910/1541, loss: 0.2632613480091095 2023-01-23 07:35:44.516255: step: 912/1541, loss: 0.8344745635986328 2023-01-23 07:35:45.408461: step: 914/1541, loss: 2.04886531829834 2023-01-23 07:35:46.301973: step: 916/1541, loss: 0.5797562003135681 2023-01-23 07:35:47.177944: step: 918/1541, loss: 0.4554595947265625 2023-01-23 07:35:48.112274: step: 920/1541, loss: 2.7801249027252197 2023-01-23 07:35:49.081555: step: 922/1541, loss: 0.5338180065155029 2023-01-23 07:35:49.962261: step: 924/1541, loss: 3.4457993507385254 2023-01-23 07:35:50.848888: step: 926/1541, loss: 0.2356952279806137 2023-01-23 07:35:51.748013: step: 928/1541, loss: 0.781859815120697 2023-01-23 07:35:52.687131: step: 930/1541, loss: 0.21223798394203186 2023-01-23 07:35:53.623714: step: 932/1541, loss: 1.017479419708252 2023-01-23 07:35:54.580751: step: 934/1541, loss: 0.34725669026374817 2023-01-23 07:35:55.464576: step: 936/1541, loss: 0.10230568796396255 2023-01-23 07:35:56.376626: step: 938/1541, loss: 0.6401623487472534 2023-01-23 07:35:57.272770: step: 940/1541, loss: 0.28686606884002686 2023-01-23 07:35:58.283398: step: 942/1541, loss: 0.22817867994308472 2023-01-23 07:35:59.169202: step: 944/1541, loss: 0.2418433427810669 2023-01-23 07:36:00.083487: step: 946/1541, loss: 0.44537290930747986 2023-01-23 07:36:01.004037: step: 948/1541, loss: 0.4259825050830841 2023-01-23 07:36:01.892318: step: 950/1541, loss: 1.0799678564071655 2023-01-23 07:36:02.774040: step: 952/1541, loss: 0.4374207556247711 2023-01-23 07:36:03.685657: step: 954/1541, loss: 1.15109384059906 2023-01-23 07:36:04.580461: step: 956/1541, loss: 0.3016016185283661 2023-01-23 07:36:05.499904: step: 958/1541, loss: 1.7758619785308838 2023-01-23 07:36:06.399067: step: 960/1541, loss: 0.39634400606155396 2023-01-23 07:36:07.290227: step: 962/1541, loss: 0.7258965373039246 2023-01-23 07:36:08.231509: step: 964/1541, loss: 0.22947531938552856 2023-01-23 07:36:09.266372: step: 966/1541, loss: 0.25345972180366516 2023-01-23 07:36:10.162114: step: 968/1541, loss: 0.4889308214187622 2023-01-23 07:36:11.038823: step: 970/1541, loss: 0.3609386086463928 2023-01-23 07:36:12.044408: step: 972/1541, loss: 0.34476932883262634 2023-01-23 07:36:12.950883: step: 974/1541, loss: 2.7162017822265625 2023-01-23 07:36:13.831118: step: 976/1541, loss: 0.6947752833366394 2023-01-23 07:36:14.769987: step: 978/1541, loss: 0.41056814789772034 2023-01-23 07:36:15.624532: step: 980/1541, loss: 3.4516541957855225 2023-01-23 07:36:16.566373: step: 982/1541, loss: 0.7445786595344543 2023-01-23 07:36:17.439516: step: 984/1541, loss: 0.7263025045394897 2023-01-23 07:36:18.349374: step: 986/1541, loss: 0.5466316342353821 2023-01-23 07:36:19.219807: step: 988/1541, loss: 0.8063783049583435 2023-01-23 07:36:20.111452: step: 990/1541, loss: 0.7906944751739502 2023-01-23 07:36:20.983730: step: 992/1541, loss: 0.18509134650230408 2023-01-23 07:36:21.869875: step: 994/1541, loss: 0.5561310648918152 2023-01-23 07:36:22.799567: step: 996/1541, loss: 0.3927188217639923 2023-01-23 07:36:23.669910: step: 998/1541, loss: 0.47916659712791443 2023-01-23 07:36:24.549306: step: 1000/1541, loss: 0.4942193925380707 2023-01-23 07:36:25.452486: step: 1002/1541, loss: 0.3566565215587616 2023-01-23 07:36:26.370485: step: 1004/1541, loss: 3.343271493911743 2023-01-23 07:36:27.251068: step: 1006/1541, loss: 0.3570817708969116 2023-01-23 07:36:28.128440: step: 1008/1541, loss: 0.5698543190956116 2023-01-23 07:36:29.019996: step: 1010/1541, loss: 0.5940348505973816 2023-01-23 07:36:29.906016: step: 1012/1541, loss: 0.15990616381168365 2023-01-23 07:36:30.916049: step: 1014/1541, loss: 0.11668329685926437 2023-01-23 07:36:31.793436: step: 1016/1541, loss: 0.860272228717804 2023-01-23 07:36:32.680784: step: 1018/1541, loss: 0.7826528549194336 2023-01-23 07:36:33.620633: step: 1020/1541, loss: 0.20190156996250153 2023-01-23 07:36:34.538396: step: 1022/1541, loss: 0.3348049223423004 2023-01-23 07:36:35.418039: step: 1024/1541, loss: 0.4455721378326416 2023-01-23 07:36:36.278659: step: 1026/1541, loss: 0.44219541549682617 2023-01-23 07:36:37.159657: step: 1028/1541, loss: 0.5935437083244324 2023-01-23 07:36:38.053210: step: 1030/1541, loss: 0.5856661200523376 2023-01-23 07:36:38.948405: step: 1032/1541, loss: 0.09067773818969727 2023-01-23 07:36:39.846357: step: 1034/1541, loss: 0.669896125793457 2023-01-23 07:36:40.735807: step: 1036/1541, loss: 0.6033347845077515 2023-01-23 07:36:41.615613: step: 1038/1541, loss: 0.8281624913215637 2023-01-23 07:36:42.502995: step: 1040/1541, loss: 0.7439713478088379 2023-01-23 07:36:43.383674: step: 1042/1541, loss: 0.7827363610267639 2023-01-23 07:36:44.289447: step: 1044/1541, loss: 2.5513949394226074 2023-01-23 07:36:45.178467: step: 1046/1541, loss: 0.4544568657875061 2023-01-23 07:36:46.122389: step: 1048/1541, loss: 0.9045029282569885 2023-01-23 07:36:47.056905: step: 1050/1541, loss: 0.752305269241333 2023-01-23 07:36:47.933626: step: 1052/1541, loss: 0.11627599596977234 2023-01-23 07:36:48.870684: step: 1054/1541, loss: 0.7777702808380127 2023-01-23 07:36:49.744759: step: 1056/1541, loss: 5.260656833648682 2023-01-23 07:36:50.605375: step: 1058/1541, loss: 1.1494892835617065 2023-01-23 07:36:51.537180: step: 1060/1541, loss: 0.2884897291660309 2023-01-23 07:36:52.437620: step: 1062/1541, loss: 0.9772645235061646 2023-01-23 07:36:53.366342: step: 1064/1541, loss: 0.2515645921230316 2023-01-23 07:36:54.228627: step: 1066/1541, loss: 0.41256651282310486 2023-01-23 07:36:55.221676: step: 1068/1541, loss: 0.7231655716896057 2023-01-23 07:36:56.117033: step: 1070/1541, loss: 1.7784807682037354 2023-01-23 07:36:56.959253: step: 1072/1541, loss: 0.7518859505653381 2023-01-23 07:36:57.889675: step: 1074/1541, loss: 1.1050477027893066 2023-01-23 07:36:58.775741: step: 1076/1541, loss: 0.6274323463439941 2023-01-23 07:36:59.652916: step: 1078/1541, loss: 2.5760889053344727 2023-01-23 07:37:00.533453: step: 1080/1541, loss: 0.6022012233734131 2023-01-23 07:37:01.408538: step: 1082/1541, loss: 0.4579090178012848 2023-01-23 07:37:02.356179: step: 1084/1541, loss: 0.31383267045021057 2023-01-23 07:37:03.284204: step: 1086/1541, loss: 0.9515392184257507 2023-01-23 07:37:04.173317: step: 1088/1541, loss: 0.3204362392425537 2023-01-23 07:37:05.136553: step: 1090/1541, loss: 0.23692597448825836 2023-01-23 07:37:06.018982: step: 1092/1541, loss: 0.3729360103607178 2023-01-23 07:37:06.907676: step: 1094/1541, loss: 0.3071056604385376 2023-01-23 07:37:07.820121: step: 1096/1541, loss: 0.4494170546531677 2023-01-23 07:37:08.700226: step: 1098/1541, loss: 0.21378082036972046 2023-01-23 07:37:09.585416: step: 1100/1541, loss: 0.26828885078430176 2023-01-23 07:37:10.474437: step: 1102/1541, loss: 0.2949698269367218 2023-01-23 07:37:11.371732: step: 1104/1541, loss: 0.312873899936676 2023-01-23 07:37:12.283273: step: 1106/1541, loss: 0.33266013860702515 2023-01-23 07:37:13.224511: step: 1108/1541, loss: 0.48245131969451904 2023-01-23 07:37:14.154362: step: 1110/1541, loss: 0.7071015238761902 2023-01-23 07:37:15.056349: step: 1112/1541, loss: 0.5664901733398438 2023-01-23 07:37:15.935874: step: 1114/1541, loss: 0.14100973308086395 2023-01-23 07:37:16.828501: step: 1116/1541, loss: 0.7930102348327637 2023-01-23 07:37:17.719049: step: 1118/1541, loss: 0.6491302251815796 2023-01-23 07:37:18.615259: step: 1120/1541, loss: 0.8862331509590149 2023-01-23 07:37:19.549205: step: 1122/1541, loss: 0.6725467443466187 2023-01-23 07:37:20.465744: step: 1124/1541, loss: 0.6601250171661377 2023-01-23 07:37:21.436114: step: 1126/1541, loss: 1.057313323020935 2023-01-23 07:37:22.332723: step: 1128/1541, loss: 0.34777745604515076 2023-01-23 07:37:23.209317: step: 1130/1541, loss: 0.18097764253616333 2023-01-23 07:37:24.069708: step: 1132/1541, loss: 0.6565943360328674 2023-01-23 07:37:24.968880: step: 1134/1541, loss: 0.8931524157524109 2023-01-23 07:37:25.884710: step: 1136/1541, loss: 0.7064542174339294 2023-01-23 07:37:26.765733: step: 1138/1541, loss: 0.46629437804222107 2023-01-23 07:37:27.682236: step: 1140/1541, loss: 0.6157916784286499 2023-01-23 07:37:28.576541: step: 1142/1541, loss: 0.5076134204864502 2023-01-23 07:37:29.478526: step: 1144/1541, loss: 0.3612079322338104 2023-01-23 07:37:30.385711: step: 1146/1541, loss: 0.44980329275131226 2023-01-23 07:37:31.308661: step: 1148/1541, loss: 0.7188858985900879 2023-01-23 07:37:32.266420: step: 1150/1541, loss: 0.9763041734695435 2023-01-23 07:37:33.139026: step: 1152/1541, loss: 0.41714194416999817 2023-01-23 07:37:34.017492: step: 1154/1541, loss: 0.4692496657371521 2023-01-23 07:37:34.904681: step: 1156/1541, loss: 0.6360200643539429 2023-01-23 07:37:35.915985: step: 1158/1541, loss: 3.2300963401794434 2023-01-23 07:37:36.809171: step: 1160/1541, loss: 0.4230053126811981 2023-01-23 07:37:37.674545: step: 1162/1541, loss: 0.5327027440071106 2023-01-23 07:37:38.553275: step: 1164/1541, loss: 0.42528027296066284 2023-01-23 07:37:39.408823: step: 1166/1541, loss: 0.39429038763046265 2023-01-23 07:37:40.359518: step: 1168/1541, loss: 0.46338340640068054 2023-01-23 07:37:41.281752: step: 1170/1541, loss: 0.3408656120300293 2023-01-23 07:37:42.195642: step: 1172/1541, loss: 0.5638906359672546 2023-01-23 07:37:43.126935: step: 1174/1541, loss: 0.38081246614456177 2023-01-23 07:37:43.995089: step: 1176/1541, loss: 0.4796012341976166 2023-01-23 07:37:44.926107: step: 1178/1541, loss: 0.205643892288208 2023-01-23 07:37:45.817907: step: 1180/1541, loss: 0.8027034401893616 2023-01-23 07:37:46.681091: step: 1182/1541, loss: 0.31363287568092346 2023-01-23 07:37:47.607070: step: 1184/1541, loss: 0.4812757074832916 2023-01-23 07:37:48.519898: step: 1186/1541, loss: 3.3730552196502686 2023-01-23 07:37:49.396463: step: 1188/1541, loss: 1.2639845609664917 2023-01-23 07:37:50.276299: step: 1190/1541, loss: 0.5311660766601562 2023-01-23 07:37:51.140231: step: 1192/1541, loss: 0.36534005403518677 2023-01-23 07:37:52.030018: step: 1194/1541, loss: 0.4665931463241577 2023-01-23 07:37:52.915155: step: 1196/1541, loss: 0.8587096929550171 2023-01-23 07:37:53.854255: step: 1198/1541, loss: 0.5670616030693054 2023-01-23 07:37:54.784637: step: 1200/1541, loss: 0.9824780225753784 2023-01-23 07:37:55.696260: step: 1202/1541, loss: 0.7411944270133972 2023-01-23 07:37:56.566826: step: 1204/1541, loss: 0.26517707109451294 2023-01-23 07:37:57.496994: step: 1206/1541, loss: 0.40082159638404846 2023-01-23 07:37:58.409695: step: 1208/1541, loss: 0.4275926947593689 2023-01-23 07:37:59.295608: step: 1210/1541, loss: 0.22526341676712036 2023-01-23 07:38:00.171832: step: 1212/1541, loss: 0.36628714203834534 2023-01-23 07:38:01.063471: step: 1214/1541, loss: 0.5922456383705139 2023-01-23 07:38:01.983944: step: 1216/1541, loss: 0.29867321252822876 2023-01-23 07:38:02.915886: step: 1218/1541, loss: 0.1311783343553543 2023-01-23 07:38:03.864000: step: 1220/1541, loss: 0.320320188999176 2023-01-23 07:38:04.779877: step: 1222/1541, loss: 0.28932952880859375 2023-01-23 07:38:05.680809: step: 1224/1541, loss: 0.48185446858406067 2023-01-23 07:38:06.575007: step: 1226/1541, loss: 0.4356159269809723 2023-01-23 07:38:07.495540: step: 1228/1541, loss: 0.5327005386352539 2023-01-23 07:38:08.358435: step: 1230/1541, loss: 0.6700252890586853 2023-01-23 07:38:09.278732: step: 1232/1541, loss: 0.6600300669670105 2023-01-23 07:38:10.156668: step: 1234/1541, loss: 0.7800655961036682 2023-01-23 07:38:11.105397: step: 1236/1541, loss: 0.3295007646083832 2023-01-23 07:38:12.005900: step: 1238/1541, loss: 0.3904189169406891 2023-01-23 07:38:12.870771: step: 1240/1541, loss: 0.2924191355705261 2023-01-23 07:38:13.724253: step: 1242/1541, loss: 0.43101298809051514 2023-01-23 07:38:14.590556: step: 1244/1541, loss: 0.9181525111198425 2023-01-23 07:38:15.473357: step: 1246/1541, loss: 0.34277647733688354 2023-01-23 07:38:16.457368: step: 1248/1541, loss: 0.3014097213745117 2023-01-23 07:38:17.383035: step: 1250/1541, loss: 2.0154049396514893 2023-01-23 07:38:18.244036: step: 1252/1541, loss: 0.7040720582008362 2023-01-23 07:38:19.138100: step: 1254/1541, loss: 0.6427429914474487 2023-01-23 07:38:20.054413: step: 1256/1541, loss: 1.074622392654419 2023-01-23 07:38:20.964841: step: 1258/1541, loss: 0.3047507405281067 2023-01-23 07:38:21.888531: step: 1260/1541, loss: 0.548299252986908 2023-01-23 07:38:22.782562: step: 1262/1541, loss: 0.4496154487133026 2023-01-23 07:38:23.700509: step: 1264/1541, loss: 0.8399937152862549 2023-01-23 07:38:24.639719: step: 1266/1541, loss: 0.4888854920864105 2023-01-23 07:38:25.520219: step: 1268/1541, loss: 6.78651237487793 2023-01-23 07:38:26.421790: step: 1270/1541, loss: 0.35305047035217285 2023-01-23 07:38:27.338960: step: 1272/1541, loss: 0.20726226270198822 2023-01-23 07:38:28.329181: step: 1274/1541, loss: 0.5576119422912598 2023-01-23 07:38:29.227380: step: 1276/1541, loss: 0.6775772571563721 2023-01-23 07:38:30.088893: step: 1278/1541, loss: 1.11936354637146 2023-01-23 07:38:30.979486: step: 1280/1541, loss: 0.7246091365814209 2023-01-23 07:38:31.874043: step: 1282/1541, loss: 0.5001809000968933 2023-01-23 07:38:32.788477: step: 1284/1541, loss: 0.5657103657722473 2023-01-23 07:38:33.751056: step: 1286/1541, loss: 0.2783762812614441 2023-01-23 07:38:34.697775: step: 1288/1541, loss: 0.360976904630661 2023-01-23 07:38:35.625675: step: 1290/1541, loss: 0.31284815073013306 2023-01-23 07:38:36.528574: step: 1292/1541, loss: 0.5728170871734619 2023-01-23 07:38:37.430550: step: 1294/1541, loss: 0.5537997484207153 2023-01-23 07:38:38.278412: step: 1296/1541, loss: 0.39191192388534546 2023-01-23 07:38:39.161540: step: 1298/1541, loss: 0.39525550603866577 2023-01-23 07:38:40.055277: step: 1300/1541, loss: 0.5906915068626404 2023-01-23 07:38:40.935127: step: 1302/1541, loss: 0.24047698080539703 2023-01-23 07:38:41.856026: step: 1304/1541, loss: 1.6463847160339355 2023-01-23 07:38:42.775748: step: 1306/1541, loss: 0.26109105348587036 2023-01-23 07:38:43.693115: step: 1308/1541, loss: 0.48667922616004944 2023-01-23 07:38:44.570154: step: 1310/1541, loss: 0.634810745716095 2023-01-23 07:38:45.493618: step: 1312/1541, loss: 0.4260971248149872 2023-01-23 07:38:46.450081: step: 1314/1541, loss: 0.5299062132835388 2023-01-23 07:38:47.344158: step: 1316/1541, loss: 0.5659303665161133 2023-01-23 07:38:48.296054: step: 1318/1541, loss: 0.3393058180809021 2023-01-23 07:38:49.199770: step: 1320/1541, loss: 0.3880079984664917 2023-01-23 07:38:50.120109: step: 1322/1541, loss: 0.7633254528045654 2023-01-23 07:38:51.014895: step: 1324/1541, loss: 0.14275532960891724 2023-01-23 07:38:51.892213: step: 1326/1541, loss: 0.10541525483131409 2023-01-23 07:38:52.810998: step: 1328/1541, loss: 1.014773964881897 2023-01-23 07:38:53.725316: step: 1330/1541, loss: 0.5553672313690186 2023-01-23 07:38:54.651701: step: 1332/1541, loss: 3.0846946239471436 2023-01-23 07:38:55.531903: step: 1334/1541, loss: 0.37686586380004883 2023-01-23 07:38:56.435431: step: 1336/1541, loss: 0.33143556118011475 2023-01-23 07:38:57.368590: step: 1338/1541, loss: 0.3340452015399933 2023-01-23 07:38:58.262970: step: 1340/1541, loss: 0.352801650762558 2023-01-23 07:38:59.137154: step: 1342/1541, loss: 0.5140126347541809 2023-01-23 07:39:00.027202: step: 1344/1541, loss: 4.933694362640381 2023-01-23 07:39:00.936925: step: 1346/1541, loss: 0.6043404340744019 2023-01-23 07:39:01.780625: step: 1348/1541, loss: 0.6080958843231201 2023-01-23 07:39:02.663747: step: 1350/1541, loss: 0.6256071925163269 2023-01-23 07:39:03.578956: step: 1352/1541, loss: 0.5295577049255371 2023-01-23 07:39:04.487156: step: 1354/1541, loss: 0.6247133016586304 2023-01-23 07:39:05.384338: step: 1356/1541, loss: 0.6012261509895325 2023-01-23 07:39:06.277183: step: 1358/1541, loss: 0.2682637870311737 2023-01-23 07:39:07.177839: step: 1360/1541, loss: 0.46150150895118713 2023-01-23 07:39:08.083907: step: 1362/1541, loss: 0.16406215727329254 2023-01-23 07:39:09.100092: step: 1364/1541, loss: 0.7621626853942871 2023-01-23 07:39:09.980025: step: 1366/1541, loss: 0.446837842464447 2023-01-23 07:39:10.991393: step: 1368/1541, loss: 0.7418548464775085 2023-01-23 07:39:11.930501: step: 1370/1541, loss: 0.30874115228652954 2023-01-23 07:39:12.781445: step: 1372/1541, loss: 0.21921543776988983 2023-01-23 07:39:13.690222: step: 1374/1541, loss: 0.2936200499534607 2023-01-23 07:39:14.550327: step: 1376/1541, loss: 0.47333425283432007 2023-01-23 07:39:15.435312: step: 1378/1541, loss: 0.6369785070419312 2023-01-23 07:39:16.291574: step: 1380/1541, loss: 0.5638763308525085 2023-01-23 07:39:17.243406: step: 1382/1541, loss: 0.8528102040290833 2023-01-23 07:39:18.170979: step: 1384/1541, loss: 0.4444093406200409 2023-01-23 07:39:19.066009: step: 1386/1541, loss: 0.7324382066726685 2023-01-23 07:39:19.931224: step: 1388/1541, loss: 0.6748558878898621 2023-01-23 07:39:20.815417: step: 1390/1541, loss: 0.6387847661972046 2023-01-23 07:39:21.812321: step: 1392/1541, loss: 0.5114825963973999 2023-01-23 07:39:22.699520: step: 1394/1541, loss: 0.9804747700691223 2023-01-23 07:39:23.610932: step: 1396/1541, loss: 0.4335254728794098 2023-01-23 07:39:24.500555: step: 1398/1541, loss: 1.1557689905166626 2023-01-23 07:39:25.416961: step: 1400/1541, loss: 0.7568908929824829 2023-01-23 07:39:26.262935: step: 1402/1541, loss: 0.5998796224594116 2023-01-23 07:39:27.176599: step: 1404/1541, loss: 0.13679426908493042 2023-01-23 07:39:28.044598: step: 1406/1541, loss: 0.9320364594459534 2023-01-23 07:39:28.943205: step: 1408/1541, loss: 1.0838992595672607 2023-01-23 07:39:29.819863: step: 1410/1541, loss: 0.18021908402442932 2023-01-23 07:39:30.718818: step: 1412/1541, loss: 0.4555767774581909 2023-01-23 07:39:31.645914: step: 1414/1541, loss: 0.6847048997879028 2023-01-23 07:39:32.573843: step: 1416/1541, loss: 0.4043191373348236 2023-01-23 07:39:33.511701: step: 1418/1541, loss: 0.8880006670951843 2023-01-23 07:39:34.422104: step: 1420/1541, loss: 0.3275037705898285 2023-01-23 07:39:35.319066: step: 1422/1541, loss: 3.0006608963012695 2023-01-23 07:39:36.219069: step: 1424/1541, loss: 0.41119661927223206 2023-01-23 07:39:37.188431: step: 1426/1541, loss: 0.5253315567970276 2023-01-23 07:39:38.069789: step: 1428/1541, loss: 0.3060051202774048 2023-01-23 07:39:38.943565: step: 1430/1541, loss: 0.6331658959388733 2023-01-23 07:39:39.868790: step: 1432/1541, loss: 0.20616474747657776 2023-01-23 07:39:40.733142: step: 1434/1541, loss: 0.3582926094532013 2023-01-23 07:39:41.658283: step: 1436/1541, loss: 0.2369866520166397 2023-01-23 07:39:42.523076: step: 1438/1541, loss: 0.2875358760356903 2023-01-23 07:39:43.435354: step: 1440/1541, loss: 0.40218380093574524 2023-01-23 07:39:44.394435: step: 1442/1541, loss: 0.4083295464515686 2023-01-23 07:39:45.283157: step: 1444/1541, loss: 0.21769268810749054 2023-01-23 07:39:46.170400: step: 1446/1541, loss: 1.2197574377059937 2023-01-23 07:39:47.078330: step: 1448/1541, loss: 0.3881017565727234 2023-01-23 07:39:47.987538: step: 1450/1541, loss: 0.15203574299812317 2023-01-23 07:39:48.878323: step: 1452/1541, loss: 0.6045504212379456 2023-01-23 07:39:49.797532: step: 1454/1541, loss: 0.4208999276161194 2023-01-23 07:39:50.683941: step: 1456/1541, loss: 1.0094655752182007 2023-01-23 07:39:51.560441: step: 1458/1541, loss: 0.3930668532848358 2023-01-23 07:39:52.557163: step: 1460/1541, loss: 0.22725845873355865 2023-01-23 07:39:53.473860: step: 1462/1541, loss: 0.49215343594551086 2023-01-23 07:39:54.432796: step: 1464/1541, loss: 0.33826228976249695 2023-01-23 07:39:55.390752: step: 1466/1541, loss: 0.3415832817554474 2023-01-23 07:39:56.289025: step: 1468/1541, loss: 0.577085018157959 2023-01-23 07:39:57.166168: step: 1470/1541, loss: 0.3390585780143738 2023-01-23 07:39:58.044911: step: 1472/1541, loss: 0.8366755247116089 2023-01-23 07:39:58.948194: step: 1474/1541, loss: 0.9019534587860107 2023-01-23 07:39:59.810225: step: 1476/1541, loss: 0.28179091215133667 2023-01-23 07:40:00.703830: step: 1478/1541, loss: 0.7666825652122498 2023-01-23 07:40:01.598264: step: 1480/1541, loss: 0.8776026368141174 2023-01-23 07:40:02.539734: step: 1482/1541, loss: 1.7366726398468018 2023-01-23 07:40:03.407584: step: 1484/1541, loss: 0.4535895884037018 2023-01-23 07:40:04.329409: step: 1486/1541, loss: 0.2853890359401703 2023-01-23 07:40:05.222732: step: 1488/1541, loss: 0.4535808563232422 2023-01-23 07:40:06.087395: step: 1490/1541, loss: 6.734591960906982 2023-01-23 07:40:06.939750: step: 1492/1541, loss: 0.6970463991165161 2023-01-23 07:40:07.817111: step: 1494/1541, loss: 0.4643147587776184 2023-01-23 07:40:08.720585: step: 1496/1541, loss: 5.621339797973633 2023-01-23 07:40:09.649235: step: 1498/1541, loss: 0.4883018136024475 2023-01-23 07:40:10.634622: step: 1500/1541, loss: 0.6850176453590393 2023-01-23 07:40:11.494149: step: 1502/1541, loss: 0.8645907640457153 2023-01-23 07:40:12.400464: step: 1504/1541, loss: 0.40511295199394226 2023-01-23 07:40:13.387564: step: 1506/1541, loss: 0.4773784279823303 2023-01-23 07:40:14.366329: step: 1508/1541, loss: 0.4860825538635254 2023-01-23 07:40:15.296648: step: 1510/1541, loss: 0.42023736238479614 2023-01-23 07:40:16.172420: step: 1512/1541, loss: 0.4416618049144745 2023-01-23 07:40:17.107794: step: 1514/1541, loss: 0.17788131535053253 2023-01-23 07:40:18.033933: step: 1516/1541, loss: 3.3933298587799072 2023-01-23 07:40:18.884891: step: 1518/1541, loss: 0.6042893528938293 2023-01-23 07:40:19.756587: step: 1520/1541, loss: 0.27926769852638245 2023-01-23 07:40:20.668538: step: 1522/1541, loss: 0.45620018243789673 2023-01-23 07:40:21.590092: step: 1524/1541, loss: 0.11441903561353683 2023-01-23 07:40:22.503283: step: 1526/1541, loss: 0.45958852767944336 2023-01-23 07:40:23.396090: step: 1528/1541, loss: 0.4643900394439697 2023-01-23 07:40:24.300185: step: 1530/1541, loss: 0.7697243690490723 2023-01-23 07:40:25.191259: step: 1532/1541, loss: 0.1989557445049286 2023-01-23 07:40:26.129774: step: 1534/1541, loss: 0.4608638286590576 2023-01-23 07:40:27.064968: step: 1536/1541, loss: 0.9047723412513733 2023-01-23 07:40:27.953545: step: 1538/1541, loss: 0.36415237188339233 2023-01-23 07:40:28.863289: step: 1540/1541, loss: 0.3601270914077759 2023-01-23 07:40:29.774097: step: 1542/1541, loss: 0.26049691438674927 2023-01-23 07:40:30.690975: step: 1544/1541, loss: 1.2196725606918335 2023-01-23 07:40:31.590889: step: 1546/1541, loss: 0.6354082226753235 2023-01-23 07:40:32.548134: step: 1548/1541, loss: 0.6900285482406616 2023-01-23 07:40:33.487548: step: 1550/1541, loss: 0.454179048538208 2023-01-23 07:40:34.426710: step: 1552/1541, loss: 1.0477672815322876 2023-01-23 07:40:35.354105: step: 1554/1541, loss: 0.9201082587242126 2023-01-23 07:40:36.321177: step: 1556/1541, loss: 0.8219138383865356 2023-01-23 07:40:37.181463: step: 1558/1541, loss: 0.3166661560535431 2023-01-23 07:40:38.093000: step: 1560/1541, loss: 0.33498865365982056 2023-01-23 07:40:39.009364: step: 1562/1541, loss: 1.2725926637649536 2023-01-23 07:40:39.919671: step: 1564/1541, loss: 0.33711379766464233 2023-01-23 07:40:40.820798: step: 1566/1541, loss: 0.2887072265148163 2023-01-23 07:40:41.741724: step: 1568/1541, loss: 0.31898391246795654 2023-01-23 07:40:42.639834: step: 1570/1541, loss: 0.5710314512252808 2023-01-23 07:40:43.502779: step: 1572/1541, loss: 1.918823480606079 2023-01-23 07:40:44.355223: step: 1574/1541, loss: 0.5563428401947021 2023-01-23 07:40:45.269603: step: 1576/1541, loss: 0.5720658898353577 2023-01-23 07:40:46.119582: step: 1578/1541, loss: 0.7809340953826904 2023-01-23 07:40:47.010824: step: 1580/1541, loss: 0.2648933231830597 2023-01-23 07:40:47.914587: step: 1582/1541, loss: 0.410857617855072 2023-01-23 07:40:48.834882: step: 1584/1541, loss: 0.8163661956787109 2023-01-23 07:40:49.774190: step: 1586/1541, loss: 0.7654340267181396 2023-01-23 07:40:50.632831: step: 1588/1541, loss: 0.3004572093486786 2023-01-23 07:40:51.517978: step: 1590/1541, loss: 0.40549084544181824 2023-01-23 07:40:52.417202: step: 1592/1541, loss: 0.15213072299957275 2023-01-23 07:40:53.344080: step: 1594/1541, loss: 0.3027664124965668 2023-01-23 07:40:54.212356: step: 1596/1541, loss: 0.33833086490631104 2023-01-23 07:40:55.093687: step: 1598/1541, loss: 0.3103741407394409 2023-01-23 07:40:56.016954: step: 1600/1541, loss: 0.6985669136047363 2023-01-23 07:40:56.904432: step: 1602/1541, loss: 1.6687273979187012 2023-01-23 07:40:57.780021: step: 1604/1541, loss: 0.21348711848258972 2023-01-23 07:40:58.651041: step: 1606/1541, loss: 0.4532705545425415 2023-01-23 07:40:59.570906: step: 1608/1541, loss: 0.4606952965259552 2023-01-23 07:41:00.514927: step: 1610/1541, loss: 0.42469215393066406 2023-01-23 07:41:01.521591: step: 1612/1541, loss: 0.3437226414680481 2023-01-23 07:41:02.418835: step: 1614/1541, loss: 0.6915025115013123 2023-01-23 07:41:03.289791: step: 1616/1541, loss: 1.281031608581543 2023-01-23 07:41:04.164530: step: 1618/1541, loss: 0.1017153188586235 2023-01-23 07:41:05.068455: step: 1620/1541, loss: 0.8560947179794312 2023-01-23 07:41:05.970866: step: 1622/1541, loss: 0.12532447278499603 2023-01-23 07:41:06.899275: step: 1624/1541, loss: 0.32414254546165466 2023-01-23 07:41:07.784327: step: 1626/1541, loss: 0.5178545713424683 2023-01-23 07:41:08.713708: step: 1628/1541, loss: 0.45803302526474 2023-01-23 07:41:09.647185: step: 1630/1541, loss: 0.4447392225265503 2023-01-23 07:41:10.559500: step: 1632/1541, loss: 0.4354960322380066 2023-01-23 07:41:11.411865: step: 1634/1541, loss: 0.7226677536964417 2023-01-23 07:41:12.341499: step: 1636/1541, loss: 0.43609803915023804 2023-01-23 07:41:13.218600: step: 1638/1541, loss: 0.22875414788722992 2023-01-23 07:41:14.089031: step: 1640/1541, loss: 0.13523606956005096 2023-01-23 07:41:14.990624: step: 1642/1541, loss: 0.7990602850914001 2023-01-23 07:41:15.904497: step: 1644/1541, loss: 0.31313249468803406 2023-01-23 07:41:16.873586: step: 1646/1541, loss: 0.2819855213165283 2023-01-23 07:41:17.760007: step: 1648/1541, loss: 0.32781657576560974 2023-01-23 07:41:18.663734: step: 1650/1541, loss: 0.7360637784004211 2023-01-23 07:41:19.538445: step: 1652/1541, loss: 0.2901712656021118 2023-01-23 07:41:20.444620: step: 1654/1541, loss: 2.201932191848755 2023-01-23 07:41:21.350738: step: 1656/1541, loss: 1.615654706954956 2023-01-23 07:41:22.223483: step: 1658/1541, loss: 0.35233065485954285 2023-01-23 07:41:23.099795: step: 1660/1541, loss: 0.24419216811656952 2023-01-23 07:41:24.075442: step: 1662/1541, loss: 0.9921004772186279 2023-01-23 07:41:24.996499: step: 1664/1541, loss: 5.789556503295898 2023-01-23 07:41:25.916490: step: 1666/1541, loss: 0.3383881449699402 2023-01-23 07:41:26.818962: step: 1668/1541, loss: 0.4602433443069458 2023-01-23 07:41:27.782270: step: 1670/1541, loss: 0.6596420407295227 2023-01-23 07:41:28.657200: step: 1672/1541, loss: 5.564754486083984 2023-01-23 07:41:29.523105: step: 1674/1541, loss: 0.5499577522277832 2023-01-23 07:41:30.419915: step: 1676/1541, loss: 0.4784996211528778 2023-01-23 07:41:31.315169: step: 1678/1541, loss: 0.16125409305095673 2023-01-23 07:41:32.244262: step: 1680/1541, loss: 1.2044631242752075 2023-01-23 07:41:33.141054: step: 1682/1541, loss: 0.7629898190498352 2023-01-23 07:41:34.043599: step: 1684/1541, loss: 0.32251060009002686 2023-01-23 07:41:34.929672: step: 1686/1541, loss: 0.7302801609039307 2023-01-23 07:41:35.837671: step: 1688/1541, loss: 0.7595292329788208 2023-01-23 07:41:36.745459: step: 1690/1541, loss: 0.19095435738563538 2023-01-23 07:41:37.601860: step: 1692/1541, loss: 1.126164436340332 2023-01-23 07:41:38.516278: step: 1694/1541, loss: 0.3286440074443817 2023-01-23 07:41:39.424885: step: 1696/1541, loss: 0.7128986716270447 2023-01-23 07:41:40.308710: step: 1698/1541, loss: 0.4973658621311188 2023-01-23 07:41:41.179868: step: 1700/1541, loss: 0.20955026149749756 2023-01-23 07:41:42.106672: step: 1702/1541, loss: 0.23987101018428802 2023-01-23 07:41:43.006176: step: 1704/1541, loss: 0.6588242053985596 2023-01-23 07:41:43.938687: step: 1706/1541, loss: 0.4027997851371765 2023-01-23 07:41:44.867113: step: 1708/1541, loss: 0.5908401012420654 2023-01-23 07:41:45.767595: step: 1710/1541, loss: 0.896230161190033 2023-01-23 07:41:46.704154: step: 1712/1541, loss: 1.2422130107879639 2023-01-23 07:41:47.595384: step: 1714/1541, loss: 0.43976062536239624 2023-01-23 07:41:48.480787: step: 1716/1541, loss: 0.6974009871482849 2023-01-23 07:41:49.372712: step: 1718/1541, loss: 1.2038114070892334 2023-01-23 07:41:50.253418: step: 1720/1541, loss: 0.808393657207489 2023-01-23 07:41:51.110632: step: 1722/1541, loss: 0.31050392985343933 2023-01-23 07:41:52.063461: step: 1724/1541, loss: 0.5559132695198059 2023-01-23 07:41:52.965891: step: 1726/1541, loss: 0.5840342044830322 2023-01-23 07:41:53.962171: step: 1728/1541, loss: 1.9780733585357666 2023-01-23 07:41:54.855323: step: 1730/1541, loss: 0.20606349408626556 2023-01-23 07:41:55.748135: step: 1732/1541, loss: 0.4596444368362427 2023-01-23 07:41:56.627552: step: 1734/1541, loss: 0.15559157729148865 2023-01-23 07:41:57.543772: step: 1736/1541, loss: 0.19681721925735474 2023-01-23 07:41:58.474539: step: 1738/1541, loss: 6.145877838134766 2023-01-23 07:41:59.422803: step: 1740/1541, loss: 0.6210438013076782 2023-01-23 07:42:00.303629: step: 1742/1541, loss: 0.6264740824699402 2023-01-23 07:42:01.248971: step: 1744/1541, loss: 1.773123860359192 2023-01-23 07:42:02.160566: step: 1746/1541, loss: 0.3454970121383667 2023-01-23 07:42:03.094884: step: 1748/1541, loss: 0.2175564467906952 2023-01-23 07:42:03.983631: step: 1750/1541, loss: 0.6460862755775452 2023-01-23 07:42:04.890543: step: 1752/1541, loss: 0.9130722880363464 2023-01-23 07:42:05.808220: step: 1754/1541, loss: 0.10155247896909714 2023-01-23 07:42:06.690000: step: 1756/1541, loss: 1.0671299695968628 2023-01-23 07:42:07.613093: step: 1758/1541, loss: 0.3074726462364197 2023-01-23 07:42:08.475278: step: 1760/1541, loss: 0.5494972467422485 2023-01-23 07:42:09.348449: step: 1762/1541, loss: 0.4528439939022064 2023-01-23 07:42:10.328288: step: 1764/1541, loss: 0.4237462282180786 2023-01-23 07:42:11.224022: step: 1766/1541, loss: 0.46598321199417114 2023-01-23 07:42:12.148233: step: 1768/1541, loss: 3.6076951026916504 2023-01-23 07:42:13.092983: step: 1770/1541, loss: 0.9771192669868469 2023-01-23 07:42:14.012333: step: 1772/1541, loss: 0.5338645577430725 2023-01-23 07:42:15.015503: step: 1774/1541, loss: 0.8781645894050598 2023-01-23 07:42:15.956261: step: 1776/1541, loss: 0.9223822355270386 2023-01-23 07:42:16.863484: step: 1778/1541, loss: 0.7650018930435181 2023-01-23 07:42:17.756312: step: 1780/1541, loss: 0.31915104389190674 2023-01-23 07:42:18.621973: step: 1782/1541, loss: 0.5141400098800659 2023-01-23 07:42:19.550266: step: 1784/1541, loss: 0.1984638273715973 2023-01-23 07:42:20.477659: step: 1786/1541, loss: 0.5182473063468933 2023-01-23 07:42:21.378756: step: 1788/1541, loss: 3.1701600551605225 2023-01-23 07:42:22.252400: step: 1790/1541, loss: 0.4186931252479553 2023-01-23 07:42:23.214234: step: 1792/1541, loss: 0.47495773434638977 2023-01-23 07:42:24.086574: step: 1794/1541, loss: 0.4946298599243164 2023-01-23 07:42:24.979117: step: 1796/1541, loss: 0.21021610498428345 2023-01-23 07:42:25.884932: step: 1798/1541, loss: 0.6010640859603882 2023-01-23 07:42:26.783229: step: 1800/1541, loss: 0.8504059314727783 2023-01-23 07:42:27.731539: step: 1802/1541, loss: 1.3051040172576904 2023-01-23 07:42:28.603376: step: 1804/1541, loss: 0.8363548517227173 2023-01-23 07:42:29.491717: step: 1806/1541, loss: 1.1930220127105713 2023-01-23 07:42:30.452858: step: 1808/1541, loss: 1.166940450668335 2023-01-23 07:42:31.351474: step: 1810/1541, loss: 0.16268429160118103 2023-01-23 07:42:32.236058: step: 1812/1541, loss: 0.9066449403762817 2023-01-23 07:42:33.156319: step: 1814/1541, loss: 0.2815547287464142 2023-01-23 07:42:34.071851: step: 1816/1541, loss: 1.6792711019515991 2023-01-23 07:42:35.007548: step: 1818/1541, loss: 1.3685059547424316 2023-01-23 07:42:35.893444: step: 1820/1541, loss: 0.355087548494339 2023-01-23 07:42:36.799456: step: 1822/1541, loss: 0.8660469651222229 2023-01-23 07:42:37.664763: step: 1824/1541, loss: 0.4410354793071747 2023-01-23 07:42:38.567806: step: 1826/1541, loss: 0.6412169933319092 2023-01-23 07:42:39.441161: step: 1828/1541, loss: 0.2582867443561554 2023-01-23 07:42:40.338561: step: 1830/1541, loss: 0.4927237629890442 2023-01-23 07:42:41.236328: step: 1832/1541, loss: 0.6614893674850464 2023-01-23 07:42:42.169791: step: 1834/1541, loss: 0.292115181684494 2023-01-23 07:42:43.109323: step: 1836/1541, loss: 1.0301507711410522 2023-01-23 07:42:44.023590: step: 1838/1541, loss: 0.30186188220977783 2023-01-23 07:42:44.931163: step: 1840/1541, loss: 0.57594895362854 2023-01-23 07:42:45.820996: step: 1842/1541, loss: 0.44874998927116394 2023-01-23 07:42:46.711028: step: 1844/1541, loss: 0.5292829275131226 2023-01-23 07:42:47.613973: step: 1846/1541, loss: 0.42340975999832153 2023-01-23 07:42:48.498967: step: 1848/1541, loss: 0.3950263559818268 2023-01-23 07:42:49.495996: step: 1850/1541, loss: 0.3906857669353485 2023-01-23 07:42:50.374528: step: 1852/1541, loss: 1.5112637281417847 2023-01-23 07:42:51.393242: step: 1854/1541, loss: 0.5465317368507385 2023-01-23 07:42:52.357061: step: 1856/1541, loss: 0.5869617462158203 2023-01-23 07:42:53.245884: step: 1858/1541, loss: 0.5065699219703674 2023-01-23 07:42:54.118298: step: 1860/1541, loss: 0.13098500669002533 2023-01-23 07:42:55.006085: step: 1862/1541, loss: 0.5376110076904297 2023-01-23 07:42:55.915481: step: 1864/1541, loss: 0.3426596224308014 2023-01-23 07:42:56.803207: step: 1866/1541, loss: 0.5097033977508545 2023-01-23 07:42:57.705008: step: 1868/1541, loss: 0.4617578089237213 2023-01-23 07:42:58.591969: step: 1870/1541, loss: 0.45811551809310913 2023-01-23 07:42:59.507239: step: 1872/1541, loss: 0.5816367864608765 2023-01-23 07:43:00.373921: step: 1874/1541, loss: 0.6765580773353577 2023-01-23 07:43:01.262593: step: 1876/1541, loss: 2.8664870262145996 2023-01-23 07:43:02.243028: step: 1878/1541, loss: 0.843685507774353 2023-01-23 07:43:03.108481: step: 1880/1541, loss: 0.6726648807525635 2023-01-23 07:43:04.033304: step: 1882/1541, loss: 0.29700714349746704 2023-01-23 07:43:04.915607: step: 1884/1541, loss: 0.8103625774383545 2023-01-23 07:43:05.869891: step: 1886/1541, loss: 0.8184390664100647 2023-01-23 07:43:06.839750: step: 1888/1541, loss: 0.35479849576950073 2023-01-23 07:43:07.743314: step: 1890/1541, loss: 1.4569694995880127 2023-01-23 07:43:08.593435: step: 1892/1541, loss: 0.4612143933773041 2023-01-23 07:43:09.464851: step: 1894/1541, loss: 0.7124918699264526 2023-01-23 07:43:10.386037: step: 1896/1541, loss: 0.6138018369674683 2023-01-23 07:43:11.309487: step: 1898/1541, loss: 0.39116576313972473 2023-01-23 07:43:12.182614: step: 1900/1541, loss: 0.5978565812110901 2023-01-23 07:43:13.121648: step: 1902/1541, loss: 0.8814130425453186 2023-01-23 07:43:14.023233: step: 1904/1541, loss: 0.4212057590484619 2023-01-23 07:43:14.925667: step: 1906/1541, loss: 0.11431492865085602 2023-01-23 07:43:15.822425: step: 1908/1541, loss: 0.25627294182777405 2023-01-23 07:43:16.709743: step: 1910/1541, loss: 0.5438554883003235 2023-01-23 07:43:17.577075: step: 1912/1541, loss: 0.8317638039588928 2023-01-23 07:43:18.476562: step: 1914/1541, loss: 0.7788376212120056 2023-01-23 07:43:19.376276: step: 1916/1541, loss: 0.5430362224578857 2023-01-23 07:43:20.261140: step: 1918/1541, loss: 0.24590010941028595 2023-01-23 07:43:21.151401: step: 1920/1541, loss: 0.7153972387313843 2023-01-23 07:43:22.048364: step: 1922/1541, loss: 0.5010850429534912 2023-01-23 07:43:22.927315: step: 1924/1541, loss: 0.6590531468391418 2023-01-23 07:43:23.764320: step: 1926/1541, loss: 0.47396785020828247 2023-01-23 07:43:24.648688: step: 1928/1541, loss: 0.16933603584766388 2023-01-23 07:43:25.627151: step: 1930/1541, loss: 1.9608027935028076 2023-01-23 07:43:26.526286: step: 1932/1541, loss: 0.3757156729698181 2023-01-23 07:43:27.452667: step: 1934/1541, loss: 0.6985293030738831 2023-01-23 07:43:28.326291: step: 1936/1541, loss: 0.7588340044021606 2023-01-23 07:43:29.222416: step: 1938/1541, loss: 0.6213301420211792 2023-01-23 07:43:30.142197: step: 1940/1541, loss: 0.8532950282096863 2023-01-23 07:43:30.998178: step: 1942/1541, loss: 0.3770001232624054 2023-01-23 07:43:31.900395: step: 1944/1541, loss: 0.7138398289680481 2023-01-23 07:43:32.859789: step: 1946/1541, loss: 0.6911877989768982 2023-01-23 07:43:33.813349: step: 1948/1541, loss: 0.2780746817588806 2023-01-23 07:43:34.806043: step: 1950/1541, loss: 0.21034617722034454 2023-01-23 07:43:35.681721: step: 1952/1541, loss: 0.28796839714050293 2023-01-23 07:43:36.562156: step: 1954/1541, loss: 0.3092373013496399 2023-01-23 07:43:37.485383: step: 1956/1541, loss: 0.43689221143722534 2023-01-23 07:43:38.387072: step: 1958/1541, loss: 0.2802369296550751 2023-01-23 07:43:39.281305: step: 1960/1541, loss: 0.2668226957321167 2023-01-23 07:43:40.214594: step: 1962/1541, loss: 1.1262402534484863 2023-01-23 07:43:41.116350: step: 1964/1541, loss: 0.3157096207141876 2023-01-23 07:43:42.104730: step: 1966/1541, loss: 0.2219172716140747 2023-01-23 07:43:42.957712: step: 1968/1541, loss: 0.35385188460350037 2023-01-23 07:43:43.919595: step: 1970/1541, loss: 0.28604856133461 2023-01-23 07:43:44.828724: step: 1972/1541, loss: 0.7322498559951782 2023-01-23 07:43:45.758231: step: 1974/1541, loss: 0.37419483065605164 2023-01-23 07:43:46.662175: step: 1976/1541, loss: 1.1844884157180786 2023-01-23 07:43:47.609605: step: 1978/1541, loss: 0.5609995722770691 2023-01-23 07:43:48.492456: step: 1980/1541, loss: 0.8219643235206604 2023-01-23 07:43:49.380651: step: 1982/1541, loss: 0.574321985244751 2023-01-23 07:43:50.300397: step: 1984/1541, loss: 0.22759869694709778 2023-01-23 07:43:51.228476: step: 1986/1541, loss: 1.1268138885498047 2023-01-23 07:43:52.195667: step: 1988/1541, loss: 1.1557213068008423 2023-01-23 07:43:53.224869: step: 1990/1541, loss: 3.5978000164031982 2023-01-23 07:43:54.118597: step: 1992/1541, loss: 0.37134653329849243 2023-01-23 07:43:55.039772: step: 1994/1541, loss: 0.6320157051086426 2023-01-23 07:43:55.937161: step: 1996/1541, loss: 0.5841609239578247 2023-01-23 07:43:56.825705: step: 1998/1541, loss: 0.4809640645980835 2023-01-23 07:43:57.709239: step: 2000/1541, loss: 0.31978195905685425 2023-01-23 07:43:58.619641: step: 2002/1541, loss: 0.28415751457214355 2023-01-23 07:43:59.491097: step: 2004/1541, loss: 0.4370911717414856 2023-01-23 07:44:00.366129: step: 2006/1541, loss: 0.9792312383651733 2023-01-23 07:44:01.246998: step: 2008/1541, loss: 0.4896540641784668 2023-01-23 07:44:02.176894: step: 2010/1541, loss: 0.5530422329902649 2023-01-23 07:44:03.099403: step: 2012/1541, loss: 0.501767098903656 2023-01-23 07:44:03.981118: step: 2014/1541, loss: 0.5961095690727234 2023-01-23 07:44:04.981721: step: 2016/1541, loss: 0.3285530209541321 2023-01-23 07:44:05.869183: step: 2018/1541, loss: 0.5313197374343872 2023-01-23 07:44:06.746587: step: 2020/1541, loss: 0.7122678160667419 2023-01-23 07:44:07.645814: step: 2022/1541, loss: 0.40945520997047424 2023-01-23 07:44:08.533691: step: 2024/1541, loss: 0.5401651263237 2023-01-23 07:44:09.430923: step: 2026/1541, loss: 0.38006874918937683 2023-01-23 07:44:10.321920: step: 2028/1541, loss: 0.2986798882484436 2023-01-23 07:44:11.204707: step: 2030/1541, loss: 0.6191384196281433 2023-01-23 07:44:12.078704: step: 2032/1541, loss: 0.2818322479724884 2023-01-23 07:44:13.037173: step: 2034/1541, loss: 0.13613861799240112 2023-01-23 07:44:13.900359: step: 2036/1541, loss: 0.36990728974342346 2023-01-23 07:44:14.787308: step: 2038/1541, loss: 0.36774516105651855 2023-01-23 07:44:15.716685: step: 2040/1541, loss: 0.8260318636894226 2023-01-23 07:44:16.633299: step: 2042/1541, loss: 0.28167402744293213 2023-01-23 07:44:17.516768: step: 2044/1541, loss: 0.8935283422470093 2023-01-23 07:44:18.382764: step: 2046/1541, loss: 0.34172406792640686 2023-01-23 07:44:19.267997: step: 2048/1541, loss: 0.48828649520874023 2023-01-23 07:44:20.133015: step: 2050/1541, loss: 0.39945751428604126 2023-01-23 07:44:21.047151: step: 2052/1541, loss: 0.8293439745903015 2023-01-23 07:44:21.931434: step: 2054/1541, loss: 0.6242957711219788 2023-01-23 07:44:22.839684: step: 2056/1541, loss: 0.26336637139320374 2023-01-23 07:44:23.703306: step: 2058/1541, loss: 0.18462279438972473 2023-01-23 07:44:24.576176: step: 2060/1541, loss: 0.2465660274028778 2023-01-23 07:44:25.489653: step: 2062/1541, loss: 1.3769887685775757 2023-01-23 07:44:26.401913: step: 2064/1541, loss: 4.938366413116455 2023-01-23 07:44:27.287450: step: 2066/1541, loss: 0.6231110692024231 2023-01-23 07:44:28.192266: step: 2068/1541, loss: 1.3646267652511597 2023-01-23 07:44:29.124808: step: 2070/1541, loss: 0.5012185573577881 2023-01-23 07:44:30.006792: step: 2072/1541, loss: 0.44385063648223877 2023-01-23 07:44:30.874399: step: 2074/1541, loss: 0.6012184023857117 2023-01-23 07:44:31.724017: step: 2076/1541, loss: 0.4074247181415558 2023-01-23 07:44:32.611456: step: 2078/1541, loss: 0.48283690214157104 2023-01-23 07:44:33.584042: step: 2080/1541, loss: 0.4333173930644989 2023-01-23 07:44:34.490890: step: 2082/1541, loss: 0.49859729409217834 2023-01-23 07:44:35.351881: step: 2084/1541, loss: 0.591911256313324 2023-01-23 07:44:36.207282: step: 2086/1541, loss: 0.48742759227752686 2023-01-23 07:44:37.082712: step: 2088/1541, loss: 0.37059134244918823 2023-01-23 07:44:37.952587: step: 2090/1541, loss: 0.3389706015586853 2023-01-23 07:44:38.853127: step: 2092/1541, loss: 0.6776812076568604 2023-01-23 07:44:39.762037: step: 2094/1541, loss: 0.3373723328113556 2023-01-23 07:44:40.634135: step: 2096/1541, loss: 0.4138197600841522 2023-01-23 07:44:41.533960: step: 2098/1541, loss: 0.23077982664108276 2023-01-23 07:44:42.431386: step: 2100/1541, loss: 0.38590607047080994 2023-01-23 07:44:43.335042: step: 2102/1541, loss: 0.5323180556297302 2023-01-23 07:44:44.236446: step: 2104/1541, loss: 0.19357557594776154 2023-01-23 07:44:45.088157: step: 2106/1541, loss: 0.14994385838508606 2023-01-23 07:44:45.974436: step: 2108/1541, loss: 0.733130156993866 2023-01-23 07:44:46.874503: step: 2110/1541, loss: 0.6588382124900818 2023-01-23 07:44:47.759539: step: 2112/1541, loss: 2.3025333881378174 2023-01-23 07:44:48.654087: step: 2114/1541, loss: 0.6714917421340942 2023-01-23 07:44:49.527163: step: 2116/1541, loss: 0.23417216539382935 2023-01-23 07:44:50.442202: step: 2118/1541, loss: 0.501509428024292 2023-01-23 07:44:51.365316: step: 2120/1541, loss: 0.6105886101722717 2023-01-23 07:44:52.214671: step: 2122/1541, loss: 0.8168858885765076 2023-01-23 07:44:53.141218: step: 2124/1541, loss: 0.3690465986728668 2023-01-23 07:44:54.017960: step: 2126/1541, loss: 0.5129752159118652 2023-01-23 07:44:54.917666: step: 2128/1541, loss: 0.879729688167572 2023-01-23 07:44:55.830904: step: 2130/1541, loss: 0.5433777570724487 2023-01-23 07:44:56.701232: step: 2132/1541, loss: 0.3671041429042816 2023-01-23 07:44:57.642529: step: 2134/1541, loss: 0.5811433792114258 2023-01-23 07:44:58.532660: step: 2136/1541, loss: 0.7223327159881592 2023-01-23 07:44:59.385997: step: 2138/1541, loss: 0.3696821331977844 2023-01-23 07:45:00.402386: step: 2140/1541, loss: 0.5526097416877747 2023-01-23 07:45:01.305295: step: 2142/1541, loss: 0.31707727909088135 2023-01-23 07:45:02.228447: step: 2144/1541, loss: 1.6041951179504395 2023-01-23 07:45:03.143186: step: 2146/1541, loss: 0.5504120588302612 2023-01-23 07:45:04.058498: step: 2148/1541, loss: 0.9219138026237488 2023-01-23 07:45:04.977253: step: 2150/1541, loss: 0.3412356972694397 2023-01-23 07:45:05.853925: step: 2152/1541, loss: 0.29843688011169434 2023-01-23 07:45:06.716778: step: 2154/1541, loss: 0.6634551286697388 2023-01-23 07:45:07.628007: step: 2156/1541, loss: 0.6644515991210938 2023-01-23 07:45:08.549662: step: 2158/1541, loss: 0.2986348271369934 2023-01-23 07:45:09.421112: step: 2160/1541, loss: 0.346907377243042 2023-01-23 07:45:10.297371: step: 2162/1541, loss: 0.6158930063247681 2023-01-23 07:45:11.202548: step: 2164/1541, loss: 0.3768528401851654 2023-01-23 07:45:12.148822: step: 2166/1541, loss: 0.3486008048057556 2023-01-23 07:45:13.082293: step: 2168/1541, loss: 0.527012288570404 2023-01-23 07:45:13.964641: step: 2170/1541, loss: 0.364910751581192 2023-01-23 07:45:14.893385: step: 2172/1541, loss: 0.4951816201210022 2023-01-23 07:45:15.801924: step: 2174/1541, loss: 0.14842495322227478 2023-01-23 07:45:16.749137: step: 2176/1541, loss: 0.3224737346172333 2023-01-23 07:45:17.706712: step: 2178/1541, loss: 0.35562413930892944 2023-01-23 07:45:18.557440: step: 2180/1541, loss: 0.9541021585464478 2023-01-23 07:45:19.442213: step: 2182/1541, loss: 1.5730279684066772 2023-01-23 07:45:20.320080: step: 2184/1541, loss: 0.31351813673973083 2023-01-23 07:45:21.248242: step: 2186/1541, loss: 0.5864546298980713 2023-01-23 07:45:22.212849: step: 2188/1541, loss: 0.32286232709884644 2023-01-23 07:45:23.071948: step: 2190/1541, loss: 3.0458335876464844 2023-01-23 07:45:23.989873: step: 2192/1541, loss: 0.48389071226119995 2023-01-23 07:45:24.888646: step: 2194/1541, loss: 0.4111628234386444 2023-01-23 07:45:25.794187: step: 2196/1541, loss: 0.68147212266922 2023-01-23 07:45:26.706640: step: 2198/1541, loss: 0.6486172676086426 2023-01-23 07:45:27.587691: step: 2200/1541, loss: 0.7520788908004761 2023-01-23 07:45:28.475158: step: 2202/1541, loss: 0.31319621205329895 2023-01-23 07:45:29.362249: step: 2204/1541, loss: 0.28722527623176575 2023-01-23 07:45:30.243009: step: 2206/1541, loss: 0.9534121751785278 2023-01-23 07:45:31.108552: step: 2208/1541, loss: 0.188206285238266 2023-01-23 07:45:32.009540: step: 2210/1541, loss: 0.33697232604026794 2023-01-23 07:45:32.937635: step: 2212/1541, loss: 0.6362572312355042 2023-01-23 07:45:33.836313: step: 2214/1541, loss: 0.20995117723941803 2023-01-23 07:45:34.718303: step: 2216/1541, loss: 1.2235164642333984 2023-01-23 07:45:35.626952: step: 2218/1541, loss: 0.5402291417121887 2023-01-23 07:45:36.584774: step: 2220/1541, loss: 0.5931715965270996 2023-01-23 07:45:37.535604: step: 2222/1541, loss: 0.8796653747558594 2023-01-23 07:45:38.468729: step: 2224/1541, loss: 0.49936944246292114 2023-01-23 07:45:39.400752: step: 2226/1541, loss: 0.25694572925567627 2023-01-23 07:45:40.349256: step: 2228/1541, loss: 0.5360348224639893 2023-01-23 07:45:41.245502: step: 2230/1541, loss: 0.25441253185272217 2023-01-23 07:45:42.135301: step: 2232/1541, loss: 0.6043915748596191 2023-01-23 07:45:43.022734: step: 2234/1541, loss: 0.28921937942504883 2023-01-23 07:45:43.934156: step: 2236/1541, loss: 0.6184629201889038 2023-01-23 07:45:44.859400: step: 2238/1541, loss: 0.3172924816608429 2023-01-23 07:45:45.763147: step: 2240/1541, loss: 0.7732107043266296 2023-01-23 07:45:46.659478: step: 2242/1541, loss: 0.6262187957763672 2023-01-23 07:45:47.555376: step: 2244/1541, loss: 0.27435359358787537 2023-01-23 07:45:48.442011: step: 2246/1541, loss: 1.5382922887802124 2023-01-23 07:45:49.387630: step: 2248/1541, loss: 0.4309486746788025 2023-01-23 07:45:50.288548: step: 2250/1541, loss: 1.293752670288086 2023-01-23 07:45:51.229506: step: 2252/1541, loss: 0.7434728741645813 2023-01-23 07:45:52.119080: step: 2254/1541, loss: 0.29091984033584595 2023-01-23 07:45:52.997677: step: 2256/1541, loss: 0.9924074411392212 2023-01-23 07:45:53.935825: step: 2258/1541, loss: 0.9667911529541016 2023-01-23 07:45:54.917556: step: 2260/1541, loss: 0.22365036606788635 2023-01-23 07:45:55.860732: step: 2262/1541, loss: 0.2367747575044632 2023-01-23 07:45:56.744964: step: 2264/1541, loss: 0.8383045792579651 2023-01-23 07:45:57.623457: step: 2266/1541, loss: 0.5761051177978516 2023-01-23 07:45:58.534114: step: 2268/1541, loss: 0.9582459926605225 2023-01-23 07:45:59.478406: step: 2270/1541, loss: 0.2485913634300232 2023-01-23 07:46:00.332600: step: 2272/1541, loss: 0.3290536403656006 2023-01-23 07:46:01.242341: step: 2274/1541, loss: 0.6106522679328918 2023-01-23 07:46:02.139670: step: 2276/1541, loss: 0.8581429123878479 2023-01-23 07:46:03.035207: step: 2278/1541, loss: 0.5930966734886169 2023-01-23 07:46:03.896638: step: 2280/1541, loss: 2.2508013248443604 2023-01-23 07:46:04.835200: step: 2282/1541, loss: 0.8818466663360596 2023-01-23 07:46:05.754876: step: 2284/1541, loss: 0.41809216141700745 2023-01-23 07:46:06.662336: step: 2286/1541, loss: 0.54686039686203 2023-01-23 07:46:07.593585: step: 2288/1541, loss: 0.2845124900341034 2023-01-23 07:46:08.477859: step: 2290/1541, loss: 0.5565572381019592 2023-01-23 07:46:09.389120: step: 2292/1541, loss: 0.2844977080821991 2023-01-23 07:46:10.279342: step: 2294/1541, loss: 0.4609750211238861 2023-01-23 07:46:11.229259: step: 2296/1541, loss: 0.6035617589950562 2023-01-23 07:46:12.126484: step: 2298/1541, loss: 0.904788613319397 2023-01-23 07:46:13.094402: step: 2300/1541, loss: 0.5786496996879578 2023-01-23 07:46:13.981021: step: 2302/1541, loss: 0.3821730613708496 2023-01-23 07:46:14.890131: step: 2304/1541, loss: 1.8671455383300781 2023-01-23 07:46:15.781044: step: 2306/1541, loss: 0.46152263879776 2023-01-23 07:46:16.685314: step: 2308/1541, loss: 0.7130335569381714 2023-01-23 07:46:17.554887: step: 2310/1541, loss: 0.3251636326313019 2023-01-23 07:46:18.422734: step: 2312/1541, loss: 1.1854100227355957 2023-01-23 07:46:19.325713: step: 2314/1541, loss: 0.38959571719169617 2023-01-23 07:46:20.204685: step: 2316/1541, loss: 0.41056156158447266 2023-01-23 07:46:21.139619: step: 2318/1541, loss: 0.32733333110809326 2023-01-23 07:46:22.043929: step: 2320/1541, loss: 0.5987414121627808 2023-01-23 07:46:22.965362: step: 2322/1541, loss: 0.45869868993759155 2023-01-23 07:46:23.858033: step: 2324/1541, loss: 0.8425505757331848 2023-01-23 07:46:24.715769: step: 2326/1541, loss: 0.6095768213272095 2023-01-23 07:46:25.629243: step: 2328/1541, loss: 0.8077930808067322 2023-01-23 07:46:26.568654: step: 2330/1541, loss: 0.519715428352356 2023-01-23 07:46:27.549538: step: 2332/1541, loss: 0.49635693430900574 2023-01-23 07:46:28.515133: step: 2334/1541, loss: 0.6946970820426941 2023-01-23 07:46:29.388042: step: 2336/1541, loss: 0.2276373654603958 2023-01-23 07:46:30.280312: step: 2338/1541, loss: 0.6693138480186462 2023-01-23 07:46:31.203824: step: 2340/1541, loss: 1.173377513885498 2023-01-23 07:46:32.086703: step: 2342/1541, loss: 0.1266472190618515 2023-01-23 07:46:32.963513: step: 2344/1541, loss: 0.9910854697227478 2023-01-23 07:46:33.857941: step: 2346/1541, loss: 0.4293586015701294 2023-01-23 07:46:34.835065: step: 2348/1541, loss: 0.2317144125699997 2023-01-23 07:46:35.772520: step: 2350/1541, loss: 0.17591947317123413 2023-01-23 07:46:36.634869: step: 2352/1541, loss: 0.4201642572879791 2023-01-23 07:46:37.528709: step: 2354/1541, loss: 0.43200987577438354 2023-01-23 07:46:38.404319: step: 2356/1541, loss: 0.15007607638835907 2023-01-23 07:46:39.315284: step: 2358/1541, loss: 0.21705077588558197 2023-01-23 07:46:40.304962: step: 2360/1541, loss: 0.33947131037712097 2023-01-23 07:46:41.177187: step: 2362/1541, loss: 0.45264697074890137 2023-01-23 07:46:42.062056: step: 2364/1541, loss: 0.6817949414253235 2023-01-23 07:46:42.945158: step: 2366/1541, loss: 0.23796576261520386 2023-01-23 07:46:43.844582: step: 2368/1541, loss: 0.6107715964317322 2023-01-23 07:46:44.713617: step: 2370/1541, loss: 0.753879964351654 2023-01-23 07:46:45.614494: step: 2372/1541, loss: 1.635825753211975 2023-01-23 07:46:46.533379: step: 2374/1541, loss: 0.7957173585891724 2023-01-23 07:46:47.411962: step: 2376/1541, loss: 1.5120786428451538 2023-01-23 07:46:48.290865: step: 2378/1541, loss: 0.2999172508716583 2023-01-23 07:46:49.187964: step: 2380/1541, loss: 0.2623540163040161 2023-01-23 07:46:50.097997: step: 2382/1541, loss: 0.42221421003341675 2023-01-23 07:46:50.996956: step: 2384/1541, loss: 0.4745548665523529 2023-01-23 07:46:51.979205: step: 2386/1541, loss: 0.6459061503410339 2023-01-23 07:46:52.880714: step: 2388/1541, loss: 0.27937254309654236 2023-01-23 07:46:53.732641: step: 2390/1541, loss: 0.2561326026916504 2023-01-23 07:46:54.625285: step: 2392/1541, loss: 0.43015503883361816 2023-01-23 07:46:55.488357: step: 2394/1541, loss: 0.22806058824062347 2023-01-23 07:46:56.377431: step: 2396/1541, loss: 0.2827138304710388 2023-01-23 07:46:57.283006: step: 2398/1541, loss: 0.5997292399406433 2023-01-23 07:46:58.154326: step: 2400/1541, loss: 0.4585961401462555 2023-01-23 07:46:59.098600: step: 2402/1541, loss: 0.6435250639915466 2023-01-23 07:46:59.983641: step: 2404/1541, loss: 0.5780887603759766 2023-01-23 07:47:00.887770: step: 2406/1541, loss: 0.7975496053695679 2023-01-23 07:47:01.770435: step: 2408/1541, loss: 1.5633654594421387 2023-01-23 07:47:02.673525: step: 2410/1541, loss: 0.29638218879699707 2023-01-23 07:47:03.531366: step: 2412/1541, loss: 1.0325651168823242 2023-01-23 07:47:04.423293: step: 2414/1541, loss: 0.40138474106788635 2023-01-23 07:47:05.269164: step: 2416/1541, loss: 0.44053056836128235 2023-01-23 07:47:06.168253: step: 2418/1541, loss: 0.3867958188056946 2023-01-23 07:47:07.122732: step: 2420/1541, loss: 0.5312348008155823 2023-01-23 07:47:07.957544: step: 2422/1541, loss: 0.3182715177536011 2023-01-23 07:47:08.816846: step: 2424/1541, loss: 1.218884825706482 2023-01-23 07:47:09.731603: step: 2426/1541, loss: 0.4880599081516266 2023-01-23 07:47:10.609481: step: 2428/1541, loss: 0.33166956901550293 2023-01-23 07:47:11.467722: step: 2430/1541, loss: 0.20079296827316284 2023-01-23 07:47:12.351024: step: 2432/1541, loss: 1.0963107347488403 2023-01-23 07:47:13.340975: step: 2434/1541, loss: 1.3988749980926514 2023-01-23 07:47:14.207163: step: 2436/1541, loss: 1.8013558387756348 2023-01-23 07:47:15.097816: step: 2438/1541, loss: 0.5373835563659668 2023-01-23 07:47:16.036277: step: 2440/1541, loss: 0.3849133551120758 2023-01-23 07:47:16.873431: step: 2442/1541, loss: 1.4432027339935303 2023-01-23 07:47:17.724401: step: 2444/1541, loss: 0.6232843399047852 2023-01-23 07:47:18.590954: step: 2446/1541, loss: 0.18729738891124725 2023-01-23 07:47:19.549300: step: 2448/1541, loss: 0.1971040964126587 2023-01-23 07:47:20.418596: step: 2450/1541, loss: 0.5734424591064453 2023-01-23 07:47:21.323159: step: 2452/1541, loss: 0.8974398970603943 2023-01-23 07:47:22.198390: step: 2454/1541, loss: 0.3670901954174042 2023-01-23 07:47:23.090443: step: 2456/1541, loss: 0.9547520279884338 2023-01-23 07:47:23.998899: step: 2458/1541, loss: 0.44365614652633667 2023-01-23 07:47:24.906466: step: 2460/1541, loss: 0.6221643686294556 2023-01-23 07:47:25.833243: step: 2462/1541, loss: 0.5835106372833252 2023-01-23 07:47:26.727926: step: 2464/1541, loss: 0.41811442375183105 2023-01-23 07:47:27.627114: step: 2466/1541, loss: 0.3281129002571106 2023-01-23 07:47:28.507315: step: 2468/1541, loss: 2.658700942993164 2023-01-23 07:47:29.375330: step: 2470/1541, loss: 0.5695252418518066 2023-01-23 07:47:30.268711: step: 2472/1541, loss: 0.38377901911735535 2023-01-23 07:47:31.165341: step: 2474/1541, loss: 3.409857988357544 2023-01-23 07:47:32.079447: step: 2476/1541, loss: 0.8149561285972595 2023-01-23 07:47:33.089121: step: 2478/1541, loss: 0.3633635342121124 2023-01-23 07:47:33.994680: step: 2480/1541, loss: 0.6450610160827637 2023-01-23 07:47:34.893624: step: 2482/1541, loss: 0.6548206210136414 2023-01-23 07:47:35.818720: step: 2484/1541, loss: 0.19018378853797913 2023-01-23 07:47:36.702095: step: 2486/1541, loss: 0.4004094898700714 2023-01-23 07:47:37.709161: step: 2488/1541, loss: 0.05881465598940849 2023-01-23 07:47:38.622377: step: 2490/1541, loss: 1.5799211263656616 2023-01-23 07:47:39.563615: step: 2492/1541, loss: 0.580580472946167 2023-01-23 07:47:40.443361: step: 2494/1541, loss: 0.5836788415908813 2023-01-23 07:47:41.302446: step: 2496/1541, loss: 1.8130683898925781 2023-01-23 07:47:42.168183: step: 2498/1541, loss: 2.8573782444000244 2023-01-23 07:47:43.089346: step: 2500/1541, loss: 0.26112112402915955 2023-01-23 07:47:44.019797: step: 2502/1541, loss: 0.5092446208000183 2023-01-23 07:47:44.935902: step: 2504/1541, loss: 0.5330967903137207 2023-01-23 07:47:45.836084: step: 2506/1541, loss: 0.6685094833374023 2023-01-23 07:47:46.740637: step: 2508/1541, loss: 0.5803960561752319 2023-01-23 07:47:47.632562: step: 2510/1541, loss: 0.6456398367881775 2023-01-23 07:47:48.529766: step: 2512/1541, loss: 0.19412089884281158 2023-01-23 07:47:49.448392: step: 2514/1541, loss: 0.766574501991272 2023-01-23 07:47:50.362906: step: 2516/1541, loss: 0.8391995429992676 2023-01-23 07:47:51.254152: step: 2518/1541, loss: 0.8298868536949158 2023-01-23 07:47:52.133171: step: 2520/1541, loss: 0.1627437323331833 2023-01-23 07:47:53.075021: step: 2522/1541, loss: 0.6375270485877991 2023-01-23 07:47:53.973142: step: 2524/1541, loss: 0.22804497182369232 2023-01-23 07:47:54.864930: step: 2526/1541, loss: 0.542543888092041 2023-01-23 07:47:55.759739: step: 2528/1541, loss: 1.433788776397705 2023-01-23 07:47:56.638513: step: 2530/1541, loss: 0.44640740752220154 2023-01-23 07:47:57.538525: step: 2532/1541, loss: 0.43025681376457214 2023-01-23 07:47:58.462998: step: 2534/1541, loss: 0.24571484327316284 2023-01-23 07:47:59.375294: step: 2536/1541, loss: 0.46485602855682373 2023-01-23 07:48:00.277120: step: 2538/1541, loss: 0.2661459147930145 2023-01-23 07:48:01.161519: step: 2540/1541, loss: 0.5421260595321655 2023-01-23 07:48:02.048388: step: 2542/1541, loss: 0.38150906562805176 2023-01-23 07:48:02.957701: step: 2544/1541, loss: 0.18406787514686584 2023-01-23 07:48:03.848767: step: 2546/1541, loss: 0.4440987706184387 2023-01-23 07:48:04.722862: step: 2548/1541, loss: 0.5075664520263672 2023-01-23 07:48:05.592283: step: 2550/1541, loss: 0.5971873998641968 2023-01-23 07:48:06.510352: step: 2552/1541, loss: 0.6478073000907898 2023-01-23 07:48:07.370445: step: 2554/1541, loss: 0.3903367221355438 2023-01-23 07:48:08.264328: step: 2556/1541, loss: 3.025042772293091 2023-01-23 07:48:09.106762: step: 2558/1541, loss: 0.38085034489631653 2023-01-23 07:48:10.018404: step: 2560/1541, loss: 0.8969613909721375 2023-01-23 07:48:10.933264: step: 2562/1541, loss: 0.46213212609291077 2023-01-23 07:48:11.856686: step: 2564/1541, loss: 0.33423665165901184 2023-01-23 07:48:12.790780: step: 2566/1541, loss: 0.31715303659439087 2023-01-23 07:48:13.718017: step: 2568/1541, loss: 0.21123430132865906 2023-01-23 07:48:14.683374: step: 2570/1541, loss: 0.25076618790626526 2023-01-23 07:48:15.581565: step: 2572/1541, loss: 0.9905939698219299 2023-01-23 07:48:16.522253: step: 2574/1541, loss: 0.37290528416633606 2023-01-23 07:48:17.414064: step: 2576/1541, loss: 0.4429464340209961 2023-01-23 07:48:18.330510: step: 2578/1541, loss: 0.4627114236354828 2023-01-23 07:48:19.266252: step: 2580/1541, loss: 0.41424453258514404 2023-01-23 07:48:20.142990: step: 2582/1541, loss: 0.2591556906700134 2023-01-23 07:48:21.001239: step: 2584/1541, loss: 3.17602801322937 2023-01-23 07:48:21.885004: step: 2586/1541, loss: 0.24383369088172913 2023-01-23 07:48:22.785159: step: 2588/1541, loss: 0.8951126933097839 2023-01-23 07:48:23.685484: step: 2590/1541, loss: 0.30006664991378784 2023-01-23 07:48:24.603900: step: 2592/1541, loss: 7.148742198944092 2023-01-23 07:48:25.477515: step: 2594/1541, loss: 0.29983770847320557 2023-01-23 07:48:26.434615: step: 2596/1541, loss: 0.8318884372711182 2023-01-23 07:48:27.312568: step: 2598/1541, loss: 6.686332702636719 2023-01-23 07:48:28.168483: step: 2600/1541, loss: 0.5490114092826843 2023-01-23 07:48:29.086707: step: 2602/1541, loss: 0.27872738242149353 2023-01-23 07:48:29.948277: step: 2604/1541, loss: 0.2965629994869232 2023-01-23 07:48:30.809802: step: 2606/1541, loss: 0.2904204726219177 2023-01-23 07:48:31.735878: step: 2608/1541, loss: 2.769334554672241 2023-01-23 07:48:32.609834: step: 2610/1541, loss: 0.1499885618686676 2023-01-23 07:48:33.513129: step: 2612/1541, loss: 0.6938329935073853 2023-01-23 07:48:34.407834: step: 2614/1541, loss: 2.540172576904297 2023-01-23 07:48:35.262159: step: 2616/1541, loss: 0.9211788773536682 2023-01-23 07:48:36.257375: step: 2618/1541, loss: 0.6909791231155396 2023-01-23 07:48:37.168265: step: 2620/1541, loss: 0.5834067463874817 2023-01-23 07:48:38.074318: step: 2622/1541, loss: 2.803541898727417 2023-01-23 07:48:38.958268: step: 2624/1541, loss: 0.9557598233222961 2023-01-23 07:48:39.922772: step: 2626/1541, loss: 0.46531492471694946 2023-01-23 07:48:40.796571: step: 2628/1541, loss: 2.994083881378174 2023-01-23 07:48:41.749997: step: 2630/1541, loss: 0.7624329924583435 2023-01-23 07:48:42.623294: step: 2632/1541, loss: 0.5804133415222168 2023-01-23 07:48:43.513006: step: 2634/1541, loss: 0.7132129669189453 2023-01-23 07:48:44.407381: step: 2636/1541, loss: 0.13404716551303864 2023-01-23 07:48:45.299713: step: 2638/1541, loss: 0.43742769956588745 2023-01-23 07:48:46.199452: step: 2640/1541, loss: 1.2374625205993652 2023-01-23 07:48:47.093836: step: 2642/1541, loss: 0.3960713744163513 2023-01-23 07:48:48.094252: step: 2644/1541, loss: 0.7798621654510498 2023-01-23 07:48:49.033571: step: 2646/1541, loss: 1.759106993675232 2023-01-23 07:48:49.887277: step: 2648/1541, loss: 0.829462468624115 2023-01-23 07:48:50.852263: step: 2650/1541, loss: 7.881776332855225 2023-01-23 07:48:51.723979: step: 2652/1541, loss: 0.23551499843597412 2023-01-23 07:48:52.601503: step: 2654/1541, loss: 0.9403271079063416 2023-01-23 07:48:53.565753: step: 2656/1541, loss: 2.200477123260498 2023-01-23 07:48:54.509476: step: 2658/1541, loss: 0.5581628084182739 2023-01-23 07:48:55.372125: step: 2660/1541, loss: 0.36955806612968445 2023-01-23 07:48:56.253128: step: 2662/1541, loss: 0.4018784761428833 2023-01-23 07:48:57.217435: step: 2664/1541, loss: 0.5713895559310913 2023-01-23 07:48:58.097539: step: 2666/1541, loss: 0.1313004046678543 2023-01-23 07:48:58.990906: step: 2668/1541, loss: 1.0491623878479004 2023-01-23 07:48:59.916016: step: 2670/1541, loss: 0.33410221338272095 2023-01-23 07:49:00.820544: step: 2672/1541, loss: 0.48724469542503357 2023-01-23 07:49:01.691280: step: 2674/1541, loss: 0.3941655158996582 2023-01-23 07:49:02.608776: step: 2676/1541, loss: 0.6323310136795044 2023-01-23 07:49:03.529168: step: 2678/1541, loss: 0.6968972682952881 2023-01-23 07:49:04.474027: step: 2680/1541, loss: 0.28634896874427795 2023-01-23 07:49:05.411690: step: 2682/1541, loss: 2.661118507385254 2023-01-23 07:49:06.326818: step: 2684/1541, loss: 1.0970723628997803 2023-01-23 07:49:07.178975: step: 2686/1541, loss: 0.5947811603546143 2023-01-23 07:49:08.108203: step: 2688/1541, loss: 0.21835392713546753 2023-01-23 07:49:08.989540: step: 2690/1541, loss: 0.6744481921195984 2023-01-23 07:49:09.922202: step: 2692/1541, loss: 0.6827616095542908 2023-01-23 07:49:10.842323: step: 2694/1541, loss: 0.891444206237793 2023-01-23 07:49:11.711231: step: 2696/1541, loss: 0.5182383060455322 2023-01-23 07:49:12.650552: step: 2698/1541, loss: 0.4009312093257904 2023-01-23 07:49:13.555952: step: 2700/1541, loss: 0.19606293737888336 2023-01-23 07:49:14.488120: step: 2702/1541, loss: 1.0697649717330933 2023-01-23 07:49:15.404294: step: 2704/1541, loss: 0.7481158375740051 2023-01-23 07:49:16.292102: step: 2706/1541, loss: 0.8890470862388611 2023-01-23 07:49:17.253588: step: 2708/1541, loss: 0.6098788380622864 2023-01-23 07:49:18.168863: step: 2710/1541, loss: 0.5135411620140076 2023-01-23 07:49:19.066289: step: 2712/1541, loss: 0.4709668457508087 2023-01-23 07:49:19.935711: step: 2714/1541, loss: 0.13676024973392487 2023-01-23 07:49:20.844119: step: 2716/1541, loss: 0.6307987570762634 2023-01-23 07:49:21.750337: step: 2718/1541, loss: 0.3302210569381714 2023-01-23 07:49:22.640013: step: 2720/1541, loss: 0.1836952418088913 2023-01-23 07:49:23.548157: step: 2722/1541, loss: 0.47469091415405273 2023-01-23 07:49:24.454552: step: 2724/1541, loss: 0.9561082124710083 2023-01-23 07:49:25.346986: step: 2726/1541, loss: 2.2484004497528076 2023-01-23 07:49:26.204119: step: 2728/1541, loss: 0.7231213450431824 2023-01-23 07:49:27.076288: step: 2730/1541, loss: 1.06607186794281 2023-01-23 07:49:27.982878: step: 2732/1541, loss: 0.5975326895713806 2023-01-23 07:49:28.883345: step: 2734/1541, loss: 0.4757048487663269 2023-01-23 07:49:29.799024: step: 2736/1541, loss: 0.5504804849624634 2023-01-23 07:49:30.717269: step: 2738/1541, loss: 2.285722017288208 2023-01-23 07:49:31.583266: step: 2740/1541, loss: 0.41910719871520996 2023-01-23 07:49:32.423818: step: 2742/1541, loss: 0.41657623648643494 2023-01-23 07:49:33.387420: step: 2744/1541, loss: 0.4868828058242798 2023-01-23 07:49:34.273826: step: 2746/1541, loss: 0.23225651681423187 2023-01-23 07:49:35.220691: step: 2748/1541, loss: 0.36105477809906006 2023-01-23 07:49:36.171301: step: 2750/1541, loss: 0.24842335283756256 2023-01-23 07:49:37.037175: step: 2752/1541, loss: 0.5889865756034851 2023-01-23 07:49:37.918857: step: 2754/1541, loss: 1.0405395030975342 2023-01-23 07:49:38.803194: step: 2756/1541, loss: 0.2835591435432434 2023-01-23 07:49:39.705917: step: 2758/1541, loss: 0.28534117341041565 2023-01-23 07:49:40.582545: step: 2760/1541, loss: 1.6880669593811035 2023-01-23 07:49:41.457861: step: 2762/1541, loss: 0.205967515707016 2023-01-23 07:49:42.353133: step: 2764/1541, loss: 0.7806691527366638 2023-01-23 07:49:43.217280: step: 2766/1541, loss: 0.6306967735290527 2023-01-23 07:49:44.147282: step: 2768/1541, loss: 0.3808109760284424 2023-01-23 07:49:45.122864: step: 2770/1541, loss: 0.5889930725097656 2023-01-23 07:49:45.991046: step: 2772/1541, loss: 0.6213759183883667 2023-01-23 07:49:46.926175: step: 2774/1541, loss: 2.1192593574523926 2023-01-23 07:49:47.842631: step: 2776/1541, loss: 0.6211602687835693 2023-01-23 07:49:48.707014: step: 2778/1541, loss: 0.5136162638664246 2023-01-23 07:49:49.607683: step: 2780/1541, loss: 0.520929753780365 2023-01-23 07:49:50.522648: step: 2782/1541, loss: 0.38986724615097046 2023-01-23 07:49:51.402356: step: 2784/1541, loss: 0.8323750495910645 2023-01-23 07:49:52.284856: step: 2786/1541, loss: 0.5448789596557617 2023-01-23 07:49:53.175966: step: 2788/1541, loss: 0.7725962400436401 2023-01-23 07:49:54.056769: step: 2790/1541, loss: 0.44614383578300476 2023-01-23 07:49:54.929155: step: 2792/1541, loss: 0.38411372900009155 2023-01-23 07:49:55.844561: step: 2794/1541, loss: 0.6907215714454651 2023-01-23 07:49:56.757453: step: 2796/1541, loss: 0.9107602834701538 2023-01-23 07:49:57.666024: step: 2798/1541, loss: 0.35842204093933105 2023-01-23 07:49:58.542393: step: 2800/1541, loss: 0.4262738823890686 2023-01-23 07:49:59.408909: step: 2802/1541, loss: 0.6638651490211487 2023-01-23 07:50:00.267036: step: 2804/1541, loss: 0.25866881012916565 2023-01-23 07:50:01.184482: step: 2806/1541, loss: 0.3232879936695099 2023-01-23 07:50:02.104110: step: 2808/1541, loss: 0.16896623373031616 2023-01-23 07:50:03.029161: step: 2810/1541, loss: 0.32700595259666443 2023-01-23 07:50:04.007577: step: 2812/1541, loss: 0.4153127074241638 2023-01-23 07:50:04.963154: step: 2814/1541, loss: 0.4146006107330322 2023-01-23 07:50:05.823798: step: 2816/1541, loss: 0.2884581983089447 2023-01-23 07:50:06.690300: step: 2818/1541, loss: 0.44731634855270386 2023-01-23 07:50:07.568012: step: 2820/1541, loss: 0.31171995401382446 2023-01-23 07:50:08.470978: step: 2822/1541, loss: 0.5659456253051758 2023-01-23 07:50:09.371341: step: 2824/1541, loss: 0.28809139132499695 2023-01-23 07:50:10.231778: step: 2826/1541, loss: 0.18087248504161835 2023-01-23 07:50:11.142160: step: 2828/1541, loss: 0.8762319087982178 2023-01-23 07:50:12.020895: step: 2830/1541, loss: 1.0840364694595337 2023-01-23 07:50:12.981122: step: 2832/1541, loss: 0.48709049820899963 2023-01-23 07:50:13.874876: step: 2834/1541, loss: 0.5796306133270264 2023-01-23 07:50:14.798721: step: 2836/1541, loss: 0.3966221511363983 2023-01-23 07:50:15.726395: step: 2838/1541, loss: 0.7902419567108154 2023-01-23 07:50:16.579921: step: 2840/1541, loss: 0.7466058135032654 2023-01-23 07:50:17.527636: step: 2842/1541, loss: 0.3851079046726227 2023-01-23 07:50:18.437238: step: 2844/1541, loss: 0.21067722141742706 2023-01-23 07:50:19.340524: step: 2846/1541, loss: 0.1876518875360489 2023-01-23 07:50:20.220212: step: 2848/1541, loss: 0.6654484272003174 2023-01-23 07:50:21.113925: step: 2850/1541, loss: 0.5183159112930298 2023-01-23 07:50:22.020333: step: 2852/1541, loss: 0.2638576030731201 2023-01-23 07:50:22.920129: step: 2854/1541, loss: 0.12464739382266998 2023-01-23 07:50:23.820423: step: 2856/1541, loss: 0.8753559589385986 2023-01-23 07:50:24.696580: step: 2858/1541, loss: 0.3281480371952057 2023-01-23 07:50:25.600131: step: 2860/1541, loss: 0.15429198741912842 2023-01-23 07:50:26.521160: step: 2862/1541, loss: 0.6728405952453613 2023-01-23 07:50:27.475322: step: 2864/1541, loss: 0.9187719821929932 2023-01-23 07:50:28.337080: step: 2866/1541, loss: 2.569481134414673 2023-01-23 07:50:29.265978: step: 2868/1541, loss: 0.5044126510620117 2023-01-23 07:50:30.201895: step: 2870/1541, loss: 0.5714678764343262 2023-01-23 07:50:31.110581: step: 2872/1541, loss: 0.386359304189682 2023-01-23 07:50:31.996766: step: 2874/1541, loss: 0.8175901174545288 2023-01-23 07:50:32.854179: step: 2876/1541, loss: 0.21900492906570435 2023-01-23 07:50:33.721318: step: 2878/1541, loss: 0.7117804884910583 2023-01-23 07:50:34.669315: step: 2880/1541, loss: 0.7579796314239502 2023-01-23 07:50:35.550587: step: 2882/1541, loss: 0.09015877544879913 2023-01-23 07:50:36.472192: step: 2884/1541, loss: 0.36341238021850586 2023-01-23 07:50:37.395408: step: 2886/1541, loss: 0.5830059051513672 2023-01-23 07:50:38.338820: step: 2888/1541, loss: 1.043920636177063 2023-01-23 07:50:39.291868: step: 2890/1541, loss: 0.11291003227233887 2023-01-23 07:50:40.144450: step: 2892/1541, loss: 0.5142529606819153 2023-01-23 07:50:41.051949: step: 2894/1541, loss: 0.26218223571777344 2023-01-23 07:50:41.958073: step: 2896/1541, loss: 1.1673959493637085 2023-01-23 07:50:42.932794: step: 2898/1541, loss: 0.5418306589126587 2023-01-23 07:50:43.810531: step: 2900/1541, loss: 0.1520148366689682 2023-01-23 07:50:44.690908: step: 2902/1541, loss: 1.3541585206985474 2023-01-23 07:50:45.566013: step: 2904/1541, loss: 0.45219936966896057 2023-01-23 07:50:46.498592: step: 2906/1541, loss: 0.9237143397331238 2023-01-23 07:50:47.353025: step: 2908/1541, loss: 1.0841646194458008 2023-01-23 07:50:48.256515: step: 2910/1541, loss: 0.3286813795566559 2023-01-23 07:50:49.157925: step: 2912/1541, loss: 2.130981922149658 2023-01-23 07:50:50.021976: step: 2914/1541, loss: 1.203633427619934 2023-01-23 07:50:50.901872: step: 2916/1541, loss: 0.4071211814880371 2023-01-23 07:50:51.835540: step: 2918/1541, loss: 0.37644514441490173 2023-01-23 07:50:52.763792: step: 2920/1541, loss: 0.5131845474243164 2023-01-23 07:50:53.685700: step: 2922/1541, loss: 0.5801596641540527 2023-01-23 07:50:54.551434: step: 2924/1541, loss: 0.3797016143798828 2023-01-23 07:50:55.479012: step: 2926/1541, loss: 0.5892748832702637 2023-01-23 07:50:56.402919: step: 2928/1541, loss: 0.31754201650619507 2023-01-23 07:50:57.311477: step: 2930/1541, loss: 0.41500288248062134 2023-01-23 07:50:58.219040: step: 2932/1541, loss: 0.8041058778762817 2023-01-23 07:50:59.087631: step: 2934/1541, loss: 0.7887074947357178 2023-01-23 07:50:59.985320: step: 2936/1541, loss: 0.6628605127334595 2023-01-23 07:51:00.899610: step: 2938/1541, loss: 0.8191405534744263 2023-01-23 07:51:01.771478: step: 2940/1541, loss: 0.08771321177482605 2023-01-23 07:51:02.686453: step: 2942/1541, loss: 0.3845948278903961 2023-01-23 07:51:03.582932: step: 2944/1541, loss: 1.0757856369018555 2023-01-23 07:51:04.456792: step: 2946/1541, loss: 0.35152778029441833 2023-01-23 07:51:05.370836: step: 2948/1541, loss: 0.1346496045589447 2023-01-23 07:51:06.251144: step: 2950/1541, loss: 0.15881846845149994 2023-01-23 07:51:07.095894: step: 2952/1541, loss: 0.9095388054847717 2023-01-23 07:51:08.057209: step: 2954/1541, loss: 0.6389191746711731 2023-01-23 07:51:08.943040: step: 2956/1541, loss: 0.8113831281661987 2023-01-23 07:51:09.891435: step: 2958/1541, loss: 0.5551766157150269 2023-01-23 07:51:10.739550: step: 2960/1541, loss: 0.45873522758483887 2023-01-23 07:51:11.604810: step: 2962/1541, loss: 0.16222529113292694 2023-01-23 07:51:12.486854: step: 2964/1541, loss: 2.4652113914489746 2023-01-23 07:51:13.355280: step: 2966/1541, loss: 0.36901333928108215 2023-01-23 07:51:14.224278: step: 2968/1541, loss: 0.2416210174560547 2023-01-23 07:51:15.102397: step: 2970/1541, loss: 0.49566081166267395 2023-01-23 07:51:15.989524: step: 2972/1541, loss: 0.7834751605987549 2023-01-23 07:51:16.854096: step: 2974/1541, loss: 0.5760067105293274 2023-01-23 07:51:17.774323: step: 2976/1541, loss: 0.20605245232582092 2023-01-23 07:51:18.670624: step: 2978/1541, loss: 0.37243977189064026 2023-01-23 07:51:19.574608: step: 2980/1541, loss: 0.33396175503730774 2023-01-23 07:51:20.422955: step: 2982/1541, loss: 0.5624947547912598 2023-01-23 07:51:21.328254: step: 2984/1541, loss: 0.7774866819381714 2023-01-23 07:51:22.224621: step: 2986/1541, loss: 0.2250281125307083 2023-01-23 07:51:23.157820: step: 2988/1541, loss: 0.5495020747184753 2023-01-23 07:51:24.089720: step: 2990/1541, loss: 0.2895275354385376 2023-01-23 07:51:25.006488: step: 2992/1541, loss: 3.0548126697540283 2023-01-23 07:51:25.887155: step: 2994/1541, loss: 0.41702914237976074 2023-01-23 07:51:26.789476: step: 2996/1541, loss: 1.1989141702651978 2023-01-23 07:51:27.691896: step: 2998/1541, loss: 0.5165621638298035 2023-01-23 07:51:28.583311: step: 3000/1541, loss: 0.3710134029388428 2023-01-23 07:51:29.528349: step: 3002/1541, loss: 0.5695631504058838 2023-01-23 07:51:30.407004: step: 3004/1541, loss: 0.16864141821861267 2023-01-23 07:51:31.351341: step: 3006/1541, loss: 0.49398794770240784 2023-01-23 07:51:32.231272: step: 3008/1541, loss: 0.40170496702194214 2023-01-23 07:51:33.112063: step: 3010/1541, loss: 0.9228140115737915 2023-01-23 07:51:34.081501: step: 3012/1541, loss: 0.2987677752971649 2023-01-23 07:51:34.967086: step: 3014/1541, loss: 0.4252917170524597 2023-01-23 07:51:35.825829: step: 3016/1541, loss: 0.21117399632930756 2023-01-23 07:51:36.705187: step: 3018/1541, loss: 0.3214211165904999 2023-01-23 07:51:37.585873: step: 3020/1541, loss: 0.5826590657234192 2023-01-23 07:51:38.492703: step: 3022/1541, loss: 0.2412843108177185 2023-01-23 07:51:39.451358: step: 3024/1541, loss: 3.0692780017852783 2023-01-23 07:51:40.309936: step: 3026/1541, loss: 0.44308072328567505 2023-01-23 07:51:41.236267: step: 3028/1541, loss: 0.6333487033843994 2023-01-23 07:51:42.103651: step: 3030/1541, loss: 0.7202054858207703 2023-01-23 07:51:43.065946: step: 3032/1541, loss: 0.9089996814727783 2023-01-23 07:51:43.963297: step: 3034/1541, loss: 0.21607564389705658 2023-01-23 07:51:44.894371: step: 3036/1541, loss: 1.3540687561035156 2023-01-23 07:51:45.766591: step: 3038/1541, loss: 2.0132968425750732 2023-01-23 07:51:46.645359: step: 3040/1541, loss: 1.058578372001648 2023-01-23 07:51:47.496924: step: 3042/1541, loss: 0.21354851126670837 2023-01-23 07:51:48.427588: step: 3044/1541, loss: 2.3483686447143555 2023-01-23 07:51:49.333029: step: 3046/1541, loss: 0.4823003113269806 2023-01-23 07:51:50.235696: step: 3048/1541, loss: 0.5542384386062622 2023-01-23 07:51:51.110298: step: 3050/1541, loss: 0.25004711747169495 2023-01-23 07:51:52.093147: step: 3052/1541, loss: 0.5144152641296387 2023-01-23 07:51:53.052848: step: 3054/1541, loss: 0.47575637698173523 2023-01-23 07:51:53.990330: step: 3056/1541, loss: 0.6124234199523926 2023-01-23 07:51:54.861504: step: 3058/1541, loss: 0.3586718440055847 2023-01-23 07:51:55.720956: step: 3060/1541, loss: 0.1471165418624878 2023-01-23 07:51:56.591025: step: 3062/1541, loss: 0.5283329486846924 2023-01-23 07:51:57.523307: step: 3064/1541, loss: 0.9339786767959595 2023-01-23 07:51:58.409360: step: 3066/1541, loss: 0.42564815282821655 2023-01-23 07:51:59.291917: step: 3068/1541, loss: 0.6590883135795593 2023-01-23 07:52:00.156138: step: 3070/1541, loss: 0.44256308674812317 2023-01-23 07:52:01.045317: step: 3072/1541, loss: 2.967658042907715 2023-01-23 07:52:01.952836: step: 3074/1541, loss: 5.605173110961914 2023-01-23 07:52:02.849263: step: 3076/1541, loss: 0.5434237718582153 2023-01-23 07:52:03.805220: step: 3078/1541, loss: 0.5258774757385254 2023-01-23 07:52:04.684009: step: 3080/1541, loss: 0.48012810945510864 2023-01-23 07:52:05.589166: step: 3082/1541, loss: 0.3737598955631256 2023-01-23 07:52:06.196281: step: 3084/1541, loss: 1.2170095443725586 ================================================== Loss: 0.710 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Dev Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6322458242188157, 'r': 0.6853109328512413, 'f1': 0.6577097696348575}, 'combined': 0.4382255946748345, 'epoch': 5} Test Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.6886317030465187, 'r': 0.7331063374307094, 'f1': 0.7101733952193441}, 'combined': 0.4818697101621034, 'epoch': 5} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6417200356729511, 'r': 0.6867615174220715, 'f1': 0.663477222445676}, 'combined': 0.4600663373196976, 'epoch': 5} Test Russian: {'event': {'p': 0.6483516483516484, 'r': 0.7423606950269622, 'f1': 0.6921787709497207}, 'argument': {'p': 0.6895321419510843, 'r': 0.7221785704796148, 'f1': 0.7054778747362179}, 'combined': 0.4883168082671363, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} Korean: {'event': {'p': 0.5625, 'r': 0.5714285714285714, 'f1': 0.5669291338582677}, 'argument': {'p': 0.5587606837606838, 'r': 0.37250712250712253, 'f1': 0.44700854700854703}, 'combined': 0.2534221683827983, 'epoch': 5} Russian: {'event': {'p': 0.47058823529411764, 'r': 0.4444444444444444, 'f1': 0.45714285714285713}, 'argument': {'p': 0.54672127441524, 'r': 0.6102935156263144, 'f1': 0.5767609048776159}, 'combined': 0.26366212794405297, 'epoch': 5} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test for Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} -------------------- Dev for Russian: {'event': {'p': 0.6575829383886256, 'r': 0.7390146471371505, 'f1': 0.695924764890282}, 'argument': {'p': 0.6561752698674627, 'r': 0.6438411482534127, 'f1': 0.6499496980470882}, 'combined': 0.45231609080392965, 'epoch': 4} Test for Russian: {'event': {'p': 0.6530612244897959, 'r': 0.7477531455961653, 'f1': 0.6972067039106145}, 'argument': {'p': 0.6933810842434333, 'r': 0.694128262135937, 'f1': 0.6937544720109311}, 'combined': 0.4836902687539899, 'epoch': 4} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5928679962013296, 'r': 0.6204432518386007, 'f1': 0.6063422688422688}, 'combined': 0.31183316683316686, 'epoch': 4} ****************************** Epoch: 6 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 07:57:40.798065: step: 2/1541, loss: 0.500507652759552 2023-01-23 07:57:41.675448: step: 4/1541, loss: 0.35091614723205566 2023-01-23 07:57:42.550487: step: 6/1541, loss: 0.267930269241333 2023-01-23 07:57:43.460792: step: 8/1541, loss: 0.3100453317165375 2023-01-23 07:57:44.406034: step: 10/1541, loss: 1.5742079019546509 2023-01-23 07:57:45.355543: step: 12/1541, loss: 0.37256312370300293 2023-01-23 07:57:46.339363: step: 14/1541, loss: 0.3211887776851654 2023-01-23 07:57:47.201938: step: 16/1541, loss: 0.5251731872558594 2023-01-23 07:57:48.188445: step: 18/1541, loss: 0.5288754105567932 2023-01-23 07:57:49.190698: step: 20/1541, loss: 0.20092666149139404 2023-01-23 07:57:50.075006: step: 22/1541, loss: 0.11596532166004181 2023-01-23 07:57:50.924783: step: 24/1541, loss: 0.15237167477607727 2023-01-23 07:57:51.791368: step: 26/1541, loss: 0.3538958430290222 2023-01-23 07:57:52.685253: step: 28/1541, loss: 0.3781517744064331 2023-01-23 07:57:53.651462: step: 30/1541, loss: 0.2017519772052765 2023-01-23 07:57:54.550724: step: 32/1541, loss: 0.30284494161605835 2023-01-23 07:57:55.418573: step: 34/1541, loss: 0.1905328631401062 2023-01-23 07:57:56.304190: step: 36/1541, loss: 0.15878772735595703 2023-01-23 07:57:57.195514: step: 38/1541, loss: 0.2544558048248291 2023-01-23 07:57:58.059866: step: 40/1541, loss: 0.29590511322021484 2023-01-23 07:57:58.949084: step: 42/1541, loss: 0.32204118371009827 2023-01-23 07:57:59.825908: step: 44/1541, loss: 0.2614116668701172 2023-01-23 07:58:00.725804: step: 46/1541, loss: 0.5014024376869202 2023-01-23 07:58:01.641671: step: 48/1541, loss: 0.20857501029968262 2023-01-23 07:58:02.519595: step: 50/1541, loss: 0.27716511487960815 2023-01-23 07:58:03.469178: step: 52/1541, loss: 0.31351786851882935 2023-01-23 07:58:04.380797: step: 54/1541, loss: 0.06620324403047562 2023-01-23 07:58:05.268812: step: 56/1541, loss: 0.19074636697769165 2023-01-23 07:58:06.161387: step: 58/1541, loss: 0.278435617685318 2023-01-23 07:58:07.096611: step: 60/1541, loss: 0.30477654933929443 2023-01-23 07:58:07.977045: step: 62/1541, loss: 0.19252918660640717 2023-01-23 07:58:08.868146: step: 64/1541, loss: 0.08956751972436905 2023-01-23 07:58:09.730788: step: 66/1541, loss: 0.15038342773914337 2023-01-23 07:58:10.611063: step: 68/1541, loss: 0.5245330333709717 2023-01-23 07:58:11.521796: step: 70/1541, loss: 0.3480131924152374 2023-01-23 07:58:12.476609: step: 72/1541, loss: 0.15348027646541595 2023-01-23 07:58:13.411375: step: 74/1541, loss: 0.16681115329265594 2023-01-23 07:58:14.266706: step: 76/1541, loss: 0.29656854271888733 2023-01-23 07:58:15.188655: step: 78/1541, loss: 0.5623155832290649 2023-01-23 07:58:16.094703: step: 80/1541, loss: 2.2392656803131104 2023-01-23 07:58:17.003802: step: 82/1541, loss: 0.18926332890987396 2023-01-23 07:58:17.966375: step: 84/1541, loss: 0.380234956741333 2023-01-23 07:58:18.853391: step: 86/1541, loss: 0.4684787094593048 2023-01-23 07:58:19.769990: step: 88/1541, loss: 0.09561269730329514 2023-01-23 07:58:20.717515: step: 90/1541, loss: 0.6057644486427307 2023-01-23 07:58:21.566500: step: 92/1541, loss: 0.37934818863868713 2023-01-23 07:58:22.498896: step: 94/1541, loss: 0.5647006034851074 2023-01-23 07:58:23.350335: step: 96/1541, loss: 0.677927553653717 2023-01-23 07:58:24.205575: step: 98/1541, loss: 0.4568807780742645 2023-01-23 07:58:25.075097: step: 100/1541, loss: 0.710163950920105 2023-01-23 07:58:25.961676: step: 102/1541, loss: 0.33039629459381104 2023-01-23 07:58:26.856400: step: 104/1541, loss: 0.3413335084915161 2023-01-23 07:58:27.762065: step: 106/1541, loss: 0.14569145441055298 2023-01-23 07:58:28.660963: step: 108/1541, loss: 0.29151976108551025 2023-01-23 07:58:29.550691: step: 110/1541, loss: 0.08945393562316895 2023-01-23 07:58:30.476227: step: 112/1541, loss: 0.1874619871377945 2023-01-23 07:58:31.361291: step: 114/1541, loss: 0.605301558971405 2023-01-23 07:58:32.256850: step: 116/1541, loss: 0.3425743281841278 2023-01-23 07:58:33.120855: step: 118/1541, loss: 0.39613476395606995 2023-01-23 07:58:34.003644: step: 120/1541, loss: 1.1774067878723145 2023-01-23 07:58:34.929912: step: 122/1541, loss: 6.471045970916748 2023-01-23 07:58:35.805733: step: 124/1541, loss: 0.9343530535697937 2023-01-23 07:58:36.722884: step: 126/1541, loss: 0.5145600438117981 2023-01-23 07:58:37.637192: step: 128/1541, loss: 2.757112979888916 2023-01-23 07:58:38.588091: step: 130/1541, loss: 0.49889618158340454 2023-01-23 07:58:39.537118: step: 132/1541, loss: 0.47047239542007446 2023-01-23 07:58:40.440358: step: 134/1541, loss: 0.43715396523475647 2023-01-23 07:58:41.388471: step: 136/1541, loss: 0.11898164451122284 2023-01-23 07:58:42.293095: step: 138/1541, loss: 0.18063297867774963 2023-01-23 07:58:43.184852: step: 140/1541, loss: 0.24882866442203522 2023-01-23 07:58:44.110550: step: 142/1541, loss: 0.7862454652786255 2023-01-23 07:58:45.044498: step: 144/1541, loss: 0.22852513194084167 2023-01-23 07:58:45.952720: step: 146/1541, loss: 0.25809445977211 2023-01-23 07:58:46.838916: step: 148/1541, loss: 0.5076236724853516 2023-01-23 07:58:47.733949: step: 150/1541, loss: 0.20079393684864044 2023-01-23 07:58:48.634471: step: 152/1541, loss: 0.27114859223365784 2023-01-23 07:58:49.509839: step: 154/1541, loss: 0.3361905813217163 2023-01-23 07:58:50.402984: step: 156/1541, loss: 0.21301060914993286 2023-01-23 07:58:51.310868: step: 158/1541, loss: 0.3379330039024353 2023-01-23 07:58:52.218360: step: 160/1541, loss: 0.27415069937705994 2023-01-23 07:58:53.120686: step: 162/1541, loss: 0.17262092232704163 2023-01-23 07:58:54.006066: step: 164/1541, loss: 0.1359281837940216 2023-01-23 07:58:54.916086: step: 166/1541, loss: 0.11521398276090622 2023-01-23 07:58:55.761719: step: 168/1541, loss: 0.15671707689762115 2023-01-23 07:58:56.735782: step: 170/1541, loss: 0.279765248298645 2023-01-23 07:58:57.609482: step: 172/1541, loss: 0.9932923316955566 2023-01-23 07:58:58.503175: step: 174/1541, loss: 0.44228315353393555 2023-01-23 07:58:59.398164: step: 176/1541, loss: 0.38992077112197876 2023-01-23 07:59:00.274732: step: 178/1541, loss: 2.519167423248291 2023-01-23 07:59:01.214336: step: 180/1541, loss: 5.133694648742676 2023-01-23 07:59:02.114849: step: 182/1541, loss: 0.15861962735652924 2023-01-23 07:59:03.065182: step: 184/1541, loss: 2.4549009799957275 2023-01-23 07:59:03.991101: step: 186/1541, loss: 0.1423036903142929 2023-01-23 07:59:04.861919: step: 188/1541, loss: 0.5594958066940308 2023-01-23 07:59:05.794683: step: 190/1541, loss: 0.8152865171432495 2023-01-23 07:59:06.716576: step: 192/1541, loss: 0.8251644968986511 2023-01-23 07:59:07.634460: step: 194/1541, loss: 0.7729272246360779 2023-01-23 07:59:08.517299: step: 196/1541, loss: 0.14108231663703918 2023-01-23 07:59:09.416295: step: 198/1541, loss: 0.7232607007026672 2023-01-23 07:59:10.355308: step: 200/1541, loss: 0.3250674605369568 2023-01-23 07:59:11.282819: step: 202/1541, loss: 0.5978620648384094 2023-01-23 07:59:12.182622: step: 204/1541, loss: 0.28110063076019287 2023-01-23 07:59:13.184627: step: 206/1541, loss: 0.3630102872848511 2023-01-23 07:59:14.086551: step: 208/1541, loss: 0.4355606734752655 2023-01-23 07:59:14.984881: step: 210/1541, loss: 2.4699418544769287 2023-01-23 07:59:15.864385: step: 212/1541, loss: 0.7364445328712463 2023-01-23 07:59:16.730205: step: 214/1541, loss: 1.8073146343231201 2023-01-23 07:59:17.611960: step: 216/1541, loss: 0.3285864293575287 2023-01-23 07:59:18.497091: step: 218/1541, loss: 0.5411184430122375 2023-01-23 07:59:19.421026: step: 220/1541, loss: 0.4004988968372345 2023-01-23 07:59:20.421609: step: 222/1541, loss: 0.5398527383804321 2023-01-23 07:59:21.284736: step: 224/1541, loss: 0.2226548194885254 2023-01-23 07:59:22.183113: step: 226/1541, loss: 0.548373818397522 2023-01-23 07:59:23.083256: step: 228/1541, loss: 0.3933064639568329 2023-01-23 07:59:23.976885: step: 230/1541, loss: 0.3323669731616974 2023-01-23 07:59:24.873285: step: 232/1541, loss: 0.23367758095264435 2023-01-23 07:59:25.846740: step: 234/1541, loss: 0.38712430000305176 2023-01-23 07:59:26.728519: step: 236/1541, loss: 0.22893044352531433 2023-01-23 07:59:27.693010: step: 238/1541, loss: 0.5218448042869568 2023-01-23 07:59:28.635266: step: 240/1541, loss: 0.7026140093803406 2023-01-23 07:59:29.512915: step: 242/1541, loss: 0.8510549068450928 2023-01-23 07:59:30.402625: step: 244/1541, loss: 0.4879894554615021 2023-01-23 07:59:31.317696: step: 246/1541, loss: 0.3634643256664276 2023-01-23 07:59:32.256050: step: 248/1541, loss: 0.8016541600227356 2023-01-23 07:59:33.142185: step: 250/1541, loss: 0.24145808815956116 2023-01-23 07:59:33.997654: step: 252/1541, loss: 0.23631155490875244 2023-01-23 07:59:34.925880: step: 254/1541, loss: 0.517349123954773 2023-01-23 07:59:35.799234: step: 256/1541, loss: 0.47147780656814575 2023-01-23 07:59:36.665926: step: 258/1541, loss: 0.5417463183403015 2023-01-23 07:59:37.563521: step: 260/1541, loss: 1.0960354804992676 2023-01-23 07:59:38.444262: step: 262/1541, loss: 0.34444668889045715 2023-01-23 07:59:39.308829: step: 264/1541, loss: 1.9368404150009155 2023-01-23 07:59:40.301106: step: 266/1541, loss: 0.48701366782188416 2023-01-23 07:59:41.189371: step: 268/1541, loss: 0.36948907375335693 2023-01-23 07:59:42.081949: step: 270/1541, loss: 0.7693370580673218 2023-01-23 07:59:42.961333: step: 272/1541, loss: 1.4262962341308594 2023-01-23 07:59:43.867447: step: 274/1541, loss: 0.6987676620483398 2023-01-23 07:59:44.755507: step: 276/1541, loss: 0.33181700110435486 2023-01-23 07:59:45.641407: step: 278/1541, loss: 0.574029803276062 2023-01-23 07:59:46.565049: step: 280/1541, loss: 0.41238266229629517 2023-01-23 07:59:47.461627: step: 282/1541, loss: 0.5183268785476685 2023-01-23 07:59:48.401044: step: 284/1541, loss: 0.2344377189874649 2023-01-23 07:59:49.288874: step: 286/1541, loss: 0.3896227180957794 2023-01-23 07:59:50.167406: step: 288/1541, loss: 0.18413016200065613 2023-01-23 07:59:51.030284: step: 290/1541, loss: 1.125159502029419 2023-01-23 07:59:51.991967: step: 292/1541, loss: 0.3108871579170227 2023-01-23 07:59:52.867819: step: 294/1541, loss: 0.6694685816764832 2023-01-23 07:59:53.753333: step: 296/1541, loss: 0.2964819669723511 2023-01-23 07:59:54.637612: step: 298/1541, loss: 0.47084900736808777 2023-01-23 07:59:55.528911: step: 300/1541, loss: 0.9026095867156982 2023-01-23 07:59:56.497994: step: 302/1541, loss: 0.4857015311717987 2023-01-23 07:59:57.384510: step: 304/1541, loss: 0.4092303514480591 2023-01-23 07:59:58.248854: step: 306/1541, loss: 0.3634371757507324 2023-01-23 07:59:59.151816: step: 308/1541, loss: 0.3198845088481903 2023-01-23 08:00:00.050943: step: 310/1541, loss: 0.4734328091144562 2023-01-23 08:00:00.923568: step: 312/1541, loss: 0.23022279143333435 2023-01-23 08:00:01.807006: step: 314/1541, loss: 0.49513155221939087 2023-01-23 08:00:02.743394: step: 316/1541, loss: 0.2671811580657959 2023-01-23 08:00:03.631459: step: 318/1541, loss: 0.42583364248275757 2023-01-23 08:00:04.504766: step: 320/1541, loss: 0.13578461110591888 2023-01-23 08:00:05.449414: step: 322/1541, loss: 0.38985249400138855 2023-01-23 08:00:06.355049: step: 324/1541, loss: 0.368411660194397 2023-01-23 08:00:07.277863: step: 326/1541, loss: 0.5089240074157715 2023-01-23 08:00:08.218463: step: 328/1541, loss: 0.2455664426088333 2023-01-23 08:00:09.094485: step: 330/1541, loss: 0.5683654546737671 2023-01-23 08:00:10.003988: step: 332/1541, loss: 0.2801283597946167 2023-01-23 08:00:10.900292: step: 334/1541, loss: 0.3468043804168701 2023-01-23 08:00:11.769497: step: 336/1541, loss: 0.27761852741241455 2023-01-23 08:00:12.679635: step: 338/1541, loss: 0.2434716820716858 2023-01-23 08:00:13.575734: step: 340/1541, loss: 0.7322372794151306 2023-01-23 08:00:14.467014: step: 342/1541, loss: 0.42338553071022034 2023-01-23 08:00:15.425644: step: 344/1541, loss: 0.7187339067459106 2023-01-23 08:00:16.329545: step: 346/1541, loss: 0.41098228096961975 2023-01-23 08:00:17.207977: step: 348/1541, loss: 0.29642969369888306 2023-01-23 08:00:18.067444: step: 350/1541, loss: 0.13020926713943481 2023-01-23 08:00:18.969304: step: 352/1541, loss: 0.6814112663269043 2023-01-23 08:00:19.898992: step: 354/1541, loss: 0.16409561038017273 2023-01-23 08:00:20.795177: step: 356/1541, loss: 0.5913775563240051 2023-01-23 08:00:21.679252: step: 358/1541, loss: 6.734096050262451 2023-01-23 08:00:22.597800: step: 360/1541, loss: 0.3419862687587738 2023-01-23 08:00:23.551669: step: 362/1541, loss: 0.2198079228401184 2023-01-23 08:00:24.489194: step: 364/1541, loss: 0.4992237985134125 2023-01-23 08:00:25.357877: step: 366/1541, loss: 0.3134104609489441 2023-01-23 08:00:26.255810: step: 368/1541, loss: 0.45748791098594666 2023-01-23 08:00:27.189658: step: 370/1541, loss: 2.709622621536255 2023-01-23 08:00:28.096625: step: 372/1541, loss: 0.20819003880023956 2023-01-23 08:00:28.983426: step: 374/1541, loss: 1.4754332304000854 2023-01-23 08:00:29.868504: step: 376/1541, loss: 0.38337305188179016 2023-01-23 08:00:30.747852: step: 378/1541, loss: 0.3490101993083954 2023-01-23 08:00:31.624754: step: 380/1541, loss: 0.37776294350624084 2023-01-23 08:00:32.562864: step: 382/1541, loss: 0.36900559067726135 2023-01-23 08:00:33.462046: step: 384/1541, loss: 1.0975899696350098 2023-01-23 08:00:34.372740: step: 386/1541, loss: 0.39791634678840637 2023-01-23 08:00:35.292149: step: 388/1541, loss: 0.7617560625076294 2023-01-23 08:00:36.251437: step: 390/1541, loss: 0.2070271223783493 2023-01-23 08:00:37.159954: step: 392/1541, loss: 0.5415173172950745 2023-01-23 08:00:38.059274: step: 394/1541, loss: 0.36216863989830017 2023-01-23 08:00:39.038305: step: 396/1541, loss: 0.44330811500549316 2023-01-23 08:00:39.934241: step: 398/1541, loss: 4.7159295082092285 2023-01-23 08:00:40.807241: step: 400/1541, loss: 0.5363852381706238 2023-01-23 08:00:41.778167: step: 402/1541, loss: 0.8135156035423279 2023-01-23 08:00:42.652439: step: 404/1541, loss: 0.1081734448671341 2023-01-23 08:00:43.523828: step: 406/1541, loss: 0.3973347544670105 2023-01-23 08:00:44.439016: step: 408/1541, loss: 0.15471100807189941 2023-01-23 08:00:45.308665: step: 410/1541, loss: 0.30473262071609497 2023-01-23 08:00:46.207122: step: 412/1541, loss: 0.2646850645542145 2023-01-23 08:00:47.062305: step: 414/1541, loss: 0.1684950292110443 2023-01-23 08:00:47.977507: step: 416/1541, loss: 0.1977977305650711 2023-01-23 08:00:48.891571: step: 418/1541, loss: 0.8516599535942078 2023-01-23 08:00:49.762686: step: 420/1541, loss: 0.24947595596313477 2023-01-23 08:00:50.679068: step: 422/1541, loss: 0.7103024125099182 2023-01-23 08:00:51.580494: step: 424/1541, loss: 0.5467776656150818 2023-01-23 08:00:52.479833: step: 426/1541, loss: 0.5151442289352417 2023-01-23 08:00:53.348410: step: 428/1541, loss: 0.2425050586462021 2023-01-23 08:00:54.248578: step: 430/1541, loss: 0.16990582644939423 2023-01-23 08:00:55.114605: step: 432/1541, loss: 0.4526803195476532 2023-01-23 08:00:56.023415: step: 434/1541, loss: 0.23862804472446442 2023-01-23 08:00:56.958193: step: 436/1541, loss: 0.44223183393478394 2023-01-23 08:00:57.876858: step: 438/1541, loss: 0.926773190498352 2023-01-23 08:00:58.744421: step: 440/1541, loss: 0.6094517111778259 2023-01-23 08:00:59.707767: step: 442/1541, loss: 0.3441835343837738 2023-01-23 08:01:00.619357: step: 444/1541, loss: 0.7516029477119446 2023-01-23 08:01:01.517814: step: 446/1541, loss: 0.3610179126262665 2023-01-23 08:01:02.571066: step: 448/1541, loss: 0.4897823929786682 2023-01-23 08:01:03.453334: step: 450/1541, loss: 0.4006584882736206 2023-01-23 08:01:04.328921: step: 452/1541, loss: 0.34259167313575745 2023-01-23 08:01:05.225792: step: 454/1541, loss: 0.23508480191230774 2023-01-23 08:01:06.114854: step: 456/1541, loss: 0.3784990608692169 2023-01-23 08:01:07.026733: step: 458/1541, loss: 0.5179835557937622 2023-01-23 08:01:07.926623: step: 460/1541, loss: 0.11767146736383438 2023-01-23 08:01:08.826754: step: 462/1541, loss: 0.5375985503196716 2023-01-23 08:01:09.712319: step: 464/1541, loss: 0.6022869944572449 2023-01-23 08:01:10.582025: step: 466/1541, loss: 0.45381268858909607 2023-01-23 08:01:11.474569: step: 468/1541, loss: 0.4353902041912079 2023-01-23 08:01:12.379551: step: 470/1541, loss: 0.6596392393112183 2023-01-23 08:01:13.315205: step: 472/1541, loss: 0.3420249819755554 2023-01-23 08:01:14.288415: step: 474/1541, loss: 0.16034764051437378 2023-01-23 08:01:15.182573: step: 476/1541, loss: 0.2635491192340851 2023-01-23 08:01:16.077850: step: 478/1541, loss: 0.5665384531021118 2023-01-23 08:01:16.984742: step: 480/1541, loss: 2.153563976287842 2023-01-23 08:01:17.926402: step: 482/1541, loss: 0.6783477663993835 2023-01-23 08:01:18.903335: step: 484/1541, loss: 0.30933985114097595 2023-01-23 08:01:19.775959: step: 486/1541, loss: 0.42668285965919495 2023-01-23 08:01:20.684733: step: 488/1541, loss: 1.2182286977767944 2023-01-23 08:01:21.567852: step: 490/1541, loss: 0.28363993763923645 2023-01-23 08:01:22.473425: step: 492/1541, loss: 0.37658271193504333 2023-01-23 08:01:23.383775: step: 494/1541, loss: 0.6546881198883057 2023-01-23 08:01:24.287382: step: 496/1541, loss: 0.719887912273407 2023-01-23 08:01:25.201652: step: 498/1541, loss: 0.6217540502548218 2023-01-23 08:01:26.102553: step: 500/1541, loss: 1.023913025856018 2023-01-23 08:01:27.000092: step: 502/1541, loss: 0.3697237968444824 2023-01-23 08:01:27.875692: step: 504/1541, loss: 0.12795498967170715 2023-01-23 08:01:28.829168: step: 506/1541, loss: 0.17195260524749756 2023-01-23 08:01:29.734292: step: 508/1541, loss: 0.060005299746990204 2023-01-23 08:01:30.622839: step: 510/1541, loss: 0.5148152709007263 2023-01-23 08:01:31.476836: step: 512/1541, loss: 0.14306561648845673 2023-01-23 08:01:32.387351: step: 514/1541, loss: 0.8711773753166199 2023-01-23 08:01:33.258650: step: 516/1541, loss: 0.29792067408561707 2023-01-23 08:01:34.110756: step: 518/1541, loss: 0.17234192788600922 2023-01-23 08:01:35.039203: step: 520/1541, loss: 1.588883638381958 2023-01-23 08:01:35.982495: step: 522/1541, loss: 0.6639774441719055 2023-01-23 08:01:36.914494: step: 524/1541, loss: 0.8196728825569153 2023-01-23 08:01:37.804842: step: 526/1541, loss: 0.3413257896900177 2023-01-23 08:01:38.674507: step: 528/1541, loss: 6.015714645385742 2023-01-23 08:01:39.606258: step: 530/1541, loss: 0.5364513397216797 2023-01-23 08:01:40.537891: step: 532/1541, loss: 0.32578402757644653 2023-01-23 08:01:41.456887: step: 534/1541, loss: 0.125862717628479 2023-01-23 08:01:42.310206: step: 536/1541, loss: 0.4823387563228607 2023-01-23 08:01:43.217106: step: 538/1541, loss: 0.30902862548828125 2023-01-23 08:01:44.110578: step: 540/1541, loss: 0.20098133385181427 2023-01-23 08:01:45.007131: step: 542/1541, loss: 0.36365655064582825 2023-01-23 08:01:45.930187: step: 544/1541, loss: 0.14494836330413818 2023-01-23 08:01:46.847440: step: 546/1541, loss: 0.08853714913129807 2023-01-23 08:01:47.707888: step: 548/1541, loss: 0.24762560427188873 2023-01-23 08:01:48.604416: step: 550/1541, loss: 0.3656854033470154 2023-01-23 08:01:49.456552: step: 552/1541, loss: 0.3737451732158661 2023-01-23 08:01:50.369804: step: 554/1541, loss: 0.2665591835975647 2023-01-23 08:01:51.261496: step: 556/1541, loss: 0.3023959994316101 2023-01-23 08:01:52.129043: step: 558/1541, loss: 0.4702352285385132 2023-01-23 08:01:52.999380: step: 560/1541, loss: 0.3383077383041382 2023-01-23 08:01:53.954139: step: 562/1541, loss: 1.045571208000183 2023-01-23 08:01:54.808691: step: 564/1541, loss: 0.4340536296367645 2023-01-23 08:01:55.703458: step: 566/1541, loss: 0.13151392340660095 2023-01-23 08:01:56.669265: step: 568/1541, loss: 0.39763346314430237 2023-01-23 08:01:57.587760: step: 570/1541, loss: 0.5530874133110046 2023-01-23 08:01:58.459178: step: 572/1541, loss: 0.436620831489563 2023-01-23 08:01:59.350496: step: 574/1541, loss: 0.2939145863056183 2023-01-23 08:02:00.274044: step: 576/1541, loss: 2.8317770957946777 2023-01-23 08:02:01.167229: step: 578/1541, loss: 0.30225977301597595 2023-01-23 08:02:02.114657: step: 580/1541, loss: 0.28455349802970886 2023-01-23 08:02:03.043877: step: 582/1541, loss: 0.36056166887283325 2023-01-23 08:02:03.925899: step: 584/1541, loss: 0.36365190148353577 2023-01-23 08:02:04.817851: step: 586/1541, loss: 1.70823335647583 2023-01-23 08:02:05.803395: step: 588/1541, loss: 0.8214101791381836 2023-01-23 08:02:06.765971: step: 590/1541, loss: 0.22921834886074066 2023-01-23 08:02:07.638528: step: 592/1541, loss: 0.2518177330493927 2023-01-23 08:02:08.507558: step: 594/1541, loss: 0.2180071622133255 2023-01-23 08:02:09.453748: step: 596/1541, loss: 6.370394229888916 2023-01-23 08:02:10.333012: step: 598/1541, loss: 0.41584527492523193 2023-01-23 08:02:11.224581: step: 600/1541, loss: 0.599894642829895 2023-01-23 08:02:12.135123: step: 602/1541, loss: 0.4958551228046417 2023-01-23 08:02:13.031007: step: 604/1541, loss: 0.5631144046783447 2023-01-23 08:02:13.924505: step: 606/1541, loss: 1.0454453229904175 2023-01-23 08:02:14.836515: step: 608/1541, loss: 0.3138551414012909 2023-01-23 08:02:15.695846: step: 610/1541, loss: 0.06449386477470398 2023-01-23 08:02:16.564924: step: 612/1541, loss: 0.5057158470153809 2023-01-23 08:02:17.454657: step: 614/1541, loss: 0.21097849309444427 2023-01-23 08:02:18.354394: step: 616/1541, loss: 0.8915258049964905 2023-01-23 08:02:19.296733: step: 618/1541, loss: 0.25977492332458496 2023-01-23 08:02:20.296579: step: 620/1541, loss: 0.16072003543376923 2023-01-23 08:02:21.282624: step: 622/1541, loss: 0.19545307755470276 2023-01-23 08:02:22.177054: step: 624/1541, loss: 0.15033531188964844 2023-01-23 08:02:23.111985: step: 626/1541, loss: 0.4854429364204407 2023-01-23 08:02:23.988619: step: 628/1541, loss: 0.599012017250061 2023-01-23 08:02:24.839849: step: 630/1541, loss: 0.22617727518081665 2023-01-23 08:02:25.719972: step: 632/1541, loss: 1.9792183637619019 2023-01-23 08:02:26.582566: step: 634/1541, loss: 0.4038866460323334 2023-01-23 08:02:27.564260: step: 636/1541, loss: 0.23420816659927368 2023-01-23 08:02:28.488369: step: 638/1541, loss: 0.3534984588623047 2023-01-23 08:02:29.375109: step: 640/1541, loss: 0.25808990001678467 2023-01-23 08:02:30.341343: step: 642/1541, loss: 2.1491594314575195 2023-01-23 08:02:31.268323: step: 644/1541, loss: 0.48934394121170044 2023-01-23 08:02:32.243481: step: 646/1541, loss: 0.44705164432525635 2023-01-23 08:02:33.089187: step: 648/1541, loss: 0.4731636941432953 2023-01-23 08:02:33.989425: step: 650/1541, loss: 0.15272469818592072 2023-01-23 08:02:34.887425: step: 652/1541, loss: 0.2533891797065735 2023-01-23 08:02:35.777414: step: 654/1541, loss: 0.3597196340560913 2023-01-23 08:02:36.657306: step: 656/1541, loss: 0.7363870143890381 2023-01-23 08:02:37.564904: step: 658/1541, loss: 1.4088890552520752 2023-01-23 08:02:38.417261: step: 660/1541, loss: 0.49380242824554443 2023-01-23 08:02:39.307273: step: 662/1541, loss: 0.15322981774806976 2023-01-23 08:02:40.210110: step: 664/1541, loss: 0.37386608123779297 2023-01-23 08:02:41.087404: step: 666/1541, loss: 0.3542101979255676 2023-01-23 08:02:41.985220: step: 668/1541, loss: 0.0664055198431015 2023-01-23 08:02:42.832646: step: 670/1541, loss: 1.065393328666687 2023-01-23 08:02:43.743482: step: 672/1541, loss: 0.11436162889003754 2023-01-23 08:02:44.614290: step: 674/1541, loss: 0.2833995819091797 2023-01-23 08:02:45.501103: step: 676/1541, loss: 2.698692798614502 2023-01-23 08:02:46.407652: step: 678/1541, loss: 0.33399492502212524 2023-01-23 08:02:47.263382: step: 680/1541, loss: 0.08963709324598312 2023-01-23 08:02:48.175970: step: 682/1541, loss: 0.2604145407676697 2023-01-23 08:02:49.054205: step: 684/1541, loss: 0.21819472312927246 2023-01-23 08:02:49.930808: step: 686/1541, loss: 0.4121764302253723 2023-01-23 08:02:50.812720: step: 688/1541, loss: 0.16777923703193665 2023-01-23 08:02:51.670344: step: 690/1541, loss: 0.13156044483184814 2023-01-23 08:02:52.655849: step: 692/1541, loss: 0.5859478116035461 2023-01-23 08:02:53.627967: step: 694/1541, loss: 0.6729094386100769 2023-01-23 08:02:54.520716: step: 696/1541, loss: 1.9342354536056519 2023-01-23 08:02:55.423917: step: 698/1541, loss: 2.3937737941741943 2023-01-23 08:02:56.293152: step: 700/1541, loss: 0.33693966269493103 2023-01-23 08:02:57.211703: step: 702/1541, loss: 0.9971305727958679 2023-01-23 08:02:58.140392: step: 704/1541, loss: 0.35445743799209595 2023-01-23 08:02:59.102366: step: 706/1541, loss: 0.232361301779747 2023-01-23 08:03:00.026336: step: 708/1541, loss: 0.4413670599460602 2023-01-23 08:03:00.918873: step: 710/1541, loss: 0.39874476194381714 2023-01-23 08:03:01.866205: step: 712/1541, loss: 0.4774099290370941 2023-01-23 08:03:02.837205: step: 714/1541, loss: 0.17228519916534424 2023-01-23 08:03:03.724212: step: 716/1541, loss: 0.6465630531311035 2023-01-23 08:03:04.626438: step: 718/1541, loss: 0.33075952529907227 2023-01-23 08:03:05.521507: step: 720/1541, loss: 2.6160523891448975 2023-01-23 08:03:06.403524: step: 722/1541, loss: 0.7621169090270996 2023-01-23 08:03:07.355978: step: 724/1541, loss: 1.1449213027954102 2023-01-23 08:03:08.244796: step: 726/1541, loss: 0.4483550786972046 2023-01-23 08:03:09.128680: step: 728/1541, loss: 0.2866792678833008 2023-01-23 08:03:10.039806: step: 730/1541, loss: 0.38806992769241333 2023-01-23 08:03:10.997128: step: 732/1541, loss: 1.3381233215332031 2023-01-23 08:03:11.871296: step: 734/1541, loss: 0.17871347069740295 2023-01-23 08:03:12.735560: step: 736/1541, loss: 0.9115318655967712 2023-01-23 08:03:13.650822: step: 738/1541, loss: 0.17706939578056335 2023-01-23 08:03:14.592756: step: 740/1541, loss: 0.5413894653320312 2023-01-23 08:03:15.484211: step: 742/1541, loss: 0.7459205985069275 2023-01-23 08:03:16.355421: step: 744/1541, loss: 0.16716526448726654 2023-01-23 08:03:17.236568: step: 746/1541, loss: 0.24285627901554108 2023-01-23 08:03:18.128395: step: 748/1541, loss: 0.34173092246055603 2023-01-23 08:03:19.007496: step: 750/1541, loss: 0.37710174918174744 2023-01-23 08:03:19.952047: step: 752/1541, loss: 0.15724965929985046 2023-01-23 08:03:20.901328: step: 754/1541, loss: 0.16739164292812347 2023-01-23 08:03:21.775272: step: 756/1541, loss: 0.7210153937339783 2023-01-23 08:03:22.704719: step: 758/1541, loss: 0.23122604191303253 2023-01-23 08:03:23.666410: step: 760/1541, loss: 0.38111844658851624 2023-01-23 08:03:24.574403: step: 762/1541, loss: 0.7611818313598633 2023-01-23 08:03:25.484761: step: 764/1541, loss: 0.4451899528503418 2023-01-23 08:03:26.337527: step: 766/1541, loss: 0.11531468480825424 2023-01-23 08:03:27.281194: step: 768/1541, loss: 0.12763148546218872 2023-01-23 08:03:28.176502: step: 770/1541, loss: 0.3126509189605713 2023-01-23 08:03:29.047490: step: 772/1541, loss: 0.5679978728294373 2023-01-23 08:03:29.973819: step: 774/1541, loss: 0.3738196790218353 2023-01-23 08:03:30.858144: step: 776/1541, loss: 0.6587554812431335 2023-01-23 08:03:31.798390: step: 778/1541, loss: 0.266900897026062 2023-01-23 08:03:32.725742: step: 780/1541, loss: 0.3926103711128235 2023-01-23 08:03:33.630521: step: 782/1541, loss: 0.5913467407226562 2023-01-23 08:03:34.521827: step: 784/1541, loss: 0.5661932229995728 2023-01-23 08:03:35.450651: step: 786/1541, loss: 0.5876968502998352 2023-01-23 08:03:36.352926: step: 788/1541, loss: 0.5589649677276611 2023-01-23 08:03:37.243789: step: 790/1541, loss: 0.4271033704280853 2023-01-23 08:03:38.163909: step: 792/1541, loss: 0.41635146737098694 2023-01-23 08:03:39.075649: step: 794/1541, loss: 0.3314535319805145 2023-01-23 08:03:40.062903: step: 796/1541, loss: 0.6780869364738464 2023-01-23 08:03:41.047520: step: 798/1541, loss: 0.3082748055458069 2023-01-23 08:03:41.966361: step: 800/1541, loss: 1.0103105306625366 2023-01-23 08:03:42.838285: step: 802/1541, loss: 0.35150429606437683 2023-01-23 08:03:43.717771: step: 804/1541, loss: 0.03887397050857544 2023-01-23 08:03:44.591968: step: 806/1541, loss: 0.27873679995536804 2023-01-23 08:03:45.478915: step: 808/1541, loss: 0.49573424458503723 2023-01-23 08:03:46.368819: step: 810/1541, loss: 0.6877084374427795 2023-01-23 08:03:47.248629: step: 812/1541, loss: 0.18027205765247345 2023-01-23 08:03:48.134616: step: 814/1541, loss: 0.36840757727622986 2023-01-23 08:03:49.131070: step: 816/1541, loss: 0.2554815113544464 2023-01-23 08:03:50.053636: step: 818/1541, loss: 0.5034499168395996 2023-01-23 08:03:50.950178: step: 820/1541, loss: 0.41872918605804443 2023-01-23 08:03:51.849826: step: 822/1541, loss: 0.7804913520812988 2023-01-23 08:03:52.745593: step: 824/1541, loss: 0.19372114539146423 2023-01-23 08:03:53.637517: step: 826/1541, loss: 0.417525976896286 2023-01-23 08:03:54.578195: step: 828/1541, loss: 0.2762351632118225 2023-01-23 08:03:55.515253: step: 830/1541, loss: 0.9263100624084473 2023-01-23 08:03:56.427172: step: 832/1541, loss: 0.41235873103141785 2023-01-23 08:03:57.319690: step: 834/1541, loss: 0.7295686602592468 2023-01-23 08:03:58.241771: step: 836/1541, loss: 0.8307045698165894 2023-01-23 08:03:59.103881: step: 838/1541, loss: 0.3706996440887451 2023-01-23 08:03:59.966007: step: 840/1541, loss: 0.24040736258029938 2023-01-23 08:04:00.977506: step: 842/1541, loss: 0.2943774163722992 2023-01-23 08:04:01.892520: step: 844/1541, loss: 0.5358448028564453 2023-01-23 08:04:02.793758: step: 846/1541, loss: 0.5927435159683228 2023-01-23 08:04:03.701220: step: 848/1541, loss: 0.03468070179224014 2023-01-23 08:04:04.643417: step: 850/1541, loss: 0.2974211573600769 2023-01-23 08:04:05.541526: step: 852/1541, loss: 0.15076890587806702 2023-01-23 08:04:06.416223: step: 854/1541, loss: 0.6955006718635559 2023-01-23 08:04:07.289383: step: 856/1541, loss: 0.6062856912612915 2023-01-23 08:04:08.184918: step: 858/1541, loss: 0.17292103171348572 2023-01-23 08:04:09.079002: step: 860/1541, loss: 0.5730358958244324 2023-01-23 08:04:09.999583: step: 862/1541, loss: 0.3197302520275116 2023-01-23 08:04:10.899575: step: 864/1541, loss: 0.2400435358285904 2023-01-23 08:04:11.786169: step: 866/1541, loss: 0.7946978211402893 2023-01-23 08:04:12.616194: step: 868/1541, loss: 0.17531928420066833 2023-01-23 08:04:13.512015: step: 870/1541, loss: 0.2082819938659668 2023-01-23 08:04:14.424622: step: 872/1541, loss: 0.3259665071964264 2023-01-23 08:04:15.292713: step: 874/1541, loss: 0.2123008668422699 2023-01-23 08:04:16.172118: step: 876/1541, loss: 0.37069353461265564 2023-01-23 08:04:17.128766: step: 878/1541, loss: 0.6953471302986145 2023-01-23 08:04:18.043015: step: 880/1541, loss: 0.2801518142223358 2023-01-23 08:04:18.938436: step: 882/1541, loss: 0.1992969512939453 2023-01-23 08:04:19.825057: step: 884/1541, loss: 0.6295225620269775 2023-01-23 08:04:20.691977: step: 886/1541, loss: 0.3483502268791199 2023-01-23 08:04:21.624579: step: 888/1541, loss: 0.5140961408615112 2023-01-23 08:04:22.532786: step: 890/1541, loss: 0.3904166519641876 2023-01-23 08:04:23.416492: step: 892/1541, loss: 0.23267509043216705 2023-01-23 08:04:24.316152: step: 894/1541, loss: 0.38980305194854736 2023-01-23 08:04:25.247107: step: 896/1541, loss: 0.22482477128505707 2023-01-23 08:04:26.160411: step: 898/1541, loss: 0.5987287759780884 2023-01-23 08:04:27.071890: step: 900/1541, loss: 0.7914019823074341 2023-01-23 08:04:27.972397: step: 902/1541, loss: 0.25442561507225037 2023-01-23 08:04:28.938092: step: 904/1541, loss: 0.29306840896606445 2023-01-23 08:04:29.815772: step: 906/1541, loss: 0.5451247096061707 2023-01-23 08:04:30.752709: step: 908/1541, loss: 0.46888452768325806 2023-01-23 08:04:31.650226: step: 910/1541, loss: 0.37538471817970276 2023-01-23 08:04:32.723757: step: 912/1541, loss: 1.1380623579025269 2023-01-23 08:04:33.613618: step: 914/1541, loss: 0.47591719031333923 2023-01-23 08:04:34.481333: step: 916/1541, loss: 0.31571197509765625 2023-01-23 08:04:35.346328: step: 918/1541, loss: 0.4004712402820587 2023-01-23 08:04:36.286340: step: 920/1541, loss: 0.3605577051639557 2023-01-23 08:04:37.169667: step: 922/1541, loss: 0.2855992913246155 2023-01-23 08:04:38.084134: step: 924/1541, loss: 0.4085249900817871 2023-01-23 08:04:38.975058: step: 926/1541, loss: 1.921816349029541 2023-01-23 08:04:39.895504: step: 928/1541, loss: 0.44360262155532837 2023-01-23 08:04:40.805434: step: 930/1541, loss: 0.2565287947654724 2023-01-23 08:04:41.705512: step: 932/1541, loss: 0.243856742978096 2023-01-23 08:04:42.601338: step: 934/1541, loss: 0.20210887491703033 2023-01-23 08:04:43.587197: step: 936/1541, loss: 0.45769912004470825 2023-01-23 08:04:44.479353: step: 938/1541, loss: 0.6384503841400146 2023-01-23 08:04:45.377113: step: 940/1541, loss: 0.24649399518966675 2023-01-23 08:04:46.294240: step: 942/1541, loss: 0.3643210828304291 2023-01-23 08:04:47.198871: step: 944/1541, loss: 0.3404299020767212 2023-01-23 08:04:48.083995: step: 946/1541, loss: 0.1754627227783203 2023-01-23 08:04:49.003680: step: 948/1541, loss: 0.2953585088253021 2023-01-23 08:04:49.898764: step: 950/1541, loss: 0.284496545791626 2023-01-23 08:04:50.829806: step: 952/1541, loss: 0.25366032123565674 2023-01-23 08:04:51.734025: step: 954/1541, loss: 0.23478244245052338 2023-01-23 08:04:52.685436: step: 956/1541, loss: 0.5787911415100098 2023-01-23 08:04:53.581749: step: 958/1541, loss: 0.7105829119682312 2023-01-23 08:04:54.486892: step: 960/1541, loss: 0.10412629693746567 2023-01-23 08:04:55.377451: step: 962/1541, loss: 0.3334192633628845 2023-01-23 08:04:56.221745: step: 964/1541, loss: 0.20940253138542175 2023-01-23 08:04:57.113366: step: 966/1541, loss: 0.08696302026510239 2023-01-23 08:04:58.002141: step: 968/1541, loss: 0.48487791419029236 2023-01-23 08:04:58.890499: step: 970/1541, loss: 0.46122342348098755 2023-01-23 08:04:59.791455: step: 972/1541, loss: 0.701758623123169 2023-01-23 08:05:00.667627: step: 974/1541, loss: 1.036554217338562 2023-01-23 08:05:01.659610: step: 976/1541, loss: 0.41769492626190186 2023-01-23 08:05:02.627199: step: 978/1541, loss: 0.670282781124115 2023-01-23 08:05:03.540392: step: 980/1541, loss: 0.3488547205924988 2023-01-23 08:05:04.425452: step: 982/1541, loss: 0.8054803609848022 2023-01-23 08:05:05.386237: step: 984/1541, loss: 0.13544903695583344 2023-01-23 08:05:06.327734: step: 986/1541, loss: 0.4760056436061859 2023-01-23 08:05:07.241091: step: 988/1541, loss: 0.3277527093887329 2023-01-23 08:05:08.197137: step: 990/1541, loss: 0.402970552444458 2023-01-23 08:05:09.120615: step: 992/1541, loss: 0.37850889563560486 2023-01-23 08:05:10.041129: step: 994/1541, loss: 0.1388314813375473 2023-01-23 08:05:10.932605: step: 996/1541, loss: 0.5825993418693542 2023-01-23 08:05:11.801901: step: 998/1541, loss: 0.2661283314228058 2023-01-23 08:05:12.708635: step: 1000/1541, loss: 0.587837278842926 2023-01-23 08:05:13.599942: step: 1002/1541, loss: 0.4747266173362732 2023-01-23 08:05:14.462902: step: 1004/1541, loss: 0.6876859068870544 2023-01-23 08:05:15.397630: step: 1006/1541, loss: 0.5383603572845459 2023-01-23 08:05:16.321655: step: 1008/1541, loss: 2.5568161010742188 2023-01-23 08:05:17.233426: step: 1010/1541, loss: 0.2735758423805237 2023-01-23 08:05:18.144467: step: 1012/1541, loss: 0.2740701735019684 2023-01-23 08:05:19.024697: step: 1014/1541, loss: 7.33635950088501 2023-01-23 08:05:19.931548: step: 1016/1541, loss: 0.5669887065887451 2023-01-23 08:05:20.825592: step: 1018/1541, loss: 0.9318906664848328 2023-01-23 08:05:21.750132: step: 1020/1541, loss: 0.5291247963905334 2023-01-23 08:05:22.567429: step: 1022/1541, loss: 0.3857899606227875 2023-01-23 08:05:23.447061: step: 1024/1541, loss: 0.2239380180835724 2023-01-23 08:05:24.381236: step: 1026/1541, loss: 2.768073081970215 2023-01-23 08:05:25.232845: step: 1028/1541, loss: 0.26628541946411133 2023-01-23 08:05:26.115515: step: 1030/1541, loss: 0.5203501582145691 2023-01-23 08:05:26.984983: step: 1032/1541, loss: 0.4125344455242157 2023-01-23 08:05:27.844447: step: 1034/1541, loss: 0.39805808663368225 2023-01-23 08:05:28.792560: step: 1036/1541, loss: 0.257047563791275 2023-01-23 08:05:29.718168: step: 1038/1541, loss: 0.4647129476070404 2023-01-23 08:05:30.603154: step: 1040/1541, loss: 0.8649576306343079 2023-01-23 08:05:31.469440: step: 1042/1541, loss: 0.2208942323923111 2023-01-23 08:05:32.391418: step: 1044/1541, loss: 0.3012198507785797 2023-01-23 08:05:33.341852: step: 1046/1541, loss: 0.34588807821273804 2023-01-23 08:05:34.293548: step: 1048/1541, loss: 0.2413737177848816 2023-01-23 08:05:35.174021: step: 1050/1541, loss: 0.7384702563285828 2023-01-23 08:05:36.064236: step: 1052/1541, loss: 0.23253315687179565 2023-01-23 08:05:36.976095: step: 1054/1541, loss: 0.17230026423931122 2023-01-23 08:05:37.877373: step: 1056/1541, loss: 6.432274341583252 2023-01-23 08:05:38.780246: step: 1058/1541, loss: 0.2000306099653244 2023-01-23 08:05:39.647073: step: 1060/1541, loss: 0.6083076596260071 2023-01-23 08:05:40.555424: step: 1062/1541, loss: 0.253540575504303 2023-01-23 08:05:41.460702: step: 1064/1541, loss: 0.47667554020881653 2023-01-23 08:05:42.329054: step: 1066/1541, loss: 0.6532435417175293 2023-01-23 08:05:43.190247: step: 1068/1541, loss: 0.18778279423713684 2023-01-23 08:05:44.137329: step: 1070/1541, loss: 1.0841236114501953 2023-01-23 08:05:45.011699: step: 1072/1541, loss: 0.3341166079044342 2023-01-23 08:05:45.860740: step: 1074/1541, loss: 0.49246859550476074 2023-01-23 08:05:46.757757: step: 1076/1541, loss: 0.4378662705421448 2023-01-23 08:05:47.762523: step: 1078/1541, loss: 0.28363037109375 2023-01-23 08:05:48.700423: step: 1080/1541, loss: 2.5233101844787598 2023-01-23 08:05:49.612476: step: 1082/1541, loss: 0.26537859439849854 2023-01-23 08:05:50.492955: step: 1084/1541, loss: 0.5256766080856323 2023-01-23 08:05:51.366234: step: 1086/1541, loss: 0.25019264221191406 2023-01-23 08:05:52.251674: step: 1088/1541, loss: 0.8238048553466797 2023-01-23 08:05:53.165660: step: 1090/1541, loss: 0.30101853609085083 2023-01-23 08:05:54.029295: step: 1092/1541, loss: 0.3395684063434601 2023-01-23 08:05:54.962687: step: 1094/1541, loss: 0.2771054804325104 2023-01-23 08:05:55.839890: step: 1096/1541, loss: 0.488614946603775 2023-01-23 08:05:56.742823: step: 1098/1541, loss: 0.2574039399623871 2023-01-23 08:05:57.593374: step: 1100/1541, loss: 0.47091948986053467 2023-01-23 08:05:58.504458: step: 1102/1541, loss: 0.2867676615715027 2023-01-23 08:05:59.413878: step: 1104/1541, loss: 0.1530173122882843 2023-01-23 08:06:00.278809: step: 1106/1541, loss: 0.3496604263782501 2023-01-23 08:06:01.175731: step: 1108/1541, loss: 0.1604013293981552 2023-01-23 08:06:02.065748: step: 1110/1541, loss: 0.3701089024543762 2023-01-23 08:06:02.981744: step: 1112/1541, loss: 0.2143677920103073 2023-01-23 08:06:03.833563: step: 1114/1541, loss: 0.2379864603281021 2023-01-23 08:06:04.914129: step: 1116/1541, loss: 0.8583307862281799 2023-01-23 08:06:05.853649: step: 1118/1541, loss: 0.5550974607467651 2023-01-23 08:06:06.720462: step: 1120/1541, loss: 0.490281343460083 2023-01-23 08:06:07.639063: step: 1122/1541, loss: 0.18793761730194092 2023-01-23 08:06:08.594454: step: 1124/1541, loss: 0.5500059127807617 2023-01-23 08:06:09.519512: step: 1126/1541, loss: 0.7743628025054932 2023-01-23 08:06:10.382113: step: 1128/1541, loss: 0.4433215856552124 2023-01-23 08:06:11.234613: step: 1130/1541, loss: 0.23658595979213715 2023-01-23 08:06:12.075495: step: 1132/1541, loss: 0.41161197423934937 2023-01-23 08:06:13.034007: step: 1134/1541, loss: 0.2969895899295807 2023-01-23 08:06:13.945644: step: 1136/1541, loss: 2.6580841541290283 2023-01-23 08:06:14.826867: step: 1138/1541, loss: 1.2683525085449219 2023-01-23 08:06:15.774289: step: 1140/1541, loss: 0.8289112448692322 2023-01-23 08:06:16.690877: step: 1142/1541, loss: 0.42904314398765564 2023-01-23 08:06:17.592217: step: 1144/1541, loss: 0.5085129141807556 2023-01-23 08:06:18.555199: step: 1146/1541, loss: 0.36965370178222656 2023-01-23 08:06:19.484947: step: 1148/1541, loss: 0.4377092719078064 2023-01-23 08:06:20.400269: step: 1150/1541, loss: 0.45383593440055847 2023-01-23 08:06:21.294065: step: 1152/1541, loss: 3.227900505065918 2023-01-23 08:06:22.212876: step: 1154/1541, loss: 0.48005664348602295 2023-01-23 08:06:23.096390: step: 1156/1541, loss: 0.6380975842475891 2023-01-23 08:06:23.970322: step: 1158/1541, loss: 0.3900451064109802 2023-01-23 08:06:24.880655: step: 1160/1541, loss: 0.25851747393608093 2023-01-23 08:06:25.795455: step: 1162/1541, loss: 0.43616923689842224 2023-01-23 08:06:26.673262: step: 1164/1541, loss: 0.11079046875238419 2023-01-23 08:06:27.615087: step: 1166/1541, loss: 0.31026148796081543 2023-01-23 08:06:28.482585: step: 1168/1541, loss: 0.48712289333343506 2023-01-23 08:06:29.368447: step: 1170/1541, loss: 0.3518558442592621 2023-01-23 08:06:30.295545: step: 1172/1541, loss: 2.860590934753418 2023-01-23 08:06:31.231066: step: 1174/1541, loss: 0.37542247772216797 2023-01-23 08:06:32.124183: step: 1176/1541, loss: 0.5232885479927063 2023-01-23 08:06:33.085946: step: 1178/1541, loss: 0.42394542694091797 2023-01-23 08:06:34.021682: step: 1180/1541, loss: 0.426658034324646 2023-01-23 08:06:34.907058: step: 1182/1541, loss: 0.43682679533958435 2023-01-23 08:06:35.778673: step: 1184/1541, loss: 0.16070260107517242 2023-01-23 08:06:36.730804: step: 1186/1541, loss: 0.6886507272720337 2023-01-23 08:06:37.645469: step: 1188/1541, loss: 0.46931231021881104 2023-01-23 08:06:38.634996: step: 1190/1541, loss: 0.12738724052906036 2023-01-23 08:06:39.515276: step: 1192/1541, loss: 0.3494041860103607 2023-01-23 08:06:40.405900: step: 1194/1541, loss: 0.6692934036254883 2023-01-23 08:06:41.387997: step: 1196/1541, loss: 0.5423597693443298 2023-01-23 08:06:42.292736: step: 1198/1541, loss: 0.2638467252254486 2023-01-23 08:06:43.180259: step: 1200/1541, loss: 0.4571537971496582 2023-01-23 08:06:44.093151: step: 1202/1541, loss: 0.3884555399417877 2023-01-23 08:06:44.994290: step: 1204/1541, loss: 0.49335455894470215 2023-01-23 08:06:45.920714: step: 1206/1541, loss: 0.49223852157592773 2023-01-23 08:06:46.819063: step: 1208/1541, loss: 0.5491306185722351 2023-01-23 08:06:47.690540: step: 1210/1541, loss: 0.44607114791870117 2023-01-23 08:06:48.579722: step: 1212/1541, loss: 0.4028584957122803 2023-01-23 08:06:49.494787: step: 1214/1541, loss: 0.5518827438354492 2023-01-23 08:06:50.380508: step: 1216/1541, loss: 1.308434009552002 2023-01-23 08:06:51.225356: step: 1218/1541, loss: 0.9668014645576477 2023-01-23 08:06:52.101200: step: 1220/1541, loss: 0.34575366973876953 2023-01-23 08:06:53.008506: step: 1222/1541, loss: 1.8316726684570312 2023-01-23 08:06:53.919584: step: 1224/1541, loss: 0.6578618288040161 2023-01-23 08:06:54.828073: step: 1226/1541, loss: 0.2728276550769806 2023-01-23 08:06:55.710417: step: 1228/1541, loss: 0.2972695827484131 2023-01-23 08:06:56.557276: step: 1230/1541, loss: 0.13338817656040192 2023-01-23 08:06:57.515205: step: 1232/1541, loss: 0.31447190046310425 2023-01-23 08:06:58.360328: step: 1234/1541, loss: 0.4039250910282135 2023-01-23 08:06:59.310813: step: 1236/1541, loss: 0.4601166248321533 2023-01-23 08:07:00.173106: step: 1238/1541, loss: 0.25321173667907715 2023-01-23 08:07:01.072066: step: 1240/1541, loss: 0.42623424530029297 2023-01-23 08:07:01.958002: step: 1242/1541, loss: 1.2845683097839355 2023-01-23 08:07:02.865635: step: 1244/1541, loss: 0.6020932197570801 2023-01-23 08:07:03.732625: step: 1246/1541, loss: 0.7893603444099426 2023-01-23 08:07:04.611460: step: 1248/1541, loss: 0.42154183983802795 2023-01-23 08:07:05.499962: step: 1250/1541, loss: 0.2440425306558609 2023-01-23 08:07:06.453002: step: 1252/1541, loss: 0.39828187227249146 2023-01-23 08:07:07.363114: step: 1254/1541, loss: 0.9992613196372986 2023-01-23 08:07:08.233132: step: 1256/1541, loss: 0.24108870327472687 2023-01-23 08:07:09.131340: step: 1258/1541, loss: 0.3356333076953888 2023-01-23 08:07:10.043720: step: 1260/1541, loss: 0.31026583909988403 2023-01-23 08:07:10.924927: step: 1262/1541, loss: 0.16001558303833008 2023-01-23 08:07:11.822858: step: 1264/1541, loss: 0.29582229256629944 2023-01-23 08:07:12.748243: step: 1266/1541, loss: 0.15201877057552338 2023-01-23 08:07:13.691227: step: 1268/1541, loss: 0.356123149394989 2023-01-23 08:07:14.571796: step: 1270/1541, loss: 0.3601122498512268 2023-01-23 08:07:15.527382: step: 1272/1541, loss: 0.45582664012908936 2023-01-23 08:07:16.411527: step: 1274/1541, loss: 0.5335782170295715 2023-01-23 08:07:17.299398: step: 1276/1541, loss: 4.515788555145264 2023-01-23 08:07:18.250525: step: 1278/1541, loss: 0.22767332196235657 2023-01-23 08:07:19.166376: step: 1280/1541, loss: 0.16251027584075928 2023-01-23 08:07:20.043484: step: 1282/1541, loss: 0.2721185088157654 2023-01-23 08:07:20.951958: step: 1284/1541, loss: 0.35295379161834717 2023-01-23 08:07:21.820012: step: 1286/1541, loss: 0.6813206672668457 2023-01-23 08:07:22.702834: step: 1288/1541, loss: 0.44827961921691895 2023-01-23 08:07:23.619697: step: 1290/1541, loss: 0.35779422521591187 2023-01-23 08:07:24.508006: step: 1292/1541, loss: 0.1183958426117897 2023-01-23 08:07:25.402041: step: 1294/1541, loss: 0.24387283623218536 2023-01-23 08:07:26.296509: step: 1296/1541, loss: 0.07062135636806488 2023-01-23 08:07:27.162975: step: 1298/1541, loss: 0.29755067825317383 2023-01-23 08:07:28.126581: step: 1300/1541, loss: 0.5464934706687927 2023-01-23 08:07:29.006255: step: 1302/1541, loss: 0.5662743449211121 2023-01-23 08:07:29.901961: step: 1304/1541, loss: 0.07722057402133942 2023-01-23 08:07:30.824054: step: 1306/1541, loss: 0.18508227169513702 2023-01-23 08:07:31.724782: step: 1308/1541, loss: 1.6872336864471436 2023-01-23 08:07:32.640769: step: 1310/1541, loss: 0.31675079464912415 2023-01-23 08:07:33.568741: step: 1312/1541, loss: 0.14444632828235626 2023-01-23 08:07:34.438889: step: 1314/1541, loss: 0.6658889055252075 2023-01-23 08:07:35.333452: step: 1316/1541, loss: 1.176300048828125 2023-01-23 08:07:36.193845: step: 1318/1541, loss: 0.5283682346343994 2023-01-23 08:07:37.081072: step: 1320/1541, loss: 0.5436858534812927 2023-01-23 08:07:37.978714: step: 1322/1541, loss: 0.3856370747089386 2023-01-23 08:07:38.869812: step: 1324/1541, loss: 0.9394755959510803 2023-01-23 08:07:39.810916: step: 1326/1541, loss: 0.39258697628974915 2023-01-23 08:07:40.792942: step: 1328/1541, loss: 1.201114535331726 2023-01-23 08:07:41.677260: step: 1330/1541, loss: 0.10832346975803375 2023-01-23 08:07:42.570468: step: 1332/1541, loss: 0.12254363298416138 2023-01-23 08:07:43.430752: step: 1334/1541, loss: 0.2659445106983185 2023-01-23 08:07:44.377442: step: 1336/1541, loss: 2.8948216438293457 2023-01-23 08:07:45.280978: step: 1338/1541, loss: 0.3928520083427429 2023-01-23 08:07:46.205084: step: 1340/1541, loss: 0.2505321204662323 2023-01-23 08:07:47.111093: step: 1342/1541, loss: 0.16268473863601685 2023-01-23 08:07:48.003118: step: 1344/1541, loss: 0.19923780858516693 2023-01-23 08:07:48.867007: step: 1346/1541, loss: 0.17007365822792053 2023-01-23 08:07:49.732680: step: 1348/1541, loss: 4.699063777923584 2023-01-23 08:07:50.598384: step: 1350/1541, loss: 0.42743438482284546 2023-01-23 08:07:51.537983: step: 1352/1541, loss: 0.9173306822776794 2023-01-23 08:07:52.501333: step: 1354/1541, loss: 0.869073212146759 2023-01-23 08:07:53.398333: step: 1356/1541, loss: 0.5463532209396362 2023-01-23 08:07:54.294233: step: 1358/1541, loss: 0.7317110300064087 2023-01-23 08:07:55.177638: step: 1360/1541, loss: 0.16724605858325958 2023-01-23 08:07:56.143406: step: 1362/1541, loss: 0.1406833976507187 2023-01-23 08:07:57.020586: step: 1364/1541, loss: 0.34723961353302 2023-01-23 08:07:57.933262: step: 1366/1541, loss: 0.7159176468849182 2023-01-23 08:07:58.798941: step: 1368/1541, loss: 0.15231749415397644 2023-01-23 08:07:59.676995: step: 1370/1541, loss: 0.09520896524190903 2023-01-23 08:08:00.593423: step: 1372/1541, loss: 1.6215941905975342 2023-01-23 08:08:01.503416: step: 1374/1541, loss: 0.11812170594930649 2023-01-23 08:08:02.525931: step: 1376/1541, loss: 0.21112239360809326 2023-01-23 08:08:03.438035: step: 1378/1541, loss: 0.13626526296138763 2023-01-23 08:08:04.361636: step: 1380/1541, loss: 0.37429195642471313 2023-01-23 08:08:05.263451: step: 1382/1541, loss: 0.4670664072036743 2023-01-23 08:08:06.165281: step: 1384/1541, loss: 3.4917948246002197 2023-01-23 08:08:07.036316: step: 1386/1541, loss: 0.9997674822807312 2023-01-23 08:08:07.938457: step: 1388/1541, loss: 0.23070518672466278 2023-01-23 08:08:08.858615: step: 1390/1541, loss: 3.033806562423706 2023-01-23 08:08:09.790168: step: 1392/1541, loss: 0.6389195919036865 2023-01-23 08:08:10.675966: step: 1394/1541, loss: 0.2891182005405426 2023-01-23 08:08:11.563034: step: 1396/1541, loss: 0.25527840852737427 2023-01-23 08:08:12.488591: step: 1398/1541, loss: 0.44451451301574707 2023-01-23 08:08:13.352090: step: 1400/1541, loss: 0.18523533642292023 2023-01-23 08:08:14.248380: step: 1402/1541, loss: 0.460363507270813 2023-01-23 08:08:15.212695: step: 1404/1541, loss: 0.7614783048629761 2023-01-23 08:08:16.101492: step: 1406/1541, loss: 0.20105844736099243 2023-01-23 08:08:16.988468: step: 1408/1541, loss: 0.3210218846797943 2023-01-23 08:08:17.919142: step: 1410/1541, loss: 0.7947766780853271 2023-01-23 08:08:18.829768: step: 1412/1541, loss: 0.5070210099220276 2023-01-23 08:08:19.704849: step: 1414/1541, loss: 4.9049811363220215 2023-01-23 08:08:20.601001: step: 1416/1541, loss: 0.6157891750335693 2023-01-23 08:08:21.549135: step: 1418/1541, loss: 1.240328311920166 2023-01-23 08:08:22.426012: step: 1420/1541, loss: 0.4867849051952362 2023-01-23 08:08:23.334580: step: 1422/1541, loss: 0.2951143682003021 2023-01-23 08:08:24.226795: step: 1424/1541, loss: 0.31082743406295776 2023-01-23 08:08:25.139405: step: 1426/1541, loss: 0.4674395024776459 2023-01-23 08:08:26.016472: step: 1428/1541, loss: 0.4037292003631592 2023-01-23 08:08:26.909110: step: 1430/1541, loss: 0.4171442985534668 2023-01-23 08:08:27.907037: step: 1432/1541, loss: 0.29033544659614563 2023-01-23 08:08:28.823959: step: 1434/1541, loss: 0.4215807020664215 2023-01-23 08:08:29.684487: step: 1436/1541, loss: 0.814064085483551 2023-01-23 08:08:30.588404: step: 1438/1541, loss: 0.648278534412384 2023-01-23 08:08:31.516725: step: 1440/1541, loss: 1.103895902633667 2023-01-23 08:08:32.450355: step: 1442/1541, loss: 0.35585474967956543 2023-01-23 08:08:33.368773: step: 1444/1541, loss: 0.4254433214664459 2023-01-23 08:08:34.236745: step: 1446/1541, loss: 0.12388462573289871 2023-01-23 08:08:35.131177: step: 1448/1541, loss: 0.7717466950416565 2023-01-23 08:08:35.995782: step: 1450/1541, loss: 0.3055998384952545 2023-01-23 08:08:36.892681: step: 1452/1541, loss: 0.25178074836730957 2023-01-23 08:08:37.741747: step: 1454/1541, loss: 0.46635282039642334 2023-01-23 08:08:38.645809: step: 1456/1541, loss: 0.36511871218681335 2023-01-23 08:08:39.568271: step: 1458/1541, loss: 0.1992773711681366 2023-01-23 08:08:40.422221: step: 1460/1541, loss: 0.13046585023403168 2023-01-23 08:08:41.307410: step: 1462/1541, loss: 0.28979167342185974 2023-01-23 08:08:42.208519: step: 1464/1541, loss: 0.592157781124115 2023-01-23 08:08:43.097349: step: 1466/1541, loss: 0.2991087734699249 2023-01-23 08:08:43.973941: step: 1468/1541, loss: 0.3025141656398773 2023-01-23 08:08:44.824260: step: 1470/1541, loss: 0.13918721675872803 2023-01-23 08:08:45.699375: step: 1472/1541, loss: 0.5022915601730347 2023-01-23 08:08:46.592322: step: 1474/1541, loss: 1.0522947311401367 2023-01-23 08:08:47.595086: step: 1476/1541, loss: 0.8561190366744995 2023-01-23 08:08:48.511038: step: 1478/1541, loss: 0.28851044178009033 2023-01-23 08:08:49.407592: step: 1480/1541, loss: 0.21742379665374756 2023-01-23 08:08:50.300090: step: 1482/1541, loss: 1.0457919836044312 2023-01-23 08:08:51.206057: step: 1484/1541, loss: 0.12185253947973251 2023-01-23 08:08:52.102628: step: 1486/1541, loss: 1.8798619508743286 2023-01-23 08:08:53.020096: step: 1488/1541, loss: 0.46604084968566895 2023-01-23 08:08:53.886544: step: 1490/1541, loss: 0.21549098193645477 2023-01-23 08:08:54.777507: step: 1492/1541, loss: 0.4875069558620453 2023-01-23 08:08:55.665722: step: 1494/1541, loss: 0.12393373250961304 2023-01-23 08:08:56.638605: step: 1496/1541, loss: 0.3794592320919037 2023-01-23 08:08:57.534965: step: 1498/1541, loss: 0.3662279546260834 2023-01-23 08:08:58.480545: step: 1500/1541, loss: 0.5055462121963501 2023-01-23 08:08:59.364259: step: 1502/1541, loss: 0.473184734582901 2023-01-23 08:09:00.265316: step: 1504/1541, loss: 0.6825792789459229 2023-01-23 08:09:01.159786: step: 1506/1541, loss: 0.12140989303588867 2023-01-23 08:09:02.053739: step: 1508/1541, loss: 0.7157256603240967 2023-01-23 08:09:02.991286: step: 1510/1541, loss: 0.2552787959575653 2023-01-23 08:09:03.882770: step: 1512/1541, loss: 0.6916846036911011 2023-01-23 08:09:04.813186: step: 1514/1541, loss: 0.3168066740036011 2023-01-23 08:09:05.802274: step: 1516/1541, loss: 0.4814586937427521 2023-01-23 08:09:06.718453: step: 1518/1541, loss: 0.3260924518108368 2023-01-23 08:09:07.593957: step: 1520/1541, loss: 0.46098577976226807 2023-01-23 08:09:08.480116: step: 1522/1541, loss: 1.124732494354248 2023-01-23 08:09:09.412673: step: 1524/1541, loss: 0.1887371689081192 2023-01-23 08:09:10.316331: step: 1526/1541, loss: 0.38815832138061523 2023-01-23 08:09:11.243747: step: 1528/1541, loss: 1.3729760646820068 2023-01-23 08:09:12.175185: step: 1530/1541, loss: 0.1724107414484024 2023-01-23 08:09:13.109152: step: 1532/1541, loss: 0.33856308460235596 2023-01-23 08:09:14.055780: step: 1534/1541, loss: 0.526258647441864 2023-01-23 08:09:14.972356: step: 1536/1541, loss: 0.31828173995018005 2023-01-23 08:09:15.877419: step: 1538/1541, loss: 0.46547731757164 2023-01-23 08:09:16.792013: step: 1540/1541, loss: 0.2913901209831238 2023-01-23 08:09:17.686683: step: 1542/1541, loss: 0.15991784632205963 2023-01-23 08:09:18.560087: step: 1544/1541, loss: 0.4026738405227661 2023-01-23 08:09:19.443547: step: 1546/1541, loss: 0.5085973143577576 2023-01-23 08:09:20.347088: step: 1548/1541, loss: 0.5442731380462646 2023-01-23 08:09:21.244928: step: 1550/1541, loss: 0.2827763557434082 2023-01-23 08:09:22.155732: step: 1552/1541, loss: 0.2696071267127991 2023-01-23 08:09:23.060759: step: 1554/1541, loss: 0.5832716822624207 2023-01-23 08:09:23.937243: step: 1556/1541, loss: 0.0731586366891861 2023-01-23 08:09:24.826965: step: 1558/1541, loss: 0.38053226470947266 2023-01-23 08:09:25.725098: step: 1560/1541, loss: 0.090195432305336 2023-01-23 08:09:26.631792: step: 1562/1541, loss: 0.6481048464775085 2023-01-23 08:09:27.496562: step: 1564/1541, loss: 0.302064448595047 2023-01-23 08:09:28.466285: step: 1566/1541, loss: 0.31314533948898315 2023-01-23 08:09:29.376223: step: 1568/1541, loss: 0.4487544894218445 2023-01-23 08:09:30.327112: step: 1570/1541, loss: 4.094618797302246 2023-01-23 08:09:31.254818: step: 1572/1541, loss: 1.426599383354187 2023-01-23 08:09:32.223804: step: 1574/1541, loss: 0.2510005831718445 2023-01-23 08:09:33.170545: step: 1576/1541, loss: 2.2623579502105713 2023-01-23 08:09:34.099757: step: 1578/1541, loss: 0.39370912313461304 2023-01-23 08:09:35.020544: step: 1580/1541, loss: 0.25355154275894165 2023-01-23 08:09:35.984861: step: 1582/1541, loss: 0.3027038872241974 2023-01-23 08:09:36.913179: step: 1584/1541, loss: 0.3077659606933594 2023-01-23 08:09:37.782629: step: 1586/1541, loss: 0.22662651538848877 2023-01-23 08:09:38.686961: step: 1588/1541, loss: 2.040287733078003 2023-01-23 08:09:39.576406: step: 1590/1541, loss: 0.12021895498037338 2023-01-23 08:09:40.453577: step: 1592/1541, loss: 0.21474123001098633 2023-01-23 08:09:41.327015: step: 1594/1541, loss: 0.04873400926589966 2023-01-23 08:09:42.287361: step: 1596/1541, loss: 0.35377195477485657 2023-01-23 08:09:43.152012: step: 1598/1541, loss: 0.9300166368484497 2023-01-23 08:09:44.054299: step: 1600/1541, loss: 0.6654055118560791 2023-01-23 08:09:44.930566: step: 1602/1541, loss: 0.14870871603488922 2023-01-23 08:09:45.832935: step: 1604/1541, loss: 0.2454858422279358 2023-01-23 08:09:46.742655: step: 1606/1541, loss: 0.517952561378479 2023-01-23 08:09:47.613394: step: 1608/1541, loss: 0.18238888680934906 2023-01-23 08:09:48.508606: step: 1610/1541, loss: 0.12312717735767365 2023-01-23 08:09:49.430751: step: 1612/1541, loss: 0.3461286127567291 2023-01-23 08:09:50.308128: step: 1614/1541, loss: 0.07638813555240631 2023-01-23 08:09:51.192226: step: 1616/1541, loss: 0.10343046486377716 2023-01-23 08:09:52.110404: step: 1618/1541, loss: 0.6030557751655579 2023-01-23 08:09:53.020849: step: 1620/1541, loss: 0.3337254822254181 2023-01-23 08:09:53.892180: step: 1622/1541, loss: 0.26383689045906067 2023-01-23 08:09:54.776541: step: 1624/1541, loss: 0.5391681790351868 2023-01-23 08:09:55.639264: step: 1626/1541, loss: 0.40140271186828613 2023-01-23 08:09:56.548133: step: 1628/1541, loss: 2.6833877563476562 2023-01-23 08:09:57.434054: step: 1630/1541, loss: 0.574533224105835 2023-01-23 08:09:58.342781: step: 1632/1541, loss: 0.5443352460861206 2023-01-23 08:09:59.245278: step: 1634/1541, loss: 0.8032952547073364 2023-01-23 08:10:00.166975: step: 1636/1541, loss: 0.07158435136079788 2023-01-23 08:10:01.099223: step: 1638/1541, loss: 0.2770496606826782 2023-01-23 08:10:01.994604: step: 1640/1541, loss: 0.4940796196460724 2023-01-23 08:10:02.905483: step: 1642/1541, loss: 0.48761293292045593 2023-01-23 08:10:03.818331: step: 1644/1541, loss: 0.6213646531105042 2023-01-23 08:10:04.719869: step: 1646/1541, loss: 0.11613161116838455 2023-01-23 08:10:05.613267: step: 1648/1541, loss: 0.22067156434059143 2023-01-23 08:10:06.531911: step: 1650/1541, loss: 0.1821424812078476 2023-01-23 08:10:07.432538: step: 1652/1541, loss: 0.45530828833580017 2023-01-23 08:10:08.322334: step: 1654/1541, loss: 0.5064567923545837 2023-01-23 08:10:09.186894: step: 1656/1541, loss: 0.4928196966648102 2023-01-23 08:10:10.073424: step: 1658/1541, loss: 0.698341429233551 2023-01-23 08:10:10.957617: step: 1660/1541, loss: 0.5419192910194397 2023-01-23 08:10:11.878569: step: 1662/1541, loss: 0.2140466868877411 2023-01-23 08:10:12.823778: step: 1664/1541, loss: 0.3653814196586609 2023-01-23 08:10:13.721931: step: 1666/1541, loss: 1.081039547920227 2023-01-23 08:10:14.609318: step: 1668/1541, loss: 0.2627686858177185 2023-01-23 08:10:15.499699: step: 1670/1541, loss: 0.14035992324352264 2023-01-23 08:10:16.405461: step: 1672/1541, loss: 0.19915711879730225 2023-01-23 08:10:17.285143: step: 1674/1541, loss: 0.1565428525209427 2023-01-23 08:10:18.229223: step: 1676/1541, loss: 0.6519702672958374 2023-01-23 08:10:19.102964: step: 1678/1541, loss: 0.6196927428245544 2023-01-23 08:10:20.015798: step: 1680/1541, loss: 0.8471708297729492 2023-01-23 08:10:20.908604: step: 1682/1541, loss: 0.5266394019126892 2023-01-23 08:10:21.887588: step: 1684/1541, loss: 1.4853723049163818 2023-01-23 08:10:22.817917: step: 1686/1541, loss: 0.593542218208313 2023-01-23 08:10:23.687952: step: 1688/1541, loss: 0.2550717890262604 2023-01-23 08:10:24.603528: step: 1690/1541, loss: 0.6854516863822937 2023-01-23 08:10:25.542490: step: 1692/1541, loss: 0.7127408981323242 2023-01-23 08:10:26.429860: step: 1694/1541, loss: 0.6967059373855591 2023-01-23 08:10:27.342600: step: 1696/1541, loss: 0.21782267093658447 2023-01-23 08:10:28.308330: step: 1698/1541, loss: 0.3847789168357849 2023-01-23 08:10:29.210782: step: 1700/1541, loss: 0.35444366931915283 2023-01-23 08:10:30.155753: step: 1702/1541, loss: 0.5638971924781799 2023-01-23 08:10:31.029062: step: 1704/1541, loss: 0.9421465992927551 2023-01-23 08:10:31.952875: step: 1706/1541, loss: 1.4487799406051636 2023-01-23 08:10:32.815100: step: 1708/1541, loss: 0.12188050895929337 2023-01-23 08:10:33.741380: step: 1710/1541, loss: 0.31764841079711914 2023-01-23 08:10:34.645543: step: 1712/1541, loss: 0.18556039035320282 2023-01-23 08:10:35.587128: step: 1714/1541, loss: 0.9724331498146057 2023-01-23 08:10:36.468103: step: 1716/1541, loss: 0.2311372607946396 2023-01-23 08:10:37.323790: step: 1718/1541, loss: 0.7424470782279968 2023-01-23 08:10:38.199696: step: 1720/1541, loss: 0.12929148972034454 2023-01-23 08:10:39.084431: step: 1722/1541, loss: 0.29430079460144043 2023-01-23 08:10:39.969578: step: 1724/1541, loss: 0.7749097347259521 2023-01-23 08:10:40.844731: step: 1726/1541, loss: 0.37104928493499756 2023-01-23 08:10:41.727001: step: 1728/1541, loss: 0.1500546932220459 2023-01-23 08:10:42.617503: step: 1730/1541, loss: 6.3805131912231445 2023-01-23 08:10:43.591631: step: 1732/1541, loss: 0.3533031940460205 2023-01-23 08:10:44.511709: step: 1734/1541, loss: 0.1606772243976593 2023-01-23 08:10:45.517422: step: 1736/1541, loss: 1.9073394536972046 2023-01-23 08:10:46.448264: step: 1738/1541, loss: 0.40120208263397217 2023-01-23 08:10:47.331059: step: 1740/1541, loss: 0.3694324493408203 2023-01-23 08:10:48.322249: step: 1742/1541, loss: 0.08487772941589355 2023-01-23 08:10:49.236745: step: 1744/1541, loss: 0.3264440894126892 2023-01-23 08:10:50.124171: step: 1746/1541, loss: 0.23353765904903412 2023-01-23 08:10:51.160547: step: 1748/1541, loss: 1.2122468948364258 2023-01-23 08:10:52.065268: step: 1750/1541, loss: 0.30692705512046814 2023-01-23 08:10:52.982264: step: 1752/1541, loss: 0.3421045243740082 2023-01-23 08:10:53.879307: step: 1754/1541, loss: 0.7973811030387878 2023-01-23 08:10:54.825822: step: 1756/1541, loss: 0.37619441747665405 2023-01-23 08:10:55.734205: step: 1758/1541, loss: 0.18830746412277222 2023-01-23 08:10:56.626905: step: 1760/1541, loss: 0.4930318593978882 2023-01-23 08:10:57.485112: step: 1762/1541, loss: 0.33675533533096313 2023-01-23 08:10:58.403145: step: 1764/1541, loss: 0.2746330797672272 2023-01-23 08:10:59.269536: step: 1766/1541, loss: 0.6306996941566467 2023-01-23 08:11:00.178913: step: 1768/1541, loss: 0.3049590587615967 2023-01-23 08:11:01.093191: step: 1770/1541, loss: 0.1275632083415985 2023-01-23 08:11:01.994526: step: 1772/1541, loss: 0.5178795456886292 2023-01-23 08:11:02.866402: step: 1774/1541, loss: 0.47180360555648804 2023-01-23 08:11:03.753786: step: 1776/1541, loss: 0.5803499221801758 2023-01-23 08:11:04.616008: step: 1778/1541, loss: 0.9570881724357605 2023-01-23 08:11:05.485035: step: 1780/1541, loss: 0.1984235644340515 2023-01-23 08:11:06.381287: step: 1782/1541, loss: 0.7167362570762634 2023-01-23 08:11:07.250112: step: 1784/1541, loss: 0.43983906507492065 2023-01-23 08:11:08.123136: step: 1786/1541, loss: 0.1715337038040161 2023-01-23 08:11:09.053878: step: 1788/1541, loss: 0.41167065501213074 2023-01-23 08:11:09.955126: step: 1790/1541, loss: 2.7641348838806152 2023-01-23 08:11:10.856178: step: 1792/1541, loss: 0.41833966970443726 2023-01-23 08:11:11.747690: step: 1794/1541, loss: 0.41295814514160156 2023-01-23 08:11:12.628804: step: 1796/1541, loss: 0.21053987741470337 2023-01-23 08:11:13.501043: step: 1798/1541, loss: 0.4137973487377167 2023-01-23 08:11:14.427176: step: 1800/1541, loss: 0.46118617057800293 2023-01-23 08:11:15.319431: step: 1802/1541, loss: 0.15275631844997406 2023-01-23 08:11:16.204049: step: 1804/1541, loss: 0.42150065302848816 2023-01-23 08:11:17.107180: step: 1806/1541, loss: 0.622704267501831 2023-01-23 08:11:18.008789: step: 1808/1541, loss: 0.5261006951332092 2023-01-23 08:11:18.900701: step: 1810/1541, loss: 0.23235934972763062 2023-01-23 08:11:19.763521: step: 1812/1541, loss: 0.30753621459007263 2023-01-23 08:11:20.643416: step: 1814/1541, loss: 0.20987842977046967 2023-01-23 08:11:21.527004: step: 1816/1541, loss: 0.7734481692314148 2023-01-23 08:11:22.401409: step: 1818/1541, loss: 0.13386915624141693 2023-01-23 08:11:23.259476: step: 1820/1541, loss: 0.578964352607727 2023-01-23 08:11:24.136174: step: 1822/1541, loss: 0.46745043992996216 2023-01-23 08:11:25.045326: step: 1824/1541, loss: 0.48281729221343994 2023-01-23 08:11:25.939675: step: 1826/1541, loss: 0.8261612057685852 2023-01-23 08:11:26.822684: step: 1828/1541, loss: 0.4423884153366089 2023-01-23 08:11:27.704471: step: 1830/1541, loss: 0.3945164382457733 2023-01-23 08:11:28.598355: step: 1832/1541, loss: 0.3215019702911377 2023-01-23 08:11:29.514222: step: 1834/1541, loss: 0.743295431137085 2023-01-23 08:11:30.386957: step: 1836/1541, loss: 0.22782263159751892 2023-01-23 08:11:31.367450: step: 1838/1541, loss: 0.30856049060821533 2023-01-23 08:11:32.258081: step: 1840/1541, loss: 0.18022757768630981 2023-01-23 08:11:33.133644: step: 1842/1541, loss: 2.9668869972229004 2023-01-23 08:11:34.087139: step: 1844/1541, loss: 0.1301700323820114 2023-01-23 08:11:35.054069: step: 1846/1541, loss: 0.27784594893455505 2023-01-23 08:11:35.987098: step: 1848/1541, loss: 0.21994799375534058 2023-01-23 08:11:36.921157: step: 1850/1541, loss: 0.18069951236248016 2023-01-23 08:11:37.805195: step: 1852/1541, loss: 0.5730895400047302 2023-01-23 08:11:38.710108: step: 1854/1541, loss: 0.5605595111846924 2023-01-23 08:11:39.633511: step: 1856/1541, loss: 0.19338136911392212 2023-01-23 08:11:40.540076: step: 1858/1541, loss: 1.9561212062835693 2023-01-23 08:11:41.431147: step: 1860/1541, loss: 0.21856805682182312 2023-01-23 08:11:42.353268: step: 1862/1541, loss: 0.1892652064561844 2023-01-23 08:11:43.245003: step: 1864/1541, loss: 0.33041176199913025 2023-01-23 08:11:44.165864: step: 1866/1541, loss: 0.35939157009124756 2023-01-23 08:11:45.129457: step: 1868/1541, loss: 0.2855517566204071 2023-01-23 08:11:46.007599: step: 1870/1541, loss: 0.1265704184770584 2023-01-23 08:11:46.899736: step: 1872/1541, loss: 2.0167810916900635 2023-01-23 08:11:47.794314: step: 1874/1541, loss: 0.12903626263141632 2023-01-23 08:11:48.682975: step: 1876/1541, loss: 1.9192875623703003 2023-01-23 08:11:49.578625: step: 1878/1541, loss: 0.18942326307296753 2023-01-23 08:11:50.495279: step: 1880/1541, loss: 0.20389850437641144 2023-01-23 08:11:51.359840: step: 1882/1541, loss: 0.366043359041214 2023-01-23 08:11:52.315928: step: 1884/1541, loss: 0.15637211501598358 2023-01-23 08:11:53.215802: step: 1886/1541, loss: 0.44786393642425537 2023-01-23 08:11:54.105365: step: 1888/1541, loss: 0.4012938141822815 2023-01-23 08:11:54.965014: step: 1890/1541, loss: 0.3843314051628113 2023-01-23 08:11:55.898215: step: 1892/1541, loss: 1.9039593935012817 2023-01-23 08:11:56.790574: step: 1894/1541, loss: 0.25778916478157043 2023-01-23 08:11:57.659419: step: 1896/1541, loss: 0.23149433732032776 2023-01-23 08:11:58.557359: step: 1898/1541, loss: 0.2231021523475647 2023-01-23 08:11:59.456786: step: 1900/1541, loss: 0.5095162391662598 2023-01-23 08:12:00.357767: step: 1902/1541, loss: 0.32144322991371155 2023-01-23 08:12:01.273880: step: 1904/1541, loss: 0.45037609338760376 2023-01-23 08:12:02.214300: step: 1906/1541, loss: 0.35783907771110535 2023-01-23 08:12:03.126424: step: 1908/1541, loss: 0.1255483776330948 2023-01-23 08:12:04.024609: step: 1910/1541, loss: 0.5776127576828003 2023-01-23 08:12:04.931064: step: 1912/1541, loss: 0.31916192173957825 2023-01-23 08:12:05.907374: step: 1914/1541, loss: 0.9212679266929626 2023-01-23 08:12:06.825046: step: 1916/1541, loss: 0.4374299943447113 2023-01-23 08:12:07.711662: step: 1918/1541, loss: 0.5092484354972839 2023-01-23 08:12:08.580195: step: 1920/1541, loss: 0.24420537054538727 2023-01-23 08:12:09.503501: step: 1922/1541, loss: 1.1992372274398804 2023-01-23 08:12:10.420378: step: 1924/1541, loss: 0.578252375125885 2023-01-23 08:12:11.339121: step: 1926/1541, loss: 0.6845478415489197 2023-01-23 08:12:12.228244: step: 1928/1541, loss: 0.550835371017456 2023-01-23 08:12:13.082571: step: 1930/1541, loss: 0.17312145233154297 2023-01-23 08:12:13.982141: step: 1932/1541, loss: 0.7207087874412537 2023-01-23 08:12:14.877665: step: 1934/1541, loss: 1.0596437454223633 2023-01-23 08:12:15.767880: step: 1936/1541, loss: 0.5228402018547058 2023-01-23 08:12:16.683770: step: 1938/1541, loss: 0.8408893942832947 2023-01-23 08:12:17.594572: step: 1940/1541, loss: 0.7478454113006592 2023-01-23 08:12:18.565450: step: 1942/1541, loss: 0.4956376850605011 2023-01-23 08:12:19.457846: step: 1944/1541, loss: 0.3363392949104309 2023-01-23 08:12:20.345507: step: 1946/1541, loss: 0.3426463007926941 2023-01-23 08:12:21.218787: step: 1948/1541, loss: 0.3291395604610443 2023-01-23 08:12:22.069564: step: 1950/1541, loss: 0.174386665225029 2023-01-23 08:12:23.018584: step: 1952/1541, loss: 0.6457948684692383 2023-01-23 08:12:23.934249: step: 1954/1541, loss: 0.14186276495456696 2023-01-23 08:12:24.873740: step: 1956/1541, loss: 0.34753894805908203 2023-01-23 08:12:25.771174: step: 1958/1541, loss: 0.8727962374687195 2023-01-23 08:12:26.682347: step: 1960/1541, loss: 0.4014109969139099 2023-01-23 08:12:27.597842: step: 1962/1541, loss: 2.603300094604492 2023-01-23 08:12:28.463977: step: 1964/1541, loss: 0.5909099578857422 2023-01-23 08:12:29.360827: step: 1966/1541, loss: 2.1943747997283936 2023-01-23 08:12:30.236123: step: 1968/1541, loss: 0.8897308707237244 2023-01-23 08:12:31.166016: step: 1970/1541, loss: 0.827364444732666 2023-01-23 08:12:32.050966: step: 1972/1541, loss: 0.38852590322494507 2023-01-23 08:12:32.920649: step: 1974/1541, loss: 0.38406965136528015 2023-01-23 08:12:33.859837: step: 1976/1541, loss: 0.3305070102214813 2023-01-23 08:12:34.826760: step: 1978/1541, loss: 1.1814570426940918 2023-01-23 08:12:35.736995: step: 1980/1541, loss: 0.9676007628440857 2023-01-23 08:12:36.647693: step: 1982/1541, loss: 0.3401069641113281 2023-01-23 08:12:37.514766: step: 1984/1541, loss: 0.4355493187904358 2023-01-23 08:12:38.428297: step: 1986/1541, loss: 0.2401835024356842 2023-01-23 08:12:39.339681: step: 1988/1541, loss: 0.1874694526195526 2023-01-23 08:12:40.237320: step: 1990/1541, loss: 0.19759489595890045 2023-01-23 08:12:41.215335: step: 1992/1541, loss: 0.1901656985282898 2023-01-23 08:12:42.098970: step: 1994/1541, loss: 0.37754663825035095 2023-01-23 08:12:42.943707: step: 1996/1541, loss: 0.08067886531352997 2023-01-23 08:12:43.839540: step: 1998/1541, loss: 0.8779184818267822 2023-01-23 08:12:44.800840: step: 2000/1541, loss: 0.37494614720344543 2023-01-23 08:12:45.686891: step: 2002/1541, loss: 0.6032521724700928 2023-01-23 08:12:46.581651: step: 2004/1541, loss: 0.9011502861976624 2023-01-23 08:12:47.451355: step: 2006/1541, loss: 0.771431565284729 2023-01-23 08:12:48.372565: step: 2008/1541, loss: 0.5954151749610901 2023-01-23 08:12:49.269707: step: 2010/1541, loss: 0.1061248779296875 2023-01-23 08:12:50.239250: step: 2012/1541, loss: 0.7599334716796875 2023-01-23 08:12:51.128859: step: 2014/1541, loss: 0.19542542099952698 2023-01-23 08:12:52.103921: step: 2016/1541, loss: 1.8331454992294312 2023-01-23 08:12:52.951241: step: 2018/1541, loss: 0.36870861053466797 2023-01-23 08:12:53.873088: step: 2020/1541, loss: 0.3996354341506958 2023-01-23 08:12:54.813965: step: 2022/1541, loss: 0.38184306025505066 2023-01-23 08:12:55.675833: step: 2024/1541, loss: 0.8622600436210632 2023-01-23 08:12:56.576264: step: 2026/1541, loss: 0.3413827121257782 2023-01-23 08:12:57.459449: step: 2028/1541, loss: 0.703942060470581 2023-01-23 08:12:58.369381: step: 2030/1541, loss: 0.22201552987098694 2023-01-23 08:12:59.301980: step: 2032/1541, loss: 0.3346216380596161 2023-01-23 08:13:00.227446: step: 2034/1541, loss: 1.221659779548645 2023-01-23 08:13:01.133674: step: 2036/1541, loss: 0.17263972759246826 2023-01-23 08:13:02.008321: step: 2038/1541, loss: 0.17381680011749268 2023-01-23 08:13:02.876415: step: 2040/1541, loss: 0.5262687802314758 2023-01-23 08:13:03.750519: step: 2042/1541, loss: 0.33701246976852417 2023-01-23 08:13:04.749844: step: 2044/1541, loss: 0.1891414076089859 2023-01-23 08:13:05.685518: step: 2046/1541, loss: 0.8200699090957642 2023-01-23 08:13:06.535187: step: 2048/1541, loss: 3.09039306640625 2023-01-23 08:13:07.440112: step: 2050/1541, loss: 0.19774585962295532 2023-01-23 08:13:08.350113: step: 2052/1541, loss: 0.20590990781784058 2023-01-23 08:13:09.218865: step: 2054/1541, loss: 1.1804803609848022 2023-01-23 08:13:10.138511: step: 2056/1541, loss: 0.2165505439043045 2023-01-23 08:13:11.025875: step: 2058/1541, loss: 0.5142730474472046 2023-01-23 08:13:11.896772: step: 2060/1541, loss: 0.2656327486038208 2023-01-23 08:13:12.773090: step: 2062/1541, loss: 0.30951741337776184 2023-01-23 08:13:13.711380: step: 2064/1541, loss: 0.4499209225177765 2023-01-23 08:13:14.623671: step: 2066/1541, loss: 0.14491242170333862 2023-01-23 08:13:15.506349: step: 2068/1541, loss: 0.1977548748254776 2023-01-23 08:13:16.383969: step: 2070/1541, loss: 0.20839262008666992 2023-01-23 08:13:17.274172: step: 2072/1541, loss: 0.7247453927993774 2023-01-23 08:13:18.195093: step: 2074/1541, loss: 0.3582303524017334 2023-01-23 08:13:19.133547: step: 2076/1541, loss: 0.13087011873722076 2023-01-23 08:13:20.004402: step: 2078/1541, loss: 0.6270990371704102 2023-01-23 08:13:20.903000: step: 2080/1541, loss: 0.2907748818397522 2023-01-23 08:13:21.814786: step: 2082/1541, loss: 0.16422948241233826 2023-01-23 08:13:22.718149: step: 2084/1541, loss: 0.6772700548171997 2023-01-23 08:13:23.596645: step: 2086/1541, loss: 1.2779960632324219 2023-01-23 08:13:24.525916: step: 2088/1541, loss: 0.31076571345329285 2023-01-23 08:13:25.430023: step: 2090/1541, loss: 0.2998512089252472 2023-01-23 08:13:26.307080: step: 2092/1541, loss: 0.7942596077919006 2023-01-23 08:13:27.199878: step: 2094/1541, loss: 0.3274824023246765 2023-01-23 08:13:28.080601: step: 2096/1541, loss: 0.3708515167236328 2023-01-23 08:13:29.007578: step: 2098/1541, loss: 0.45679593086242676 2023-01-23 08:13:29.915628: step: 2100/1541, loss: 0.28677767515182495 2023-01-23 08:13:30.851698: step: 2102/1541, loss: 0.32683128118515015 2023-01-23 08:13:31.754350: step: 2104/1541, loss: 0.5229082703590393 2023-01-23 08:13:32.665928: step: 2106/1541, loss: 0.29575225710868835 2023-01-23 08:13:33.544716: step: 2108/1541, loss: 0.708440899848938 2023-01-23 08:13:34.406593: step: 2110/1541, loss: 0.2939382493495941 2023-01-23 08:13:35.322508: step: 2112/1541, loss: 0.3038008511066437 2023-01-23 08:13:36.249874: step: 2114/1541, loss: 0.7603515982627869 2023-01-23 08:13:37.122998: step: 2116/1541, loss: 0.46666377782821655 2023-01-23 08:13:38.011850: step: 2118/1541, loss: 0.23953543603420258 2023-01-23 08:13:38.929535: step: 2120/1541, loss: 0.47807347774505615 2023-01-23 08:13:39.799179: step: 2122/1541, loss: 1.829309344291687 2023-01-23 08:13:40.704064: step: 2124/1541, loss: 0.85727858543396 2023-01-23 08:13:41.634934: step: 2126/1541, loss: 0.18524307012557983 2023-01-23 08:13:42.589997: step: 2128/1541, loss: 0.222581148147583 2023-01-23 08:13:43.511653: step: 2130/1541, loss: 0.17694586515426636 2023-01-23 08:13:44.377467: step: 2132/1541, loss: 0.7531309723854065 2023-01-23 08:13:45.264510: step: 2134/1541, loss: 0.26946309208869934 2023-01-23 08:13:46.136338: step: 2136/1541, loss: 0.9307460188865662 2023-01-23 08:13:47.025882: step: 2138/1541, loss: 0.17104892432689667 2023-01-23 08:13:47.849337: step: 2140/1541, loss: 0.3284353017807007 2023-01-23 08:13:48.764646: step: 2142/1541, loss: 0.2511015832424164 2023-01-23 08:13:49.643714: step: 2144/1541, loss: 0.5611926317214966 2023-01-23 08:13:50.555372: step: 2146/1541, loss: 0.5495691895484924 2023-01-23 08:13:51.433547: step: 2148/1541, loss: 0.9434720873832703 2023-01-23 08:13:52.333395: step: 2150/1541, loss: 0.14755043387413025 2023-01-23 08:13:53.210476: step: 2152/1541, loss: 0.29402679204940796 2023-01-23 08:13:54.178939: step: 2154/1541, loss: 0.2899833619594574 2023-01-23 08:13:55.062661: step: 2156/1541, loss: 0.31260937452316284 2023-01-23 08:13:55.956160: step: 2158/1541, loss: 0.25983378291130066 2023-01-23 08:13:56.923941: step: 2160/1541, loss: 0.34930652379989624 2023-01-23 08:13:57.794394: step: 2162/1541, loss: 0.2278863489627838 2023-01-23 08:13:58.753533: step: 2164/1541, loss: 0.9037919640541077 2023-01-23 08:13:59.702312: step: 2166/1541, loss: 0.197544127702713 2023-01-23 08:14:00.561580: step: 2168/1541, loss: 1.1624962091445923 2023-01-23 08:14:01.431355: step: 2170/1541, loss: 2.8459525108337402 2023-01-23 08:14:02.323940: step: 2172/1541, loss: 1.078050971031189 2023-01-23 08:14:03.217333: step: 2174/1541, loss: 0.29300057888031006 2023-01-23 08:14:04.130272: step: 2176/1541, loss: 0.37265491485595703 2023-01-23 08:14:05.052618: step: 2178/1541, loss: 0.3063987195491791 2023-01-23 08:14:06.006968: step: 2180/1541, loss: 0.5973526835441589 2023-01-23 08:14:06.893780: step: 2182/1541, loss: 0.2951985001564026 2023-01-23 08:14:07.770027: step: 2184/1541, loss: 0.1865624189376831 2023-01-23 08:14:08.762135: step: 2186/1541, loss: 0.4538460075855255 2023-01-23 08:14:09.703915: step: 2188/1541, loss: 5.408015251159668 2023-01-23 08:14:10.580552: step: 2190/1541, loss: 0.5049492716789246 2023-01-23 08:14:11.482027: step: 2192/1541, loss: 0.43929868936538696 2023-01-23 08:14:12.382907: step: 2194/1541, loss: 0.1757894605398178 2023-01-23 08:14:13.291657: step: 2196/1541, loss: 0.27726808190345764 2023-01-23 08:14:14.222849: step: 2198/1541, loss: 0.21720314025878906 2023-01-23 08:14:15.139341: step: 2200/1541, loss: 0.4699747562408447 2023-01-23 08:14:16.043982: step: 2202/1541, loss: 0.6674829721450806 2023-01-23 08:14:16.977112: step: 2204/1541, loss: 0.2577476501464844 2023-01-23 08:14:17.860420: step: 2206/1541, loss: 0.10521957278251648 2023-01-23 08:14:18.743148: step: 2208/1541, loss: 0.3259929418563843 2023-01-23 08:14:19.606614: step: 2210/1541, loss: 0.5821354389190674 2023-01-23 08:14:20.496894: step: 2212/1541, loss: 1.0383723974227905 2023-01-23 08:14:21.389316: step: 2214/1541, loss: 1.1998943090438843 2023-01-23 08:14:22.330872: step: 2216/1541, loss: 0.3945557177066803 2023-01-23 08:14:23.198686: step: 2218/1541, loss: 0.12839902937412262 2023-01-23 08:14:24.122510: step: 2220/1541, loss: 0.19056475162506104 2023-01-23 08:14:24.953592: step: 2222/1541, loss: 0.5152431726455688 2023-01-23 08:14:25.865620: step: 2224/1541, loss: 0.6472674608230591 2023-01-23 08:14:26.748268: step: 2226/1541, loss: 0.32214024662971497 2023-01-23 08:14:27.606878: step: 2228/1541, loss: 0.2369455248117447 2023-01-23 08:14:28.510809: step: 2230/1541, loss: 0.06562794744968414 2023-01-23 08:14:29.376559: step: 2232/1541, loss: 0.9433298707008362 2023-01-23 08:14:30.301171: step: 2234/1541, loss: 0.2614002227783203 2023-01-23 08:14:31.240407: step: 2236/1541, loss: 0.7540631890296936 2023-01-23 08:14:32.190682: step: 2238/1541, loss: 0.7071923613548279 2023-01-23 08:14:33.095814: step: 2240/1541, loss: 0.2777601480484009 2023-01-23 08:14:34.009873: step: 2242/1541, loss: 0.33888953924179077 2023-01-23 08:14:34.905502: step: 2244/1541, loss: 0.24205933511257172 2023-01-23 08:14:35.848072: step: 2246/1541, loss: 0.446771502494812 2023-01-23 08:14:36.739036: step: 2248/1541, loss: 0.5080034136772156 2023-01-23 08:14:37.678617: step: 2250/1541, loss: 0.7076008319854736 2023-01-23 08:14:38.540018: step: 2252/1541, loss: 0.2336609959602356 2023-01-23 08:14:39.394713: step: 2254/1541, loss: 0.5657228827476501 2023-01-23 08:14:40.284311: step: 2256/1541, loss: 0.20113377273082733 2023-01-23 08:14:41.221273: step: 2258/1541, loss: 0.33789119124412537 2023-01-23 08:14:42.123022: step: 2260/1541, loss: 0.21975889801979065 2023-01-23 08:14:42.998493: step: 2262/1541, loss: 0.2924900949001312 2023-01-23 08:14:43.898889: step: 2264/1541, loss: 0.10638170689344406 2023-01-23 08:14:44.768079: step: 2266/1541, loss: 1.0550544261932373 2023-01-23 08:14:45.632269: step: 2268/1541, loss: 0.4694465100765228 2023-01-23 08:14:46.543528: step: 2270/1541, loss: 0.09446468204259872 2023-01-23 08:14:47.439169: step: 2272/1541, loss: 0.2370164692401886 2023-01-23 08:14:48.320369: step: 2274/1541, loss: 0.633026123046875 2023-01-23 08:14:49.217425: step: 2276/1541, loss: 0.10303410142660141 2023-01-23 08:14:50.142885: step: 2278/1541, loss: 0.2297496497631073 2023-01-23 08:14:51.001527: step: 2280/1541, loss: 0.4575269818305969 2023-01-23 08:14:51.894022: step: 2282/1541, loss: 0.6322730779647827 2023-01-23 08:14:52.840372: step: 2284/1541, loss: 0.37269970774650574 2023-01-23 08:14:53.767741: step: 2286/1541, loss: 0.3094158172607422 2023-01-23 08:14:54.646765: step: 2288/1541, loss: 1.2406145334243774 2023-01-23 08:14:55.502477: step: 2290/1541, loss: 0.10751339793205261 2023-01-23 08:14:56.381425: step: 2292/1541, loss: 0.3772156834602356 2023-01-23 08:14:57.274699: step: 2294/1541, loss: 0.2253021001815796 2023-01-23 08:14:58.164794: step: 2296/1541, loss: 1.2976384162902832 2023-01-23 08:14:59.061284: step: 2298/1541, loss: 0.8557093739509583 2023-01-23 08:14:59.956100: step: 2300/1541, loss: 0.21377751231193542 2023-01-23 08:15:00.877549: step: 2302/1541, loss: 0.45715636014938354 2023-01-23 08:15:01.889835: step: 2304/1541, loss: 0.8232914209365845 2023-01-23 08:15:02.790618: step: 2306/1541, loss: 0.3027379810810089 2023-01-23 08:15:03.692916: step: 2308/1541, loss: 0.47753918170928955 2023-01-23 08:15:04.577172: step: 2310/1541, loss: 0.45889827609062195 2023-01-23 08:15:05.481999: step: 2312/1541, loss: 0.2833932042121887 2023-01-23 08:15:06.383774: step: 2314/1541, loss: 1.5725370645523071 2023-01-23 08:15:07.251540: step: 2316/1541, loss: 0.18132096529006958 2023-01-23 08:15:08.102842: step: 2318/1541, loss: 0.6151345372200012 2023-01-23 08:15:09.023631: step: 2320/1541, loss: 0.3792611360549927 2023-01-23 08:15:09.915417: step: 2322/1541, loss: 0.7154936790466309 2023-01-23 08:15:10.767182: step: 2324/1541, loss: 0.2134302705526352 2023-01-23 08:15:11.706851: step: 2326/1541, loss: 0.5239444375038147 2023-01-23 08:15:12.612381: step: 2328/1541, loss: 0.21578523516654968 2023-01-23 08:15:13.476620: step: 2330/1541, loss: 0.42290982604026794 2023-01-23 08:15:14.363158: step: 2332/1541, loss: 0.4022929072380066 2023-01-23 08:15:15.295155: step: 2334/1541, loss: 0.4764786660671234 2023-01-23 08:15:16.199188: step: 2336/1541, loss: 0.35197898745536804 2023-01-23 08:15:17.168889: step: 2338/1541, loss: 0.8903753757476807 2023-01-23 08:15:18.026047: step: 2340/1541, loss: 0.4492586553096771 2023-01-23 08:15:18.952710: step: 2342/1541, loss: 1.4296019077301025 2023-01-23 08:15:19.842123: step: 2344/1541, loss: 0.15995270013809204 2023-01-23 08:15:20.788255: step: 2346/1541, loss: 0.6593777537345886 2023-01-23 08:15:21.726276: step: 2348/1541, loss: 1.660221815109253 2023-01-23 08:15:22.666610: step: 2350/1541, loss: 1.0757331848144531 2023-01-23 08:15:23.580038: step: 2352/1541, loss: 0.45130011439323425 2023-01-23 08:15:24.472536: step: 2354/1541, loss: 0.31820303201675415 2023-01-23 08:15:25.407614: step: 2356/1541, loss: 0.9968035221099854 2023-01-23 08:15:26.289602: step: 2358/1541, loss: 0.2868610620498657 2023-01-23 08:15:27.182264: step: 2360/1541, loss: 0.2427171915769577 2023-01-23 08:15:28.046357: step: 2362/1541, loss: 0.24064801633358002 2023-01-23 08:15:28.924344: step: 2364/1541, loss: 0.35737913846969604 2023-01-23 08:15:29.864922: step: 2366/1541, loss: 0.28629040718078613 2023-01-23 08:15:30.735350: step: 2368/1541, loss: 0.3833003342151642 2023-01-23 08:15:31.636033: step: 2370/1541, loss: 0.2298727035522461 2023-01-23 08:15:32.591993: step: 2372/1541, loss: 0.6637483239173889 2023-01-23 08:15:33.483010: step: 2374/1541, loss: 0.5959676504135132 2023-01-23 08:15:34.389392: step: 2376/1541, loss: 0.35025352239608765 2023-01-23 08:15:35.321572: step: 2378/1541, loss: 0.26199302077293396 2023-01-23 08:15:36.232300: step: 2380/1541, loss: 0.7131326794624329 2023-01-23 08:15:37.171605: step: 2382/1541, loss: 0.27044135332107544 2023-01-23 08:15:38.075596: step: 2384/1541, loss: 0.5264255404472351 2023-01-23 08:15:38.983496: step: 2386/1541, loss: 0.5265430808067322 2023-01-23 08:15:39.879278: step: 2388/1541, loss: 1.4314311742782593 2023-01-23 08:15:40.814532: step: 2390/1541, loss: 0.21118289232254028 2023-01-23 08:15:41.690114: step: 2392/1541, loss: 0.3093354105949402 2023-01-23 08:15:42.585493: step: 2394/1541, loss: 0.452451229095459 2023-01-23 08:15:43.521607: step: 2396/1541, loss: 0.4869271516799927 2023-01-23 08:15:44.437320: step: 2398/1541, loss: 0.6034398078918457 2023-01-23 08:15:45.344447: step: 2400/1541, loss: 0.3765832781791687 2023-01-23 08:15:46.272344: step: 2402/1541, loss: 0.6270300149917603 2023-01-23 08:15:47.198113: step: 2404/1541, loss: 0.2202858179807663 2023-01-23 08:15:48.164653: step: 2406/1541, loss: 3.4819798469543457 2023-01-23 08:15:49.041209: step: 2408/1541, loss: 0.4979615807533264 2023-01-23 08:15:49.959970: step: 2410/1541, loss: 0.41165822744369507 2023-01-23 08:15:50.837972: step: 2412/1541, loss: 0.27582621574401855 2023-01-23 08:15:51.758326: step: 2414/1541, loss: 0.3619877099990845 2023-01-23 08:15:52.634906: step: 2416/1541, loss: 1.0140975713729858 2023-01-23 08:15:53.521578: step: 2418/1541, loss: 0.38894355297088623 2023-01-23 08:15:54.418401: step: 2420/1541, loss: 0.3088950216770172 2023-01-23 08:15:55.253186: step: 2422/1541, loss: 0.29405105113983154 2023-01-23 08:15:56.152946: step: 2424/1541, loss: 0.23790881037712097 2023-01-23 08:15:57.035882: step: 2426/1541, loss: 0.5583164691925049 2023-01-23 08:15:57.930143: step: 2428/1541, loss: 0.6943160891532898 2023-01-23 08:15:58.794700: step: 2430/1541, loss: 1.337069034576416 2023-01-23 08:15:59.672404: step: 2432/1541, loss: 0.5324947834014893 2023-01-23 08:16:00.649516: step: 2434/1541, loss: 0.5383636951446533 2023-01-23 08:16:01.504979: step: 2436/1541, loss: 0.9083061814308167 2023-01-23 08:16:02.433780: step: 2438/1541, loss: 0.6371046304702759 2023-01-23 08:16:03.293174: step: 2440/1541, loss: 0.7018256187438965 2023-01-23 08:16:04.228884: step: 2442/1541, loss: 4.1311798095703125 2023-01-23 08:16:05.145352: step: 2444/1541, loss: 0.5486370921134949 2023-01-23 08:16:06.067075: step: 2446/1541, loss: 0.6000863313674927 2023-01-23 08:16:06.991740: step: 2448/1541, loss: 0.3837604522705078 2023-01-23 08:16:07.854821: step: 2450/1541, loss: 0.22511935234069824 2023-01-23 08:16:08.725471: step: 2452/1541, loss: 0.11245474964380264 2023-01-23 08:16:09.637848: step: 2454/1541, loss: 0.509580135345459 2023-01-23 08:16:10.526900: step: 2456/1541, loss: 0.26752182841300964 2023-01-23 08:16:11.419757: step: 2458/1541, loss: 0.3914578855037689 2023-01-23 08:16:12.311759: step: 2460/1541, loss: 0.3182029724121094 2023-01-23 08:16:13.255246: step: 2462/1541, loss: 0.5318822264671326 2023-01-23 08:16:14.144162: step: 2464/1541, loss: 0.10105490684509277 2023-01-23 08:16:15.021298: step: 2466/1541, loss: 0.3681928813457489 2023-01-23 08:16:15.967047: step: 2468/1541, loss: 0.3074917197227478 2023-01-23 08:16:16.862537: step: 2470/1541, loss: 0.36758705973625183 2023-01-23 08:16:17.868632: step: 2472/1541, loss: 0.2939111292362213 2023-01-23 08:16:18.756441: step: 2474/1541, loss: 0.3810208737850189 2023-01-23 08:16:19.656733: step: 2476/1541, loss: 0.5922797322273254 2023-01-23 08:16:20.624674: step: 2478/1541, loss: 1.2189526557922363 2023-01-23 08:16:21.506065: step: 2480/1541, loss: 0.19897696375846863 2023-01-23 08:16:22.382588: step: 2482/1541, loss: 1.1608086824417114 2023-01-23 08:16:23.320814: step: 2484/1541, loss: 0.5219589471817017 2023-01-23 08:16:24.307152: step: 2486/1541, loss: 0.3631427586078644 2023-01-23 08:16:25.239210: step: 2488/1541, loss: 0.3742224872112274 2023-01-23 08:16:26.145424: step: 2490/1541, loss: 0.5578489303588867 2023-01-23 08:16:27.021560: step: 2492/1541, loss: 0.5493852496147156 2023-01-23 08:16:27.899631: step: 2494/1541, loss: 0.4827592968940735 2023-01-23 08:16:28.741623: step: 2496/1541, loss: 0.2987029254436493 2023-01-23 08:16:29.626432: step: 2498/1541, loss: 2.0250182151794434 2023-01-23 08:16:30.496489: step: 2500/1541, loss: 1.444091558456421 2023-01-23 08:16:31.402448: step: 2502/1541, loss: 0.6860601305961609 2023-01-23 08:16:32.315523: step: 2504/1541, loss: 0.1689060628414154 2023-01-23 08:16:33.189730: step: 2506/1541, loss: 0.5259720087051392 2023-01-23 08:16:34.080732: step: 2508/1541, loss: 0.4987666606903076 2023-01-23 08:16:34.968855: step: 2510/1541, loss: 0.47401249408721924 2023-01-23 08:16:35.902509: step: 2512/1541, loss: 2.9559648036956787 2023-01-23 08:16:36.779792: step: 2514/1541, loss: 0.4894676208496094 2023-01-23 08:16:37.685862: step: 2516/1541, loss: 0.7534530758857727 2023-01-23 08:16:38.544364: step: 2518/1541, loss: 0.40719717741012573 2023-01-23 08:16:39.411003: step: 2520/1541, loss: 0.1949417144060135 2023-01-23 08:16:40.314813: step: 2522/1541, loss: 1.0050588846206665 2023-01-23 08:16:41.236198: step: 2524/1541, loss: 2.890892267227173 2023-01-23 08:16:42.151191: step: 2526/1541, loss: 0.46157681941986084 2023-01-23 08:16:43.076790: step: 2528/1541, loss: 0.08505836129188538 2023-01-23 08:16:43.950467: step: 2530/1541, loss: 0.20933003723621368 2023-01-23 08:16:44.843218: step: 2532/1541, loss: 0.3245677947998047 2023-01-23 08:16:45.767200: step: 2534/1541, loss: 0.3890281021595001 2023-01-23 08:16:46.629343: step: 2536/1541, loss: 0.49329739809036255 2023-01-23 08:16:47.506759: step: 2538/1541, loss: 0.31598812341690063 2023-01-23 08:16:48.420015: step: 2540/1541, loss: 0.7554256916046143 2023-01-23 08:16:49.352936: step: 2542/1541, loss: 0.09741850197315216 2023-01-23 08:16:50.221068: step: 2544/1541, loss: 0.4451141953468323 2023-01-23 08:16:51.133725: step: 2546/1541, loss: 0.06661173701286316 2023-01-23 08:16:52.031497: step: 2548/1541, loss: 0.4275400936603546 2023-01-23 08:16:52.952191: step: 2550/1541, loss: 0.8161649107933044 2023-01-23 08:16:53.863548: step: 2552/1541, loss: 0.30927780270576477 2023-01-23 08:16:54.783838: step: 2554/1541, loss: 0.16097821295261383 2023-01-23 08:16:55.771482: step: 2556/1541, loss: 0.24781450629234314 2023-01-23 08:16:56.730733: step: 2558/1541, loss: 0.32843297719955444 2023-01-23 08:16:57.715404: step: 2560/1541, loss: 0.596539318561554 2023-01-23 08:16:58.657240: step: 2562/1541, loss: 0.11232232302427292 2023-01-23 08:16:59.580516: step: 2564/1541, loss: 0.09299303591251373 2023-01-23 08:17:00.454919: step: 2566/1541, loss: 0.12168044596910477 2023-01-23 08:17:01.373741: step: 2568/1541, loss: 0.7553045153617859 2023-01-23 08:17:02.266423: step: 2570/1541, loss: 0.6948060393333435 2023-01-23 08:17:03.155028: step: 2572/1541, loss: 0.370410680770874 2023-01-23 08:17:04.052253: step: 2574/1541, loss: 0.0413714274764061 2023-01-23 08:17:04.953310: step: 2576/1541, loss: 0.39163413643836975 2023-01-23 08:17:05.836879: step: 2578/1541, loss: 0.2875373065471649 2023-01-23 08:17:06.709871: step: 2580/1541, loss: 0.8042882680892944 2023-01-23 08:17:07.620275: step: 2582/1541, loss: 0.20667371153831482 2023-01-23 08:17:08.511129: step: 2584/1541, loss: 0.480561226606369 2023-01-23 08:17:09.493143: step: 2586/1541, loss: 0.7435701489448547 2023-01-23 08:17:10.416032: step: 2588/1541, loss: 0.4974386990070343 2023-01-23 08:17:11.309107: step: 2590/1541, loss: 0.1007733941078186 2023-01-23 08:17:12.245659: step: 2592/1541, loss: 0.19914785027503967 2023-01-23 08:17:13.133463: step: 2594/1541, loss: 0.5125964283943176 2023-01-23 08:17:14.051370: step: 2596/1541, loss: 0.19911642372608185 2023-01-23 08:17:14.935667: step: 2598/1541, loss: 0.46639159321784973 2023-01-23 08:17:15.869048: step: 2600/1541, loss: 0.5610818266868591 2023-01-23 08:17:16.840238: step: 2602/1541, loss: 0.3130478858947754 2023-01-23 08:17:17.740853: step: 2604/1541, loss: 0.5314725637435913 2023-01-23 08:17:18.610086: step: 2606/1541, loss: 3.045461654663086 2023-01-23 08:17:19.530368: step: 2608/1541, loss: 0.650795042514801 2023-01-23 08:17:20.401034: step: 2610/1541, loss: 0.4002058804035187 2023-01-23 08:17:21.271603: step: 2612/1541, loss: 0.15212808549404144 2023-01-23 08:17:22.177732: step: 2614/1541, loss: 0.41163790225982666 2023-01-23 08:17:23.083980: step: 2616/1541, loss: 0.38238802552223206 2023-01-23 08:17:23.960564: step: 2618/1541, loss: 0.14184576272964478 2023-01-23 08:17:24.861383: step: 2620/1541, loss: 1.0672186613082886 2023-01-23 08:17:25.719041: step: 2622/1541, loss: 0.5093117356300354 2023-01-23 08:17:26.677897: step: 2624/1541, loss: 0.5087106823921204 2023-01-23 08:17:27.560129: step: 2626/1541, loss: 0.6794245839118958 2023-01-23 08:17:28.476505: step: 2628/1541, loss: 0.3614633083343506 2023-01-23 08:17:29.349934: step: 2630/1541, loss: 0.22724072635173798 2023-01-23 08:17:30.263423: step: 2632/1541, loss: 0.3360174298286438 2023-01-23 08:17:31.174903: step: 2634/1541, loss: 0.20562320947647095 2023-01-23 08:17:32.075800: step: 2636/1541, loss: 0.5588959455490112 2023-01-23 08:17:32.972480: step: 2638/1541, loss: 0.25569090247154236 2023-01-23 08:17:33.895765: step: 2640/1541, loss: 0.7446404695510864 2023-01-23 08:17:34.777053: step: 2642/1541, loss: 0.5160866379737854 2023-01-23 08:17:35.713049: step: 2644/1541, loss: 0.49094071984291077 2023-01-23 08:17:36.583180: step: 2646/1541, loss: 0.746354877948761 2023-01-23 08:17:37.482891: step: 2648/1541, loss: 0.5160349011421204 2023-01-23 08:17:38.397228: step: 2650/1541, loss: 0.3393181562423706 2023-01-23 08:17:39.293423: step: 2652/1541, loss: 0.425957590341568 2023-01-23 08:17:40.139233: step: 2654/1541, loss: 0.26617783308029175 2023-01-23 08:17:41.073837: step: 2656/1541, loss: 0.1296987235546112 2023-01-23 08:17:41.993488: step: 2658/1541, loss: 2.695450782775879 2023-01-23 08:17:42.857357: step: 2660/1541, loss: 0.05855937302112579 2023-01-23 08:17:43.791866: step: 2662/1541, loss: 0.6484640836715698 2023-01-23 08:17:44.681422: step: 2664/1541, loss: 0.5116156935691833 2023-01-23 08:17:45.560776: step: 2666/1541, loss: 0.5472994446754456 2023-01-23 08:17:46.439340: step: 2668/1541, loss: 0.44732606410980225 2023-01-23 08:17:47.311515: step: 2670/1541, loss: 0.28267085552215576 2023-01-23 08:17:48.180461: step: 2672/1541, loss: 0.39071786403656006 2023-01-23 08:17:49.078428: step: 2674/1541, loss: 0.9140645265579224 2023-01-23 08:17:50.017081: step: 2676/1541, loss: 0.5654433369636536 2023-01-23 08:17:50.942780: step: 2678/1541, loss: 0.22274208068847656 2023-01-23 08:17:51.836213: step: 2680/1541, loss: 0.33373114466667175 2023-01-23 08:17:52.761227: step: 2682/1541, loss: 0.21486611664295197 2023-01-23 08:17:53.662324: step: 2684/1541, loss: 0.5885992646217346 2023-01-23 08:17:54.582811: step: 2686/1541, loss: 1.079018235206604 2023-01-23 08:17:55.493596: step: 2688/1541, loss: 0.23274166882038116 2023-01-23 08:17:56.397035: step: 2690/1541, loss: 0.2751084268093109 2023-01-23 08:17:57.284603: step: 2692/1541, loss: 0.28637221455574036 2023-01-23 08:17:58.165156: step: 2694/1541, loss: 0.3898649215698242 2023-01-23 08:17:59.049388: step: 2696/1541, loss: 0.99503093957901 2023-01-23 08:17:59.954794: step: 2698/1541, loss: 0.2188851684331894 2023-01-23 08:18:00.868378: step: 2700/1541, loss: 0.49827417731285095 2023-01-23 08:18:01.731371: step: 2702/1541, loss: 0.6567983627319336 2023-01-23 08:18:02.635171: step: 2704/1541, loss: 0.1222076267004013 2023-01-23 08:18:03.532117: step: 2706/1541, loss: 0.5982767343521118 2023-01-23 08:18:04.405659: step: 2708/1541, loss: 1.0173170566558838 2023-01-23 08:18:05.352451: step: 2710/1541, loss: 0.1552242934703827 2023-01-23 08:18:06.234136: step: 2712/1541, loss: 0.36681339144706726 2023-01-23 08:18:07.130121: step: 2714/1541, loss: 0.2911258041858673 2023-01-23 08:18:08.029986: step: 2716/1541, loss: 1.0895599126815796 2023-01-23 08:18:08.914863: step: 2718/1541, loss: 0.45269763469696045 2023-01-23 08:18:09.806068: step: 2720/1541, loss: 0.5924204587936401 2023-01-23 08:18:10.721285: step: 2722/1541, loss: 1.6550719738006592 2023-01-23 08:18:11.653989: step: 2724/1541, loss: 0.4262261390686035 2023-01-23 08:18:12.507624: step: 2726/1541, loss: 0.2604774832725525 2023-01-23 08:18:13.453706: step: 2728/1541, loss: 0.5609034895896912 2023-01-23 08:18:14.364300: step: 2730/1541, loss: 0.9172269701957703 2023-01-23 08:18:15.232110: step: 2732/1541, loss: 1.9584486484527588 2023-01-23 08:18:16.184035: step: 2734/1541, loss: 0.7872551083564758 2023-01-23 08:18:17.059449: step: 2736/1541, loss: 0.523343026638031 2023-01-23 08:18:17.966247: step: 2738/1541, loss: 0.3168150782585144 2023-01-23 08:18:18.854608: step: 2740/1541, loss: 4.935987949371338 2023-01-23 08:18:19.809346: step: 2742/1541, loss: 0.2545333206653595 2023-01-23 08:18:20.729895: step: 2744/1541, loss: 0.7484838366508484 2023-01-23 08:18:21.629301: step: 2746/1541, loss: 0.8104331493377686 2023-01-23 08:18:22.527087: step: 2748/1541, loss: 0.23071765899658203 2023-01-23 08:18:23.421142: step: 2750/1541, loss: 0.2178056538105011 2023-01-23 08:18:24.278899: step: 2752/1541, loss: 0.3566720187664032 2023-01-23 08:18:25.207554: step: 2754/1541, loss: 0.075624480843544 2023-01-23 08:18:26.101398: step: 2756/1541, loss: 0.6506491899490356 2023-01-23 08:18:26.989141: step: 2758/1541, loss: 0.2962110936641693 2023-01-23 08:18:27.882433: step: 2760/1541, loss: 0.12602218985557556 2023-01-23 08:18:28.763045: step: 2762/1541, loss: 0.37188854813575745 2023-01-23 08:18:29.710614: step: 2764/1541, loss: 0.1570844203233719 2023-01-23 08:18:30.549942: step: 2766/1541, loss: 0.1626659780740738 2023-01-23 08:18:31.457451: step: 2768/1541, loss: 0.2826482057571411 2023-01-23 08:18:32.333291: step: 2770/1541, loss: 0.6721881628036499 2023-01-23 08:18:33.263548: step: 2772/1541, loss: 0.19929471611976624 2023-01-23 08:18:34.199930: step: 2774/1541, loss: 0.28152522444725037 2023-01-23 08:18:35.072345: step: 2776/1541, loss: 1.4459084272384644 2023-01-23 08:18:35.956424: step: 2778/1541, loss: 0.4399173855781555 2023-01-23 08:18:36.862809: step: 2780/1541, loss: 0.4291118383407593 2023-01-23 08:18:37.769083: step: 2782/1541, loss: 0.4374447762966156 2023-01-23 08:18:38.649826: step: 2784/1541, loss: 0.358429491519928 2023-01-23 08:18:39.577401: step: 2786/1541, loss: 0.7848423719406128 2023-01-23 08:18:40.493490: step: 2788/1541, loss: 0.8004155158996582 2023-01-23 08:18:41.412437: step: 2790/1541, loss: 0.4481789171695709 2023-01-23 08:18:42.346331: step: 2792/1541, loss: 0.30845269560813904 2023-01-23 08:18:43.321169: step: 2794/1541, loss: 0.21521735191345215 2023-01-23 08:18:44.230046: step: 2796/1541, loss: 0.527890145778656 2023-01-23 08:18:45.151907: step: 2798/1541, loss: 1.2102043628692627 2023-01-23 08:18:46.091758: step: 2800/1541, loss: 0.8190504312515259 2023-01-23 08:18:46.991796: step: 2802/1541, loss: 0.8713911175727844 2023-01-23 08:18:47.837613: step: 2804/1541, loss: 0.3126477599143982 2023-01-23 08:18:48.694123: step: 2806/1541, loss: 0.47505971789360046 2023-01-23 08:18:49.620075: step: 2808/1541, loss: 0.3879315257072449 2023-01-23 08:18:50.490171: step: 2810/1541, loss: 0.4827576279640198 2023-01-23 08:18:51.376400: step: 2812/1541, loss: 0.21920500695705414 2023-01-23 08:18:52.267010: step: 2814/1541, loss: 0.6004037857055664 2023-01-23 08:18:53.156734: step: 2816/1541, loss: 0.41227197647094727 2023-01-23 08:18:54.079175: step: 2818/1541, loss: 0.3899393677711487 2023-01-23 08:18:54.978666: step: 2820/1541, loss: 2.852163791656494 2023-01-23 08:18:55.917033: step: 2822/1541, loss: 0.4338768422603607 2023-01-23 08:18:56.802152: step: 2824/1541, loss: 0.4513006806373596 2023-01-23 08:18:57.694010: step: 2826/1541, loss: 0.31895533204078674 2023-01-23 08:18:58.562308: step: 2828/1541, loss: 0.26491931080818176 2023-01-23 08:18:59.465258: step: 2830/1541, loss: 0.5767761468887329 2023-01-23 08:19:00.341524: step: 2832/1541, loss: 0.3173013925552368 2023-01-23 08:19:01.248543: step: 2834/1541, loss: 0.9124114513397217 2023-01-23 08:19:02.155543: step: 2836/1541, loss: 0.40037232637405396 2023-01-23 08:19:03.073138: step: 2838/1541, loss: 2.761157989501953 2023-01-23 08:19:03.952583: step: 2840/1541, loss: 0.2249407172203064 2023-01-23 08:19:04.831063: step: 2842/1541, loss: 0.14282286167144775 2023-01-23 08:19:05.712561: step: 2844/1541, loss: 0.3354862332344055 2023-01-23 08:19:06.639777: step: 2846/1541, loss: 0.47263509035110474 2023-01-23 08:19:07.499902: step: 2848/1541, loss: 0.2427360564470291 2023-01-23 08:19:08.382150: step: 2850/1541, loss: 1.6927896738052368 2023-01-23 08:19:09.256812: step: 2852/1541, loss: 0.6404045820236206 2023-01-23 08:19:10.183123: step: 2854/1541, loss: 0.42759400606155396 2023-01-23 08:19:11.052302: step: 2856/1541, loss: 2.946695566177368 2023-01-23 08:19:11.921730: step: 2858/1541, loss: 0.39658769965171814 2023-01-23 08:19:12.812306: step: 2860/1541, loss: 0.21681484580039978 2023-01-23 08:19:13.684450: step: 2862/1541, loss: 0.31748509407043457 2023-01-23 08:19:14.644420: step: 2864/1541, loss: 0.8376338481903076 2023-01-23 08:19:15.520148: step: 2866/1541, loss: 0.6023551821708679 2023-01-23 08:19:16.422890: step: 2868/1541, loss: 3.1157965660095215 2023-01-23 08:19:17.309952: step: 2870/1541, loss: 0.670340895652771 2023-01-23 08:19:18.157072: step: 2872/1541, loss: 0.35883456468582153 2023-01-23 08:19:19.033578: step: 2874/1541, loss: 0.4592936038970947 2023-01-23 08:19:19.952111: step: 2876/1541, loss: 0.2538592219352722 2023-01-23 08:19:20.794211: step: 2878/1541, loss: 0.7265527248382568 2023-01-23 08:19:21.770020: step: 2880/1541, loss: 0.8792724609375 2023-01-23 08:19:22.704210: step: 2882/1541, loss: 0.738746166229248 2023-01-23 08:19:23.566344: step: 2884/1541, loss: 0.15256236493587494 2023-01-23 08:19:24.424213: step: 2886/1541, loss: 0.842243492603302 2023-01-23 08:19:25.396463: step: 2888/1541, loss: 0.16923481225967407 2023-01-23 08:19:26.307779: step: 2890/1541, loss: 0.7615208625793457 2023-01-23 08:19:27.217965: step: 2892/1541, loss: 0.4112745225429535 2023-01-23 08:19:28.088827: step: 2894/1541, loss: 0.6240286827087402 2023-01-23 08:19:28.987612: step: 2896/1541, loss: 0.14702272415161133 2023-01-23 08:19:29.914928: step: 2898/1541, loss: 0.4155255854129791 2023-01-23 08:19:30.820336: step: 2900/1541, loss: 0.27667757868766785 2023-01-23 08:19:31.722195: step: 2902/1541, loss: 0.1865798383951187 2023-01-23 08:19:32.638604: step: 2904/1541, loss: 0.5667209029197693 2023-01-23 08:19:33.543132: step: 2906/1541, loss: 0.4336584210395813 2023-01-23 08:19:34.421481: step: 2908/1541, loss: 0.5776082873344421 2023-01-23 08:19:35.331688: step: 2910/1541, loss: 0.4636707901954651 2023-01-23 08:19:36.210549: step: 2912/1541, loss: 0.06498751789331436 2023-01-23 08:19:37.084898: step: 2914/1541, loss: 0.33142897486686707 2023-01-23 08:19:37.965395: step: 2916/1541, loss: 0.4506511688232422 2023-01-23 08:19:38.831261: step: 2918/1541, loss: 0.2490890920162201 2023-01-23 08:19:39.734950: step: 2920/1541, loss: 0.42526480555534363 2023-01-23 08:19:40.710716: step: 2922/1541, loss: 0.2141733318567276 2023-01-23 08:19:41.568609: step: 2924/1541, loss: 0.4117315113544464 2023-01-23 08:19:42.439933: step: 2926/1541, loss: 0.5281999111175537 2023-01-23 08:19:43.360568: step: 2928/1541, loss: 0.5379652380943298 2023-01-23 08:19:44.279639: step: 2930/1541, loss: 0.15913569927215576 2023-01-23 08:19:45.236199: step: 2932/1541, loss: 0.752458930015564 2023-01-23 08:19:46.125714: step: 2934/1541, loss: 1.0511993169784546 2023-01-23 08:19:47.056371: step: 2936/1541, loss: 0.4670538008213043 2023-01-23 08:19:47.940183: step: 2938/1541, loss: 1.8797932863235474 2023-01-23 08:19:48.808741: step: 2940/1541, loss: 0.16291335225105286 2023-01-23 08:19:49.695480: step: 2942/1541, loss: 0.30556637048721313 2023-01-23 08:19:50.577888: step: 2944/1541, loss: 6.746592044830322 2023-01-23 08:19:51.434380: step: 2946/1541, loss: 0.5648263096809387 2023-01-23 08:19:52.361495: step: 2948/1541, loss: 0.369114488363266 2023-01-23 08:19:53.255158: step: 2950/1541, loss: 0.18774299323558807 2023-01-23 08:19:54.124555: step: 2952/1541, loss: 0.2861602008342743 2023-01-23 08:19:55.099476: step: 2954/1541, loss: 0.42782121896743774 2023-01-23 08:19:55.960076: step: 2956/1541, loss: 0.6700953245162964 2023-01-23 08:19:56.833683: step: 2958/1541, loss: 0.21008075773715973 2023-01-23 08:19:57.759434: step: 2960/1541, loss: 0.37176215648651123 2023-01-23 08:19:58.657066: step: 2962/1541, loss: 1.0723779201507568 2023-01-23 08:19:59.582491: step: 2964/1541, loss: 0.3081706464290619 2023-01-23 08:20:00.460056: step: 2966/1541, loss: 0.3139583468437195 2023-01-23 08:20:01.332769: step: 2968/1541, loss: 0.4168941080570221 2023-01-23 08:20:02.242649: step: 2970/1541, loss: 0.197774276137352 2023-01-23 08:20:03.187316: step: 2972/1541, loss: 1.02774977684021 2023-01-23 08:20:04.081280: step: 2974/1541, loss: 0.5368696451187134 2023-01-23 08:20:04.965398: step: 2976/1541, loss: 0.5826009511947632 2023-01-23 08:20:05.864310: step: 2978/1541, loss: 0.3105542063713074 2023-01-23 08:20:06.741771: step: 2980/1541, loss: 0.3880952298641205 2023-01-23 08:20:07.649791: step: 2982/1541, loss: 0.2610984742641449 2023-01-23 08:20:08.507804: step: 2984/1541, loss: 0.8989841938018799 2023-01-23 08:20:09.353871: step: 2986/1541, loss: 0.27873265743255615 2023-01-23 08:20:10.227114: step: 2988/1541, loss: 0.44223955273628235 2023-01-23 08:20:11.113505: step: 2990/1541, loss: 0.46815556287765503 2023-01-23 08:20:12.034108: step: 2992/1541, loss: 0.4209692180156708 2023-01-23 08:20:12.937237: step: 2994/1541, loss: 0.43794941902160645 2023-01-23 08:20:13.833403: step: 2996/1541, loss: 0.10971313714981079 2023-01-23 08:20:14.725419: step: 2998/1541, loss: 0.6902456283569336 2023-01-23 08:20:15.613333: step: 3000/1541, loss: 0.8427921533584595 2023-01-23 08:20:16.513211: step: 3002/1541, loss: 0.2503439486026764 2023-01-23 08:20:17.396964: step: 3004/1541, loss: 0.3947385251522064 2023-01-23 08:20:18.264075: step: 3006/1541, loss: 0.22591744363307953 2023-01-23 08:20:19.151047: step: 3008/1541, loss: 3.078430652618408 2023-01-23 08:20:20.040613: step: 3010/1541, loss: 0.6604998707771301 2023-01-23 08:20:20.926573: step: 3012/1541, loss: 0.17101141810417175 2023-01-23 08:20:21.807375: step: 3014/1541, loss: 2.3351857662200928 2023-01-23 08:20:22.713974: step: 3016/1541, loss: 0.27140122652053833 2023-01-23 08:20:23.586365: step: 3018/1541, loss: 0.8198812007904053 2023-01-23 08:20:24.491822: step: 3020/1541, loss: 0.16209684312343597 2023-01-23 08:20:25.374537: step: 3022/1541, loss: 0.8859480023384094 2023-01-23 08:20:26.290616: step: 3024/1541, loss: 2.89910626411438 2023-01-23 08:20:27.228381: step: 3026/1541, loss: 0.3184930086135864 2023-01-23 08:20:28.135492: step: 3028/1541, loss: 0.4536515772342682 2023-01-23 08:20:29.011390: step: 3030/1541, loss: 1.0025537014007568 2023-01-23 08:20:29.898197: step: 3032/1541, loss: 0.4787933826446533 2023-01-23 08:20:30.782863: step: 3034/1541, loss: 0.38623735308647156 2023-01-23 08:20:31.649053: step: 3036/1541, loss: 0.6923116445541382 2023-01-23 08:20:32.605267: step: 3038/1541, loss: 0.6744558215141296 2023-01-23 08:20:33.483765: step: 3040/1541, loss: 0.3620207905769348 2023-01-23 08:20:34.374191: step: 3042/1541, loss: 0.09328323602676392 2023-01-23 08:20:35.297971: step: 3044/1541, loss: 0.547307014465332 2023-01-23 08:20:36.202531: step: 3046/1541, loss: 0.25897449254989624 2023-01-23 08:20:37.051348: step: 3048/1541, loss: 0.8650431632995605 2023-01-23 08:20:37.947271: step: 3050/1541, loss: 0.788163423538208 2023-01-23 08:20:38.923767: step: 3052/1541, loss: 0.7731156349182129 2023-01-23 08:20:39.803302: step: 3054/1541, loss: 0.5013137459754944 2023-01-23 08:20:40.692569: step: 3056/1541, loss: 0.5345146656036377 2023-01-23 08:20:41.590563: step: 3058/1541, loss: 0.15929993987083435 2023-01-23 08:20:42.520814: step: 3060/1541, loss: 0.4128381609916687 2023-01-23 08:20:43.395095: step: 3062/1541, loss: 0.410574346780777 2023-01-23 08:20:44.305136: step: 3064/1541, loss: 0.5719864964485168 2023-01-23 08:20:45.195933: step: 3066/1541, loss: 0.4258204996585846 2023-01-23 08:20:46.095040: step: 3068/1541, loss: 0.3244175314903259 2023-01-23 08:20:47.022449: step: 3070/1541, loss: 0.11442817747592926 2023-01-23 08:20:47.946425: step: 3072/1541, loss: 0.45432591438293457 2023-01-23 08:20:48.871029: step: 3074/1541, loss: 0.07743653655052185 2023-01-23 08:20:49.752993: step: 3076/1541, loss: 0.43917447328567505 2023-01-23 08:20:50.647130: step: 3078/1541, loss: 0.13430707156658173 2023-01-23 08:20:51.512222: step: 3080/1541, loss: 0.6464106440544128 2023-01-23 08:20:52.412798: step: 3082/1541, loss: 0.6587355732917786 2023-01-23 08:20:53.038694: step: 3084/1541, loss: 0.10745313763618469 ================================================== Loss: 0.581 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.674168746877286, 'r': 0.6563723669131339, 'f1': 0.6651515409788485}, 'combined': 0.4398042304960901, 'epoch': 6} Test Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7174883835585704, 'r': 0.6942085927686642, 'f1': 0.7056565388046863}, 'combined': 0.4604255769854156, 'epoch': 6} Dev Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.6733013090338247, 'r': 0.671367924083261, 'f1': 0.6723332266340636}, 'combined': 0.44189490446191415, 'epoch': 6} Test Korean: {'event': {'p': 0.6129647283126788, 'r': 0.7705212702216896, 'f1': 0.6827714361560924}, 'argument': {'p': 0.719809576245395, 'r': 0.7112103051307276, 'f1': 0.715484103358802}, 'combined': 0.48851210879714324, 'epoch': 6} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6749027227273954, 'r': 0.6759176140397674, 'f1': 0.6754097871321493}, 'combined': 0.467493670842566, 'epoch': 6} Test Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7154362539653685, 'r': 0.7041128024637584, 'f1': 0.7097293657901553}, 'combined': 0.49443157493872836, 'epoch': 6} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6476210705182667, 'r': 0.6132341110217215, 'f1': 0.6299586776859504}, 'combined': 0.4522780250052977, 'epoch': 6} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5402494331065759, 'r': 0.33938746438746437, 'f1': 0.41688538932633423}, 'combined': 0.24290959693030498, 'epoch': 6} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5630713489409142, 'r': 0.6023553965414431, 'f1': 0.5820512820512821}, 'combined': 0.2993406593406594, 'epoch': 6} New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test for Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} -------------------- Dev for Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6749027227273954, 'r': 0.6759176140397674, 'f1': 0.6754097871321493}, 'combined': 0.467493670842566, 'epoch': 6} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7154362539653685, 'r': 0.7041128024637584, 'f1': 0.7097293657901553}, 'combined': 0.49443157493872836, 'epoch': 6} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5630713489409142, 'r': 0.6023553965414431, 'f1': 0.5820512820512821}, 'combined': 0.2993406593406594, 'epoch': 6} ****************************** Epoch: 7 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 08:26:23.358780: step: 2/1541, loss: 0.2579868733882904 2023-01-23 08:26:24.306996: step: 4/1541, loss: 0.3112909197807312 2023-01-23 08:26:25.213853: step: 6/1541, loss: 0.19454462826251984 2023-01-23 08:26:26.099082: step: 8/1541, loss: 0.20036593079566956 2023-01-23 08:26:26.958285: step: 10/1541, loss: 0.2667955160140991 2023-01-23 08:26:27.838572: step: 12/1541, loss: 0.2849726974964142 2023-01-23 08:26:28.753734: step: 14/1541, loss: 0.2859853208065033 2023-01-23 08:26:29.668606: step: 16/1541, loss: 0.2911447286605835 2023-01-23 08:26:30.570967: step: 18/1541, loss: 0.5702903866767883 2023-01-23 08:26:31.473537: step: 20/1541, loss: 0.24346494674682617 2023-01-23 08:26:32.396202: step: 22/1541, loss: 0.5249785780906677 2023-01-23 08:26:33.284188: step: 24/1541, loss: 0.5091520547866821 2023-01-23 08:26:34.149266: step: 26/1541, loss: 0.7474247217178345 2023-01-23 08:26:35.026318: step: 28/1541, loss: 0.21959027647972107 2023-01-23 08:26:35.920723: step: 30/1541, loss: 0.195523202419281 2023-01-23 08:26:36.760867: step: 32/1541, loss: 0.12691530585289001 2023-01-23 08:26:37.640522: step: 34/1541, loss: 0.26039090752601624 2023-01-23 08:26:38.470803: step: 36/1541, loss: 0.21303991973400116 2023-01-23 08:26:39.338679: step: 38/1541, loss: 0.09901204705238342 2023-01-23 08:26:40.252111: step: 40/1541, loss: 0.260595440864563 2023-01-23 08:26:41.119335: step: 42/1541, loss: 0.02686426229774952 2023-01-23 08:26:42.010464: step: 44/1541, loss: 0.33071988821029663 2023-01-23 08:26:42.935238: step: 46/1541, loss: 0.6816605925559998 2023-01-23 08:26:43.834097: step: 48/1541, loss: 0.28850430250167847 2023-01-23 08:26:44.754832: step: 50/1541, loss: 1.0197232961654663 2023-01-23 08:26:45.603014: step: 52/1541, loss: 1.3067376613616943 2023-01-23 08:26:46.478335: step: 54/1541, loss: 2.507636308670044 2023-01-23 08:26:47.385233: step: 56/1541, loss: 0.38445988297462463 2023-01-23 08:26:48.253831: step: 58/1541, loss: 0.5914362668991089 2023-01-23 08:26:49.146272: step: 60/1541, loss: 0.24124006927013397 2023-01-23 08:26:50.038401: step: 62/1541, loss: 0.3070181608200073 2023-01-23 08:26:51.034841: step: 64/1541, loss: 0.5058199763298035 2023-01-23 08:26:51.905151: step: 66/1541, loss: 0.28303250670433044 2023-01-23 08:26:52.823222: step: 68/1541, loss: 2.995687961578369 2023-01-23 08:26:53.784881: step: 70/1541, loss: 0.6866189241409302 2023-01-23 08:26:54.758355: step: 72/1541, loss: 0.11763561517000198 2023-01-23 08:26:55.636412: step: 74/1541, loss: 0.5432503819465637 2023-01-23 08:26:56.545487: step: 76/1541, loss: 0.7442798018455505 2023-01-23 08:26:57.431311: step: 78/1541, loss: 0.39380913972854614 2023-01-23 08:26:58.388272: step: 80/1541, loss: 0.3768186867237091 2023-01-23 08:26:59.423360: step: 82/1541, loss: 0.5372965335845947 2023-01-23 08:27:00.346970: step: 84/1541, loss: 0.4120064079761505 2023-01-23 08:27:01.272327: step: 86/1541, loss: 0.7660772800445557 2023-01-23 08:27:02.200186: step: 88/1541, loss: 0.18100562691688538 2023-01-23 08:27:03.100394: step: 90/1541, loss: 0.9102276563644409 2023-01-23 08:27:03.990303: step: 92/1541, loss: 0.2516002058982849 2023-01-23 08:27:04.869610: step: 94/1541, loss: 0.21865513920783997 2023-01-23 08:27:05.774682: step: 96/1541, loss: 0.3413856625556946 2023-01-23 08:27:06.738611: step: 98/1541, loss: 0.16463804244995117 2023-01-23 08:27:07.613889: step: 100/1541, loss: 0.011443693190813065 2023-01-23 08:27:08.511921: step: 102/1541, loss: 0.3352484405040741 2023-01-23 08:27:09.412540: step: 104/1541, loss: 0.10072481632232666 2023-01-23 08:27:10.326059: step: 106/1541, loss: 0.12936626374721527 2023-01-23 08:27:11.195020: step: 108/1541, loss: 0.19323241710662842 2023-01-23 08:27:12.112362: step: 110/1541, loss: 0.4644816219806671 2023-01-23 08:27:13.027146: step: 112/1541, loss: 0.40150773525238037 2023-01-23 08:27:13.942674: step: 114/1541, loss: 0.1350114345550537 2023-01-23 08:27:14.867546: step: 116/1541, loss: 0.1921389400959015 2023-01-23 08:27:15.740171: step: 118/1541, loss: 0.16776296496391296 2023-01-23 08:27:16.645622: step: 120/1541, loss: 0.3189740777015686 2023-01-23 08:27:17.533271: step: 122/1541, loss: 0.9888404011726379 2023-01-23 08:27:18.413875: step: 124/1541, loss: 0.2898557186126709 2023-01-23 08:27:19.339803: step: 126/1541, loss: 0.24769869446754456 2023-01-23 08:27:20.236099: step: 128/1541, loss: 5.799822807312012 2023-01-23 08:27:21.131254: step: 130/1541, loss: 0.35392826795578003 2023-01-23 08:27:22.026801: step: 132/1541, loss: 0.07290142774581909 2023-01-23 08:27:22.927774: step: 134/1541, loss: 0.30579203367233276 2023-01-23 08:27:23.851745: step: 136/1541, loss: 0.23729072511196136 2023-01-23 08:27:24.738514: step: 138/1541, loss: 1.5848329067230225 2023-01-23 08:27:25.646066: step: 140/1541, loss: 0.5560625195503235 2023-01-23 08:27:26.573463: step: 142/1541, loss: 0.3065751791000366 2023-01-23 08:27:27.482975: step: 144/1541, loss: 0.12699942290782928 2023-01-23 08:27:28.362920: step: 146/1541, loss: 0.23076719045639038 2023-01-23 08:27:29.377908: step: 148/1541, loss: 0.41583770513534546 2023-01-23 08:27:30.297479: step: 150/1541, loss: 0.21387813985347748 2023-01-23 08:27:31.225840: step: 152/1541, loss: 0.49229568243026733 2023-01-23 08:27:32.258549: step: 154/1541, loss: 0.6294094920158386 2023-01-23 08:27:33.150006: step: 156/1541, loss: 0.40746909379959106 2023-01-23 08:27:34.184253: step: 158/1541, loss: 0.2892521321773529 2023-01-23 08:27:35.074599: step: 160/1541, loss: 0.29808762669563293 2023-01-23 08:27:35.941507: step: 162/1541, loss: 0.875683069229126 2023-01-23 08:27:36.820926: step: 164/1541, loss: 0.4833061695098877 2023-01-23 08:27:37.695085: step: 166/1541, loss: 0.4335256814956665 2023-01-23 08:27:38.577753: step: 168/1541, loss: 0.7704944610595703 2023-01-23 08:27:39.469331: step: 170/1541, loss: 0.2012990266084671 2023-01-23 08:27:40.362621: step: 172/1541, loss: 0.17984017729759216 2023-01-23 08:27:41.224956: step: 174/1541, loss: 0.15501579642295837 2023-01-23 08:27:42.143331: step: 176/1541, loss: 0.5073744654655457 2023-01-23 08:27:42.996771: step: 178/1541, loss: 0.16515709459781647 2023-01-23 08:27:43.902242: step: 180/1541, loss: 0.0960150957107544 2023-01-23 08:27:44.796737: step: 182/1541, loss: 0.163369283080101 2023-01-23 08:27:45.687175: step: 184/1541, loss: 0.14943847060203552 2023-01-23 08:27:46.550613: step: 186/1541, loss: 0.19953438639640808 2023-01-23 08:27:47.398136: step: 188/1541, loss: 0.3014092445373535 2023-01-23 08:27:48.285697: step: 190/1541, loss: 5.95229434967041 2023-01-23 08:27:49.222536: step: 192/1541, loss: 0.5793501734733582 2023-01-23 08:27:50.114230: step: 194/1541, loss: 0.253137469291687 2023-01-23 08:27:51.022254: step: 196/1541, loss: 0.12551692128181458 2023-01-23 08:27:51.945509: step: 198/1541, loss: 0.5133922696113586 2023-01-23 08:27:52.856435: step: 200/1541, loss: 0.13721275329589844 2023-01-23 08:27:53.805570: step: 202/1541, loss: 0.39521482586860657 2023-01-23 08:27:54.678597: step: 204/1541, loss: 0.3520086407661438 2023-01-23 08:27:55.572138: step: 206/1541, loss: 0.5080333948135376 2023-01-23 08:27:56.489718: step: 208/1541, loss: 0.2133118212223053 2023-01-23 08:27:57.435360: step: 210/1541, loss: 0.24512256681919098 2023-01-23 08:27:58.322438: step: 212/1541, loss: 0.2026044875383377 2023-01-23 08:27:59.263768: step: 214/1541, loss: 0.45961588621139526 2023-01-23 08:28:00.162514: step: 216/1541, loss: 1.014976978302002 2023-01-23 08:28:01.052056: step: 218/1541, loss: 0.4315493106842041 2023-01-23 08:28:01.948547: step: 220/1541, loss: 0.18767286837100983 2023-01-23 08:28:02.845402: step: 222/1541, loss: 0.34766581654548645 2023-01-23 08:28:03.742312: step: 224/1541, loss: 0.49199947714805603 2023-01-23 08:28:04.618399: step: 226/1541, loss: 0.2898993492126465 2023-01-23 08:28:05.541376: step: 228/1541, loss: 0.5220848917961121 2023-01-23 08:28:06.486098: step: 230/1541, loss: 0.23349371552467346 2023-01-23 08:28:07.385249: step: 232/1541, loss: 0.45449119806289673 2023-01-23 08:28:08.267263: step: 234/1541, loss: 0.25531965494155884 2023-01-23 08:28:09.169686: step: 236/1541, loss: 0.0563228502869606 2023-01-23 08:28:10.054279: step: 238/1541, loss: 0.6879651546478271 2023-01-23 08:28:10.925080: step: 240/1541, loss: 0.22322702407836914 2023-01-23 08:28:11.863842: step: 242/1541, loss: 0.06621050834655762 2023-01-23 08:28:12.748408: step: 244/1541, loss: 0.2405666559934616 2023-01-23 08:28:13.629586: step: 246/1541, loss: 0.2709241509437561 2023-01-23 08:28:14.586074: step: 248/1541, loss: 0.1756470799446106 2023-01-23 08:28:15.443449: step: 250/1541, loss: 0.21960696578025818 2023-01-23 08:28:16.358835: step: 252/1541, loss: 0.327443391084671 2023-01-23 08:28:17.270195: step: 254/1541, loss: 0.2742615044116974 2023-01-23 08:28:18.191839: step: 256/1541, loss: 0.017471445724368095 2023-01-23 08:28:19.125436: step: 258/1541, loss: 0.383104145526886 2023-01-23 08:28:20.040706: step: 260/1541, loss: 3.0665433406829834 2023-01-23 08:28:20.952546: step: 262/1541, loss: 0.3512595295906067 2023-01-23 08:28:21.830577: step: 264/1541, loss: 0.4373190104961395 2023-01-23 08:28:22.737640: step: 266/1541, loss: 0.4605315029621124 2023-01-23 08:28:23.652707: step: 268/1541, loss: 1.0388996601104736 2023-01-23 08:28:24.636089: step: 270/1541, loss: 0.459468275308609 2023-01-23 08:28:25.479969: step: 272/1541, loss: 0.4399876594543457 2023-01-23 08:28:26.402687: step: 274/1541, loss: 0.5556464195251465 2023-01-23 08:28:27.327348: step: 276/1541, loss: 0.49865007400512695 2023-01-23 08:28:28.247831: step: 278/1541, loss: 0.8816655874252319 2023-01-23 08:28:29.135514: step: 280/1541, loss: 0.0702686607837677 2023-01-23 08:28:30.020706: step: 282/1541, loss: 0.49585720896720886 2023-01-23 08:28:30.970250: step: 284/1541, loss: 3.2273566722869873 2023-01-23 08:28:31.892311: step: 286/1541, loss: 0.24767212569713593 2023-01-23 08:28:32.797009: step: 288/1541, loss: 0.18136970698833466 2023-01-23 08:28:33.703843: step: 290/1541, loss: 0.3989066481590271 2023-01-23 08:28:34.593211: step: 292/1541, loss: 0.12109224498271942 2023-01-23 08:28:35.522297: step: 294/1541, loss: 0.6205105781555176 2023-01-23 08:28:36.394328: step: 296/1541, loss: 6.787278175354004 2023-01-23 08:28:37.339243: step: 298/1541, loss: 0.21714048087596893 2023-01-23 08:28:38.281956: step: 300/1541, loss: 0.32577332854270935 2023-01-23 08:28:39.192277: step: 302/1541, loss: 0.08281299471855164 2023-01-23 08:28:40.104764: step: 304/1541, loss: 1.3422305583953857 2023-01-23 08:28:40.961965: step: 306/1541, loss: 0.19603011012077332 2023-01-23 08:28:41.951957: step: 308/1541, loss: 0.44443175196647644 2023-01-23 08:28:42.874918: step: 310/1541, loss: 0.266818106174469 2023-01-23 08:28:43.752470: step: 312/1541, loss: 0.34786778688430786 2023-01-23 08:28:44.677439: step: 314/1541, loss: 2.2741754055023193 2023-01-23 08:28:45.581658: step: 316/1541, loss: 0.3950643837451935 2023-01-23 08:28:46.461617: step: 318/1541, loss: 0.3384542763233185 2023-01-23 08:28:47.367532: step: 320/1541, loss: 1.0678051710128784 2023-01-23 08:28:48.240845: step: 322/1541, loss: 0.24772965908050537 2023-01-23 08:28:49.235947: step: 324/1541, loss: 0.18143102526664734 2023-01-23 08:28:50.123876: step: 326/1541, loss: 0.43713483214378357 2023-01-23 08:28:51.032376: step: 328/1541, loss: 0.571273148059845 2023-01-23 08:28:51.979914: step: 330/1541, loss: 0.15658316016197205 2023-01-23 08:28:52.914348: step: 332/1541, loss: 0.24066106975078583 2023-01-23 08:28:53.761995: step: 334/1541, loss: 0.3567242920398712 2023-01-23 08:28:54.608685: step: 336/1541, loss: 0.20732437074184418 2023-01-23 08:28:55.488786: step: 338/1541, loss: 0.1386740654706955 2023-01-23 08:28:56.373651: step: 340/1541, loss: 0.06989392638206482 2023-01-23 08:28:57.302490: step: 342/1541, loss: 0.37804320454597473 2023-01-23 08:28:58.209372: step: 344/1541, loss: 0.3783661127090454 2023-01-23 08:28:59.141756: step: 346/1541, loss: 0.43097394704818726 2023-01-23 08:29:00.045552: step: 348/1541, loss: 0.45231297612190247 2023-01-23 08:29:00.985017: step: 350/1541, loss: 0.2359803318977356 2023-01-23 08:29:01.896187: step: 352/1541, loss: 0.1537838578224182 2023-01-23 08:29:02.880995: step: 354/1541, loss: 0.25892677903175354 2023-01-23 08:29:03.832481: step: 356/1541, loss: 0.07267072051763535 2023-01-23 08:29:04.744744: step: 358/1541, loss: 0.2805868983268738 2023-01-23 08:29:05.612900: step: 360/1541, loss: 0.16907362639904022 2023-01-23 08:29:06.542720: step: 362/1541, loss: 0.4856979548931122 2023-01-23 08:29:07.399979: step: 364/1541, loss: 0.17920894920825958 2023-01-23 08:29:08.308958: step: 366/1541, loss: 0.2735346555709839 2023-01-23 08:29:09.214418: step: 368/1541, loss: 0.16371436417102814 2023-01-23 08:29:10.157813: step: 370/1541, loss: 0.5101383924484253 2023-01-23 08:29:11.058938: step: 372/1541, loss: 0.8571751117706299 2023-01-23 08:29:12.011680: step: 374/1541, loss: 0.8259065747261047 2023-01-23 08:29:12.893462: step: 376/1541, loss: 2.4326486587524414 2023-01-23 08:29:13.803961: step: 378/1541, loss: 0.46349045634269714 2023-01-23 08:29:14.700735: step: 380/1541, loss: 0.39658451080322266 2023-01-23 08:29:15.609417: step: 382/1541, loss: 0.391915500164032 2023-01-23 08:29:16.494755: step: 384/1541, loss: 0.2724573612213135 2023-01-23 08:29:17.397606: step: 386/1541, loss: 0.3474920094013214 2023-01-23 08:29:18.262460: step: 388/1541, loss: 0.13698600232601166 2023-01-23 08:29:19.193109: step: 390/1541, loss: 0.39944037795066833 2023-01-23 08:29:20.093036: step: 392/1541, loss: 0.8012515306472778 2023-01-23 08:29:20.983146: step: 394/1541, loss: 0.5748049020767212 2023-01-23 08:29:21.888629: step: 396/1541, loss: 0.6397297382354736 2023-01-23 08:29:22.740547: step: 398/1541, loss: 0.35809463262557983 2023-01-23 08:29:23.603481: step: 400/1541, loss: 0.07820772379636765 2023-01-23 08:29:24.481570: step: 402/1541, loss: 0.2578110694885254 2023-01-23 08:29:25.381657: step: 404/1541, loss: 0.4215601086616516 2023-01-23 08:29:26.268028: step: 406/1541, loss: 0.344975084066391 2023-01-23 08:29:27.146122: step: 408/1541, loss: 0.18576180934906006 2023-01-23 08:29:28.036893: step: 410/1541, loss: 0.274728924036026 2023-01-23 08:29:28.952189: step: 412/1541, loss: 0.20173729956150055 2023-01-23 08:29:29.855716: step: 414/1541, loss: 0.1006726324558258 2023-01-23 08:29:30.752093: step: 416/1541, loss: 1.1978679895401 2023-01-23 08:29:31.645747: step: 418/1541, loss: 0.2037152796983719 2023-01-23 08:29:32.623585: step: 420/1541, loss: 0.7937273383140564 2023-01-23 08:29:33.510076: step: 422/1541, loss: 0.25827687978744507 2023-01-23 08:29:34.410308: step: 424/1541, loss: 0.14855742454528809 2023-01-23 08:29:35.319280: step: 426/1541, loss: 0.38144344091415405 2023-01-23 08:29:36.214304: step: 428/1541, loss: 0.1392955332994461 2023-01-23 08:29:37.125513: step: 430/1541, loss: 0.7863103747367859 2023-01-23 08:29:38.019091: step: 432/1541, loss: 1.5677496194839478 2023-01-23 08:29:38.906342: step: 434/1541, loss: 0.23640485107898712 2023-01-23 08:29:39.780462: step: 436/1541, loss: 6.682746410369873 2023-01-23 08:29:40.705483: step: 438/1541, loss: 0.5333479642868042 2023-01-23 08:29:41.632051: step: 440/1541, loss: 0.4462403953075409 2023-01-23 08:29:42.478324: step: 442/1541, loss: 0.21430566906929016 2023-01-23 08:29:43.331788: step: 444/1541, loss: 0.32456985116004944 2023-01-23 08:29:44.179676: step: 446/1541, loss: 0.20582491159439087 2023-01-23 08:29:45.070755: step: 448/1541, loss: 0.484720379114151 2023-01-23 08:29:45.961011: step: 450/1541, loss: 0.5268211364746094 2023-01-23 08:29:46.910632: step: 452/1541, loss: 2.1941208839416504 2023-01-23 08:29:47.797790: step: 454/1541, loss: 0.21621575951576233 2023-01-23 08:29:48.697783: step: 456/1541, loss: 0.4416951537132263 2023-01-23 08:29:49.575075: step: 458/1541, loss: 3.0485496520996094 2023-01-23 08:29:50.441180: step: 460/1541, loss: 1.7346560955047607 2023-01-23 08:29:51.326764: step: 462/1541, loss: 0.2685045897960663 2023-01-23 08:29:52.181114: step: 464/1541, loss: 0.3650919198989868 2023-01-23 08:29:53.020911: step: 466/1541, loss: 0.13363616168498993 2023-01-23 08:29:53.923767: step: 468/1541, loss: 0.13520768284797668 2023-01-23 08:29:54.797522: step: 470/1541, loss: 0.48177021741867065 2023-01-23 08:29:55.657959: step: 472/1541, loss: 0.1882304549217224 2023-01-23 08:29:56.552946: step: 474/1541, loss: 1.5592635869979858 2023-01-23 08:29:57.424729: step: 476/1541, loss: 0.36058205366134644 2023-01-23 08:29:58.338594: step: 478/1541, loss: 0.16266925632953644 2023-01-23 08:29:59.222992: step: 480/1541, loss: 0.1960209310054779 2023-01-23 08:30:00.092740: step: 482/1541, loss: 0.3955203592777252 2023-01-23 08:30:01.030340: step: 484/1541, loss: 0.24714389443397522 2023-01-23 08:30:01.947506: step: 486/1541, loss: 0.4978666305541992 2023-01-23 08:30:02.835939: step: 488/1541, loss: 0.3672775328159332 2023-01-23 08:30:03.743091: step: 490/1541, loss: 0.19905082881450653 2023-01-23 08:30:04.619097: step: 492/1541, loss: 0.6397659182548523 2023-01-23 08:30:05.518831: step: 494/1541, loss: 0.10973998159170151 2023-01-23 08:30:06.417532: step: 496/1541, loss: 0.5213236808776855 2023-01-23 08:30:07.313681: step: 498/1541, loss: 0.47932180762290955 2023-01-23 08:30:08.187666: step: 500/1541, loss: 0.4322316646575928 2023-01-23 08:30:09.146549: step: 502/1541, loss: 0.11883854120969772 2023-01-23 08:30:10.069751: step: 504/1541, loss: 0.04202457144856453 2023-01-23 08:30:10.938446: step: 506/1541, loss: 0.1869208663702011 2023-01-23 08:30:11.828355: step: 508/1541, loss: 0.24159564077854156 2023-01-23 08:30:12.728693: step: 510/1541, loss: 0.2433539479970932 2023-01-23 08:30:13.650613: step: 512/1541, loss: 0.36284831166267395 2023-01-23 08:30:14.578075: step: 514/1541, loss: 0.2882801592350006 2023-01-23 08:30:15.504647: step: 516/1541, loss: 0.7294763326644897 2023-01-23 08:30:16.373063: step: 518/1541, loss: 0.20194245874881744 2023-01-23 08:30:17.271827: step: 520/1541, loss: 0.19908370077610016 2023-01-23 08:30:18.170543: step: 522/1541, loss: 0.41796013712882996 2023-01-23 08:30:19.075616: step: 524/1541, loss: 0.23977124691009521 2023-01-23 08:30:19.970554: step: 526/1541, loss: 0.4369518756866455 2023-01-23 08:30:20.949364: step: 528/1541, loss: 0.2187320739030838 2023-01-23 08:30:21.835733: step: 530/1541, loss: 0.45716825127601624 2023-01-23 08:30:22.724236: step: 532/1541, loss: 0.17617806792259216 2023-01-23 08:30:23.591510: step: 534/1541, loss: 0.37426337599754333 2023-01-23 08:30:24.446194: step: 536/1541, loss: 0.22789603471755981 2023-01-23 08:30:25.332805: step: 538/1541, loss: 0.18400344252586365 2023-01-23 08:30:26.260986: step: 540/1541, loss: 0.09015686810016632 2023-01-23 08:30:27.161807: step: 542/1541, loss: 0.08141035586595535 2023-01-23 08:30:28.113049: step: 544/1541, loss: 0.3763066530227661 2023-01-23 08:30:28.981872: step: 546/1541, loss: 0.19467346370220184 2023-01-23 08:30:29.883058: step: 548/1541, loss: 0.6898791193962097 2023-01-23 08:30:30.796050: step: 550/1541, loss: 0.31362757086753845 2023-01-23 08:30:31.690867: step: 552/1541, loss: 0.6027659177780151 2023-01-23 08:30:32.609490: step: 554/1541, loss: 0.3650335967540741 2023-01-23 08:30:33.491516: step: 556/1541, loss: 0.4224291741847992 2023-01-23 08:30:34.364640: step: 558/1541, loss: 0.7707064747810364 2023-01-23 08:30:35.283960: step: 560/1541, loss: 0.32461389899253845 2023-01-23 08:30:36.232043: step: 562/1541, loss: 0.09881287068128586 2023-01-23 08:30:37.180203: step: 564/1541, loss: 0.3992837071418762 2023-01-23 08:30:38.047692: step: 566/1541, loss: 0.32323217391967773 2023-01-23 08:30:38.980386: step: 568/1541, loss: 0.4090976417064667 2023-01-23 08:30:39.909334: step: 570/1541, loss: 0.15114207565784454 2023-01-23 08:30:40.867038: step: 572/1541, loss: 0.16840362548828125 2023-01-23 08:30:41.782760: step: 574/1541, loss: 0.18898136913776398 2023-01-23 08:30:42.669759: step: 576/1541, loss: 0.051270026713609695 2023-01-23 08:30:43.622491: step: 578/1541, loss: 0.35568898916244507 2023-01-23 08:30:44.546962: step: 580/1541, loss: 0.09289038181304932 2023-01-23 08:30:45.474595: step: 582/1541, loss: 0.24647611379623413 2023-01-23 08:30:46.386048: step: 584/1541, loss: 0.27983155846595764 2023-01-23 08:30:47.343052: step: 586/1541, loss: 1.0221837759017944 2023-01-23 08:30:48.248222: step: 588/1541, loss: 0.1509568989276886 2023-01-23 08:30:49.117688: step: 590/1541, loss: 0.3104735314846039 2023-01-23 08:30:50.029137: step: 592/1541, loss: 1.4532992839813232 2023-01-23 08:30:50.953753: step: 594/1541, loss: 0.7861902713775635 2023-01-23 08:30:51.910101: step: 596/1541, loss: 0.4640372395515442 2023-01-23 08:30:52.782774: step: 598/1541, loss: 0.3820495009422302 2023-01-23 08:30:53.713585: step: 600/1541, loss: 0.09101098030805588 2023-01-23 08:30:54.571760: step: 602/1541, loss: 0.7845120429992676 2023-01-23 08:30:55.529973: step: 604/1541, loss: 0.5452173352241516 2023-01-23 08:30:56.377208: step: 606/1541, loss: 0.4203261733055115 2023-01-23 08:30:57.264098: step: 608/1541, loss: 0.22583241760730743 2023-01-23 08:30:58.142163: step: 610/1541, loss: 0.8402143716812134 2023-01-23 08:30:59.122981: step: 612/1541, loss: 0.32172200083732605 2023-01-23 08:31:00.005235: step: 614/1541, loss: 0.6810522079467773 2023-01-23 08:31:00.940874: step: 616/1541, loss: 0.243846595287323 2023-01-23 08:31:01.867340: step: 618/1541, loss: 0.6745726466178894 2023-01-23 08:31:02.833639: step: 620/1541, loss: 0.530299186706543 2023-01-23 08:31:03.723191: step: 622/1541, loss: 0.2856835424900055 2023-01-23 08:31:04.614277: step: 624/1541, loss: 0.2768542170524597 2023-01-23 08:31:05.525178: step: 626/1541, loss: 0.3699389100074768 2023-01-23 08:31:06.444154: step: 628/1541, loss: 0.03845856338739395 2023-01-23 08:31:07.400112: step: 630/1541, loss: 0.15958641469478607 2023-01-23 08:31:08.297512: step: 632/1541, loss: 0.28781193494796753 2023-01-23 08:31:09.220446: step: 634/1541, loss: 0.712994396686554 2023-01-23 08:31:10.143622: step: 636/1541, loss: 0.4011119604110718 2023-01-23 08:31:11.034937: step: 638/1541, loss: 0.20299585163593292 2023-01-23 08:31:11.967639: step: 640/1541, loss: 0.6453062295913696 2023-01-23 08:31:12.861395: step: 642/1541, loss: 0.10512293130159378 2023-01-23 08:31:13.743011: step: 644/1541, loss: 0.11156363785266876 2023-01-23 08:31:14.601757: step: 646/1541, loss: 0.23377631604671478 2023-01-23 08:31:15.568373: step: 648/1541, loss: 0.4704728424549103 2023-01-23 08:31:16.516366: step: 650/1541, loss: 0.2469640076160431 2023-01-23 08:31:17.397528: step: 652/1541, loss: 0.21659035980701447 2023-01-23 08:31:18.353928: step: 654/1541, loss: 0.2814634144306183 2023-01-23 08:31:19.197744: step: 656/1541, loss: 0.18545600771903992 2023-01-23 08:31:20.051681: step: 658/1541, loss: 0.6186196804046631 2023-01-23 08:31:21.008991: step: 660/1541, loss: 0.3037784695625305 2023-01-23 08:31:21.852307: step: 662/1541, loss: 1.1168867349624634 2023-01-23 08:31:22.732545: step: 664/1541, loss: 0.27956852316856384 2023-01-23 08:31:23.659955: step: 666/1541, loss: 0.3349986672401428 2023-01-23 08:31:24.521852: step: 668/1541, loss: 0.3103780746459961 2023-01-23 08:31:25.426123: step: 670/1541, loss: 0.31018802523612976 2023-01-23 08:31:26.305851: step: 672/1541, loss: 0.23334205150604248 2023-01-23 08:31:27.192212: step: 674/1541, loss: 0.13397960364818573 2023-01-23 08:31:28.105779: step: 676/1541, loss: 0.24775786697864532 2023-01-23 08:31:29.001173: step: 678/1541, loss: 0.054243117570877075 2023-01-23 08:31:29.905010: step: 680/1541, loss: 0.27819493412971497 2023-01-23 08:31:30.800619: step: 682/1541, loss: 0.14233683049678802 2023-01-23 08:31:31.673401: step: 684/1541, loss: 0.8290404677391052 2023-01-23 08:31:32.619823: step: 686/1541, loss: 0.07500151544809341 2023-01-23 08:31:33.544432: step: 688/1541, loss: 1.004034161567688 2023-01-23 08:31:34.476374: step: 690/1541, loss: 0.26544108986854553 2023-01-23 08:31:35.346696: step: 692/1541, loss: 0.15782786905765533 2023-01-23 08:31:36.241902: step: 694/1541, loss: 0.8817427158355713 2023-01-23 08:31:37.148994: step: 696/1541, loss: 0.056677885353565216 2023-01-23 08:31:38.090408: step: 698/1541, loss: 0.10722021758556366 2023-01-23 08:31:38.943516: step: 700/1541, loss: 0.10385723412036896 2023-01-23 08:31:39.820904: step: 702/1541, loss: 0.8616011142730713 2023-01-23 08:31:40.737520: step: 704/1541, loss: 0.21695345640182495 2023-01-23 08:31:41.669776: step: 706/1541, loss: 0.8210453391075134 2023-01-23 08:31:42.568413: step: 708/1541, loss: 0.3399941921234131 2023-01-23 08:31:43.461009: step: 710/1541, loss: 0.27620694041252136 2023-01-23 08:31:44.350961: step: 712/1541, loss: 0.12839655578136444 2023-01-23 08:31:45.269151: step: 714/1541, loss: 2.2951109409332275 2023-01-23 08:31:46.230408: step: 716/1541, loss: 0.4268982410430908 2023-01-23 08:31:47.178801: step: 718/1541, loss: 0.5679402351379395 2023-01-23 08:31:48.068138: step: 720/1541, loss: 0.4327657222747803 2023-01-23 08:31:48.986542: step: 722/1541, loss: 0.4732581377029419 2023-01-23 08:31:49.881941: step: 724/1541, loss: 0.14840345084667206 2023-01-23 08:31:50.793051: step: 726/1541, loss: 0.42364075779914856 2023-01-23 08:31:51.742549: step: 728/1541, loss: 0.4982748329639435 2023-01-23 08:31:52.675134: step: 730/1541, loss: 0.8702011704444885 2023-01-23 08:31:53.639169: step: 732/1541, loss: 0.05972221866250038 2023-01-23 08:31:54.541562: step: 734/1541, loss: 0.13852082192897797 2023-01-23 08:31:55.381257: step: 736/1541, loss: 0.2889558672904968 2023-01-23 08:31:56.296590: step: 738/1541, loss: 0.23100398480892181 2023-01-23 08:31:57.197396: step: 740/1541, loss: 0.03512898460030556 2023-01-23 08:31:58.113803: step: 742/1541, loss: 0.3834770917892456 2023-01-23 08:31:59.051150: step: 744/1541, loss: 0.23369812965393066 2023-01-23 08:31:59.924996: step: 746/1541, loss: 1.2577942609786987 2023-01-23 08:32:00.850714: step: 748/1541, loss: 0.27194279432296753 2023-01-23 08:32:01.771703: step: 750/1541, loss: 0.3688798248767853 2023-01-23 08:32:02.682704: step: 752/1541, loss: 0.33492979407310486 2023-01-23 08:32:03.599698: step: 754/1541, loss: 0.44792285561561584 2023-01-23 08:32:04.496081: step: 756/1541, loss: 0.47783607244491577 2023-01-23 08:32:05.382527: step: 758/1541, loss: 0.6042757034301758 2023-01-23 08:32:06.289971: step: 760/1541, loss: 0.17332983016967773 2023-01-23 08:32:07.188895: step: 762/1541, loss: 0.8555347323417664 2023-01-23 08:32:08.091056: step: 764/1541, loss: 0.06774295121431351 2023-01-23 08:32:09.008628: step: 766/1541, loss: 0.18698076903820038 2023-01-23 08:32:09.907535: step: 768/1541, loss: 0.12717026472091675 2023-01-23 08:32:10.907017: step: 770/1541, loss: 0.5633063912391663 2023-01-23 08:32:11.794860: step: 772/1541, loss: 0.10527928918600082 2023-01-23 08:32:12.666120: step: 774/1541, loss: 0.18219217658042908 2023-01-23 08:32:13.560949: step: 776/1541, loss: 0.49104151129722595 2023-01-23 08:32:14.437394: step: 778/1541, loss: 0.10098573565483093 2023-01-23 08:32:15.347669: step: 780/1541, loss: 0.15657080709934235 2023-01-23 08:32:16.231449: step: 782/1541, loss: 0.23773601651191711 2023-01-23 08:32:17.145877: step: 784/1541, loss: 0.11670130491256714 2023-01-23 08:32:18.030032: step: 786/1541, loss: 0.7490885853767395 2023-01-23 08:32:18.924197: step: 788/1541, loss: 0.9559379816055298 2023-01-23 08:32:19.806918: step: 790/1541, loss: 0.25439316034317017 2023-01-23 08:32:20.684195: step: 792/1541, loss: 0.15857601165771484 2023-01-23 08:32:21.582000: step: 794/1541, loss: 0.16463524103164673 2023-01-23 08:32:22.451520: step: 796/1541, loss: 0.4128653109073639 2023-01-23 08:32:23.381085: step: 798/1541, loss: 0.12038800120353699 2023-01-23 08:32:24.262475: step: 800/1541, loss: 0.18641377985477448 2023-01-23 08:32:25.201884: step: 802/1541, loss: 0.48713311553001404 2023-01-23 08:32:26.066659: step: 804/1541, loss: 0.13032253086566925 2023-01-23 08:32:26.991563: step: 806/1541, loss: 0.08155515044927597 2023-01-23 08:32:27.883547: step: 808/1541, loss: 1.7395902872085571 2023-01-23 08:32:28.775583: step: 810/1541, loss: 0.31235790252685547 2023-01-23 08:32:29.688611: step: 812/1541, loss: 0.4996591806411743 2023-01-23 08:32:30.568202: step: 814/1541, loss: 0.4564872980117798 2023-01-23 08:32:31.448926: step: 816/1541, loss: 0.2423122674226761 2023-01-23 08:32:32.356713: step: 818/1541, loss: 0.40414443612098694 2023-01-23 08:32:33.233017: step: 820/1541, loss: 0.3642255663871765 2023-01-23 08:32:34.134278: step: 822/1541, loss: 2.3704423904418945 2023-01-23 08:32:35.078394: step: 824/1541, loss: 0.441689670085907 2023-01-23 08:32:35.969953: step: 826/1541, loss: 0.6363021731376648 2023-01-23 08:32:36.881448: step: 828/1541, loss: 0.6839260458946228 2023-01-23 08:32:37.775848: step: 830/1541, loss: 0.2381812483072281 2023-01-23 08:32:38.687055: step: 832/1541, loss: 0.16840089857578278 2023-01-23 08:32:39.553915: step: 834/1541, loss: 0.10771824419498444 2023-01-23 08:32:40.573930: step: 836/1541, loss: 0.07294340431690216 2023-01-23 08:32:41.435353: step: 838/1541, loss: 0.13562335073947906 2023-01-23 08:32:42.389478: step: 840/1541, loss: 0.2573380470275879 2023-01-23 08:32:43.304193: step: 842/1541, loss: 0.20665988326072693 2023-01-23 08:32:44.208826: step: 844/1541, loss: 0.8580670356750488 2023-01-23 08:32:45.123060: step: 846/1541, loss: 0.181913360953331 2023-01-23 08:32:45.984142: step: 848/1541, loss: 0.33348268270492554 2023-01-23 08:32:46.842871: step: 850/1541, loss: 0.3474941551685333 2023-01-23 08:32:47.716080: step: 852/1541, loss: 0.7065410614013672 2023-01-23 08:32:48.595260: step: 854/1541, loss: 0.20562924444675446 2023-01-23 08:32:49.482870: step: 856/1541, loss: 0.15198832750320435 2023-01-23 08:32:50.354900: step: 858/1541, loss: 0.7981809377670288 2023-01-23 08:32:51.259579: step: 860/1541, loss: 0.613726794719696 2023-01-23 08:32:52.201200: step: 862/1541, loss: 0.6841071248054504 2023-01-23 08:32:53.062777: step: 864/1541, loss: 0.3022190034389496 2023-01-23 08:32:53.982060: step: 866/1541, loss: 0.15435321629047394 2023-01-23 08:32:55.027261: step: 868/1541, loss: 0.3733799159526825 2023-01-23 08:32:55.891434: step: 870/1541, loss: 0.256309449672699 2023-01-23 08:32:56.789129: step: 872/1541, loss: 0.40228721499443054 2023-01-23 08:32:57.668810: step: 874/1541, loss: 0.33935362100601196 2023-01-23 08:32:58.545645: step: 876/1541, loss: 0.22340454161167145 2023-01-23 08:32:59.443634: step: 878/1541, loss: 0.1715954691171646 2023-01-23 08:33:00.344624: step: 880/1541, loss: 0.369717001914978 2023-01-23 08:33:01.298346: step: 882/1541, loss: 0.5501030683517456 2023-01-23 08:33:02.220763: step: 884/1541, loss: 0.20789161324501038 2023-01-23 08:33:03.130161: step: 886/1541, loss: 0.2301330864429474 2023-01-23 08:33:04.007770: step: 888/1541, loss: 0.14301452040672302 2023-01-23 08:33:04.921165: step: 890/1541, loss: 0.4622572660446167 2023-01-23 08:33:05.870822: step: 892/1541, loss: 0.18752002716064453 2023-01-23 08:33:06.766249: step: 894/1541, loss: 0.10893110930919647 2023-01-23 08:33:07.640356: step: 896/1541, loss: 0.043070584535598755 2023-01-23 08:33:08.512479: step: 898/1541, loss: 0.20609423518180847 2023-01-23 08:33:09.403953: step: 900/1541, loss: 0.7390761971473694 2023-01-23 08:33:10.347844: step: 902/1541, loss: 0.29839375615119934 2023-01-23 08:33:11.368709: step: 904/1541, loss: 0.2254573553800583 2023-01-23 08:33:12.268993: step: 906/1541, loss: 0.2319883406162262 2023-01-23 08:33:13.175340: step: 908/1541, loss: 0.5462487936019897 2023-01-23 08:33:14.051918: step: 910/1541, loss: 0.28919467329978943 2023-01-23 08:33:14.944017: step: 912/1541, loss: 0.06495305150747299 2023-01-23 08:33:15.829130: step: 914/1541, loss: 0.19169417023658752 2023-01-23 08:33:16.751086: step: 916/1541, loss: 0.6844806671142578 2023-01-23 08:33:17.636856: step: 918/1541, loss: 0.2526741027832031 2023-01-23 08:33:18.558198: step: 920/1541, loss: 1.5737882852554321 2023-01-23 08:33:19.419681: step: 922/1541, loss: 0.2873963713645935 2023-01-23 08:33:20.287152: step: 924/1541, loss: 0.33577579259872437 2023-01-23 08:33:21.205501: step: 926/1541, loss: 0.1714704930782318 2023-01-23 08:33:22.088713: step: 928/1541, loss: 0.31668221950531006 2023-01-23 08:33:23.062363: step: 930/1541, loss: 0.38787582516670227 2023-01-23 08:33:23.958750: step: 932/1541, loss: 2.1952991485595703 2023-01-23 08:33:24.851453: step: 934/1541, loss: 0.23791129887104034 2023-01-23 08:33:25.752750: step: 936/1541, loss: 0.29705098271369934 2023-01-23 08:33:26.637899: step: 938/1541, loss: 0.24616776406764984 2023-01-23 08:33:27.468260: step: 940/1541, loss: 0.33777371048927307 2023-01-23 08:33:28.372263: step: 942/1541, loss: 0.6387158036231995 2023-01-23 08:33:29.294181: step: 944/1541, loss: 0.22414058446884155 2023-01-23 08:33:30.177709: step: 946/1541, loss: 0.3406793773174286 2023-01-23 08:33:31.070709: step: 948/1541, loss: 0.6640609502792358 2023-01-23 08:33:31.999102: step: 950/1541, loss: 0.3584485352039337 2023-01-23 08:33:32.991286: step: 952/1541, loss: 0.44927358627319336 2023-01-23 08:33:33.853442: step: 954/1541, loss: 0.3329017758369446 2023-01-23 08:33:34.855907: step: 956/1541, loss: 0.18487165868282318 2023-01-23 08:33:35.833878: step: 958/1541, loss: 0.3191041648387909 2023-01-23 08:33:36.744900: step: 960/1541, loss: 0.291284441947937 2023-01-23 08:33:37.677083: step: 962/1541, loss: 0.8550196886062622 2023-01-23 08:33:38.594803: step: 964/1541, loss: 0.304932177066803 2023-01-23 08:33:39.503009: step: 966/1541, loss: 0.08526482433080673 2023-01-23 08:33:40.377156: step: 968/1541, loss: 0.05775326490402222 2023-01-23 08:33:41.288289: step: 970/1541, loss: 0.06938868016004562 2023-01-23 08:33:42.275753: step: 972/1541, loss: 0.2347739189863205 2023-01-23 08:33:43.177941: step: 974/1541, loss: 0.42807164788246155 2023-01-23 08:33:44.064470: step: 976/1541, loss: 0.23820118606090546 2023-01-23 08:33:44.994598: step: 978/1541, loss: 0.2347961962223053 2023-01-23 08:33:45.871363: step: 980/1541, loss: 0.505066990852356 2023-01-23 08:33:46.741298: step: 982/1541, loss: 0.08242340385913849 2023-01-23 08:33:47.665971: step: 984/1541, loss: 0.23973217606544495 2023-01-23 08:33:48.521551: step: 986/1541, loss: 0.21527667343616486 2023-01-23 08:33:49.415458: step: 988/1541, loss: 0.883880615234375 2023-01-23 08:33:50.289312: step: 990/1541, loss: 0.12345483154058456 2023-01-23 08:33:51.183775: step: 992/1541, loss: 0.5530388951301575 2023-01-23 08:33:52.100888: step: 994/1541, loss: 0.1348269283771515 2023-01-23 08:33:53.014547: step: 996/1541, loss: 0.4397215247154236 2023-01-23 08:33:53.924875: step: 998/1541, loss: 0.3771911561489105 2023-01-23 08:33:54.818318: step: 1000/1541, loss: 0.35827815532684326 2023-01-23 08:33:55.659184: step: 1002/1541, loss: 0.2890794277191162 2023-01-23 08:33:56.518978: step: 1004/1541, loss: 0.21495068073272705 2023-01-23 08:33:57.486309: step: 1006/1541, loss: 0.39130890369415283 2023-01-23 08:33:58.363992: step: 1008/1541, loss: 0.0797007605433464 2023-01-23 08:33:59.295750: step: 1010/1541, loss: 0.251566618680954 2023-01-23 08:34:00.198699: step: 1012/1541, loss: 0.5177432894706726 2023-01-23 08:34:01.068772: step: 1014/1541, loss: 0.5503188967704773 2023-01-23 08:34:01.966060: step: 1016/1541, loss: 0.15749038755893707 2023-01-23 08:34:02.929477: step: 1018/1541, loss: 0.30708467960357666 2023-01-23 08:34:03.820969: step: 1020/1541, loss: 0.6466546654701233 2023-01-23 08:34:04.728822: step: 1022/1541, loss: 0.7868139743804932 2023-01-23 08:34:05.613700: step: 1024/1541, loss: 0.28918367624282837 2023-01-23 08:34:06.618961: step: 1026/1541, loss: 0.2505778670310974 2023-01-23 08:34:07.501126: step: 1028/1541, loss: 0.2464020997285843 2023-01-23 08:34:08.416407: step: 1030/1541, loss: 0.20665672421455383 2023-01-23 08:34:09.342335: step: 1032/1541, loss: 0.41766926646232605 2023-01-23 08:34:10.223637: step: 1034/1541, loss: 0.245699942111969 2023-01-23 08:34:11.111166: step: 1036/1541, loss: 0.2478688657283783 2023-01-23 08:34:12.027069: step: 1038/1541, loss: 0.3928442895412445 2023-01-23 08:34:12.905457: step: 1040/1541, loss: 0.0872078537940979 2023-01-23 08:34:13.781781: step: 1042/1541, loss: 0.4928225874900818 2023-01-23 08:34:14.719076: step: 1044/1541, loss: 0.5507192015647888 2023-01-23 08:34:15.558204: step: 1046/1541, loss: 0.1648120880126953 2023-01-23 08:34:16.480331: step: 1048/1541, loss: 0.563238263130188 2023-01-23 08:34:17.377129: step: 1050/1541, loss: 0.2177170217037201 2023-01-23 08:34:18.299716: step: 1052/1541, loss: 0.32968631386756897 2023-01-23 08:34:19.187796: step: 1054/1541, loss: 0.476439505815506 2023-01-23 08:34:20.061155: step: 1056/1541, loss: 0.18091759085655212 2023-01-23 08:34:20.926018: step: 1058/1541, loss: 0.3608913719654083 2023-01-23 08:34:21.829278: step: 1060/1541, loss: 0.480849951505661 2023-01-23 08:34:22.681164: step: 1062/1541, loss: 0.39912649989128113 2023-01-23 08:34:23.573273: step: 1064/1541, loss: 0.23174653947353363 2023-01-23 08:34:24.458451: step: 1066/1541, loss: 0.37582162022590637 2023-01-23 08:34:25.336798: step: 1068/1541, loss: 0.13542242348194122 2023-01-23 08:34:26.250272: step: 1070/1541, loss: 0.2542039752006531 2023-01-23 08:34:27.143919: step: 1072/1541, loss: 0.2662966549396515 2023-01-23 08:34:28.074140: step: 1074/1541, loss: 0.09999299794435501 2023-01-23 08:34:28.971297: step: 1076/1541, loss: 0.6352642178535461 2023-01-23 08:34:29.815957: step: 1078/1541, loss: 0.1773248016834259 2023-01-23 08:34:30.687983: step: 1080/1541, loss: 0.2096543312072754 2023-01-23 08:34:31.577901: step: 1082/1541, loss: 0.2870991826057434 2023-01-23 08:34:32.545215: step: 1084/1541, loss: 0.13007821142673492 2023-01-23 08:34:33.516141: step: 1086/1541, loss: 0.2609751224517822 2023-01-23 08:34:34.425129: step: 1088/1541, loss: 0.43803301453590393 2023-01-23 08:34:35.339890: step: 1090/1541, loss: 0.4591968357563019 2023-01-23 08:34:36.223086: step: 1092/1541, loss: 0.7317426204681396 2023-01-23 08:34:37.104886: step: 1094/1541, loss: 0.3549436032772064 2023-01-23 08:34:37.980837: step: 1096/1541, loss: 0.389618456363678 2023-01-23 08:34:38.885873: step: 1098/1541, loss: 0.5937238931655884 2023-01-23 08:34:39.750306: step: 1100/1541, loss: 0.6287765502929688 2023-01-23 08:34:40.620337: step: 1102/1541, loss: 0.4034788906574249 2023-01-23 08:34:41.528483: step: 1104/1541, loss: 0.08048451691865921 2023-01-23 08:34:42.442110: step: 1106/1541, loss: 0.35189691185951233 2023-01-23 08:34:43.341149: step: 1108/1541, loss: 0.2806164026260376 2023-01-23 08:34:44.242587: step: 1110/1541, loss: 0.37241435050964355 2023-01-23 08:34:45.140342: step: 1112/1541, loss: 0.6181018948554993 2023-01-23 08:34:46.053303: step: 1114/1541, loss: 0.41607069969177246 2023-01-23 08:34:46.981463: step: 1116/1541, loss: 0.49896642565727234 2023-01-23 08:34:47.942585: step: 1118/1541, loss: 0.18803562223911285 2023-01-23 08:34:48.837483: step: 1120/1541, loss: 0.6014017462730408 2023-01-23 08:34:49.752402: step: 1122/1541, loss: 0.24998259544372559 2023-01-23 08:34:50.623418: step: 1124/1541, loss: 0.33035239577293396 2023-01-23 08:34:51.542782: step: 1126/1541, loss: 0.23334823548793793 2023-01-23 08:34:52.411106: step: 1128/1541, loss: 0.6676085591316223 2023-01-23 08:34:53.312740: step: 1130/1541, loss: 0.13788385689258575 2023-01-23 08:34:54.264698: step: 1132/1541, loss: 0.8407596349716187 2023-01-23 08:34:55.171557: step: 1134/1541, loss: 0.05388827249407768 2023-01-23 08:34:56.126783: step: 1136/1541, loss: 0.14324751496315002 2023-01-23 08:34:57.005712: step: 1138/1541, loss: 0.4026322364807129 2023-01-23 08:34:57.914395: step: 1140/1541, loss: 0.1255100965499878 2023-01-23 08:34:58.825009: step: 1142/1541, loss: 0.28043609857559204 2023-01-23 08:34:59.747740: step: 1144/1541, loss: 0.21465568244457245 2023-01-23 08:35:00.582880: step: 1146/1541, loss: 0.24014264345169067 2023-01-23 08:35:01.484258: step: 1148/1541, loss: 0.24367225170135498 2023-01-23 08:35:02.391692: step: 1150/1541, loss: 0.18713566660881042 2023-01-23 08:35:03.293175: step: 1152/1541, loss: 0.2914345860481262 2023-01-23 08:35:04.163498: step: 1154/1541, loss: 0.4228455126285553 2023-01-23 08:35:05.056433: step: 1156/1541, loss: 0.3068799376487732 2023-01-23 08:35:05.920157: step: 1158/1541, loss: 0.13234184682369232 2023-01-23 08:35:06.815749: step: 1160/1541, loss: 0.8396160006523132 2023-01-23 08:35:07.713900: step: 1162/1541, loss: 0.11079775542020798 2023-01-23 08:35:08.571765: step: 1164/1541, loss: 0.303480327129364 2023-01-23 08:35:09.490032: step: 1166/1541, loss: 0.1865474134683609 2023-01-23 08:35:10.393506: step: 1168/1541, loss: 0.15451611578464508 2023-01-23 08:35:11.324913: step: 1170/1541, loss: 0.3254542350769043 2023-01-23 08:35:12.181375: step: 1172/1541, loss: 0.8040919899940491 2023-01-23 08:35:13.074914: step: 1174/1541, loss: 0.46917039155960083 2023-01-23 08:35:13.964431: step: 1176/1541, loss: 0.11041927337646484 2023-01-23 08:35:14.833105: step: 1178/1541, loss: 0.662546694278717 2023-01-23 08:35:15.720715: step: 1180/1541, loss: 0.36037567257881165 2023-01-23 08:35:16.579164: step: 1182/1541, loss: 0.32077476382255554 2023-01-23 08:35:17.581253: step: 1184/1541, loss: 0.5724806785583496 2023-01-23 08:35:18.498970: step: 1186/1541, loss: 0.5406301021575928 2023-01-23 08:35:19.397987: step: 1188/1541, loss: 0.2957630157470703 2023-01-23 08:35:20.281075: step: 1190/1541, loss: 0.4246107339859009 2023-01-23 08:35:21.171983: step: 1192/1541, loss: 0.2360759824514389 2023-01-23 08:35:22.047080: step: 1194/1541, loss: 0.3574409782886505 2023-01-23 08:35:22.968835: step: 1196/1541, loss: 0.06381846964359283 2023-01-23 08:35:23.868980: step: 1198/1541, loss: 0.11947043240070343 2023-01-23 08:35:24.793448: step: 1200/1541, loss: 0.8118222951889038 2023-01-23 08:35:25.661746: step: 1202/1541, loss: 0.2529664635658264 2023-01-23 08:35:26.590659: step: 1204/1541, loss: 0.8353080153465271 2023-01-23 08:35:27.496668: step: 1206/1541, loss: 0.20431536436080933 2023-01-23 08:35:28.430346: step: 1208/1541, loss: 0.33704233169555664 2023-01-23 08:35:29.318977: step: 1210/1541, loss: 0.31771379709243774 2023-01-23 08:35:30.195162: step: 1212/1541, loss: 0.11969727277755737 2023-01-23 08:35:31.087690: step: 1214/1541, loss: 0.5515315532684326 2023-01-23 08:35:31.962887: step: 1216/1541, loss: 0.38952094316482544 2023-01-23 08:35:32.988313: step: 1218/1541, loss: 0.7978955507278442 2023-01-23 08:35:33.868933: step: 1220/1541, loss: 0.13736270368099213 2023-01-23 08:35:34.842677: step: 1222/1541, loss: 0.24598652124404907 2023-01-23 08:35:35.765698: step: 1224/1541, loss: 0.12441441416740417 2023-01-23 08:35:36.642913: step: 1226/1541, loss: 0.33494552969932556 2023-01-23 08:35:37.576678: step: 1228/1541, loss: 0.25519996881484985 2023-01-23 08:35:38.443895: step: 1230/1541, loss: 0.24666538834571838 2023-01-23 08:35:39.324678: step: 1232/1541, loss: 0.43826112151145935 2023-01-23 08:35:40.211857: step: 1234/1541, loss: 0.22097806632518768 2023-01-23 08:35:41.133902: step: 1236/1541, loss: 0.3092517852783203 2023-01-23 08:35:42.006980: step: 1238/1541, loss: 0.2845372259616852 2023-01-23 08:35:42.898211: step: 1240/1541, loss: 0.3623139560222626 2023-01-23 08:35:43.806675: step: 1242/1541, loss: 0.37559449672698975 2023-01-23 08:35:44.698898: step: 1244/1541, loss: 0.8646624088287354 2023-01-23 08:35:45.563086: step: 1246/1541, loss: 0.5978426337242126 2023-01-23 08:35:46.449318: step: 1248/1541, loss: 0.24442169070243835 2023-01-23 08:35:47.340763: step: 1250/1541, loss: 0.13546520471572876 2023-01-23 08:35:48.219994: step: 1252/1541, loss: 0.6540672183036804 2023-01-23 08:35:49.078245: step: 1254/1541, loss: 0.39073312282562256 2023-01-23 08:35:49.968523: step: 1256/1541, loss: 0.49506330490112305 2023-01-23 08:35:50.856926: step: 1258/1541, loss: 0.30767327547073364 2023-01-23 08:35:51.794700: step: 1260/1541, loss: 0.18526460230350494 2023-01-23 08:35:52.675771: step: 1262/1541, loss: 0.059662822633981705 2023-01-23 08:35:53.551698: step: 1264/1541, loss: 0.9550172686576843 2023-01-23 08:35:54.419399: step: 1266/1541, loss: 0.20004987716674805 2023-01-23 08:35:55.345838: step: 1268/1541, loss: 0.47127941250801086 2023-01-23 08:35:56.230987: step: 1270/1541, loss: 0.17785249650478363 2023-01-23 08:35:57.203959: step: 1272/1541, loss: 0.973026692867279 2023-01-23 08:35:58.107132: step: 1274/1541, loss: 0.4812963008880615 2023-01-23 08:35:58.987610: step: 1276/1541, loss: 0.2112557590007782 2023-01-23 08:35:59.871974: step: 1278/1541, loss: 0.6579480767250061 2023-01-23 08:36:00.775562: step: 1280/1541, loss: 0.2628234624862671 2023-01-23 08:36:01.714682: step: 1282/1541, loss: 0.21819737553596497 2023-01-23 08:36:02.651524: step: 1284/1541, loss: 0.23399683833122253 2023-01-23 08:36:03.585433: step: 1286/1541, loss: 0.15375976264476776 2023-01-23 08:36:04.472963: step: 1288/1541, loss: 0.9729387164115906 2023-01-23 08:36:05.480971: step: 1290/1541, loss: 0.6592154502868652 2023-01-23 08:36:06.435751: step: 1292/1541, loss: 0.9296486377716064 2023-01-23 08:36:07.311322: step: 1294/1541, loss: 0.31457728147506714 2023-01-23 08:36:08.176622: step: 1296/1541, loss: 0.28909122943878174 2023-01-23 08:36:09.034382: step: 1298/1541, loss: 0.17825832962989807 2023-01-23 08:36:09.919674: step: 1300/1541, loss: 2.5730977058410645 2023-01-23 08:36:10.810311: step: 1302/1541, loss: 0.42690956592559814 2023-01-23 08:36:11.694835: step: 1304/1541, loss: 0.31871268153190613 2023-01-23 08:36:12.558038: step: 1306/1541, loss: 0.3044290840625763 2023-01-23 08:36:13.441610: step: 1308/1541, loss: 0.7143298387527466 2023-01-23 08:36:14.355045: step: 1310/1541, loss: 5.70650577545166 2023-01-23 08:36:15.248385: step: 1312/1541, loss: 0.48668909072875977 2023-01-23 08:36:16.168516: step: 1314/1541, loss: 0.5591357350349426 2023-01-23 08:36:17.154686: step: 1316/1541, loss: 0.33276498317718506 2023-01-23 08:36:18.018959: step: 1318/1541, loss: 0.4152753949165344 2023-01-23 08:36:18.920470: step: 1320/1541, loss: 0.33674514293670654 2023-01-23 08:36:19.777026: step: 1322/1541, loss: 0.37288811802864075 2023-01-23 08:36:20.718401: step: 1324/1541, loss: 0.08342667669057846 2023-01-23 08:36:21.622265: step: 1326/1541, loss: 0.5239245891571045 2023-01-23 08:36:22.534668: step: 1328/1541, loss: 0.23211202025413513 2023-01-23 08:36:23.406154: step: 1330/1541, loss: 0.053385406732559204 2023-01-23 08:36:24.293002: step: 1332/1541, loss: 0.11654054373502731 2023-01-23 08:36:25.205351: step: 1334/1541, loss: 0.4182012975215912 2023-01-23 08:36:26.054979: step: 1336/1541, loss: 0.34045371413230896 2023-01-23 08:36:26.921969: step: 1338/1541, loss: 0.20453403890132904 2023-01-23 08:36:27.789621: step: 1340/1541, loss: 0.23172733187675476 2023-01-23 08:36:28.718458: step: 1342/1541, loss: 0.6193944215774536 2023-01-23 08:36:29.661832: step: 1344/1541, loss: 0.2326865792274475 2023-01-23 08:36:30.565328: step: 1346/1541, loss: 0.24087798595428467 2023-01-23 08:36:31.475702: step: 1348/1541, loss: 6.5349907875061035 2023-01-23 08:36:32.397692: step: 1350/1541, loss: 0.3635169267654419 2023-01-23 08:36:33.303958: step: 1352/1541, loss: 0.3335595726966858 2023-01-23 08:36:34.214189: step: 1354/1541, loss: 0.12568721175193787 2023-01-23 08:36:35.119770: step: 1356/1541, loss: 0.28722840547561646 2023-01-23 08:36:35.979022: step: 1358/1541, loss: 0.6660380363464355 2023-01-23 08:36:36.897287: step: 1360/1541, loss: 0.22261960804462433 2023-01-23 08:36:37.796141: step: 1362/1541, loss: 0.5108412504196167 2023-01-23 08:36:38.684033: step: 1364/1541, loss: 0.23665672540664673 2023-01-23 08:36:39.604185: step: 1366/1541, loss: 0.48853063583374023 2023-01-23 08:36:40.489877: step: 1368/1541, loss: 0.25398150086402893 2023-01-23 08:36:41.366239: step: 1370/1541, loss: 0.7445465922355652 2023-01-23 08:36:42.254275: step: 1372/1541, loss: 0.6558852791786194 2023-01-23 08:36:43.116806: step: 1374/1541, loss: 0.5328109264373779 2023-01-23 08:36:43.975239: step: 1376/1541, loss: 0.6069847941398621 2023-01-23 08:36:44.853435: step: 1378/1541, loss: 0.2788829803466797 2023-01-23 08:36:45.757610: step: 1380/1541, loss: 0.3581686317920685 2023-01-23 08:36:46.645188: step: 1382/1541, loss: 0.8145251870155334 2023-01-23 08:36:47.514357: step: 1384/1541, loss: 0.43482738733291626 2023-01-23 08:36:48.371847: step: 1386/1541, loss: 0.10163240879774094 2023-01-23 08:36:49.242275: step: 1388/1541, loss: 0.4682327210903168 2023-01-23 08:36:50.127736: step: 1390/1541, loss: 0.22677722573280334 2023-01-23 08:36:50.997793: step: 1392/1541, loss: 0.3996405005455017 2023-01-23 08:36:51.902211: step: 1394/1541, loss: 0.375044047832489 2023-01-23 08:36:52.814343: step: 1396/1541, loss: 0.32310035824775696 2023-01-23 08:36:53.688461: step: 1398/1541, loss: 0.2907094359397888 2023-01-23 08:36:54.585194: step: 1400/1541, loss: 0.514477014541626 2023-01-23 08:36:55.463108: step: 1402/1541, loss: 0.4515198767185211 2023-01-23 08:36:56.379768: step: 1404/1541, loss: 0.5355935096740723 2023-01-23 08:36:57.312171: step: 1406/1541, loss: 0.15118126571178436 2023-01-23 08:36:58.225473: step: 1408/1541, loss: 0.1599118858575821 2023-01-23 08:36:59.102306: step: 1410/1541, loss: 0.05651652812957764 2023-01-23 08:36:59.985550: step: 1412/1541, loss: 1.388922929763794 2023-01-23 08:37:00.846231: step: 1414/1541, loss: 0.9286857843399048 2023-01-23 08:37:01.733280: step: 1416/1541, loss: 1.5245795249938965 2023-01-23 08:37:02.656140: step: 1418/1541, loss: 0.45383909344673157 2023-01-23 08:37:03.540373: step: 1420/1541, loss: 0.61452317237854 2023-01-23 08:37:04.430478: step: 1422/1541, loss: 0.15614885091781616 2023-01-23 08:37:05.349128: step: 1424/1541, loss: 0.6164962649345398 2023-01-23 08:37:06.243876: step: 1426/1541, loss: 0.15266206860542297 2023-01-23 08:37:07.130380: step: 1428/1541, loss: 0.15426164865493774 2023-01-23 08:37:08.018689: step: 1430/1541, loss: 0.1173129752278328 2023-01-23 08:37:08.953298: step: 1432/1541, loss: 0.1356419026851654 2023-01-23 08:37:09.895557: step: 1434/1541, loss: 0.29822590947151184 2023-01-23 08:37:10.840836: step: 1436/1541, loss: 0.2121669501066208 2023-01-23 08:37:11.721171: step: 1438/1541, loss: 0.39645275473594666 2023-01-23 08:37:12.615724: step: 1440/1541, loss: 0.46632349491119385 2023-01-23 08:37:13.533897: step: 1442/1541, loss: 0.3340807855129242 2023-01-23 08:37:14.458413: step: 1444/1541, loss: 0.41715267300605774 2023-01-23 08:37:15.371296: step: 1446/1541, loss: 0.20168247818946838 2023-01-23 08:37:16.243309: step: 1448/1541, loss: 3.07072377204895 2023-01-23 08:37:17.157918: step: 1450/1541, loss: 0.18472370505332947 2023-01-23 08:37:18.053301: step: 1452/1541, loss: 0.1378912329673767 2023-01-23 08:37:18.965236: step: 1454/1541, loss: 0.21300868690013885 2023-01-23 08:37:19.833749: step: 1456/1541, loss: 0.2209184765815735 2023-01-23 08:37:20.716447: step: 1458/1541, loss: 0.17803436517715454 2023-01-23 08:37:21.636396: step: 1460/1541, loss: 0.48274609446525574 2023-01-23 08:37:22.557324: step: 1462/1541, loss: 0.2495938092470169 2023-01-23 08:37:23.417474: step: 1464/1541, loss: 0.2055354118347168 2023-01-23 08:37:24.330857: step: 1466/1541, loss: 0.46917062997817993 2023-01-23 08:37:25.217219: step: 1468/1541, loss: 0.04115413874387741 2023-01-23 08:37:26.104091: step: 1470/1541, loss: 0.578141987323761 2023-01-23 08:37:26.976031: step: 1472/1541, loss: 0.17361211776733398 2023-01-23 08:37:27.849683: step: 1474/1541, loss: 0.22547468543052673 2023-01-23 08:37:28.715071: step: 1476/1541, loss: 0.48529288172721863 2023-01-23 08:37:29.640447: step: 1478/1541, loss: 0.8478308916091919 2023-01-23 08:37:30.540414: step: 1480/1541, loss: 0.34507107734680176 2023-01-23 08:37:31.443595: step: 1482/1541, loss: 0.02140633761882782 2023-01-23 08:37:32.309666: step: 1484/1541, loss: 0.0643143355846405 2023-01-23 08:37:33.179712: step: 1486/1541, loss: 0.6336548924446106 2023-01-23 08:37:34.126379: step: 1488/1541, loss: 0.3145975172519684 2023-01-23 08:37:35.017984: step: 1490/1541, loss: 0.29805630445480347 2023-01-23 08:37:35.986255: step: 1492/1541, loss: 0.08937878161668777 2023-01-23 08:37:36.887502: step: 1494/1541, loss: 0.46448418498039246 2023-01-23 08:37:37.766184: step: 1496/1541, loss: 0.16359254717826843 2023-01-23 08:37:38.670597: step: 1498/1541, loss: 0.2730698585510254 2023-01-23 08:37:39.560926: step: 1500/1541, loss: 0.4020162522792816 2023-01-23 08:37:40.472648: step: 1502/1541, loss: 0.3995341658592224 2023-01-23 08:37:41.438085: step: 1504/1541, loss: 0.36614304780960083 2023-01-23 08:37:42.314725: step: 1506/1541, loss: 0.3947961926460266 2023-01-23 08:37:43.260923: step: 1508/1541, loss: 0.37174466252326965 2023-01-23 08:37:44.161975: step: 1510/1541, loss: 0.22610744833946228 2023-01-23 08:37:45.018124: step: 1512/1541, loss: 0.05340229719877243 2023-01-23 08:37:45.913933: step: 1514/1541, loss: 0.3958969712257385 2023-01-23 08:37:46.817693: step: 1516/1541, loss: 0.37998732924461365 2023-01-23 08:37:47.734629: step: 1518/1541, loss: 0.24764081835746765 2023-01-23 08:37:48.659122: step: 1520/1541, loss: 0.5959717631340027 2023-01-23 08:37:49.545020: step: 1522/1541, loss: 0.13943125307559967 2023-01-23 08:37:50.391813: step: 1524/1541, loss: 0.4691806733608246 2023-01-23 08:37:51.278905: step: 1526/1541, loss: 0.31875479221343994 2023-01-23 08:37:52.194045: step: 1528/1541, loss: 1.608276128768921 2023-01-23 08:37:53.077908: step: 1530/1541, loss: 0.39372238516807556 2023-01-23 08:37:53.946916: step: 1532/1541, loss: 0.07101384550333023 2023-01-23 08:37:54.850138: step: 1534/1541, loss: 0.33321821689605713 2023-01-23 08:37:55.701186: step: 1536/1541, loss: 0.14626802504062653 2023-01-23 08:37:56.558952: step: 1538/1541, loss: 0.28987208008766174 2023-01-23 08:37:57.555845: step: 1540/1541, loss: 0.12783828377723694 2023-01-23 08:37:58.440415: step: 1542/1541, loss: 0.07345323264598846 2023-01-23 08:37:59.372318: step: 1544/1541, loss: 0.2187182605266571 2023-01-23 08:38:00.297284: step: 1546/1541, loss: 0.35277196764945984 2023-01-23 08:38:01.202778: step: 1548/1541, loss: 0.045481156557798386 2023-01-23 08:38:02.088816: step: 1550/1541, loss: 1.328108787536621 2023-01-23 08:38:03.010776: step: 1552/1541, loss: 0.3369249701499939 2023-01-23 08:38:03.942681: step: 1554/1541, loss: 0.1661584973335266 2023-01-23 08:38:04.847770: step: 1556/1541, loss: 0.5658454895019531 2023-01-23 08:38:05.759234: step: 1558/1541, loss: 0.9586595892906189 2023-01-23 08:38:06.676623: step: 1560/1541, loss: 0.23506903648376465 2023-01-23 08:38:07.545436: step: 1562/1541, loss: 0.4441637694835663 2023-01-23 08:38:08.469168: step: 1564/1541, loss: 0.156511589884758 2023-01-23 08:38:09.404393: step: 1566/1541, loss: 0.3005983829498291 2023-01-23 08:38:10.294227: step: 1568/1541, loss: 0.07048358768224716 2023-01-23 08:38:11.193716: step: 1570/1541, loss: 0.27366453409194946 2023-01-23 08:38:12.090579: step: 1572/1541, loss: 0.17554965615272522 2023-01-23 08:38:12.981421: step: 1574/1541, loss: 0.15288925170898438 2023-01-23 08:38:13.908130: step: 1576/1541, loss: 2.636324644088745 2023-01-23 08:38:14.754935: step: 1578/1541, loss: 0.392859548330307 2023-01-23 08:38:15.752011: step: 1580/1541, loss: 0.14294105768203735 2023-01-23 08:38:16.667646: step: 1582/1541, loss: 0.22159676253795624 2023-01-23 08:38:17.565316: step: 1584/1541, loss: 0.74299156665802 2023-01-23 08:38:18.452415: step: 1586/1541, loss: 0.340656042098999 2023-01-23 08:38:19.351400: step: 1588/1541, loss: 0.2871818542480469 2023-01-23 08:38:20.225361: step: 1590/1541, loss: 0.13121475279331207 2023-01-23 08:38:21.080491: step: 1592/1541, loss: 0.28397175669670105 2023-01-23 08:38:21.938617: step: 1594/1541, loss: 0.18689614534378052 2023-01-23 08:38:22.817654: step: 1596/1541, loss: 0.7040547728538513 2023-01-23 08:38:23.746972: step: 1598/1541, loss: 0.4295247197151184 2023-01-23 08:38:24.673554: step: 1600/1541, loss: 0.30938225984573364 2023-01-23 08:38:25.591150: step: 1602/1541, loss: 0.5149834156036377 2023-01-23 08:38:26.508477: step: 1604/1541, loss: 0.2731139659881592 2023-01-23 08:38:27.425849: step: 1606/1541, loss: 0.26095399260520935 2023-01-23 08:38:28.399676: step: 1608/1541, loss: 0.2496783435344696 2023-01-23 08:38:29.308505: step: 1610/1541, loss: 0.22413252294063568 2023-01-23 08:38:30.176382: step: 1612/1541, loss: 0.23736155033111572 2023-01-23 08:38:31.067418: step: 1614/1541, loss: 1.0473871231079102 2023-01-23 08:38:32.001819: step: 1616/1541, loss: 0.14380177855491638 2023-01-23 08:38:32.897820: step: 1618/1541, loss: 0.09788968414068222 2023-01-23 08:38:33.777881: step: 1620/1541, loss: 0.20035184919834137 2023-01-23 08:38:34.657179: step: 1622/1541, loss: 0.31642085313796997 2023-01-23 08:38:35.584509: step: 1624/1541, loss: 0.4712423086166382 2023-01-23 08:38:36.651918: step: 1626/1541, loss: 0.5291101336479187 2023-01-23 08:38:37.527253: step: 1628/1541, loss: 0.2295064628124237 2023-01-23 08:38:38.426255: step: 1630/1541, loss: 0.2886195182800293 2023-01-23 08:38:39.289134: step: 1632/1541, loss: 0.16207768023014069 2023-01-23 08:38:40.144749: step: 1634/1541, loss: 0.1638546735048294 2023-01-23 08:38:41.107117: step: 1636/1541, loss: 0.3166096806526184 2023-01-23 08:38:41.970377: step: 1638/1541, loss: 0.19401293992996216 2023-01-23 08:38:42.848284: step: 1640/1541, loss: 0.5776307582855225 2023-01-23 08:38:43.767181: step: 1642/1541, loss: 0.45582646131515503 2023-01-23 08:38:44.648274: step: 1644/1541, loss: 0.1872657984495163 2023-01-23 08:38:45.575102: step: 1646/1541, loss: 0.12086701393127441 2023-01-23 08:38:46.485969: step: 1648/1541, loss: 0.5407492518424988 2023-01-23 08:38:47.418634: step: 1650/1541, loss: 0.5849198698997498 2023-01-23 08:38:48.311909: step: 1652/1541, loss: 0.2705681025981903 2023-01-23 08:38:49.211328: step: 1654/1541, loss: 0.49976983666419983 2023-01-23 08:38:50.160183: step: 1656/1541, loss: 0.30706578493118286 2023-01-23 08:38:51.062550: step: 1658/1541, loss: 0.11011076718568802 2023-01-23 08:38:51.983330: step: 1660/1541, loss: 0.11094848066568375 2023-01-23 08:38:52.930140: step: 1662/1541, loss: 0.9470345377922058 2023-01-23 08:38:53.815054: step: 1664/1541, loss: 0.2339661717414856 2023-01-23 08:38:54.726602: step: 1666/1541, loss: 0.4574819803237915 2023-01-23 08:38:55.663540: step: 1668/1541, loss: 0.12269807606935501 2023-01-23 08:38:56.511516: step: 1670/1541, loss: 0.07816260308027267 2023-01-23 08:38:57.500316: step: 1672/1541, loss: 0.3836249113082886 2023-01-23 08:38:58.377381: step: 1674/1541, loss: 0.1500188112258911 2023-01-23 08:38:59.226857: step: 1676/1541, loss: 0.49625974893569946 2023-01-23 08:39:00.112661: step: 1678/1541, loss: 0.5408392548561096 2023-01-23 08:39:00.999995: step: 1680/1541, loss: 0.3192576766014099 2023-01-23 08:39:01.889101: step: 1682/1541, loss: 2.2818214893341064 2023-01-23 08:39:02.752142: step: 1684/1541, loss: 0.2264879196882248 2023-01-23 08:39:03.650559: step: 1686/1541, loss: 0.2121826559305191 2023-01-23 08:39:04.557102: step: 1688/1541, loss: 0.6993222236633301 2023-01-23 08:39:05.439140: step: 1690/1541, loss: 0.2428724616765976 2023-01-23 08:39:06.327162: step: 1692/1541, loss: 0.7717711925506592 2023-01-23 08:39:07.278262: step: 1694/1541, loss: 0.16197603940963745 2023-01-23 08:39:08.128010: step: 1696/1541, loss: 0.35116812586784363 2023-01-23 08:39:08.989274: step: 1698/1541, loss: 0.23948219418525696 2023-01-23 08:39:09.854265: step: 1700/1541, loss: 0.8779803514480591 2023-01-23 08:39:10.764572: step: 1702/1541, loss: 0.14915156364440918 2023-01-23 08:39:11.677075: step: 1704/1541, loss: 0.5099900960922241 2023-01-23 08:39:12.610235: step: 1706/1541, loss: 1.120344638824463 2023-01-23 08:39:13.525306: step: 1708/1541, loss: 0.1530326008796692 2023-01-23 08:39:14.391402: step: 1710/1541, loss: 0.42148908972740173 2023-01-23 08:39:15.298933: step: 1712/1541, loss: 0.1295679360628128 2023-01-23 08:39:16.202073: step: 1714/1541, loss: 0.23366494476795197 2023-01-23 08:39:17.097161: step: 1716/1541, loss: 0.37598028779029846 2023-01-23 08:39:17.955571: step: 1718/1541, loss: 0.11597274243831635 2023-01-23 08:39:18.849196: step: 1720/1541, loss: 0.102633036673069 2023-01-23 08:39:19.761872: step: 1722/1541, loss: 0.4153258800506592 2023-01-23 08:39:20.710035: step: 1724/1541, loss: 0.33611148595809937 2023-01-23 08:39:21.571414: step: 1726/1541, loss: 0.2520520091056824 2023-01-23 08:39:22.446903: step: 1728/1541, loss: 0.23120605945587158 2023-01-23 08:39:23.309656: step: 1730/1541, loss: 0.3314112722873688 2023-01-23 08:39:24.216874: step: 1732/1541, loss: 0.12127629667520523 2023-01-23 08:39:25.097429: step: 1734/1541, loss: 0.18815664947032928 2023-01-23 08:39:25.968964: step: 1736/1541, loss: 0.11198629438877106 2023-01-23 08:39:26.853650: step: 1738/1541, loss: 0.2847248613834381 2023-01-23 08:39:27.717621: step: 1740/1541, loss: 0.1179942861199379 2023-01-23 08:39:28.565049: step: 1742/1541, loss: 0.08582418411970139 2023-01-23 08:39:29.473791: step: 1744/1541, loss: 0.8753372430801392 2023-01-23 08:39:30.383901: step: 1746/1541, loss: 0.15589889883995056 2023-01-23 08:39:31.336104: step: 1748/1541, loss: 0.31136199831962585 2023-01-23 08:39:32.237529: step: 1750/1541, loss: 0.42567694187164307 2023-01-23 08:39:33.099948: step: 1752/1541, loss: 0.20999875664710999 2023-01-23 08:39:34.048603: step: 1754/1541, loss: 1.0583170652389526 2023-01-23 08:39:34.903315: step: 1756/1541, loss: 0.36861205101013184 2023-01-23 08:39:35.843322: step: 1758/1541, loss: 0.7219881415367126 2023-01-23 08:39:36.804921: step: 1760/1541, loss: 0.2589603066444397 2023-01-23 08:39:37.694367: step: 1762/1541, loss: 0.5033439993858337 2023-01-23 08:39:38.636680: step: 1764/1541, loss: 0.09476900100708008 2023-01-23 08:39:39.529922: step: 1766/1541, loss: 0.07314592599868774 2023-01-23 08:39:40.494904: step: 1768/1541, loss: 0.21053218841552734 2023-01-23 08:39:41.346274: step: 1770/1541, loss: 0.06287243217229843 2023-01-23 08:39:42.281838: step: 1772/1541, loss: 0.5812272429466248 2023-01-23 08:39:43.185275: step: 1774/1541, loss: 0.21244171261787415 2023-01-23 08:39:44.069892: step: 1776/1541, loss: 0.12914085388183594 2023-01-23 08:39:44.959806: step: 1778/1541, loss: 0.25127699971199036 2023-01-23 08:39:45.900450: step: 1780/1541, loss: 0.361451655626297 2023-01-23 08:39:46.781490: step: 1782/1541, loss: 0.027878044173121452 2023-01-23 08:39:47.658120: step: 1784/1541, loss: 0.6155533194541931 2023-01-23 08:39:48.596391: step: 1786/1541, loss: 0.18399056792259216 2023-01-23 08:39:49.449013: step: 1788/1541, loss: 0.3452518880367279 2023-01-23 08:39:50.323326: step: 1790/1541, loss: 0.20288248360157013 2023-01-23 08:39:51.233158: step: 1792/1541, loss: 0.45852789282798767 2023-01-23 08:39:52.116630: step: 1794/1541, loss: 1.6928319931030273 2023-01-23 08:39:53.043833: step: 1796/1541, loss: 0.120839424431324 2023-01-23 08:39:53.937557: step: 1798/1541, loss: 0.14099030196666718 2023-01-23 08:39:54.887474: step: 1800/1541, loss: 1.0332958698272705 2023-01-23 08:39:55.809516: step: 1802/1541, loss: 0.33075594902038574 2023-01-23 08:39:56.712832: step: 1804/1541, loss: 1.4131231307983398 2023-01-23 08:39:57.672254: step: 1806/1541, loss: 0.24633903801441193 2023-01-23 08:39:58.603739: step: 1808/1541, loss: 0.39419370889663696 2023-01-23 08:39:59.511137: step: 1810/1541, loss: 0.4598924219608307 2023-01-23 08:40:00.418891: step: 1812/1541, loss: 0.26240110397338867 2023-01-23 08:40:01.411320: step: 1814/1541, loss: 0.6132620573043823 2023-01-23 08:40:02.322228: step: 1816/1541, loss: 0.4458830952644348 2023-01-23 08:40:03.193635: step: 1818/1541, loss: 2.678124189376831 2023-01-23 08:40:04.064233: step: 1820/1541, loss: 3.3001301288604736 2023-01-23 08:40:04.953465: step: 1822/1541, loss: 0.28663334250450134 2023-01-23 08:40:05.856638: step: 1824/1541, loss: 0.42263737320899963 2023-01-23 08:40:06.729924: step: 1826/1541, loss: 0.7184319496154785 2023-01-23 08:40:07.597949: step: 1828/1541, loss: 0.09113439172506332 2023-01-23 08:40:08.490474: step: 1830/1541, loss: 0.5262944102287292 2023-01-23 08:40:09.418456: step: 1832/1541, loss: 3.037912130355835 2023-01-23 08:40:10.330266: step: 1834/1541, loss: 0.7961606383323669 2023-01-23 08:40:11.239284: step: 1836/1541, loss: 0.42054736614227295 2023-01-23 08:40:12.131887: step: 1838/1541, loss: 0.12348917871713638 2023-01-23 08:40:13.031435: step: 1840/1541, loss: 0.36358341574668884 2023-01-23 08:40:13.938032: step: 1842/1541, loss: 0.34891337156295776 2023-01-23 08:40:14.847795: step: 1844/1541, loss: 0.5716712474822998 2023-01-23 08:40:15.828604: step: 1846/1541, loss: 0.3733435869216919 2023-01-23 08:40:16.751693: step: 1848/1541, loss: 6.5909318923950195 2023-01-23 08:40:17.637312: step: 1850/1541, loss: 1.765024185180664 2023-01-23 08:40:18.546236: step: 1852/1541, loss: 0.14654621481895447 2023-01-23 08:40:19.482899: step: 1854/1541, loss: 0.5765149593353271 2023-01-23 08:40:20.354602: step: 1856/1541, loss: 0.21690644323825836 2023-01-23 08:40:21.251700: step: 1858/1541, loss: 0.6653696298599243 2023-01-23 08:40:22.142089: step: 1860/1541, loss: 0.21826018393039703 2023-01-23 08:40:23.037484: step: 1862/1541, loss: 0.29685917496681213 2023-01-23 08:40:23.909618: step: 1864/1541, loss: 0.8270794153213501 2023-01-23 08:40:24.859431: step: 1866/1541, loss: 0.10833743214607239 2023-01-23 08:40:25.775071: step: 1868/1541, loss: 0.1083211749792099 2023-01-23 08:40:26.649556: step: 1870/1541, loss: 0.26496830582618713 2023-01-23 08:40:27.487659: step: 1872/1541, loss: 0.08761674165725708 2023-01-23 08:40:28.389273: step: 1874/1541, loss: 0.07300181686878204 2023-01-23 08:40:29.300922: step: 1876/1541, loss: 0.3720702528953552 2023-01-23 08:40:30.267935: step: 1878/1541, loss: 0.4814613461494446 2023-01-23 08:40:31.157677: step: 1880/1541, loss: 0.22369085252285004 2023-01-23 08:40:32.030463: step: 1882/1541, loss: 0.059808049350976944 2023-01-23 08:40:32.890227: step: 1884/1541, loss: 0.3595752418041229 2023-01-23 08:40:33.768327: step: 1886/1541, loss: 0.6015540361404419 2023-01-23 08:40:34.694848: step: 1888/1541, loss: 0.7216262817382812 2023-01-23 08:40:35.652279: step: 1890/1541, loss: 0.1343783587217331 2023-01-23 08:40:36.546657: step: 1892/1541, loss: 0.3338821232318878 2023-01-23 08:40:37.488276: step: 1894/1541, loss: 0.3091242015361786 2023-01-23 08:40:38.400365: step: 1896/1541, loss: 0.26844385266304016 2023-01-23 08:40:39.359060: step: 1898/1541, loss: 0.4188350439071655 2023-01-23 08:40:40.256539: step: 1900/1541, loss: 0.24210718274116516 2023-01-23 08:40:41.131333: step: 1902/1541, loss: 0.20306086540222168 2023-01-23 08:40:42.007788: step: 1904/1541, loss: 0.5116848349571228 2023-01-23 08:40:42.897284: step: 1906/1541, loss: 0.08630871027708054 2023-01-23 08:40:43.790813: step: 1908/1541, loss: 0.49427530169487 2023-01-23 08:40:44.641565: step: 1910/1541, loss: 0.18967001140117645 2023-01-23 08:40:45.570981: step: 1912/1541, loss: 0.1441437005996704 2023-01-23 08:40:46.456663: step: 1914/1541, loss: 0.39146363735198975 2023-01-23 08:40:47.417638: step: 1916/1541, loss: 0.13844230771064758 2023-01-23 08:40:48.290391: step: 1918/1541, loss: 0.26899102330207825 2023-01-23 08:40:49.164941: step: 1920/1541, loss: 0.22967523336410522 2023-01-23 08:40:50.028071: step: 1922/1541, loss: 0.06466708332300186 2023-01-23 08:40:50.921162: step: 1924/1541, loss: 0.21066109836101532 2023-01-23 08:40:51.795409: step: 1926/1541, loss: 0.10050169378519058 2023-01-23 08:40:52.720724: step: 1928/1541, loss: 0.4394597113132477 2023-01-23 08:40:53.636667: step: 1930/1541, loss: 0.4202926456928253 2023-01-23 08:40:54.531945: step: 1932/1541, loss: 0.35205167531967163 2023-01-23 08:40:55.452860: step: 1934/1541, loss: 0.7811616063117981 2023-01-23 08:40:56.351548: step: 1936/1541, loss: 0.6250223517417908 2023-01-23 08:40:57.235940: step: 1938/1541, loss: 0.29542332887649536 2023-01-23 08:40:58.119433: step: 1940/1541, loss: 0.4277648329734802 2023-01-23 08:40:59.014088: step: 1942/1541, loss: 0.17471811175346375 2023-01-23 08:40:59.916894: step: 1944/1541, loss: 0.17055368423461914 2023-01-23 08:41:00.806718: step: 1946/1541, loss: 0.3485257625579834 2023-01-23 08:41:01.706524: step: 1948/1541, loss: 0.7119833827018738 2023-01-23 08:41:02.685498: step: 1950/1541, loss: 0.27477598190307617 2023-01-23 08:41:03.551017: step: 1952/1541, loss: 0.2970939576625824 2023-01-23 08:41:04.423102: step: 1954/1541, loss: 0.1341252624988556 2023-01-23 08:41:05.295033: step: 1956/1541, loss: 0.16931083798408508 2023-01-23 08:41:06.223266: step: 1958/1541, loss: 0.28700441122055054 2023-01-23 08:41:07.111344: step: 1960/1541, loss: 0.2897406816482544 2023-01-23 08:41:08.034670: step: 1962/1541, loss: 0.11484561860561371 2023-01-23 08:41:08.973678: step: 1964/1541, loss: 0.47575414180755615 2023-01-23 08:41:09.890092: step: 1966/1541, loss: 0.38784873485565186 2023-01-23 08:41:10.837811: step: 1968/1541, loss: 0.9653695821762085 2023-01-23 08:41:11.805934: step: 1970/1541, loss: 0.5624181032180786 2023-01-23 08:41:12.719150: step: 1972/1541, loss: 0.11884024739265442 2023-01-23 08:41:13.602543: step: 1974/1541, loss: 0.12356608361005783 2023-01-23 08:41:14.495795: step: 1976/1541, loss: 0.29293930530548096 2023-01-23 08:41:15.348650: step: 1978/1541, loss: 0.4581310749053955 2023-01-23 08:41:16.242609: step: 1980/1541, loss: 0.03161611035466194 2023-01-23 08:41:17.199653: step: 1982/1541, loss: 0.2849188446998596 2023-01-23 08:41:18.118069: step: 1984/1541, loss: 0.6086403727531433 2023-01-23 08:41:19.038473: step: 1986/1541, loss: 0.28121334314346313 2023-01-23 08:41:19.970036: step: 1988/1541, loss: 0.4743480384349823 2023-01-23 08:41:20.850044: step: 1990/1541, loss: 0.13661164045333862 2023-01-23 08:41:21.803447: step: 1992/1541, loss: 0.14679566025733948 2023-01-23 08:41:22.746634: step: 1994/1541, loss: 0.8136275410652161 2023-01-23 08:41:23.615924: step: 1996/1541, loss: 0.37765517830848694 2023-01-23 08:41:24.526227: step: 1998/1541, loss: 0.15248726308345795 2023-01-23 08:41:25.461186: step: 2000/1541, loss: 0.7815495133399963 2023-01-23 08:41:26.370115: step: 2002/1541, loss: 1.5811805725097656 2023-01-23 08:41:27.236887: step: 2004/1541, loss: 0.7177900671958923 2023-01-23 08:41:28.123082: step: 2006/1541, loss: 1.211712121963501 2023-01-23 08:41:28.979961: step: 2008/1541, loss: 0.5257464647293091 2023-01-23 08:41:29.893328: step: 2010/1541, loss: 0.6389713883399963 2023-01-23 08:41:30.798088: step: 2012/1541, loss: 0.38326436281204224 2023-01-23 08:41:31.648119: step: 2014/1541, loss: 0.30857256054878235 2023-01-23 08:41:32.537791: step: 2016/1541, loss: 0.5009207725524902 2023-01-23 08:41:33.417318: step: 2018/1541, loss: 0.2978295683860779 2023-01-23 08:41:34.308376: step: 2020/1541, loss: 0.7282541990280151 2023-01-23 08:41:35.217076: step: 2022/1541, loss: 0.16067461669445038 2023-01-23 08:41:36.149585: step: 2024/1541, loss: 0.3776209354400635 2023-01-23 08:41:37.024204: step: 2026/1541, loss: 0.5650154948234558 2023-01-23 08:41:37.937096: step: 2028/1541, loss: 0.27200791239738464 2023-01-23 08:41:38.824193: step: 2030/1541, loss: 0.3095071613788605 2023-01-23 08:41:39.693148: step: 2032/1541, loss: 0.31708604097366333 2023-01-23 08:41:40.563038: step: 2034/1541, loss: 0.46485939621925354 2023-01-23 08:41:41.464414: step: 2036/1541, loss: 0.2636399567127228 2023-01-23 08:41:42.350547: step: 2038/1541, loss: 0.2602059841156006 2023-01-23 08:41:43.222072: step: 2040/1541, loss: 3.116335391998291 2023-01-23 08:41:44.079029: step: 2042/1541, loss: 0.6464486122131348 2023-01-23 08:41:45.004650: step: 2044/1541, loss: 0.9015677571296692 2023-01-23 08:41:45.912096: step: 2046/1541, loss: 0.302144318819046 2023-01-23 08:41:46.842134: step: 2048/1541, loss: 0.5732295513153076 2023-01-23 08:41:47.796517: step: 2050/1541, loss: 1.3775526285171509 2023-01-23 08:41:48.688846: step: 2052/1541, loss: 0.6441189050674438 2023-01-23 08:41:49.530388: step: 2054/1541, loss: 0.18883420526981354 2023-01-23 08:41:50.481096: step: 2056/1541, loss: 0.44407138228416443 2023-01-23 08:41:51.398771: step: 2058/1541, loss: 0.24786917865276337 2023-01-23 08:41:52.258507: step: 2060/1541, loss: 0.9105544090270996 2023-01-23 08:41:53.170664: step: 2062/1541, loss: 0.2540925145149231 2023-01-23 08:41:54.029571: step: 2064/1541, loss: 0.31085866689682007 2023-01-23 08:41:54.914859: step: 2066/1541, loss: 0.2688819169998169 2023-01-23 08:41:55.805874: step: 2068/1541, loss: 0.24375221133232117 2023-01-23 08:41:56.737436: step: 2070/1541, loss: 2.1689400672912598 2023-01-23 08:41:57.636666: step: 2072/1541, loss: 0.61872398853302 2023-01-23 08:41:58.524201: step: 2074/1541, loss: 0.27283164858818054 2023-01-23 08:41:59.390104: step: 2076/1541, loss: 0.20400191843509674 2023-01-23 08:42:00.353658: step: 2078/1541, loss: 0.0975201427936554 2023-01-23 08:42:01.241001: step: 2080/1541, loss: 0.07524192333221436 2023-01-23 08:42:02.208279: step: 2082/1541, loss: 0.649480402469635 2023-01-23 08:42:03.088262: step: 2084/1541, loss: 0.6558796763420105 2023-01-23 08:42:04.043818: step: 2086/1541, loss: 0.28322094678878784 2023-01-23 08:42:04.932967: step: 2088/1541, loss: 0.04825965315103531 2023-01-23 08:42:05.826683: step: 2090/1541, loss: 0.35484209656715393 2023-01-23 08:42:06.717364: step: 2092/1541, loss: 0.33023199439048767 2023-01-23 08:42:07.591290: step: 2094/1541, loss: 0.17157693207263947 2023-01-23 08:42:08.459681: step: 2096/1541, loss: 0.5462715029716492 2023-01-23 08:42:09.312976: step: 2098/1541, loss: 0.09145782142877579 2023-01-23 08:42:10.232935: step: 2100/1541, loss: 0.6375144124031067 2023-01-23 08:42:11.203372: step: 2102/1541, loss: 0.3471088409423828 2023-01-23 08:42:12.077053: step: 2104/1541, loss: 0.3640731871128082 2023-01-23 08:42:12.977284: step: 2106/1541, loss: 0.20674046874046326 2023-01-23 08:42:13.894986: step: 2108/1541, loss: 0.47436150908470154 2023-01-23 08:42:14.753484: step: 2110/1541, loss: 0.07288578152656555 2023-01-23 08:42:15.685289: step: 2112/1541, loss: 0.08506342023611069 2023-01-23 08:42:16.604269: step: 2114/1541, loss: 0.2755962312221527 2023-01-23 08:42:17.468703: step: 2116/1541, loss: 0.15878069400787354 2023-01-23 08:42:18.417716: step: 2118/1541, loss: 0.5393216609954834 2023-01-23 08:42:19.318890: step: 2120/1541, loss: 2.431621789932251 2023-01-23 08:42:20.282580: step: 2122/1541, loss: 1.0077521800994873 2023-01-23 08:42:21.318701: step: 2124/1541, loss: 0.11929725855588913 2023-01-23 08:42:22.236321: step: 2126/1541, loss: 0.19274598360061646 2023-01-23 08:42:23.146109: step: 2128/1541, loss: 0.46566036343574524 2023-01-23 08:42:24.039453: step: 2130/1541, loss: 0.23189280927181244 2023-01-23 08:42:24.978525: step: 2132/1541, loss: 1.2097551822662354 2023-01-23 08:42:25.876272: step: 2134/1541, loss: 0.22102104127407074 2023-01-23 08:42:26.746609: step: 2136/1541, loss: 0.1781720519065857 2023-01-23 08:42:27.632866: step: 2138/1541, loss: 0.2788478434085846 2023-01-23 08:42:28.502943: step: 2140/1541, loss: 0.10973258316516876 2023-01-23 08:42:29.442499: step: 2142/1541, loss: 0.13068686425685883 2023-01-23 08:42:30.315680: step: 2144/1541, loss: 0.22836558520793915 2023-01-23 08:42:31.171540: step: 2146/1541, loss: 0.23776298761367798 2023-01-23 08:42:32.086465: step: 2148/1541, loss: 0.0871429592370987 2023-01-23 08:42:32.947151: step: 2150/1541, loss: 0.6404004693031311 2023-01-23 08:42:33.837227: step: 2152/1541, loss: 1.9553484916687012 2023-01-23 08:42:34.724012: step: 2154/1541, loss: 0.44649314880371094 2023-01-23 08:42:35.662389: step: 2156/1541, loss: 0.7455390691757202 2023-01-23 08:42:36.583514: step: 2158/1541, loss: 3.1563639640808105 2023-01-23 08:42:37.572609: step: 2160/1541, loss: 0.7671181559562683 2023-01-23 08:42:38.422132: step: 2162/1541, loss: 0.1477026492357254 2023-01-23 08:42:39.364153: step: 2164/1541, loss: 0.7822679877281189 2023-01-23 08:42:40.243653: step: 2166/1541, loss: 0.45982587337493896 2023-01-23 08:42:41.138300: step: 2168/1541, loss: 1.4704889059066772 2023-01-23 08:42:42.038237: step: 2170/1541, loss: 0.4957258999347687 2023-01-23 08:42:42.941869: step: 2172/1541, loss: 0.27133244276046753 2023-01-23 08:42:43.870725: step: 2174/1541, loss: 0.09684135019779205 2023-01-23 08:42:44.761036: step: 2176/1541, loss: 0.26608389616012573 2023-01-23 08:42:45.644875: step: 2178/1541, loss: 0.8111787438392639 2023-01-23 08:42:46.600089: step: 2180/1541, loss: 0.4811553359031677 2023-01-23 08:42:47.537524: step: 2182/1541, loss: 0.45045992732048035 2023-01-23 08:42:48.408637: step: 2184/1541, loss: 0.1316881626844406 2023-01-23 08:42:49.316136: step: 2186/1541, loss: 0.20468896627426147 2023-01-23 08:42:50.179625: step: 2188/1541, loss: 0.6926272511482239 2023-01-23 08:42:51.115950: step: 2190/1541, loss: 0.30313634872436523 2023-01-23 08:42:52.064815: step: 2192/1541, loss: 0.17268404364585876 2023-01-23 08:42:52.946951: step: 2194/1541, loss: 0.46872973442077637 2023-01-23 08:42:53.817665: step: 2196/1541, loss: 1.295058012008667 2023-01-23 08:42:54.768012: step: 2198/1541, loss: 0.7914263010025024 2023-01-23 08:42:55.637184: step: 2200/1541, loss: 0.21398721635341644 2023-01-23 08:42:56.584164: step: 2202/1541, loss: 0.4847568869590759 2023-01-23 08:42:57.500488: step: 2204/1541, loss: 0.4403842091560364 2023-01-23 08:42:58.431118: step: 2206/1541, loss: 0.2987613379955292 2023-01-23 08:42:59.359207: step: 2208/1541, loss: 0.23694536089897156 2023-01-23 08:43:00.246157: step: 2210/1541, loss: 0.18121486902236938 2023-01-23 08:43:01.141160: step: 2212/1541, loss: 0.5672374963760376 2023-01-23 08:43:02.024012: step: 2214/1541, loss: 1.3083536624908447 2023-01-23 08:43:02.907583: step: 2216/1541, loss: 0.16892744600772858 2023-01-23 08:43:03.809094: step: 2218/1541, loss: 7.112881660461426 2023-01-23 08:43:04.764658: step: 2220/1541, loss: 0.09177269786596298 2023-01-23 08:43:05.679449: step: 2222/1541, loss: 0.7267053127288818 2023-01-23 08:43:06.573574: step: 2224/1541, loss: 0.43907269835472107 2023-01-23 08:43:07.445630: step: 2226/1541, loss: 0.3741980195045471 2023-01-23 08:43:08.306438: step: 2228/1541, loss: 0.30413979291915894 2023-01-23 08:43:09.190603: step: 2230/1541, loss: 0.2637301981449127 2023-01-23 08:43:10.104500: step: 2232/1541, loss: 0.25711703300476074 2023-01-23 08:43:10.990400: step: 2234/1541, loss: 0.13774393498897552 2023-01-23 08:43:11.862470: step: 2236/1541, loss: 0.22333279252052307 2023-01-23 08:43:12.784492: step: 2238/1541, loss: 1.1398361921310425 2023-01-23 08:43:13.663950: step: 2240/1541, loss: 2.1705586910247803 2023-01-23 08:43:14.518963: step: 2242/1541, loss: 0.7086838483810425 2023-01-23 08:43:15.396392: step: 2244/1541, loss: 0.17775027453899384 2023-01-23 08:43:16.270941: step: 2246/1541, loss: 0.4577482044696808 2023-01-23 08:43:17.186987: step: 2248/1541, loss: 0.14972837269306183 2023-01-23 08:43:18.124479: step: 2250/1541, loss: 1.4836394786834717 2023-01-23 08:43:19.014113: step: 2252/1541, loss: 0.29988110065460205 2023-01-23 08:43:19.947942: step: 2254/1541, loss: 0.17109106481075287 2023-01-23 08:43:21.001484: step: 2256/1541, loss: 0.40885481238365173 2023-01-23 08:43:21.878836: step: 2258/1541, loss: 0.1717814803123474 2023-01-23 08:43:22.780850: step: 2260/1541, loss: 0.06545763462781906 2023-01-23 08:43:23.703386: step: 2262/1541, loss: 0.2564588785171509 2023-01-23 08:43:24.644714: step: 2264/1541, loss: 0.3786143660545349 2023-01-23 08:43:25.518439: step: 2266/1541, loss: 0.09707748144865036 2023-01-23 08:43:26.407161: step: 2268/1541, loss: 0.9002615213394165 2023-01-23 08:43:27.294928: step: 2270/1541, loss: 0.48918986320495605 2023-01-23 08:43:28.224616: step: 2272/1541, loss: 0.2832319438457489 2023-01-23 08:43:29.201892: step: 2274/1541, loss: 0.44146135449409485 2023-01-23 08:43:30.116800: step: 2276/1541, loss: 0.2600003182888031 2023-01-23 08:43:31.061645: step: 2278/1541, loss: 1.590244174003601 2023-01-23 08:43:31.968512: step: 2280/1541, loss: 0.6839343905448914 2023-01-23 08:43:32.870539: step: 2282/1541, loss: 0.5285193920135498 2023-01-23 08:43:33.750176: step: 2284/1541, loss: 0.38344061374664307 2023-01-23 08:43:34.642968: step: 2286/1541, loss: 0.19469647109508514 2023-01-23 08:43:35.522880: step: 2288/1541, loss: 0.43375706672668457 2023-01-23 08:43:36.408106: step: 2290/1541, loss: 0.09420255571603775 2023-01-23 08:43:37.286923: step: 2292/1541, loss: 0.4373776912689209 2023-01-23 08:43:38.236557: step: 2294/1541, loss: 0.5805098414421082 2023-01-23 08:43:39.156820: step: 2296/1541, loss: 0.48910510540008545 2023-01-23 08:43:40.029228: step: 2298/1541, loss: 0.8200961351394653 2023-01-23 08:43:40.920123: step: 2300/1541, loss: 0.5247272849082947 2023-01-23 08:43:41.891764: step: 2302/1541, loss: 0.3650725781917572 2023-01-23 08:43:42.784382: step: 2304/1541, loss: 0.9533417820930481 2023-01-23 08:43:43.638683: step: 2306/1541, loss: 0.6031295657157898 2023-01-23 08:43:44.513825: step: 2308/1541, loss: 0.195201113820076 2023-01-23 08:43:45.430668: step: 2310/1541, loss: 0.4959094524383545 2023-01-23 08:43:46.307450: step: 2312/1541, loss: 0.34700876474380493 2023-01-23 08:43:47.187730: step: 2314/1541, loss: 0.35324519872665405 2023-01-23 08:43:48.104169: step: 2316/1541, loss: 0.3771945536136627 2023-01-23 08:43:49.026407: step: 2318/1541, loss: 0.46257463097572327 2023-01-23 08:43:49.886764: step: 2320/1541, loss: 0.4242256283760071 2023-01-23 08:43:50.785370: step: 2322/1541, loss: 0.36749720573425293 2023-01-23 08:43:51.672142: step: 2324/1541, loss: 0.5312297344207764 2023-01-23 08:43:52.627922: step: 2326/1541, loss: 0.9114644527435303 2023-01-23 08:43:53.576747: step: 2328/1541, loss: 0.2154645025730133 2023-01-23 08:43:54.493909: step: 2330/1541, loss: 0.30247342586517334 2023-01-23 08:43:55.433203: step: 2332/1541, loss: 6.998466491699219 2023-01-23 08:43:56.325641: step: 2334/1541, loss: 0.5159568190574646 2023-01-23 08:43:57.184059: step: 2336/1541, loss: 0.908668041229248 2023-01-23 08:43:58.082999: step: 2338/1541, loss: 0.4787433445453644 2023-01-23 08:43:58.988241: step: 2340/1541, loss: 0.38345280289649963 2023-01-23 08:43:59.883343: step: 2342/1541, loss: 0.13771216571331024 2023-01-23 08:44:00.771259: step: 2344/1541, loss: 0.15504616498947144 2023-01-23 08:44:01.684961: step: 2346/1541, loss: 0.2937473654747009 2023-01-23 08:44:02.655502: step: 2348/1541, loss: 0.7096284627914429 2023-01-23 08:44:03.597750: step: 2350/1541, loss: 0.32115891575813293 2023-01-23 08:44:04.539303: step: 2352/1541, loss: 0.4481918513774872 2023-01-23 08:44:05.456664: step: 2354/1541, loss: 0.12879736721515656 2023-01-23 08:44:06.343654: step: 2356/1541, loss: 0.15287061035633087 2023-01-23 08:44:07.261455: step: 2358/1541, loss: 0.6187535524368286 2023-01-23 08:44:08.147380: step: 2360/1541, loss: 0.3934434652328491 2023-01-23 08:44:08.971102: step: 2362/1541, loss: 0.09378928691148758 2023-01-23 08:44:09.913393: step: 2364/1541, loss: 0.7108938694000244 2023-01-23 08:44:10.821440: step: 2366/1541, loss: 0.8377087712287903 2023-01-23 08:44:11.720557: step: 2368/1541, loss: 0.4195309579372406 2023-01-23 08:44:12.579016: step: 2370/1541, loss: 0.4814087748527527 2023-01-23 08:44:13.490786: step: 2372/1541, loss: 0.5487820506095886 2023-01-23 08:44:14.357219: step: 2374/1541, loss: 3.509823799133301 2023-01-23 08:44:15.285844: step: 2376/1541, loss: 1.4785007238388062 2023-01-23 08:44:16.164193: step: 2378/1541, loss: 0.8106158375740051 2023-01-23 08:44:17.067374: step: 2380/1541, loss: 0.445271372795105 2023-01-23 08:44:17.966318: step: 2382/1541, loss: 2.987663984298706 2023-01-23 08:44:18.864554: step: 2384/1541, loss: 0.8138993382453918 2023-01-23 08:44:19.800821: step: 2386/1541, loss: 0.5907180309295654 2023-01-23 08:44:20.711978: step: 2388/1541, loss: 0.05744149535894394 2023-01-23 08:44:21.601370: step: 2390/1541, loss: 0.1395740956068039 2023-01-23 08:44:22.473950: step: 2392/1541, loss: 0.36702898144721985 2023-01-23 08:44:23.349252: step: 2394/1541, loss: 0.6484761238098145 2023-01-23 08:44:24.249177: step: 2396/1541, loss: 0.2076699584722519 2023-01-23 08:44:25.220627: step: 2398/1541, loss: 0.7828734517097473 2023-01-23 08:44:26.126382: step: 2400/1541, loss: 0.22474981844425201 2023-01-23 08:44:27.112499: step: 2402/1541, loss: 0.3589324951171875 2023-01-23 08:44:28.016540: step: 2404/1541, loss: 2.0633058547973633 2023-01-23 08:44:28.920534: step: 2406/1541, loss: 0.45256105065345764 2023-01-23 08:44:29.817741: step: 2408/1541, loss: 0.1863170713186264 2023-01-23 08:44:30.705528: step: 2410/1541, loss: 0.15425992012023926 2023-01-23 08:44:31.601677: step: 2412/1541, loss: 0.2558766007423401 2023-01-23 08:44:32.557611: step: 2414/1541, loss: 0.32351765036582947 2023-01-23 08:44:33.454725: step: 2416/1541, loss: 0.1271723210811615 2023-01-23 08:44:34.358149: step: 2418/1541, loss: 0.27193647623062134 2023-01-23 08:44:35.274916: step: 2420/1541, loss: 0.18641388416290283 2023-01-23 08:44:36.186057: step: 2422/1541, loss: 0.42715245485305786 2023-01-23 08:44:37.075274: step: 2424/1541, loss: 0.1728404462337494 2023-01-23 08:44:37.996368: step: 2426/1541, loss: 0.348870187997818 2023-01-23 08:44:38.883859: step: 2428/1541, loss: 0.40825581550598145 2023-01-23 08:44:39.764272: step: 2430/1541, loss: 0.287274032831192 2023-01-23 08:44:40.722018: step: 2432/1541, loss: 0.10970395803451538 2023-01-23 08:44:41.617774: step: 2434/1541, loss: 0.019741004332900047 2023-01-23 08:44:42.546127: step: 2436/1541, loss: 0.4310489296913147 2023-01-23 08:44:43.429573: step: 2438/1541, loss: 0.4531339406967163 2023-01-23 08:44:44.357703: step: 2440/1541, loss: 0.38500407338142395 2023-01-23 08:44:45.332546: step: 2442/1541, loss: 0.34484097361564636 2023-01-23 08:44:46.191891: step: 2444/1541, loss: 0.5558748245239258 2023-01-23 08:44:47.120523: step: 2446/1541, loss: 0.4133989214897156 2023-01-23 08:44:48.057289: step: 2448/1541, loss: 0.2932298481464386 2023-01-23 08:44:48.952884: step: 2450/1541, loss: 0.11933828890323639 2023-01-23 08:44:49.842680: step: 2452/1541, loss: 4.2662529945373535 2023-01-23 08:44:50.756636: step: 2454/1541, loss: 0.2503391206264496 2023-01-23 08:44:51.640533: step: 2456/1541, loss: 0.0475214384496212 2023-01-23 08:44:52.540151: step: 2458/1541, loss: 0.3252633213996887 2023-01-23 08:44:53.427712: step: 2460/1541, loss: 0.642236053943634 2023-01-23 08:44:54.326319: step: 2462/1541, loss: 0.41269993782043457 2023-01-23 08:44:55.241580: step: 2464/1541, loss: 0.4353215992450714 2023-01-23 08:44:56.173543: step: 2466/1541, loss: 0.3979428708553314 2023-01-23 08:44:57.129250: step: 2468/1541, loss: 0.3020991086959839 2023-01-23 08:44:57.973657: step: 2470/1541, loss: 0.09131471067667007 2023-01-23 08:44:58.877744: step: 2472/1541, loss: 0.32506221532821655 2023-01-23 08:44:59.750111: step: 2474/1541, loss: 0.3278367221355438 2023-01-23 08:45:00.642310: step: 2476/1541, loss: 0.22739863395690918 2023-01-23 08:45:01.542564: step: 2478/1541, loss: 0.17320485413074493 2023-01-23 08:45:02.418604: step: 2480/1541, loss: 0.1453019678592682 2023-01-23 08:45:03.309457: step: 2482/1541, loss: 0.33958134055137634 2023-01-23 08:45:04.218571: step: 2484/1541, loss: 0.19369065761566162 2023-01-23 08:45:05.145334: step: 2486/1541, loss: 0.3993438482284546 2023-01-23 08:45:06.016811: step: 2488/1541, loss: 0.35700881481170654 2023-01-23 08:45:06.881858: step: 2490/1541, loss: 0.536776065826416 2023-01-23 08:45:07.836548: step: 2492/1541, loss: 1.2465729713439941 2023-01-23 08:45:08.805158: step: 2494/1541, loss: 0.057297833263874054 2023-01-23 08:45:09.681493: step: 2496/1541, loss: 0.09657078236341476 2023-01-23 08:45:10.570006: step: 2498/1541, loss: 0.8617656826972961 2023-01-23 08:45:11.448685: step: 2500/1541, loss: 0.3254818320274353 2023-01-23 08:45:12.391319: step: 2502/1541, loss: 0.14461784064769745 2023-01-23 08:45:13.256753: step: 2504/1541, loss: 0.27458450198173523 2023-01-23 08:45:14.121015: step: 2506/1541, loss: 0.522658109664917 2023-01-23 08:45:15.045736: step: 2508/1541, loss: 0.3959309458732605 2023-01-23 08:45:15.918972: step: 2510/1541, loss: 0.8341986536979675 2023-01-23 08:45:16.818955: step: 2512/1541, loss: 0.36911624670028687 2023-01-23 08:45:17.694210: step: 2514/1541, loss: 0.23020891845226288 2023-01-23 08:45:18.620574: step: 2516/1541, loss: 0.14656758308410645 2023-01-23 08:45:19.479369: step: 2518/1541, loss: 0.32615891098976135 2023-01-23 08:45:20.353684: step: 2520/1541, loss: 0.20968355238437653 2023-01-23 08:45:21.270525: step: 2522/1541, loss: 0.04204924777150154 2023-01-23 08:45:22.217716: step: 2524/1541, loss: 0.16659902036190033 2023-01-23 08:45:23.129646: step: 2526/1541, loss: 0.5736224055290222 2023-01-23 08:45:24.023545: step: 2528/1541, loss: 0.11113576591014862 2023-01-23 08:45:24.943564: step: 2530/1541, loss: 0.20678605139255524 2023-01-23 08:45:25.848015: step: 2532/1541, loss: 0.5210680961608887 2023-01-23 08:45:26.773640: step: 2534/1541, loss: 0.22691072523593903 2023-01-23 08:45:27.622285: step: 2536/1541, loss: 3.914980173110962 2023-01-23 08:45:28.534650: step: 2538/1541, loss: 0.2831028401851654 2023-01-23 08:45:29.417978: step: 2540/1541, loss: 0.447029709815979 2023-01-23 08:45:30.275422: step: 2542/1541, loss: 0.30014118552207947 2023-01-23 08:45:31.136138: step: 2544/1541, loss: 0.6399809122085571 2023-01-23 08:45:31.974011: step: 2546/1541, loss: 0.2573177218437195 2023-01-23 08:45:32.852384: step: 2548/1541, loss: 0.27049222588539124 2023-01-23 08:45:33.746711: step: 2550/1541, loss: 0.15321049094200134 2023-01-23 08:45:34.629786: step: 2552/1541, loss: 0.3094770610332489 2023-01-23 08:45:35.492936: step: 2554/1541, loss: 0.2217947095632553 2023-01-23 08:45:36.393097: step: 2556/1541, loss: 0.2588229179382324 2023-01-23 08:45:37.326138: step: 2558/1541, loss: 0.27166661620140076 2023-01-23 08:45:38.219414: step: 2560/1541, loss: 0.47720953822135925 2023-01-23 08:45:39.094303: step: 2562/1541, loss: 0.8794222474098206 2023-01-23 08:45:39.992858: step: 2564/1541, loss: 0.11906193941831589 2023-01-23 08:45:40.937262: step: 2566/1541, loss: 0.27549949288368225 2023-01-23 08:45:41.829381: step: 2568/1541, loss: 0.7940478324890137 2023-01-23 08:45:42.679807: step: 2570/1541, loss: 0.27774566411972046 2023-01-23 08:45:43.593301: step: 2572/1541, loss: 1.016179084777832 2023-01-23 08:45:44.488353: step: 2574/1541, loss: 0.5272226333618164 2023-01-23 08:45:45.362037: step: 2576/1541, loss: 0.14738354086875916 2023-01-23 08:45:46.334582: step: 2578/1541, loss: 2.1615686416625977 2023-01-23 08:45:47.228864: step: 2580/1541, loss: 0.41334405541419983 2023-01-23 08:45:48.175221: step: 2582/1541, loss: 0.1776949167251587 2023-01-23 08:45:49.041116: step: 2584/1541, loss: 0.7775675654411316 2023-01-23 08:45:49.981472: step: 2586/1541, loss: 0.2635655701160431 2023-01-23 08:45:50.871490: step: 2588/1541, loss: 0.5401977896690369 2023-01-23 08:45:51.798687: step: 2590/1541, loss: 0.2968725562095642 2023-01-23 08:45:52.705307: step: 2592/1541, loss: 0.38869091868400574 2023-01-23 08:45:53.603515: step: 2594/1541, loss: 0.6044213771820068 2023-01-23 08:45:54.495723: step: 2596/1541, loss: 0.3553915321826935 2023-01-23 08:45:55.402506: step: 2598/1541, loss: 0.850558340549469 2023-01-23 08:45:56.320596: step: 2600/1541, loss: 0.30987977981567383 2023-01-23 08:45:57.214156: step: 2602/1541, loss: 0.37742123007774353 2023-01-23 08:45:58.151401: step: 2604/1541, loss: 0.3864361643791199 2023-01-23 08:45:59.065944: step: 2606/1541, loss: 0.5637118220329285 2023-01-23 08:45:59.915216: step: 2608/1541, loss: 0.15855011343955994 2023-01-23 08:46:00.828582: step: 2610/1541, loss: 0.3676816523075104 2023-01-23 08:46:01.751082: step: 2612/1541, loss: 0.5146892070770264 2023-01-23 08:46:02.613816: step: 2614/1541, loss: 0.10158680379390717 2023-01-23 08:46:03.529642: step: 2616/1541, loss: 0.26921868324279785 2023-01-23 08:46:04.414171: step: 2618/1541, loss: 1.1246919631958008 2023-01-23 08:46:05.299332: step: 2620/1541, loss: 0.27318060398101807 2023-01-23 08:46:06.188658: step: 2622/1541, loss: 0.28630533814430237 2023-01-23 08:46:07.111588: step: 2624/1541, loss: 0.7034803032875061 2023-01-23 08:46:07.987710: step: 2626/1541, loss: 0.2217896580696106 2023-01-23 08:46:08.877964: step: 2628/1541, loss: 0.6922611594200134 2023-01-23 08:46:09.805293: step: 2630/1541, loss: 1.015041470527649 2023-01-23 08:46:10.701613: step: 2632/1541, loss: 0.7084100842475891 2023-01-23 08:46:11.585962: step: 2634/1541, loss: 0.14237162470817566 2023-01-23 08:46:12.466820: step: 2636/1541, loss: 0.5667481422424316 2023-01-23 08:46:13.329340: step: 2638/1541, loss: 0.21578213572502136 2023-01-23 08:46:14.245504: step: 2640/1541, loss: 0.6642374396324158 2023-01-23 08:46:15.122848: step: 2642/1541, loss: 0.23515425622463226 2023-01-23 08:46:16.070045: step: 2644/1541, loss: 0.25279566645622253 2023-01-23 08:46:16.933570: step: 2646/1541, loss: 0.41577214002609253 2023-01-23 08:46:17.884731: step: 2648/1541, loss: 0.4370065927505493 2023-01-23 08:46:18.806154: step: 2650/1541, loss: 1.0290411710739136 2023-01-23 08:46:19.694386: step: 2652/1541, loss: 0.2927994430065155 2023-01-23 08:46:20.607274: step: 2654/1541, loss: 0.21931298077106476 2023-01-23 08:46:21.588739: step: 2656/1541, loss: 0.5623241066932678 2023-01-23 08:46:22.504073: step: 2658/1541, loss: 0.4138188362121582 2023-01-23 08:46:23.432474: step: 2660/1541, loss: 0.7298594117164612 2023-01-23 08:46:24.324014: step: 2662/1541, loss: 0.43434497714042664 2023-01-23 08:46:25.222104: step: 2664/1541, loss: 0.44901177287101746 2023-01-23 08:46:26.122899: step: 2666/1541, loss: 0.14821451902389526 2023-01-23 08:46:27.008945: step: 2668/1541, loss: 0.2259255051612854 2023-01-23 08:46:27.951065: step: 2670/1541, loss: 0.49473217129707336 2023-01-23 08:46:28.794703: step: 2672/1541, loss: 0.08306942880153656 2023-01-23 08:46:29.679671: step: 2674/1541, loss: 0.29192227125167847 2023-01-23 08:46:30.668643: step: 2676/1541, loss: 0.27128851413726807 2023-01-23 08:46:31.554861: step: 2678/1541, loss: 0.6122020483016968 2023-01-23 08:46:32.442877: step: 2680/1541, loss: 0.40871208906173706 2023-01-23 08:46:33.336662: step: 2682/1541, loss: 0.540895402431488 2023-01-23 08:46:34.212090: step: 2684/1541, loss: 0.8980310559272766 2023-01-23 08:46:35.145700: step: 2686/1541, loss: 0.5789536833763123 2023-01-23 08:46:36.044509: step: 2688/1541, loss: 0.3259032666683197 2023-01-23 08:46:37.041446: step: 2690/1541, loss: 0.09351566433906555 2023-01-23 08:46:37.960925: step: 2692/1541, loss: 0.09663309156894684 2023-01-23 08:46:38.825124: step: 2694/1541, loss: 0.4355930685997009 2023-01-23 08:46:39.748090: step: 2696/1541, loss: 0.2470693141222 2023-01-23 08:46:40.656601: step: 2698/1541, loss: 1.5100090503692627 2023-01-23 08:46:41.574829: step: 2700/1541, loss: 0.275439590215683 2023-01-23 08:46:42.419863: step: 2702/1541, loss: 0.41235435009002686 2023-01-23 08:46:43.320220: step: 2704/1541, loss: 0.3157893419265747 2023-01-23 08:46:44.193377: step: 2706/1541, loss: 1.1320383548736572 2023-01-23 08:46:45.105368: step: 2708/1541, loss: 0.1922101378440857 2023-01-23 08:46:45.991596: step: 2710/1541, loss: 0.15797816216945648 2023-01-23 08:46:46.868771: step: 2712/1541, loss: 1.271747350692749 2023-01-23 08:46:47.793047: step: 2714/1541, loss: 0.163770854473114 2023-01-23 08:46:48.691779: step: 2716/1541, loss: 1.0467568635940552 2023-01-23 08:46:49.574618: step: 2718/1541, loss: 0.0501798577606678 2023-01-23 08:46:50.451431: step: 2720/1541, loss: 0.8538199663162231 2023-01-23 08:46:51.362706: step: 2722/1541, loss: 1.4250762462615967 2023-01-23 08:46:52.247842: step: 2724/1541, loss: 0.4781339466571808 2023-01-23 08:46:53.150997: step: 2726/1541, loss: 0.07999666035175323 2023-01-23 08:46:54.026532: step: 2728/1541, loss: 0.2770761251449585 2023-01-23 08:46:54.922059: step: 2730/1541, loss: 0.42548421025276184 2023-01-23 08:46:55.869656: step: 2732/1541, loss: 0.7069023847579956 2023-01-23 08:46:56.781546: step: 2734/1541, loss: 0.3152773380279541 2023-01-23 08:46:57.684911: step: 2736/1541, loss: 0.26251643896102905 2023-01-23 08:46:58.568092: step: 2738/1541, loss: 0.4306727945804596 2023-01-23 08:46:59.468993: step: 2740/1541, loss: 0.22170878946781158 2023-01-23 08:47:00.372797: step: 2742/1541, loss: 0.16103531420230865 2023-01-23 08:47:01.261376: step: 2744/1541, loss: 0.38389822840690613 2023-01-23 08:47:02.179003: step: 2746/1541, loss: 0.21200604736804962 2023-01-23 08:47:03.078601: step: 2748/1541, loss: 0.28855499625205994 2023-01-23 08:47:03.950721: step: 2750/1541, loss: 0.24875664710998535 2023-01-23 08:47:04.825174: step: 2752/1541, loss: 0.3533601462841034 2023-01-23 08:47:05.689373: step: 2754/1541, loss: 0.2902499735355377 2023-01-23 08:47:06.591950: step: 2756/1541, loss: 0.5265021920204163 2023-01-23 08:47:07.504515: step: 2758/1541, loss: 0.2891033887863159 2023-01-23 08:47:08.411245: step: 2760/1541, loss: 0.18731889128684998 2023-01-23 08:47:09.301665: step: 2762/1541, loss: 0.32807213068008423 2023-01-23 08:47:10.263287: step: 2764/1541, loss: 0.3242992162704468 2023-01-23 08:47:11.133903: step: 2766/1541, loss: 0.4800828695297241 2023-01-23 08:47:12.065756: step: 2768/1541, loss: 0.5091508626937866 2023-01-23 08:47:12.989755: step: 2770/1541, loss: 0.6080043911933899 2023-01-23 08:47:13.917570: step: 2772/1541, loss: 1.0060502290725708 2023-01-23 08:47:14.800971: step: 2774/1541, loss: 0.2717355191707611 2023-01-23 08:47:15.693171: step: 2776/1541, loss: 0.1505589336156845 2023-01-23 08:47:16.613363: step: 2778/1541, loss: 0.28090548515319824 2023-01-23 08:47:17.543085: step: 2780/1541, loss: 0.11628617346286774 2023-01-23 08:47:18.457044: step: 2782/1541, loss: 0.08841924369335175 2023-01-23 08:47:19.314791: step: 2784/1541, loss: 0.31778475642204285 2023-01-23 08:47:20.192437: step: 2786/1541, loss: 0.2654852271080017 2023-01-23 08:47:21.066010: step: 2788/1541, loss: 0.3232026994228363 2023-01-23 08:47:21.965607: step: 2790/1541, loss: 0.9756520390510559 2023-01-23 08:47:22.883161: step: 2792/1541, loss: 0.6181508302688599 2023-01-23 08:47:23.791887: step: 2794/1541, loss: 0.6176692247390747 2023-01-23 08:47:24.704284: step: 2796/1541, loss: 0.19222073256969452 2023-01-23 08:47:25.618501: step: 2798/1541, loss: 0.2993812561035156 2023-01-23 08:47:26.508158: step: 2800/1541, loss: 0.3309221565723419 2023-01-23 08:47:27.555967: step: 2802/1541, loss: 3.0514824390411377 2023-01-23 08:47:28.426455: step: 2804/1541, loss: 0.19470177590847015 2023-01-23 08:47:29.318544: step: 2806/1541, loss: 0.35636478662490845 2023-01-23 08:47:30.233396: step: 2808/1541, loss: 0.5871233940124512 2023-01-23 08:47:31.183452: step: 2810/1541, loss: 0.16174176335334778 2023-01-23 08:47:32.070103: step: 2812/1541, loss: 0.05421967804431915 2023-01-23 08:47:32.977035: step: 2814/1541, loss: 0.31481364369392395 2023-01-23 08:47:33.879363: step: 2816/1541, loss: 0.14762625098228455 2023-01-23 08:47:34.762653: step: 2818/1541, loss: 0.5151282548904419 2023-01-23 08:47:35.659368: step: 2820/1541, loss: 2.3756625652313232 2023-01-23 08:47:36.568408: step: 2822/1541, loss: 0.8063677549362183 2023-01-23 08:47:37.507128: step: 2824/1541, loss: 0.27282020449638367 2023-01-23 08:47:38.435430: step: 2826/1541, loss: 0.4668744206428528 2023-01-23 08:47:39.350880: step: 2828/1541, loss: 0.14044354856014252 2023-01-23 08:47:40.269720: step: 2830/1541, loss: 0.506291389465332 2023-01-23 08:47:41.144874: step: 2832/1541, loss: 1.2949838638305664 2023-01-23 08:47:42.048452: step: 2834/1541, loss: 0.40165483951568604 2023-01-23 08:47:42.895173: step: 2836/1541, loss: 0.4821920096874237 2023-01-23 08:47:43.786220: step: 2838/1541, loss: 0.6983499526977539 2023-01-23 08:47:44.679162: step: 2840/1541, loss: 0.48420169949531555 2023-01-23 08:47:45.606267: step: 2842/1541, loss: 0.16095636785030365 2023-01-23 08:47:46.498945: step: 2844/1541, loss: 0.8426284790039062 2023-01-23 08:47:47.497710: step: 2846/1541, loss: 0.2926345467567444 2023-01-23 08:47:48.386744: step: 2848/1541, loss: 2.5824341773986816 2023-01-23 08:47:49.300359: step: 2850/1541, loss: 0.6838439702987671 2023-01-23 08:47:50.182825: step: 2852/1541, loss: 0.2895483374595642 2023-01-23 08:47:51.081845: step: 2854/1541, loss: 0.7401331663131714 2023-01-23 08:47:51.952506: step: 2856/1541, loss: 0.3385723829269409 2023-01-23 08:47:52.804442: step: 2858/1541, loss: 0.5105559229850769 2023-01-23 08:47:53.681852: step: 2860/1541, loss: 0.5701456069946289 2023-01-23 08:47:54.546110: step: 2862/1541, loss: 0.20775505900382996 2023-01-23 08:47:55.523042: step: 2864/1541, loss: 0.48817625641822815 2023-01-23 08:47:56.416306: step: 2866/1541, loss: 0.4055022895336151 2023-01-23 08:47:57.325846: step: 2868/1541, loss: 0.41734015941619873 2023-01-23 08:47:58.214230: step: 2870/1541, loss: 0.2126019150018692 2023-01-23 08:47:59.080150: step: 2872/1541, loss: 0.1076497957110405 2023-01-23 08:47:59.968230: step: 2874/1541, loss: 0.12766830623149872 2023-01-23 08:48:00.859087: step: 2876/1541, loss: 0.32231420278549194 2023-01-23 08:48:01.756077: step: 2878/1541, loss: 0.36491072177886963 2023-01-23 08:48:02.667919: step: 2880/1541, loss: 0.3494873344898224 2023-01-23 08:48:03.538504: step: 2882/1541, loss: 2.648960828781128 2023-01-23 08:48:04.411245: step: 2884/1541, loss: 0.22576072812080383 2023-01-23 08:48:05.311350: step: 2886/1541, loss: 0.8264789581298828 2023-01-23 08:48:06.175551: step: 2888/1541, loss: 0.20193257927894592 2023-01-23 08:48:07.052909: step: 2890/1541, loss: 0.3488505184650421 2023-01-23 08:48:07.941285: step: 2892/1541, loss: 0.35895901918411255 2023-01-23 08:48:08.842760: step: 2894/1541, loss: 0.21958452463150024 2023-01-23 08:48:09.735945: step: 2896/1541, loss: 0.11155732721090317 2023-01-23 08:48:10.629427: step: 2898/1541, loss: 0.46389085054397583 2023-01-23 08:48:11.613821: step: 2900/1541, loss: 0.1141156479716301 2023-01-23 08:48:12.481647: step: 2902/1541, loss: 0.27213147282600403 2023-01-23 08:48:13.370548: step: 2904/1541, loss: 0.181068554520607 2023-01-23 08:48:14.296905: step: 2906/1541, loss: 0.23251523077487946 2023-01-23 08:48:15.212282: step: 2908/1541, loss: 0.37934011220932007 2023-01-23 08:48:16.109217: step: 2910/1541, loss: 0.1854988932609558 2023-01-23 08:48:17.066701: step: 2912/1541, loss: 0.5580707788467407 2023-01-23 08:48:18.010793: step: 2914/1541, loss: 6.529485702514648 2023-01-23 08:48:18.904818: step: 2916/1541, loss: 0.08076023310422897 2023-01-23 08:48:19.804759: step: 2918/1541, loss: 0.12509246170520782 2023-01-23 08:48:20.716818: step: 2920/1541, loss: 0.18337488174438477 2023-01-23 08:48:21.568506: step: 2922/1541, loss: 0.35074669122695923 2023-01-23 08:48:22.468114: step: 2924/1541, loss: 0.9174584150314331 2023-01-23 08:48:23.383967: step: 2926/1541, loss: 0.25028306245803833 2023-01-23 08:48:24.275229: step: 2928/1541, loss: 0.3745024502277374 2023-01-23 08:48:25.140299: step: 2930/1541, loss: 0.5159785747528076 2023-01-23 08:48:26.017390: step: 2932/1541, loss: 0.3782102167606354 2023-01-23 08:48:26.947106: step: 2934/1541, loss: 0.2698003053665161 2023-01-23 08:48:27.876608: step: 2936/1541, loss: 0.4574465751647949 2023-01-23 08:48:28.752671: step: 2938/1541, loss: 0.7038713097572327 2023-01-23 08:48:29.641616: step: 2940/1541, loss: 0.24750155210494995 2023-01-23 08:48:30.528845: step: 2942/1541, loss: 0.610650897026062 2023-01-23 08:48:31.388802: step: 2944/1541, loss: 0.4263595640659332 2023-01-23 08:48:32.353960: step: 2946/1541, loss: 0.43445104360580444 2023-01-23 08:48:33.240117: step: 2948/1541, loss: 0.23274995386600494 2023-01-23 08:48:34.164667: step: 2950/1541, loss: 0.20349109172821045 2023-01-23 08:48:35.078067: step: 2952/1541, loss: 0.24008063971996307 2023-01-23 08:48:36.050681: step: 2954/1541, loss: 0.12781324982643127 2023-01-23 08:48:36.945459: step: 2956/1541, loss: 0.4657706618309021 2023-01-23 08:48:37.848522: step: 2958/1541, loss: 6.297757625579834 2023-01-23 08:48:38.762821: step: 2960/1541, loss: 0.5114642977714539 2023-01-23 08:48:39.645367: step: 2962/1541, loss: 0.14329704642295837 2023-01-23 08:48:40.558110: step: 2964/1541, loss: 1.8480336666107178 2023-01-23 08:48:41.542086: step: 2966/1541, loss: 0.7811606526374817 2023-01-23 08:48:42.426548: step: 2968/1541, loss: 0.15707898139953613 2023-01-23 08:48:43.336945: step: 2970/1541, loss: 0.6814414262771606 2023-01-23 08:48:44.191349: step: 2972/1541, loss: 0.10705244541168213 2023-01-23 08:48:45.064170: step: 2974/1541, loss: 0.0754353255033493 2023-01-23 08:48:45.968786: step: 2976/1541, loss: 0.1127442866563797 2023-01-23 08:48:46.877950: step: 2978/1541, loss: 0.20250388979911804 2023-01-23 08:48:47.762524: step: 2980/1541, loss: 0.11417242139577866 2023-01-23 08:48:48.662870: step: 2982/1541, loss: 0.4153313934803009 2023-01-23 08:48:49.564120: step: 2984/1541, loss: 0.3637051284313202 2023-01-23 08:48:50.488371: step: 2986/1541, loss: 0.32101964950561523 2023-01-23 08:48:51.353076: step: 2988/1541, loss: 0.728821337223053 2023-01-23 08:48:52.252580: step: 2990/1541, loss: 0.3661695718765259 2023-01-23 08:48:53.196664: step: 2992/1541, loss: 1.8911995887756348 2023-01-23 08:48:54.101350: step: 2994/1541, loss: 0.3543197810649872 2023-01-23 08:48:55.055889: step: 2996/1541, loss: 0.10071176290512085 2023-01-23 08:48:55.926442: step: 2998/1541, loss: 0.4615675210952759 2023-01-23 08:48:56.814540: step: 3000/1541, loss: 0.44055312871932983 2023-01-23 08:48:57.711567: step: 3002/1541, loss: 0.3080785870552063 2023-01-23 08:48:58.579358: step: 3004/1541, loss: 0.37081682682037354 2023-01-23 08:48:59.489221: step: 3006/1541, loss: 0.22164539992809296 2023-01-23 08:49:00.372032: step: 3008/1541, loss: 0.6316171884536743 2023-01-23 08:49:01.268991: step: 3010/1541, loss: 0.39418235421180725 2023-01-23 08:49:02.220805: step: 3012/1541, loss: 0.4380565285682678 2023-01-23 08:49:03.127024: step: 3014/1541, loss: 0.28880590200424194 2023-01-23 08:49:04.009680: step: 3016/1541, loss: 0.5694687366485596 2023-01-23 08:49:04.903251: step: 3018/1541, loss: 2.8570146560668945 2023-01-23 08:49:05.745000: step: 3020/1541, loss: 0.15646231174468994 2023-01-23 08:49:06.604810: step: 3022/1541, loss: 0.3992617130279541 2023-01-23 08:49:07.508966: step: 3024/1541, loss: 0.13979263603687286 2023-01-23 08:49:08.420239: step: 3026/1541, loss: 0.34074491262435913 2023-01-23 08:49:09.333173: step: 3028/1541, loss: 0.4539744555950165 2023-01-23 08:49:10.210512: step: 3030/1541, loss: 0.45043835043907166 2023-01-23 08:49:11.082392: step: 3032/1541, loss: 0.805532693862915 2023-01-23 08:49:12.039313: step: 3034/1541, loss: 0.36475273966789246 2023-01-23 08:49:12.939503: step: 3036/1541, loss: 0.08202151209115982 2023-01-23 08:49:13.829591: step: 3038/1541, loss: 0.48514431715011597 2023-01-23 08:49:14.776791: step: 3040/1541, loss: 0.3627646565437317 2023-01-23 08:49:15.706788: step: 3042/1541, loss: 0.5026212334632874 2023-01-23 08:49:16.703835: step: 3044/1541, loss: 0.40191197395324707 2023-01-23 08:49:17.584253: step: 3046/1541, loss: 0.26678621768951416 2023-01-23 08:49:18.520677: step: 3048/1541, loss: 0.5204678177833557 2023-01-23 08:49:19.429072: step: 3050/1541, loss: 0.5666854381561279 2023-01-23 08:49:20.328212: step: 3052/1541, loss: 0.44629761576652527 2023-01-23 08:49:21.215150: step: 3054/1541, loss: 0.21641314029693604 2023-01-23 08:49:22.087169: step: 3056/1541, loss: 0.22346769273281097 2023-01-23 08:49:22.955613: step: 3058/1541, loss: 0.26228857040405273 2023-01-23 08:49:23.859032: step: 3060/1541, loss: 5.788942813873291 2023-01-23 08:49:24.737422: step: 3062/1541, loss: 0.8568452000617981 2023-01-23 08:49:25.598628: step: 3064/1541, loss: 0.06983695924282074 2023-01-23 08:49:26.498676: step: 3066/1541, loss: 0.39691534638404846 2023-01-23 08:49:27.388197: step: 3068/1541, loss: 0.3640490770339966 2023-01-23 08:49:28.256868: step: 3070/1541, loss: 0.18515080213546753 2023-01-23 08:49:29.147405: step: 3072/1541, loss: 0.4327467679977417 2023-01-23 08:49:30.068416: step: 3074/1541, loss: 0.19945192337036133 2023-01-23 08:49:31.080411: step: 3076/1541, loss: 0.6251460909843445 2023-01-23 08:49:31.967806: step: 3078/1541, loss: 0.16112688183784485 2023-01-23 08:49:32.869087: step: 3080/1541, loss: 0.37015676498413086 2023-01-23 08:49:33.742500: step: 3082/1541, loss: 0.548466682434082 2023-01-23 08:49:34.371461: step: 3084/1541, loss: 0.30947452783584595 ================================================== Loss: 0.484 -------------------- Dev Chinese: {'event': {'p': 0.6296296296296297, 'r': 0.7017310252996005, 'f1': 0.663727959697733}, 'argument': {'p': 0.6672884936953917, 'r': 0.6595293251640499, 'f1': 0.6633862218024361}, 'combined': 0.4403079834885187, 'epoch': 7} Test Chinese: {'event': {'p': 0.6088328075709779, 'r': 0.6938286399041342, 'f1': 0.6485578269392327}, 'argument': {'p': 0.7017525808926439, 'r': 0.6928081743626939, 'f1': 0.6972516938977305}, 'combined': 0.4522080434240111, 'epoch': 7} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.658943151930715, 'r': 0.6660438324472097, 'f1': 0.6624744657138378}, 'combined': 0.43616331283022863, 'epoch': 7} Test Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7013697662599169, 'r': 0.7122207753019127, 'f1': 0.7067536235026656}, 'combined': 0.4795493128942961, 'epoch': 7} Dev Russian: {'event': {'p': 0.6528436018957346, 'r': 0.7336884154460719, 'f1': 0.6909090909090909}, 'argument': {'p': 0.6689314588101222, 'r': 0.6699381276721389, 'f1': 0.66943441479419}, 'combined': 0.4625183229487131, 'epoch': 7} Test Russian: {'event': {'p': 0.6457352171637886, 'r': 0.7393648891551827, 'f1': 0.6893854748603351}, 'argument': {'p': 0.7036239331831553, 'r': 0.7031191814376867, 'f1': 0.7033714667557484}, 'combined': 0.48489407261262196, 'epoch': 7} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6462391774891775, 'r': 0.6405202467149369, 'f1': 0.6433670033670033}, 'combined': 0.4619045152378485, 'epoch': 7} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.4995136943666355, 'r': 0.3396693121693122, 'f1': 0.40436822877299067}, 'combined': 0.24198413690352194, 'epoch': 7} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5776661508989095, 'r': 0.5373638613013112, 'f1': 0.5567866514688284}, 'combined': 0.28634742075539754, 'epoch': 7} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test for Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} -------------------- Dev for Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6749027227273954, 'r': 0.6759176140397674, 'f1': 0.6754097871321493}, 'combined': 0.467493670842566, 'epoch': 6} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7154362539653685, 'r': 0.7041128024637584, 'f1': 0.7097293657901553}, 'combined': 0.49443157493872836, 'epoch': 6} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5630713489409142, 'r': 0.6023553965414431, 'f1': 0.5820512820512821}, 'combined': 0.2993406593406594, 'epoch': 6} ****************************** Epoch: 8 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 08:54:57.777047: step: 2/1541, loss: 0.7092674970626831 2023-01-23 08:54:58.736105: step: 4/1541, loss: 0.8989720940589905 2023-01-23 08:54:59.674156: step: 6/1541, loss: 0.38708606362342834 2023-01-23 08:55:00.601058: step: 8/1541, loss: 1.0592925548553467 2023-01-23 08:55:01.508641: step: 10/1541, loss: 0.6187146902084351 2023-01-23 08:55:02.392272: step: 12/1541, loss: 0.5760719776153564 2023-01-23 08:55:03.272507: step: 14/1541, loss: 0.08038970082998276 2023-01-23 08:55:04.235133: step: 16/1541, loss: 0.3100259304046631 2023-01-23 08:55:05.192610: step: 18/1541, loss: 0.2853802442550659 2023-01-23 08:55:06.079883: step: 20/1541, loss: 0.34769922494888306 2023-01-23 08:55:07.058843: step: 22/1541, loss: 0.4201951324939728 2023-01-23 08:55:07.986904: step: 24/1541, loss: 0.26736530661582947 2023-01-23 08:55:08.839986: step: 26/1541, loss: 0.3183107376098633 2023-01-23 08:55:09.714826: step: 28/1541, loss: 0.157777339220047 2023-01-23 08:55:10.727867: step: 30/1541, loss: 0.17534734308719635 2023-01-23 08:55:11.653646: step: 32/1541, loss: 0.37610870599746704 2023-01-23 08:55:12.509325: step: 34/1541, loss: 0.49154403805732727 2023-01-23 08:55:13.393806: step: 36/1541, loss: 0.5306994915008545 2023-01-23 08:55:14.264575: step: 38/1541, loss: 0.31633856892585754 2023-01-23 08:55:15.151905: step: 40/1541, loss: 0.049112606793642044 2023-01-23 08:55:16.122222: step: 42/1541, loss: 0.2843547761440277 2023-01-23 08:55:17.030974: step: 44/1541, loss: 0.1026160791516304 2023-01-23 08:55:17.931899: step: 46/1541, loss: 0.04339149594306946 2023-01-23 08:55:18.822113: step: 48/1541, loss: 0.07928504794836044 2023-01-23 08:55:19.672523: step: 50/1541, loss: 0.25001800060272217 2023-01-23 08:55:20.558418: step: 52/1541, loss: 0.14484870433807373 2023-01-23 08:55:21.490294: step: 54/1541, loss: 0.08899285644292831 2023-01-23 08:55:22.352602: step: 56/1541, loss: 0.0953286811709404 2023-01-23 08:55:23.208204: step: 58/1541, loss: 0.32407665252685547 2023-01-23 08:55:24.098515: step: 60/1541, loss: 0.4188701808452606 2023-01-23 08:55:25.006807: step: 62/1541, loss: 0.38136976957321167 2023-01-23 08:55:25.938456: step: 64/1541, loss: 0.1255464106798172 2023-01-23 08:55:26.932355: step: 66/1541, loss: 0.240033358335495 2023-01-23 08:55:27.848938: step: 68/1541, loss: 0.07123233377933502 2023-01-23 08:55:28.748768: step: 70/1541, loss: 0.1821332573890686 2023-01-23 08:55:29.647181: step: 72/1541, loss: 0.1078438013792038 2023-01-23 08:55:30.579792: step: 74/1541, loss: 0.8596268892288208 2023-01-23 08:55:31.453601: step: 76/1541, loss: 0.455793559551239 2023-01-23 08:55:32.391212: step: 78/1541, loss: 0.11640194803476334 2023-01-23 08:55:33.284659: step: 80/1541, loss: 0.09530476480722427 2023-01-23 08:55:34.204931: step: 82/1541, loss: 0.09931390732526779 2023-01-23 08:55:35.100624: step: 84/1541, loss: 0.18135854601860046 2023-01-23 08:55:35.964912: step: 86/1541, loss: 0.4101509153842926 2023-01-23 08:55:36.866930: step: 88/1541, loss: 0.08508282154798508 2023-01-23 08:55:37.824776: step: 90/1541, loss: 0.5067468881607056 2023-01-23 08:55:38.724412: step: 92/1541, loss: 0.878486692905426 2023-01-23 08:55:39.615996: step: 94/1541, loss: 0.2273377925157547 2023-01-23 08:55:40.528431: step: 96/1541, loss: 0.08918411284685135 2023-01-23 08:55:41.429670: step: 98/1541, loss: 0.3156835436820984 2023-01-23 08:55:42.297820: step: 100/1541, loss: 0.2290678173303604 2023-01-23 08:55:43.198503: step: 102/1541, loss: 0.26319336891174316 2023-01-23 08:55:44.077293: step: 104/1541, loss: 0.1367424726486206 2023-01-23 08:55:44.969869: step: 106/1541, loss: 5.353947162628174 2023-01-23 08:55:45.900438: step: 108/1541, loss: 0.19372783601284027 2023-01-23 08:55:46.854990: step: 110/1541, loss: 0.23189949989318848 2023-01-23 08:55:47.723077: step: 112/1541, loss: 0.06389416754245758 2023-01-23 08:55:48.592899: step: 114/1541, loss: 0.02550284005701542 2023-01-23 08:55:49.496719: step: 116/1541, loss: 0.2925412654876709 2023-01-23 08:55:50.422708: step: 118/1541, loss: 0.36789241433143616 2023-01-23 08:55:51.297505: step: 120/1541, loss: 0.5154916048049927 2023-01-23 08:55:52.166972: step: 122/1541, loss: 0.6093747615814209 2023-01-23 08:55:53.099863: step: 124/1541, loss: 0.35260146856307983 2023-01-23 08:55:53.990761: step: 126/1541, loss: 0.085757315158844 2023-01-23 08:55:54.889107: step: 128/1541, loss: 0.2661469876766205 2023-01-23 08:55:55.857475: step: 130/1541, loss: 0.3302815854549408 2023-01-23 08:55:56.743038: step: 132/1541, loss: 0.20152193307876587 2023-01-23 08:55:57.595173: step: 134/1541, loss: 0.18849922716617584 2023-01-23 08:55:58.476681: step: 136/1541, loss: 0.09884797781705856 2023-01-23 08:55:59.365886: step: 138/1541, loss: 0.2806706726551056 2023-01-23 08:56:00.243868: step: 140/1541, loss: 0.014226775616407394 2023-01-23 08:56:01.175068: step: 142/1541, loss: 0.40434762835502625 2023-01-23 08:56:02.150749: step: 144/1541, loss: 0.31449735164642334 2023-01-23 08:56:03.079084: step: 146/1541, loss: 0.048693783581256866 2023-01-23 08:56:03.948988: step: 148/1541, loss: 0.2551950216293335 2023-01-23 08:56:04.877501: step: 150/1541, loss: 0.2282411903142929 2023-01-23 08:56:05.785256: step: 152/1541, loss: 0.15666227042675018 2023-01-23 08:56:06.719455: step: 154/1541, loss: 2.232457399368286 2023-01-23 08:56:07.583601: step: 156/1541, loss: 0.31320199370384216 2023-01-23 08:56:08.480988: step: 158/1541, loss: 0.052275147289037704 2023-01-23 08:56:09.360443: step: 160/1541, loss: 0.12383747100830078 2023-01-23 08:56:10.238221: step: 162/1541, loss: 0.20275279879570007 2023-01-23 08:56:11.090420: step: 164/1541, loss: 0.044961102306842804 2023-01-23 08:56:12.006670: step: 166/1541, loss: 0.11048642545938492 2023-01-23 08:56:12.890784: step: 168/1541, loss: 2.7381553649902344 2023-01-23 08:56:13.777539: step: 170/1541, loss: 0.4771283268928528 2023-01-23 08:56:14.658018: step: 172/1541, loss: 6.1607160568237305 2023-01-23 08:56:15.581396: step: 174/1541, loss: 0.18761813640594482 2023-01-23 08:56:16.548209: step: 176/1541, loss: 0.1718984693288803 2023-01-23 08:56:17.438502: step: 178/1541, loss: 0.13167248666286469 2023-01-23 08:56:18.373433: step: 180/1541, loss: 0.20299032330513 2023-01-23 08:56:19.301298: step: 182/1541, loss: 1.4784116744995117 2023-01-23 08:56:20.191854: step: 184/1541, loss: 0.06670089066028595 2023-01-23 08:56:21.084349: step: 186/1541, loss: 0.15766103565692902 2023-01-23 08:56:21.967146: step: 188/1541, loss: 0.5126956701278687 2023-01-23 08:56:22.841774: step: 190/1541, loss: 0.04687628895044327 2023-01-23 08:56:23.727983: step: 192/1541, loss: 0.03833435848355293 2023-01-23 08:56:24.632228: step: 194/1541, loss: 0.23557037115097046 2023-01-23 08:56:25.621074: step: 196/1541, loss: 0.23864218592643738 2023-01-23 08:56:26.531503: step: 198/1541, loss: 0.13233543932437897 2023-01-23 08:56:27.473777: step: 200/1541, loss: 0.09751345217227936 2023-01-23 08:56:28.467410: step: 202/1541, loss: 1.3517712354660034 2023-01-23 08:56:29.414226: step: 204/1541, loss: 0.24223341047763824 2023-01-23 08:56:30.333622: step: 206/1541, loss: 0.35545867681503296 2023-01-23 08:56:31.265681: step: 208/1541, loss: 0.11671481281518936 2023-01-23 08:56:32.194537: step: 210/1541, loss: 0.37659910321235657 2023-01-23 08:56:33.161222: step: 212/1541, loss: 0.3385368883609772 2023-01-23 08:56:34.045734: step: 214/1541, loss: 0.2778233289718628 2023-01-23 08:56:34.955913: step: 216/1541, loss: 0.28336918354034424 2023-01-23 08:56:35.843803: step: 218/1541, loss: 0.2107166051864624 2023-01-23 08:56:36.738781: step: 220/1541, loss: 0.23825520277023315 2023-01-23 08:56:37.625032: step: 222/1541, loss: 0.0549149289727211 2023-01-23 08:56:38.523580: step: 224/1541, loss: 5.615252494812012 2023-01-23 08:56:39.429795: step: 226/1541, loss: 0.3404209017753601 2023-01-23 08:56:40.373106: step: 228/1541, loss: 0.3749641180038452 2023-01-23 08:56:41.299018: step: 230/1541, loss: 0.08973979949951172 2023-01-23 08:56:42.163414: step: 232/1541, loss: 0.02360268123447895 2023-01-23 08:56:43.041201: step: 234/1541, loss: 0.15761211514472961 2023-01-23 08:56:43.913569: step: 236/1541, loss: 0.6572771668434143 2023-01-23 08:56:44.780178: step: 238/1541, loss: 0.16515272855758667 2023-01-23 08:56:45.680895: step: 240/1541, loss: 0.46333903074264526 2023-01-23 08:56:46.586244: step: 242/1541, loss: 0.044936370104551315 2023-01-23 08:56:47.493881: step: 244/1541, loss: 0.06705182045698166 2023-01-23 08:56:48.425004: step: 246/1541, loss: 0.15811660885810852 2023-01-23 08:56:49.302108: step: 248/1541, loss: 0.25634077191352844 2023-01-23 08:56:50.178748: step: 250/1541, loss: 0.7750280499458313 2023-01-23 08:56:51.065825: step: 252/1541, loss: 0.19902460277080536 2023-01-23 08:56:51.957000: step: 254/1541, loss: 0.13752858340740204 2023-01-23 08:56:52.823822: step: 256/1541, loss: 0.46626564860343933 2023-01-23 08:56:53.738022: step: 258/1541, loss: 0.5727975368499756 2023-01-23 08:56:54.626447: step: 260/1541, loss: 0.03630806505680084 2023-01-23 08:56:55.540377: step: 262/1541, loss: 0.10763192176818848 2023-01-23 08:56:56.431216: step: 264/1541, loss: 0.28117865324020386 2023-01-23 08:56:57.352778: step: 266/1541, loss: 0.23738402128219604 2023-01-23 08:56:58.214968: step: 268/1541, loss: 0.1855504810810089 2023-01-23 08:56:59.155036: step: 270/1541, loss: 0.12931253015995026 2023-01-23 08:57:00.115823: step: 272/1541, loss: 0.16479912400245667 2023-01-23 08:57:01.007813: step: 274/1541, loss: 0.28705647587776184 2023-01-23 08:57:01.942864: step: 276/1541, loss: 0.6646198630332947 2023-01-23 08:57:02.831402: step: 278/1541, loss: 0.18054261803627014 2023-01-23 08:57:03.706329: step: 280/1541, loss: 0.1924961805343628 2023-01-23 08:57:04.607075: step: 282/1541, loss: 1.5481404066085815 2023-01-23 08:57:05.510849: step: 284/1541, loss: 1.1320140361785889 2023-01-23 08:57:06.412200: step: 286/1541, loss: 0.13247694075107574 2023-01-23 08:57:07.288368: step: 288/1541, loss: 0.049541134387254715 2023-01-23 08:57:08.203740: step: 290/1541, loss: 0.30350974202156067 2023-01-23 08:57:09.092719: step: 292/1541, loss: 0.18737363815307617 2023-01-23 08:57:09.998308: step: 294/1541, loss: 0.2972704768180847 2023-01-23 08:57:10.917054: step: 296/1541, loss: 0.17920398712158203 2023-01-23 08:57:11.803548: step: 298/1541, loss: 0.1229490414261818 2023-01-23 08:57:12.689313: step: 300/1541, loss: 0.13637027144432068 2023-01-23 08:57:13.553349: step: 302/1541, loss: 0.17220300436019897 2023-01-23 08:57:14.446984: step: 304/1541, loss: 0.1762782335281372 2023-01-23 08:57:15.325403: step: 306/1541, loss: 4.114156723022461 2023-01-23 08:57:16.210544: step: 308/1541, loss: 0.29962676763534546 2023-01-23 08:57:17.109310: step: 310/1541, loss: 0.1584121286869049 2023-01-23 08:57:18.004844: step: 312/1541, loss: 0.6425752639770508 2023-01-23 08:57:18.881133: step: 314/1541, loss: 0.04406866803765297 2023-01-23 08:57:19.770466: step: 316/1541, loss: 0.19815613329410553 2023-01-23 08:57:20.679806: step: 318/1541, loss: 0.17441494762897491 2023-01-23 08:57:21.556867: step: 320/1541, loss: 0.2848162353038788 2023-01-23 08:57:22.475724: step: 322/1541, loss: 0.3242846131324768 2023-01-23 08:57:23.366940: step: 324/1541, loss: 0.5200057029724121 2023-01-23 08:57:24.276998: step: 326/1541, loss: 0.08059714734554291 2023-01-23 08:57:25.137389: step: 328/1541, loss: 0.48898816108703613 2023-01-23 08:57:26.054794: step: 330/1541, loss: 2.5878937244415283 2023-01-23 08:57:26.970679: step: 332/1541, loss: 0.20846489071846008 2023-01-23 08:57:27.899250: step: 334/1541, loss: 0.17729796469211578 2023-01-23 08:57:28.793990: step: 336/1541, loss: 0.112240269780159 2023-01-23 08:57:29.691952: step: 338/1541, loss: 0.37833142280578613 2023-01-23 08:57:30.544722: step: 340/1541, loss: 0.11195075511932373 2023-01-23 08:57:31.453440: step: 342/1541, loss: 0.5010582804679871 2023-01-23 08:57:32.378163: step: 344/1541, loss: 0.6016409993171692 2023-01-23 08:57:33.267829: step: 346/1541, loss: 0.11339890956878662 2023-01-23 08:57:34.266066: step: 348/1541, loss: 0.27418074011802673 2023-01-23 08:57:35.239457: step: 350/1541, loss: 0.16279388964176178 2023-01-23 08:57:36.144425: step: 352/1541, loss: 0.3997231423854828 2023-01-23 08:57:37.038644: step: 354/1541, loss: 0.14592915773391724 2023-01-23 08:57:37.955886: step: 356/1541, loss: 0.2681649923324585 2023-01-23 08:57:38.940215: step: 358/1541, loss: 0.13302797079086304 2023-01-23 08:57:39.851889: step: 360/1541, loss: 0.5911234021186829 2023-01-23 08:57:40.766655: step: 362/1541, loss: 0.10504402965307236 2023-01-23 08:57:41.636195: step: 364/1541, loss: 0.14873850345611572 2023-01-23 08:57:42.528839: step: 366/1541, loss: 0.26978400349617004 2023-01-23 08:57:43.402516: step: 368/1541, loss: 0.5241570472717285 2023-01-23 08:57:44.278454: step: 370/1541, loss: 0.15118063986301422 2023-01-23 08:57:45.179351: step: 372/1541, loss: 0.502505362033844 2023-01-23 08:57:46.067177: step: 374/1541, loss: 0.31317293643951416 2023-01-23 08:57:46.959837: step: 376/1541, loss: 0.27866068482398987 2023-01-23 08:57:47.853047: step: 378/1541, loss: 0.27184563875198364 2023-01-23 08:57:48.740928: step: 380/1541, loss: 0.10243668407201767 2023-01-23 08:57:49.624023: step: 382/1541, loss: 0.24722279608249664 2023-01-23 08:57:50.483583: step: 384/1541, loss: 0.6838850378990173 2023-01-23 08:57:51.395089: step: 386/1541, loss: 0.08703355491161346 2023-01-23 08:57:52.310519: step: 388/1541, loss: 0.13622938096523285 2023-01-23 08:57:53.178200: step: 390/1541, loss: 0.23301544785499573 2023-01-23 08:57:54.106576: step: 392/1541, loss: 0.4438494145870209 2023-01-23 08:57:55.022524: step: 394/1541, loss: 0.01725609228014946 2023-01-23 08:57:55.904746: step: 396/1541, loss: 0.7871732711791992 2023-01-23 08:57:56.833556: step: 398/1541, loss: 0.24501067399978638 2023-01-23 08:57:57.759345: step: 400/1541, loss: 0.4253551661968231 2023-01-23 08:57:58.700204: step: 402/1541, loss: 0.10430236905813217 2023-01-23 08:57:59.598343: step: 404/1541, loss: 0.6730362772941589 2023-01-23 08:58:00.497380: step: 406/1541, loss: 0.5114786624908447 2023-01-23 08:58:01.402416: step: 408/1541, loss: 0.5121461749076843 2023-01-23 08:58:02.339178: step: 410/1541, loss: 0.20432120561599731 2023-01-23 08:58:03.236441: step: 412/1541, loss: 0.2469378113746643 2023-01-23 08:58:04.203752: step: 414/1541, loss: 0.2799513041973114 2023-01-23 08:58:05.076690: step: 416/1541, loss: 0.5066489577293396 2023-01-23 08:58:05.981277: step: 418/1541, loss: 0.35418739914894104 2023-01-23 08:58:06.877631: step: 420/1541, loss: 0.24089951813220978 2023-01-23 08:58:07.783580: step: 422/1541, loss: 0.16809476912021637 2023-01-23 08:58:08.682889: step: 424/1541, loss: 0.22231924533843994 2023-01-23 08:58:09.569423: step: 426/1541, loss: 0.32364189624786377 2023-01-23 08:58:10.470760: step: 428/1541, loss: 0.23848989605903625 2023-01-23 08:58:11.545795: step: 430/1541, loss: 0.14477884769439697 2023-01-23 08:58:12.422606: step: 432/1541, loss: 0.16194921731948853 2023-01-23 08:58:13.301440: step: 434/1541, loss: 0.20634527504444122 2023-01-23 08:58:14.206125: step: 436/1541, loss: 0.11853683739900589 2023-01-23 08:58:15.095104: step: 438/1541, loss: 0.336585134267807 2023-01-23 08:58:15.997909: step: 440/1541, loss: 0.21320147812366486 2023-01-23 08:58:16.892543: step: 442/1541, loss: 0.14536575973033905 2023-01-23 08:58:17.732913: step: 444/1541, loss: 0.06089679151773453 2023-01-23 08:58:18.658051: step: 446/1541, loss: 0.20457032322883606 2023-01-23 08:58:19.531488: step: 448/1541, loss: 0.3148142099380493 2023-01-23 08:58:20.415331: step: 450/1541, loss: 0.062558114528656 2023-01-23 08:58:21.302019: step: 452/1541, loss: 0.14517183601856232 2023-01-23 08:58:22.249677: step: 454/1541, loss: 0.7344353199005127 2023-01-23 08:58:23.185533: step: 456/1541, loss: 0.17343170940876007 2023-01-23 08:58:24.058112: step: 458/1541, loss: 0.31388533115386963 2023-01-23 08:58:24.993407: step: 460/1541, loss: 0.24720583856105804 2023-01-23 08:58:25.919497: step: 462/1541, loss: 0.6868103742599487 2023-01-23 08:58:26.770824: step: 464/1541, loss: 0.23598308861255646 2023-01-23 08:58:27.634669: step: 466/1541, loss: 0.10375763475894928 2023-01-23 08:58:28.541686: step: 468/1541, loss: 0.25479480624198914 2023-01-23 08:58:29.443628: step: 470/1541, loss: 0.15068168938159943 2023-01-23 08:58:30.337661: step: 472/1541, loss: 0.5590219497680664 2023-01-23 08:58:31.219482: step: 474/1541, loss: 0.9867967367172241 2023-01-23 08:58:32.201682: step: 476/1541, loss: 0.20740726590156555 2023-01-23 08:58:33.100198: step: 478/1541, loss: 0.12433119863271713 2023-01-23 08:58:33.989925: step: 480/1541, loss: 0.2942602336406708 2023-01-23 08:58:34.892852: step: 482/1541, loss: 0.39902493357658386 2023-01-23 08:58:35.770745: step: 484/1541, loss: 0.10264185070991516 2023-01-23 08:58:36.669713: step: 486/1541, loss: 0.28230515122413635 2023-01-23 08:58:37.554519: step: 488/1541, loss: 0.32199278473854065 2023-01-23 08:58:38.457474: step: 490/1541, loss: 0.23195865750312805 2023-01-23 08:58:39.339730: step: 492/1541, loss: 0.2723443806171417 2023-01-23 08:58:40.229328: step: 494/1541, loss: 0.28980758786201477 2023-01-23 08:58:41.112027: step: 496/1541, loss: 0.041938841342926025 2023-01-23 08:58:42.000972: step: 498/1541, loss: 0.3795255422592163 2023-01-23 08:58:42.847013: step: 500/1541, loss: 0.2255391776561737 2023-01-23 08:58:43.783136: step: 502/1541, loss: 0.1279078871011734 2023-01-23 08:58:44.646792: step: 504/1541, loss: 0.34781375527381897 2023-01-23 08:58:45.528991: step: 506/1541, loss: 0.028361212462186813 2023-01-23 08:58:46.397662: step: 508/1541, loss: 0.11886262148618698 2023-01-23 08:58:47.356511: step: 510/1541, loss: 0.027150390669703484 2023-01-23 08:58:48.255175: step: 512/1541, loss: 0.32966887950897217 2023-01-23 08:58:49.186098: step: 514/1541, loss: 0.06847831606864929 2023-01-23 08:58:50.047182: step: 516/1541, loss: 0.08668810874223709 2023-01-23 08:58:50.949736: step: 518/1541, loss: 0.2365998923778534 2023-01-23 08:58:51.839825: step: 520/1541, loss: 0.09254540503025055 2023-01-23 08:58:52.733940: step: 522/1541, loss: 0.4489545226097107 2023-01-23 08:58:53.599504: step: 524/1541, loss: 0.3731232285499573 2023-01-23 08:58:54.481063: step: 526/1541, loss: 0.05709047615528107 2023-01-23 08:58:55.305412: step: 528/1541, loss: 0.13722339272499084 2023-01-23 08:58:56.182754: step: 530/1541, loss: 0.1603531539440155 2023-01-23 08:58:57.053579: step: 532/1541, loss: 0.2117634117603302 2023-01-23 08:58:57.928453: step: 534/1541, loss: 0.935532808303833 2023-01-23 08:58:58.805476: step: 536/1541, loss: 0.7019410133361816 2023-01-23 08:58:59.674377: step: 538/1541, loss: 0.2789885997772217 2023-01-23 08:59:00.562177: step: 540/1541, loss: 0.17654861509799957 2023-01-23 08:59:01.443277: step: 542/1541, loss: 0.27958112955093384 2023-01-23 08:59:02.323530: step: 544/1541, loss: 0.3898446261882782 2023-01-23 08:59:03.205791: step: 546/1541, loss: 0.13723713159561157 2023-01-23 08:59:04.106849: step: 548/1541, loss: 0.26803386211395264 2023-01-23 08:59:04.989320: step: 550/1541, loss: 0.14282524585723877 2023-01-23 08:59:05.865651: step: 552/1541, loss: 0.13901060819625854 2023-01-23 08:59:06.764248: step: 554/1541, loss: 0.4717334508895874 2023-01-23 08:59:07.636741: step: 556/1541, loss: 0.2339152991771698 2023-01-23 08:59:08.541445: step: 558/1541, loss: 0.1418565958738327 2023-01-23 08:59:09.433655: step: 560/1541, loss: 0.10317815095186234 2023-01-23 08:59:10.331214: step: 562/1541, loss: 0.22424864768981934 2023-01-23 08:59:11.279329: step: 564/1541, loss: 0.17089471220970154 2023-01-23 08:59:12.188970: step: 566/1541, loss: 0.6673729419708252 2023-01-23 08:59:13.043349: step: 568/1541, loss: 0.267341285943985 2023-01-23 08:59:13.983134: step: 570/1541, loss: 0.2849971354007721 2023-01-23 08:59:14.859162: step: 572/1541, loss: 0.2280297577381134 2023-01-23 08:59:15.751873: step: 574/1541, loss: 0.1883089393377304 2023-01-23 08:59:16.644524: step: 576/1541, loss: 0.12055852264165878 2023-01-23 08:59:17.590548: step: 578/1541, loss: 0.1734188199043274 2023-01-23 08:59:18.494854: step: 580/1541, loss: 0.15690867602825165 2023-01-23 08:59:19.385879: step: 582/1541, loss: 0.4131505489349365 2023-01-23 08:59:20.269398: step: 584/1541, loss: 0.535994827747345 2023-01-23 08:59:21.138700: step: 586/1541, loss: 0.105524942278862 2023-01-23 08:59:22.001828: step: 588/1541, loss: 0.1927644908428192 2023-01-23 08:59:22.883812: step: 590/1541, loss: 0.35470348596572876 2023-01-23 08:59:23.777513: step: 592/1541, loss: 0.2165147364139557 2023-01-23 08:59:24.726680: step: 594/1541, loss: 0.23340237140655518 2023-01-23 08:59:25.631929: step: 596/1541, loss: 0.06422393769025803 2023-01-23 08:59:26.496939: step: 598/1541, loss: 0.24087177217006683 2023-01-23 08:59:27.442671: step: 600/1541, loss: 0.11723830550909042 2023-01-23 08:59:28.323078: step: 602/1541, loss: 0.13340166211128235 2023-01-23 08:59:29.199318: step: 604/1541, loss: 0.2219773232936859 2023-01-23 08:59:30.121029: step: 606/1541, loss: 0.25711590051651 2023-01-23 08:59:31.024574: step: 608/1541, loss: 0.1427733600139618 2023-01-23 08:59:31.930130: step: 610/1541, loss: 0.3120322823524475 2023-01-23 08:59:32.857430: step: 612/1541, loss: 3.1628706455230713 2023-01-23 08:59:33.727481: step: 614/1541, loss: 0.573024570941925 2023-01-23 08:59:34.609267: step: 616/1541, loss: 0.1640225201845169 2023-01-23 08:59:35.474301: step: 618/1541, loss: 0.06583130359649658 2023-01-23 08:59:36.449442: step: 620/1541, loss: 0.3837348520755768 2023-01-23 08:59:37.399328: step: 622/1541, loss: 0.3417341411113739 2023-01-23 08:59:38.266060: step: 624/1541, loss: 0.331610769033432 2023-01-23 08:59:39.163286: step: 626/1541, loss: 0.8629190921783447 2023-01-23 08:59:40.076521: step: 628/1541, loss: 0.26418471336364746 2023-01-23 08:59:40.932794: step: 630/1541, loss: 0.35485130548477173 2023-01-23 08:59:41.851445: step: 632/1541, loss: 0.13848480582237244 2023-01-23 08:59:42.745344: step: 634/1541, loss: 0.3678683042526245 2023-01-23 08:59:43.637681: step: 636/1541, loss: 1.2961747646331787 2023-01-23 08:59:44.526051: step: 638/1541, loss: 0.4509340822696686 2023-01-23 08:59:45.416355: step: 640/1541, loss: 0.25069382786750793 2023-01-23 08:59:46.357544: step: 642/1541, loss: 0.07764411717653275 2023-01-23 08:59:47.218376: step: 644/1541, loss: 0.33933988213539124 2023-01-23 08:59:48.133670: step: 646/1541, loss: 0.4113904535770416 2023-01-23 08:59:49.046294: step: 648/1541, loss: 0.3510264754295349 2023-01-23 08:59:49.984884: step: 650/1541, loss: 0.5162007212638855 2023-01-23 08:59:50.877755: step: 652/1541, loss: 0.08358964323997498 2023-01-23 08:59:51.730856: step: 654/1541, loss: 0.267295241355896 2023-01-23 08:59:52.628326: step: 656/1541, loss: 0.32688966393470764 2023-01-23 08:59:53.507669: step: 658/1541, loss: 0.07727637887001038 2023-01-23 08:59:54.370792: step: 660/1541, loss: 0.14434053003787994 2023-01-23 08:59:55.292950: step: 662/1541, loss: 0.2054741084575653 2023-01-23 08:59:56.170353: step: 664/1541, loss: 0.41536760330200195 2023-01-23 08:59:57.060712: step: 666/1541, loss: 0.28894492983818054 2023-01-23 08:59:57.937018: step: 668/1541, loss: 1.4805420637130737 2023-01-23 08:59:58.854696: step: 670/1541, loss: 0.7664957046508789 2023-01-23 08:59:59.786978: step: 672/1541, loss: 0.5723371505737305 2023-01-23 09:00:00.653380: step: 674/1541, loss: 0.10014176368713379 2023-01-23 09:00:01.545098: step: 676/1541, loss: 0.03459944948554039 2023-01-23 09:00:02.465718: step: 678/1541, loss: 0.25770923495292664 2023-01-23 09:00:03.382232: step: 680/1541, loss: 0.13279281556606293 2023-01-23 09:00:04.290608: step: 682/1541, loss: 0.05217217281460762 2023-01-23 09:00:05.169509: step: 684/1541, loss: 0.3082217872142792 2023-01-23 09:00:06.089262: step: 686/1541, loss: 0.4289914667606354 2023-01-23 09:00:06.950005: step: 688/1541, loss: 0.4104774296283722 2023-01-23 09:00:07.836188: step: 690/1541, loss: 0.23686490952968597 2023-01-23 09:00:08.774068: step: 692/1541, loss: 0.76156085729599 2023-01-23 09:00:09.669078: step: 694/1541, loss: 0.19556811451911926 2023-01-23 09:00:10.569502: step: 696/1541, loss: 0.44717568159103394 2023-01-23 09:00:11.455613: step: 698/1541, loss: 0.28095537424087524 2023-01-23 09:00:12.378702: step: 700/1541, loss: 0.3327099084854126 2023-01-23 09:00:13.254433: step: 702/1541, loss: 2.0842702388763428 2023-01-23 09:00:14.190597: step: 704/1541, loss: 0.10384751111268997 2023-01-23 09:00:15.081524: step: 706/1541, loss: 0.46085771918296814 2023-01-23 09:00:16.017266: step: 708/1541, loss: 0.08412142097949982 2023-01-23 09:00:16.873899: step: 710/1541, loss: 0.27391961216926575 2023-01-23 09:00:17.772755: step: 712/1541, loss: 1.3946641683578491 2023-01-23 09:00:18.674234: step: 714/1541, loss: 0.27037736773490906 2023-01-23 09:00:19.536336: step: 716/1541, loss: 0.400256484746933 2023-01-23 09:00:20.454999: step: 718/1541, loss: 0.3715636730194092 2023-01-23 09:00:21.354032: step: 720/1541, loss: 0.5654448866844177 2023-01-23 09:00:22.272098: step: 722/1541, loss: 0.1279812604188919 2023-01-23 09:00:23.148447: step: 724/1541, loss: 0.3619259297847748 2023-01-23 09:00:24.051873: step: 726/1541, loss: 0.11784109473228455 2023-01-23 09:00:24.943494: step: 728/1541, loss: 0.18772360682487488 2023-01-23 09:00:25.865404: step: 730/1541, loss: 0.16780604422092438 2023-01-23 09:00:26.745959: step: 732/1541, loss: 0.2969396114349365 2023-01-23 09:00:27.651562: step: 734/1541, loss: 0.017602022737264633 2023-01-23 09:00:28.603930: step: 736/1541, loss: 0.4443768858909607 2023-01-23 09:00:29.489293: step: 738/1541, loss: 0.3947889506816864 2023-01-23 09:00:30.406872: step: 740/1541, loss: 0.09203210473060608 2023-01-23 09:00:31.295083: step: 742/1541, loss: 0.11011610925197601 2023-01-23 09:00:32.208358: step: 744/1541, loss: 0.14592063426971436 2023-01-23 09:00:33.102242: step: 746/1541, loss: 0.14793990552425385 2023-01-23 09:00:33.979708: step: 748/1541, loss: 0.16475842893123627 2023-01-23 09:00:34.906052: step: 750/1541, loss: 0.04989202320575714 2023-01-23 09:00:35.827585: step: 752/1541, loss: 0.2082868367433548 2023-01-23 09:00:36.714400: step: 754/1541, loss: 0.06723321974277496 2023-01-23 09:00:37.603033: step: 756/1541, loss: 0.15014812350273132 2023-01-23 09:00:38.491994: step: 758/1541, loss: 0.21786350011825562 2023-01-23 09:00:39.383594: step: 760/1541, loss: 0.05235901474952698 2023-01-23 09:00:40.248993: step: 762/1541, loss: 0.2466575652360916 2023-01-23 09:00:41.164470: step: 764/1541, loss: 0.2749221622943878 2023-01-23 09:00:42.112953: step: 766/1541, loss: 0.3417665362358093 2023-01-23 09:00:43.030383: step: 768/1541, loss: 0.2253231257200241 2023-01-23 09:00:43.922325: step: 770/1541, loss: 0.3732249438762665 2023-01-23 09:00:44.824701: step: 772/1541, loss: 0.1889524906873703 2023-01-23 09:00:45.746100: step: 774/1541, loss: 0.5758458375930786 2023-01-23 09:00:46.665851: step: 776/1541, loss: 0.312568724155426 2023-01-23 09:00:47.614826: step: 778/1541, loss: 0.6710711121559143 2023-01-23 09:00:48.461655: step: 780/1541, loss: 0.2583059072494507 2023-01-23 09:00:49.409875: step: 782/1541, loss: 0.7300339341163635 2023-01-23 09:00:50.269059: step: 784/1541, loss: 0.10472672432661057 2023-01-23 09:00:51.311464: step: 786/1541, loss: 0.1472710818052292 2023-01-23 09:00:52.286143: step: 788/1541, loss: 0.1931128054857254 2023-01-23 09:00:53.260098: step: 790/1541, loss: 0.07773993909358978 2023-01-23 09:00:54.243501: step: 792/1541, loss: 0.14281587302684784 2023-01-23 09:00:55.129630: step: 794/1541, loss: 0.06897807121276855 2023-01-23 09:00:56.015430: step: 796/1541, loss: 0.12431960552930832 2023-01-23 09:00:56.959751: step: 798/1541, loss: 0.09060949832201004 2023-01-23 09:00:57.850526: step: 800/1541, loss: 0.3322908580303192 2023-01-23 09:00:58.767652: step: 802/1541, loss: 0.12348435819149017 2023-01-23 09:00:59.689612: step: 804/1541, loss: 0.07012984156608582 2023-01-23 09:01:00.593405: step: 806/1541, loss: 0.3104954957962036 2023-01-23 09:01:01.582374: step: 808/1541, loss: 0.8441667556762695 2023-01-23 09:01:02.540891: step: 810/1541, loss: 0.14578984677791595 2023-01-23 09:01:03.432892: step: 812/1541, loss: 0.4660199284553528 2023-01-23 09:01:04.298681: step: 814/1541, loss: 0.2132277935743332 2023-01-23 09:01:05.283576: step: 816/1541, loss: 0.036176204681396484 2023-01-23 09:01:06.210032: step: 818/1541, loss: 0.18837371468544006 2023-01-23 09:01:07.120053: step: 820/1541, loss: 0.39951127767562866 2023-01-23 09:01:08.023747: step: 822/1541, loss: 0.1523778736591339 2023-01-23 09:01:08.898488: step: 824/1541, loss: 0.22080984711647034 2023-01-23 09:01:09.844873: step: 826/1541, loss: 0.038705974817276 2023-01-23 09:01:10.757845: step: 828/1541, loss: 0.3045610189437866 2023-01-23 09:01:11.656409: step: 830/1541, loss: 0.31715109944343567 2023-01-23 09:01:12.528114: step: 832/1541, loss: 0.17403289675712585 2023-01-23 09:01:13.418279: step: 834/1541, loss: 0.19593225419521332 2023-01-23 09:01:14.306346: step: 836/1541, loss: 0.2538262903690338 2023-01-23 09:01:15.167918: step: 838/1541, loss: 0.14544810354709625 2023-01-23 09:01:16.057535: step: 840/1541, loss: 0.17134276032447815 2023-01-23 09:01:16.984486: step: 842/1541, loss: 0.3153569996356964 2023-01-23 09:01:17.903764: step: 844/1541, loss: 0.3979543447494507 2023-01-23 09:01:18.844189: step: 846/1541, loss: 0.030768748372793198 2023-01-23 09:01:19.772471: step: 848/1541, loss: 0.21108311414718628 2023-01-23 09:01:20.685049: step: 850/1541, loss: 0.139243945479393 2023-01-23 09:01:21.548775: step: 852/1541, loss: 0.49013572931289673 2023-01-23 09:01:22.439341: step: 854/1541, loss: 0.21392719447612762 2023-01-23 09:01:23.302948: step: 856/1541, loss: 0.17514900863170624 2023-01-23 09:01:24.265852: step: 858/1541, loss: 0.2789953947067261 2023-01-23 09:01:25.161178: step: 860/1541, loss: 0.14631973206996918 2023-01-23 09:01:26.135694: step: 862/1541, loss: 0.44938918948173523 2023-01-23 09:01:27.009849: step: 864/1541, loss: 0.04130754992365837 2023-01-23 09:01:27.865923: step: 866/1541, loss: 0.11637964099645615 2023-01-23 09:01:28.797212: step: 868/1541, loss: 0.1276969611644745 2023-01-23 09:01:29.690867: step: 870/1541, loss: 1.3175559043884277 2023-01-23 09:01:30.607256: step: 872/1541, loss: 0.11711776256561279 2023-01-23 09:01:31.582823: step: 874/1541, loss: 0.2790093719959259 2023-01-23 09:01:32.452760: step: 876/1541, loss: 0.21923306584358215 2023-01-23 09:01:33.380252: step: 878/1541, loss: 0.18338677287101746 2023-01-23 09:01:34.300916: step: 880/1541, loss: 0.5767916440963745 2023-01-23 09:01:35.207624: step: 882/1541, loss: 0.18558937311172485 2023-01-23 09:01:36.104261: step: 884/1541, loss: 0.6982947587966919 2023-01-23 09:01:37.055479: step: 886/1541, loss: 0.05277671664953232 2023-01-23 09:01:38.042084: step: 888/1541, loss: 0.5176936984062195 2023-01-23 09:01:38.961808: step: 890/1541, loss: 0.25163573026657104 2023-01-23 09:01:39.862413: step: 892/1541, loss: 0.22499710321426392 2023-01-23 09:01:40.731755: step: 894/1541, loss: 0.1145203560590744 2023-01-23 09:01:41.650695: step: 896/1541, loss: 0.19865915179252625 2023-01-23 09:01:42.575357: step: 898/1541, loss: 0.5642952919006348 2023-01-23 09:01:43.444933: step: 900/1541, loss: 0.27136391401290894 2023-01-23 09:01:44.331889: step: 902/1541, loss: 0.5017983317375183 2023-01-23 09:01:45.241947: step: 904/1541, loss: 0.30752742290496826 2023-01-23 09:01:46.112533: step: 906/1541, loss: 0.14372090995311737 2023-01-23 09:01:47.082736: step: 908/1541, loss: 0.11875452101230621 2023-01-23 09:01:47.990118: step: 910/1541, loss: 0.12513580918312073 2023-01-23 09:01:48.875831: step: 912/1541, loss: 0.0193899255245924 2023-01-23 09:01:49.808155: step: 914/1541, loss: 0.3006480038166046 2023-01-23 09:01:50.663549: step: 916/1541, loss: 0.47228652238845825 2023-01-23 09:01:51.575255: step: 918/1541, loss: 0.28863710165023804 2023-01-23 09:01:52.454593: step: 920/1541, loss: 0.118068166077137 2023-01-23 09:01:53.363609: step: 922/1541, loss: 0.4244133532047272 2023-01-23 09:01:54.240621: step: 924/1541, loss: 0.1281542330980301 2023-01-23 09:01:55.110496: step: 926/1541, loss: 0.37754303216934204 2023-01-23 09:01:55.962274: step: 928/1541, loss: 0.12538829445838928 2023-01-23 09:01:56.810971: step: 930/1541, loss: 0.22629056870937347 2023-01-23 09:01:57.686938: step: 932/1541, loss: 0.08328033238649368 2023-01-23 09:01:58.578482: step: 934/1541, loss: 0.5021407008171082 2023-01-23 09:01:59.458495: step: 936/1541, loss: 0.3451068699359894 2023-01-23 09:02:00.322060: step: 938/1541, loss: 0.21503278613090515 2023-01-23 09:02:01.342254: step: 940/1541, loss: 0.47299715876579285 2023-01-23 09:02:02.219118: step: 942/1541, loss: 0.23840612173080444 2023-01-23 09:02:03.104818: step: 944/1541, loss: 0.2206846922636032 2023-01-23 09:02:03.950799: step: 946/1541, loss: 0.05052908882498741 2023-01-23 09:02:04.847965: step: 948/1541, loss: 0.36651620268821716 2023-01-23 09:02:05.782138: step: 950/1541, loss: 0.037198275327682495 2023-01-23 09:02:06.671079: step: 952/1541, loss: 0.3116895258426666 2023-01-23 09:02:07.538246: step: 954/1541, loss: 0.0942276194691658 2023-01-23 09:02:08.438759: step: 956/1541, loss: 0.5378003716468811 2023-01-23 09:02:09.338112: step: 958/1541, loss: 0.25590577721595764 2023-01-23 09:02:10.270208: step: 960/1541, loss: 0.40703922510147095 2023-01-23 09:02:11.152645: step: 962/1541, loss: 2.0952770709991455 2023-01-23 09:02:12.053693: step: 964/1541, loss: 0.3364919424057007 2023-01-23 09:02:12.976031: step: 966/1541, loss: 0.14618776738643646 2023-01-23 09:02:13.859858: step: 968/1541, loss: 0.20529524981975555 2023-01-23 09:02:14.741237: step: 970/1541, loss: 0.6164051294326782 2023-01-23 09:02:15.657969: step: 972/1541, loss: 0.25023341178894043 2023-01-23 09:02:16.564826: step: 974/1541, loss: 0.22412723302841187 2023-01-23 09:02:17.455044: step: 976/1541, loss: 0.3247217535972595 2023-01-23 09:02:18.333995: step: 978/1541, loss: 0.2290809452533722 2023-01-23 09:02:19.206808: step: 980/1541, loss: 0.31088924407958984 2023-01-23 09:02:20.204196: step: 982/1541, loss: 2.7000038623809814 2023-01-23 09:02:21.089387: step: 984/1541, loss: 0.45086556673049927 2023-01-23 09:02:22.002855: step: 986/1541, loss: 0.07382301986217499 2023-01-23 09:02:22.902690: step: 988/1541, loss: 0.16215567290782928 2023-01-23 09:02:23.785104: step: 990/1541, loss: 0.17826691269874573 2023-01-23 09:02:24.683567: step: 992/1541, loss: 0.07513023167848587 2023-01-23 09:02:25.591041: step: 994/1541, loss: 0.12508635222911835 2023-01-23 09:02:26.467864: step: 996/1541, loss: 0.0926673412322998 2023-01-23 09:02:27.352466: step: 998/1541, loss: 0.33914831280708313 2023-01-23 09:02:28.250068: step: 1000/1541, loss: 0.5552700161933899 2023-01-23 09:02:29.201108: step: 1002/1541, loss: 2.860319137573242 2023-01-23 09:02:30.087413: step: 1004/1541, loss: 0.11927483975887299 2023-01-23 09:02:30.973785: step: 1006/1541, loss: 0.17764678597450256 2023-01-23 09:02:31.947195: step: 1008/1541, loss: 0.6294629573822021 2023-01-23 09:02:32.867482: step: 1010/1541, loss: 0.048417117446660995 2023-01-23 09:02:33.747422: step: 1012/1541, loss: 0.5354798436164856 2023-01-23 09:02:34.643131: step: 1014/1541, loss: 0.24024765193462372 2023-01-23 09:02:35.481427: step: 1016/1541, loss: 0.14350134134292603 2023-01-23 09:02:36.315988: step: 1018/1541, loss: 0.12302000820636749 2023-01-23 09:02:37.214002: step: 1020/1541, loss: 0.05177362635731697 2023-01-23 09:02:38.097579: step: 1022/1541, loss: 0.3048599660396576 2023-01-23 09:02:38.981980: step: 1024/1541, loss: 0.5705740451812744 2023-01-23 09:02:39.915065: step: 1026/1541, loss: 0.1447829306125641 2023-01-23 09:02:40.800462: step: 1028/1541, loss: 0.2789214849472046 2023-01-23 09:02:41.706017: step: 1030/1541, loss: 0.11631966382265091 2023-01-23 09:02:42.723784: step: 1032/1541, loss: 0.462648868560791 2023-01-23 09:02:43.615555: step: 1034/1541, loss: 0.1569337546825409 2023-01-23 09:02:44.497105: step: 1036/1541, loss: 0.5990718007087708 2023-01-23 09:02:45.415463: step: 1038/1541, loss: 0.08771166205406189 2023-01-23 09:02:46.267064: step: 1040/1541, loss: 0.545964777469635 2023-01-23 09:02:47.176512: step: 1042/1541, loss: 0.26331233978271484 2023-01-23 09:02:48.055238: step: 1044/1541, loss: 0.4375384449958801 2023-01-23 09:02:48.924091: step: 1046/1541, loss: 0.18368643522262573 2023-01-23 09:02:49.808401: step: 1048/1541, loss: 0.1786356270313263 2023-01-23 09:02:50.731062: step: 1050/1541, loss: 0.4940832257270813 2023-01-23 09:02:51.634768: step: 1052/1541, loss: 0.13508009910583496 2023-01-23 09:02:52.510337: step: 1054/1541, loss: 0.04129781201481819 2023-01-23 09:02:53.407612: step: 1056/1541, loss: 0.20675742626190186 2023-01-23 09:02:54.301076: step: 1058/1541, loss: 0.48036912083625793 2023-01-23 09:02:55.197857: step: 1060/1541, loss: 0.181174173951149 2023-01-23 09:02:56.087675: step: 1062/1541, loss: 0.3179694712162018 2023-01-23 09:02:57.113100: step: 1064/1541, loss: 0.23812101781368256 2023-01-23 09:02:58.036983: step: 1066/1541, loss: 0.5655955076217651 2023-01-23 09:02:58.918615: step: 1068/1541, loss: 0.1307605803012848 2023-01-23 09:02:59.813522: step: 1070/1541, loss: 0.3554949164390564 2023-01-23 09:03:00.700425: step: 1072/1541, loss: 0.12349610030651093 2023-01-23 09:03:01.577379: step: 1074/1541, loss: 0.16179193556308746 2023-01-23 09:03:02.462100: step: 1076/1541, loss: 0.12441294640302658 2023-01-23 09:03:03.352623: step: 1078/1541, loss: 0.31493374705314636 2023-01-23 09:03:04.242892: step: 1080/1541, loss: 0.11727207899093628 2023-01-23 09:03:05.108667: step: 1082/1541, loss: 0.8298004865646362 2023-01-23 09:03:05.978775: step: 1084/1541, loss: 0.6018333435058594 2023-01-23 09:03:06.924177: step: 1086/1541, loss: 0.2098284810781479 2023-01-23 09:03:07.797404: step: 1088/1541, loss: 0.12035094946622849 2023-01-23 09:03:08.738284: step: 1090/1541, loss: 0.09265388548374176 2023-01-23 09:03:09.652434: step: 1092/1541, loss: 0.3167804777622223 2023-01-23 09:03:10.561614: step: 1094/1541, loss: 0.5232741236686707 2023-01-23 09:03:11.455721: step: 1096/1541, loss: 0.7635761499404907 2023-01-23 09:03:12.372144: step: 1098/1541, loss: 0.030391426756978035 2023-01-23 09:03:13.291459: step: 1100/1541, loss: 0.2759545147418976 2023-01-23 09:03:14.179641: step: 1102/1541, loss: 0.23467017710208893 2023-01-23 09:03:15.098926: step: 1104/1541, loss: 0.07231434434652328 2023-01-23 09:03:15.988247: step: 1106/1541, loss: 0.23580245673656464 2023-01-23 09:03:16.886081: step: 1108/1541, loss: 0.29032647609710693 2023-01-23 09:03:17.766834: step: 1110/1541, loss: 0.11443343013525009 2023-01-23 09:03:18.679000: step: 1112/1541, loss: 0.4167023003101349 2023-01-23 09:03:19.544855: step: 1114/1541, loss: 0.3791022002696991 2023-01-23 09:03:20.444743: step: 1116/1541, loss: 1.10447359085083 2023-01-23 09:03:21.441479: step: 1118/1541, loss: 0.6088427305221558 2023-01-23 09:03:22.371548: step: 1120/1541, loss: 0.11298108100891113 2023-01-23 09:03:23.248353: step: 1122/1541, loss: 0.25600770115852356 2023-01-23 09:03:24.138006: step: 1124/1541, loss: 0.19295823574066162 2023-01-23 09:03:25.055705: step: 1126/1541, loss: 0.16079553961753845 2023-01-23 09:03:25.951653: step: 1128/1541, loss: 0.17034828662872314 2023-01-23 09:03:26.843403: step: 1130/1541, loss: 0.40280771255493164 2023-01-23 09:03:27.699819: step: 1132/1541, loss: 0.2788561284542084 2023-01-23 09:03:28.572693: step: 1134/1541, loss: 0.41007721424102783 2023-01-23 09:03:29.474731: step: 1136/1541, loss: 0.2304895967245102 2023-01-23 09:03:30.330193: step: 1138/1541, loss: 0.2223493754863739 2023-01-23 09:03:31.195257: step: 1140/1541, loss: 0.2303449511528015 2023-01-23 09:03:32.076974: step: 1142/1541, loss: 0.4681335985660553 2023-01-23 09:03:32.985902: step: 1144/1541, loss: 0.27100905776023865 2023-01-23 09:03:33.862739: step: 1146/1541, loss: 0.3017275035381317 2023-01-23 09:03:34.775409: step: 1148/1541, loss: 0.3755943179130554 2023-01-23 09:03:35.646639: step: 1150/1541, loss: 0.24946200847625732 2023-01-23 09:03:36.539451: step: 1152/1541, loss: 0.20885556936264038 2023-01-23 09:03:37.471182: step: 1154/1541, loss: 0.9257455468177795 2023-01-23 09:03:38.319997: step: 1156/1541, loss: 0.04105904698371887 2023-01-23 09:03:39.214204: step: 1158/1541, loss: 0.7386103868484497 2023-01-23 09:03:40.088236: step: 1160/1541, loss: 0.22729259729385376 2023-01-23 09:03:40.982335: step: 1162/1541, loss: 0.24831314384937286 2023-01-23 09:03:41.859021: step: 1164/1541, loss: 0.4692046046257019 2023-01-23 09:03:42.744422: step: 1166/1541, loss: 0.10201333463191986 2023-01-23 09:03:43.650884: step: 1168/1541, loss: 0.15686500072479248 2023-01-23 09:03:44.539307: step: 1170/1541, loss: 0.14583277702331543 2023-01-23 09:03:45.430538: step: 1172/1541, loss: 0.11325076222419739 2023-01-23 09:03:46.299921: step: 1174/1541, loss: 0.4039437174797058 2023-01-23 09:03:47.201821: step: 1176/1541, loss: 0.07500102370977402 2023-01-23 09:03:48.073098: step: 1178/1541, loss: 0.15522345900535583 2023-01-23 09:03:49.001763: step: 1180/1541, loss: 0.12913398444652557 2023-01-23 09:03:49.898840: step: 1182/1541, loss: 0.5198084712028503 2023-01-23 09:03:50.792159: step: 1184/1541, loss: 0.16138452291488647 2023-01-23 09:03:51.705582: step: 1186/1541, loss: 0.38517096638679504 2023-01-23 09:03:52.608271: step: 1188/1541, loss: 1.0309265851974487 2023-01-23 09:03:53.484486: step: 1190/1541, loss: 0.17880472540855408 2023-01-23 09:03:54.380308: step: 1192/1541, loss: 0.4249650239944458 2023-01-23 09:03:55.293220: step: 1194/1541, loss: 0.07639306783676147 2023-01-23 09:03:56.170967: step: 1196/1541, loss: 0.08034177124500275 2023-01-23 09:03:57.034838: step: 1198/1541, loss: 0.6297215819358826 2023-01-23 09:03:57.948922: step: 1200/1541, loss: 0.04531107097864151 2023-01-23 09:03:58.851316: step: 1202/1541, loss: 0.35390642285346985 2023-01-23 09:03:59.735907: step: 1204/1541, loss: 0.40880194306373596 2023-01-23 09:04:00.590857: step: 1206/1541, loss: 0.32089468836784363 2023-01-23 09:04:01.593744: step: 1208/1541, loss: 0.2455144226551056 2023-01-23 09:04:02.485347: step: 1210/1541, loss: 0.36539897322654724 2023-01-23 09:04:03.339087: step: 1212/1541, loss: 0.29531556367874146 2023-01-23 09:04:04.206520: step: 1214/1541, loss: 0.05364980548620224 2023-01-23 09:04:05.146893: step: 1216/1541, loss: 0.12103310227394104 2023-01-23 09:04:06.051585: step: 1218/1541, loss: 0.11411546915769577 2023-01-23 09:04:06.927153: step: 1220/1541, loss: 0.12936106324195862 2023-01-23 09:04:07.863459: step: 1222/1541, loss: 0.7065333724021912 2023-01-23 09:04:08.757065: step: 1224/1541, loss: 0.05020981654524803 2023-01-23 09:04:09.677206: step: 1226/1541, loss: 0.0970168188214302 2023-01-23 09:04:10.580880: step: 1228/1541, loss: 5.8890557289123535 2023-01-23 09:04:11.469636: step: 1230/1541, loss: 0.10054980218410492 2023-01-23 09:04:12.324284: step: 1232/1541, loss: 0.3572758436203003 2023-01-23 09:04:13.222595: step: 1234/1541, loss: 0.21362270414829254 2023-01-23 09:04:14.170338: step: 1236/1541, loss: 0.5132401585578918 2023-01-23 09:04:15.088527: step: 1238/1541, loss: 0.11742381006479263 2023-01-23 09:04:15.974581: step: 1240/1541, loss: 1.0878351926803589 2023-01-23 09:04:16.923431: step: 1242/1541, loss: 0.14776445925235748 2023-01-23 09:04:17.959671: step: 1244/1541, loss: 0.10418710112571716 2023-01-23 09:04:18.838092: step: 1246/1541, loss: 0.39260393381118774 2023-01-23 09:04:19.740545: step: 1248/1541, loss: 0.578518271446228 2023-01-23 09:04:20.655203: step: 1250/1541, loss: 0.5121132731437683 2023-01-23 09:04:21.586031: step: 1252/1541, loss: 0.2725147306919098 2023-01-23 09:04:22.538393: step: 1254/1541, loss: 0.5009443163871765 2023-01-23 09:04:23.477632: step: 1256/1541, loss: 0.3903195559978485 2023-01-23 09:04:24.380194: step: 1258/1541, loss: 0.4770912826061249 2023-01-23 09:04:25.274070: step: 1260/1541, loss: 0.15723836421966553 2023-01-23 09:04:26.164775: step: 1262/1541, loss: 0.13259807229042053 2023-01-23 09:04:27.043307: step: 1264/1541, loss: 0.2899843752384186 2023-01-23 09:04:27.935458: step: 1266/1541, loss: 2.7200381755828857 2023-01-23 09:04:28.830906: step: 1268/1541, loss: 0.04518859088420868 2023-01-23 09:04:29.759270: step: 1270/1541, loss: 0.3649928569793701 2023-01-23 09:04:30.649054: step: 1272/1541, loss: 0.3672371804714203 2023-01-23 09:04:31.544207: step: 1274/1541, loss: 0.07631847262382507 2023-01-23 09:04:32.405322: step: 1276/1541, loss: 0.5566667318344116 2023-01-23 09:04:33.255208: step: 1278/1541, loss: 0.7055189609527588 2023-01-23 09:04:34.115985: step: 1280/1541, loss: 0.32574811577796936 2023-01-23 09:04:35.156277: step: 1282/1541, loss: 0.6312763094902039 2023-01-23 09:04:36.030101: step: 1284/1541, loss: 0.13628222048282623 2023-01-23 09:04:36.915437: step: 1286/1541, loss: 0.858329713344574 2023-01-23 09:04:37.834272: step: 1288/1541, loss: 0.1847183257341385 2023-01-23 09:04:38.729251: step: 1290/1541, loss: 0.32233256101608276 2023-01-23 09:04:39.584839: step: 1292/1541, loss: 0.9799516201019287 2023-01-23 09:04:40.489329: step: 1294/1541, loss: 0.4099602997303009 2023-01-23 09:04:41.391689: step: 1296/1541, loss: 0.19952435791492462 2023-01-23 09:04:42.277679: step: 1298/1541, loss: 0.27838823199272156 2023-01-23 09:04:43.144185: step: 1300/1541, loss: 0.2843322157859802 2023-01-23 09:04:44.030250: step: 1302/1541, loss: 2.3917181491851807 2023-01-23 09:04:44.959778: step: 1304/1541, loss: 0.23297400772571564 2023-01-23 09:04:45.839163: step: 1306/1541, loss: 0.1553962677717209 2023-01-23 09:04:46.725767: step: 1308/1541, loss: 0.5410343408584595 2023-01-23 09:04:47.588464: step: 1310/1541, loss: 0.30465617775917053 2023-01-23 09:04:48.509529: step: 1312/1541, loss: 0.34324875473976135 2023-01-23 09:04:49.421597: step: 1314/1541, loss: 0.20564399659633636 2023-01-23 09:04:50.374452: step: 1316/1541, loss: 0.08096019178628922 2023-01-23 09:04:51.245246: step: 1318/1541, loss: 0.5900781750679016 2023-01-23 09:04:52.111288: step: 1320/1541, loss: 0.16541516780853271 2023-01-23 09:04:53.020903: step: 1322/1541, loss: 0.4655025899410248 2023-01-23 09:04:53.895343: step: 1324/1541, loss: 0.25810080766677856 2023-01-23 09:04:54.820673: step: 1326/1541, loss: 0.4015727937221527 2023-01-23 09:04:55.731494: step: 1328/1541, loss: 0.04106136038899422 2023-01-23 09:04:56.649039: step: 1330/1541, loss: 0.02401904948055744 2023-01-23 09:04:57.525852: step: 1332/1541, loss: 0.6785838007926941 2023-01-23 09:04:58.401088: step: 1334/1541, loss: 0.7778465747833252 2023-01-23 09:04:59.312325: step: 1336/1541, loss: 0.057694029062986374 2023-01-23 09:05:00.191328: step: 1338/1541, loss: 0.7183762788772583 2023-01-23 09:05:01.063763: step: 1340/1541, loss: 0.6136292219161987 2023-01-23 09:05:01.974257: step: 1342/1541, loss: 0.16966712474822998 2023-01-23 09:05:02.851940: step: 1344/1541, loss: 0.5475000143051147 2023-01-23 09:05:03.749744: step: 1346/1541, loss: 0.5677876472473145 2023-01-23 09:05:04.635325: step: 1348/1541, loss: 0.1866384595632553 2023-01-23 09:05:05.515277: step: 1350/1541, loss: 0.1958206295967102 2023-01-23 09:05:06.404481: step: 1352/1541, loss: 0.37264925241470337 2023-01-23 09:05:07.263736: step: 1354/1541, loss: 0.09920679777860641 2023-01-23 09:05:08.110865: step: 1356/1541, loss: 3.2297701835632324 2023-01-23 09:05:09.019058: step: 1358/1541, loss: 0.13891273736953735 2023-01-23 09:05:09.924505: step: 1360/1541, loss: 0.467173308134079 2023-01-23 09:05:10.889832: step: 1362/1541, loss: 0.16531886160373688 2023-01-23 09:05:11.795720: step: 1364/1541, loss: 0.919691264629364 2023-01-23 09:05:12.693917: step: 1366/1541, loss: 0.3502064347267151 2023-01-23 09:05:13.590797: step: 1368/1541, loss: 0.09208566695451736 2023-01-23 09:05:14.475212: step: 1370/1541, loss: 0.42855995893478394 2023-01-23 09:05:15.399274: step: 1372/1541, loss: 0.08867386728525162 2023-01-23 09:05:16.254222: step: 1374/1541, loss: 0.6000205278396606 2023-01-23 09:05:17.173259: step: 1376/1541, loss: 0.33365398645401 2023-01-23 09:05:18.031722: step: 1378/1541, loss: 0.1820664256811142 2023-01-23 09:05:18.952562: step: 1380/1541, loss: 0.34618398547172546 2023-01-23 09:05:19.863249: step: 1382/1541, loss: 3.2218737602233887 2023-01-23 09:05:20.760488: step: 1384/1541, loss: 0.24653363227844238 2023-01-23 09:05:21.633677: step: 1386/1541, loss: 0.1698988974094391 2023-01-23 09:05:22.543347: step: 1388/1541, loss: 0.12640295922756195 2023-01-23 09:05:23.435086: step: 1390/1541, loss: 0.09399135410785675 2023-01-23 09:05:24.368885: step: 1392/1541, loss: 1.4806290864944458 2023-01-23 09:05:25.216347: step: 1394/1541, loss: 0.3122166395187378 2023-01-23 09:05:26.182404: step: 1396/1541, loss: 0.1530916541814804 2023-01-23 09:05:27.166984: step: 1398/1541, loss: 0.4404471814632416 2023-01-23 09:05:28.064986: step: 1400/1541, loss: 0.299938827753067 2023-01-23 09:05:28.971194: step: 1402/1541, loss: 0.4630556106567383 2023-01-23 09:05:29.870601: step: 1404/1541, loss: 0.04556753486394882 2023-01-23 09:05:30.768818: step: 1406/1541, loss: 0.24312321841716766 2023-01-23 09:05:31.635607: step: 1408/1541, loss: 0.5969514846801758 2023-01-23 09:05:32.579336: step: 1410/1541, loss: 0.14200256764888763 2023-01-23 09:05:33.452039: step: 1412/1541, loss: 0.2039196491241455 2023-01-23 09:05:34.325102: step: 1414/1541, loss: 0.338051438331604 2023-01-23 09:05:35.258305: step: 1416/1541, loss: 0.40465450286865234 2023-01-23 09:05:36.190581: step: 1418/1541, loss: 0.2318529486656189 2023-01-23 09:05:37.099333: step: 1420/1541, loss: 0.8091174960136414 2023-01-23 09:05:38.000907: step: 1422/1541, loss: 0.18784043192863464 2023-01-23 09:05:38.893572: step: 1424/1541, loss: 0.25198620557785034 2023-01-23 09:05:39.737028: step: 1426/1541, loss: 0.20980480313301086 2023-01-23 09:05:40.607132: step: 1428/1541, loss: 0.2568448483943939 2023-01-23 09:05:41.454643: step: 1430/1541, loss: 0.15628716349601746 2023-01-23 09:05:42.352720: step: 1432/1541, loss: 0.32791000604629517 2023-01-23 09:05:43.241009: step: 1434/1541, loss: 0.25104498863220215 2023-01-23 09:05:44.125624: step: 1436/1541, loss: 0.26576822996139526 2023-01-23 09:05:45.056627: step: 1438/1541, loss: 0.10982480645179749 2023-01-23 09:05:45.929815: step: 1440/1541, loss: 0.16887040436267853 2023-01-23 09:05:46.869483: step: 1442/1541, loss: 0.7564423680305481 2023-01-23 09:05:47.843934: step: 1444/1541, loss: 0.2205217033624649 2023-01-23 09:05:48.723665: step: 1446/1541, loss: 0.07152381539344788 2023-01-23 09:05:49.637845: step: 1448/1541, loss: 0.18054452538490295 2023-01-23 09:05:50.552608: step: 1450/1541, loss: 0.22854705154895782 2023-01-23 09:05:51.419652: step: 1452/1541, loss: 0.29404568672180176 2023-01-23 09:05:52.315079: step: 1454/1541, loss: 0.37390077114105225 2023-01-23 09:05:53.228227: step: 1456/1541, loss: 0.15028201043605804 2023-01-23 09:05:54.125882: step: 1458/1541, loss: 0.26791664958000183 2023-01-23 09:05:55.041323: step: 1460/1541, loss: 0.20377297699451447 2023-01-23 09:05:55.905362: step: 1462/1541, loss: 0.12819457054138184 2023-01-23 09:05:56.772489: step: 1464/1541, loss: 0.25015363097190857 2023-01-23 09:05:57.669192: step: 1466/1541, loss: 0.7870298027992249 2023-01-23 09:05:58.599780: step: 1468/1541, loss: 0.05047915130853653 2023-01-23 09:05:59.498330: step: 1470/1541, loss: 0.34597504138946533 2023-01-23 09:06:00.400641: step: 1472/1541, loss: 0.15243873000144958 2023-01-23 09:06:01.305274: step: 1474/1541, loss: 0.21206170320510864 2023-01-23 09:06:02.259186: step: 1476/1541, loss: 0.282589852809906 2023-01-23 09:06:03.165371: step: 1478/1541, loss: 0.1517818719148636 2023-01-23 09:06:04.075481: step: 1480/1541, loss: 0.6855144500732422 2023-01-23 09:06:05.145898: step: 1482/1541, loss: 1.3373395204544067 2023-01-23 09:06:06.058058: step: 1484/1541, loss: 0.3130142390727997 2023-01-23 09:06:06.958177: step: 1486/1541, loss: 0.24349245429039001 2023-01-23 09:06:07.868748: step: 1488/1541, loss: 0.3544529676437378 2023-01-23 09:06:08.823857: step: 1490/1541, loss: 0.21878556907176971 2023-01-23 09:06:09.748591: step: 1492/1541, loss: 0.42255890369415283 2023-01-23 09:06:10.641052: step: 1494/1541, loss: 0.2949022948741913 2023-01-23 09:06:11.528262: step: 1496/1541, loss: 0.18336351215839386 2023-01-23 09:06:12.406672: step: 1498/1541, loss: 0.06450431048870087 2023-01-23 09:06:13.314969: step: 1500/1541, loss: 0.16090720891952515 2023-01-23 09:06:14.246391: step: 1502/1541, loss: 0.37485718727111816 2023-01-23 09:06:15.135565: step: 1504/1541, loss: 0.7896736264228821 2023-01-23 09:06:16.035391: step: 1506/1541, loss: 0.23848704993724823 2023-01-23 09:06:16.916403: step: 1508/1541, loss: 0.19833926856517792 2023-01-23 09:06:17.785735: step: 1510/1541, loss: 0.3704019784927368 2023-01-23 09:06:18.696663: step: 1512/1541, loss: 0.14536352455615997 2023-01-23 09:06:19.603028: step: 1514/1541, loss: 0.02074240893125534 2023-01-23 09:06:20.468393: step: 1516/1541, loss: 0.4963991641998291 2023-01-23 09:06:21.359180: step: 1518/1541, loss: 0.28941184282302856 2023-01-23 09:06:22.215063: step: 1520/1541, loss: 0.2314903736114502 2023-01-23 09:06:23.086551: step: 1522/1541, loss: 0.1898326575756073 2023-01-23 09:06:24.064473: step: 1524/1541, loss: 0.11566522717475891 2023-01-23 09:06:24.963212: step: 1526/1541, loss: 0.7352291941642761 2023-01-23 09:06:25.910556: step: 1528/1541, loss: 0.27956071496009827 2023-01-23 09:06:26.845582: step: 1530/1541, loss: 0.8043093681335449 2023-01-23 09:06:27.720839: step: 1532/1541, loss: 0.17066238820552826 2023-01-23 09:06:28.699327: step: 1534/1541, loss: 0.287367045879364 2023-01-23 09:06:29.548601: step: 1536/1541, loss: 0.1925763189792633 2023-01-23 09:06:30.457708: step: 1538/1541, loss: 0.16528978943824768 2023-01-23 09:06:31.321937: step: 1540/1541, loss: 0.07448031008243561 2023-01-23 09:06:32.225189: step: 1542/1541, loss: 0.11729364842176437 2023-01-23 09:06:33.132741: step: 1544/1541, loss: 0.2082531452178955 2023-01-23 09:06:34.081929: step: 1546/1541, loss: 0.3891673982143402 2023-01-23 09:06:34.983196: step: 1548/1541, loss: 0.11647490411996841 2023-01-23 09:06:35.914669: step: 1550/1541, loss: 0.4611850678920746 2023-01-23 09:06:36.848796: step: 1552/1541, loss: 0.22879661619663239 2023-01-23 09:06:37.749630: step: 1554/1541, loss: 0.3967829942703247 2023-01-23 09:06:38.652302: step: 1556/1541, loss: 0.6409009099006653 2023-01-23 09:06:39.586602: step: 1558/1541, loss: 0.1782601922750473 2023-01-23 09:06:40.478444: step: 1560/1541, loss: 0.46329376101493835 2023-01-23 09:06:41.351081: step: 1562/1541, loss: 0.28868845105171204 2023-01-23 09:06:42.301542: step: 1564/1541, loss: 0.25204938650131226 2023-01-23 09:06:43.240142: step: 1566/1541, loss: 0.6222215294837952 2023-01-23 09:06:44.129082: step: 1568/1541, loss: 0.3136622905731201 2023-01-23 09:06:45.056031: step: 1570/1541, loss: 0.17431849241256714 2023-01-23 09:06:45.983968: step: 1572/1541, loss: 0.2539958357810974 2023-01-23 09:06:46.905835: step: 1574/1541, loss: 0.0427895188331604 2023-01-23 09:06:47.853844: step: 1576/1541, loss: 0.22278422117233276 2023-01-23 09:06:48.749282: step: 1578/1541, loss: 0.3164202868938446 2023-01-23 09:06:49.646880: step: 1580/1541, loss: 0.822379469871521 2023-01-23 09:06:50.569451: step: 1582/1541, loss: 0.06537691503763199 2023-01-23 09:06:51.456704: step: 1584/1541, loss: 0.931003987789154 2023-01-23 09:06:52.364064: step: 1586/1541, loss: 0.1486918330192566 2023-01-23 09:06:53.329892: step: 1588/1541, loss: 0.28752678632736206 2023-01-23 09:06:54.210674: step: 1590/1541, loss: 0.15049031376838684 2023-01-23 09:06:55.102885: step: 1592/1541, loss: 0.3993586301803589 2023-01-23 09:06:56.015531: step: 1594/1541, loss: 0.16964013874530792 2023-01-23 09:06:56.895049: step: 1596/1541, loss: 0.28585246205329895 2023-01-23 09:06:57.799442: step: 1598/1541, loss: 0.09459560364484787 2023-01-23 09:06:58.690882: step: 1600/1541, loss: 0.14442411065101624 2023-01-23 09:06:59.573565: step: 1602/1541, loss: 0.9252886176109314 2023-01-23 09:07:00.466325: step: 1604/1541, loss: 0.16414666175842285 2023-01-23 09:07:01.327694: step: 1606/1541, loss: 0.3276593089103699 2023-01-23 09:07:02.322986: step: 1608/1541, loss: 0.16733111441135406 2023-01-23 09:07:03.217621: step: 1610/1541, loss: 0.12275604903697968 2023-01-23 09:07:04.096471: step: 1612/1541, loss: 0.5402511358261108 2023-01-23 09:07:05.023207: step: 1614/1541, loss: 0.0686907097697258 2023-01-23 09:07:05.924586: step: 1616/1541, loss: 0.2368481606245041 2023-01-23 09:07:06.820298: step: 1618/1541, loss: 0.39695072174072266 2023-01-23 09:07:07.732759: step: 1620/1541, loss: 0.6848029494285583 2023-01-23 09:07:08.611575: step: 1622/1541, loss: 0.26814842224121094 2023-01-23 09:07:09.519278: step: 1624/1541, loss: 0.3469451069831848 2023-01-23 09:07:10.434715: step: 1626/1541, loss: 0.4641973078250885 2023-01-23 09:07:11.316511: step: 1628/1541, loss: 0.07180790603160858 2023-01-23 09:07:12.211593: step: 1630/1541, loss: 0.3057088255882263 2023-01-23 09:07:13.108152: step: 1632/1541, loss: 0.21425020694732666 2023-01-23 09:07:14.011286: step: 1634/1541, loss: 0.34023311734199524 2023-01-23 09:07:14.912998: step: 1636/1541, loss: 2.1587443351745605 2023-01-23 09:07:15.800572: step: 1638/1541, loss: 0.7468168139457703 2023-01-23 09:07:16.700629: step: 1640/1541, loss: 1.5374023914337158 2023-01-23 09:07:17.592715: step: 1642/1541, loss: 0.2758271396160126 2023-01-23 09:07:18.504961: step: 1644/1541, loss: 0.15095637738704681 2023-01-23 09:07:19.365461: step: 1646/1541, loss: 0.33348146080970764 2023-01-23 09:07:20.306798: step: 1648/1541, loss: 0.043896690011024475 2023-01-23 09:07:21.206202: step: 1650/1541, loss: 0.20861375331878662 2023-01-23 09:07:22.089268: step: 1652/1541, loss: 0.20285271108150482 2023-01-23 09:07:22.982761: step: 1654/1541, loss: 0.3522891402244568 2023-01-23 09:07:23.825178: step: 1656/1541, loss: 0.05708305537700653 2023-01-23 09:07:24.726130: step: 1658/1541, loss: 1.1171683073043823 2023-01-23 09:07:25.608343: step: 1660/1541, loss: 0.1245218887925148 2023-01-23 09:07:26.493153: step: 1662/1541, loss: 0.33753347396850586 2023-01-23 09:07:27.395645: step: 1664/1541, loss: 0.182266965508461 2023-01-23 09:07:28.258484: step: 1666/1541, loss: 0.3287879228591919 2023-01-23 09:07:29.115792: step: 1668/1541, loss: 0.17591705918312073 2023-01-23 09:07:30.105775: step: 1670/1541, loss: 0.2955394387245178 2023-01-23 09:07:30.999847: step: 1672/1541, loss: 0.2058495283126831 2023-01-23 09:07:31.891097: step: 1674/1541, loss: 1.826682448387146 2023-01-23 09:07:32.761515: step: 1676/1541, loss: 0.033852823078632355 2023-01-23 09:07:33.688607: step: 1678/1541, loss: 0.5080146193504333 2023-01-23 09:07:34.574735: step: 1680/1541, loss: 0.24224413931369781 2023-01-23 09:07:35.449126: step: 1682/1541, loss: 0.13259832561016083 2023-01-23 09:07:36.325448: step: 1684/1541, loss: 0.22362272441387177 2023-01-23 09:07:37.202743: step: 1686/1541, loss: 0.09263278543949127 2023-01-23 09:07:38.069255: step: 1688/1541, loss: 0.09201755374670029 2023-01-23 09:07:38.975105: step: 1690/1541, loss: 0.10790093243122101 2023-01-23 09:07:39.906158: step: 1692/1541, loss: 0.3982125222682953 2023-01-23 09:07:40.818278: step: 1694/1541, loss: 0.22650814056396484 2023-01-23 09:07:41.687096: step: 1696/1541, loss: 0.16733583807945251 2023-01-23 09:07:42.578650: step: 1698/1541, loss: 0.07612943649291992 2023-01-23 09:07:43.474291: step: 1700/1541, loss: 0.10518229007720947 2023-01-23 09:07:44.333097: step: 1702/1541, loss: 0.12214167416095734 2023-01-23 09:07:45.235965: step: 1704/1541, loss: 0.3309216797351837 2023-01-23 09:07:46.113955: step: 1706/1541, loss: 0.2783180773258209 2023-01-23 09:07:47.050341: step: 1708/1541, loss: 0.04939375817775726 2023-01-23 09:07:47.921751: step: 1710/1541, loss: 1.3317455053329468 2023-01-23 09:07:48.828801: step: 1712/1541, loss: 0.3103322684764862 2023-01-23 09:07:49.716012: step: 1714/1541, loss: 0.1128225326538086 2023-01-23 09:07:50.618659: step: 1716/1541, loss: 0.19656962156295776 2023-01-23 09:07:51.516630: step: 1718/1541, loss: 0.20830288529396057 2023-01-23 09:07:52.483610: step: 1720/1541, loss: 0.2050950527191162 2023-01-23 09:07:53.390290: step: 1722/1541, loss: 0.20813225209712982 2023-01-23 09:07:54.285117: step: 1724/1541, loss: 0.15275956690311432 2023-01-23 09:07:55.153623: step: 1726/1541, loss: 0.13620907068252563 2023-01-23 09:07:56.024057: step: 1728/1541, loss: 0.1687186360359192 2023-01-23 09:07:56.984198: step: 1730/1541, loss: 0.44360312819480896 2023-01-23 09:07:57.864135: step: 1732/1541, loss: 0.63536536693573 2023-01-23 09:07:58.735854: step: 1734/1541, loss: 0.0976312905550003 2023-01-23 09:07:59.657499: step: 1736/1541, loss: 0.8556305766105652 2023-01-23 09:08:00.554065: step: 1738/1541, loss: 0.37110695242881775 2023-01-23 09:08:01.471373: step: 1740/1541, loss: 0.16710986196994781 2023-01-23 09:08:02.332362: step: 1742/1541, loss: 0.3528805673122406 2023-01-23 09:08:03.262409: step: 1744/1541, loss: 1.4107189178466797 2023-01-23 09:08:04.172941: step: 1746/1541, loss: 0.06571301072835922 2023-01-23 09:08:05.064509: step: 1748/1541, loss: 0.17893928289413452 2023-01-23 09:08:05.950003: step: 1750/1541, loss: 0.4647967219352722 2023-01-23 09:08:06.826425: step: 1752/1541, loss: 0.26481255888938904 2023-01-23 09:08:07.682914: step: 1754/1541, loss: 0.20460501313209534 2023-01-23 09:08:08.576592: step: 1756/1541, loss: 0.19211503863334656 2023-01-23 09:08:09.437293: step: 1758/1541, loss: 0.12405713647603989 2023-01-23 09:08:10.325979: step: 1760/1541, loss: 0.22577841579914093 2023-01-23 09:08:11.210646: step: 1762/1541, loss: 0.6700439453125 2023-01-23 09:08:12.114825: step: 1764/1541, loss: 0.13650697469711304 2023-01-23 09:08:13.013587: step: 1766/1541, loss: 0.4730009436607361 2023-01-23 09:08:13.877333: step: 1768/1541, loss: 0.34301331639289856 2023-01-23 09:08:14.778831: step: 1770/1541, loss: 0.7214612364768982 2023-01-23 09:08:15.681923: step: 1772/1541, loss: 0.20291444659233093 2023-01-23 09:08:16.612513: step: 1774/1541, loss: 0.5972603559494019 2023-01-23 09:08:17.582398: step: 1776/1541, loss: 0.5399450063705444 2023-01-23 09:08:18.496800: step: 1778/1541, loss: 0.2238154411315918 2023-01-23 09:08:19.387912: step: 1780/1541, loss: 0.657554566860199 2023-01-23 09:08:20.304765: step: 1782/1541, loss: 0.17826101183891296 2023-01-23 09:08:21.211017: step: 1784/1541, loss: 0.3762268126010895 2023-01-23 09:08:22.048267: step: 1786/1541, loss: 0.09318701177835464 2023-01-23 09:08:22.973012: step: 1788/1541, loss: 0.25539645552635193 2023-01-23 09:08:23.846929: step: 1790/1541, loss: 0.2909674048423767 2023-01-23 09:08:24.751125: step: 1792/1541, loss: 0.9560630917549133 2023-01-23 09:08:25.592983: step: 1794/1541, loss: 0.3498843312263489 2023-01-23 09:08:26.452418: step: 1796/1541, loss: 0.3023238182067871 2023-01-23 09:08:27.333848: step: 1798/1541, loss: 0.1859573870897293 2023-01-23 09:08:28.209200: step: 1800/1541, loss: 0.15284587442874908 2023-01-23 09:08:29.087700: step: 1802/1541, loss: 0.08904241025447845 2023-01-23 09:08:29.948800: step: 1804/1541, loss: 0.24499870836734772 2023-01-23 09:08:30.848022: step: 1806/1541, loss: 0.48345571756362915 2023-01-23 09:08:31.757828: step: 1808/1541, loss: 1.455467700958252 2023-01-23 09:08:32.672923: step: 1810/1541, loss: 0.4203127920627594 2023-01-23 09:08:33.526529: step: 1812/1541, loss: 0.1505056619644165 2023-01-23 09:08:34.413612: step: 1814/1541, loss: 0.07236674427986145 2023-01-23 09:08:35.280299: step: 1816/1541, loss: 0.5695028305053711 2023-01-23 09:08:36.160163: step: 1818/1541, loss: 0.20901168882846832 2023-01-23 09:08:37.033125: step: 1820/1541, loss: 0.28404727578163147 2023-01-23 09:08:37.900309: step: 1822/1541, loss: 0.11344753205776215 2023-01-23 09:08:38.814415: step: 1824/1541, loss: 0.5339322090148926 2023-01-23 09:08:39.682834: step: 1826/1541, loss: 0.13400506973266602 2023-01-23 09:08:40.540759: step: 1828/1541, loss: 0.21319210529327393 2023-01-23 09:08:41.444624: step: 1830/1541, loss: 0.8969838619232178 2023-01-23 09:08:42.302543: step: 1832/1541, loss: 0.4505656361579895 2023-01-23 09:08:43.169139: step: 1834/1541, loss: 0.2151792347431183 2023-01-23 09:08:44.091574: step: 1836/1541, loss: 0.3936416208744049 2023-01-23 09:08:45.002368: step: 1838/1541, loss: 0.521467924118042 2023-01-23 09:08:45.930349: step: 1840/1541, loss: 0.20744803547859192 2023-01-23 09:08:46.814789: step: 1842/1541, loss: 1.6572908163070679 2023-01-23 09:08:47.712337: step: 1844/1541, loss: 0.2051784098148346 2023-01-23 09:08:48.634888: step: 1846/1541, loss: 0.19381102919578552 2023-01-23 09:08:49.500136: step: 1848/1541, loss: 0.29657211899757385 2023-01-23 09:08:50.343589: step: 1850/1541, loss: 0.016418440267443657 2023-01-23 09:08:51.271529: step: 1852/1541, loss: 0.538013756275177 2023-01-23 09:08:52.177242: step: 1854/1541, loss: 0.07014857977628708 2023-01-23 09:08:53.059489: step: 1856/1541, loss: 0.18566209077835083 2023-01-23 09:08:53.976923: step: 1858/1541, loss: 0.22616703808307648 2023-01-23 09:08:54.841611: step: 1860/1541, loss: 0.3049183785915375 2023-01-23 09:08:55.691907: step: 1862/1541, loss: 0.009280430153012276 2023-01-23 09:08:56.585415: step: 1864/1541, loss: 0.24498215317726135 2023-01-23 09:08:57.451670: step: 1866/1541, loss: 0.47133758664131165 2023-01-23 09:08:58.308447: step: 1868/1541, loss: 1.9862133264541626 2023-01-23 09:08:59.182736: step: 1870/1541, loss: 0.20481520891189575 2023-01-23 09:09:00.095546: step: 1872/1541, loss: 0.07982293516397476 2023-01-23 09:09:00.947463: step: 1874/1541, loss: 0.23515887558460236 2023-01-23 09:09:01.813023: step: 1876/1541, loss: 0.35351061820983887 2023-01-23 09:09:02.732483: step: 1878/1541, loss: 2.648242950439453 2023-01-23 09:09:03.589236: step: 1880/1541, loss: 0.12746207416057587 2023-01-23 09:09:04.474042: step: 1882/1541, loss: 0.27111324667930603 2023-01-23 09:09:05.313907: step: 1884/1541, loss: 0.23234188556671143 2023-01-23 09:09:06.214103: step: 1886/1541, loss: 0.2190835326910019 2023-01-23 09:09:07.082258: step: 1888/1541, loss: 1.4578757286071777 2023-01-23 09:09:08.014673: step: 1890/1541, loss: 0.284073144197464 2023-01-23 09:09:08.917358: step: 1892/1541, loss: 0.2895537316799164 2023-01-23 09:09:09.799635: step: 1894/1541, loss: 0.021739978343248367 2023-01-23 09:09:10.666709: step: 1896/1541, loss: 0.2158127725124359 2023-01-23 09:09:11.575034: step: 1898/1541, loss: 0.19537357985973358 2023-01-23 09:09:12.440395: step: 1900/1541, loss: 0.15650957822799683 2023-01-23 09:09:13.326462: step: 1902/1541, loss: 0.24004751443862915 2023-01-23 09:09:14.229294: step: 1904/1541, loss: 0.4488022029399872 2023-01-23 09:09:15.109172: step: 1906/1541, loss: 0.33730998635292053 2023-01-23 09:09:15.979310: step: 1908/1541, loss: 0.06693227589130402 2023-01-23 09:09:16.894959: step: 1910/1541, loss: 0.8968831896781921 2023-01-23 09:09:17.791050: step: 1912/1541, loss: 0.36484506726264954 2023-01-23 09:09:18.669996: step: 1914/1541, loss: 0.2051963210105896 2023-01-23 09:09:19.576685: step: 1916/1541, loss: 0.12724177539348602 2023-01-23 09:09:20.443701: step: 1918/1541, loss: 0.13555209338665009 2023-01-23 09:09:21.361163: step: 1920/1541, loss: 0.0961809828877449 2023-01-23 09:09:22.279400: step: 1922/1541, loss: 0.23579111695289612 2023-01-23 09:09:23.210946: step: 1924/1541, loss: 0.11512596905231476 2023-01-23 09:09:24.080556: step: 1926/1541, loss: 0.43141481280326843 2023-01-23 09:09:24.968212: step: 1928/1541, loss: 0.33527904748916626 2023-01-23 09:09:25.885483: step: 1930/1541, loss: 0.43961572647094727 2023-01-23 09:09:26.786320: step: 1932/1541, loss: 0.48678267002105713 2023-01-23 09:09:27.644816: step: 1934/1541, loss: 0.07666558772325516 2023-01-23 09:09:28.543725: step: 1936/1541, loss: 0.20989564061164856 2023-01-23 09:09:29.438838: step: 1938/1541, loss: 0.32551324367523193 2023-01-23 09:09:30.304358: step: 1940/1541, loss: 0.19585387408733368 2023-01-23 09:09:31.170046: step: 1942/1541, loss: 0.3075013756752014 2023-01-23 09:09:32.146085: step: 1944/1541, loss: 0.3008163869380951 2023-01-23 09:09:32.993974: step: 1946/1541, loss: 0.16124066710472107 2023-01-23 09:09:33.890587: step: 1948/1541, loss: 0.3239748179912567 2023-01-23 09:09:34.858464: step: 1950/1541, loss: 0.22689004242420197 2023-01-23 09:09:35.771220: step: 1952/1541, loss: 0.37020057439804077 2023-01-23 09:09:36.674168: step: 1954/1541, loss: 0.15919195115566254 2023-01-23 09:09:37.528935: step: 1956/1541, loss: 0.050450194627046585 2023-01-23 09:09:38.393955: step: 1958/1541, loss: 0.07290790975093842 2023-01-23 09:09:39.282827: step: 1960/1541, loss: 0.13088972866535187 2023-01-23 09:09:40.213305: step: 1962/1541, loss: 0.08899350464344025 2023-01-23 09:09:41.087843: step: 1964/1541, loss: 0.25511181354522705 2023-01-23 09:09:41.976334: step: 1966/1541, loss: 1.0184946060180664 2023-01-23 09:09:42.887918: step: 1968/1541, loss: 0.4438803791999817 2023-01-23 09:09:43.793309: step: 1970/1541, loss: 0.18809741735458374 2023-01-23 09:09:44.686639: step: 1972/1541, loss: 0.06044226139783859 2023-01-23 09:09:45.553468: step: 1974/1541, loss: 0.16691981256008148 2023-01-23 09:09:46.422292: step: 1976/1541, loss: 0.15310253202915192 2023-01-23 09:09:47.288118: step: 1978/1541, loss: 0.15711967647075653 2023-01-23 09:09:48.140988: step: 1980/1541, loss: 0.385616272687912 2023-01-23 09:09:49.094955: step: 1982/1541, loss: 0.5126733183860779 2023-01-23 09:09:50.008485: step: 1984/1541, loss: 0.09630247950553894 2023-01-23 09:09:50.896003: step: 1986/1541, loss: 0.22816576063632965 2023-01-23 09:09:51.791670: step: 1988/1541, loss: 0.1412990242242813 2023-01-23 09:09:52.769091: step: 1990/1541, loss: 0.43201589584350586 2023-01-23 09:09:53.727794: step: 1992/1541, loss: 0.396751344203949 2023-01-23 09:09:54.602996: step: 1994/1541, loss: 1.9251513481140137 2023-01-23 09:09:55.539535: step: 1996/1541, loss: 0.2745111882686615 2023-01-23 09:09:56.436457: step: 1998/1541, loss: 0.0749640241265297 2023-01-23 09:09:57.344311: step: 2000/1541, loss: 0.21891336143016815 2023-01-23 09:09:58.239440: step: 2002/1541, loss: 0.18483759462833405 2023-01-23 09:09:59.134097: step: 2004/1541, loss: 0.0801093801856041 2023-01-23 09:10:00.035772: step: 2006/1541, loss: 0.08496332913637161 2023-01-23 09:10:00.888222: step: 2008/1541, loss: 0.7354941964149475 2023-01-23 09:10:01.863756: step: 2010/1541, loss: 0.26759058237075806 2023-01-23 09:10:02.726755: step: 2012/1541, loss: 0.7614687085151672 2023-01-23 09:10:03.627881: step: 2014/1541, loss: 0.22530913352966309 2023-01-23 09:10:04.496993: step: 2016/1541, loss: 0.5505266189575195 2023-01-23 09:10:05.367756: step: 2018/1541, loss: 0.06213703379034996 2023-01-23 09:10:06.253174: step: 2020/1541, loss: 0.24043568968772888 2023-01-23 09:10:07.159067: step: 2022/1541, loss: 0.16677792370319366 2023-01-23 09:10:08.047331: step: 2024/1541, loss: 0.19095227122306824 2023-01-23 09:10:08.945008: step: 2026/1541, loss: 0.3368196189403534 2023-01-23 09:10:09.840521: step: 2028/1541, loss: 0.23364537954330444 2023-01-23 09:10:10.778290: step: 2030/1541, loss: 0.2632630169391632 2023-01-23 09:10:11.701139: step: 2032/1541, loss: 0.23278504610061646 2023-01-23 09:10:12.594767: step: 2034/1541, loss: 0.184975728392601 2023-01-23 09:10:13.516561: step: 2036/1541, loss: 0.3175411820411682 2023-01-23 09:10:14.413049: step: 2038/1541, loss: 0.2970467209815979 2023-01-23 09:10:15.292850: step: 2040/1541, loss: 1.1266025304794312 2023-01-23 09:10:16.152773: step: 2042/1541, loss: 0.21760134398937225 2023-01-23 09:10:17.009063: step: 2044/1541, loss: 0.055789317935705185 2023-01-23 09:10:17.875973: step: 2046/1541, loss: 0.19083862006664276 2023-01-23 09:10:18.748984: step: 2048/1541, loss: 0.23012124001979828 2023-01-23 09:10:19.624556: step: 2050/1541, loss: 0.17022539675235748 2023-01-23 09:10:20.497909: step: 2052/1541, loss: 0.1464495062828064 2023-01-23 09:10:21.347128: step: 2054/1541, loss: 0.1371002197265625 2023-01-23 09:10:22.227122: step: 2056/1541, loss: 0.18116825819015503 2023-01-23 09:10:23.061344: step: 2058/1541, loss: 0.1453987956047058 2023-01-23 09:10:23.954800: step: 2060/1541, loss: 0.09382222592830658 2023-01-23 09:10:24.820098: step: 2062/1541, loss: 1.2501182556152344 2023-01-23 09:10:25.712124: step: 2064/1541, loss: 0.14983129501342773 2023-01-23 09:10:26.596999: step: 2066/1541, loss: 0.26943206787109375 2023-01-23 09:10:27.477167: step: 2068/1541, loss: 0.21857410669326782 2023-01-23 09:10:28.354804: step: 2070/1541, loss: 0.16446933150291443 2023-01-23 09:10:29.257777: step: 2072/1541, loss: 0.21540264785289764 2023-01-23 09:10:30.141816: step: 2074/1541, loss: 0.1385381668806076 2023-01-23 09:10:31.053499: step: 2076/1541, loss: 0.21847085654735565 2023-01-23 09:10:31.957389: step: 2078/1541, loss: 0.11215614527463913 2023-01-23 09:10:32.823788: step: 2080/1541, loss: 0.34375742077827454 2023-01-23 09:10:33.753148: step: 2082/1541, loss: 0.25590553879737854 2023-01-23 09:10:34.680861: step: 2084/1541, loss: 0.20048271119594574 2023-01-23 09:10:35.601825: step: 2086/1541, loss: 0.03584030643105507 2023-01-23 09:10:36.476794: step: 2088/1541, loss: 0.168910413980484 2023-01-23 09:10:37.366753: step: 2090/1541, loss: 0.04654385894536972 2023-01-23 09:10:38.268033: step: 2092/1541, loss: 0.278812974691391 2023-01-23 09:10:39.127141: step: 2094/1541, loss: 0.12919223308563232 2023-01-23 09:10:40.079548: step: 2096/1541, loss: 0.5965428948402405 2023-01-23 09:10:40.963656: step: 2098/1541, loss: 0.4028666019439697 2023-01-23 09:10:41.845920: step: 2100/1541, loss: 0.42224353551864624 2023-01-23 09:10:42.707667: step: 2102/1541, loss: 3.074824094772339 2023-01-23 09:10:43.653745: step: 2104/1541, loss: 0.24831849336624146 2023-01-23 09:10:44.577915: step: 2106/1541, loss: 0.43793153762817383 2023-01-23 09:10:45.405629: step: 2108/1541, loss: 0.05542680621147156 2023-01-23 09:10:46.303381: step: 2110/1541, loss: 0.2446688711643219 2023-01-23 09:10:47.184163: step: 2112/1541, loss: 0.4026852548122406 2023-01-23 09:10:48.083426: step: 2114/1541, loss: 0.9338136315345764 2023-01-23 09:10:48.971203: step: 2116/1541, loss: 0.471125066280365 2023-01-23 09:10:49.839795: step: 2118/1541, loss: 0.05813908576965332 2023-01-23 09:10:50.723392: step: 2120/1541, loss: 0.364787220954895 2023-01-23 09:10:51.601549: step: 2122/1541, loss: 0.5048458576202393 2023-01-23 09:10:52.522167: step: 2124/1541, loss: 0.16210903227329254 2023-01-23 09:10:53.526463: step: 2126/1541, loss: 0.2042125165462494 2023-01-23 09:10:54.388725: step: 2128/1541, loss: 0.1784154325723648 2023-01-23 09:10:55.277094: step: 2130/1541, loss: 0.3925095796585083 2023-01-23 09:10:56.139276: step: 2132/1541, loss: 0.437311589717865 2023-01-23 09:10:57.047590: step: 2134/1541, loss: 0.3132435381412506 2023-01-23 09:10:57.961469: step: 2136/1541, loss: 0.44780999422073364 2023-01-23 09:10:58.834949: step: 2138/1541, loss: 0.20827755331993103 2023-01-23 09:10:59.690699: step: 2140/1541, loss: 0.1733761727809906 2023-01-23 09:11:00.598761: step: 2142/1541, loss: 0.08732642233371735 2023-01-23 09:11:01.498926: step: 2144/1541, loss: 0.3530410826206207 2023-01-23 09:11:02.369369: step: 2146/1541, loss: 0.46422046422958374 2023-01-23 09:11:03.263654: step: 2148/1541, loss: 3.3724184036254883 2023-01-23 09:11:04.183956: step: 2150/1541, loss: 0.8506219983100891 2023-01-23 09:11:05.072256: step: 2152/1541, loss: 1.6042686700820923 2023-01-23 09:11:06.010737: step: 2154/1541, loss: 0.24072995781898499 2023-01-23 09:11:06.907480: step: 2156/1541, loss: 0.4191143810749054 2023-01-23 09:11:07.829386: step: 2158/1541, loss: 0.6291268467903137 2023-01-23 09:11:08.696166: step: 2160/1541, loss: 0.2990776598453522 2023-01-23 09:11:09.603572: step: 2162/1541, loss: 0.3367736339569092 2023-01-23 09:11:10.513696: step: 2164/1541, loss: 0.16761262714862823 2023-01-23 09:11:11.404432: step: 2166/1541, loss: 0.40945208072662354 2023-01-23 09:11:12.295551: step: 2168/1541, loss: 0.8822054862976074 2023-01-23 09:11:13.188628: step: 2170/1541, loss: 0.18861033022403717 2023-01-23 09:11:14.041898: step: 2172/1541, loss: 0.12632562220096588 2023-01-23 09:11:14.918775: step: 2174/1541, loss: 0.11271718144416809 2023-01-23 09:11:15.820372: step: 2176/1541, loss: 0.13071954250335693 2023-01-23 09:11:16.733805: step: 2178/1541, loss: 0.04493161663413048 2023-01-23 09:11:17.649735: step: 2180/1541, loss: 0.24023652076721191 2023-01-23 09:11:18.609345: step: 2182/1541, loss: 0.039995111525058746 2023-01-23 09:11:19.460027: step: 2184/1541, loss: 0.21362626552581787 2023-01-23 09:11:20.400429: step: 2186/1541, loss: 0.2738499343395233 2023-01-23 09:11:21.323752: step: 2188/1541, loss: 0.8678939938545227 2023-01-23 09:11:22.162503: step: 2190/1541, loss: 0.05729668214917183 2023-01-23 09:11:23.083085: step: 2192/1541, loss: 0.45816177129745483 2023-01-23 09:11:23.983074: step: 2194/1541, loss: 0.4072459638118744 2023-01-23 09:11:24.852513: step: 2196/1541, loss: 0.1978493332862854 2023-01-23 09:11:25.766649: step: 2198/1541, loss: 0.15298864245414734 2023-01-23 09:11:26.685540: step: 2200/1541, loss: 0.3719974756240845 2023-01-23 09:11:27.538475: step: 2202/1541, loss: 0.8863948583602905 2023-01-23 09:11:28.440603: step: 2204/1541, loss: 0.4419066905975342 2023-01-23 09:11:29.352409: step: 2206/1541, loss: 0.37761810421943665 2023-01-23 09:11:30.236881: step: 2208/1541, loss: 0.23988905549049377 2023-01-23 09:11:31.128860: step: 2210/1541, loss: 0.09192762523889542 2023-01-23 09:11:32.023301: step: 2212/1541, loss: 1.619320034980774 2023-01-23 09:11:32.947062: step: 2214/1541, loss: 0.8991813063621521 2023-01-23 09:11:33.845260: step: 2216/1541, loss: 0.2581828534603119 2023-01-23 09:11:34.750489: step: 2218/1541, loss: 0.4804997146129608 2023-01-23 09:11:35.624402: step: 2220/1541, loss: 0.13264307379722595 2023-01-23 09:11:36.478893: step: 2222/1541, loss: 0.16173705458641052 2023-01-23 09:11:37.340024: step: 2224/1541, loss: 0.30312487483024597 2023-01-23 09:11:38.203216: step: 2226/1541, loss: 0.26233336329460144 2023-01-23 09:11:39.083101: step: 2228/1541, loss: 0.2009860873222351 2023-01-23 09:11:40.031080: step: 2230/1541, loss: 0.24641132354736328 2023-01-23 09:11:40.926078: step: 2232/1541, loss: 0.5081836581230164 2023-01-23 09:11:41.817962: step: 2234/1541, loss: 0.374525249004364 2023-01-23 09:11:42.720252: step: 2236/1541, loss: 0.21878555417060852 2023-01-23 09:11:43.580847: step: 2238/1541, loss: 0.12203916907310486 2023-01-23 09:11:44.507391: step: 2240/1541, loss: 0.16830554604530334 2023-01-23 09:11:45.387212: step: 2242/1541, loss: 0.17766843736171722 2023-01-23 09:11:46.225195: step: 2244/1541, loss: 0.12424632161855698 2023-01-23 09:11:47.091719: step: 2246/1541, loss: 0.19852185249328613 2023-01-23 09:11:47.992796: step: 2248/1541, loss: 0.6301100254058838 2023-01-23 09:11:48.917309: step: 2250/1541, loss: 0.23144643008708954 2023-01-23 09:11:49.814924: step: 2252/1541, loss: 1.7081663608551025 2023-01-23 09:11:50.716863: step: 2254/1541, loss: 0.14617688953876495 2023-01-23 09:11:51.688104: step: 2256/1541, loss: 0.13831296563148499 2023-01-23 09:11:52.578503: step: 2258/1541, loss: 0.21111565828323364 2023-01-23 09:11:53.481666: step: 2260/1541, loss: 0.16069379448890686 2023-01-23 09:11:54.372217: step: 2262/1541, loss: 0.3696594536304474 2023-01-23 09:11:55.273632: step: 2264/1541, loss: 0.28170377016067505 2023-01-23 09:11:56.171695: step: 2266/1541, loss: 0.023678649216890335 2023-01-23 09:11:57.051653: step: 2268/1541, loss: 0.40038177371025085 2023-01-23 09:11:57.926420: step: 2270/1541, loss: 0.6098918914794922 2023-01-23 09:11:58.840664: step: 2272/1541, loss: 0.45533159375190735 2023-01-23 09:11:59.715314: step: 2274/1541, loss: 0.46711471676826477 2023-01-23 09:12:00.593968: step: 2276/1541, loss: 0.2501513957977295 2023-01-23 09:12:01.480771: step: 2278/1541, loss: 0.16798731684684753 2023-01-23 09:12:02.468941: step: 2280/1541, loss: 0.04170709848403931 2023-01-23 09:12:03.346179: step: 2282/1541, loss: 0.01448519341647625 2023-01-23 09:12:04.241988: step: 2284/1541, loss: 0.17489536106586456 2023-01-23 09:12:05.113438: step: 2286/1541, loss: 0.41862961649894714 2023-01-23 09:12:05.970007: step: 2288/1541, loss: 0.5044803023338318 2023-01-23 09:12:06.845629: step: 2290/1541, loss: 0.15603584051132202 2023-01-23 09:12:07.701577: step: 2292/1541, loss: 0.08861694484949112 2023-01-23 09:12:08.572529: step: 2294/1541, loss: 0.7632233500480652 2023-01-23 09:12:09.431680: step: 2296/1541, loss: 0.07919758558273315 2023-01-23 09:12:10.307098: step: 2298/1541, loss: 0.023161137476563454 2023-01-23 09:12:11.197612: step: 2300/1541, loss: 0.265720009803772 2023-01-23 09:12:12.079777: step: 2302/1541, loss: 0.6487032175064087 2023-01-23 09:12:12.943726: step: 2304/1541, loss: 0.40289708971977234 2023-01-23 09:12:13.943537: step: 2306/1541, loss: 0.2848949134349823 2023-01-23 09:12:14.886702: step: 2308/1541, loss: 6.51893424987793 2023-01-23 09:12:15.889643: step: 2310/1541, loss: 0.2656910717487335 2023-01-23 09:12:16.756816: step: 2312/1541, loss: 0.20525343716144562 2023-01-23 09:12:17.631448: step: 2314/1541, loss: 0.23521703481674194 2023-01-23 09:12:18.524521: step: 2316/1541, loss: 0.4065929055213928 2023-01-23 09:12:19.430740: step: 2318/1541, loss: 0.1769087314605713 2023-01-23 09:12:20.335944: step: 2320/1541, loss: 0.05008334666490555 2023-01-23 09:12:21.265990: step: 2322/1541, loss: 0.12437966465950012 2023-01-23 09:12:22.119708: step: 2324/1541, loss: 0.34315404295921326 2023-01-23 09:12:23.036003: step: 2326/1541, loss: 0.0442512147128582 2023-01-23 09:12:23.911168: step: 2328/1541, loss: 0.5826465487480164 2023-01-23 09:12:24.790385: step: 2330/1541, loss: 0.025415968149900436 2023-01-23 09:12:25.663918: step: 2332/1541, loss: 0.04355845972895622 2023-01-23 09:12:26.569014: step: 2334/1541, loss: 0.7111067175865173 2023-01-23 09:12:27.455912: step: 2336/1541, loss: 0.7149801254272461 2023-01-23 09:12:28.315351: step: 2338/1541, loss: 0.7351207137107849 2023-01-23 09:12:29.248365: step: 2340/1541, loss: 0.23256511986255646 2023-01-23 09:12:30.141237: step: 2342/1541, loss: 0.3679976761341095 2023-01-23 09:12:31.088518: step: 2344/1541, loss: 0.60556960105896 2023-01-23 09:12:31.971127: step: 2346/1541, loss: 0.6011631488800049 2023-01-23 09:12:32.854607: step: 2348/1541, loss: 0.20943695306777954 2023-01-23 09:12:33.754980: step: 2350/1541, loss: 0.5674219727516174 2023-01-23 09:12:34.610700: step: 2352/1541, loss: 2.9201297760009766 2023-01-23 09:12:35.455961: step: 2354/1541, loss: 0.10205285996198654 2023-01-23 09:12:36.332918: step: 2356/1541, loss: 0.5187875032424927 2023-01-23 09:12:37.207791: step: 2358/1541, loss: 0.24536103010177612 2023-01-23 09:12:38.090295: step: 2360/1541, loss: 1.9952980279922485 2023-01-23 09:12:39.062377: step: 2362/1541, loss: 0.6396872997283936 2023-01-23 09:12:39.972551: step: 2364/1541, loss: 0.17067544162273407 2023-01-23 09:12:40.839024: step: 2366/1541, loss: 0.07982224225997925 2023-01-23 09:12:41.751345: step: 2368/1541, loss: 0.12915587425231934 2023-01-23 09:12:42.634534: step: 2370/1541, loss: 0.2300691157579422 2023-01-23 09:12:43.561333: step: 2372/1541, loss: 0.1714499145746231 2023-01-23 09:12:44.449330: step: 2374/1541, loss: 0.377383291721344 2023-01-23 09:12:45.301867: step: 2376/1541, loss: 0.19157497584819794 2023-01-23 09:12:46.179985: step: 2378/1541, loss: 0.1488531529903412 2023-01-23 09:12:47.089346: step: 2380/1541, loss: 0.520780086517334 2023-01-23 09:12:48.009301: step: 2382/1541, loss: 0.2041529417037964 2023-01-23 09:12:48.894665: step: 2384/1541, loss: 0.3295779228210449 2023-01-23 09:12:49.805173: step: 2386/1541, loss: 0.36834537982940674 2023-01-23 09:12:50.747709: step: 2388/1541, loss: 0.9380020499229431 2023-01-23 09:12:51.605978: step: 2390/1541, loss: 0.05024319142103195 2023-01-23 09:12:52.601599: step: 2392/1541, loss: 0.0636867806315422 2023-01-23 09:12:53.551694: step: 2394/1541, loss: 0.11656740307807922 2023-01-23 09:12:54.509778: step: 2396/1541, loss: 0.25592124462127686 2023-01-23 09:12:55.436897: step: 2398/1541, loss: 0.16037817299365997 2023-01-23 09:12:56.315141: step: 2400/1541, loss: 0.09645339846611023 2023-01-23 09:12:57.174004: step: 2402/1541, loss: 0.31400811672210693 2023-01-23 09:12:58.051678: step: 2404/1541, loss: 0.2516227066516876 2023-01-23 09:12:58.929066: step: 2406/1541, loss: 0.1599937081336975 2023-01-23 09:12:59.785127: step: 2408/1541, loss: 0.2998466491699219 2023-01-23 09:13:00.657989: step: 2410/1541, loss: 0.1105344220995903 2023-01-23 09:13:01.541203: step: 2412/1541, loss: 0.2872244715690613 2023-01-23 09:13:02.454430: step: 2414/1541, loss: 0.16467714309692383 2023-01-23 09:13:03.373327: step: 2416/1541, loss: 0.12887562811374664 2023-01-23 09:13:04.246488: step: 2418/1541, loss: 0.349771112203598 2023-01-23 09:13:05.123297: step: 2420/1541, loss: 0.05565398931503296 2023-01-23 09:13:06.057987: step: 2422/1541, loss: 0.237355038523674 2023-01-23 09:13:06.935010: step: 2424/1541, loss: 0.4901267886161804 2023-01-23 09:13:07.799585: step: 2426/1541, loss: 0.1553451269865036 2023-01-23 09:13:08.741551: step: 2428/1541, loss: 0.3482005298137665 2023-01-23 09:13:09.660721: step: 2430/1541, loss: 0.3066772222518921 2023-01-23 09:13:10.507298: step: 2432/1541, loss: 0.5300337076187134 2023-01-23 09:13:11.452176: step: 2434/1541, loss: 0.37878894805908203 2023-01-23 09:13:12.333662: step: 2436/1541, loss: 0.3578142821788788 2023-01-23 09:13:13.170496: step: 2438/1541, loss: 0.08005336672067642 2023-01-23 09:13:14.066440: step: 2440/1541, loss: 0.11576230823993683 2023-01-23 09:13:14.937973: step: 2442/1541, loss: 0.07539256662130356 2023-01-23 09:13:15.799461: step: 2444/1541, loss: 0.060289837419986725 2023-01-23 09:13:16.715881: step: 2446/1541, loss: 0.42832016944885254 2023-01-23 09:13:17.609928: step: 2448/1541, loss: 0.29067039489746094 2023-01-23 09:13:18.520016: step: 2450/1541, loss: 0.288535475730896 2023-01-23 09:13:19.421458: step: 2452/1541, loss: 0.0975586473941803 2023-01-23 09:13:20.349343: step: 2454/1541, loss: 4.516404151916504 2023-01-23 09:13:21.234806: step: 2456/1541, loss: 0.1466895192861557 2023-01-23 09:13:22.128951: step: 2458/1541, loss: 0.2665511667728424 2023-01-23 09:13:23.011153: step: 2460/1541, loss: 0.15944379568099976 2023-01-23 09:13:23.889487: step: 2462/1541, loss: 0.40491169691085815 2023-01-23 09:13:24.739070: step: 2464/1541, loss: 0.24316924810409546 2023-01-23 09:13:25.606584: step: 2466/1541, loss: 0.2815352976322174 2023-01-23 09:13:26.541476: step: 2468/1541, loss: 0.6819792985916138 2023-01-23 09:13:27.495813: step: 2470/1541, loss: 0.21136321127414703 2023-01-23 09:13:28.456060: step: 2472/1541, loss: 0.16844795644283295 2023-01-23 09:13:29.478566: step: 2474/1541, loss: 0.45965588092803955 2023-01-23 09:13:30.372556: step: 2476/1541, loss: 0.4765174388885498 2023-01-23 09:13:31.268870: step: 2478/1541, loss: 0.19260558485984802 2023-01-23 09:13:32.208782: step: 2480/1541, loss: 0.2994953989982605 2023-01-23 09:13:33.110147: step: 2482/1541, loss: 0.4688156545162201 2023-01-23 09:13:34.008299: step: 2484/1541, loss: 0.14383617043495178 2023-01-23 09:13:34.948164: step: 2486/1541, loss: 0.660780668258667 2023-01-23 09:13:35.823920: step: 2488/1541, loss: 0.14811848104000092 2023-01-23 09:13:36.706957: step: 2490/1541, loss: 0.05201767385005951 2023-01-23 09:13:37.676998: step: 2492/1541, loss: 0.23530830442905426 2023-01-23 09:13:38.578945: step: 2494/1541, loss: 0.14330437779426575 2023-01-23 09:13:39.465402: step: 2496/1541, loss: 0.39433905482292175 2023-01-23 09:13:40.383542: step: 2498/1541, loss: 0.4840487837791443 2023-01-23 09:13:41.239546: step: 2500/1541, loss: 0.12503403425216675 2023-01-23 09:13:42.139191: step: 2502/1541, loss: 0.8084648847579956 2023-01-23 09:13:43.069130: step: 2504/1541, loss: 0.2352672666311264 2023-01-23 09:13:43.977536: step: 2506/1541, loss: 0.0929010808467865 2023-01-23 09:13:44.836964: step: 2508/1541, loss: 0.4021932780742645 2023-01-23 09:13:45.702302: step: 2510/1541, loss: 0.6720964908599854 2023-01-23 09:13:46.593495: step: 2512/1541, loss: 0.1661195009946823 2023-01-23 09:13:47.462826: step: 2514/1541, loss: 0.4847918152809143 2023-01-23 09:13:48.349113: step: 2516/1541, loss: 0.3176623582839966 2023-01-23 09:13:49.224518: step: 2518/1541, loss: 0.34325945377349854 2023-01-23 09:13:50.094614: step: 2520/1541, loss: 2.8082640171051025 2023-01-23 09:13:50.985030: step: 2522/1541, loss: 0.10347620397806168 2023-01-23 09:13:51.854041: step: 2524/1541, loss: 0.035432543605566025 2023-01-23 09:13:52.797554: step: 2526/1541, loss: 0.2105982005596161 2023-01-23 09:13:53.694647: step: 2528/1541, loss: 0.15435576438903809 2023-01-23 09:13:54.562642: step: 2530/1541, loss: 0.4350910484790802 2023-01-23 09:13:55.483759: step: 2532/1541, loss: 5.737583637237549 2023-01-23 09:13:56.318270: step: 2534/1541, loss: 0.07745540887117386 2023-01-23 09:13:57.175761: step: 2536/1541, loss: 0.8459224700927734 2023-01-23 09:13:58.068765: step: 2538/1541, loss: 1.1537723541259766 2023-01-23 09:13:58.947962: step: 2540/1541, loss: 0.05733073130249977 2023-01-23 09:13:59.834250: step: 2542/1541, loss: 0.6201778650283813 2023-01-23 09:14:00.689993: step: 2544/1541, loss: 0.5771917700767517 2023-01-23 09:14:01.616906: step: 2546/1541, loss: 0.1853787899017334 2023-01-23 09:14:02.614263: step: 2548/1541, loss: 0.14852455258369446 2023-01-23 09:14:03.617544: step: 2550/1541, loss: 0.7682462334632874 2023-01-23 09:14:04.488362: step: 2552/1541, loss: 0.10311951488256454 2023-01-23 09:14:05.362743: step: 2554/1541, loss: 0.23586313426494598 2023-01-23 09:14:06.217160: step: 2556/1541, loss: 0.4680333435535431 2023-01-23 09:14:07.093019: step: 2558/1541, loss: 0.20509198307991028 2023-01-23 09:14:07.969890: step: 2560/1541, loss: 0.18400262296199799 2023-01-23 09:14:08.880337: step: 2562/1541, loss: 0.15447023510932922 2023-01-23 09:14:09.770413: step: 2564/1541, loss: 0.5834512710571289 2023-01-23 09:14:10.685194: step: 2566/1541, loss: 0.808861494064331 2023-01-23 09:14:11.570017: step: 2568/1541, loss: 0.574787974357605 2023-01-23 09:14:12.476101: step: 2570/1541, loss: 0.15409083664417267 2023-01-23 09:14:13.331513: step: 2572/1541, loss: 0.1078348159790039 2023-01-23 09:14:14.218878: step: 2574/1541, loss: 0.3307148218154907 2023-01-23 09:14:15.214282: step: 2576/1541, loss: 0.1771690845489502 2023-01-23 09:14:16.109531: step: 2578/1541, loss: 0.2031843513250351 2023-01-23 09:14:17.050083: step: 2580/1541, loss: 0.48766106367111206 2023-01-23 09:14:17.936592: step: 2582/1541, loss: 0.6463031768798828 2023-01-23 09:14:18.873092: step: 2584/1541, loss: 0.41723424196243286 2023-01-23 09:14:19.722175: step: 2586/1541, loss: 0.3314477801322937 2023-01-23 09:14:20.583488: step: 2588/1541, loss: 0.37270388007164 2023-01-23 09:14:21.395768: step: 2590/1541, loss: 0.06828682124614716 2023-01-23 09:14:22.274563: step: 2592/1541, loss: 0.19847610592842102 2023-01-23 09:14:23.123442: step: 2594/1541, loss: 0.1796649992465973 2023-01-23 09:14:24.029850: step: 2596/1541, loss: 0.4197980761528015 2023-01-23 09:14:24.906227: step: 2598/1541, loss: 0.07867169380187988 2023-01-23 09:14:25.842887: step: 2600/1541, loss: 0.1513620913028717 2023-01-23 09:14:26.736427: step: 2602/1541, loss: 0.10024703294038773 2023-01-23 09:14:27.603166: step: 2604/1541, loss: 0.5506578683853149 2023-01-23 09:14:28.485308: step: 2606/1541, loss: 0.10536038875579834 2023-01-23 09:14:29.421199: step: 2608/1541, loss: 0.21679756045341492 2023-01-23 09:14:30.365986: step: 2610/1541, loss: 0.47749796509742737 2023-01-23 09:14:31.265051: step: 2612/1541, loss: 1.2942805290222168 2023-01-23 09:14:32.142042: step: 2614/1541, loss: 0.1911001205444336 2023-01-23 09:14:33.016773: step: 2616/1541, loss: 0.20678021013736725 2023-01-23 09:14:33.889553: step: 2618/1541, loss: 0.06370878964662552 2023-01-23 09:14:34.742646: step: 2620/1541, loss: 0.0194966159760952 2023-01-23 09:14:35.629857: step: 2622/1541, loss: 0.25799867510795593 2023-01-23 09:14:36.533342: step: 2624/1541, loss: 0.23774340748786926 2023-01-23 09:14:37.447830: step: 2626/1541, loss: 0.6164515018463135 2023-01-23 09:14:38.337996: step: 2628/1541, loss: 0.3202398121356964 2023-01-23 09:14:39.251924: step: 2630/1541, loss: 0.2030401974916458 2023-01-23 09:14:40.163191: step: 2632/1541, loss: 0.2150663584470749 2023-01-23 09:14:41.001424: step: 2634/1541, loss: 0.2187080681324005 2023-01-23 09:14:41.869011: step: 2636/1541, loss: 0.2841675281524658 2023-01-23 09:14:42.726688: step: 2638/1541, loss: 0.323889821767807 2023-01-23 09:14:43.618458: step: 2640/1541, loss: 0.161396786570549 2023-01-23 09:14:44.478835: step: 2642/1541, loss: 0.3817737102508545 2023-01-23 09:14:45.415778: step: 2644/1541, loss: 0.3766334354877472 2023-01-23 09:14:46.316894: step: 2646/1541, loss: 0.14684870839118958 2023-01-23 09:14:47.293892: step: 2648/1541, loss: 0.2236715406179428 2023-01-23 09:14:48.179619: step: 2650/1541, loss: 0.43384024500846863 2023-01-23 09:14:49.074996: step: 2652/1541, loss: 0.677995502948761 2023-01-23 09:14:49.998802: step: 2654/1541, loss: 5.051350116729736 2023-01-23 09:14:50.868840: step: 2656/1541, loss: 0.35234472155570984 2023-01-23 09:14:51.772060: step: 2658/1541, loss: 0.3274671137332916 2023-01-23 09:14:52.640740: step: 2660/1541, loss: 0.28917640447616577 2023-01-23 09:14:53.559050: step: 2662/1541, loss: 0.14351120591163635 2023-01-23 09:14:54.476110: step: 2664/1541, loss: 0.5229008197784424 2023-01-23 09:14:55.383445: step: 2666/1541, loss: 0.06337198615074158 2023-01-23 09:14:56.334452: step: 2668/1541, loss: 0.38258060812950134 2023-01-23 09:14:57.260317: step: 2670/1541, loss: 0.49077656865119934 2023-01-23 09:14:58.122525: step: 2672/1541, loss: 0.5808050632476807 2023-01-23 09:14:59.056216: step: 2674/1541, loss: 0.28445112705230713 2023-01-23 09:14:59.927843: step: 2676/1541, loss: 0.13654828071594238 2023-01-23 09:15:00.779297: step: 2678/1541, loss: 0.4073385000228882 2023-01-23 09:15:01.745872: step: 2680/1541, loss: 0.7030133008956909 2023-01-23 09:15:02.641847: step: 2682/1541, loss: 0.19077840447425842 2023-01-23 09:15:03.529652: step: 2684/1541, loss: 0.2896243929862976 2023-01-23 09:15:04.590095: step: 2686/1541, loss: 0.44073814153671265 2023-01-23 09:15:05.554132: step: 2688/1541, loss: 2.420675277709961 2023-01-23 09:15:06.442088: step: 2690/1541, loss: 0.3099508285522461 2023-01-23 09:15:07.344981: step: 2692/1541, loss: 0.4153916835784912 2023-01-23 09:15:08.253955: step: 2694/1541, loss: 1.1942065954208374 2023-01-23 09:15:09.152593: step: 2696/1541, loss: 0.10212957113981247 2023-01-23 09:15:10.049487: step: 2698/1541, loss: 0.6741876602172852 2023-01-23 09:15:10.910723: step: 2700/1541, loss: 0.35176485776901245 2023-01-23 09:15:11.796030: step: 2702/1541, loss: 0.6200026273727417 2023-01-23 09:15:12.658015: step: 2704/1541, loss: 0.12848897278308868 2023-01-23 09:15:13.539026: step: 2706/1541, loss: 0.17578034102916718 2023-01-23 09:15:14.407184: step: 2708/1541, loss: 0.3294285237789154 2023-01-23 09:15:15.327546: step: 2710/1541, loss: 0.30779823660850525 2023-01-23 09:15:16.200235: step: 2712/1541, loss: 0.21615399420261383 2023-01-23 09:15:17.073425: step: 2714/1541, loss: 0.3461429476737976 2023-01-23 09:15:17.972783: step: 2716/1541, loss: 0.47559654712677 2023-01-23 09:15:18.861593: step: 2718/1541, loss: 0.14621871709823608 2023-01-23 09:15:19.712935: step: 2720/1541, loss: 0.13280998170375824 2023-01-23 09:15:20.654237: step: 2722/1541, loss: 0.913946270942688 2023-01-23 09:15:21.515924: step: 2724/1541, loss: 0.2321770042181015 2023-01-23 09:15:22.410491: step: 2726/1541, loss: 0.1963253915309906 2023-01-23 09:15:23.298353: step: 2728/1541, loss: 0.803600013256073 2023-01-23 09:15:24.242272: step: 2730/1541, loss: 0.15790820121765137 2023-01-23 09:15:25.221230: step: 2732/1541, loss: 0.38159167766571045 2023-01-23 09:15:26.104971: step: 2734/1541, loss: 1.169169545173645 2023-01-23 09:15:27.027726: step: 2736/1541, loss: 0.3470412492752075 2023-01-23 09:15:27.914991: step: 2738/1541, loss: 0.09738810360431671 2023-01-23 09:15:28.809580: step: 2740/1541, loss: 0.22061413526535034 2023-01-23 09:15:29.678741: step: 2742/1541, loss: 0.7989923357963562 2023-01-23 09:15:30.574988: step: 2744/1541, loss: 0.3861831724643707 2023-01-23 09:15:31.429260: step: 2746/1541, loss: 0.2654147148132324 2023-01-23 09:15:32.326775: step: 2748/1541, loss: 0.5741081237792969 2023-01-23 09:15:33.198794: step: 2750/1541, loss: 0.18891552090644836 2023-01-23 09:15:34.073796: step: 2752/1541, loss: 0.17700476944446564 2023-01-23 09:15:34.933697: step: 2754/1541, loss: 0.16117407381534576 2023-01-23 09:15:35.788661: step: 2756/1541, loss: 0.2812223434448242 2023-01-23 09:15:36.655155: step: 2758/1541, loss: 0.24237419664859772 2023-01-23 09:15:37.602045: step: 2760/1541, loss: 0.07996194809675217 2023-01-23 09:15:38.526050: step: 2762/1541, loss: 0.37924647331237793 2023-01-23 09:15:39.406611: step: 2764/1541, loss: 0.24488775432109833 2023-01-23 09:15:40.293075: step: 2766/1541, loss: 0.3890034258365631 2023-01-23 09:15:41.207039: step: 2768/1541, loss: 0.42322152853012085 2023-01-23 09:15:42.072264: step: 2770/1541, loss: 0.09200447052717209 2023-01-23 09:15:42.960977: step: 2772/1541, loss: 0.7180778384208679 2023-01-23 09:15:43.850333: step: 2774/1541, loss: 0.11019064486026764 2023-01-23 09:15:44.739307: step: 2776/1541, loss: 0.04365178942680359 2023-01-23 09:15:45.641221: step: 2778/1541, loss: 0.332001656293869 2023-01-23 09:15:46.481916: step: 2780/1541, loss: 0.680924117565155 2023-01-23 09:15:47.434824: step: 2782/1541, loss: 0.14638057351112366 2023-01-23 09:15:48.306117: step: 2784/1541, loss: 0.11132407933473587 2023-01-23 09:15:49.228530: step: 2786/1541, loss: 0.7168728709220886 2023-01-23 09:15:50.131821: step: 2788/1541, loss: 0.34945762157440186 2023-01-23 09:15:51.136941: step: 2790/1541, loss: 0.14809376001358032 2023-01-23 09:15:52.019017: step: 2792/1541, loss: 0.76646888256073 2023-01-23 09:15:52.927096: step: 2794/1541, loss: 0.36552178859710693 2023-01-23 09:15:53.842345: step: 2796/1541, loss: 0.5123617053031921 2023-01-23 09:15:54.771714: step: 2798/1541, loss: 0.5222800970077515 2023-01-23 09:15:55.647311: step: 2800/1541, loss: 0.33372098207473755 2023-01-23 09:15:56.550960: step: 2802/1541, loss: 0.6011630296707153 2023-01-23 09:15:57.434554: step: 2804/1541, loss: 0.41256752610206604 2023-01-23 09:15:58.332484: step: 2806/1541, loss: 0.15015873312950134 2023-01-23 09:15:59.190193: step: 2808/1541, loss: 2.388155937194824 2023-01-23 09:16:00.079615: step: 2810/1541, loss: 0.6652228832244873 2023-01-23 09:16:01.026761: step: 2812/1541, loss: 0.11077040433883667 2023-01-23 09:16:01.937345: step: 2814/1541, loss: 1.2276757955551147 2023-01-23 09:16:02.824251: step: 2816/1541, loss: 0.25512367486953735 2023-01-23 09:16:03.712338: step: 2818/1541, loss: 0.3672495186328888 2023-01-23 09:16:04.635026: step: 2820/1541, loss: 0.18039683997631073 2023-01-23 09:16:05.503354: step: 2822/1541, loss: 0.37588298320770264 2023-01-23 09:16:06.372372: step: 2824/1541, loss: 0.22103293240070343 2023-01-23 09:16:07.252530: step: 2826/1541, loss: 1.5382379293441772 2023-01-23 09:16:08.149655: step: 2828/1541, loss: 0.18759629130363464 2023-01-23 09:16:09.017308: step: 2830/1541, loss: 0.438467800617218 2023-01-23 09:16:09.915919: step: 2832/1541, loss: 0.29191604256629944 2023-01-23 09:16:10.790121: step: 2834/1541, loss: 0.5865563154220581 2023-01-23 09:16:11.649255: step: 2836/1541, loss: 0.06864462047815323 2023-01-23 09:16:12.520583: step: 2838/1541, loss: 2.523850679397583 2023-01-23 09:16:13.391918: step: 2840/1541, loss: 0.39849650859832764 2023-01-23 09:16:14.256789: step: 2842/1541, loss: 0.2092827409505844 2023-01-23 09:16:15.224993: step: 2844/1541, loss: 0.11749592423439026 2023-01-23 09:16:16.101364: step: 2846/1541, loss: 0.13906921446323395 2023-01-23 09:16:16.991162: step: 2848/1541, loss: 0.13803310692310333 2023-01-23 09:16:17.867544: step: 2850/1541, loss: 0.08551476150751114 2023-01-23 09:16:18.747203: step: 2852/1541, loss: 0.3332498073577881 2023-01-23 09:16:19.616351: step: 2854/1541, loss: 0.12107348442077637 2023-01-23 09:16:20.502077: step: 2856/1541, loss: 0.13114136457443237 2023-01-23 09:16:21.389120: step: 2858/1541, loss: 0.4661577343940735 2023-01-23 09:16:22.296203: step: 2860/1541, loss: 0.13229872286319733 2023-01-23 09:16:23.197700: step: 2862/1541, loss: 0.2185334712266922 2023-01-23 09:16:24.097936: step: 2864/1541, loss: 0.12819397449493408 2023-01-23 09:16:24.985420: step: 2866/1541, loss: 0.3134589195251465 2023-01-23 09:16:25.892651: step: 2868/1541, loss: 0.28910142183303833 2023-01-23 09:16:26.794823: step: 2870/1541, loss: 0.5841655731201172 2023-01-23 09:16:27.657700: step: 2872/1541, loss: 0.8240686058998108 2023-01-23 09:16:28.513492: step: 2874/1541, loss: 0.21985085308551788 2023-01-23 09:16:29.418811: step: 2876/1541, loss: 0.1204453781247139 2023-01-23 09:16:30.327661: step: 2878/1541, loss: 0.6012529730796814 2023-01-23 09:16:31.219141: step: 2880/1541, loss: 1.0791785717010498 2023-01-23 09:16:32.140424: step: 2882/1541, loss: 0.5196380615234375 2023-01-23 09:16:33.037191: step: 2884/1541, loss: 0.2847828269004822 2023-01-23 09:16:33.957882: step: 2886/1541, loss: 0.7430177330970764 2023-01-23 09:16:34.822790: step: 2888/1541, loss: 0.6399762630462646 2023-01-23 09:16:35.684996: step: 2890/1541, loss: 0.4867386221885681 2023-01-23 09:16:36.611491: step: 2892/1541, loss: 0.3954136073589325 2023-01-23 09:16:37.506980: step: 2894/1541, loss: 0.28413140773773193 2023-01-23 09:16:38.434205: step: 2896/1541, loss: 0.263794481754303 2023-01-23 09:16:39.303328: step: 2898/1541, loss: 0.12701226770877838 2023-01-23 09:16:40.176972: step: 2900/1541, loss: 0.30815428495407104 2023-01-23 09:16:41.070012: step: 2902/1541, loss: 0.13255995512008667 2023-01-23 09:16:41.945956: step: 2904/1541, loss: 0.2845686376094818 2023-01-23 09:16:42.834372: step: 2906/1541, loss: 0.7445847988128662 2023-01-23 09:16:43.736646: step: 2908/1541, loss: 0.26967084407806396 2023-01-23 09:16:44.628305: step: 2910/1541, loss: 0.10161785781383514 2023-01-23 09:16:45.528228: step: 2912/1541, loss: 0.29293587803840637 2023-01-23 09:16:46.435408: step: 2914/1541, loss: 0.13158322870731354 2023-01-23 09:16:47.317737: step: 2916/1541, loss: 0.1889778971672058 2023-01-23 09:16:48.207399: step: 2918/1541, loss: 0.38392385840415955 2023-01-23 09:16:49.081840: step: 2920/1541, loss: 0.037800855934619904 2023-01-23 09:16:49.995207: step: 2922/1541, loss: 0.19816021621227264 2023-01-23 09:16:50.898872: step: 2924/1541, loss: 0.2875034809112549 2023-01-23 09:16:51.828563: step: 2926/1541, loss: 0.6376743316650391 2023-01-23 09:16:52.754307: step: 2928/1541, loss: 0.1928943544626236 2023-01-23 09:16:53.748696: step: 2930/1541, loss: 0.3047928512096405 2023-01-23 09:16:54.677217: step: 2932/1541, loss: 0.7914055585861206 2023-01-23 09:16:55.593608: step: 2934/1541, loss: 0.1119471862912178 2023-01-23 09:16:56.502045: step: 2936/1541, loss: 0.16823293268680573 2023-01-23 09:16:57.385574: step: 2938/1541, loss: 0.762241780757904 2023-01-23 09:16:58.265322: step: 2940/1541, loss: 0.3507826030254364 2023-01-23 09:16:59.124899: step: 2942/1541, loss: 0.1752697080373764 2023-01-23 09:17:00.006702: step: 2944/1541, loss: 0.48214295506477356 2023-01-23 09:17:00.903335: step: 2946/1541, loss: 0.21703548729419708 2023-01-23 09:17:01.794779: step: 2948/1541, loss: 0.20315539836883545 2023-01-23 09:17:02.672813: step: 2950/1541, loss: 0.18917186558246613 2023-01-23 09:17:03.558497: step: 2952/1541, loss: 0.8255267143249512 2023-01-23 09:17:04.438702: step: 2954/1541, loss: 0.296637624502182 2023-01-23 09:17:05.299484: step: 2956/1541, loss: 0.5726242065429688 2023-01-23 09:17:06.202754: step: 2958/1541, loss: 0.18714632093906403 2023-01-23 09:17:07.099789: step: 2960/1541, loss: 1.1045899391174316 2023-01-23 09:17:08.009296: step: 2962/1541, loss: 2.184030771255493 2023-01-23 09:17:08.825269: step: 2964/1541, loss: 0.38983041048049927 2023-01-23 09:17:09.732188: step: 2966/1541, loss: 1.0044190883636475 2023-01-23 09:17:10.648680: step: 2968/1541, loss: 0.40031158924102783 2023-01-23 09:17:11.539688: step: 2970/1541, loss: 0.25363871455192566 2023-01-23 09:17:12.423201: step: 2972/1541, loss: 0.22495338320732117 2023-01-23 09:17:13.316084: step: 2974/1541, loss: 0.5664472579956055 2023-01-23 09:17:14.252074: step: 2976/1541, loss: 0.27431678771972656 2023-01-23 09:17:15.123313: step: 2978/1541, loss: 0.27987998723983765 2023-01-23 09:17:16.051763: step: 2980/1541, loss: 0.3705157935619354 2023-01-23 09:17:16.943561: step: 2982/1541, loss: 0.16222050786018372 2023-01-23 09:17:17.943143: step: 2984/1541, loss: 0.4955909848213196 2023-01-23 09:17:18.862487: step: 2986/1541, loss: 0.1252453625202179 2023-01-23 09:17:19.747624: step: 2988/1541, loss: 0.12274228036403656 2023-01-23 09:17:20.579301: step: 2990/1541, loss: 0.18815977871418 2023-01-23 09:17:21.483801: step: 2992/1541, loss: 0.6314312219619751 2023-01-23 09:17:22.416618: step: 2994/1541, loss: 1.8238776922225952 2023-01-23 09:17:23.308490: step: 2996/1541, loss: 0.26576292514801025 2023-01-23 09:17:24.206286: step: 2998/1541, loss: 0.183059424161911 2023-01-23 09:17:25.056765: step: 3000/1541, loss: 0.02354094572365284 2023-01-23 09:17:25.913364: step: 3002/1541, loss: 0.18382838368415833 2023-01-23 09:17:26.836476: step: 3004/1541, loss: 0.2455272227525711 2023-01-23 09:17:27.702758: step: 3006/1541, loss: 0.43833738565444946 2023-01-23 09:17:28.620634: step: 3008/1541, loss: 0.13477562367916107 2023-01-23 09:17:29.523085: step: 3010/1541, loss: 0.42483049631118774 2023-01-23 09:17:30.468592: step: 3012/1541, loss: 2.9746174812316895 2023-01-23 09:17:31.344487: step: 3014/1541, loss: 0.16829361021518707 2023-01-23 09:17:32.267807: step: 3016/1541, loss: 0.03099575638771057 2023-01-23 09:17:33.136795: step: 3018/1541, loss: 0.13281305134296417 2023-01-23 09:17:34.019452: step: 3020/1541, loss: 0.07837405800819397 2023-01-23 09:17:34.946061: step: 3022/1541, loss: 0.9582601189613342 2023-01-23 09:17:35.800219: step: 3024/1541, loss: 0.10956444591283798 2023-01-23 09:17:36.686044: step: 3026/1541, loss: 0.1932496726512909 2023-01-23 09:17:37.557159: step: 3028/1541, loss: 0.07860324531793594 2023-01-23 09:17:38.474855: step: 3030/1541, loss: 0.20121784508228302 2023-01-23 09:17:39.397834: step: 3032/1541, loss: 0.32276788353919983 2023-01-23 09:17:40.315614: step: 3034/1541, loss: 0.15568336844444275 2023-01-23 09:17:41.222385: step: 3036/1541, loss: 0.38444018363952637 2023-01-23 09:17:42.094075: step: 3038/1541, loss: 0.3647812604904175 2023-01-23 09:17:42.944399: step: 3040/1541, loss: 0.22981008887290955 2023-01-23 09:17:43.881150: step: 3042/1541, loss: 0.4960915446281433 2023-01-23 09:17:44.722276: step: 3044/1541, loss: 0.2961905896663666 2023-01-23 09:17:45.671723: step: 3046/1541, loss: 0.663835883140564 2023-01-23 09:17:46.604400: step: 3048/1541, loss: 0.3816401958465576 2023-01-23 09:17:47.538962: step: 3050/1541, loss: 0.19384220242500305 2023-01-23 09:17:48.499828: step: 3052/1541, loss: 0.24292251467704773 2023-01-23 09:17:49.375237: step: 3054/1541, loss: 0.24915672838687897 2023-01-23 09:17:50.245828: step: 3056/1541, loss: 0.22219668328762054 2023-01-23 09:17:51.127621: step: 3058/1541, loss: 0.3064993619918823 2023-01-23 09:17:51.986596: step: 3060/1541, loss: 0.42448821663856506 2023-01-23 09:17:52.934765: step: 3062/1541, loss: 0.1749090850353241 2023-01-23 09:17:53.819867: step: 3064/1541, loss: 6.138085842132568 2023-01-23 09:17:54.725541: step: 3066/1541, loss: 0.9669855237007141 2023-01-23 09:17:55.642019: step: 3068/1541, loss: 0.572438657283783 2023-01-23 09:17:56.545090: step: 3070/1541, loss: 0.21623533964157104 2023-01-23 09:17:57.481538: step: 3072/1541, loss: 0.19184665381908417 2023-01-23 09:17:58.371376: step: 3074/1541, loss: 0.1812366545200348 2023-01-23 09:17:59.253492: step: 3076/1541, loss: 2.373581886291504 2023-01-23 09:18:00.219560: step: 3078/1541, loss: 0.1833885759115219 2023-01-23 09:18:01.099011: step: 3080/1541, loss: 0.162556454539299 2023-01-23 09:18:02.057197: step: 3082/1541, loss: 0.5613314509391785 2023-01-23 09:18:02.673918: step: 3084/1541, loss: 0.41520756483078003 ================================================== Loss: 0.386 -------------------- Dev Chinese: {'event': {'p': 0.6308243727598566, 'r': 0.7030625832223701, 'f1': 0.6649874055415615}, 'argument': {'p': 0.6804945183739395, 'r': 0.6503560665112241, 'f1': 0.6650840346443746}, 'combined': 0.4422725066652767, 'epoch': 8} Test Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.7145227646598844, 'r': 0.6991594836413227, 'f1': 0.7067576434377564}, 'combined': 0.46074819208151696, 'epoch': 8} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6718530148667579, 'r': 0.6646184167167211, 'f1': 0.6682161345986232}, 'combined': 0.43994354203387614, 'epoch': 8} Test Korean: {'event': {'p': 0.6110581506196378, 'r': 0.7681246255242661, 'f1': 0.6806477302893549}, 'argument': {'p': 0.7112547814187644, 'r': 0.7160064835449799, 'f1': 0.7136227226921052}, 'combined': 0.4857256864832911, 'epoch': 8} Dev Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.6774181562069249, 'r': 0.6646463206826616, 'f1': 0.6709714664828391}, 'combined': 0.46189759887031806, 'epoch': 8} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.714479757993221, 'r': 0.7047310216184483, 'f1': 0.7095719071610188}, 'combined': 0.49194342837252764, 'epoch': 8} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6370694479420241, 'r': 0.6427072306671748, 'f1': 0.6398759212809757}, 'combined': 0.45939809732993125, 'epoch': 8} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.5609939325971934, 'r': 0.3590361168622038, 'f1': 0.4378489230026876}, 'combined': 0.26201982793861617, 'epoch': 8} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5501329207191277, 'r': 0.6396894426966601, 'f1': 0.591540774966804}, 'combined': 0.3042209699829278, 'epoch': 8} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5803921568627451, 'r': 0.7882822902796272, 'f1': 0.6685488424618858}, 'argument': {'p': 0.6535413679604629, 'r': 0.6483879951665094, 'f1': 0.6509544823729979}, 'combined': 0.4351948656858438, 'epoch': 4} Test for Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.6941652832691442, 'r': 0.7002565897211375, 'f1': 0.6971976320515535}, 'combined': 0.4763967891958319, 'epoch': 4} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6318317099567099, 'r': 0.365797305764411, 'f1': 0.4633432539682539}, 'combined': 0.2772762779652543, 'epoch': 4} -------------------- Dev for Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6749027227273954, 'r': 0.6759176140397674, 'f1': 0.6754097871321493}, 'combined': 0.467493670842566, 'epoch': 6} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7154362539653685, 'r': 0.7041128024637584, 'f1': 0.7097293657901553}, 'combined': 0.49443157493872836, 'epoch': 6} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5630713489409142, 'r': 0.6023553965414431, 'f1': 0.5820512820512821}, 'combined': 0.2993406593406594, 'epoch': 6} ****************************** Epoch: 9 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 09:23:20.147714: step: 2/1541, loss: 0.08849950134754181 2023-01-23 09:23:21.024067: step: 4/1541, loss: 0.3818768560886383 2023-01-23 09:23:21.892382: step: 6/1541, loss: 0.1831766963005066 2023-01-23 09:23:22.754997: step: 8/1541, loss: 0.4335534870624542 2023-01-23 09:23:23.666297: step: 10/1541, loss: 0.12297455966472626 2023-01-23 09:23:24.638279: step: 12/1541, loss: 0.1698128581047058 2023-01-23 09:23:25.488420: step: 14/1541, loss: 0.3063061535358429 2023-01-23 09:23:26.356454: step: 16/1541, loss: 0.16007713973522186 2023-01-23 09:23:27.240245: step: 18/1541, loss: 0.8173515796661377 2023-01-23 09:23:28.126169: step: 20/1541, loss: 0.22326132655143738 2023-01-23 09:23:28.957725: step: 22/1541, loss: 0.2017570436000824 2023-01-23 09:23:29.916601: step: 24/1541, loss: 0.2269849181175232 2023-01-23 09:23:30.776156: step: 26/1541, loss: 0.06522899866104126 2023-01-23 09:23:31.682295: step: 28/1541, loss: 0.061657585203647614 2023-01-23 09:23:32.595770: step: 30/1541, loss: 0.03461162745952606 2023-01-23 09:23:33.483639: step: 32/1541, loss: 0.24730491638183594 2023-01-23 09:23:34.407140: step: 34/1541, loss: 0.0533130019903183 2023-01-23 09:23:35.314642: step: 36/1541, loss: 0.035560764372348785 2023-01-23 09:23:36.206871: step: 38/1541, loss: 0.15141895413398743 2023-01-23 09:23:37.066637: step: 40/1541, loss: 0.6802422404289246 2023-01-23 09:23:37.980431: step: 42/1541, loss: 0.9321190714836121 2023-01-23 09:23:38.886321: step: 44/1541, loss: 0.2061871588230133 2023-01-23 09:23:39.761151: step: 46/1541, loss: 0.06406951695680618 2023-01-23 09:23:40.650746: step: 48/1541, loss: 0.3250621259212494 2023-01-23 09:23:41.551981: step: 50/1541, loss: 0.6815181970596313 2023-01-23 09:23:42.500429: step: 52/1541, loss: 0.06462164968252182 2023-01-23 09:23:43.476543: step: 54/1541, loss: 0.9669224619865417 2023-01-23 09:23:44.341061: step: 56/1541, loss: 0.4909720718860626 2023-01-23 09:23:45.220399: step: 58/1541, loss: 0.17208082973957062 2023-01-23 09:23:46.197138: step: 60/1541, loss: 0.24275562167167664 2023-01-23 09:23:47.072613: step: 62/1541, loss: 0.4083239436149597 2023-01-23 09:23:47.952619: step: 64/1541, loss: 0.8105455636978149 2023-01-23 09:23:48.864861: step: 66/1541, loss: 0.13859625160694122 2023-01-23 09:23:49.764066: step: 68/1541, loss: 0.15874768793582916 2023-01-23 09:23:50.642883: step: 70/1541, loss: 0.9745830297470093 2023-01-23 09:23:51.514611: step: 72/1541, loss: 0.1605963110923767 2023-01-23 09:23:52.416066: step: 74/1541, loss: 0.21721839904785156 2023-01-23 09:23:53.306407: step: 76/1541, loss: 0.7228643298149109 2023-01-23 09:23:54.218743: step: 78/1541, loss: 0.5253440141677856 2023-01-23 09:23:55.103608: step: 80/1541, loss: 0.18331216275691986 2023-01-23 09:23:55.973527: step: 82/1541, loss: 0.2752654552459717 2023-01-23 09:23:56.894044: step: 84/1541, loss: 0.37541037797927856 2023-01-23 09:23:57.789332: step: 86/1541, loss: 0.14044196903705597 2023-01-23 09:23:58.662933: step: 88/1541, loss: 0.9838200807571411 2023-01-23 09:23:59.623299: step: 90/1541, loss: 0.11835242807865143 2023-01-23 09:24:00.493467: step: 92/1541, loss: 0.1367492377758026 2023-01-23 09:24:01.429344: step: 94/1541, loss: 0.17120957374572754 2023-01-23 09:24:02.317244: step: 96/1541, loss: 0.3103199303150177 2023-01-23 09:24:03.197015: step: 98/1541, loss: 0.30147358775138855 2023-01-23 09:24:04.075225: step: 100/1541, loss: 0.18248197436332703 2023-01-23 09:24:04.953313: step: 102/1541, loss: 0.1500866711139679 2023-01-23 09:24:05.830780: step: 104/1541, loss: 1.1941094398498535 2023-01-23 09:24:06.716811: step: 106/1541, loss: 0.17991721630096436 2023-01-23 09:24:07.596588: step: 108/1541, loss: 0.3394630253314972 2023-01-23 09:24:08.465823: step: 110/1541, loss: 0.17129595577716827 2023-01-23 09:24:09.358226: step: 112/1541, loss: 0.13033278286457062 2023-01-23 09:24:10.272293: step: 114/1541, loss: 0.1318882256746292 2023-01-23 09:24:11.147662: step: 116/1541, loss: 0.06817492097616196 2023-01-23 09:24:12.020634: step: 118/1541, loss: 0.06256018579006195 2023-01-23 09:24:12.858972: step: 120/1541, loss: 0.2401001900434494 2023-01-23 09:24:13.733075: step: 122/1541, loss: 0.13759872317314148 2023-01-23 09:24:14.588308: step: 124/1541, loss: 0.1245851069688797 2023-01-23 09:24:15.473448: step: 126/1541, loss: 0.013286438770592213 2023-01-23 09:24:16.388922: step: 128/1541, loss: 0.11984191834926605 2023-01-23 09:24:17.299250: step: 130/1541, loss: 0.44518786668777466 2023-01-23 09:24:18.236345: step: 132/1541, loss: 0.09595832973718643 2023-01-23 09:24:19.106129: step: 134/1541, loss: 0.15139366686344147 2023-01-23 09:24:19.975028: step: 136/1541, loss: 0.7643557786941528 2023-01-23 09:24:20.857610: step: 138/1541, loss: 0.10923217982053757 2023-01-23 09:24:21.736565: step: 140/1541, loss: 0.14095909893512726 2023-01-23 09:24:22.637549: step: 142/1541, loss: 0.16588452458381653 2023-01-23 09:24:23.522002: step: 144/1541, loss: 0.0553138330578804 2023-01-23 09:24:24.415143: step: 146/1541, loss: 0.14413033425807953 2023-01-23 09:24:25.289672: step: 148/1541, loss: 0.2762118875980377 2023-01-23 09:24:26.263171: step: 150/1541, loss: 0.16289490461349487 2023-01-23 09:24:27.134160: step: 152/1541, loss: 0.41082704067230225 2023-01-23 09:24:28.061588: step: 154/1541, loss: 0.206671804189682 2023-01-23 09:24:28.951518: step: 156/1541, loss: 0.14828188717365265 2023-01-23 09:24:29.874044: step: 158/1541, loss: 0.21582669019699097 2023-01-23 09:24:30.761269: step: 160/1541, loss: 0.3847080171108246 2023-01-23 09:24:31.658916: step: 162/1541, loss: 0.10612446069717407 2023-01-23 09:24:32.589437: step: 164/1541, loss: 1.3898890018463135 2023-01-23 09:24:33.470849: step: 166/1541, loss: 0.18937847018241882 2023-01-23 09:24:34.339048: step: 168/1541, loss: 0.34558799862861633 2023-01-23 09:24:35.225905: step: 170/1541, loss: 0.45837122201919556 2023-01-23 09:24:36.140905: step: 172/1541, loss: 0.6627717614173889 2023-01-23 09:24:37.035677: step: 174/1541, loss: 0.018535396084189415 2023-01-23 09:24:37.969239: step: 176/1541, loss: 0.3143318295478821 2023-01-23 09:24:38.858219: step: 178/1541, loss: 0.046122174710035324 2023-01-23 09:24:39.729858: step: 180/1541, loss: 0.21139885485172272 2023-01-23 09:24:40.644066: step: 182/1541, loss: 0.1471918225288391 2023-01-23 09:24:41.525339: step: 184/1541, loss: 2.6129589080810547 2023-01-23 09:24:42.431284: step: 186/1541, loss: 0.1444685310125351 2023-01-23 09:24:43.305388: step: 188/1541, loss: 0.07672528922557831 2023-01-23 09:24:44.195879: step: 190/1541, loss: 0.1024586409330368 2023-01-23 09:24:45.096787: step: 192/1541, loss: 0.25463616847991943 2023-01-23 09:24:45.968988: step: 194/1541, loss: 0.31195327639579773 2023-01-23 09:24:46.874939: step: 196/1541, loss: 0.18103601038455963 2023-01-23 09:24:47.763869: step: 198/1541, loss: 0.07657942920923233 2023-01-23 09:24:48.615824: step: 200/1541, loss: 0.3553549647331238 2023-01-23 09:24:49.508500: step: 202/1541, loss: 0.304115891456604 2023-01-23 09:24:50.369987: step: 204/1541, loss: 0.12321966141462326 2023-01-23 09:24:51.252392: step: 206/1541, loss: 0.27048546075820923 2023-01-23 09:24:52.159526: step: 208/1541, loss: 0.11285960674285889 2023-01-23 09:24:53.016571: step: 210/1541, loss: 3.451056718826294 2023-01-23 09:24:53.967724: step: 212/1541, loss: 0.3376152515411377 2023-01-23 09:24:54.855562: step: 214/1541, loss: 0.1818607747554779 2023-01-23 09:24:55.766870: step: 216/1541, loss: 0.27834340929985046 2023-01-23 09:24:56.654401: step: 218/1541, loss: 0.2931063175201416 2023-01-23 09:24:57.571116: step: 220/1541, loss: 0.20536525547504425 2023-01-23 09:24:58.524060: step: 222/1541, loss: 1.1001957654953003 2023-01-23 09:24:59.433121: step: 224/1541, loss: 0.4779075086116791 2023-01-23 09:25:00.369352: step: 226/1541, loss: 0.037064384669065475 2023-01-23 09:25:01.270131: step: 228/1541, loss: 0.036667484790086746 2023-01-23 09:25:02.211088: step: 230/1541, loss: 0.28543421626091003 2023-01-23 09:25:03.058677: step: 232/1541, loss: 0.022566258907318115 2023-01-23 09:25:03.963776: step: 234/1541, loss: 0.23865529894828796 2023-01-23 09:25:04.848541: step: 236/1541, loss: 0.2020459771156311 2023-01-23 09:25:05.713441: step: 238/1541, loss: 0.3515109419822693 2023-01-23 09:25:06.584042: step: 240/1541, loss: 0.3588036298751831 2023-01-23 09:25:07.477664: step: 242/1541, loss: 0.8248891234397888 2023-01-23 09:25:08.367746: step: 244/1541, loss: 0.15460717678070068 2023-01-23 09:25:09.260874: step: 246/1541, loss: 0.09541217237710953 2023-01-23 09:25:10.159222: step: 248/1541, loss: 0.4784385859966278 2023-01-23 09:25:11.058114: step: 250/1541, loss: 3.0983664989471436 2023-01-23 09:25:11.957060: step: 252/1541, loss: 0.8365470170974731 2023-01-23 09:25:12.841770: step: 254/1541, loss: 0.09223765134811401 2023-01-23 09:25:13.696494: step: 256/1541, loss: 0.1387873888015747 2023-01-23 09:25:14.642917: step: 258/1541, loss: 0.5801163911819458 2023-01-23 09:25:15.505142: step: 260/1541, loss: 0.334931880235672 2023-01-23 09:25:16.391652: step: 262/1541, loss: 0.39161500334739685 2023-01-23 09:25:17.271149: step: 264/1541, loss: 0.19352374970912933 2023-01-23 09:25:18.198213: step: 266/1541, loss: 0.19825930893421173 2023-01-23 09:25:19.097706: step: 268/1541, loss: 0.33586248755455017 2023-01-23 09:25:20.000197: step: 270/1541, loss: 0.016258466988801956 2023-01-23 09:25:20.854212: step: 272/1541, loss: 0.1571401059627533 2023-01-23 09:25:21.778200: step: 274/1541, loss: 0.13477002084255219 2023-01-23 09:25:22.673253: step: 276/1541, loss: 0.12927083671092987 2023-01-23 09:25:23.577596: step: 278/1541, loss: 0.7181580662727356 2023-01-23 09:25:24.435799: step: 280/1541, loss: 0.12027833610773087 2023-01-23 09:25:25.292631: step: 282/1541, loss: 0.24783192574977875 2023-01-23 09:25:26.205091: step: 284/1541, loss: 0.656017541885376 2023-01-23 09:25:27.034928: step: 286/1541, loss: 0.4658767282962799 2023-01-23 09:25:27.917989: step: 288/1541, loss: 0.5186198353767395 2023-01-23 09:25:28.780941: step: 290/1541, loss: 0.35442325472831726 2023-01-23 09:25:29.644549: step: 292/1541, loss: 0.13488280773162842 2023-01-23 09:25:30.530702: step: 294/1541, loss: 0.2767036557197571 2023-01-23 09:25:31.430300: step: 296/1541, loss: 0.14557631313800812 2023-01-23 09:25:32.342001: step: 298/1541, loss: 0.6376498341560364 2023-01-23 09:25:33.212889: step: 300/1541, loss: 0.3282111585140228 2023-01-23 09:25:34.130783: step: 302/1541, loss: 0.17363348603248596 2023-01-23 09:25:35.006294: step: 304/1541, loss: 0.11293661594390869 2023-01-23 09:25:35.901898: step: 306/1541, loss: 0.5269594788551331 2023-01-23 09:25:36.802024: step: 308/1541, loss: 0.14204546809196472 2023-01-23 09:25:37.712648: step: 310/1541, loss: 0.3097150921821594 2023-01-23 09:25:38.605105: step: 312/1541, loss: 0.06838434189558029 2023-01-23 09:25:39.520104: step: 314/1541, loss: 0.40955179929733276 2023-01-23 09:25:40.422319: step: 316/1541, loss: 0.04131755232810974 2023-01-23 09:25:41.309062: step: 318/1541, loss: 0.44231119751930237 2023-01-23 09:25:42.198636: step: 320/1541, loss: 0.27357587218284607 2023-01-23 09:25:43.043454: step: 322/1541, loss: 0.12308917939662933 2023-01-23 09:25:43.900699: step: 324/1541, loss: 0.10686862468719482 2023-01-23 09:25:44.745715: step: 326/1541, loss: 0.13698187470436096 2023-01-23 09:25:45.661579: step: 328/1541, loss: 0.14809712767601013 2023-01-23 09:25:46.524956: step: 330/1541, loss: 0.16281560063362122 2023-01-23 09:25:47.424215: step: 332/1541, loss: 0.09277968853712082 2023-01-23 09:25:48.370859: step: 334/1541, loss: 0.06206349655985832 2023-01-23 09:25:49.229017: step: 336/1541, loss: 0.074163518846035 2023-01-23 09:25:50.128933: step: 338/1541, loss: 0.28913843631744385 2023-01-23 09:25:51.076194: step: 340/1541, loss: 0.26801222562789917 2023-01-23 09:25:51.944479: step: 342/1541, loss: 0.22803114354610443 2023-01-23 09:25:52.818815: step: 344/1541, loss: 0.1951894760131836 2023-01-23 09:25:53.698349: step: 346/1541, loss: 0.09171034395694733 2023-01-23 09:25:54.598091: step: 348/1541, loss: 0.09227226674556732 2023-01-23 09:25:55.444149: step: 350/1541, loss: 0.13122323155403137 2023-01-23 09:25:56.308437: step: 352/1541, loss: 0.3783782422542572 2023-01-23 09:25:57.177598: step: 354/1541, loss: 0.13052228093147278 2023-01-23 09:25:58.071374: step: 356/1541, loss: 0.02123410254716873 2023-01-23 09:25:59.041039: step: 358/1541, loss: 0.14220567047595978 2023-01-23 09:25:59.991372: step: 360/1541, loss: 0.06270617991685867 2023-01-23 09:26:00.861849: step: 362/1541, loss: 0.09543919563293457 2023-01-23 09:26:01.799275: step: 364/1541, loss: 0.05690629780292511 2023-01-23 09:26:02.676353: step: 366/1541, loss: 0.0733509510755539 2023-01-23 09:26:03.557552: step: 368/1541, loss: 0.3238367438316345 2023-01-23 09:26:04.451327: step: 370/1541, loss: 0.18956232070922852 2023-01-23 09:26:05.309546: step: 372/1541, loss: 0.03597695752978325 2023-01-23 09:26:06.219918: step: 374/1541, loss: 0.11905527114868164 2023-01-23 09:26:07.101474: step: 376/1541, loss: 0.5611344575881958 2023-01-23 09:26:08.040168: step: 378/1541, loss: 0.2002776712179184 2023-01-23 09:26:08.971537: step: 380/1541, loss: 0.05807005614042282 2023-01-23 09:26:09.891518: step: 382/1541, loss: 0.8874881267547607 2023-01-23 09:26:10.746280: step: 384/1541, loss: 0.37317878007888794 2023-01-23 09:26:11.573521: step: 386/1541, loss: 0.07991878688335419 2023-01-23 09:26:12.439232: step: 388/1541, loss: 0.20329561829566956 2023-01-23 09:26:13.364503: step: 390/1541, loss: 0.05267823114991188 2023-01-23 09:26:14.354275: step: 392/1541, loss: 0.2383064329624176 2023-01-23 09:26:15.241979: step: 394/1541, loss: 0.3360752761363983 2023-01-23 09:26:16.146749: step: 396/1541, loss: 0.19697462022304535 2023-01-23 09:26:17.046710: step: 398/1541, loss: 0.6298373937606812 2023-01-23 09:26:17.957131: step: 400/1541, loss: 0.15028689801692963 2023-01-23 09:26:18.824086: step: 402/1541, loss: 0.10913603007793427 2023-01-23 09:26:19.713167: step: 404/1541, loss: 0.13117271661758423 2023-01-23 09:26:20.577060: step: 406/1541, loss: 0.17766320705413818 2023-01-23 09:26:21.484900: step: 408/1541, loss: 0.28360220789909363 2023-01-23 09:26:22.381014: step: 410/1541, loss: 0.1090322881937027 2023-01-23 09:26:23.259872: step: 412/1541, loss: 0.405957967042923 2023-01-23 09:26:24.163476: step: 414/1541, loss: 0.07792535424232483 2023-01-23 09:26:25.046254: step: 416/1541, loss: 0.3359549641609192 2023-01-23 09:26:25.932095: step: 418/1541, loss: 0.12524640560150146 2023-01-23 09:26:26.822821: step: 420/1541, loss: 0.078191377222538 2023-01-23 09:26:27.716148: step: 422/1541, loss: 0.1347796469926834 2023-01-23 09:26:28.608043: step: 424/1541, loss: 0.1405165195465088 2023-01-23 09:26:29.487326: step: 426/1541, loss: 0.6634395122528076 2023-01-23 09:26:30.437954: step: 428/1541, loss: 0.19486750662326813 2023-01-23 09:26:31.282643: step: 430/1541, loss: 0.47370755672454834 2023-01-23 09:26:32.203894: step: 432/1541, loss: 0.1742035299539566 2023-01-23 09:26:33.098252: step: 434/1541, loss: 0.18999925255775452 2023-01-23 09:26:33.941453: step: 436/1541, loss: 0.12112574279308319 2023-01-23 09:26:34.832191: step: 438/1541, loss: 0.2824372351169586 2023-01-23 09:26:35.767173: step: 440/1541, loss: 0.1455850601196289 2023-01-23 09:26:36.669134: step: 442/1541, loss: 2.759300708770752 2023-01-23 09:26:37.531325: step: 444/1541, loss: 0.08684539049863815 2023-01-23 09:26:38.371126: step: 446/1541, loss: 0.1965583711862564 2023-01-23 09:26:39.234890: step: 448/1541, loss: 0.06895361095666885 2023-01-23 09:26:40.097937: step: 450/1541, loss: 0.03447265177965164 2023-01-23 09:26:40.978262: step: 452/1541, loss: 0.1070137470960617 2023-01-23 09:26:41.889232: step: 454/1541, loss: 0.5106670260429382 2023-01-23 09:26:42.793824: step: 456/1541, loss: 0.9519891142845154 2023-01-23 09:26:43.698538: step: 458/1541, loss: 0.09971962124109268 2023-01-23 09:26:44.602083: step: 460/1541, loss: 0.23203226923942566 2023-01-23 09:26:45.495170: step: 462/1541, loss: 0.07363779097795486 2023-01-23 09:26:46.354907: step: 464/1541, loss: 0.020710846409201622 2023-01-23 09:26:47.261382: step: 466/1541, loss: 0.14575082063674927 2023-01-23 09:26:48.115779: step: 468/1541, loss: 0.20324769616127014 2023-01-23 09:26:49.018908: step: 470/1541, loss: 0.32155805826187134 2023-01-23 09:26:49.978753: step: 472/1541, loss: 1.1377322673797607 2023-01-23 09:26:50.879226: step: 474/1541, loss: 0.10374592244625092 2023-01-23 09:26:51.749542: step: 476/1541, loss: 0.26250019669532776 2023-01-23 09:26:52.652451: step: 478/1541, loss: 0.2683311700820923 2023-01-23 09:26:53.559343: step: 480/1541, loss: 0.0843125656247139 2023-01-23 09:26:54.457643: step: 482/1541, loss: 0.07624998688697815 2023-01-23 09:26:55.363345: step: 484/1541, loss: 0.07041385769844055 2023-01-23 09:26:56.236027: step: 486/1541, loss: 0.15790799260139465 2023-01-23 09:26:57.109645: step: 488/1541, loss: 0.13001233339309692 2023-01-23 09:26:57.973706: step: 490/1541, loss: 0.14234766364097595 2023-01-23 09:26:58.882402: step: 492/1541, loss: 0.050360020250082016 2023-01-23 09:26:59.784636: step: 494/1541, loss: 0.2772422730922699 2023-01-23 09:27:00.665394: step: 496/1541, loss: 3.6981709003448486 2023-01-23 09:27:01.568944: step: 498/1541, loss: 0.16660889983177185 2023-01-23 09:27:02.467942: step: 500/1541, loss: 0.3357832729816437 2023-01-23 09:27:03.350569: step: 502/1541, loss: 0.07522409409284592 2023-01-23 09:27:04.315629: step: 504/1541, loss: 0.38274821639060974 2023-01-23 09:27:05.186970: step: 506/1541, loss: 1.0866475105285645 2023-01-23 09:27:06.065213: step: 508/1541, loss: 0.2603122889995575 2023-01-23 09:27:07.017043: step: 510/1541, loss: 0.2740657329559326 2023-01-23 09:27:07.912552: step: 512/1541, loss: 0.07122129201889038 2023-01-23 09:27:08.780792: step: 514/1541, loss: 0.10574494302272797 2023-01-23 09:27:09.656887: step: 516/1541, loss: 0.1692340224981308 2023-01-23 09:27:10.548016: step: 518/1541, loss: 0.2091754525899887 2023-01-23 09:27:11.433533: step: 520/1541, loss: 0.4043773412704468 2023-01-23 09:27:12.299654: step: 522/1541, loss: 0.5537917613983154 2023-01-23 09:27:13.188401: step: 524/1541, loss: 0.2057144045829773 2023-01-23 09:27:14.016708: step: 526/1541, loss: 0.31159067153930664 2023-01-23 09:27:14.884505: step: 528/1541, loss: 0.15632714331150055 2023-01-23 09:27:15.751533: step: 530/1541, loss: 0.4796421229839325 2023-01-23 09:27:16.734008: step: 532/1541, loss: 0.38382697105407715 2023-01-23 09:27:17.618332: step: 534/1541, loss: 0.27740737795829773 2023-01-23 09:27:18.513434: step: 536/1541, loss: 0.07464627921581268 2023-01-23 09:27:19.385322: step: 538/1541, loss: 0.06713549047708511 2023-01-23 09:27:20.287531: step: 540/1541, loss: 0.7892381548881531 2023-01-23 09:27:21.151835: step: 542/1541, loss: 0.20634999871253967 2023-01-23 09:27:22.043895: step: 544/1541, loss: 0.0908367782831192 2023-01-23 09:27:22.939149: step: 546/1541, loss: 0.2120552659034729 2023-01-23 09:27:23.798743: step: 548/1541, loss: 0.047315556555986404 2023-01-23 09:27:24.677479: step: 550/1541, loss: 0.4097866415977478 2023-01-23 09:27:25.557568: step: 552/1541, loss: 0.2664259374141693 2023-01-23 09:27:26.512732: step: 554/1541, loss: 0.1258765310049057 2023-01-23 09:27:27.400592: step: 556/1541, loss: 0.12027312070131302 2023-01-23 09:27:28.338189: step: 558/1541, loss: 0.15566347539424896 2023-01-23 09:27:29.203140: step: 560/1541, loss: 0.040281567722558975 2023-01-23 09:27:30.071272: step: 562/1541, loss: 0.11437418311834335 2023-01-23 09:27:30.930418: step: 564/1541, loss: 0.06373283267021179 2023-01-23 09:27:31.848603: step: 566/1541, loss: 0.1336960345506668 2023-01-23 09:27:32.738267: step: 568/1541, loss: 0.10302471369504929 2023-01-23 09:27:33.603401: step: 570/1541, loss: 0.1660638153553009 2023-01-23 09:27:34.459130: step: 572/1541, loss: 0.047428809106349945 2023-01-23 09:27:35.337709: step: 574/1541, loss: 0.13064055144786835 2023-01-23 09:27:36.185172: step: 576/1541, loss: 0.7123777270317078 2023-01-23 09:27:37.114869: step: 578/1541, loss: 0.08293230086565018 2023-01-23 09:27:37.996451: step: 580/1541, loss: 0.17088936269283295 2023-01-23 09:27:38.918811: step: 582/1541, loss: 0.09173554182052612 2023-01-23 09:27:39.862582: step: 584/1541, loss: 0.3242572546005249 2023-01-23 09:27:40.739091: step: 586/1541, loss: 0.05350404232740402 2023-01-23 09:27:41.593855: step: 588/1541, loss: 0.24149899184703827 2023-01-23 09:27:42.463919: step: 590/1541, loss: 0.14970514178276062 2023-01-23 09:27:43.328131: step: 592/1541, loss: 0.59630286693573 2023-01-23 09:27:44.211299: step: 594/1541, loss: 0.22070206701755524 2023-01-23 09:27:45.095406: step: 596/1541, loss: 0.09192207455635071 2023-01-23 09:27:46.015492: step: 598/1541, loss: 0.2955693006515503 2023-01-23 09:27:46.951591: step: 600/1541, loss: 0.08855314552783966 2023-01-23 09:27:47.926752: step: 602/1541, loss: 0.27530282735824585 2023-01-23 09:27:48.821636: step: 604/1541, loss: 0.3314485251903534 2023-01-23 09:27:49.727857: step: 606/1541, loss: 0.1490289270877838 2023-01-23 09:27:50.625167: step: 608/1541, loss: 0.11797115951776505 2023-01-23 09:27:51.544767: step: 610/1541, loss: 0.11661265045404434 2023-01-23 09:27:52.432065: step: 612/1541, loss: 0.12182460725307465 2023-01-23 09:27:53.303911: step: 614/1541, loss: 0.02836156263947487 2023-01-23 09:27:54.191248: step: 616/1541, loss: 0.18925994634628296 2023-01-23 09:27:55.085611: step: 618/1541, loss: 0.06563887000083923 2023-01-23 09:27:55.964358: step: 620/1541, loss: 0.09230674058198929 2023-01-23 09:27:56.820153: step: 622/1541, loss: 0.13122184574604034 2023-01-23 09:27:57.711732: step: 624/1541, loss: 0.21549035608768463 2023-01-23 09:27:58.561888: step: 626/1541, loss: 0.016666458919644356 2023-01-23 09:27:59.421826: step: 628/1541, loss: 0.32329612970352173 2023-01-23 09:28:00.323876: step: 630/1541, loss: 0.267677903175354 2023-01-23 09:28:01.185641: step: 632/1541, loss: 0.04642551392316818 2023-01-23 09:28:02.169602: step: 634/1541, loss: 0.10810843110084534 2023-01-23 09:28:03.074601: step: 636/1541, loss: 0.1817188560962677 2023-01-23 09:28:03.920579: step: 638/1541, loss: 0.06168406829237938 2023-01-23 09:28:04.820923: step: 640/1541, loss: 0.49765321612358093 2023-01-23 09:28:05.716411: step: 642/1541, loss: 0.9355195164680481 2023-01-23 09:28:06.597282: step: 644/1541, loss: 0.19406484067440033 2023-01-23 09:28:07.490122: step: 646/1541, loss: 0.5174009799957275 2023-01-23 09:28:08.362028: step: 648/1541, loss: 0.21677149832248688 2023-01-23 09:28:09.235611: step: 650/1541, loss: 0.13368189334869385 2023-01-23 09:28:10.168030: step: 652/1541, loss: 0.16495098173618317 2023-01-23 09:28:11.026963: step: 654/1541, loss: 0.10058344900608063 2023-01-23 09:28:11.929780: step: 656/1541, loss: 0.3419042229652405 2023-01-23 09:28:12.846409: step: 658/1541, loss: 0.2687222361564636 2023-01-23 09:28:13.707850: step: 660/1541, loss: 0.313256174325943 2023-01-23 09:28:14.578460: step: 662/1541, loss: 0.4811215400695801 2023-01-23 09:28:15.490552: step: 664/1541, loss: 0.4686118960380554 2023-01-23 09:28:16.365150: step: 666/1541, loss: 0.16966687142848969 2023-01-23 09:28:17.276783: step: 668/1541, loss: 0.32929083704948425 2023-01-23 09:28:18.184801: step: 670/1541, loss: 0.26809805631637573 2023-01-23 09:28:19.071515: step: 672/1541, loss: 0.11051356047391891 2023-01-23 09:28:19.983455: step: 674/1541, loss: 2.42803955078125 2023-01-23 09:28:20.861520: step: 676/1541, loss: 0.14292046427726746 2023-01-23 09:28:21.736101: step: 678/1541, loss: 0.03488488867878914 2023-01-23 09:28:22.648919: step: 680/1541, loss: 2.8323497772216797 2023-01-23 09:28:23.512184: step: 682/1541, loss: 0.14627328515052795 2023-01-23 09:28:24.410044: step: 684/1541, loss: 0.20011520385742188 2023-01-23 09:28:25.318263: step: 686/1541, loss: 0.12082941085100174 2023-01-23 09:28:26.228529: step: 688/1541, loss: 0.10834495723247528 2023-01-23 09:28:27.141509: step: 690/1541, loss: 2.95354962348938 2023-01-23 09:28:28.034809: step: 692/1541, loss: 0.6568156480789185 2023-01-23 09:28:28.904437: step: 694/1541, loss: 0.08545096963644028 2023-01-23 09:28:29.868074: step: 696/1541, loss: 0.4965648353099823 2023-01-23 09:28:30.738917: step: 698/1541, loss: 0.26038655638694763 2023-01-23 09:28:31.631415: step: 700/1541, loss: 0.2832585871219635 2023-01-23 09:28:32.573091: step: 702/1541, loss: 0.2690350115299225 2023-01-23 09:28:33.482194: step: 704/1541, loss: 0.12295836955308914 2023-01-23 09:28:34.360244: step: 706/1541, loss: 0.265652596950531 2023-01-23 09:28:35.241128: step: 708/1541, loss: 0.14550244808197021 2023-01-23 09:28:36.141360: step: 710/1541, loss: 0.14302238821983337 2023-01-23 09:28:37.085105: step: 712/1541, loss: 0.3060433566570282 2023-01-23 09:28:37.946569: step: 714/1541, loss: 0.63747638463974 2023-01-23 09:28:38.817074: step: 716/1541, loss: 0.026901233941316605 2023-01-23 09:28:39.704631: step: 718/1541, loss: 0.08116761595010757 2023-01-23 09:28:40.614376: step: 720/1541, loss: 0.8338356614112854 2023-01-23 09:28:41.525025: step: 722/1541, loss: 0.2727735936641693 2023-01-23 09:28:42.429572: step: 724/1541, loss: 0.3946680426597595 2023-01-23 09:28:43.350843: step: 726/1541, loss: 0.42981112003326416 2023-01-23 09:28:44.235891: step: 728/1541, loss: 0.2744418978691101 2023-01-23 09:28:45.097834: step: 730/1541, loss: 0.39270153641700745 2023-01-23 09:28:46.006279: step: 732/1541, loss: 0.2225494384765625 2023-01-23 09:28:46.866020: step: 734/1541, loss: 0.2905536890029907 2023-01-23 09:28:47.814602: step: 736/1541, loss: 0.48735612630844116 2023-01-23 09:28:48.702619: step: 738/1541, loss: 0.7443857789039612 2023-01-23 09:28:49.563822: step: 740/1541, loss: 0.1419381946325302 2023-01-23 09:28:50.426483: step: 742/1541, loss: 0.20353910326957703 2023-01-23 09:28:51.343270: step: 744/1541, loss: 0.17446132004261017 2023-01-23 09:28:52.226853: step: 746/1541, loss: 0.20718713104724884 2023-01-23 09:28:53.118020: step: 748/1541, loss: 0.9089257717132568 2023-01-23 09:28:53.981197: step: 750/1541, loss: 0.046666670590639114 2023-01-23 09:28:54.917923: step: 752/1541, loss: 0.06938852369785309 2023-01-23 09:28:55.812003: step: 754/1541, loss: 0.373489111661911 2023-01-23 09:28:56.673246: step: 756/1541, loss: 0.43893977999687195 2023-01-23 09:28:57.576357: step: 758/1541, loss: 0.03221005201339722 2023-01-23 09:28:58.484393: step: 760/1541, loss: 0.14426293969154358 2023-01-23 09:28:59.336243: step: 762/1541, loss: 0.03655238449573517 2023-01-23 09:29:00.254234: step: 764/1541, loss: 0.18293841183185577 2023-01-23 09:29:01.090676: step: 766/1541, loss: 0.49208560585975647 2023-01-23 09:29:01.944532: step: 768/1541, loss: 0.2536385655403137 2023-01-23 09:29:02.865593: step: 770/1541, loss: 0.1306973546743393 2023-01-23 09:29:03.757764: step: 772/1541, loss: 0.10045012086629868 2023-01-23 09:29:04.654739: step: 774/1541, loss: 0.0750202164053917 2023-01-23 09:29:05.524171: step: 776/1541, loss: 0.06845349073410034 2023-01-23 09:29:06.418374: step: 778/1541, loss: 0.31013280153274536 2023-01-23 09:29:07.353529: step: 780/1541, loss: 0.43157801032066345 2023-01-23 09:29:08.252781: step: 782/1541, loss: 0.2578630745410919 2023-01-23 09:29:09.128620: step: 784/1541, loss: 0.19877862930297852 2023-01-23 09:29:09.986137: step: 786/1541, loss: 0.4726126492023468 2023-01-23 09:29:10.837140: step: 788/1541, loss: 0.23883090913295746 2023-01-23 09:29:11.722084: step: 790/1541, loss: 0.06416235864162445 2023-01-23 09:29:12.603187: step: 792/1541, loss: 0.10584954917430878 2023-01-23 09:29:13.508853: step: 794/1541, loss: 0.15291620790958405 2023-01-23 09:29:14.401071: step: 796/1541, loss: 0.22278562188148499 2023-01-23 09:29:15.289850: step: 798/1541, loss: 0.36481615900993347 2023-01-23 09:29:16.178299: step: 800/1541, loss: 0.07773931324481964 2023-01-23 09:29:17.049323: step: 802/1541, loss: 0.4076418876647949 2023-01-23 09:29:17.916442: step: 804/1541, loss: 0.27578940987586975 2023-01-23 09:29:18.782297: step: 806/1541, loss: 0.24015580117702484 2023-01-23 09:29:19.686842: step: 808/1541, loss: 0.24530406296253204 2023-01-23 09:29:20.589193: step: 810/1541, loss: 0.13177993893623352 2023-01-23 09:29:21.471194: step: 812/1541, loss: 0.556718647480011 2023-01-23 09:29:22.390763: step: 814/1541, loss: 0.49292221665382385 2023-01-23 09:29:23.276186: step: 816/1541, loss: 0.17663435637950897 2023-01-23 09:29:24.188568: step: 818/1541, loss: 0.13270118832588196 2023-01-23 09:29:25.107864: step: 820/1541, loss: 0.07503288239240646 2023-01-23 09:29:26.018351: step: 822/1541, loss: 0.0602244958281517 2023-01-23 09:29:26.918598: step: 824/1541, loss: 0.03913363069295883 2023-01-23 09:29:27.818054: step: 826/1541, loss: 0.12237777560949326 2023-01-23 09:29:28.680358: step: 828/1541, loss: 0.10824333876371384 2023-01-23 09:29:29.607453: step: 830/1541, loss: 0.4840984344482422 2023-01-23 09:29:30.487114: step: 832/1541, loss: 0.13489030301570892 2023-01-23 09:29:31.364877: step: 834/1541, loss: 0.06253604590892792 2023-01-23 09:29:32.265797: step: 836/1541, loss: 0.11453165113925934 2023-01-23 09:29:33.149259: step: 838/1541, loss: 0.335276335477829 2023-01-23 09:29:34.061871: step: 840/1541, loss: 0.21576905250549316 2023-01-23 09:29:34.966734: step: 842/1541, loss: 0.2958153188228607 2023-01-23 09:29:35.851386: step: 844/1541, loss: 0.22386875748634338 2023-01-23 09:29:36.733810: step: 846/1541, loss: 0.09266981482505798 2023-01-23 09:29:37.595924: step: 848/1541, loss: 0.08618690818548203 2023-01-23 09:29:38.470353: step: 850/1541, loss: 0.061094313859939575 2023-01-23 09:29:39.329334: step: 852/1541, loss: 0.04503752663731575 2023-01-23 09:29:40.227085: step: 854/1541, loss: 0.33244550228118896 2023-01-23 09:29:41.118792: step: 856/1541, loss: 0.0649273693561554 2023-01-23 09:29:42.058268: step: 858/1541, loss: 0.16388224065303802 2023-01-23 09:29:42.946745: step: 860/1541, loss: 0.24587924778461456 2023-01-23 09:29:43.870302: step: 862/1541, loss: 0.4617248475551605 2023-01-23 09:29:44.739092: step: 864/1541, loss: 0.17269855737686157 2023-01-23 09:29:45.645206: step: 866/1541, loss: 0.3807412087917328 2023-01-23 09:29:46.520077: step: 868/1541, loss: 0.09470195323228836 2023-01-23 09:29:47.414564: step: 870/1541, loss: 0.17047548294067383 2023-01-23 09:29:48.321153: step: 872/1541, loss: 2.6380739212036133 2023-01-23 09:29:49.245852: step: 874/1541, loss: 0.4602656364440918 2023-01-23 09:29:50.218126: step: 876/1541, loss: 0.3481162488460541 2023-01-23 09:29:51.099361: step: 878/1541, loss: 0.09600889682769775 2023-01-23 09:29:52.031757: step: 880/1541, loss: 0.18936070799827576 2023-01-23 09:29:52.921225: step: 882/1541, loss: 3.6581742763519287 2023-01-23 09:29:53.892902: step: 884/1541, loss: 0.14214502274990082 2023-01-23 09:29:54.796682: step: 886/1541, loss: 0.3186121881008148 2023-01-23 09:29:55.699641: step: 888/1541, loss: 0.192325621843338 2023-01-23 09:29:56.644003: step: 890/1541, loss: 0.11869687587022781 2023-01-23 09:29:57.515263: step: 892/1541, loss: 0.12090513855218887 2023-01-23 09:29:58.412945: step: 894/1541, loss: 0.09309802204370499 2023-01-23 09:29:59.286777: step: 896/1541, loss: 0.1123320460319519 2023-01-23 09:30:00.191283: step: 898/1541, loss: 0.2198963463306427 2023-01-23 09:30:01.111233: step: 900/1541, loss: 0.31485864520072937 2023-01-23 09:30:01.998583: step: 902/1541, loss: 0.026976915076375008 2023-01-23 09:30:02.917465: step: 904/1541, loss: 0.30743861198425293 2023-01-23 09:30:03.783274: step: 906/1541, loss: 0.2368983030319214 2023-01-23 09:30:04.634214: step: 908/1541, loss: 0.0662633404135704 2023-01-23 09:30:05.548512: step: 910/1541, loss: 0.14501552283763885 2023-01-23 09:30:06.436589: step: 912/1541, loss: 0.09008108824491501 2023-01-23 09:30:07.329209: step: 914/1541, loss: 0.2748074531555176 2023-01-23 09:30:08.219014: step: 916/1541, loss: 0.40472856163978577 2023-01-23 09:30:09.089866: step: 918/1541, loss: 0.11708708852529526 2023-01-23 09:30:09.958626: step: 920/1541, loss: 0.8348321914672852 2023-01-23 09:30:10.838373: step: 922/1541, loss: 0.3040390908718109 2023-01-23 09:30:11.727098: step: 924/1541, loss: 0.20293010771274567 2023-01-23 09:30:12.657610: step: 926/1541, loss: 0.1084696501493454 2023-01-23 09:30:13.570538: step: 928/1541, loss: 0.08042187243700027 2023-01-23 09:30:14.418385: step: 930/1541, loss: 0.2863194942474365 2023-01-23 09:30:15.295702: step: 932/1541, loss: 0.22140765190124512 2023-01-23 09:30:16.171913: step: 934/1541, loss: 0.2696683704853058 2023-01-23 09:30:17.034669: step: 936/1541, loss: 0.16470788419246674 2023-01-23 09:30:17.934803: step: 938/1541, loss: 0.0915156900882721 2023-01-23 09:30:18.831193: step: 940/1541, loss: 0.19595728814601898 2023-01-23 09:30:19.737326: step: 942/1541, loss: 0.1321089118719101 2023-01-23 09:30:20.657472: step: 944/1541, loss: 0.12080419808626175 2023-01-23 09:30:21.512245: step: 946/1541, loss: 0.2550188899040222 2023-01-23 09:30:22.387877: step: 948/1541, loss: 0.3225822150707245 2023-01-23 09:30:23.228872: step: 950/1541, loss: 0.27678558230400085 2023-01-23 09:30:24.119043: step: 952/1541, loss: 0.13193489611148834 2023-01-23 09:30:24.973382: step: 954/1541, loss: 0.07208768278360367 2023-01-23 09:30:25.928815: step: 956/1541, loss: 0.525526225566864 2023-01-23 09:30:26.856255: step: 958/1541, loss: 0.37465569376945496 2023-01-23 09:30:27.757959: step: 960/1541, loss: 0.11308063566684723 2023-01-23 09:30:28.631967: step: 962/1541, loss: 0.292889803647995 2023-01-23 09:30:29.593015: step: 964/1541, loss: 0.7167654037475586 2023-01-23 09:30:30.502576: step: 966/1541, loss: 0.4756660759449005 2023-01-23 09:30:31.364656: step: 968/1541, loss: 0.6008126735687256 2023-01-23 09:30:32.229776: step: 970/1541, loss: 0.3481917977333069 2023-01-23 09:30:33.109235: step: 972/1541, loss: 0.07396048307418823 2023-01-23 09:30:33.991947: step: 974/1541, loss: 0.12833771109580994 2023-01-23 09:30:34.870474: step: 976/1541, loss: 0.15678606927394867 2023-01-23 09:30:35.737991: step: 978/1541, loss: 0.2941652536392212 2023-01-23 09:30:36.590607: step: 980/1541, loss: 2.4832088947296143 2023-01-23 09:30:37.463701: step: 982/1541, loss: 0.4334222674369812 2023-01-23 09:30:38.350818: step: 984/1541, loss: 0.18131409585475922 2023-01-23 09:30:39.216925: step: 986/1541, loss: 0.21195754408836365 2023-01-23 09:30:40.101243: step: 988/1541, loss: 0.054286979138851166 2023-01-23 09:30:41.035693: step: 990/1541, loss: 0.39178740978240967 2023-01-23 09:30:42.000452: step: 992/1541, loss: 0.279953271150589 2023-01-23 09:30:42.867529: step: 994/1541, loss: 0.07835876941680908 2023-01-23 09:30:43.791287: step: 996/1541, loss: 0.06549602001905441 2023-01-23 09:30:44.656124: step: 998/1541, loss: 0.04545709490776062 2023-01-23 09:30:45.540542: step: 1000/1541, loss: 0.5606716275215149 2023-01-23 09:30:46.416175: step: 1002/1541, loss: 0.11661122739315033 2023-01-23 09:30:47.361044: step: 1004/1541, loss: 0.1450364589691162 2023-01-23 09:30:48.263218: step: 1006/1541, loss: 0.15104298293590546 2023-01-23 09:30:49.158388: step: 1008/1541, loss: 0.19327040016651154 2023-01-23 09:30:50.060559: step: 1010/1541, loss: 0.20937791466712952 2023-01-23 09:30:50.914836: step: 1012/1541, loss: 0.09495054930448532 2023-01-23 09:30:51.801558: step: 1014/1541, loss: 0.4311925768852234 2023-01-23 09:30:52.733409: step: 1016/1541, loss: 0.06182965263724327 2023-01-23 09:30:53.594048: step: 1018/1541, loss: 0.1456158608198166 2023-01-23 09:30:54.519678: step: 1020/1541, loss: 0.03274774178862572 2023-01-23 09:30:55.363387: step: 1022/1541, loss: 0.3339128792285919 2023-01-23 09:30:56.270447: step: 1024/1541, loss: 0.26554369926452637 2023-01-23 09:30:57.117666: step: 1026/1541, loss: 0.06480393558740616 2023-01-23 09:30:58.027213: step: 1028/1541, loss: 0.19306638836860657 2023-01-23 09:30:58.897044: step: 1030/1541, loss: 0.09348426014184952 2023-01-23 09:30:59.761704: step: 1032/1541, loss: 0.10266513377428055 2023-01-23 09:31:00.627876: step: 1034/1541, loss: 0.4852600395679474 2023-01-23 09:31:01.512662: step: 1036/1541, loss: 0.5647838115692139 2023-01-23 09:31:02.431547: step: 1038/1541, loss: 0.2851905822753906 2023-01-23 09:31:03.305119: step: 1040/1541, loss: 0.1025794968008995 2023-01-23 09:31:04.191311: step: 1042/1541, loss: 0.12434898316860199 2023-01-23 09:31:05.041828: step: 1044/1541, loss: 0.17248208820819855 2023-01-23 09:31:05.935380: step: 1046/1541, loss: 0.17035120725631714 2023-01-23 09:31:06.913026: step: 1048/1541, loss: 0.12836410105228424 2023-01-23 09:31:07.829957: step: 1050/1541, loss: 0.1985316425561905 2023-01-23 09:31:08.728344: step: 1052/1541, loss: 0.08309601247310638 2023-01-23 09:31:09.578487: step: 1054/1541, loss: 0.13596652448177338 2023-01-23 09:31:10.520160: step: 1056/1541, loss: 0.11871174722909927 2023-01-23 09:31:11.373187: step: 1058/1541, loss: 0.22926223278045654 2023-01-23 09:31:12.251643: step: 1060/1541, loss: 0.022072887048125267 2023-01-23 09:31:13.109390: step: 1062/1541, loss: 0.24211497604846954 2023-01-23 09:31:14.003687: step: 1064/1541, loss: 0.33171898126602173 2023-01-23 09:31:15.012917: step: 1066/1541, loss: 0.24900513887405396 2023-01-23 09:31:15.948713: step: 1068/1541, loss: 0.15964046120643616 2023-01-23 09:31:16.859516: step: 1070/1541, loss: 0.560958981513977 2023-01-23 09:31:17.746716: step: 1072/1541, loss: 0.3915262520313263 2023-01-23 09:31:18.619412: step: 1074/1541, loss: 0.17339418828487396 2023-01-23 09:31:19.519664: step: 1076/1541, loss: 0.0857517197728157 2023-01-23 09:31:20.428266: step: 1078/1541, loss: 0.17239047586917877 2023-01-23 09:31:21.316508: step: 1080/1541, loss: 0.27751192450523376 2023-01-23 09:31:22.178830: step: 1082/1541, loss: 0.09408023208379745 2023-01-23 09:31:23.071643: step: 1084/1541, loss: 0.06357160955667496 2023-01-23 09:31:23.955013: step: 1086/1541, loss: 0.15394851565361023 2023-01-23 09:31:24.832447: step: 1088/1541, loss: 0.05382506176829338 2023-01-23 09:31:25.696129: step: 1090/1541, loss: 0.08601900935173035 2023-01-23 09:31:26.595110: step: 1092/1541, loss: 0.16752049326896667 2023-01-23 09:31:27.454483: step: 1094/1541, loss: 0.5969488620758057 2023-01-23 09:31:28.315620: step: 1096/1541, loss: 0.33658257126808167 2023-01-23 09:31:29.186981: step: 1098/1541, loss: 0.09067797660827637 2023-01-23 09:31:30.088278: step: 1100/1541, loss: 0.25720635056495667 2023-01-23 09:31:30.965850: step: 1102/1541, loss: 0.06583476811647415 2023-01-23 09:31:31.881954: step: 1104/1541, loss: 0.17592307925224304 2023-01-23 09:31:32.757210: step: 1106/1541, loss: 0.05683542415499687 2023-01-23 09:31:33.667870: step: 1108/1541, loss: 0.31286337971687317 2023-01-23 09:31:34.537340: step: 1110/1541, loss: 0.3231843113899231 2023-01-23 09:31:35.401780: step: 1112/1541, loss: 0.4650062024593353 2023-01-23 09:31:36.422401: step: 1114/1541, loss: 0.32232820987701416 2023-01-23 09:31:37.280104: step: 1116/1541, loss: 0.13915176689624786 2023-01-23 09:31:38.239277: step: 1118/1541, loss: 0.3810867965221405 2023-01-23 09:31:39.125438: step: 1120/1541, loss: 0.24100397527217865 2023-01-23 09:31:40.120319: step: 1122/1541, loss: 0.2249571979045868 2023-01-23 09:31:41.048863: step: 1124/1541, loss: 0.35297149419784546 2023-01-23 09:31:41.929093: step: 1126/1541, loss: 0.16815787553787231 2023-01-23 09:31:42.857078: step: 1128/1541, loss: 0.11615581810474396 2023-01-23 09:31:43.853905: step: 1130/1541, loss: 0.8684134483337402 2023-01-23 09:31:44.751493: step: 1132/1541, loss: 0.22362904250621796 2023-01-23 09:31:45.623620: step: 1134/1541, loss: 0.2283923327922821 2023-01-23 09:31:46.460860: step: 1136/1541, loss: 0.09751953184604645 2023-01-23 09:31:47.330092: step: 1138/1541, loss: 0.07428660243749619 2023-01-23 09:31:48.204913: step: 1140/1541, loss: 0.06055280566215515 2023-01-23 09:31:49.061934: step: 1142/1541, loss: 0.08538509160280228 2023-01-23 09:31:49.941719: step: 1144/1541, loss: 0.46324777603149414 2023-01-23 09:31:50.876443: step: 1146/1541, loss: 0.3688974678516388 2023-01-23 09:31:51.747470: step: 1148/1541, loss: 0.21424759924411774 2023-01-23 09:31:52.656134: step: 1150/1541, loss: 0.24244363605976105 2023-01-23 09:31:53.561547: step: 1152/1541, loss: 0.525848388671875 2023-01-23 09:31:54.444019: step: 1154/1541, loss: 0.2006988674402237 2023-01-23 09:31:55.379951: step: 1156/1541, loss: 0.7393337488174438 2023-01-23 09:31:56.223078: step: 1158/1541, loss: 0.23839738965034485 2023-01-23 09:31:57.107912: step: 1160/1541, loss: 0.5482367873191833 2023-01-23 09:31:57.987629: step: 1162/1541, loss: 0.21838000416755676 2023-01-23 09:31:58.910994: step: 1164/1541, loss: 0.1121140569448471 2023-01-23 09:31:59.893990: step: 1166/1541, loss: 0.10142770409584045 2023-01-23 09:32:00.808690: step: 1168/1541, loss: 0.12567190825939178 2023-01-23 09:32:01.687541: step: 1170/1541, loss: 0.07081685960292816 2023-01-23 09:32:02.619024: step: 1172/1541, loss: 0.1310352087020874 2023-01-23 09:32:03.508697: step: 1174/1541, loss: 0.057648565620183945 2023-01-23 09:32:04.476515: step: 1176/1541, loss: 0.3937949240207672 2023-01-23 09:32:05.363015: step: 1178/1541, loss: 0.40244561433792114 2023-01-23 09:32:06.241204: step: 1180/1541, loss: 0.24609118700027466 2023-01-23 09:32:07.146177: step: 1182/1541, loss: 0.1297173798084259 2023-01-23 09:32:08.084977: step: 1184/1541, loss: 0.06971150636672974 2023-01-23 09:32:08.983809: step: 1186/1541, loss: 0.12693503499031067 2023-01-23 09:32:09.893934: step: 1188/1541, loss: 0.1391269862651825 2023-01-23 09:32:10.767781: step: 1190/1541, loss: 0.0663967952132225 2023-01-23 09:32:11.675078: step: 1192/1541, loss: 0.38922587037086487 2023-01-23 09:32:12.569523: step: 1194/1541, loss: 0.11293023079633713 2023-01-23 09:32:13.476912: step: 1196/1541, loss: 0.05635279789566994 2023-01-23 09:32:14.387733: step: 1198/1541, loss: 0.0608907975256443 2023-01-23 09:32:15.195885: step: 1200/1541, loss: 0.3377990424633026 2023-01-23 09:32:16.090373: step: 1202/1541, loss: 0.10905923694372177 2023-01-23 09:32:16.971575: step: 1204/1541, loss: 0.06603400409221649 2023-01-23 09:32:17.842854: step: 1206/1541, loss: 0.1507679969072342 2023-01-23 09:32:18.776981: step: 1208/1541, loss: 0.09762440621852875 2023-01-23 09:32:19.692748: step: 1210/1541, loss: 2.56103253364563 2023-01-23 09:32:20.549954: step: 1212/1541, loss: 0.443973571062088 2023-01-23 09:32:21.448224: step: 1214/1541, loss: 0.6030001044273376 2023-01-23 09:32:22.351850: step: 1216/1541, loss: 0.10680204629898071 2023-01-23 09:32:23.251649: step: 1218/1541, loss: 0.21253885328769684 2023-01-23 09:32:24.127684: step: 1220/1541, loss: 0.023111648857593536 2023-01-23 09:32:25.011716: step: 1222/1541, loss: 2.660343885421753 2023-01-23 09:32:25.889523: step: 1224/1541, loss: 0.25598978996276855 2023-01-23 09:32:26.778737: step: 1226/1541, loss: 0.20289799571037292 2023-01-23 09:32:27.656567: step: 1228/1541, loss: 0.43110284209251404 2023-01-23 09:32:28.519769: step: 1230/1541, loss: 0.24074840545654297 2023-01-23 09:32:29.418133: step: 1232/1541, loss: 0.1947442591190338 2023-01-23 09:32:30.311133: step: 1234/1541, loss: 0.3535206615924835 2023-01-23 09:32:31.211780: step: 1236/1541, loss: 0.41342076659202576 2023-01-23 09:32:32.142425: step: 1238/1541, loss: 0.09683500230312347 2023-01-23 09:32:33.067361: step: 1240/1541, loss: 0.321214497089386 2023-01-23 09:32:34.031101: step: 1242/1541, loss: 0.017291944473981857 2023-01-23 09:32:34.908314: step: 1244/1541, loss: 0.048578567802906036 2023-01-23 09:32:35.772812: step: 1246/1541, loss: 0.2609523832798004 2023-01-23 09:32:36.672827: step: 1248/1541, loss: 0.4683876037597656 2023-01-23 09:32:37.513995: step: 1250/1541, loss: 0.17866843938827515 2023-01-23 09:32:38.558944: step: 1252/1541, loss: 0.038797587156295776 2023-01-23 09:32:39.466598: step: 1254/1541, loss: 0.03610469028353691 2023-01-23 09:32:40.336232: step: 1256/1541, loss: 0.08725883066654205 2023-01-23 09:32:41.217901: step: 1258/1541, loss: 0.06346090883016586 2023-01-23 09:32:42.097741: step: 1260/1541, loss: 0.3433014452457428 2023-01-23 09:32:42.985867: step: 1262/1541, loss: 0.0675526037812233 2023-01-23 09:32:43.938001: step: 1264/1541, loss: 2.635739326477051 2023-01-23 09:32:44.818439: step: 1266/1541, loss: 0.23807375133037567 2023-01-23 09:32:45.718599: step: 1268/1541, loss: 0.14259019494056702 2023-01-23 09:32:46.553747: step: 1270/1541, loss: 0.20458854734897614 2023-01-23 09:32:47.432100: step: 1272/1541, loss: 0.22421209514141083 2023-01-23 09:32:48.293171: step: 1274/1541, loss: 0.06451942026615143 2023-01-23 09:32:49.158212: step: 1276/1541, loss: 0.5921694040298462 2023-01-23 09:32:50.022402: step: 1278/1541, loss: 0.7103195786476135 2023-01-23 09:32:50.922685: step: 1280/1541, loss: 0.033203572034835815 2023-01-23 09:32:51.795606: step: 1282/1541, loss: 0.4405272901058197 2023-01-23 09:32:52.768830: step: 1284/1541, loss: 0.2502868175506592 2023-01-23 09:32:53.631956: step: 1286/1541, loss: 0.07896176725625992 2023-01-23 09:32:54.587795: step: 1288/1541, loss: 0.14283107221126556 2023-01-23 09:32:55.484868: step: 1290/1541, loss: 0.026771973818540573 2023-01-23 09:32:56.382399: step: 1292/1541, loss: 0.07682117074728012 2023-01-23 09:32:57.358635: step: 1294/1541, loss: 0.291003942489624 2023-01-23 09:32:58.344983: step: 1296/1541, loss: 0.17538605630397797 2023-01-23 09:32:59.261012: step: 1298/1541, loss: 0.20646214485168457 2023-01-23 09:33:00.124029: step: 1300/1541, loss: 0.1961369514465332 2023-01-23 09:33:01.006385: step: 1302/1541, loss: 0.45381051301956177 2023-01-23 09:33:01.874706: step: 1304/1541, loss: 0.07273972779512405 2023-01-23 09:33:02.790093: step: 1306/1541, loss: 0.14976651966571808 2023-01-23 09:33:03.738644: step: 1308/1541, loss: 0.22500422596931458 2023-01-23 09:33:04.598320: step: 1310/1541, loss: 1.3074034452438354 2023-01-23 09:33:05.471696: step: 1312/1541, loss: 0.05319596081972122 2023-01-23 09:33:06.402173: step: 1314/1541, loss: 0.27584969997406006 2023-01-23 09:33:07.264568: step: 1316/1541, loss: 0.147754967212677 2023-01-23 09:33:08.125770: step: 1318/1541, loss: 0.18469516932964325 2023-01-23 09:33:09.060188: step: 1320/1541, loss: 0.19076307117938995 2023-01-23 09:33:09.978490: step: 1322/1541, loss: 0.33881694078445435 2023-01-23 09:33:10.865226: step: 1324/1541, loss: 0.18019762635231018 2023-01-23 09:33:11.781865: step: 1326/1541, loss: 0.13619524240493774 2023-01-23 09:33:12.678196: step: 1328/1541, loss: 0.5838195085525513 2023-01-23 09:33:13.587346: step: 1330/1541, loss: 0.11764713376760483 2023-01-23 09:33:14.562421: step: 1332/1541, loss: 0.5059939622879028 2023-01-23 09:33:15.461685: step: 1334/1541, loss: 0.5833179950714111 2023-01-23 09:33:16.371188: step: 1336/1541, loss: 0.11056886613368988 2023-01-23 09:33:17.280402: step: 1338/1541, loss: 0.41658416390419006 2023-01-23 09:33:18.123480: step: 1340/1541, loss: 0.4191717207431793 2023-01-23 09:33:19.099107: step: 1342/1541, loss: 0.12713634967803955 2023-01-23 09:33:20.064977: step: 1344/1541, loss: 0.1894146054983139 2023-01-23 09:33:20.957547: step: 1346/1541, loss: 0.2253829389810562 2023-01-23 09:33:21.862706: step: 1348/1541, loss: 0.19540734589099884 2023-01-23 09:33:22.738693: step: 1350/1541, loss: 1.0439908504486084 2023-01-23 09:33:23.640252: step: 1352/1541, loss: 0.3524302840232849 2023-01-23 09:33:24.482525: step: 1354/1541, loss: 0.30207207798957825 2023-01-23 09:33:25.340178: step: 1356/1541, loss: 0.18539386987686157 2023-01-23 09:33:26.222519: step: 1358/1541, loss: 0.12720277905464172 2023-01-23 09:33:27.071832: step: 1360/1541, loss: 0.06244327127933502 2023-01-23 09:33:27.972344: step: 1362/1541, loss: 1.0843409299850464 2023-01-23 09:33:28.898471: step: 1364/1541, loss: 0.23893025517463684 2023-01-23 09:33:29.768985: step: 1366/1541, loss: 0.05703403428196907 2023-01-23 09:33:30.669250: step: 1368/1541, loss: 0.11465064436197281 2023-01-23 09:33:31.534083: step: 1370/1541, loss: 0.2179623395204544 2023-01-23 09:33:32.420668: step: 1372/1541, loss: 0.48341724276542664 2023-01-23 09:33:33.307921: step: 1374/1541, loss: 0.3357727825641632 2023-01-23 09:33:34.171722: step: 1376/1541, loss: 0.025457216426730156 2023-01-23 09:33:35.037695: step: 1378/1541, loss: 0.18791353702545166 2023-01-23 09:33:35.920482: step: 1380/1541, loss: 0.14003315567970276 2023-01-23 09:33:36.797434: step: 1382/1541, loss: 0.20736950635910034 2023-01-23 09:33:37.685421: step: 1384/1541, loss: 0.31402310729026794 2023-01-23 09:33:38.587069: step: 1386/1541, loss: 0.15951098501682281 2023-01-23 09:33:39.452974: step: 1388/1541, loss: 0.09498119354248047 2023-01-23 09:33:40.463804: step: 1390/1541, loss: 0.26857903599739075 2023-01-23 09:33:41.433923: step: 1392/1541, loss: 0.2215355932712555 2023-01-23 09:33:42.327876: step: 1394/1541, loss: 0.1655474305152893 2023-01-23 09:33:43.189589: step: 1396/1541, loss: 0.21713784337043762 2023-01-23 09:33:44.061698: step: 1398/1541, loss: 0.0813726857304573 2023-01-23 09:33:44.965200: step: 1400/1541, loss: 0.10161161422729492 2023-01-23 09:33:45.861362: step: 1402/1541, loss: 0.16113117337226868 2023-01-23 09:33:46.772710: step: 1404/1541, loss: 0.5317211747169495 2023-01-23 09:33:47.690078: step: 1406/1541, loss: 0.393004447221756 2023-01-23 09:33:48.611650: step: 1408/1541, loss: 0.17905357480049133 2023-01-23 09:33:49.513294: step: 1410/1541, loss: 0.09311681985855103 2023-01-23 09:33:50.403619: step: 1412/1541, loss: 0.38119223713874817 2023-01-23 09:33:51.318546: step: 1414/1541, loss: 0.09836100786924362 2023-01-23 09:33:52.167084: step: 1416/1541, loss: 0.09568000584840775 2023-01-23 09:33:53.092268: step: 1418/1541, loss: 0.17666777968406677 2023-01-23 09:33:53.971111: step: 1420/1541, loss: 0.29375705122947693 2023-01-23 09:33:54.886914: step: 1422/1541, loss: 0.19515225291252136 2023-01-23 09:33:55.794252: step: 1424/1541, loss: 0.6357325911521912 2023-01-23 09:33:56.670658: step: 1426/1541, loss: 0.07450495660305023 2023-01-23 09:33:57.566662: step: 1428/1541, loss: 0.47057151794433594 2023-01-23 09:33:58.445760: step: 1430/1541, loss: 0.07830962538719177 2023-01-23 09:33:59.377858: step: 1432/1541, loss: 0.4544040858745575 2023-01-23 09:34:00.302098: step: 1434/1541, loss: 0.07881249487400055 2023-01-23 09:34:01.188935: step: 1436/1541, loss: 0.18092301487922668 2023-01-23 09:34:02.071218: step: 1438/1541, loss: 0.31649401783943176 2023-01-23 09:34:03.047178: step: 1440/1541, loss: 0.164496049284935 2023-01-23 09:34:03.959122: step: 1442/1541, loss: 0.26147598028182983 2023-01-23 09:34:04.828936: step: 1444/1541, loss: 0.07481164485216141 2023-01-23 09:34:05.744219: step: 1446/1541, loss: 0.01725071668624878 2023-01-23 09:34:06.628639: step: 1448/1541, loss: 0.3183620274066925 2023-01-23 09:34:07.504087: step: 1450/1541, loss: 0.12366121262311935 2023-01-23 09:34:08.394371: step: 1452/1541, loss: 0.36392703652381897 2023-01-23 09:34:09.284654: step: 1454/1541, loss: 0.06011415645480156 2023-01-23 09:34:10.161235: step: 1456/1541, loss: 0.4270907938480377 2023-01-23 09:34:11.057306: step: 1458/1541, loss: 0.6847031712532043 2023-01-23 09:34:11.992416: step: 1460/1541, loss: 0.2519384026527405 2023-01-23 09:34:12.976991: step: 1462/1541, loss: 0.17485564947128296 2023-01-23 09:34:13.845960: step: 1464/1541, loss: 0.2606077194213867 2023-01-23 09:34:14.746308: step: 1466/1541, loss: 0.07374773919582367 2023-01-23 09:34:15.614027: step: 1468/1541, loss: 0.17902912199497223 2023-01-23 09:34:16.517510: step: 1470/1541, loss: 0.2924749553203583 2023-01-23 09:34:17.415477: step: 1472/1541, loss: 0.3448423147201538 2023-01-23 09:34:18.323338: step: 1474/1541, loss: 0.16835004091262817 2023-01-23 09:34:19.236911: step: 1476/1541, loss: 0.25398650765419006 2023-01-23 09:34:20.158610: step: 1478/1541, loss: 0.21890613436698914 2023-01-23 09:34:21.027394: step: 1480/1541, loss: 0.05765065550804138 2023-01-23 09:34:21.891539: step: 1482/1541, loss: 0.06441443413496017 2023-01-23 09:34:22.822176: step: 1484/1541, loss: 0.04458324983716011 2023-01-23 09:34:23.682803: step: 1486/1541, loss: 0.05086435377597809 2023-01-23 09:34:24.620951: step: 1488/1541, loss: 0.24995914101600647 2023-01-23 09:34:25.488076: step: 1490/1541, loss: 0.06924313306808472 2023-01-23 09:34:26.360847: step: 1492/1541, loss: 0.18685072660446167 2023-01-23 09:34:27.261904: step: 1494/1541, loss: 0.09914394468069077 2023-01-23 09:34:28.088482: step: 1496/1541, loss: 0.20670634508132935 2023-01-23 09:34:28.953815: step: 1498/1541, loss: 0.16532482206821442 2023-01-23 09:34:29.828862: step: 1500/1541, loss: 1.4280840158462524 2023-01-23 09:34:30.737536: step: 1502/1541, loss: 0.5472580790519714 2023-01-23 09:34:31.647075: step: 1504/1541, loss: 0.3311781883239746 2023-01-23 09:34:32.523781: step: 1506/1541, loss: 0.38131555914878845 2023-01-23 09:34:33.414641: step: 1508/1541, loss: 0.26764991879463196 2023-01-23 09:34:34.249124: step: 1510/1541, loss: 0.0781201720237732 2023-01-23 09:34:35.144990: step: 1512/1541, loss: 0.14999866485595703 2023-01-23 09:34:36.056905: step: 1514/1541, loss: 0.30355948209762573 2023-01-23 09:34:36.912629: step: 1516/1541, loss: 0.8383158445358276 2023-01-23 09:34:37.786306: step: 1518/1541, loss: 0.27126675844192505 2023-01-23 09:34:38.682773: step: 1520/1541, loss: 0.48461851477622986 2023-01-23 09:34:39.596210: step: 1522/1541, loss: 0.05318940430879593 2023-01-23 09:34:40.519408: step: 1524/1541, loss: 0.5827850103378296 2023-01-23 09:34:41.424484: step: 1526/1541, loss: 0.06983567774295807 2023-01-23 09:34:42.286842: step: 1528/1541, loss: 0.4004029631614685 2023-01-23 09:34:43.287052: step: 1530/1541, loss: 0.1605294644832611 2023-01-23 09:34:44.208953: step: 1532/1541, loss: 0.2105468064546585 2023-01-23 09:34:45.071011: step: 1534/1541, loss: 0.2614946961402893 2023-01-23 09:34:45.974891: step: 1536/1541, loss: 0.24795125424861908 2023-01-23 09:34:46.850020: step: 1538/1541, loss: 0.3172835409641266 2023-01-23 09:34:47.751614: step: 1540/1541, loss: 0.08712523430585861 2023-01-23 09:34:48.674528: step: 1542/1541, loss: 0.16321933269500732 2023-01-23 09:34:49.538896: step: 1544/1541, loss: 0.09958256781101227 2023-01-23 09:34:50.424972: step: 1546/1541, loss: 0.0995885580778122 2023-01-23 09:34:51.306798: step: 1548/1541, loss: 0.25513288378715515 2023-01-23 09:34:52.185207: step: 1550/1541, loss: 0.1413261741399765 2023-01-23 09:34:53.077280: step: 1552/1541, loss: 0.06573913246393204 2023-01-23 09:34:53.961915: step: 1554/1541, loss: 0.575734555721283 2023-01-23 09:34:54.943380: step: 1556/1541, loss: 0.41452324390411377 2023-01-23 09:34:55.822322: step: 1558/1541, loss: 0.2535116374492645 2023-01-23 09:34:56.636954: step: 1560/1541, loss: 0.1657170057296753 2023-01-23 09:34:57.505233: step: 1562/1541, loss: 0.08641941845417023 2023-01-23 09:34:58.407853: step: 1564/1541, loss: 0.0611468143761158 2023-01-23 09:34:59.273610: step: 1566/1541, loss: 0.154298335313797 2023-01-23 09:35:00.136844: step: 1568/1541, loss: 0.2236993908882141 2023-01-23 09:35:00.991060: step: 1570/1541, loss: 0.8576454520225525 2023-01-23 09:35:01.892202: step: 1572/1541, loss: 0.12627620995044708 2023-01-23 09:35:02.776102: step: 1574/1541, loss: 0.16728124022483826 2023-01-23 09:35:03.654509: step: 1576/1541, loss: 0.0779535323381424 2023-01-23 09:35:04.524468: step: 1578/1541, loss: 0.322355717420578 2023-01-23 09:35:05.411211: step: 1580/1541, loss: 0.043120041489601135 2023-01-23 09:35:06.282105: step: 1582/1541, loss: 0.09650524705648422 2023-01-23 09:35:07.161681: step: 1584/1541, loss: 0.02738642506301403 2023-01-23 09:35:08.048334: step: 1586/1541, loss: 0.09908054769039154 2023-01-23 09:35:08.905334: step: 1588/1541, loss: 0.1040741577744484 2023-01-23 09:35:09.780586: step: 1590/1541, loss: 0.4166569709777832 2023-01-23 09:35:10.659052: step: 1592/1541, loss: 0.07165475189685822 2023-01-23 09:35:11.526488: step: 1594/1541, loss: 0.05516029894351959 2023-01-23 09:35:12.416197: step: 1596/1541, loss: 0.37620311975479126 2023-01-23 09:35:13.335819: step: 1598/1541, loss: 0.25950515270233154 2023-01-23 09:35:14.234921: step: 1600/1541, loss: 0.34326791763305664 2023-01-23 09:35:15.136604: step: 1602/1541, loss: 0.017869535833597183 2023-01-23 09:35:16.047290: step: 1604/1541, loss: 0.13227568566799164 2023-01-23 09:35:16.948685: step: 1606/1541, loss: 0.45054495334625244 2023-01-23 09:35:17.836483: step: 1608/1541, loss: 0.08730431646108627 2023-01-23 09:35:18.707570: step: 1610/1541, loss: 0.1587425172328949 2023-01-23 09:35:19.623344: step: 1612/1541, loss: 0.15582124888896942 2023-01-23 09:35:20.499558: step: 1614/1541, loss: 0.023945942521095276 2023-01-23 09:35:21.340929: step: 1616/1541, loss: 0.0886530727148056 2023-01-23 09:35:22.317698: step: 1618/1541, loss: 0.31007054448127747 2023-01-23 09:35:23.188111: step: 1620/1541, loss: 0.1804034262895584 2023-01-23 09:35:24.093763: step: 1622/1541, loss: 0.20646077394485474 2023-01-23 09:35:24.910153: step: 1624/1541, loss: 0.0515245720744133 2023-01-23 09:35:25.866523: step: 1626/1541, loss: 0.49136626720428467 2023-01-23 09:35:26.757382: step: 1628/1541, loss: 0.08872933685779572 2023-01-23 09:35:27.661381: step: 1630/1541, loss: 0.23828352987766266 2023-01-23 09:35:28.533104: step: 1632/1541, loss: 0.07590851187705994 2023-01-23 09:35:29.416435: step: 1634/1541, loss: 0.10091445595026016 2023-01-23 09:35:30.279708: step: 1636/1541, loss: 0.03410471975803375 2023-01-23 09:35:31.248503: step: 1638/1541, loss: 0.06077847629785538 2023-01-23 09:35:32.221789: step: 1640/1541, loss: 0.5647609829902649 2023-01-23 09:35:33.135681: step: 1642/1541, loss: 0.28936728835105896 2023-01-23 09:35:33.981160: step: 1644/1541, loss: 0.13047082722187042 2023-01-23 09:35:34.934968: step: 1646/1541, loss: 0.33953994512557983 2023-01-23 09:35:35.824175: step: 1648/1541, loss: 0.4762602746486664 2023-01-23 09:35:36.736378: step: 1650/1541, loss: 1.149779200553894 2023-01-23 09:35:37.622317: step: 1652/1541, loss: 0.38663962483406067 2023-01-23 09:35:38.513806: step: 1654/1541, loss: 0.39733466506004333 2023-01-23 09:35:39.384107: step: 1656/1541, loss: 0.4769030809402466 2023-01-23 09:35:40.287544: step: 1658/1541, loss: 0.6401698589324951 2023-01-23 09:35:41.200916: step: 1660/1541, loss: 1.1467839479446411 2023-01-23 09:35:42.103744: step: 1662/1541, loss: 0.04300631582736969 2023-01-23 09:35:42.983021: step: 1664/1541, loss: 0.71736079454422 2023-01-23 09:35:43.856108: step: 1666/1541, loss: 0.09583216905593872 2023-01-23 09:35:44.741645: step: 1668/1541, loss: 0.17247837781906128 2023-01-23 09:35:45.655387: step: 1670/1541, loss: 0.5506844520568848 2023-01-23 09:35:46.567456: step: 1672/1541, loss: 0.2673543393611908 2023-01-23 09:35:47.441344: step: 1674/1541, loss: 4.243233680725098 2023-01-23 09:35:48.318387: step: 1676/1541, loss: 0.25966402888298035 2023-01-23 09:35:49.223178: step: 1678/1541, loss: 0.09724392741918564 2023-01-23 09:35:50.101090: step: 1680/1541, loss: 0.07899001985788345 2023-01-23 09:35:50.992851: step: 1682/1541, loss: 2.571840763092041 2023-01-23 09:35:51.849915: step: 1684/1541, loss: 0.12488716095685959 2023-01-23 09:35:52.715577: step: 1686/1541, loss: 0.3399791419506073 2023-01-23 09:35:53.589920: step: 1688/1541, loss: 0.6123899221420288 2023-01-23 09:35:54.485198: step: 1690/1541, loss: 1.082729458808899 2023-01-23 09:35:55.325989: step: 1692/1541, loss: 0.22370576858520508 2023-01-23 09:35:56.217994: step: 1694/1541, loss: 0.4328305125236511 2023-01-23 09:35:57.140780: step: 1696/1541, loss: 0.20761308073997498 2023-01-23 09:35:58.098325: step: 1698/1541, loss: 0.50253826379776 2023-01-23 09:35:58.988673: step: 1700/1541, loss: 0.35749560594558716 2023-01-23 09:35:59.870130: step: 1702/1541, loss: 0.8282798528671265 2023-01-23 09:36:00.755167: step: 1704/1541, loss: 0.3155874013900757 2023-01-23 09:36:01.620451: step: 1706/1541, loss: 0.13016916811466217 2023-01-23 09:36:02.497832: step: 1708/1541, loss: 0.5141702890396118 2023-01-23 09:36:03.411952: step: 1710/1541, loss: 0.20249348878860474 2023-01-23 09:36:04.284749: step: 1712/1541, loss: 0.10515248775482178 2023-01-23 09:36:05.134719: step: 1714/1541, loss: 0.06060458719730377 2023-01-23 09:36:05.999543: step: 1716/1541, loss: 0.11855172365903854 2023-01-23 09:36:06.966823: step: 1718/1541, loss: 0.13798770308494568 2023-01-23 09:36:07.928642: step: 1720/1541, loss: 0.5028994679450989 2023-01-23 09:36:08.838900: step: 1722/1541, loss: 0.3039814233779907 2023-01-23 09:36:09.746443: step: 1724/1541, loss: 0.08186756819486618 2023-01-23 09:36:10.622358: step: 1726/1541, loss: 0.10327520966529846 2023-01-23 09:36:11.473934: step: 1728/1541, loss: 0.10719671100378036 2023-01-23 09:36:12.322293: step: 1730/1541, loss: 0.14884844422340393 2023-01-23 09:36:13.179553: step: 1732/1541, loss: 0.3104710280895233 2023-01-23 09:36:14.087243: step: 1734/1541, loss: 0.25822746753692627 2023-01-23 09:36:14.974278: step: 1736/1541, loss: 0.10533695667982101 2023-01-23 09:36:15.854048: step: 1738/1541, loss: 0.2851044535636902 2023-01-23 09:36:16.744766: step: 1740/1541, loss: 4.553930282592773 2023-01-23 09:36:17.722798: step: 1742/1541, loss: 0.05980926752090454 2023-01-23 09:36:18.606646: step: 1744/1541, loss: 2.544339179992676 2023-01-23 09:36:19.521092: step: 1746/1541, loss: 0.08052535355091095 2023-01-23 09:36:20.450691: step: 1748/1541, loss: 0.42791157960891724 2023-01-23 09:36:21.342614: step: 1750/1541, loss: 0.014388270676136017 2023-01-23 09:36:22.222256: step: 1752/1541, loss: 0.40080851316452026 2023-01-23 09:36:23.077344: step: 1754/1541, loss: 0.09339556097984314 2023-01-23 09:36:23.950463: step: 1756/1541, loss: 0.031314633786678314 2023-01-23 09:36:24.840370: step: 1758/1541, loss: 0.1540186107158661 2023-01-23 09:36:25.781950: step: 1760/1541, loss: 0.8018968105316162 2023-01-23 09:36:26.672452: step: 1762/1541, loss: 0.2291809618473053 2023-01-23 09:36:27.586626: step: 1764/1541, loss: 0.3917151987552643 2023-01-23 09:36:28.483597: step: 1766/1541, loss: 0.3390170633792877 2023-01-23 09:36:29.400367: step: 1768/1541, loss: 0.17521804571151733 2023-01-23 09:36:30.298567: step: 1770/1541, loss: 0.06710376590490341 2023-01-23 09:36:31.179734: step: 1772/1541, loss: 0.32515132427215576 2023-01-23 09:36:32.041965: step: 1774/1541, loss: 0.11223306506872177 2023-01-23 09:36:32.905374: step: 1776/1541, loss: 0.08381815254688263 2023-01-23 09:36:33.816394: step: 1778/1541, loss: 1.2516357898712158 2023-01-23 09:36:34.691903: step: 1780/1541, loss: 0.2823198139667511 2023-01-23 09:36:35.586835: step: 1782/1541, loss: 0.1866491436958313 2023-01-23 09:36:36.504822: step: 1784/1541, loss: 0.5148088335990906 2023-01-23 09:36:37.393896: step: 1786/1541, loss: 0.27484315633773804 2023-01-23 09:36:38.263877: step: 1788/1541, loss: 0.19865386188030243 2023-01-23 09:36:39.142324: step: 1790/1541, loss: 0.5482164025306702 2023-01-23 09:36:40.033617: step: 1792/1541, loss: 0.4812394678592682 2023-01-23 09:36:40.914782: step: 1794/1541, loss: 0.23213084042072296 2023-01-23 09:36:41.830475: step: 1796/1541, loss: 0.31502488255500793 2023-01-23 09:36:42.766807: step: 1798/1541, loss: 0.1967175304889679 2023-01-23 09:36:43.644988: step: 1800/1541, loss: 0.4033498167991638 2023-01-23 09:36:44.545178: step: 1802/1541, loss: 0.28225308656692505 2023-01-23 09:36:45.376818: step: 1804/1541, loss: 0.18753395974636078 2023-01-23 09:36:46.274872: step: 1806/1541, loss: 0.1880318820476532 2023-01-23 09:36:47.244401: step: 1808/1541, loss: 0.6968146562576294 2023-01-23 09:36:48.116487: step: 1810/1541, loss: 0.21195711195468903 2023-01-23 09:36:49.056675: step: 1812/1541, loss: 0.09065788239240646 2023-01-23 09:36:49.945439: step: 1814/1541, loss: 0.2732725739479065 2023-01-23 09:36:50.829598: step: 1816/1541, loss: 1.038291573524475 2023-01-23 09:36:51.802508: step: 1818/1541, loss: 0.33096474409103394 2023-01-23 09:36:52.652407: step: 1820/1541, loss: 0.072660431265831 2023-01-23 09:36:53.530994: step: 1822/1541, loss: 0.5528992414474487 2023-01-23 09:36:54.420614: step: 1824/1541, loss: 0.36143529415130615 2023-01-23 09:36:55.255568: step: 1826/1541, loss: 0.24060866236686707 2023-01-23 09:36:56.128730: step: 1828/1541, loss: 0.6401172280311584 2023-01-23 09:36:56.978163: step: 1830/1541, loss: 0.4591028094291687 2023-01-23 09:36:57.856038: step: 1832/1541, loss: 0.5113583207130432 2023-01-23 09:36:58.747393: step: 1834/1541, loss: 0.3518529236316681 2023-01-23 09:36:59.742461: step: 1836/1541, loss: 0.6810919642448425 2023-01-23 09:37:00.652258: step: 1838/1541, loss: 0.22136667370796204 2023-01-23 09:37:01.554901: step: 1840/1541, loss: 0.1331520527601242 2023-01-23 09:37:02.457166: step: 1842/1541, loss: 0.23428678512573242 2023-01-23 09:37:03.345204: step: 1844/1541, loss: 0.5783284902572632 2023-01-23 09:37:04.237257: step: 1846/1541, loss: 0.10161332786083221 2023-01-23 09:37:05.145655: step: 1848/1541, loss: 0.15115362405776978 2023-01-23 09:37:06.035252: step: 1850/1541, loss: 0.3607369363307953 2023-01-23 09:37:06.932335: step: 1852/1541, loss: 0.125215545296669 2023-01-23 09:37:07.846257: step: 1854/1541, loss: 0.13405010104179382 2023-01-23 09:37:08.707248: step: 1856/1541, loss: 0.18084149062633514 2023-01-23 09:37:09.643032: step: 1858/1541, loss: 0.0850292518734932 2023-01-23 09:37:10.551331: step: 1860/1541, loss: 0.21013206243515015 2023-01-23 09:37:11.431361: step: 1862/1541, loss: 0.16041803359985352 2023-01-23 09:37:12.373304: step: 1864/1541, loss: 0.20681187510490417 2023-01-23 09:37:13.257555: step: 1866/1541, loss: 0.1751009076833725 2023-01-23 09:37:14.149401: step: 1868/1541, loss: 0.2697794437408447 2023-01-23 09:37:15.078279: step: 1870/1541, loss: 0.42999255657196045 2023-01-23 09:37:15.910865: step: 1872/1541, loss: 0.16977714002132416 2023-01-23 09:37:16.813429: step: 1874/1541, loss: 0.28100278973579407 2023-01-23 09:37:17.698339: step: 1876/1541, loss: 0.06501683592796326 2023-01-23 09:37:18.579708: step: 1878/1541, loss: 0.5323101878166199 2023-01-23 09:37:19.437219: step: 1880/1541, loss: 0.36311131715774536 2023-01-23 09:37:20.307107: step: 1882/1541, loss: 0.4334944486618042 2023-01-23 09:37:21.183154: step: 1884/1541, loss: 0.34743908047676086 2023-01-23 09:37:22.058677: step: 1886/1541, loss: 0.693647563457489 2023-01-23 09:37:22.926654: step: 1888/1541, loss: 0.23870639503002167 2023-01-23 09:37:23.812581: step: 1890/1541, loss: 0.02590460702776909 2023-01-23 09:37:24.710633: step: 1892/1541, loss: 1.105580449104309 2023-01-23 09:37:25.618034: step: 1894/1541, loss: 0.3869720995426178 2023-01-23 09:37:26.562128: step: 1896/1541, loss: 0.2053702175617218 2023-01-23 09:37:27.481312: step: 1898/1541, loss: 6.137970447540283 2023-01-23 09:37:28.346835: step: 1900/1541, loss: 0.1926175057888031 2023-01-23 09:37:29.206028: step: 1902/1541, loss: 0.06598333269357681 2023-01-23 09:37:30.084063: step: 1904/1541, loss: 0.7698848247528076 2023-01-23 09:37:30.967849: step: 1906/1541, loss: 0.0619351789355278 2023-01-23 09:37:31.890529: step: 1908/1541, loss: 2.3361990451812744 2023-01-23 09:37:32.790690: step: 1910/1541, loss: 0.24670733511447906 2023-01-23 09:37:33.709286: step: 1912/1541, loss: 0.304941326379776 2023-01-23 09:37:34.581068: step: 1914/1541, loss: 0.11054433137178421 2023-01-23 09:37:35.480309: step: 1916/1541, loss: 0.15419155359268188 2023-01-23 09:37:36.436976: step: 1918/1541, loss: 0.10035141557455063 2023-01-23 09:37:37.363954: step: 1920/1541, loss: 0.6010708808898926 2023-01-23 09:37:38.248053: step: 1922/1541, loss: 0.2548263370990753 2023-01-23 09:37:39.133006: step: 1924/1541, loss: 0.16073299944400787 2023-01-23 09:37:40.074025: step: 1926/1541, loss: 0.4150567352771759 2023-01-23 09:37:41.003181: step: 1928/1541, loss: 0.5023674964904785 2023-01-23 09:37:41.893734: step: 1930/1541, loss: 0.1644505262374878 2023-01-23 09:37:42.828812: step: 1932/1541, loss: 0.08452413976192474 2023-01-23 09:37:43.715415: step: 1934/1541, loss: 0.18050691485404968 2023-01-23 09:37:44.597904: step: 1936/1541, loss: 0.07743747532367706 2023-01-23 09:37:45.444390: step: 1938/1541, loss: 0.1486462950706482 2023-01-23 09:37:46.353751: step: 1940/1541, loss: 0.2557002007961273 2023-01-23 09:37:47.263049: step: 1942/1541, loss: 0.09479693323373795 2023-01-23 09:37:48.246102: step: 1944/1541, loss: 1.4233802556991577 2023-01-23 09:37:49.195840: step: 1946/1541, loss: 0.13588453829288483 2023-01-23 09:37:50.116334: step: 1948/1541, loss: 0.08173694461584091 2023-01-23 09:37:51.021611: step: 1950/1541, loss: 0.5758231282234192 2023-01-23 09:37:51.924363: step: 1952/1541, loss: 0.07601810246706009 2023-01-23 09:37:52.823749: step: 1954/1541, loss: 0.14326471090316772 2023-01-23 09:37:53.751112: step: 1956/1541, loss: 0.04926428198814392 2023-01-23 09:37:54.627778: step: 1958/1541, loss: 0.20052292943000793 2023-01-23 09:37:55.508778: step: 1960/1541, loss: 0.5536402463912964 2023-01-23 09:37:56.414452: step: 1962/1541, loss: 0.6194049715995789 2023-01-23 09:37:57.359141: step: 1964/1541, loss: 0.3253398537635803 2023-01-23 09:37:58.261573: step: 1966/1541, loss: 0.22276833653450012 2023-01-23 09:37:59.182292: step: 1968/1541, loss: 0.14099344611167908 2023-01-23 09:38:00.091699: step: 1970/1541, loss: 0.5848124623298645 2023-01-23 09:38:00.939223: step: 1972/1541, loss: 0.13944724202156067 2023-01-23 09:38:01.793718: step: 1974/1541, loss: 0.25872117280960083 2023-01-23 09:38:02.693684: step: 1976/1541, loss: 0.10308411717414856 2023-01-23 09:38:03.580265: step: 1978/1541, loss: 0.30396953225135803 2023-01-23 09:38:04.447434: step: 1980/1541, loss: 0.1786070615053177 2023-01-23 09:38:05.362125: step: 1982/1541, loss: 0.3834533393383026 2023-01-23 09:38:06.221207: step: 1984/1541, loss: 0.046240467578172684 2023-01-23 09:38:07.175502: step: 1986/1541, loss: 0.7104818224906921 2023-01-23 09:38:08.048711: step: 1988/1541, loss: 0.15149785578250885 2023-01-23 09:38:09.014469: step: 1990/1541, loss: 0.11801578104496002 2023-01-23 09:38:09.930208: step: 1992/1541, loss: 0.7594659924507141 2023-01-23 09:38:10.823534: step: 1994/1541, loss: 0.5795380473136902 2023-01-23 09:38:11.720009: step: 1996/1541, loss: 0.24946822226047516 2023-01-23 09:38:12.599196: step: 1998/1541, loss: 0.07266542315483093 2023-01-23 09:38:13.482454: step: 2000/1541, loss: 0.3390006124973297 2023-01-23 09:38:14.372324: step: 2002/1541, loss: 0.07876035571098328 2023-01-23 09:38:15.255422: step: 2004/1541, loss: 0.11363330483436584 2023-01-23 09:38:16.132087: step: 2006/1541, loss: 0.694266140460968 2023-01-23 09:38:17.004483: step: 2008/1541, loss: 0.19077494740486145 2023-01-23 09:38:17.876369: step: 2010/1541, loss: 0.2764683961868286 2023-01-23 09:38:18.769966: step: 2012/1541, loss: 0.2517140507698059 2023-01-23 09:38:19.642157: step: 2014/1541, loss: 0.06704089045524597 2023-01-23 09:38:20.588880: step: 2016/1541, loss: 0.06949549168348312 2023-01-23 09:38:21.465015: step: 2018/1541, loss: 0.030724313110113144 2023-01-23 09:38:22.361319: step: 2020/1541, loss: 0.2038574516773224 2023-01-23 09:38:23.231501: step: 2022/1541, loss: 2.6024770736694336 2023-01-23 09:38:24.097098: step: 2024/1541, loss: 0.46229904890060425 2023-01-23 09:38:24.925506: step: 2026/1541, loss: 0.2117437720298767 2023-01-23 09:38:25.785600: step: 2028/1541, loss: 0.1735517978668213 2023-01-23 09:38:26.669945: step: 2030/1541, loss: 0.09016828238964081 2023-01-23 09:38:27.580755: step: 2032/1541, loss: 0.11768047511577606 2023-01-23 09:38:28.501597: step: 2034/1541, loss: 0.1651037633419037 2023-01-23 09:38:29.408388: step: 2036/1541, loss: 0.6488091349601746 2023-01-23 09:38:30.329171: step: 2038/1541, loss: 0.3028683662414551 2023-01-23 09:38:31.300992: step: 2040/1541, loss: 0.024501699954271317 2023-01-23 09:38:32.217669: step: 2042/1541, loss: 0.5248585343360901 2023-01-23 09:38:33.090654: step: 2044/1541, loss: 0.022809751331806183 2023-01-23 09:38:34.003689: step: 2046/1541, loss: 0.35326555371284485 2023-01-23 09:38:34.912544: step: 2048/1541, loss: 0.11297497153282166 2023-01-23 09:38:35.879932: step: 2050/1541, loss: 0.050709012895822525 2023-01-23 09:38:36.761972: step: 2052/1541, loss: 0.0886891633272171 2023-01-23 09:38:37.702430: step: 2054/1541, loss: 0.8011775612831116 2023-01-23 09:38:38.568493: step: 2056/1541, loss: 0.4314846694469452 2023-01-23 09:38:39.464836: step: 2058/1541, loss: 0.758881151676178 2023-01-23 09:38:40.324264: step: 2060/1541, loss: 0.12116606533527374 2023-01-23 09:38:41.211625: step: 2062/1541, loss: 0.05562435835599899 2023-01-23 09:38:42.096392: step: 2064/1541, loss: 0.4713073670864105 2023-01-23 09:38:42.999601: step: 2066/1541, loss: 0.3830731511116028 2023-01-23 09:38:43.895754: step: 2068/1541, loss: 0.329985111951828 2023-01-23 09:38:44.788441: step: 2070/1541, loss: 0.07311759889125824 2023-01-23 09:38:45.677482: step: 2072/1541, loss: 0.14033378660678864 2023-01-23 09:38:46.605368: step: 2074/1541, loss: 0.1764543354511261 2023-01-23 09:38:47.542492: step: 2076/1541, loss: 0.09703803062438965 2023-01-23 09:38:48.480913: step: 2078/1541, loss: 0.8095695972442627 2023-01-23 09:38:49.377262: step: 2080/1541, loss: 0.1744304746389389 2023-01-23 09:38:50.280328: step: 2082/1541, loss: 0.2776558995246887 2023-01-23 09:38:51.161110: step: 2084/1541, loss: 0.8111638426780701 2023-01-23 09:38:52.010649: step: 2086/1541, loss: 0.3494884967803955 2023-01-23 09:38:52.906611: step: 2088/1541, loss: 0.14721907675266266 2023-01-23 09:38:53.791332: step: 2090/1541, loss: 0.29001787304878235 2023-01-23 09:38:54.639157: step: 2092/1541, loss: 0.33349186182022095 2023-01-23 09:38:55.578625: step: 2094/1541, loss: 0.7510465979576111 2023-01-23 09:38:56.450421: step: 2096/1541, loss: 0.06653910130262375 2023-01-23 09:38:57.336042: step: 2098/1541, loss: 0.11598420888185501 2023-01-23 09:38:58.259038: step: 2100/1541, loss: 0.8000307083129883 2023-01-23 09:38:59.122262: step: 2102/1541, loss: 0.09228885173797607 2023-01-23 09:39:00.002291: step: 2104/1541, loss: 0.2312598079442978 2023-01-23 09:39:00.913786: step: 2106/1541, loss: 0.2249191701412201 2023-01-23 09:39:01.802813: step: 2108/1541, loss: 6.382078170776367 2023-01-23 09:39:02.705588: step: 2110/1541, loss: 0.41585612297058105 2023-01-23 09:39:03.645391: step: 2112/1541, loss: 0.5922414660453796 2023-01-23 09:39:04.567142: step: 2114/1541, loss: 0.04515157267451286 2023-01-23 09:39:05.474247: step: 2116/1541, loss: 0.2122543901205063 2023-01-23 09:39:06.359191: step: 2118/1541, loss: 0.7412799000740051 2023-01-23 09:39:07.260573: step: 2120/1541, loss: 0.6862457394599915 2023-01-23 09:39:08.153918: step: 2122/1541, loss: 1.0326985120773315 2023-01-23 09:39:09.013634: step: 2124/1541, loss: 0.521982729434967 2023-01-23 09:39:09.917679: step: 2126/1541, loss: 0.15911825001239777 2023-01-23 09:39:10.837661: step: 2128/1541, loss: 0.3115053176879883 2023-01-23 09:39:11.717270: step: 2130/1541, loss: 0.09009471535682678 2023-01-23 09:39:12.625347: step: 2132/1541, loss: 0.16623635590076447 2023-01-23 09:39:13.489659: step: 2134/1541, loss: 0.11468185484409332 2023-01-23 09:39:14.351752: step: 2136/1541, loss: 0.10301192849874496 2023-01-23 09:39:15.230056: step: 2138/1541, loss: 0.25942790508270264 2023-01-23 09:39:16.111607: step: 2140/1541, loss: 0.08448849618434906 2023-01-23 09:39:16.989370: step: 2142/1541, loss: 0.3796858489513397 2023-01-23 09:39:17.870645: step: 2144/1541, loss: 0.2636389136314392 2023-01-23 09:39:18.751589: step: 2146/1541, loss: 0.45449692010879517 2023-01-23 09:39:19.638997: step: 2148/1541, loss: 0.05563736706972122 2023-01-23 09:39:20.583273: step: 2150/1541, loss: 0.23123109340667725 2023-01-23 09:39:21.517106: step: 2152/1541, loss: 0.10570797324180603 2023-01-23 09:39:22.397319: step: 2154/1541, loss: 0.07320284843444824 2023-01-23 09:39:23.266230: step: 2156/1541, loss: 0.14361420273780823 2023-01-23 09:39:24.160777: step: 2158/1541, loss: 0.26889801025390625 2023-01-23 09:39:25.082640: step: 2160/1541, loss: 0.13426469266414642 2023-01-23 09:39:25.938218: step: 2162/1541, loss: 0.3992423713207245 2023-01-23 09:39:26.789169: step: 2164/1541, loss: 0.050036340951919556 2023-01-23 09:39:27.685356: step: 2166/1541, loss: 0.16986224055290222 2023-01-23 09:39:28.558066: step: 2168/1541, loss: 0.12985245883464813 2023-01-23 09:39:29.493736: step: 2170/1541, loss: 0.5709704756736755 2023-01-23 09:39:30.412386: step: 2172/1541, loss: 0.02235233224928379 2023-01-23 09:39:31.302461: step: 2174/1541, loss: 0.19080349802970886 2023-01-23 09:39:32.189168: step: 2176/1541, loss: 0.7013010382652283 2023-01-23 09:39:33.066631: step: 2178/1541, loss: 0.13114720582962036 2023-01-23 09:39:33.930033: step: 2180/1541, loss: 0.08206647634506226 2023-01-23 09:39:34.828378: step: 2182/1541, loss: 0.5219153165817261 2023-01-23 09:39:35.721999: step: 2184/1541, loss: 0.18216918408870697 2023-01-23 09:39:36.588379: step: 2186/1541, loss: 0.15750931203365326 2023-01-23 09:39:37.490771: step: 2188/1541, loss: 0.341234028339386 2023-01-23 09:39:38.339890: step: 2190/1541, loss: 0.7825143337249756 2023-01-23 09:39:39.201639: step: 2192/1541, loss: 0.3282555043697357 2023-01-23 09:39:40.096413: step: 2194/1541, loss: 0.14554019272327423 2023-01-23 09:39:40.985137: step: 2196/1541, loss: 0.4375821650028229 2023-01-23 09:39:41.863916: step: 2198/1541, loss: 0.33718857169151306 2023-01-23 09:39:42.786873: step: 2200/1541, loss: 0.03324282169342041 2023-01-23 09:39:43.638201: step: 2202/1541, loss: 0.14688155055046082 2023-01-23 09:39:44.496936: step: 2204/1541, loss: 0.3857566714286804 2023-01-23 09:39:45.406126: step: 2206/1541, loss: 0.262445330619812 2023-01-23 09:39:46.280493: step: 2208/1541, loss: 0.13835901021957397 2023-01-23 09:39:47.173637: step: 2210/1541, loss: 0.17560362815856934 2023-01-23 09:39:48.081245: step: 2212/1541, loss: 0.3877340257167816 2023-01-23 09:39:48.943310: step: 2214/1541, loss: 0.11090201139450073 2023-01-23 09:39:49.813301: step: 2216/1541, loss: 0.3901270925998688 2023-01-23 09:39:50.727582: step: 2218/1541, loss: 0.03038395196199417 2023-01-23 09:39:51.657724: step: 2220/1541, loss: 0.3543100953102112 2023-01-23 09:39:52.543452: step: 2222/1541, loss: 0.028543565422296524 2023-01-23 09:39:53.433638: step: 2224/1541, loss: 2.3428165912628174 2023-01-23 09:39:54.314556: step: 2226/1541, loss: 0.20254352688789368 2023-01-23 09:39:55.215297: step: 2228/1541, loss: 0.18456320464611053 2023-01-23 09:39:56.121129: step: 2230/1541, loss: 0.27217987179756165 2023-01-23 09:39:56.998183: step: 2232/1541, loss: 0.5628343820571899 2023-01-23 09:39:57.874417: step: 2234/1541, loss: 0.305797815322876 2023-01-23 09:39:58.806704: step: 2236/1541, loss: 0.06780172884464264 2023-01-23 09:39:59.691183: step: 2238/1541, loss: 0.302782267332077 2023-01-23 09:40:00.556000: step: 2240/1541, loss: 0.14542090892791748 2023-01-23 09:40:01.427254: step: 2242/1541, loss: 0.19596144556999207 2023-01-23 09:40:02.317015: step: 2244/1541, loss: 0.2680083215236664 2023-01-23 09:40:03.211493: step: 2246/1541, loss: 0.6320004463195801 2023-01-23 09:40:04.078474: step: 2248/1541, loss: 0.3547893166542053 2023-01-23 09:40:05.017493: step: 2250/1541, loss: 0.01303697656840086 2023-01-23 09:40:05.874759: step: 2252/1541, loss: 0.12386550009250641 2023-01-23 09:40:06.756541: step: 2254/1541, loss: 0.07782914489507675 2023-01-23 09:40:07.658795: step: 2256/1541, loss: 0.08026327192783356 2023-01-23 09:40:08.586297: step: 2258/1541, loss: 0.1617620289325714 2023-01-23 09:40:09.434968: step: 2260/1541, loss: 0.6244106888771057 2023-01-23 09:40:10.403028: step: 2262/1541, loss: 0.11956135928630829 2023-01-23 09:40:11.286858: step: 2264/1541, loss: 0.07545401155948639 2023-01-23 09:40:12.151428: step: 2266/1541, loss: 0.2762607932090759 2023-01-23 09:40:13.128324: step: 2268/1541, loss: 0.3093870282173157 2023-01-23 09:40:13.974603: step: 2270/1541, loss: 0.4201737344264984 2023-01-23 09:40:14.873071: step: 2272/1541, loss: 0.020697979256510735 2023-01-23 09:40:15.785608: step: 2274/1541, loss: 0.19356665015220642 2023-01-23 09:40:16.646664: step: 2276/1541, loss: 0.08792527765035629 2023-01-23 09:40:17.553508: step: 2278/1541, loss: 0.4176226854324341 2023-01-23 09:40:18.453435: step: 2280/1541, loss: 0.2981446087360382 2023-01-23 09:40:19.356196: step: 2282/1541, loss: 0.1982073038816452 2023-01-23 09:40:20.212648: step: 2284/1541, loss: 3.453902244567871 2023-01-23 09:40:21.079261: step: 2286/1541, loss: 0.18672354519367218 2023-01-23 09:40:21.932953: step: 2288/1541, loss: 0.2797592580318451 2023-01-23 09:40:22.900061: step: 2290/1541, loss: 0.28315865993499756 2023-01-23 09:40:23.751575: step: 2292/1541, loss: 0.09996242821216583 2023-01-23 09:40:24.625863: step: 2294/1541, loss: 0.4089702069759369 2023-01-23 09:40:25.489247: step: 2296/1541, loss: 0.15736187994480133 2023-01-23 09:40:26.379249: step: 2298/1541, loss: 0.2537373900413513 2023-01-23 09:40:27.274678: step: 2300/1541, loss: 0.3471985459327698 2023-01-23 09:40:28.165276: step: 2302/1541, loss: 0.10346130281686783 2023-01-23 09:40:29.037706: step: 2304/1541, loss: 0.12460405379533768 2023-01-23 09:40:29.890672: step: 2306/1541, loss: 0.09464776515960693 2023-01-23 09:40:30.740908: step: 2308/1541, loss: 0.2545294463634491 2023-01-23 09:40:31.597853: step: 2310/1541, loss: 0.048138827085494995 2023-01-23 09:40:32.489030: step: 2312/1541, loss: 0.2750270664691925 2023-01-23 09:40:33.419774: step: 2314/1541, loss: 0.2450491487979889 2023-01-23 09:40:34.288316: step: 2316/1541, loss: 0.7134926319122314 2023-01-23 09:40:35.144751: step: 2318/1541, loss: 0.34036317467689514 2023-01-23 09:40:36.066092: step: 2320/1541, loss: 0.12865766882896423 2023-01-23 09:40:36.948100: step: 2322/1541, loss: 0.7697577476501465 2023-01-23 09:40:37.832184: step: 2324/1541, loss: 0.1969575732946396 2023-01-23 09:40:38.753165: step: 2326/1541, loss: 0.12152724713087082 2023-01-23 09:40:39.619276: step: 2328/1541, loss: 0.3691754639148712 2023-01-23 09:40:40.489519: step: 2330/1541, loss: 4.720114707946777 2023-01-23 09:40:41.402228: step: 2332/1541, loss: 0.042505260556936264 2023-01-23 09:40:42.271236: step: 2334/1541, loss: 0.2884078323841095 2023-01-23 09:40:43.149543: step: 2336/1541, loss: 0.4213843047618866 2023-01-23 09:40:44.016691: step: 2338/1541, loss: 0.3425132930278778 2023-01-23 09:40:44.933141: step: 2340/1541, loss: 0.05832051485776901 2023-01-23 09:40:45.810028: step: 2342/1541, loss: 0.159187912940979 2023-01-23 09:40:46.718542: step: 2344/1541, loss: 0.34925931692123413 2023-01-23 09:40:47.587505: step: 2346/1541, loss: 0.3482411503791809 2023-01-23 09:40:48.484825: step: 2348/1541, loss: 0.0627594143152237 2023-01-23 09:40:49.361175: step: 2350/1541, loss: 0.1897342950105667 2023-01-23 09:40:50.271269: step: 2352/1541, loss: 0.31971845030784607 2023-01-23 09:40:51.212502: step: 2354/1541, loss: 0.312498539686203 2023-01-23 09:40:52.066675: step: 2356/1541, loss: 0.2289956957101822 2023-01-23 09:40:52.968774: step: 2358/1541, loss: 0.13524463772773743 2023-01-23 09:40:53.862290: step: 2360/1541, loss: 0.34777945280075073 2023-01-23 09:40:54.733705: step: 2362/1541, loss: 0.26814478635787964 2023-01-23 09:40:55.586858: step: 2364/1541, loss: 0.06914201378822327 2023-01-23 09:40:56.488854: step: 2366/1541, loss: 0.4413376748561859 2023-01-23 09:40:57.408656: step: 2368/1541, loss: 0.3449631333351135 2023-01-23 09:40:58.279116: step: 2370/1541, loss: 0.09218991547822952 2023-01-23 09:40:59.204367: step: 2372/1541, loss: 0.11130709946155548 2023-01-23 09:41:00.063583: step: 2374/1541, loss: 0.11810264736413956 2023-01-23 09:41:01.010177: step: 2376/1541, loss: 1.1107221841812134 2023-01-23 09:41:01.963518: step: 2378/1541, loss: 0.12084874510765076 2023-01-23 09:41:02.834195: step: 2380/1541, loss: 0.3459426760673523 2023-01-23 09:41:03.719001: step: 2382/1541, loss: 0.23884600400924683 2023-01-23 09:41:04.586967: step: 2384/1541, loss: 0.21213211119174957 2023-01-23 09:41:05.435628: step: 2386/1541, loss: 0.09516769647598267 2023-01-23 09:41:06.323888: step: 2388/1541, loss: 0.0919719859957695 2023-01-23 09:41:07.178939: step: 2390/1541, loss: 0.1256527453660965 2023-01-23 09:41:08.109187: step: 2392/1541, loss: 0.3187708854675293 2023-01-23 09:41:09.018808: step: 2394/1541, loss: 1.0023118257522583 2023-01-23 09:41:09.907340: step: 2396/1541, loss: 0.24147966504096985 2023-01-23 09:41:10.798825: step: 2398/1541, loss: 0.10039130598306656 2023-01-23 09:41:11.669121: step: 2400/1541, loss: 0.1853712946176529 2023-01-23 09:41:12.523786: step: 2402/1541, loss: 0.17448821663856506 2023-01-23 09:41:13.414041: step: 2404/1541, loss: 0.2022107094526291 2023-01-23 09:41:14.261121: step: 2406/1541, loss: 0.42861711978912354 2023-01-23 09:41:15.134005: step: 2408/1541, loss: 0.1931176483631134 2023-01-23 09:41:16.003138: step: 2410/1541, loss: 0.28870952129364014 2023-01-23 09:41:16.863820: step: 2412/1541, loss: 0.4777725636959076 2023-01-23 09:41:17.743290: step: 2414/1541, loss: 0.5971308350563049 2023-01-23 09:41:18.654568: step: 2416/1541, loss: 0.4203327000141144 2023-01-23 09:41:19.567470: step: 2418/1541, loss: 0.15599846839904785 2023-01-23 09:41:20.486225: step: 2420/1541, loss: 0.3241722583770752 2023-01-23 09:41:21.372715: step: 2422/1541, loss: 0.35241463780403137 2023-01-23 09:41:22.241759: step: 2424/1541, loss: 0.15040554106235504 2023-01-23 09:41:23.133580: step: 2426/1541, loss: 0.1763322651386261 2023-01-23 09:41:23.993871: step: 2428/1541, loss: 0.18791010975837708 2023-01-23 09:41:24.907773: step: 2430/1541, loss: 0.026877785101532936 2023-01-23 09:41:25.810565: step: 2432/1541, loss: 0.07912344485521317 2023-01-23 09:41:26.733166: step: 2434/1541, loss: 0.4642575979232788 2023-01-23 09:41:27.617846: step: 2436/1541, loss: 0.36090895533561707 2023-01-23 09:41:28.509523: step: 2438/1541, loss: 0.2705332934856415 2023-01-23 09:41:29.409291: step: 2440/1541, loss: 0.051019348204135895 2023-01-23 09:41:30.269429: step: 2442/1541, loss: 0.531570553779602 2023-01-23 09:41:31.144712: step: 2444/1541, loss: 0.13971152901649475 2023-01-23 09:41:32.036208: step: 2446/1541, loss: 0.39958930015563965 2023-01-23 09:41:32.910079: step: 2448/1541, loss: 0.19677424430847168 2023-01-23 09:41:33.873955: step: 2450/1541, loss: 0.16627031564712524 2023-01-23 09:41:34.717300: step: 2452/1541, loss: 0.18167591094970703 2023-01-23 09:41:35.613987: step: 2454/1541, loss: 0.4962911009788513 2023-01-23 09:41:36.504549: step: 2456/1541, loss: 0.1915210783481598 2023-01-23 09:41:37.389630: step: 2458/1541, loss: 0.20442254841327667 2023-01-23 09:41:38.302465: step: 2460/1541, loss: 0.17976592481136322 2023-01-23 09:41:39.220242: step: 2462/1541, loss: 0.24155743420124054 2023-01-23 09:41:40.135401: step: 2464/1541, loss: 0.189917653799057 2023-01-23 09:41:41.077948: step: 2466/1541, loss: 0.18061137199401855 2023-01-23 09:41:41.966403: step: 2468/1541, loss: 0.10474904626607895 2023-01-23 09:41:42.868479: step: 2470/1541, loss: 0.06187141686677933 2023-01-23 09:41:43.745097: step: 2472/1541, loss: 0.18154096603393555 2023-01-23 09:41:44.631489: step: 2474/1541, loss: 0.24799390137195587 2023-01-23 09:41:45.500173: step: 2476/1541, loss: 0.1255643665790558 2023-01-23 09:41:46.387281: step: 2478/1541, loss: 0.16116967797279358 2023-01-23 09:41:47.280947: step: 2480/1541, loss: 0.13322468101978302 2023-01-23 09:41:48.167611: step: 2482/1541, loss: 0.06874939799308777 2023-01-23 09:41:49.032505: step: 2484/1541, loss: 0.04593556746840477 2023-01-23 09:41:49.906704: step: 2486/1541, loss: 0.029551904648542404 2023-01-23 09:41:50.839739: step: 2488/1541, loss: 0.06868641823530197 2023-01-23 09:41:51.701712: step: 2490/1541, loss: 0.15416233241558075 2023-01-23 09:41:52.561775: step: 2492/1541, loss: 0.0816771388053894 2023-01-23 09:41:53.411774: step: 2494/1541, loss: 0.025556091219186783 2023-01-23 09:41:54.295571: step: 2496/1541, loss: 0.133463054895401 2023-01-23 09:41:55.176427: step: 2498/1541, loss: 0.5493516325950623 2023-01-23 09:41:56.066225: step: 2500/1541, loss: 0.04253830760717392 2023-01-23 09:41:56.980089: step: 2502/1541, loss: 0.3009457588195801 2023-01-23 09:41:57.823291: step: 2504/1541, loss: 0.10926332324743271 2023-01-23 09:41:58.744948: step: 2506/1541, loss: 0.2029089480638504 2023-01-23 09:41:59.630652: step: 2508/1541, loss: 0.029582429677248 2023-01-23 09:42:00.507863: step: 2510/1541, loss: 0.14467507600784302 2023-01-23 09:42:01.378688: step: 2512/1541, loss: 0.06261157244443893 2023-01-23 09:42:02.260826: step: 2514/1541, loss: 0.09188655018806458 2023-01-23 09:42:03.221963: step: 2516/1541, loss: 0.047314565628767014 2023-01-23 09:42:04.125043: step: 2518/1541, loss: 0.1058502346277237 2023-01-23 09:42:05.019392: step: 2520/1541, loss: 0.43161800503730774 2023-01-23 09:42:05.868401: step: 2522/1541, loss: 0.21898716688156128 2023-01-23 09:42:06.760518: step: 2524/1541, loss: 0.25740429759025574 2023-01-23 09:42:07.673890: step: 2526/1541, loss: 0.037449780851602554 2023-01-23 09:42:08.564651: step: 2528/1541, loss: 0.1726321130990982 2023-01-23 09:42:09.484382: step: 2530/1541, loss: 0.20636597275733948 2023-01-23 09:42:10.390073: step: 2532/1541, loss: 0.6758025288581848 2023-01-23 09:42:11.266597: step: 2534/1541, loss: 1.033253788948059 2023-01-23 09:42:12.140656: step: 2536/1541, loss: 0.39966854453086853 2023-01-23 09:42:12.978045: step: 2538/1541, loss: 0.06387469917535782 2023-01-23 09:42:13.851028: step: 2540/1541, loss: 0.2544199228286743 2023-01-23 09:42:14.729360: step: 2542/1541, loss: 0.34335947036743164 2023-01-23 09:42:15.693274: step: 2544/1541, loss: 0.1267574280500412 2023-01-23 09:42:16.613277: step: 2546/1541, loss: 0.7142511010169983 2023-01-23 09:42:17.454152: step: 2548/1541, loss: 0.07994963228702545 2023-01-23 09:42:18.310881: step: 2550/1541, loss: 0.10048442333936691 2023-01-23 09:42:19.158019: step: 2552/1541, loss: 0.0892600417137146 2023-01-23 09:42:20.054385: step: 2554/1541, loss: 0.16319209337234497 2023-01-23 09:42:20.953590: step: 2556/1541, loss: 0.26238641142845154 2023-01-23 09:42:21.828312: step: 2558/1541, loss: 0.4610028862953186 2023-01-23 09:42:22.698908: step: 2560/1541, loss: 0.08412348479032516 2023-01-23 09:42:23.625876: step: 2562/1541, loss: 0.32545000314712524 2023-01-23 09:42:24.484946: step: 2564/1541, loss: 0.2983594834804535 2023-01-23 09:42:25.337127: step: 2566/1541, loss: 0.3538817763328552 2023-01-23 09:42:26.216140: step: 2568/1541, loss: 0.05188456177711487 2023-01-23 09:42:27.146902: step: 2570/1541, loss: 0.5017791986465454 2023-01-23 09:42:27.985975: step: 2572/1541, loss: 0.294673353433609 2023-01-23 09:42:28.884283: step: 2574/1541, loss: 0.36293405294418335 2023-01-23 09:42:29.782279: step: 2576/1541, loss: 0.08309108763933182 2023-01-23 09:42:30.642122: step: 2578/1541, loss: 0.26856017112731934 2023-01-23 09:42:31.596854: step: 2580/1541, loss: 0.12798930704593658 2023-01-23 09:42:32.448830: step: 2582/1541, loss: 0.061520468443632126 2023-01-23 09:42:33.340320: step: 2584/1541, loss: 0.33071237802505493 2023-01-23 09:42:34.217823: step: 2586/1541, loss: 0.038943830877542496 2023-01-23 09:42:35.190763: step: 2588/1541, loss: 0.45212602615356445 2023-01-23 09:42:36.152447: step: 2590/1541, loss: 0.1729012429714203 2023-01-23 09:42:37.057974: step: 2592/1541, loss: 0.11702883243560791 2023-01-23 09:42:37.967782: step: 2594/1541, loss: 0.8776335716247559 2023-01-23 09:42:38.880759: step: 2596/1541, loss: 0.23221448063850403 2023-01-23 09:42:39.803966: step: 2598/1541, loss: 0.04755585268139839 2023-01-23 09:42:40.742490: step: 2600/1541, loss: 0.1633782684803009 2023-01-23 09:42:41.614607: step: 2602/1541, loss: 0.4828452467918396 2023-01-23 09:42:42.524432: step: 2604/1541, loss: 0.376944363117218 2023-01-23 09:42:43.380193: step: 2606/1541, loss: 0.29273977875709534 2023-01-23 09:42:44.231808: step: 2608/1541, loss: 0.296100914478302 2023-01-23 09:42:45.059085: step: 2610/1541, loss: 0.06057422608137131 2023-01-23 09:42:45.936633: step: 2612/1541, loss: 0.1085035651922226 2023-01-23 09:42:46.939434: step: 2614/1541, loss: 0.204148069024086 2023-01-23 09:42:47.841239: step: 2616/1541, loss: 0.29394835233688354 2023-01-23 09:42:48.738615: step: 2618/1541, loss: 0.04287354648113251 2023-01-23 09:42:49.602104: step: 2620/1541, loss: 0.10101959109306335 2023-01-23 09:42:50.496468: step: 2622/1541, loss: 0.7080071568489075 2023-01-23 09:42:51.358654: step: 2624/1541, loss: 0.22088713943958282 2023-01-23 09:42:52.262203: step: 2626/1541, loss: 0.05093829706311226 2023-01-23 09:42:53.190777: step: 2628/1541, loss: 0.017115123569965363 2023-01-23 09:42:54.083960: step: 2630/1541, loss: 0.1692960411310196 2023-01-23 09:42:54.970763: step: 2632/1541, loss: 0.5056935548782349 2023-01-23 09:42:55.862459: step: 2634/1541, loss: 0.48189377784729004 2023-01-23 09:42:56.723291: step: 2636/1541, loss: 0.036693308502435684 2023-01-23 09:42:57.603666: step: 2638/1541, loss: 0.6505148410797119 2023-01-23 09:42:58.525793: step: 2640/1541, loss: 0.5361058115959167 2023-01-23 09:42:59.427740: step: 2642/1541, loss: 0.16639362275600433 2023-01-23 09:43:00.314883: step: 2644/1541, loss: 0.22558024525642395 2023-01-23 09:43:01.248604: step: 2646/1541, loss: 0.06937853991985321 2023-01-23 09:43:02.240063: step: 2648/1541, loss: 0.7967790365219116 2023-01-23 09:43:03.121548: step: 2650/1541, loss: 0.1984306126832962 2023-01-23 09:43:04.005147: step: 2652/1541, loss: 0.16638138890266418 2023-01-23 09:43:04.851535: step: 2654/1541, loss: 0.2942207455635071 2023-01-23 09:43:05.726076: step: 2656/1541, loss: 0.2915002107620239 2023-01-23 09:43:06.695922: step: 2658/1541, loss: 0.313141405582428 2023-01-23 09:43:07.671347: step: 2660/1541, loss: 0.356119304895401 2023-01-23 09:43:08.547392: step: 2662/1541, loss: 0.29443737864494324 2023-01-23 09:43:09.493645: step: 2664/1541, loss: 0.15591202676296234 2023-01-23 09:43:10.368142: step: 2666/1541, loss: 0.25677967071533203 2023-01-23 09:43:11.244006: step: 2668/1541, loss: 0.5172672271728516 2023-01-23 09:43:12.117659: step: 2670/1541, loss: 0.23972982168197632 2023-01-23 09:43:13.014658: step: 2672/1541, loss: 0.48474612832069397 2023-01-23 09:43:13.887816: step: 2674/1541, loss: 0.1785552203655243 2023-01-23 09:43:14.826879: step: 2676/1541, loss: 0.11641836911439896 2023-01-23 09:43:15.682673: step: 2678/1541, loss: 0.11073422431945801 2023-01-23 09:43:16.567343: step: 2680/1541, loss: 0.13979923725128174 2023-01-23 09:43:17.405721: step: 2682/1541, loss: 0.9133527874946594 2023-01-23 09:43:18.296043: step: 2684/1541, loss: 0.1773366779088974 2023-01-23 09:43:19.145274: step: 2686/1541, loss: 0.4061146378517151 2023-01-23 09:43:20.091913: step: 2688/1541, loss: 0.25392717123031616 2023-01-23 09:43:20.993971: step: 2690/1541, loss: 0.040946610271930695 2023-01-23 09:43:21.847292: step: 2692/1541, loss: 0.027818119153380394 2023-01-23 09:43:22.725134: step: 2694/1541, loss: 1.53458571434021 2023-01-23 09:43:23.568825: step: 2696/1541, loss: 0.07430579513311386 2023-01-23 09:43:24.500155: step: 2698/1541, loss: 0.09436246752738953 2023-01-23 09:43:25.371182: step: 2700/1541, loss: 1.3564774990081787 2023-01-23 09:43:26.297981: step: 2702/1541, loss: 0.08041033148765564 2023-01-23 09:43:27.248336: step: 2704/1541, loss: 0.33138301968574524 2023-01-23 09:43:28.137457: step: 2706/1541, loss: 0.47633421421051025 2023-01-23 09:43:29.010357: step: 2708/1541, loss: 2.584773063659668 2023-01-23 09:43:29.906720: step: 2710/1541, loss: 0.04888160526752472 2023-01-23 09:43:30.797609: step: 2712/1541, loss: 0.21429742872714996 2023-01-23 09:43:31.718014: step: 2714/1541, loss: 0.0749923586845398 2023-01-23 09:43:32.678161: step: 2716/1541, loss: 0.18606948852539062 2023-01-23 09:43:33.543379: step: 2718/1541, loss: 0.4277445673942566 2023-01-23 09:43:34.433191: step: 2720/1541, loss: 0.1858893632888794 2023-01-23 09:43:35.298905: step: 2722/1541, loss: 0.0465201698243618 2023-01-23 09:43:36.161735: step: 2724/1541, loss: 0.11075737327337265 2023-01-23 09:43:37.040066: step: 2726/1541, loss: 0.09173691272735596 2023-01-23 09:43:37.892687: step: 2728/1541, loss: 0.09123048931360245 2023-01-23 09:43:38.825566: step: 2730/1541, loss: 0.1998068243265152 2023-01-23 09:43:39.685843: step: 2732/1541, loss: 0.025390034541487694 2023-01-23 09:43:40.610109: step: 2734/1541, loss: 0.30593544244766235 2023-01-23 09:43:41.507155: step: 2736/1541, loss: 0.13731884956359863 2023-01-23 09:43:42.347538: step: 2738/1541, loss: 0.16137446463108063 2023-01-23 09:43:43.252539: step: 2740/1541, loss: 0.11129585653543472 2023-01-23 09:43:44.118354: step: 2742/1541, loss: 0.24588221311569214 2023-01-23 09:43:44.977470: step: 2744/1541, loss: 0.358719140291214 2023-01-23 09:43:45.865876: step: 2746/1541, loss: 0.12311690300703049 2023-01-23 09:43:46.743136: step: 2748/1541, loss: 0.10310330241918564 2023-01-23 09:43:47.637115: step: 2750/1541, loss: 0.1716635823249817 2023-01-23 09:43:48.563797: step: 2752/1541, loss: 0.13054700195789337 2023-01-23 09:43:49.488169: step: 2754/1541, loss: 0.20027591288089752 2023-01-23 09:43:50.414585: step: 2756/1541, loss: 0.14163872599601746 2023-01-23 09:43:51.312645: step: 2758/1541, loss: 0.14679519832134247 2023-01-23 09:43:52.220713: step: 2760/1541, loss: 0.2299707978963852 2023-01-23 09:43:53.150477: step: 2762/1541, loss: 0.31492966413497925 2023-01-23 09:43:54.048818: step: 2764/1541, loss: 0.4679778516292572 2023-01-23 09:43:54.926381: step: 2766/1541, loss: 0.03260521590709686 2023-01-23 09:43:55.818206: step: 2768/1541, loss: 2.3702547550201416 2023-01-23 09:43:56.702769: step: 2770/1541, loss: 0.2557046115398407 2023-01-23 09:43:57.653897: step: 2772/1541, loss: 0.12688857316970825 2023-01-23 09:43:58.569421: step: 2774/1541, loss: 0.056803081184625626 2023-01-23 09:43:59.454461: step: 2776/1541, loss: 0.21403513848781586 2023-01-23 09:44:00.333967: step: 2778/1541, loss: 0.017574233934283257 2023-01-23 09:44:01.254487: step: 2780/1541, loss: 0.5320234894752502 2023-01-23 09:44:02.217316: step: 2782/1541, loss: 0.1252480149269104 2023-01-23 09:44:03.063951: step: 2784/1541, loss: 0.4895220100879669 2023-01-23 09:44:04.000641: step: 2786/1541, loss: 0.15177297592163086 2023-01-23 09:44:04.925766: step: 2788/1541, loss: 0.3406023681163788 2023-01-23 09:44:05.874479: step: 2790/1541, loss: 0.759107768535614 2023-01-23 09:44:06.753416: step: 2792/1541, loss: 1.319380760192871 2023-01-23 09:44:07.649653: step: 2794/1541, loss: 1.8059481382369995 2023-01-23 09:44:08.525305: step: 2796/1541, loss: 0.037036824971437454 2023-01-23 09:44:09.420936: step: 2798/1541, loss: 0.32359862327575684 2023-01-23 09:44:10.339112: step: 2800/1541, loss: 0.28967610001564026 2023-01-23 09:44:11.229092: step: 2802/1541, loss: 2.646009683609009 2023-01-23 09:44:12.125146: step: 2804/1541, loss: 0.24098949134349823 2023-01-23 09:44:13.016967: step: 2806/1541, loss: 0.16760976612567902 2023-01-23 09:44:13.930193: step: 2808/1541, loss: 0.05642140284180641 2023-01-23 09:44:14.822495: step: 2810/1541, loss: 0.6094708442687988 2023-01-23 09:44:15.713800: step: 2812/1541, loss: 0.10213445872068405 2023-01-23 09:44:16.628287: step: 2814/1541, loss: 0.1967180222272873 2023-01-23 09:44:17.490802: step: 2816/1541, loss: 0.31906935572624207 2023-01-23 09:44:18.421706: step: 2818/1541, loss: 0.10124746710062027 2023-01-23 09:44:19.332895: step: 2820/1541, loss: 0.26965588331222534 2023-01-23 09:44:20.194063: step: 2822/1541, loss: 0.08380629867315292 2023-01-23 09:44:21.041609: step: 2824/1541, loss: 0.05800025165081024 2023-01-23 09:44:21.959591: step: 2826/1541, loss: 0.24709193408489227 2023-01-23 09:44:22.852642: step: 2828/1541, loss: 0.19084547460079193 2023-01-23 09:44:23.761685: step: 2830/1541, loss: 2.1318793296813965 2023-01-23 09:44:24.662407: step: 2832/1541, loss: 0.31042057275772095 2023-01-23 09:44:25.630051: step: 2834/1541, loss: 0.40718695521354675 2023-01-23 09:44:26.532974: step: 2836/1541, loss: 0.7857382893562317 2023-01-23 09:44:27.401889: step: 2838/1541, loss: 0.30119848251342773 2023-01-23 09:44:28.355660: step: 2840/1541, loss: 0.07313646376132965 2023-01-23 09:44:29.239636: step: 2842/1541, loss: 0.13511043787002563 2023-01-23 09:44:30.252111: step: 2844/1541, loss: 3.0033693313598633 2023-01-23 09:44:31.125424: step: 2846/1541, loss: 0.030810408294200897 2023-01-23 09:44:32.159372: step: 2848/1541, loss: 0.1647966355085373 2023-01-23 09:44:33.014027: step: 2850/1541, loss: 0.09005265682935715 2023-01-23 09:44:33.933531: step: 2852/1541, loss: 0.4425811469554901 2023-01-23 09:44:34.781636: step: 2854/1541, loss: 0.15259455144405365 2023-01-23 09:44:35.655879: step: 2856/1541, loss: 0.3119044005870819 2023-01-23 09:44:36.509460: step: 2858/1541, loss: 0.20727509260177612 2023-01-23 09:44:37.344196: step: 2860/1541, loss: 0.40941497683525085 2023-01-23 09:44:38.189177: step: 2862/1541, loss: 0.27990710735321045 2023-01-23 09:44:39.120108: step: 2864/1541, loss: 0.08409430086612701 2023-01-23 09:44:40.082472: step: 2866/1541, loss: 0.039405740797519684 2023-01-23 09:44:40.978459: step: 2868/1541, loss: 0.05209992453455925 2023-01-23 09:44:41.888025: step: 2870/1541, loss: 0.27241402864456177 2023-01-23 09:44:42.772101: step: 2872/1541, loss: 0.23626603186130524 2023-01-23 09:44:43.680739: step: 2874/1541, loss: 0.3049314618110657 2023-01-23 09:44:44.580872: step: 2876/1541, loss: 0.27185872197151184 2023-01-23 09:44:45.513285: step: 2878/1541, loss: 3.0209436416625977 2023-01-23 09:44:46.357490: step: 2880/1541, loss: 0.2631308138370514 2023-01-23 09:44:47.210818: step: 2882/1541, loss: 0.3886012136936188 2023-01-23 09:44:48.091532: step: 2884/1541, loss: 0.641040563583374 2023-01-23 09:44:48.995304: step: 2886/1541, loss: 0.2843859791755676 2023-01-23 09:44:49.891572: step: 2888/1541, loss: 0.08811124414205551 2023-01-23 09:44:50.758702: step: 2890/1541, loss: 0.13425858318805695 2023-01-23 09:44:51.644459: step: 2892/1541, loss: 0.08158013224601746 2023-01-23 09:44:52.549825: step: 2894/1541, loss: 0.3818031847476959 2023-01-23 09:44:53.453006: step: 2896/1541, loss: 0.3071700930595398 2023-01-23 09:44:54.355427: step: 2898/1541, loss: 0.10187596082687378 2023-01-23 09:44:55.232975: step: 2900/1541, loss: 0.43592369556427 2023-01-23 09:44:56.122153: step: 2902/1541, loss: 0.2265395075082779 2023-01-23 09:44:57.041599: step: 2904/1541, loss: 0.5763894319534302 2023-01-23 09:44:57.917245: step: 2906/1541, loss: 1.0538876056671143 2023-01-23 09:44:58.839280: step: 2908/1541, loss: 0.131484255194664 2023-01-23 09:44:59.734047: step: 2910/1541, loss: 0.18038983643054962 2023-01-23 09:45:00.576883: step: 2912/1541, loss: 0.16282831132411957 2023-01-23 09:45:01.442975: step: 2914/1541, loss: 0.015143321827054024 2023-01-23 09:45:02.367890: step: 2916/1541, loss: 0.16819116473197937 2023-01-23 09:45:03.321384: step: 2918/1541, loss: 0.9164193272590637 2023-01-23 09:45:04.190549: step: 2920/1541, loss: 0.3125941753387451 2023-01-23 09:45:05.035051: step: 2922/1541, loss: 0.07605115324258804 2023-01-23 09:45:05.903580: step: 2924/1541, loss: 0.07888045907020569 2023-01-23 09:45:06.793711: step: 2926/1541, loss: 0.23779934644699097 2023-01-23 09:45:07.659696: step: 2928/1541, loss: 0.24970266222953796 2023-01-23 09:45:08.530572: step: 2930/1541, loss: 0.18074439465999603 2023-01-23 09:45:09.407979: step: 2932/1541, loss: 0.04743354022502899 2023-01-23 09:45:10.326667: step: 2934/1541, loss: 0.20878033339977264 2023-01-23 09:45:11.246794: step: 2936/1541, loss: 0.09438946098089218 2023-01-23 09:45:12.154780: step: 2938/1541, loss: 0.298721045255661 2023-01-23 09:45:13.017613: step: 2940/1541, loss: 0.07681972533464432 2023-01-23 09:45:13.910779: step: 2942/1541, loss: 0.3650657832622528 2023-01-23 09:45:14.763831: step: 2944/1541, loss: 0.02583388425409794 2023-01-23 09:45:15.689247: step: 2946/1541, loss: 0.05888133496046066 2023-01-23 09:45:16.557642: step: 2948/1541, loss: 0.08753672242164612 2023-01-23 09:45:17.496664: step: 2950/1541, loss: 0.17572790384292603 2023-01-23 09:45:18.375890: step: 2952/1541, loss: 0.5235558748245239 2023-01-23 09:45:19.245934: step: 2954/1541, loss: 0.18865923583507538 2023-01-23 09:45:20.125885: step: 2956/1541, loss: 0.14530496299266815 2023-01-23 09:45:21.014000: step: 2958/1541, loss: 0.23988527059555054 2023-01-23 09:45:21.919243: step: 2960/1541, loss: 0.1910317987203598 2023-01-23 09:45:22.783877: step: 2962/1541, loss: 0.04970977455377579 2023-01-23 09:45:23.685703: step: 2964/1541, loss: 0.26410168409347534 2023-01-23 09:45:24.566958: step: 2966/1541, loss: 0.3429359197616577 2023-01-23 09:45:25.491548: step: 2968/1541, loss: 0.1979752779006958 2023-01-23 09:45:26.439557: step: 2970/1541, loss: 4.890284061431885 2023-01-23 09:45:27.354690: step: 2972/1541, loss: 0.05754892900586128 2023-01-23 09:45:28.227472: step: 2974/1541, loss: 0.18061567842960358 2023-01-23 09:45:29.107892: step: 2976/1541, loss: 0.09177882969379425 2023-01-23 09:45:30.005286: step: 2978/1541, loss: 0.10755370557308197 2023-01-23 09:45:30.840479: step: 2980/1541, loss: 0.11671389639377594 2023-01-23 09:45:31.696143: step: 2982/1541, loss: 0.1119859367609024 2023-01-23 09:45:32.622798: step: 2984/1541, loss: 0.41539809107780457 2023-01-23 09:45:33.541767: step: 2986/1541, loss: 0.3149631917476654 2023-01-23 09:45:34.451416: step: 2988/1541, loss: 0.07496559619903564 2023-01-23 09:45:35.382709: step: 2990/1541, loss: 0.1900557577610016 2023-01-23 09:45:36.212874: step: 2992/1541, loss: 0.20685029029846191 2023-01-23 09:45:37.141337: step: 2994/1541, loss: 0.31628432869911194 2023-01-23 09:45:38.061442: step: 2996/1541, loss: 0.5018075108528137 2023-01-23 09:45:38.966509: step: 2998/1541, loss: 1.491349220275879 2023-01-23 09:45:39.815115: step: 3000/1541, loss: 0.5734071135520935 2023-01-23 09:45:40.709753: step: 3002/1541, loss: 0.13605251908302307 2023-01-23 09:45:41.629805: step: 3004/1541, loss: 0.22651809453964233 2023-01-23 09:45:42.483396: step: 3006/1541, loss: 0.7032191753387451 2023-01-23 09:45:43.371593: step: 3008/1541, loss: 0.28209805488586426 2023-01-23 09:45:44.244307: step: 3010/1541, loss: 0.1559039056301117 2023-01-23 09:45:45.124176: step: 3012/1541, loss: 0.13902291655540466 2023-01-23 09:45:46.036951: step: 3014/1541, loss: 0.6471412777900696 2023-01-23 09:45:46.960202: step: 3016/1541, loss: 0.10610763728618622 2023-01-23 09:45:47.817952: step: 3018/1541, loss: 0.055043499916791916 2023-01-23 09:45:48.741984: step: 3020/1541, loss: 0.04432721063494682 2023-01-23 09:45:49.636789: step: 3022/1541, loss: 0.13258135318756104 2023-01-23 09:45:50.523039: step: 3024/1541, loss: 0.38363564014434814 2023-01-23 09:45:51.387116: step: 3026/1541, loss: 0.14155898988246918 2023-01-23 09:45:52.273040: step: 3028/1541, loss: 0.6620336174964905 2023-01-23 09:45:53.175074: step: 3030/1541, loss: 0.057734645903110504 2023-01-23 09:45:54.053497: step: 3032/1541, loss: 0.5097474455833435 2023-01-23 09:45:54.943183: step: 3034/1541, loss: 0.11774524301290512 2023-01-23 09:45:55.827902: step: 3036/1541, loss: 0.1950455754995346 2023-01-23 09:45:56.730172: step: 3038/1541, loss: 0.27648448944091797 2023-01-23 09:45:57.626468: step: 3040/1541, loss: 0.3608171343803406 2023-01-23 09:45:58.540448: step: 3042/1541, loss: 0.016782093793153763 2023-01-23 09:45:59.495776: step: 3044/1541, loss: 0.13767871260643005 2023-01-23 09:46:00.402398: step: 3046/1541, loss: 0.26329806447029114 2023-01-23 09:46:01.290536: step: 3048/1541, loss: 0.3148276209831238 2023-01-23 09:46:02.214386: step: 3050/1541, loss: 0.16909579932689667 2023-01-23 09:46:03.108503: step: 3052/1541, loss: 0.7976105809211731 2023-01-23 09:46:04.047094: step: 3054/1541, loss: 0.7917187809944153 2023-01-23 09:46:04.935934: step: 3056/1541, loss: 0.3476805090904236 2023-01-23 09:46:05.798938: step: 3058/1541, loss: 2.010490894317627 2023-01-23 09:46:06.696378: step: 3060/1541, loss: 0.30359774827957153 2023-01-23 09:46:07.595845: step: 3062/1541, loss: 0.508926510810852 2023-01-23 09:46:08.527227: step: 3064/1541, loss: 0.5762585997581482 2023-01-23 09:46:09.396335: step: 3066/1541, loss: 0.6791622042655945 2023-01-23 09:46:10.281240: step: 3068/1541, loss: 0.5120150446891785 2023-01-23 09:46:11.143596: step: 3070/1541, loss: 0.08679717779159546 2023-01-23 09:46:12.051001: step: 3072/1541, loss: 0.3322812020778656 2023-01-23 09:46:13.017270: step: 3074/1541, loss: 0.1748136281967163 2023-01-23 09:46:13.906369: step: 3076/1541, loss: 0.3307199478149414 2023-01-23 09:46:14.840041: step: 3078/1541, loss: 0.23756501078605652 2023-01-23 09:46:15.808206: step: 3080/1541, loss: 0.3518257141113281 2023-01-23 09:46:16.735110: step: 3082/1541, loss: 0.26266393065452576 2023-01-23 09:46:17.345575: step: 3084/1541, loss: 2.0369186401367188 ================================================== Loss: 0.324 -------------------- Dev Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6662925586063192, 'r': 0.6559544273633973, 'f1': 0.6610830780855506}, 'combined': 0.44044451927866024, 'epoch': 9} Test Chinese: {'event': {'p': 0.6093585699263933, 'r': 0.6944278010784901, 'f1': 0.6491178941472977}, 'argument': {'p': 0.7064930019884798, 'r': 0.7057205971029239, 'f1': 0.7061065883137368}, 'combined': 0.45834642164974576, 'epoch': 9} Dev Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6601370733441357, 'r': 0.6626206590226313, 'f1': 0.6613765346159047}, 'combined': 0.45778037254919046, 'epoch': 9} Test Russian: {'event': {'p': 0.6499215070643642, 'r': 0.74415817855003, 'f1': 0.693854748603352}, 'argument': {'p': 0.7045800388777901, 'r': 0.7126728462159413, 'f1': 0.70860333674433}, 'combined': 0.4916677900762334, 'epoch': 9} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6380632411067193, 'r': 0.6493563958165728, 'f1': 0.6436602870813396}, 'combined': 0.47311781785465984, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5360780268488179, 'r': 0.635813473704412, 'f1': 0.5817016887082918}, 'combined': 0.29916086847855006, 'epoch': 9} New best korean model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test for Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} -------------------- Dev for Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6749027227273954, 'r': 0.6759176140397674, 'f1': 0.6754097871321493}, 'combined': 0.467493670842566, 'epoch': 6} Test for Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7154362539653685, 'r': 0.7041128024637584, 'f1': 0.7097293657901553}, 'combined': 0.49443157493872836, 'epoch': 6} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5630713489409142, 'r': 0.6023553965414431, 'f1': 0.5820512820512821}, 'combined': 0.2993406593406594, 'epoch': 6} ****************************** Epoch: 10 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 09:51:40.394128: step: 2/1541, loss: 0.2469751387834549 2023-01-23 09:51:41.318525: step: 4/1541, loss: 0.17699915170669556 2023-01-23 09:51:42.200856: step: 6/1541, loss: 0.0412934273481369 2023-01-23 09:51:43.124384: step: 8/1541, loss: 0.030747033655643463 2023-01-23 09:51:43.979370: step: 10/1541, loss: 0.0745549350976944 2023-01-23 09:51:44.843559: step: 12/1541, loss: 0.25741124153137207 2023-01-23 09:51:45.764586: step: 14/1541, loss: 0.0906505137681961 2023-01-23 09:51:46.646321: step: 16/1541, loss: 0.318698525428772 2023-01-23 09:51:47.523209: step: 18/1541, loss: 0.13808131217956543 2023-01-23 09:51:48.396661: step: 20/1541, loss: 0.3426491618156433 2023-01-23 09:51:49.246027: step: 22/1541, loss: 0.180007666349411 2023-01-23 09:51:50.121378: step: 24/1541, loss: 0.08259780704975128 2023-01-23 09:51:51.003090: step: 26/1541, loss: 0.21171510219573975 2023-01-23 09:51:51.891669: step: 28/1541, loss: 0.05125214904546738 2023-01-23 09:51:52.752937: step: 30/1541, loss: 0.052927833050489426 2023-01-23 09:51:53.591432: step: 32/1541, loss: 0.37652498483657837 2023-01-23 09:51:54.422984: step: 34/1541, loss: 0.19750723242759705 2023-01-23 09:51:55.321659: step: 36/1541, loss: 0.29222095012664795 2023-01-23 09:51:56.282601: step: 38/1541, loss: 0.11314287036657333 2023-01-23 09:51:57.150595: step: 40/1541, loss: 0.027607394382357597 2023-01-23 09:51:58.043353: step: 42/1541, loss: 0.28115716576576233 2023-01-23 09:51:58.896546: step: 44/1541, loss: 0.13708281517028809 2023-01-23 09:51:59.812931: step: 46/1541, loss: 0.14677506685256958 2023-01-23 09:52:00.710120: step: 48/1541, loss: 0.16292449831962585 2023-01-23 09:52:01.597766: step: 50/1541, loss: 0.1800725758075714 2023-01-23 09:52:02.485377: step: 52/1541, loss: 0.10712499916553497 2023-01-23 09:52:03.361489: step: 54/1541, loss: 0.1510281264781952 2023-01-23 09:52:04.207404: step: 56/1541, loss: 0.4348873496055603 2023-01-23 09:52:05.059812: step: 58/1541, loss: 0.09651239961385727 2023-01-23 09:52:05.952394: step: 60/1541, loss: 0.023153597488999367 2023-01-23 09:52:06.791986: step: 62/1541, loss: 0.03233088180422783 2023-01-23 09:52:07.684882: step: 64/1541, loss: 0.01262095756828785 2023-01-23 09:52:08.583739: step: 66/1541, loss: 0.03792193904519081 2023-01-23 09:52:09.494337: step: 68/1541, loss: 0.23889988660812378 2023-01-23 09:52:10.419452: step: 70/1541, loss: 0.07473056763410568 2023-01-23 09:52:11.317573: step: 72/1541, loss: 0.6322988271713257 2023-01-23 09:52:12.330663: step: 74/1541, loss: 0.4942486584186554 2023-01-23 09:52:13.212118: step: 76/1541, loss: 0.09868096560239792 2023-01-23 09:52:14.115572: step: 78/1541, loss: 0.06156446784734726 2023-01-23 09:52:14.996457: step: 80/1541, loss: 0.19538703560829163 2023-01-23 09:52:15.876062: step: 82/1541, loss: 0.1876184493303299 2023-01-23 09:52:16.775967: step: 84/1541, loss: 0.1495506763458252 2023-01-23 09:52:17.647993: step: 86/1541, loss: 0.11098706722259521 2023-01-23 09:52:18.634143: step: 88/1541, loss: 0.1595221906900406 2023-01-23 09:52:19.504492: step: 90/1541, loss: 0.3239764869213104 2023-01-23 09:52:20.364469: step: 92/1541, loss: 0.11338067054748535 2023-01-23 09:52:21.239886: step: 94/1541, loss: 0.0935710221529007 2023-01-23 09:52:22.095130: step: 96/1541, loss: 0.4276179373264313 2023-01-23 09:52:23.009282: step: 98/1541, loss: 0.04033844918012619 2023-01-23 09:52:23.860562: step: 100/1541, loss: 0.047648221254348755 2023-01-23 09:52:24.771186: step: 102/1541, loss: 0.4687086045742035 2023-01-23 09:52:25.658316: step: 104/1541, loss: 0.22567927837371826 2023-01-23 09:52:26.567938: step: 106/1541, loss: 0.10840418189764023 2023-01-23 09:52:27.465129: step: 108/1541, loss: 0.21243803203105927 2023-01-23 09:52:28.391463: step: 110/1541, loss: 0.13490153849124908 2023-01-23 09:52:29.300388: step: 112/1541, loss: 0.029271909967064857 2023-01-23 09:52:30.276796: step: 114/1541, loss: 0.12162879854440689 2023-01-23 09:52:31.163365: step: 116/1541, loss: 0.10663305222988129 2023-01-23 09:52:32.072975: step: 118/1541, loss: 0.504298746585846 2023-01-23 09:52:32.937133: step: 120/1541, loss: 0.20144213736057281 2023-01-23 09:52:33.841176: step: 122/1541, loss: 0.17985336482524872 2023-01-23 09:52:34.708702: step: 124/1541, loss: 0.08530545234680176 2023-01-23 09:52:35.595651: step: 126/1541, loss: 0.14294292032718658 2023-01-23 09:52:36.479088: step: 128/1541, loss: 0.2150096446275711 2023-01-23 09:52:37.348497: step: 130/1541, loss: 0.21143747866153717 2023-01-23 09:52:38.314286: step: 132/1541, loss: 0.3835648000240326 2023-01-23 09:52:39.276307: step: 134/1541, loss: 0.04485420137643814 2023-01-23 09:52:40.115953: step: 136/1541, loss: 0.5679571032524109 2023-01-23 09:52:41.022454: step: 138/1541, loss: 0.3954576849937439 2023-01-23 09:52:41.914495: step: 140/1541, loss: 0.7169243097305298 2023-01-23 09:52:42.784175: step: 142/1541, loss: 0.18198545277118683 2023-01-23 09:52:43.709477: step: 144/1541, loss: 0.1008416935801506 2023-01-23 09:52:44.577372: step: 146/1541, loss: 0.15839608013629913 2023-01-23 09:52:45.480614: step: 148/1541, loss: 0.03096730262041092 2023-01-23 09:52:46.407237: step: 150/1541, loss: 0.010787645354866982 2023-01-23 09:52:47.290038: step: 152/1541, loss: 0.17714329063892365 2023-01-23 09:52:48.218095: step: 154/1541, loss: 0.0882793590426445 2023-01-23 09:52:49.109896: step: 156/1541, loss: 0.15939557552337646 2023-01-23 09:52:49.975072: step: 158/1541, loss: 0.3693152964115143 2023-01-23 09:52:50.839126: step: 160/1541, loss: 0.07474488019943237 2023-01-23 09:52:51.729505: step: 162/1541, loss: 0.41089197993278503 2023-01-23 09:52:52.679359: step: 164/1541, loss: 0.1220540851354599 2023-01-23 09:52:53.535437: step: 166/1541, loss: 0.31101590394973755 2023-01-23 09:52:54.440092: step: 168/1541, loss: 0.0299668088555336 2023-01-23 09:52:55.306394: step: 170/1541, loss: 0.1425209939479828 2023-01-23 09:52:56.227075: step: 172/1541, loss: 1.3902770280838013 2023-01-23 09:52:57.073886: step: 174/1541, loss: 0.09348975867033005 2023-01-23 09:52:57.981105: step: 176/1541, loss: 0.32123398780822754 2023-01-23 09:52:58.884505: step: 178/1541, loss: 0.8009049892425537 2023-01-23 09:52:59.922139: step: 180/1541, loss: 0.21570613980293274 2023-01-23 09:53:00.809940: step: 182/1541, loss: 0.3176441788673401 2023-01-23 09:53:01.668994: step: 184/1541, loss: 0.25323525071144104 2023-01-23 09:53:02.606609: step: 186/1541, loss: 0.04632653668522835 2023-01-23 09:53:03.512417: step: 188/1541, loss: 0.10892833769321442 2023-01-23 09:53:04.367309: step: 190/1541, loss: 0.49976953864097595 2023-01-23 09:53:05.296178: step: 192/1541, loss: 0.13527293503284454 2023-01-23 09:53:06.187585: step: 194/1541, loss: 0.15490451455116272 2023-01-23 09:53:07.186801: step: 196/1541, loss: 0.32439854741096497 2023-01-23 09:53:08.079396: step: 198/1541, loss: 0.08472011238336563 2023-01-23 09:53:08.937648: step: 200/1541, loss: 0.3055139183998108 2023-01-23 09:53:09.809372: step: 202/1541, loss: 0.12988688051700592 2023-01-23 09:53:10.722903: step: 204/1541, loss: 0.11938577890396118 2023-01-23 09:53:11.650928: step: 206/1541, loss: 0.38551953434944153 2023-01-23 09:53:12.516296: step: 208/1541, loss: 0.03407882899045944 2023-01-23 09:53:13.376869: step: 210/1541, loss: 0.05799426510930061 2023-01-23 09:53:14.247807: step: 212/1541, loss: 0.17015038430690765 2023-01-23 09:53:15.256747: step: 214/1541, loss: 0.12146751582622528 2023-01-23 09:53:16.147339: step: 216/1541, loss: 0.1583675891160965 2023-01-23 09:53:17.017224: step: 218/1541, loss: 0.04296141862869263 2023-01-23 09:53:17.882374: step: 220/1541, loss: 0.04060254991054535 2023-01-23 09:53:18.748906: step: 222/1541, loss: 0.2932208180427551 2023-01-23 09:53:19.687748: step: 224/1541, loss: 0.039695609360933304 2023-01-23 09:53:20.573818: step: 226/1541, loss: 1.0040751695632935 2023-01-23 09:53:21.448034: step: 228/1541, loss: 0.15599559247493744 2023-01-23 09:53:22.356506: step: 230/1541, loss: 0.3445487320423126 2023-01-23 09:53:23.238022: step: 232/1541, loss: 0.2587352693080902 2023-01-23 09:53:24.111390: step: 234/1541, loss: 0.029458913952112198 2023-01-23 09:53:25.010397: step: 236/1541, loss: 0.1427193582057953 2023-01-23 09:53:25.955234: step: 238/1541, loss: 0.39952173829078674 2023-01-23 09:53:26.862150: step: 240/1541, loss: 0.09346608817577362 2023-01-23 09:53:27.790442: step: 242/1541, loss: 0.48405975103378296 2023-01-23 09:53:28.663322: step: 244/1541, loss: 0.1159655973315239 2023-01-23 09:53:29.635154: step: 246/1541, loss: 0.2512359023094177 2023-01-23 09:53:30.541839: step: 248/1541, loss: 0.11129854619503021 2023-01-23 09:53:31.428086: step: 250/1541, loss: 0.8363077044487 2023-01-23 09:53:32.296601: step: 252/1541, loss: 0.05346103012561798 2023-01-23 09:53:33.216618: step: 254/1541, loss: 0.1707412153482437 2023-01-23 09:53:34.080473: step: 256/1541, loss: 0.1986730396747589 2023-01-23 09:53:34.986153: step: 258/1541, loss: 0.6934967041015625 2023-01-23 09:53:35.851099: step: 260/1541, loss: 0.07798660546541214 2023-01-23 09:53:36.751346: step: 262/1541, loss: 0.15809552371501923 2023-01-23 09:53:37.654424: step: 264/1541, loss: 0.1195521280169487 2023-01-23 09:53:38.546671: step: 266/1541, loss: 0.19926005601882935 2023-01-23 09:53:39.564665: step: 268/1541, loss: 1.620071291923523 2023-01-23 09:53:40.542488: step: 270/1541, loss: 0.05959387868642807 2023-01-23 09:53:41.427245: step: 272/1541, loss: 2.1034271717071533 2023-01-23 09:53:42.322668: step: 274/1541, loss: 0.11068731546401978 2023-01-23 09:53:43.291201: step: 276/1541, loss: 0.35342729091644287 2023-01-23 09:53:44.166745: step: 278/1541, loss: 0.11720692366361618 2023-01-23 09:53:45.052509: step: 280/1541, loss: 0.28372591733932495 2023-01-23 09:53:45.928240: step: 282/1541, loss: 0.0535251721739769 2023-01-23 09:53:46.876432: step: 284/1541, loss: 0.329746812582016 2023-01-23 09:53:47.805028: step: 286/1541, loss: 0.16671955585479736 2023-01-23 09:53:48.722977: step: 288/1541, loss: 0.48128533363342285 2023-01-23 09:53:49.585777: step: 290/1541, loss: 0.4709831476211548 2023-01-23 09:53:50.511826: step: 292/1541, loss: 0.6373992562294006 2023-01-23 09:53:51.368588: step: 294/1541, loss: 0.042668357491493225 2023-01-23 09:53:52.335817: step: 296/1541, loss: 0.2050645649433136 2023-01-23 09:53:53.192219: step: 298/1541, loss: 0.18258965015411377 2023-01-23 09:53:54.110311: step: 300/1541, loss: 0.185882106423378 2023-01-23 09:53:55.030471: step: 302/1541, loss: 0.02396121434867382 2023-01-23 09:53:55.958965: step: 304/1541, loss: 0.07365795969963074 2023-01-23 09:53:56.906239: step: 306/1541, loss: 0.1647574007511139 2023-01-23 09:53:57.778489: step: 308/1541, loss: 0.14800061285495758 2023-01-23 09:53:58.670531: step: 310/1541, loss: 1.3552511930465698 2023-01-23 09:53:59.521027: step: 312/1541, loss: 0.13479970395565033 2023-01-23 09:54:00.426380: step: 314/1541, loss: 0.8073357343673706 2023-01-23 09:54:01.342253: step: 316/1541, loss: 0.1363389790058136 2023-01-23 09:54:02.286139: step: 318/1541, loss: 0.08208127319812775 2023-01-23 09:54:03.176221: step: 320/1541, loss: 0.14605385065078735 2023-01-23 09:54:04.163050: step: 322/1541, loss: 0.7368641495704651 2023-01-23 09:54:05.085210: step: 324/1541, loss: 0.1686534583568573 2023-01-23 09:54:05.935230: step: 326/1541, loss: 0.016760515049099922 2023-01-23 09:54:06.852449: step: 328/1541, loss: 0.4484851062297821 2023-01-23 09:54:07.722028: step: 330/1541, loss: 0.15666836500167847 2023-01-23 09:54:08.640935: step: 332/1541, loss: 0.32714059948921204 2023-01-23 09:54:09.524791: step: 334/1541, loss: 0.19846795499324799 2023-01-23 09:54:10.413987: step: 336/1541, loss: 0.03219594806432724 2023-01-23 09:54:11.283990: step: 338/1541, loss: 0.2275375872850418 2023-01-23 09:54:12.158885: step: 340/1541, loss: 0.07792522013187408 2023-01-23 09:54:13.086805: step: 342/1541, loss: 0.08658836781978607 2023-01-23 09:54:13.959873: step: 344/1541, loss: 0.4786078929901123 2023-01-23 09:54:14.819669: step: 346/1541, loss: 0.302968829870224 2023-01-23 09:54:15.717045: step: 348/1541, loss: 0.1039929986000061 2023-01-23 09:54:16.589722: step: 350/1541, loss: 0.14719825983047485 2023-01-23 09:54:17.507766: step: 352/1541, loss: 0.17428725957870483 2023-01-23 09:54:18.433770: step: 354/1541, loss: 0.30927056074142456 2023-01-23 09:54:19.298793: step: 356/1541, loss: 0.13653665781021118 2023-01-23 09:54:20.169524: step: 358/1541, loss: 0.15864905714988708 2023-01-23 09:54:21.078143: step: 360/1541, loss: 0.2822880148887634 2023-01-23 09:54:22.029474: step: 362/1541, loss: 0.1817948967218399 2023-01-23 09:54:22.947205: step: 364/1541, loss: 0.27611982822418213 2023-01-23 09:54:23.879473: step: 366/1541, loss: 0.26757755875587463 2023-01-23 09:54:24.744506: step: 368/1541, loss: 0.0310675036162138 2023-01-23 09:54:25.632564: step: 370/1541, loss: 0.10665813088417053 2023-01-23 09:54:26.526909: step: 372/1541, loss: 0.010620148852467537 2023-01-23 09:54:27.377262: step: 374/1541, loss: 0.09557775408029556 2023-01-23 09:54:28.259806: step: 376/1541, loss: 0.11175116151571274 2023-01-23 09:54:29.102418: step: 378/1541, loss: 0.3025254011154175 2023-01-23 09:54:29.951192: step: 380/1541, loss: 0.15380263328552246 2023-01-23 09:54:30.847945: step: 382/1541, loss: 0.09752964973449707 2023-01-23 09:54:31.745102: step: 384/1541, loss: 0.15831151604652405 2023-01-23 09:54:32.644314: step: 386/1541, loss: 0.7266661524772644 2023-01-23 09:54:33.516758: step: 388/1541, loss: 0.048232655972242355 2023-01-23 09:54:34.415419: step: 390/1541, loss: 0.13462884724140167 2023-01-23 09:54:35.365113: step: 392/1541, loss: 0.1282159984111786 2023-01-23 09:54:36.243301: step: 394/1541, loss: 0.5800015926361084 2023-01-23 09:54:37.112655: step: 396/1541, loss: 0.012145227752625942 2023-01-23 09:54:37.992316: step: 398/1541, loss: 0.12742657959461212 2023-01-23 09:54:38.870992: step: 400/1541, loss: 0.05869505554437637 2023-01-23 09:54:39.753471: step: 402/1541, loss: 0.10103221982717514 2023-01-23 09:54:40.751739: step: 404/1541, loss: 0.2744011878967285 2023-01-23 09:54:41.634955: step: 406/1541, loss: 0.2656240165233612 2023-01-23 09:54:42.548372: step: 408/1541, loss: 0.04957761615514755 2023-01-23 09:54:43.453505: step: 410/1541, loss: 0.1996103823184967 2023-01-23 09:54:44.301658: step: 412/1541, loss: 0.10836829245090485 2023-01-23 09:54:45.148537: step: 414/1541, loss: 0.2419140636920929 2023-01-23 09:54:46.054552: step: 416/1541, loss: 0.08858916908502579 2023-01-23 09:54:46.956562: step: 418/1541, loss: 0.1056586280465126 2023-01-23 09:54:47.824208: step: 420/1541, loss: 0.1506417840719223 2023-01-23 09:54:48.698666: step: 422/1541, loss: 0.09080414474010468 2023-01-23 09:54:49.567536: step: 424/1541, loss: 0.451128214597702 2023-01-23 09:54:50.451164: step: 426/1541, loss: 0.07170562446117401 2023-01-23 09:54:51.367526: step: 428/1541, loss: 0.033461298793554306 2023-01-23 09:54:52.256773: step: 430/1541, loss: 0.2484254091978073 2023-01-23 09:54:53.198994: step: 432/1541, loss: 0.6132885813713074 2023-01-23 09:54:54.066438: step: 434/1541, loss: 0.052568867802619934 2023-01-23 09:54:55.037774: step: 436/1541, loss: 0.03006013296544552 2023-01-23 09:54:55.933621: step: 438/1541, loss: 0.14667364954948425 2023-01-23 09:54:56.912278: step: 440/1541, loss: 0.10967006534337997 2023-01-23 09:54:57.862069: step: 442/1541, loss: 1.168108582496643 2023-01-23 09:54:58.747913: step: 444/1541, loss: 0.1809820681810379 2023-01-23 09:54:59.626471: step: 446/1541, loss: 0.024660207331180573 2023-01-23 09:55:00.526956: step: 448/1541, loss: 0.15289783477783203 2023-01-23 09:55:01.400185: step: 450/1541, loss: 0.469833105802536 2023-01-23 09:55:02.316705: step: 452/1541, loss: 0.3241220712661743 2023-01-23 09:55:03.204483: step: 454/1541, loss: 0.11534354090690613 2023-01-23 09:55:04.076778: step: 456/1541, loss: 0.13364939391613007 2023-01-23 09:55:04.945289: step: 458/1541, loss: 0.13736429810523987 2023-01-23 09:55:05.841207: step: 460/1541, loss: 0.11524131894111633 2023-01-23 09:55:06.760364: step: 462/1541, loss: 0.31453922390937805 2023-01-23 09:55:07.660064: step: 464/1541, loss: 0.09937889128923416 2023-01-23 09:55:08.547799: step: 466/1541, loss: 0.18805554509162903 2023-01-23 09:55:09.440119: step: 468/1541, loss: 0.13734833896160126 2023-01-23 09:55:10.302974: step: 470/1541, loss: 0.07215661555528641 2023-01-23 09:55:11.169783: step: 472/1541, loss: 0.07786191999912262 2023-01-23 09:55:12.053534: step: 474/1541, loss: 0.13587605953216553 2023-01-23 09:55:12.956870: step: 476/1541, loss: 0.4687683880329132 2023-01-23 09:55:13.817545: step: 478/1541, loss: 0.2472650557756424 2023-01-23 09:55:14.655533: step: 480/1541, loss: 0.0550665557384491 2023-01-23 09:55:15.545711: step: 482/1541, loss: 0.12449443340301514 2023-01-23 09:55:16.446762: step: 484/1541, loss: 0.1478307843208313 2023-01-23 09:55:17.308446: step: 486/1541, loss: 0.21574687957763672 2023-01-23 09:55:18.162380: step: 488/1541, loss: 0.19872666895389557 2023-01-23 09:55:19.021610: step: 490/1541, loss: 0.15334048867225647 2023-01-23 09:55:19.884647: step: 492/1541, loss: 0.17366349697113037 2023-01-23 09:55:20.788749: step: 494/1541, loss: 0.18415796756744385 2023-01-23 09:55:21.693378: step: 496/1541, loss: 0.8554756045341492 2023-01-23 09:55:22.602598: step: 498/1541, loss: 0.2456374168395996 2023-01-23 09:55:23.580918: step: 500/1541, loss: 0.09040428698062897 2023-01-23 09:55:24.487865: step: 502/1541, loss: 0.16741380095481873 2023-01-23 09:55:25.370882: step: 504/1541, loss: 0.1312686651945114 2023-01-23 09:55:26.277702: step: 506/1541, loss: 0.09926676750183105 2023-01-23 09:55:27.119366: step: 508/1541, loss: 0.09265372157096863 2023-01-23 09:55:28.007606: step: 510/1541, loss: 0.09500078111886978 2023-01-23 09:55:28.858479: step: 512/1541, loss: 0.05751212686300278 2023-01-23 09:55:29.750430: step: 514/1541, loss: 0.4070844054222107 2023-01-23 09:55:30.683503: step: 516/1541, loss: 0.15697415173053741 2023-01-23 09:55:31.559292: step: 518/1541, loss: 0.2039821445941925 2023-01-23 09:55:32.463723: step: 520/1541, loss: 0.06021205708384514 2023-01-23 09:55:33.363499: step: 522/1541, loss: 0.12904733419418335 2023-01-23 09:55:34.253203: step: 524/1541, loss: 0.039791006594896317 2023-01-23 09:55:35.165066: step: 526/1541, loss: 0.17162790894508362 2023-01-23 09:55:36.062813: step: 528/1541, loss: 0.20725922286510468 2023-01-23 09:55:36.999860: step: 530/1541, loss: 0.09031476080417633 2023-01-23 09:55:37.895555: step: 532/1541, loss: 0.06869525462388992 2023-01-23 09:55:38.776303: step: 534/1541, loss: 0.0634758248925209 2023-01-23 09:55:39.668531: step: 536/1541, loss: 0.0780629962682724 2023-01-23 09:55:40.572083: step: 538/1541, loss: 0.02047586627304554 2023-01-23 09:55:41.499626: step: 540/1541, loss: 0.38004621863365173 2023-01-23 09:55:42.378979: step: 542/1541, loss: 0.10914571583271027 2023-01-23 09:55:43.297835: step: 544/1541, loss: 0.09812096506357193 2023-01-23 09:55:44.182884: step: 546/1541, loss: 0.5885182619094849 2023-01-23 09:55:45.061126: step: 548/1541, loss: 0.1056840568780899 2023-01-23 09:55:45.908419: step: 550/1541, loss: 0.1283605545759201 2023-01-23 09:55:46.802477: step: 552/1541, loss: 0.5889483690261841 2023-01-23 09:55:47.670784: step: 554/1541, loss: 0.31764256954193115 2023-01-23 09:55:48.537788: step: 556/1541, loss: 0.09214136749505997 2023-01-23 09:55:49.412466: step: 558/1541, loss: 1.1455363035202026 2023-01-23 09:55:50.283625: step: 560/1541, loss: 0.08777838200330734 2023-01-23 09:55:51.164654: step: 562/1541, loss: 0.1613617092370987 2023-01-23 09:55:52.037987: step: 564/1541, loss: 0.3582864999771118 2023-01-23 09:55:52.943575: step: 566/1541, loss: 0.12068373709917068 2023-01-23 09:55:53.898630: step: 568/1541, loss: 0.10384388267993927 2023-01-23 09:55:54.774161: step: 570/1541, loss: 0.4560776352882385 2023-01-23 09:55:55.613907: step: 572/1541, loss: 0.026669561862945557 2023-01-23 09:55:56.480487: step: 574/1541, loss: 0.025415394455194473 2023-01-23 09:55:57.329883: step: 576/1541, loss: 0.14599592983722687 2023-01-23 09:55:58.212028: step: 578/1541, loss: 0.12077343463897705 2023-01-23 09:55:59.118849: step: 580/1541, loss: 0.4458065927028656 2023-01-23 09:56:00.017954: step: 582/1541, loss: 0.9607308506965637 2023-01-23 09:56:00.938942: step: 584/1541, loss: 0.2077038437128067 2023-01-23 09:56:01.881106: step: 586/1541, loss: 0.4800177812576294 2023-01-23 09:56:02.806732: step: 588/1541, loss: 0.12922559678554535 2023-01-23 09:56:03.681798: step: 590/1541, loss: 0.26474976539611816 2023-01-23 09:56:04.594469: step: 592/1541, loss: 0.3635310232639313 2023-01-23 09:56:05.477665: step: 594/1541, loss: 0.056609585881233215 2023-01-23 09:56:06.358927: step: 596/1541, loss: 0.17285500466823578 2023-01-23 09:56:07.228398: step: 598/1541, loss: 0.03776932507753372 2023-01-23 09:56:08.185395: step: 600/1541, loss: 0.053282998502254486 2023-01-23 09:56:09.133798: step: 602/1541, loss: 0.16001345217227936 2023-01-23 09:56:10.005998: step: 604/1541, loss: 0.021150708198547363 2023-01-23 09:56:10.955870: step: 606/1541, loss: 0.24671043455600739 2023-01-23 09:56:11.854746: step: 608/1541, loss: 0.1085677295923233 2023-01-23 09:56:12.713334: step: 610/1541, loss: 0.41458654403686523 2023-01-23 09:56:13.619522: step: 612/1541, loss: 0.42737117409706116 2023-01-23 09:56:14.545587: step: 614/1541, loss: 0.08166787773370743 2023-01-23 09:56:15.410930: step: 616/1541, loss: 0.20246222615242004 2023-01-23 09:56:16.300106: step: 618/1541, loss: 0.06499719619750977 2023-01-23 09:56:17.161525: step: 620/1541, loss: 0.0374876894056797 2023-01-23 09:56:18.041927: step: 622/1541, loss: 0.19328072667121887 2023-01-23 09:56:18.924692: step: 624/1541, loss: 0.09957582503557205 2023-01-23 09:56:19.792303: step: 626/1541, loss: 0.3454267978668213 2023-01-23 09:56:20.677231: step: 628/1541, loss: 0.16053324937820435 2023-01-23 09:56:21.574885: step: 630/1541, loss: 0.20322217047214508 2023-01-23 09:56:22.432470: step: 632/1541, loss: 0.1382349729537964 2023-01-23 09:56:23.330687: step: 634/1541, loss: 0.43812403082847595 2023-01-23 09:56:24.247646: step: 636/1541, loss: 0.5242509841918945 2023-01-23 09:56:25.120811: step: 638/1541, loss: 0.8143897652626038 2023-01-23 09:56:25.973194: step: 640/1541, loss: 0.1748550683259964 2023-01-23 09:56:26.883066: step: 642/1541, loss: 0.06652973592281342 2023-01-23 09:56:27.801653: step: 644/1541, loss: 0.20722702145576477 2023-01-23 09:56:28.729732: step: 646/1541, loss: 0.3434882164001465 2023-01-23 09:56:29.624633: step: 648/1541, loss: 0.05381179600954056 2023-01-23 09:56:30.528476: step: 650/1541, loss: 0.041531722992658615 2023-01-23 09:56:31.459753: step: 652/1541, loss: 0.09140176326036453 2023-01-23 09:56:32.385809: step: 654/1541, loss: 0.03153228759765625 2023-01-23 09:56:33.297787: step: 656/1541, loss: 0.08280909806489944 2023-01-23 09:56:34.178281: step: 658/1541, loss: 0.1325855255126953 2023-01-23 09:56:35.064445: step: 660/1541, loss: 0.652807354927063 2023-01-23 09:56:35.953979: step: 662/1541, loss: 0.21840102970600128 2023-01-23 09:56:36.896252: step: 664/1541, loss: 0.0442441888153553 2023-01-23 09:56:37.738993: step: 666/1541, loss: 0.1011999323964119 2023-01-23 09:56:38.574978: step: 668/1541, loss: 0.3618943691253662 2023-01-23 09:56:39.452247: step: 670/1541, loss: 0.17035990953445435 2023-01-23 09:56:40.359087: step: 672/1541, loss: 0.2845936417579651 2023-01-23 09:56:41.263718: step: 674/1541, loss: 0.028390422463417053 2023-01-23 09:56:42.147921: step: 676/1541, loss: 0.05788850784301758 2023-01-23 09:56:43.037432: step: 678/1541, loss: 0.04726305976510048 2023-01-23 09:56:43.907818: step: 680/1541, loss: 0.06825893372297287 2023-01-23 09:56:44.803491: step: 682/1541, loss: 0.11128529906272888 2023-01-23 09:56:45.657622: step: 684/1541, loss: 0.04908549040555954 2023-01-23 09:56:46.518227: step: 686/1541, loss: 0.25975313782691956 2023-01-23 09:56:47.421450: step: 688/1541, loss: 0.34708213806152344 2023-01-23 09:56:48.289195: step: 690/1541, loss: 0.01810324564576149 2023-01-23 09:56:49.175062: step: 692/1541, loss: 0.08290958404541016 2023-01-23 09:56:50.038869: step: 694/1541, loss: 0.018825463950634003 2023-01-23 09:56:50.918163: step: 696/1541, loss: 0.1330423206090927 2023-01-23 09:56:51.829554: step: 698/1541, loss: 0.6218555569648743 2023-01-23 09:56:52.706102: step: 700/1541, loss: 0.040078114718198776 2023-01-23 09:56:53.610244: step: 702/1541, loss: 0.24579866230487823 2023-01-23 09:56:54.550331: step: 704/1541, loss: 0.10466497391462326 2023-01-23 09:56:55.453097: step: 706/1541, loss: 0.03692243993282318 2023-01-23 09:56:56.325388: step: 708/1541, loss: 0.14674760401248932 2023-01-23 09:56:57.172851: step: 710/1541, loss: 0.3016733229160309 2023-01-23 09:56:58.095353: step: 712/1541, loss: 0.05815880000591278 2023-01-23 09:56:59.022430: step: 714/1541, loss: 0.13286961615085602 2023-01-23 09:56:59.905416: step: 716/1541, loss: 0.2649591267108917 2023-01-23 09:57:00.819962: step: 718/1541, loss: 0.3963492214679718 2023-01-23 09:57:01.681393: step: 720/1541, loss: 1.2364994287490845 2023-01-23 09:57:02.587650: step: 722/1541, loss: 0.5182981491088867 2023-01-23 09:57:03.434777: step: 724/1541, loss: 0.10504351556301117 2023-01-23 09:57:04.312418: step: 726/1541, loss: 0.060373205691576004 2023-01-23 09:57:05.225341: step: 728/1541, loss: 0.16789552569389343 2023-01-23 09:57:06.130617: step: 730/1541, loss: 0.052267398685216904 2023-01-23 09:57:07.037952: step: 732/1541, loss: 0.2717927396297455 2023-01-23 09:57:07.927995: step: 734/1541, loss: 0.05465281382203102 2023-01-23 09:57:08.808208: step: 736/1541, loss: 0.11196783185005188 2023-01-23 09:57:09.673732: step: 738/1541, loss: 0.016171477735042572 2023-01-23 09:57:10.600512: step: 740/1541, loss: 0.480745792388916 2023-01-23 09:57:11.465031: step: 742/1541, loss: 0.3033750057220459 2023-01-23 09:57:12.382311: step: 744/1541, loss: 0.2063264548778534 2023-01-23 09:57:13.293338: step: 746/1541, loss: 0.22227366268634796 2023-01-23 09:57:14.175505: step: 748/1541, loss: 0.4196997284889221 2023-01-23 09:57:15.048807: step: 750/1541, loss: 0.5538378953933716 2023-01-23 09:57:15.933901: step: 752/1541, loss: 0.21237358450889587 2023-01-23 09:57:16.809766: step: 754/1541, loss: 0.31981799006462097 2023-01-23 09:57:17.702698: step: 756/1541, loss: 0.17271628975868225 2023-01-23 09:57:18.565097: step: 758/1541, loss: 0.45807135105133057 2023-01-23 09:57:19.418143: step: 760/1541, loss: 0.2696119248867035 2023-01-23 09:57:20.258017: step: 762/1541, loss: 0.09987469762563705 2023-01-23 09:57:21.119790: step: 764/1541, loss: 0.2045655995607376 2023-01-23 09:57:22.024636: step: 766/1541, loss: 0.15636149048805237 2023-01-23 09:57:22.918962: step: 768/1541, loss: 0.11143192648887634 2023-01-23 09:57:23.800339: step: 770/1541, loss: 0.5898945331573486 2023-01-23 09:57:24.758416: step: 772/1541, loss: 0.14090374112129211 2023-01-23 09:57:25.632722: step: 774/1541, loss: 0.0559554286301136 2023-01-23 09:57:26.494500: step: 776/1541, loss: 0.026971058920025826 2023-01-23 09:57:27.393787: step: 778/1541, loss: 0.17190295457839966 2023-01-23 09:57:28.341398: step: 780/1541, loss: 0.0752919465303421 2023-01-23 09:57:29.229393: step: 782/1541, loss: 0.017673088237643242 2023-01-23 09:57:30.135679: step: 784/1541, loss: 0.09994824230670929 2023-01-23 09:57:31.016839: step: 786/1541, loss: 0.1424322873353958 2023-01-23 09:57:31.945386: step: 788/1541, loss: 0.016892217099666595 2023-01-23 09:57:32.828385: step: 790/1541, loss: 0.387380450963974 2023-01-23 09:57:33.784158: step: 792/1541, loss: 0.23802830278873444 2023-01-23 09:57:34.716331: step: 794/1541, loss: 0.18305906653404236 2023-01-23 09:57:35.598418: step: 796/1541, loss: 0.41498756408691406 2023-01-23 09:57:36.460189: step: 798/1541, loss: 0.41611364483833313 2023-01-23 09:57:37.336203: step: 800/1541, loss: 0.018827470019459724 2023-01-23 09:57:38.165027: step: 802/1541, loss: 0.4887574017047882 2023-01-23 09:57:39.034952: step: 804/1541, loss: 0.0897439494729042 2023-01-23 09:57:39.942539: step: 806/1541, loss: 0.5881171226501465 2023-01-23 09:57:40.922270: step: 808/1541, loss: 0.2920692563056946 2023-01-23 09:57:41.865295: step: 810/1541, loss: 0.10764341801404953 2023-01-23 09:57:42.797672: step: 812/1541, loss: 0.2449437975883484 2023-01-23 09:57:43.814645: step: 814/1541, loss: 0.03947371244430542 2023-01-23 09:57:44.718108: step: 816/1541, loss: 0.15225255489349365 2023-01-23 09:57:45.647481: step: 818/1541, loss: 0.38645440340042114 2023-01-23 09:57:46.537763: step: 820/1541, loss: 0.16920991241931915 2023-01-23 09:57:47.432526: step: 822/1541, loss: 0.45099350810050964 2023-01-23 09:57:48.317125: step: 824/1541, loss: 0.9287546873092651 2023-01-23 09:57:49.226819: step: 826/1541, loss: 0.1991133987903595 2023-01-23 09:57:50.131450: step: 828/1541, loss: 0.011471753008663654 2023-01-23 09:57:50.993266: step: 830/1541, loss: 0.09566531330347061 2023-01-23 09:57:51.949777: step: 832/1541, loss: 0.05057687684893608 2023-01-23 09:57:52.845909: step: 834/1541, loss: 2.7997148036956787 2023-01-23 09:57:53.770541: step: 836/1541, loss: 0.07441890984773636 2023-01-23 09:57:54.638247: step: 838/1541, loss: 0.04200877249240875 2023-01-23 09:57:55.583472: step: 840/1541, loss: 0.2676643431186676 2023-01-23 09:57:56.550435: step: 842/1541, loss: 0.5563794374465942 2023-01-23 09:57:57.400803: step: 844/1541, loss: 0.07606419175863266 2023-01-23 09:57:58.289537: step: 846/1541, loss: 1.623862862586975 2023-01-23 09:57:59.164230: step: 848/1541, loss: 0.17698392271995544 2023-01-23 09:58:00.026467: step: 850/1541, loss: 0.052681323140859604 2023-01-23 09:58:01.033494: step: 852/1541, loss: 0.1468668431043625 2023-01-23 09:58:01.946262: step: 854/1541, loss: 0.19907590746879578 2023-01-23 09:58:02.826619: step: 856/1541, loss: 0.10799204558134079 2023-01-23 09:58:03.702638: step: 858/1541, loss: 0.2691182792186737 2023-01-23 09:58:04.582854: step: 860/1541, loss: 0.07284499704837799 2023-01-23 09:58:05.442305: step: 862/1541, loss: 0.19948837161064148 2023-01-23 09:58:06.317941: step: 864/1541, loss: 3.142396926879883 2023-01-23 09:58:07.194064: step: 866/1541, loss: 0.11412058025598526 2023-01-23 09:58:08.062902: step: 868/1541, loss: 0.2722991704940796 2023-01-23 09:58:08.912113: step: 870/1541, loss: 0.1555134803056717 2023-01-23 09:58:09.802013: step: 872/1541, loss: 0.0771317183971405 2023-01-23 09:58:10.677162: step: 874/1541, loss: 0.1301223784685135 2023-01-23 09:58:11.575705: step: 876/1541, loss: 0.4253966212272644 2023-01-23 09:58:12.575438: step: 878/1541, loss: 0.17865894734859467 2023-01-23 09:58:13.478653: step: 880/1541, loss: 0.31324657797813416 2023-01-23 09:58:14.352613: step: 882/1541, loss: 0.08212582767009735 2023-01-23 09:58:15.198475: step: 884/1541, loss: 0.12742654979228973 2023-01-23 09:58:16.134794: step: 886/1541, loss: 0.18246705830097198 2023-01-23 09:58:17.021157: step: 888/1541, loss: 0.13081994652748108 2023-01-23 09:58:17.909884: step: 890/1541, loss: 0.2103862464427948 2023-01-23 09:58:18.760621: step: 892/1541, loss: 0.29558196663856506 2023-01-23 09:58:19.652762: step: 894/1541, loss: 0.08305089920759201 2023-01-23 09:58:20.560547: step: 896/1541, loss: 0.40226075053215027 2023-01-23 09:58:21.465932: step: 898/1541, loss: 0.040114399045705795 2023-01-23 09:58:22.356127: step: 900/1541, loss: 0.08796984702348709 2023-01-23 09:58:23.219640: step: 902/1541, loss: 0.6303611397743225 2023-01-23 09:58:24.093642: step: 904/1541, loss: 0.28145861625671387 2023-01-23 09:58:25.050739: step: 906/1541, loss: 0.15831400454044342 2023-01-23 09:58:25.946419: step: 908/1541, loss: 0.18520909547805786 2023-01-23 09:58:26.824521: step: 910/1541, loss: 0.024348227307200432 2023-01-23 09:58:27.713763: step: 912/1541, loss: 0.45748376846313477 2023-01-23 09:58:28.589176: step: 914/1541, loss: 0.23228119313716888 2023-01-23 09:58:29.514568: step: 916/1541, loss: 0.3177109658718109 2023-01-23 09:58:30.430703: step: 918/1541, loss: 0.4396885931491852 2023-01-23 09:58:31.311750: step: 920/1541, loss: 0.4814327657222748 2023-01-23 09:58:32.211501: step: 922/1541, loss: 0.0926010012626648 2023-01-23 09:58:33.080783: step: 924/1541, loss: 0.15371915698051453 2023-01-23 09:58:33.953369: step: 926/1541, loss: 0.03427673876285553 2023-01-23 09:58:34.886502: step: 928/1541, loss: 0.2841809093952179 2023-01-23 09:58:35.767720: step: 930/1541, loss: 0.1403902918100357 2023-01-23 09:58:36.620585: step: 932/1541, loss: 0.6090277433395386 2023-01-23 09:58:37.510699: step: 934/1541, loss: 0.09383281320333481 2023-01-23 09:58:38.415605: step: 936/1541, loss: 0.037992432713508606 2023-01-23 09:58:39.327774: step: 938/1541, loss: 1.6557482481002808 2023-01-23 09:58:40.230426: step: 940/1541, loss: 0.08598195761442184 2023-01-23 09:58:41.079613: step: 942/1541, loss: 0.0943470299243927 2023-01-23 09:58:41.940048: step: 944/1541, loss: 0.09498445689678192 2023-01-23 09:58:42.800536: step: 946/1541, loss: 0.09894432872533798 2023-01-23 09:58:43.688386: step: 948/1541, loss: 0.2279713749885559 2023-01-23 09:58:44.575610: step: 950/1541, loss: 0.0551806315779686 2023-01-23 09:58:45.461854: step: 952/1541, loss: 0.37211012840270996 2023-01-23 09:58:46.323196: step: 954/1541, loss: 0.25559327006340027 2023-01-23 09:58:47.186517: step: 956/1541, loss: 0.07619312405586243 2023-01-23 09:58:48.088322: step: 958/1541, loss: 0.0323590487241745 2023-01-23 09:58:48.933846: step: 960/1541, loss: 0.06240404024720192 2023-01-23 09:58:49.838979: step: 962/1541, loss: 0.18394315242767334 2023-01-23 09:58:50.715778: step: 964/1541, loss: 0.10749411582946777 2023-01-23 09:58:51.579694: step: 966/1541, loss: 0.21186526119709015 2023-01-23 09:58:52.463869: step: 968/1541, loss: 0.17450450360774994 2023-01-23 09:58:53.368465: step: 970/1541, loss: 0.32936543226242065 2023-01-23 09:58:54.242004: step: 972/1541, loss: 0.1212751492857933 2023-01-23 09:58:55.146406: step: 974/1541, loss: 0.8261388540267944 2023-01-23 09:58:56.015493: step: 976/1541, loss: 0.17210035026073456 2023-01-23 09:58:56.867187: step: 978/1541, loss: 0.3662504553794861 2023-01-23 09:58:57.860516: step: 980/1541, loss: 0.5020464062690735 2023-01-23 09:58:58.796906: step: 982/1541, loss: 0.2942039966583252 2023-01-23 09:58:59.711325: step: 984/1541, loss: 0.3926539719104767 2023-01-23 09:59:00.595625: step: 986/1541, loss: 0.47090649604797363 2023-01-23 09:59:01.463696: step: 988/1541, loss: 0.5988619327545166 2023-01-23 09:59:02.350312: step: 990/1541, loss: 0.11543093621730804 2023-01-23 09:59:03.211563: step: 992/1541, loss: 0.2839638590812683 2023-01-23 09:59:04.060054: step: 994/1541, loss: 0.28824880719184875 2023-01-23 09:59:04.941775: step: 996/1541, loss: 0.1542753130197525 2023-01-23 09:59:05.817769: step: 998/1541, loss: 0.11463537067174911 2023-01-23 09:59:06.711207: step: 1000/1541, loss: 2.42665696144104 2023-01-23 09:59:07.658724: step: 1002/1541, loss: 0.1466531604528427 2023-01-23 09:59:08.595859: step: 1004/1541, loss: 0.2816244661808014 2023-01-23 09:59:09.458348: step: 1006/1541, loss: 0.17998091876506805 2023-01-23 09:59:10.415174: step: 1008/1541, loss: 0.08281778544187546 2023-01-23 09:59:11.311040: step: 1010/1541, loss: 0.3252359926700592 2023-01-23 09:59:12.206953: step: 1012/1541, loss: 0.13849036395549774 2023-01-23 09:59:13.090678: step: 1014/1541, loss: 0.8834522366523743 2023-01-23 09:59:13.958535: step: 1016/1541, loss: 0.19293668866157532 2023-01-23 09:59:14.886703: step: 1018/1541, loss: 0.2984391748905182 2023-01-23 09:59:15.767755: step: 1020/1541, loss: 0.023137733340263367 2023-01-23 09:59:16.639316: step: 1022/1541, loss: 0.4275234639644623 2023-01-23 09:59:17.554334: step: 1024/1541, loss: 0.2617718279361725 2023-01-23 09:59:18.466669: step: 1026/1541, loss: 0.20637784898281097 2023-01-23 09:59:19.449529: step: 1028/1541, loss: 0.06616364419460297 2023-01-23 09:59:20.372528: step: 1030/1541, loss: 0.19799846410751343 2023-01-23 09:59:21.259835: step: 1032/1541, loss: 0.4291686713695526 2023-01-23 09:59:22.184629: step: 1034/1541, loss: 0.02523699402809143 2023-01-23 09:59:23.028107: step: 1036/1541, loss: 0.20577847957611084 2023-01-23 09:59:23.872210: step: 1038/1541, loss: 0.07348163425922394 2023-01-23 09:59:24.777697: step: 1040/1541, loss: 0.06354578584432602 2023-01-23 09:59:25.641757: step: 1042/1541, loss: 0.14851179718971252 2023-01-23 09:59:26.536986: step: 1044/1541, loss: 0.19087502360343933 2023-01-23 09:59:27.459257: step: 1046/1541, loss: 0.2031027227640152 2023-01-23 09:59:28.386287: step: 1048/1541, loss: 0.2347787320613861 2023-01-23 09:59:29.298631: step: 1050/1541, loss: 0.1320699155330658 2023-01-23 09:59:30.218383: step: 1052/1541, loss: 0.10245887190103531 2023-01-23 09:59:31.091014: step: 1054/1541, loss: 0.7305266261100769 2023-01-23 09:59:31.966617: step: 1056/1541, loss: 0.06497642397880554 2023-01-23 09:59:32.818612: step: 1058/1541, loss: 0.06224162131547928 2023-01-23 09:59:33.700672: step: 1060/1541, loss: 0.353915274143219 2023-01-23 09:59:34.591126: step: 1062/1541, loss: 0.028493039309978485 2023-01-23 09:59:35.465872: step: 1064/1541, loss: 0.061636313796043396 2023-01-23 09:59:36.347935: step: 1066/1541, loss: 0.1480022668838501 2023-01-23 09:59:37.252939: step: 1068/1541, loss: 0.8746756911277771 2023-01-23 09:59:38.104830: step: 1070/1541, loss: 0.07131201028823853 2023-01-23 09:59:38.973686: step: 1072/1541, loss: 0.18354979157447815 2023-01-23 09:59:39.859558: step: 1074/1541, loss: 0.22368192672729492 2023-01-23 09:59:40.834198: step: 1076/1541, loss: 0.199369415640831 2023-01-23 09:59:41.712464: step: 1078/1541, loss: 0.4087916910648346 2023-01-23 09:59:42.620012: step: 1080/1541, loss: 0.6765972375869751 2023-01-23 09:59:43.534654: step: 1082/1541, loss: 0.14994066953659058 2023-01-23 09:59:44.356393: step: 1084/1541, loss: 0.45704150199890137 2023-01-23 09:59:45.204918: step: 1086/1541, loss: 0.1753227412700653 2023-01-23 09:59:46.060825: step: 1088/1541, loss: 0.1473478078842163 2023-01-23 09:59:46.932410: step: 1090/1541, loss: 0.17896795272827148 2023-01-23 09:59:47.779976: step: 1092/1541, loss: 0.2693363130092621 2023-01-23 09:59:48.662340: step: 1094/1541, loss: 0.10465782880783081 2023-01-23 09:59:49.563108: step: 1096/1541, loss: 0.19435633718967438 2023-01-23 09:59:50.422710: step: 1098/1541, loss: 0.4061468541622162 2023-01-23 09:59:51.334593: step: 1100/1541, loss: 0.29757529497146606 2023-01-23 09:59:52.220058: step: 1102/1541, loss: 0.08330229669809341 2023-01-23 09:59:53.109098: step: 1104/1541, loss: 0.3932095766067505 2023-01-23 09:59:53.979242: step: 1106/1541, loss: 0.12220307439565659 2023-01-23 09:59:54.844063: step: 1108/1541, loss: 0.42020195722579956 2023-01-23 09:59:55.720635: step: 1110/1541, loss: 0.22671081125736237 2023-01-23 09:59:56.683439: step: 1112/1541, loss: 0.22323639690876007 2023-01-23 09:59:57.531103: step: 1114/1541, loss: 0.19976259768009186 2023-01-23 09:59:58.427822: step: 1116/1541, loss: 0.12486173212528229 2023-01-23 09:59:59.324521: step: 1118/1541, loss: 0.3156995475292206 2023-01-23 10:00:00.191131: step: 1120/1541, loss: 0.12974952161312103 2023-01-23 10:00:01.080204: step: 1122/1541, loss: 0.04042389616370201 2023-01-23 10:00:01.969589: step: 1124/1541, loss: 0.20035304129123688 2023-01-23 10:00:02.933224: step: 1126/1541, loss: 0.3967030346393585 2023-01-23 10:00:03.822459: step: 1128/1541, loss: 0.21194924414157867 2023-01-23 10:00:04.717568: step: 1130/1541, loss: 0.17145834863185883 2023-01-23 10:00:05.583725: step: 1132/1541, loss: 0.15740810334682465 2023-01-23 10:00:06.510756: step: 1134/1541, loss: 0.15415982902050018 2023-01-23 10:00:07.400386: step: 1136/1541, loss: 0.16548149287700653 2023-01-23 10:00:08.294963: step: 1138/1541, loss: 0.040734659880399704 2023-01-23 10:00:09.229237: step: 1140/1541, loss: 0.37019553780555725 2023-01-23 10:00:10.146997: step: 1142/1541, loss: 0.1431932896375656 2023-01-23 10:00:11.119962: step: 1144/1541, loss: 0.4906178116798401 2023-01-23 10:00:12.008885: step: 1146/1541, loss: 0.4906354248523712 2023-01-23 10:00:12.918150: step: 1148/1541, loss: 0.03526934236288071 2023-01-23 10:00:13.744204: step: 1150/1541, loss: 0.1718561202287674 2023-01-23 10:00:14.639142: step: 1152/1541, loss: 0.2153809517621994 2023-01-23 10:00:15.536776: step: 1154/1541, loss: 0.09280535578727722 2023-01-23 10:00:16.489460: step: 1156/1541, loss: 0.025212517008185387 2023-01-23 10:00:17.376536: step: 1158/1541, loss: 0.28779804706573486 2023-01-23 10:00:18.246503: step: 1160/1541, loss: 0.3467833399772644 2023-01-23 10:00:19.155743: step: 1162/1541, loss: 1.2504863739013672 2023-01-23 10:00:20.055855: step: 1164/1541, loss: 0.04243594408035278 2023-01-23 10:00:20.938079: step: 1166/1541, loss: 0.11771157383918762 2023-01-23 10:00:21.827544: step: 1168/1541, loss: 0.02584385871887207 2023-01-23 10:00:22.725967: step: 1170/1541, loss: 0.2319343090057373 2023-01-23 10:00:23.663207: step: 1172/1541, loss: 1.229602336883545 2023-01-23 10:00:24.552414: step: 1174/1541, loss: 0.028984209522604942 2023-01-23 10:00:25.398893: step: 1176/1541, loss: 0.280948281288147 2023-01-23 10:00:26.295589: step: 1178/1541, loss: 0.35089877247810364 2023-01-23 10:00:27.204737: step: 1180/1541, loss: 0.5148988962173462 2023-01-23 10:00:28.077110: step: 1182/1541, loss: 0.09577827900648117 2023-01-23 10:00:29.013918: step: 1184/1541, loss: 0.04830079525709152 2023-01-23 10:00:29.954906: step: 1186/1541, loss: 0.23429569602012634 2023-01-23 10:00:30.851807: step: 1188/1541, loss: 0.30480220913887024 2023-01-23 10:00:31.781069: step: 1190/1541, loss: 0.20468246936798096 2023-01-23 10:00:32.768004: step: 1192/1541, loss: 0.13233229517936707 2023-01-23 10:00:33.702075: step: 1194/1541, loss: 0.285038024187088 2023-01-23 10:00:34.627866: step: 1196/1541, loss: 0.03097529709339142 2023-01-23 10:00:35.482396: step: 1198/1541, loss: 0.1934569776058197 2023-01-23 10:00:36.389802: step: 1200/1541, loss: 0.2633362114429474 2023-01-23 10:00:37.254975: step: 1202/1541, loss: 0.25337550044059753 2023-01-23 10:00:38.162108: step: 1204/1541, loss: 0.08846218883991241 2023-01-23 10:00:39.081405: step: 1206/1541, loss: 0.871947705745697 2023-01-23 10:00:39.995845: step: 1208/1541, loss: 0.09593226760625839 2023-01-23 10:00:40.873592: step: 1210/1541, loss: 0.13824249804019928 2023-01-23 10:00:41.740848: step: 1212/1541, loss: 0.08624104410409927 2023-01-23 10:00:42.621962: step: 1214/1541, loss: 0.4548869729042053 2023-01-23 10:00:43.513962: step: 1216/1541, loss: 0.08977502584457397 2023-01-23 10:00:44.434250: step: 1218/1541, loss: 0.2198830246925354 2023-01-23 10:00:45.330704: step: 1220/1541, loss: 0.2129269689321518 2023-01-23 10:00:46.219962: step: 1222/1541, loss: 0.1347821056842804 2023-01-23 10:00:47.167414: step: 1224/1541, loss: 0.6322583556175232 2023-01-23 10:00:48.023641: step: 1226/1541, loss: 0.03033360093832016 2023-01-23 10:00:48.904794: step: 1228/1541, loss: 0.27292782068252563 2023-01-23 10:00:49.800752: step: 1230/1541, loss: 0.03352636843919754 2023-01-23 10:00:50.674088: step: 1232/1541, loss: 0.11894948035478592 2023-01-23 10:00:51.579717: step: 1234/1541, loss: 0.2582622468471527 2023-01-23 10:00:52.435825: step: 1236/1541, loss: 0.09343062341213226 2023-01-23 10:00:53.317962: step: 1238/1541, loss: 0.19234538078308105 2023-01-23 10:00:54.192302: step: 1240/1541, loss: 0.5511993765830994 2023-01-23 10:00:55.059512: step: 1242/1541, loss: 0.06843453645706177 2023-01-23 10:00:55.930505: step: 1244/1541, loss: 0.21496012806892395 2023-01-23 10:00:56.759239: step: 1246/1541, loss: 0.09417610615491867 2023-01-23 10:00:57.645872: step: 1248/1541, loss: 0.036843374371528625 2023-01-23 10:00:58.541182: step: 1250/1541, loss: 0.20664915442466736 2023-01-23 10:00:59.469817: step: 1252/1541, loss: 0.22043509781360626 2023-01-23 10:01:00.374071: step: 1254/1541, loss: 0.07266135513782501 2023-01-23 10:01:01.243595: step: 1256/1541, loss: 0.14670398831367493 2023-01-23 10:01:02.190301: step: 1258/1541, loss: 0.28050464391708374 2023-01-23 10:01:03.085261: step: 1260/1541, loss: 0.14778853952884674 2023-01-23 10:01:03.952561: step: 1262/1541, loss: 0.02984769456088543 2023-01-23 10:01:04.844829: step: 1264/1541, loss: 0.3222203552722931 2023-01-23 10:01:05.740938: step: 1266/1541, loss: 0.15047574043273926 2023-01-23 10:01:06.632872: step: 1268/1541, loss: 0.3173059821128845 2023-01-23 10:01:07.541626: step: 1270/1541, loss: 0.3158576190471649 2023-01-23 10:01:08.420228: step: 1272/1541, loss: 0.38365793228149414 2023-01-23 10:01:09.312483: step: 1274/1541, loss: 0.15133832395076752 2023-01-23 10:01:10.207231: step: 1276/1541, loss: 0.16921907663345337 2023-01-23 10:01:11.114702: step: 1278/1541, loss: 0.9161630868911743 2023-01-23 10:01:12.035063: step: 1280/1541, loss: 0.2880553901195526 2023-01-23 10:01:12.979619: step: 1282/1541, loss: 0.03316709026694298 2023-01-23 10:01:13.831142: step: 1284/1541, loss: 0.46172887086868286 2023-01-23 10:01:14.676818: step: 1286/1541, loss: 0.12012830376625061 2023-01-23 10:01:15.569640: step: 1288/1541, loss: 0.11374978721141815 2023-01-23 10:01:16.425219: step: 1290/1541, loss: 0.49088147282600403 2023-01-23 10:01:17.270112: step: 1292/1541, loss: 0.8372253775596619 2023-01-23 10:01:18.172291: step: 1294/1541, loss: 0.05674440413713455 2023-01-23 10:01:19.048184: step: 1296/1541, loss: 0.16349166631698608 2023-01-23 10:01:19.912059: step: 1298/1541, loss: 0.08377614617347717 2023-01-23 10:01:20.817802: step: 1300/1541, loss: 0.34649232029914856 2023-01-23 10:01:21.686204: step: 1302/1541, loss: 0.5677289366722107 2023-01-23 10:01:22.553913: step: 1304/1541, loss: 0.1438327133655548 2023-01-23 10:01:23.432629: step: 1306/1541, loss: 0.1634601652622223 2023-01-23 10:01:24.268320: step: 1308/1541, loss: 0.20663224160671234 2023-01-23 10:01:25.131285: step: 1310/1541, loss: 0.13120301067829132 2023-01-23 10:01:26.003712: step: 1312/1541, loss: 0.29468706250190735 2023-01-23 10:01:26.875382: step: 1314/1541, loss: 0.07021208852529526 2023-01-23 10:01:27.745637: step: 1316/1541, loss: 0.4224720597267151 2023-01-23 10:01:28.636482: step: 1318/1541, loss: 0.7435312867164612 2023-01-23 10:01:29.546757: step: 1320/1541, loss: 0.22554171085357666 2023-01-23 10:01:30.430823: step: 1322/1541, loss: 0.19700533151626587 2023-01-23 10:01:31.387869: step: 1324/1541, loss: 0.12248902767896652 2023-01-23 10:01:32.281633: step: 1326/1541, loss: 0.0973980575799942 2023-01-23 10:01:33.175007: step: 1328/1541, loss: 0.15450061857700348 2023-01-23 10:01:34.079272: step: 1330/1541, loss: 0.5507363080978394 2023-01-23 10:01:34.935770: step: 1332/1541, loss: 0.750016987323761 2023-01-23 10:01:35.814169: step: 1334/1541, loss: 0.6940247416496277 2023-01-23 10:01:36.680260: step: 1336/1541, loss: 0.265410840511322 2023-01-23 10:01:37.641825: step: 1338/1541, loss: 0.08605016767978668 2023-01-23 10:01:38.513853: step: 1340/1541, loss: 0.41109684109687805 2023-01-23 10:01:39.444069: step: 1342/1541, loss: 0.15740041434764862 2023-01-23 10:01:40.376563: step: 1344/1541, loss: 0.2975683808326721 2023-01-23 10:01:41.255234: step: 1346/1541, loss: 0.3102381229400635 2023-01-23 10:01:42.134216: step: 1348/1541, loss: 0.12938737869262695 2023-01-23 10:01:43.031621: step: 1350/1541, loss: 0.3006160855293274 2023-01-23 10:01:43.996959: step: 1352/1541, loss: 0.184710294008255 2023-01-23 10:01:44.901628: step: 1354/1541, loss: 0.090128093957901 2023-01-23 10:01:45.777874: step: 1356/1541, loss: 0.42516618967056274 2023-01-23 10:01:46.641835: step: 1358/1541, loss: 0.15965056419372559 2023-01-23 10:01:47.504617: step: 1360/1541, loss: 0.15597285330295563 2023-01-23 10:01:48.415659: step: 1362/1541, loss: 0.12968674302101135 2023-01-23 10:01:49.307494: step: 1364/1541, loss: 0.1300598382949829 2023-01-23 10:01:50.217710: step: 1366/1541, loss: 0.0719119980931282 2023-01-23 10:01:51.155403: step: 1368/1541, loss: 0.10306086391210556 2023-01-23 10:01:52.048092: step: 1370/1541, loss: 0.1578925997018814 2023-01-23 10:01:52.916172: step: 1372/1541, loss: 0.191089928150177 2023-01-23 10:01:53.808960: step: 1374/1541, loss: 0.195257306098938 2023-01-23 10:01:54.696583: step: 1376/1541, loss: 0.5715517997741699 2023-01-23 10:01:55.598662: step: 1378/1541, loss: 0.10946568846702576 2023-01-23 10:01:56.501189: step: 1380/1541, loss: 0.06764218211174011 2023-01-23 10:01:57.405420: step: 1382/1541, loss: 0.011542655527591705 2023-01-23 10:01:58.338692: step: 1384/1541, loss: 0.11842811852693558 2023-01-23 10:01:59.229626: step: 1386/1541, loss: 0.029856977984309196 2023-01-23 10:02:00.160899: step: 1388/1541, loss: 0.23224563896656036 2023-01-23 10:02:01.071185: step: 1390/1541, loss: 0.2248784303665161 2023-01-23 10:02:01.945025: step: 1392/1541, loss: 0.33939453959465027 2023-01-23 10:02:02.828259: step: 1394/1541, loss: 0.13489937782287598 2023-01-23 10:02:03.721832: step: 1396/1541, loss: 0.1690477430820465 2023-01-23 10:02:04.608676: step: 1398/1541, loss: 0.5254474878311157 2023-01-23 10:02:05.510560: step: 1400/1541, loss: 0.44383928179740906 2023-01-23 10:02:06.381578: step: 1402/1541, loss: 0.08783402293920517 2023-01-23 10:02:07.297704: step: 1404/1541, loss: 0.10614370554685593 2023-01-23 10:02:08.184292: step: 1406/1541, loss: 0.14933952689170837 2023-01-23 10:02:09.093665: step: 1408/1541, loss: 0.2641569674015045 2023-01-23 10:02:10.050138: step: 1410/1541, loss: 0.22279779613018036 2023-01-23 10:02:10.924598: step: 1412/1541, loss: 0.17148661613464355 2023-01-23 10:02:11.789801: step: 1414/1541, loss: 0.16980823874473572 2023-01-23 10:02:12.723690: step: 1416/1541, loss: 0.23447275161743164 2023-01-23 10:02:13.607110: step: 1418/1541, loss: 0.1878277212381363 2023-01-23 10:02:14.495746: step: 1420/1541, loss: 0.06776374578475952 2023-01-23 10:02:15.373966: step: 1422/1541, loss: 0.4596870541572571 2023-01-23 10:02:16.284156: step: 1424/1541, loss: 0.060334645211696625 2023-01-23 10:02:17.173305: step: 1426/1541, loss: 0.4462471902370453 2023-01-23 10:02:18.058435: step: 1428/1541, loss: 0.12452436983585358 2023-01-23 10:02:18.926815: step: 1430/1541, loss: 0.13495826721191406 2023-01-23 10:02:19.896885: step: 1432/1541, loss: 0.23905572295188904 2023-01-23 10:02:20.804901: step: 1434/1541, loss: 0.3317893445491791 2023-01-23 10:02:21.669553: step: 1436/1541, loss: 0.5702382922172546 2023-01-23 10:02:22.552962: step: 1438/1541, loss: 0.12858456373214722 2023-01-23 10:02:23.426832: step: 1440/1541, loss: 0.15035150945186615 2023-01-23 10:02:24.328753: step: 1442/1541, loss: 0.1226540133357048 2023-01-23 10:02:25.243660: step: 1444/1541, loss: 0.2506676912307739 2023-01-23 10:02:26.112271: step: 1446/1541, loss: 0.08368875086307526 2023-01-23 10:02:27.027277: step: 1448/1541, loss: 0.10226321965456009 2023-01-23 10:02:27.902784: step: 1450/1541, loss: 0.05231126770377159 2023-01-23 10:02:28.853199: step: 1452/1541, loss: 0.12786924839019775 2023-01-23 10:02:29.763939: step: 1454/1541, loss: 0.11712800711393356 2023-01-23 10:02:30.672946: step: 1456/1541, loss: 0.028889760375022888 2023-01-23 10:02:31.533214: step: 1458/1541, loss: 0.5884401202201843 2023-01-23 10:02:32.413535: step: 1460/1541, loss: 0.20800042152404785 2023-01-23 10:02:33.323855: step: 1462/1541, loss: 0.22942502796649933 2023-01-23 10:02:34.203633: step: 1464/1541, loss: 0.2611391544342041 2023-01-23 10:02:35.128617: step: 1466/1541, loss: 0.020051412284374237 2023-01-23 10:02:36.030482: step: 1468/1541, loss: 1.007437825202942 2023-01-23 10:02:36.929344: step: 1470/1541, loss: 0.2093716561794281 2023-01-23 10:02:37.807399: step: 1472/1541, loss: 0.25038984417915344 2023-01-23 10:02:38.715104: step: 1474/1541, loss: 0.3870786726474762 2023-01-23 10:02:39.592859: step: 1476/1541, loss: 0.012226935476064682 2023-01-23 10:02:40.517755: step: 1478/1541, loss: 0.5620296597480774 2023-01-23 10:02:41.453161: step: 1480/1541, loss: 0.13159282505512238 2023-01-23 10:02:42.407817: step: 1482/1541, loss: 0.04054320976138115 2023-01-23 10:02:43.324154: step: 1484/1541, loss: 0.21936288475990295 2023-01-23 10:02:44.225895: step: 1486/1541, loss: 0.5937581062316895 2023-01-23 10:02:45.119541: step: 1488/1541, loss: 0.11401878297328949 2023-01-23 10:02:45.991922: step: 1490/1541, loss: 0.22566822171211243 2023-01-23 10:02:46.882399: step: 1492/1541, loss: 0.1679385006427765 2023-01-23 10:02:47.803672: step: 1494/1541, loss: 0.196164071559906 2023-01-23 10:02:48.648617: step: 1496/1541, loss: 0.20186805725097656 2023-01-23 10:02:49.534438: step: 1498/1541, loss: 0.10251334309577942 2023-01-23 10:02:50.422251: step: 1500/1541, loss: 0.226583331823349 2023-01-23 10:02:51.410029: step: 1502/1541, loss: 0.40444549918174744 2023-01-23 10:02:52.288703: step: 1504/1541, loss: 0.03983260318636894 2023-01-23 10:02:53.154941: step: 1506/1541, loss: 0.095394067466259 2023-01-23 10:02:54.053309: step: 1508/1541, loss: 0.07620473206043243 2023-01-23 10:02:54.951015: step: 1510/1541, loss: 0.43230485916137695 2023-01-23 10:02:55.805217: step: 1512/1541, loss: 0.10618473589420319 2023-01-23 10:02:56.668286: step: 1514/1541, loss: 0.14931949973106384 2023-01-23 10:02:57.564275: step: 1516/1541, loss: 0.060113415122032166 2023-01-23 10:02:58.408230: step: 1518/1541, loss: 0.06267307698726654 2023-01-23 10:02:59.315042: step: 1520/1541, loss: 0.31806641817092896 2023-01-23 10:03:00.187145: step: 1522/1541, loss: 0.18428319692611694 2023-01-23 10:03:01.061055: step: 1524/1541, loss: 0.02816775068640709 2023-01-23 10:03:01.967509: step: 1526/1541, loss: 0.11815538257360458 2023-01-23 10:03:02.870903: step: 1528/1541, loss: 0.08442022651433945 2023-01-23 10:03:03.783551: step: 1530/1541, loss: 0.6207011342048645 2023-01-23 10:03:04.711280: step: 1532/1541, loss: 0.02823096513748169 2023-01-23 10:03:05.601538: step: 1534/1541, loss: 0.39074981212615967 2023-01-23 10:03:06.475572: step: 1536/1541, loss: 0.05891439691185951 2023-01-23 10:03:07.391941: step: 1538/1541, loss: 0.3770109713077545 2023-01-23 10:03:08.260097: step: 1540/1541, loss: 0.015158070251345634 2023-01-23 10:03:09.124714: step: 1542/1541, loss: 0.2524159252643585 2023-01-23 10:03:10.019485: step: 1544/1541, loss: 0.0491531565785408 2023-01-23 10:03:10.894296: step: 1546/1541, loss: 0.2747595012187958 2023-01-23 10:03:11.748204: step: 1548/1541, loss: 0.1802530139684677 2023-01-23 10:03:12.632210: step: 1550/1541, loss: 0.4357498288154602 2023-01-23 10:03:13.516156: step: 1552/1541, loss: 0.07233517616987228 2023-01-23 10:03:14.413318: step: 1554/1541, loss: 0.17309586703777313 2023-01-23 10:03:15.305519: step: 1556/1541, loss: 0.14884009957313538 2023-01-23 10:03:16.191502: step: 1558/1541, loss: 0.3937899172306061 2023-01-23 10:03:17.082876: step: 1560/1541, loss: 0.18472078442573547 2023-01-23 10:03:17.997308: step: 1562/1541, loss: 0.20696504414081573 2023-01-23 10:03:18.861565: step: 1564/1541, loss: 0.31096333265304565 2023-01-23 10:03:19.725216: step: 1566/1541, loss: 0.35468363761901855 2023-01-23 10:03:20.608133: step: 1568/1541, loss: 0.47868120670318604 2023-01-23 10:03:21.485174: step: 1570/1541, loss: 0.3269248902797699 2023-01-23 10:03:22.407100: step: 1572/1541, loss: 0.2641526162624359 2023-01-23 10:03:23.335931: step: 1574/1541, loss: 0.16016465425491333 2023-01-23 10:03:24.257065: step: 1576/1541, loss: 0.034494549036026 2023-01-23 10:03:25.136958: step: 1578/1541, loss: 0.07645715028047562 2023-01-23 10:03:26.016030: step: 1580/1541, loss: 0.17309001088142395 2023-01-23 10:03:26.890112: step: 1582/1541, loss: 0.05731002613902092 2023-01-23 10:03:27.795382: step: 1584/1541, loss: 0.11290565133094788 2023-01-23 10:03:28.700085: step: 1586/1541, loss: 0.09096347540616989 2023-01-23 10:03:29.609510: step: 1588/1541, loss: 0.09445777535438538 2023-01-23 10:03:30.483639: step: 1590/1541, loss: 0.21362413465976715 2023-01-23 10:03:31.355212: step: 1592/1541, loss: 0.028980743139982224 2023-01-23 10:03:32.266828: step: 1594/1541, loss: 0.11432187259197235 2023-01-23 10:03:33.182163: step: 1596/1541, loss: 0.3985253870487213 2023-01-23 10:03:34.102198: step: 1598/1541, loss: 0.3776247799396515 2023-01-23 10:03:35.063667: step: 1600/1541, loss: 0.10586008429527283 2023-01-23 10:03:35.964248: step: 1602/1541, loss: 0.019440211355686188 2023-01-23 10:03:36.846370: step: 1604/1541, loss: 0.2776131331920624 2023-01-23 10:03:37.719221: step: 1606/1541, loss: 0.09367608278989792 2023-01-23 10:03:38.585533: step: 1608/1541, loss: 0.008961817249655724 2023-01-23 10:03:39.531538: step: 1610/1541, loss: 0.34549471735954285 2023-01-23 10:03:40.457806: step: 1612/1541, loss: 0.5333029627799988 2023-01-23 10:03:41.344887: step: 1614/1541, loss: 0.2306893914937973 2023-01-23 10:03:42.257998: step: 1616/1541, loss: 0.11358075588941574 2023-01-23 10:03:43.130995: step: 1618/1541, loss: 0.07728207111358643 2023-01-23 10:03:44.027557: step: 1620/1541, loss: 0.13749590516090393 2023-01-23 10:03:44.992450: step: 1622/1541, loss: 0.05787939950823784 2023-01-23 10:03:45.943531: step: 1624/1541, loss: 0.031092317774891853 2023-01-23 10:03:46.830659: step: 1626/1541, loss: 0.6284499764442444 2023-01-23 10:03:47.709401: step: 1628/1541, loss: 0.10297174751758575 2023-01-23 10:03:48.613861: step: 1630/1541, loss: 0.24576616287231445 2023-01-23 10:03:49.516686: step: 1632/1541, loss: 0.4748353064060211 2023-01-23 10:03:50.430696: step: 1634/1541, loss: 0.27134284377098083 2023-01-23 10:03:51.338544: step: 1636/1541, loss: 0.09713345766067505 2023-01-23 10:03:52.248069: step: 1638/1541, loss: 0.03423170745372772 2023-01-23 10:03:53.114019: step: 1640/1541, loss: 0.29356837272644043 2023-01-23 10:03:53.963191: step: 1642/1541, loss: 4.930833339691162 2023-01-23 10:03:54.868792: step: 1644/1541, loss: 0.42171725630760193 2023-01-23 10:03:55.798460: step: 1646/1541, loss: 0.24943892657756805 2023-01-23 10:03:56.662796: step: 1648/1541, loss: 0.3697364628314972 2023-01-23 10:03:57.563919: step: 1650/1541, loss: 0.2577477693557739 2023-01-23 10:03:58.448091: step: 1652/1541, loss: 0.02877977304160595 2023-01-23 10:03:59.381564: step: 1654/1541, loss: 0.13337793946266174 2023-01-23 10:04:00.277319: step: 1656/1541, loss: 0.34836557507514954 2023-01-23 10:04:01.120028: step: 1658/1541, loss: 0.1353806108236313 2023-01-23 10:04:02.057365: step: 1660/1541, loss: 0.1813162863254547 2023-01-23 10:04:02.987902: step: 1662/1541, loss: 0.10135219991207123 2023-01-23 10:04:03.897754: step: 1664/1541, loss: 1.4557723999023438 2023-01-23 10:04:04.837750: step: 1666/1541, loss: 0.03511808440089226 2023-01-23 10:04:05.761023: step: 1668/1541, loss: 0.12410231679677963 2023-01-23 10:04:06.641016: step: 1670/1541, loss: 0.33718588948249817 2023-01-23 10:04:07.542887: step: 1672/1541, loss: 0.12819501757621765 2023-01-23 10:04:08.425885: step: 1674/1541, loss: 0.023365730419754982 2023-01-23 10:04:09.295241: step: 1676/1541, loss: 0.20196351408958435 2023-01-23 10:04:10.154234: step: 1678/1541, loss: 0.07798891514539719 2023-01-23 10:04:11.030841: step: 1680/1541, loss: 0.12641432881355286 2023-01-23 10:04:11.929983: step: 1682/1541, loss: 0.08539959043264389 2023-01-23 10:04:12.788829: step: 1684/1541, loss: 0.10994552820920944 2023-01-23 10:04:13.651434: step: 1686/1541, loss: 0.21395212411880493 2023-01-23 10:04:14.555381: step: 1688/1541, loss: 0.08810080587863922 2023-01-23 10:04:15.436470: step: 1690/1541, loss: 0.14784061908721924 2023-01-23 10:04:16.322980: step: 1692/1541, loss: 0.14052192866802216 2023-01-23 10:04:17.254651: step: 1694/1541, loss: 0.06995540112257004 2023-01-23 10:04:18.123537: step: 1696/1541, loss: 0.01886986568570137 2023-01-23 10:04:19.054010: step: 1698/1541, loss: 0.24606074392795563 2023-01-23 10:04:19.916340: step: 1700/1541, loss: 0.08724486082792282 2023-01-23 10:04:20.800305: step: 1702/1541, loss: 0.04426249861717224 2023-01-23 10:04:21.709394: step: 1704/1541, loss: 0.35539427399635315 2023-01-23 10:04:22.592256: step: 1706/1541, loss: 0.07161653786897659 2023-01-23 10:04:23.479570: step: 1708/1541, loss: 0.07575526833534241 2023-01-23 10:04:24.363757: step: 1710/1541, loss: 0.12281621247529984 2023-01-23 10:04:25.255632: step: 1712/1541, loss: 0.26626092195510864 2023-01-23 10:04:26.140663: step: 1714/1541, loss: 1.6938403844833374 2023-01-23 10:04:27.072181: step: 1716/1541, loss: 0.09309490025043488 2023-01-23 10:04:27.969631: step: 1718/1541, loss: 0.11687871068716049 2023-01-23 10:04:28.901009: step: 1720/1541, loss: 0.0517238974571228 2023-01-23 10:04:29.770458: step: 1722/1541, loss: 0.17980678379535675 2023-01-23 10:04:30.665939: step: 1724/1541, loss: 0.06688953936100006 2023-01-23 10:04:31.584477: step: 1726/1541, loss: 0.3143911063671112 2023-01-23 10:04:32.465529: step: 1728/1541, loss: 0.17348040640354156 2023-01-23 10:04:33.347021: step: 1730/1541, loss: 2.3993775844573975 2023-01-23 10:04:34.237133: step: 1732/1541, loss: 0.09461557120084763 2023-01-23 10:04:35.097178: step: 1734/1541, loss: 0.3272949457168579 2023-01-23 10:04:35.946717: step: 1736/1541, loss: 0.038790538907051086 2023-01-23 10:04:36.828567: step: 1738/1541, loss: 0.02766355313360691 2023-01-23 10:04:37.724460: step: 1740/1541, loss: 0.736857533454895 2023-01-23 10:04:38.624963: step: 1742/1541, loss: 0.06965532153844833 2023-01-23 10:04:39.543763: step: 1744/1541, loss: 0.3524792492389679 2023-01-23 10:04:40.388680: step: 1746/1541, loss: 0.12566234171390533 2023-01-23 10:04:41.348329: step: 1748/1541, loss: 0.06756795197725296 2023-01-23 10:04:42.243696: step: 1750/1541, loss: 0.21863842010498047 2023-01-23 10:04:43.169985: step: 1752/1541, loss: 0.12140945345163345 2023-01-23 10:04:44.050971: step: 1754/1541, loss: 0.23827087879180908 2023-01-23 10:04:44.921628: step: 1756/1541, loss: 0.5732818841934204 2023-01-23 10:04:45.844653: step: 1758/1541, loss: 0.05802568420767784 2023-01-23 10:04:46.786213: step: 1760/1541, loss: 0.39385664463043213 2023-01-23 10:04:47.703875: step: 1762/1541, loss: 0.0618375726044178 2023-01-23 10:04:48.619115: step: 1764/1541, loss: 0.30733153223991394 2023-01-23 10:04:49.501744: step: 1766/1541, loss: 0.07447945326566696 2023-01-23 10:04:50.379087: step: 1768/1541, loss: 0.794759213924408 2023-01-23 10:04:51.258623: step: 1770/1541, loss: 0.29906171560287476 2023-01-23 10:04:52.162191: step: 1772/1541, loss: 0.2628743648529053 2023-01-23 10:04:53.021217: step: 1774/1541, loss: 0.12679600715637207 2023-01-23 10:04:53.938910: step: 1776/1541, loss: 0.07892642170190811 2023-01-23 10:04:54.818593: step: 1778/1541, loss: 0.31078070402145386 2023-01-23 10:04:55.715968: step: 1780/1541, loss: 0.33132117986679077 2023-01-23 10:04:56.632796: step: 1782/1541, loss: 0.06829150021076202 2023-01-23 10:04:57.524106: step: 1784/1541, loss: 0.2476823627948761 2023-01-23 10:04:58.387656: step: 1786/1541, loss: 0.1588459312915802 2023-01-23 10:04:59.333391: step: 1788/1541, loss: 1.8815289735794067 2023-01-23 10:05:00.226301: step: 1790/1541, loss: 0.3174060583114624 2023-01-23 10:05:01.157120: step: 1792/1541, loss: 0.23459209501743317 2023-01-23 10:05:02.141706: step: 1794/1541, loss: 0.1896134912967682 2023-01-23 10:05:03.013089: step: 1796/1541, loss: 0.03706773743033409 2023-01-23 10:05:03.889392: step: 1798/1541, loss: 0.06946242600679398 2023-01-23 10:05:04.815869: step: 1800/1541, loss: 0.16240012645721436 2023-01-23 10:05:05.667666: step: 1802/1541, loss: 0.19264090061187744 2023-01-23 10:05:06.533574: step: 1804/1541, loss: 0.0303892120718956 2023-01-23 10:05:07.415349: step: 1806/1541, loss: 0.04471006989479065 2023-01-23 10:05:08.326802: step: 1808/1541, loss: 0.38064801692962646 2023-01-23 10:05:09.187551: step: 1810/1541, loss: 0.1063527762889862 2023-01-23 10:05:10.066739: step: 1812/1541, loss: 0.5222527384757996 2023-01-23 10:05:10.948733: step: 1814/1541, loss: 0.5472899675369263 2023-01-23 10:05:11.835583: step: 1816/1541, loss: 0.40369275212287903 2023-01-23 10:05:12.713397: step: 1818/1541, loss: 0.0007812640396878123 2023-01-23 10:05:13.603255: step: 1820/1541, loss: 0.16007308661937714 2023-01-23 10:05:14.498774: step: 1822/1541, loss: 0.5441920757293701 2023-01-23 10:05:15.421170: step: 1824/1541, loss: 0.38821855187416077 2023-01-23 10:05:16.244075: step: 1826/1541, loss: 0.11523570120334625 2023-01-23 10:05:17.129705: step: 1828/1541, loss: 0.3754579722881317 2023-01-23 10:05:18.064089: step: 1830/1541, loss: 0.03940008208155632 2023-01-23 10:05:18.969611: step: 1832/1541, loss: 0.14418479800224304 2023-01-23 10:05:19.869415: step: 1834/1541, loss: 0.27254220843315125 2023-01-23 10:05:20.748481: step: 1836/1541, loss: 0.22917556762695312 2023-01-23 10:05:21.586435: step: 1838/1541, loss: 0.2098105400800705 2023-01-23 10:05:22.537341: step: 1840/1541, loss: 0.06203056499361992 2023-01-23 10:05:23.473135: step: 1842/1541, loss: 0.23772591352462769 2023-01-23 10:05:24.348843: step: 1844/1541, loss: 0.30134886503219604 2023-01-23 10:05:25.239567: step: 1846/1541, loss: 0.07731509953737259 2023-01-23 10:05:26.150697: step: 1848/1541, loss: 0.05930429697036743 2023-01-23 10:05:27.049761: step: 1850/1541, loss: 0.3815804123878479 2023-01-23 10:05:27.908975: step: 1852/1541, loss: 0.3458426892757416 2023-01-23 10:05:28.900867: step: 1854/1541, loss: 0.8067540526390076 2023-01-23 10:05:29.832132: step: 1856/1541, loss: 0.6645575165748596 2023-01-23 10:05:30.684776: step: 1858/1541, loss: 0.12894563376903534 2023-01-23 10:05:31.564926: step: 1860/1541, loss: 0.298915833234787 2023-01-23 10:05:32.454693: step: 1862/1541, loss: 0.032177604734897614 2023-01-23 10:05:33.308572: step: 1864/1541, loss: 0.3054853081703186 2023-01-23 10:05:34.188498: step: 1866/1541, loss: 0.18382932245731354 2023-01-23 10:05:35.076668: step: 1868/1541, loss: 0.24140183627605438 2023-01-23 10:05:36.003610: step: 1870/1541, loss: 0.10363582521677017 2023-01-23 10:05:36.884234: step: 1872/1541, loss: 0.1294577717781067 2023-01-23 10:05:37.768666: step: 1874/1541, loss: 0.27635201811790466 2023-01-23 10:05:38.646734: step: 1876/1541, loss: 0.3236853778362274 2023-01-23 10:05:39.563153: step: 1878/1541, loss: 0.43120360374450684 2023-01-23 10:05:40.444470: step: 1880/1541, loss: 0.32784250378608704 2023-01-23 10:05:41.283027: step: 1882/1541, loss: 0.06053059920668602 2023-01-23 10:05:42.148453: step: 1884/1541, loss: 0.05588248744606972 2023-01-23 10:05:43.095122: step: 1886/1541, loss: 0.02947891689836979 2023-01-23 10:05:44.000887: step: 1888/1541, loss: 0.07083448022603989 2023-01-23 10:05:44.899281: step: 1890/1541, loss: 0.22821396589279175 2023-01-23 10:05:45.818610: step: 1892/1541, loss: 4.013778209686279 2023-01-23 10:05:46.697025: step: 1894/1541, loss: 0.68080735206604 2023-01-23 10:05:47.578307: step: 1896/1541, loss: 0.16365770995616913 2023-01-23 10:05:48.510251: step: 1898/1541, loss: 0.24133293330669403 2023-01-23 10:05:49.405584: step: 1900/1541, loss: 0.07535029947757721 2023-01-23 10:05:50.270965: step: 1902/1541, loss: 0.21632200479507446 2023-01-23 10:05:51.157388: step: 1904/1541, loss: 0.20859278738498688 2023-01-23 10:05:52.050002: step: 1906/1541, loss: 0.11619226634502411 2023-01-23 10:05:52.989309: step: 1908/1541, loss: 0.0933876484632492 2023-01-23 10:05:53.940100: step: 1910/1541, loss: 0.1568206399679184 2023-01-23 10:05:54.818417: step: 1912/1541, loss: 0.1610950082540512 2023-01-23 10:05:55.674782: step: 1914/1541, loss: 0.17376992106437683 2023-01-23 10:05:56.605266: step: 1916/1541, loss: 0.9072670936584473 2023-01-23 10:05:57.554637: step: 1918/1541, loss: 0.06199941039085388 2023-01-23 10:05:58.400377: step: 1920/1541, loss: 0.7584981918334961 2023-01-23 10:05:59.308426: step: 1922/1541, loss: 0.8913253545761108 2023-01-23 10:06:00.198242: step: 1924/1541, loss: 0.16094982624053955 2023-01-23 10:06:01.133102: step: 1926/1541, loss: 0.09150167554616928 2023-01-23 10:06:02.029602: step: 1928/1541, loss: 0.14073051512241364 2023-01-23 10:06:02.925095: step: 1930/1541, loss: 0.31639572978019714 2023-01-23 10:06:03.802016: step: 1932/1541, loss: 0.3229427635669708 2023-01-23 10:06:04.773777: step: 1934/1541, loss: 0.16777381300926208 2023-01-23 10:06:05.643660: step: 1936/1541, loss: 2.6777377128601074 2023-01-23 10:06:06.565739: step: 1938/1541, loss: 0.37703052163124084 2023-01-23 10:06:07.431119: step: 1940/1541, loss: 0.2352573424577713 2023-01-23 10:06:08.359357: step: 1942/1541, loss: 0.11480479687452316 2023-01-23 10:06:09.231515: step: 1944/1541, loss: 0.14802667498588562 2023-01-23 10:06:10.116734: step: 1946/1541, loss: 0.46576905250549316 2023-01-23 10:06:10.980675: step: 1948/1541, loss: 0.16104349493980408 2023-01-23 10:06:11.857654: step: 1950/1541, loss: 0.23017928004264832 2023-01-23 10:06:12.735383: step: 1952/1541, loss: 0.12224756181240082 2023-01-23 10:06:13.628179: step: 1954/1541, loss: 2.6655192375183105 2023-01-23 10:06:14.535577: step: 1956/1541, loss: 0.12594839930534363 2023-01-23 10:06:15.434263: step: 1958/1541, loss: 0.12304140627384186 2023-01-23 10:06:16.325939: step: 1960/1541, loss: 0.32312896847724915 2023-01-23 10:06:17.185138: step: 1962/1541, loss: 0.07807095348834991 2023-01-23 10:06:18.077242: step: 1964/1541, loss: 0.13681991398334503 2023-01-23 10:06:18.973540: step: 1966/1541, loss: 0.23221658170223236 2023-01-23 10:06:19.906089: step: 1968/1541, loss: 0.1627674102783203 2023-01-23 10:06:20.782780: step: 1970/1541, loss: 0.07266730070114136 2023-01-23 10:06:21.661762: step: 1972/1541, loss: 0.6505405306816101 2023-01-23 10:06:22.537731: step: 1974/1541, loss: 0.1723688393831253 2023-01-23 10:06:23.427380: step: 1976/1541, loss: 0.05701199546456337 2023-01-23 10:06:24.319891: step: 1978/1541, loss: 0.1996580809354782 2023-01-23 10:06:25.224698: step: 1980/1541, loss: 0.039436787366867065 2023-01-23 10:06:26.132085: step: 1982/1541, loss: 0.2886192500591278 2023-01-23 10:06:26.990155: step: 1984/1541, loss: 0.06737978756427765 2023-01-23 10:06:27.869472: step: 1986/1541, loss: 0.7763397097587585 2023-01-23 10:06:28.767377: step: 1988/1541, loss: 0.07411019504070282 2023-01-23 10:06:29.668576: step: 1990/1541, loss: 0.15350446105003357 2023-01-23 10:06:30.524845: step: 1992/1541, loss: 0.05438750609755516 2023-01-23 10:06:31.413543: step: 1994/1541, loss: 1.0337027311325073 2023-01-23 10:06:32.375624: step: 1996/1541, loss: 0.058144114911556244 2023-01-23 10:06:33.250132: step: 1998/1541, loss: 0.18872879445552826 2023-01-23 10:06:34.162006: step: 2000/1541, loss: 0.031035013496875763 2023-01-23 10:06:35.044841: step: 2002/1541, loss: 0.20361556112766266 2023-01-23 10:06:35.927055: step: 2004/1541, loss: 0.07926327735185623 2023-01-23 10:06:36.881305: step: 2006/1541, loss: 0.521790087223053 2023-01-23 10:06:37.777745: step: 2008/1541, loss: 0.3253403902053833 2023-01-23 10:06:38.671020: step: 2010/1541, loss: 0.13108675181865692 2023-01-23 10:06:39.589025: step: 2012/1541, loss: 0.3266574740409851 2023-01-23 10:06:40.458756: step: 2014/1541, loss: 0.1975729763507843 2023-01-23 10:06:41.332377: step: 2016/1541, loss: 0.20410391688346863 2023-01-23 10:06:42.233977: step: 2018/1541, loss: 0.13208259642124176 2023-01-23 10:06:43.139900: step: 2020/1541, loss: 0.23550313711166382 2023-01-23 10:06:44.027156: step: 2022/1541, loss: 1.3095450401306152 2023-01-23 10:06:44.946591: step: 2024/1541, loss: 0.17458826303482056 2023-01-23 10:06:45.817427: step: 2026/1541, loss: 0.2832958996295929 2023-01-23 10:06:46.694802: step: 2028/1541, loss: 0.17292240262031555 2023-01-23 10:06:47.564065: step: 2030/1541, loss: 0.14631740748882294 2023-01-23 10:06:48.436967: step: 2032/1541, loss: 0.14910529553890228 2023-01-23 10:06:49.286136: step: 2034/1541, loss: 0.3003441393375397 2023-01-23 10:06:50.175805: step: 2036/1541, loss: 0.3098946809768677 2023-01-23 10:06:51.073988: step: 2038/1541, loss: 0.05513467639684677 2023-01-23 10:06:51.939310: step: 2040/1541, loss: 0.127846822142601 2023-01-23 10:06:52.831924: step: 2042/1541, loss: 0.09930315613746643 2023-01-23 10:06:53.726776: step: 2044/1541, loss: 0.0766662210226059 2023-01-23 10:06:54.602038: step: 2046/1541, loss: 0.448342889547348 2023-01-23 10:06:55.487782: step: 2048/1541, loss: 0.09667807072401047 2023-01-23 10:06:56.395719: step: 2050/1541, loss: 0.211411252617836 2023-01-23 10:06:57.315692: step: 2052/1541, loss: 0.054366715252399445 2023-01-23 10:06:58.221811: step: 2054/1541, loss: 2.667447090148926 2023-01-23 10:06:59.141299: step: 2056/1541, loss: 0.22028248012065887 2023-01-23 10:06:59.994642: step: 2058/1541, loss: 0.06671981513500214 2023-01-23 10:07:00.881430: step: 2060/1541, loss: 0.07846815139055252 2023-01-23 10:07:01.808222: step: 2062/1541, loss: 0.17303234338760376 2023-01-23 10:07:02.689093: step: 2064/1541, loss: 0.007269986439496279 2023-01-23 10:07:03.574505: step: 2066/1541, loss: 0.5303552746772766 2023-01-23 10:07:04.467334: step: 2068/1541, loss: 0.1631292849779129 2023-01-23 10:07:05.323412: step: 2070/1541, loss: 0.3867970407009125 2023-01-23 10:07:06.237161: step: 2072/1541, loss: 0.09590408205986023 2023-01-23 10:07:07.116439: step: 2074/1541, loss: 0.09948544949293137 2023-01-23 10:07:08.013602: step: 2076/1541, loss: 0.09067979454994202 2023-01-23 10:07:08.960896: step: 2078/1541, loss: 0.41805532574653625 2023-01-23 10:07:09.869930: step: 2080/1541, loss: 0.8016523718833923 2023-01-23 10:07:10.786307: step: 2082/1541, loss: 0.15590624511241913 2023-01-23 10:07:11.660931: step: 2084/1541, loss: 0.090089350938797 2023-01-23 10:07:12.554143: step: 2086/1541, loss: 0.42552515864372253 2023-01-23 10:07:13.428622: step: 2088/1541, loss: 0.18994997441768646 2023-01-23 10:07:14.323431: step: 2090/1541, loss: 0.033660128712654114 2023-01-23 10:07:15.282988: step: 2092/1541, loss: 0.013630608096718788 2023-01-23 10:07:16.130689: step: 2094/1541, loss: 0.5032479166984558 2023-01-23 10:07:16.994635: step: 2096/1541, loss: 0.3778481185436249 2023-01-23 10:07:17.901350: step: 2098/1541, loss: 0.4629822075366974 2023-01-23 10:07:18.790782: step: 2100/1541, loss: 0.5062333345413208 2023-01-23 10:07:19.662798: step: 2102/1541, loss: 6.871192455291748 2023-01-23 10:07:20.539792: step: 2104/1541, loss: 0.12686027586460114 2023-01-23 10:07:21.396997: step: 2106/1541, loss: 0.4271026849746704 2023-01-23 10:07:22.247530: step: 2108/1541, loss: 0.49740782380104065 2023-01-23 10:07:23.150693: step: 2110/1541, loss: 0.06763026118278503 2023-01-23 10:07:24.044228: step: 2112/1541, loss: 0.07396601885557175 2023-01-23 10:07:24.920801: step: 2114/1541, loss: 0.15963006019592285 2023-01-23 10:07:25.833843: step: 2116/1541, loss: 0.1179879903793335 2023-01-23 10:07:26.705889: step: 2118/1541, loss: 0.04706330969929695 2023-01-23 10:07:27.573794: step: 2120/1541, loss: 0.05462836101651192 2023-01-23 10:07:28.448857: step: 2122/1541, loss: 0.012301423586905003 2023-01-23 10:07:29.377375: step: 2124/1541, loss: 0.017612287774682045 2023-01-23 10:07:30.268883: step: 2126/1541, loss: 0.14306814968585968 2023-01-23 10:07:31.154786: step: 2128/1541, loss: 0.09809056669473648 2023-01-23 10:07:32.000901: step: 2130/1541, loss: 0.4162103533744812 2023-01-23 10:07:32.933852: step: 2132/1541, loss: 0.25503861904144287 2023-01-23 10:07:33.795422: step: 2134/1541, loss: 0.11864251643419266 2023-01-23 10:07:34.707063: step: 2136/1541, loss: 0.028582096099853516 2023-01-23 10:07:35.635980: step: 2138/1541, loss: 0.12173382937908173 2023-01-23 10:07:36.499879: step: 2140/1541, loss: 0.04984813183546066 2023-01-23 10:07:37.403526: step: 2142/1541, loss: 0.4804038405418396 2023-01-23 10:07:38.309120: step: 2144/1541, loss: 0.23167483508586884 2023-01-23 10:07:39.236188: step: 2146/1541, loss: 0.0598752498626709 2023-01-23 10:07:40.125945: step: 2148/1541, loss: 0.02846798300743103 2023-01-23 10:07:41.024886: step: 2150/1541, loss: 0.3225451111793518 2023-01-23 10:07:42.022574: step: 2152/1541, loss: 0.22393785417079926 2023-01-23 10:07:42.917814: step: 2154/1541, loss: 0.04329291731119156 2023-01-23 10:07:43.800929: step: 2156/1541, loss: 0.8139138221740723 2023-01-23 10:07:44.714223: step: 2158/1541, loss: 0.07526593655347824 2023-01-23 10:07:45.579996: step: 2160/1541, loss: 0.1021074578166008 2023-01-23 10:07:46.493498: step: 2162/1541, loss: 0.13693827390670776 2023-01-23 10:07:47.404549: step: 2164/1541, loss: 0.07535558938980103 2023-01-23 10:07:48.319076: step: 2166/1541, loss: 0.7203501462936401 2023-01-23 10:07:49.222987: step: 2168/1541, loss: 0.03774705156683922 2023-01-23 10:07:50.178316: step: 2170/1541, loss: 0.04865601658821106 2023-01-23 10:07:51.168613: step: 2172/1541, loss: 0.0650583803653717 2023-01-23 10:07:52.086491: step: 2174/1541, loss: 0.12849034368991852 2023-01-23 10:07:52.988758: step: 2176/1541, loss: 0.3499852418899536 2023-01-23 10:07:53.862153: step: 2178/1541, loss: 0.09117577224969864 2023-01-23 10:07:54.764703: step: 2180/1541, loss: 0.17524836957454681 2023-01-23 10:07:55.667673: step: 2182/1541, loss: 0.13711798191070557 2023-01-23 10:07:56.537254: step: 2184/1541, loss: 0.03928158059716225 2023-01-23 10:07:57.473297: step: 2186/1541, loss: 0.0438535213470459 2023-01-23 10:07:58.418684: step: 2188/1541, loss: 0.22365602850914001 2023-01-23 10:07:59.296750: step: 2190/1541, loss: 0.5327985882759094 2023-01-23 10:08:00.175628: step: 2192/1541, loss: 0.05278198793530464 2023-01-23 10:08:01.074663: step: 2194/1541, loss: 0.17124098539352417 2023-01-23 10:08:02.020976: step: 2196/1541, loss: 0.23916839063167572 2023-01-23 10:08:02.913042: step: 2198/1541, loss: 0.3911369740962982 2023-01-23 10:08:03.804507: step: 2200/1541, loss: 0.09961455315351486 2023-01-23 10:08:04.683573: step: 2202/1541, loss: 0.1445411592721939 2023-01-23 10:08:05.556001: step: 2204/1541, loss: 0.08041870594024658 2023-01-23 10:08:06.442651: step: 2206/1541, loss: 0.12883935868740082 2023-01-23 10:08:07.323149: step: 2208/1541, loss: 0.14517249166965485 2023-01-23 10:08:08.222239: step: 2210/1541, loss: 0.6671572923660278 2023-01-23 10:08:09.112641: step: 2212/1541, loss: 0.14508451521396637 2023-01-23 10:08:09.998739: step: 2214/1541, loss: 0.028539003804326057 2023-01-23 10:08:10.914385: step: 2216/1541, loss: 0.09978733956813812 2023-01-23 10:08:11.797071: step: 2218/1541, loss: 0.17435237765312195 2023-01-23 10:08:12.648213: step: 2220/1541, loss: 0.17074070870876312 2023-01-23 10:08:13.527714: step: 2222/1541, loss: 0.18602225184440613 2023-01-23 10:08:14.399825: step: 2224/1541, loss: 0.8501771688461304 2023-01-23 10:08:15.312617: step: 2226/1541, loss: 0.20145444571971893 2023-01-23 10:08:16.170339: step: 2228/1541, loss: 0.05240466818213463 2023-01-23 10:08:17.078774: step: 2230/1541, loss: 0.5256107449531555 2023-01-23 10:08:18.065042: step: 2232/1541, loss: 0.1571916937828064 2023-01-23 10:08:18.950835: step: 2234/1541, loss: 0.011033602058887482 2023-01-23 10:08:19.918019: step: 2236/1541, loss: 0.31336510181427 2023-01-23 10:08:20.854080: step: 2238/1541, loss: 2.274803638458252 2023-01-23 10:08:21.709933: step: 2240/1541, loss: 0.15675726532936096 2023-01-23 10:08:22.636373: step: 2242/1541, loss: 0.22228902578353882 2023-01-23 10:08:23.499589: step: 2244/1541, loss: 0.1612531840801239 2023-01-23 10:08:24.351761: step: 2246/1541, loss: 0.025879614055156708 2023-01-23 10:08:25.266383: step: 2248/1541, loss: 0.21653762459754944 2023-01-23 10:08:26.134838: step: 2250/1541, loss: 0.09453490376472473 2023-01-23 10:08:27.022260: step: 2252/1541, loss: 0.32114750146865845 2023-01-23 10:08:27.889300: step: 2254/1541, loss: 0.33229923248291016 2023-01-23 10:08:28.789588: step: 2256/1541, loss: 0.17780597507953644 2023-01-23 10:08:29.681195: step: 2258/1541, loss: 0.3019281327724457 2023-01-23 10:08:30.519926: step: 2260/1541, loss: 0.30357885360717773 2023-01-23 10:08:31.447927: step: 2262/1541, loss: 0.16400137543678284 2023-01-23 10:08:32.348155: step: 2264/1541, loss: 0.06485030800104141 2023-01-23 10:08:33.209953: step: 2266/1541, loss: 0.08801700174808502 2023-01-23 10:08:34.117386: step: 2268/1541, loss: 0.02462923526763916 2023-01-23 10:08:34.939396: step: 2270/1541, loss: 0.16800528764724731 2023-01-23 10:08:35.802588: step: 2272/1541, loss: 0.1740894466638565 2023-01-23 10:08:36.687042: step: 2274/1541, loss: 0.18416792154312134 2023-01-23 10:08:37.547349: step: 2276/1541, loss: 0.27032873034477234 2023-01-23 10:08:38.443240: step: 2278/1541, loss: 0.6975831389427185 2023-01-23 10:08:39.373370: step: 2280/1541, loss: 0.3794836401939392 2023-01-23 10:08:40.246126: step: 2282/1541, loss: 0.07863181829452515 2023-01-23 10:08:41.119725: step: 2284/1541, loss: 1.2088948488235474 2023-01-23 10:08:42.061696: step: 2286/1541, loss: 0.25704213976860046 2023-01-23 10:08:43.010721: step: 2288/1541, loss: 0.2243809849023819 2023-01-23 10:08:43.879195: step: 2290/1541, loss: 0.12039123475551605 2023-01-23 10:08:44.737105: step: 2292/1541, loss: 0.10350792109966278 2023-01-23 10:08:45.604100: step: 2294/1541, loss: 0.25263696908950806 2023-01-23 10:08:46.494395: step: 2296/1541, loss: 0.05844896286725998 2023-01-23 10:08:47.369459: step: 2298/1541, loss: 0.3419080674648285 2023-01-23 10:08:48.224669: step: 2300/1541, loss: 0.4363268315792084 2023-01-23 10:08:49.088469: step: 2302/1541, loss: 0.3966640830039978 2023-01-23 10:08:49.931558: step: 2304/1541, loss: 0.5330513119697571 2023-01-23 10:08:50.869560: step: 2306/1541, loss: 0.06113972142338753 2023-01-23 10:08:51.710929: step: 2308/1541, loss: 0.027508946135640144 2023-01-23 10:08:52.649631: step: 2310/1541, loss: 0.3952196538448334 2023-01-23 10:08:53.504501: step: 2312/1541, loss: 0.36734598875045776 2023-01-23 10:08:54.375319: step: 2314/1541, loss: 0.20684275031089783 2023-01-23 10:08:55.244956: step: 2316/1541, loss: 0.05673838406801224 2023-01-23 10:08:56.104268: step: 2318/1541, loss: 0.05074680969119072 2023-01-23 10:08:56.982255: step: 2320/1541, loss: 0.13302189111709595 2023-01-23 10:08:57.887674: step: 2322/1541, loss: 0.2455388307571411 2023-01-23 10:08:58.830184: step: 2324/1541, loss: 0.18000854551792145 2023-01-23 10:08:59.712965: step: 2326/1541, loss: 0.25911572575569153 2023-01-23 10:09:00.608942: step: 2328/1541, loss: 0.19298206269741058 2023-01-23 10:09:01.531040: step: 2330/1541, loss: 0.21264424920082092 2023-01-23 10:09:02.433341: step: 2332/1541, loss: 0.32592594623565674 2023-01-23 10:09:03.307337: step: 2334/1541, loss: 0.16792166233062744 2023-01-23 10:09:04.182565: step: 2336/1541, loss: 0.04787774384021759 2023-01-23 10:09:05.079625: step: 2338/1541, loss: 0.08198204636573792 2023-01-23 10:09:05.983714: step: 2340/1541, loss: 0.08749619871377945 2023-01-23 10:09:06.880452: step: 2342/1541, loss: 0.15211406350135803 2023-01-23 10:09:07.739582: step: 2344/1541, loss: 0.0687873438000679 2023-01-23 10:09:08.620025: step: 2346/1541, loss: 0.10234348475933075 2023-01-23 10:09:09.520992: step: 2348/1541, loss: 0.16326770186424255 2023-01-23 10:09:10.411672: step: 2350/1541, loss: 0.09649834036827087 2023-01-23 10:09:11.326918: step: 2352/1541, loss: 0.26848945021629333 2023-01-23 10:09:12.254444: step: 2354/1541, loss: 0.06767235696315765 2023-01-23 10:09:13.207733: step: 2356/1541, loss: 0.24443107843399048 2023-01-23 10:09:14.081006: step: 2358/1541, loss: 0.616755485534668 2023-01-23 10:09:14.994779: step: 2360/1541, loss: 0.10138267278671265 2023-01-23 10:09:15.937255: step: 2362/1541, loss: 0.3767945170402527 2023-01-23 10:09:16.798858: step: 2364/1541, loss: 0.23802128434181213 2023-01-23 10:09:17.701928: step: 2366/1541, loss: 0.20390693843364716 2023-01-23 10:09:18.618972: step: 2368/1541, loss: 0.3761470913887024 2023-01-23 10:09:19.468028: step: 2370/1541, loss: 0.2581820785999298 2023-01-23 10:09:20.333683: step: 2372/1541, loss: 0.09561117738485336 2023-01-23 10:09:21.257301: step: 2374/1541, loss: 0.12467093765735626 2023-01-23 10:09:22.167894: step: 2376/1541, loss: 0.4359615445137024 2023-01-23 10:09:23.024518: step: 2378/1541, loss: 1.810438632965088 2023-01-23 10:09:23.931968: step: 2380/1541, loss: 2.1002273559570312 2023-01-23 10:09:24.815337: step: 2382/1541, loss: 0.6869134902954102 2023-01-23 10:09:25.722244: step: 2384/1541, loss: 0.03718483820557594 2023-01-23 10:09:26.604605: step: 2386/1541, loss: 0.214650496840477 2023-01-23 10:09:27.474709: step: 2388/1541, loss: 0.06994795054197311 2023-01-23 10:09:28.384187: step: 2390/1541, loss: 0.3733657896518707 2023-01-23 10:09:29.267781: step: 2392/1541, loss: 0.22609661519527435 2023-01-23 10:09:30.145436: step: 2394/1541, loss: 0.4323694109916687 2023-01-23 10:09:31.019454: step: 2396/1541, loss: 0.0897093340754509 2023-01-23 10:09:31.923468: step: 2398/1541, loss: 0.023687629029154778 2023-01-23 10:09:32.829820: step: 2400/1541, loss: 0.14389744400978088 2023-01-23 10:09:33.728081: step: 2402/1541, loss: 0.011191200464963913 2023-01-23 10:09:34.626117: step: 2404/1541, loss: 0.18359529972076416 2023-01-23 10:09:35.615081: step: 2406/1541, loss: 0.2635972797870636 2023-01-23 10:09:36.492592: step: 2408/1541, loss: 0.2443736344575882 2023-01-23 10:09:37.390291: step: 2410/1541, loss: 0.13496068120002747 2023-01-23 10:09:38.325290: step: 2412/1541, loss: 0.27331316471099854 2023-01-23 10:09:39.245609: step: 2414/1541, loss: 0.2428807020187378 2023-01-23 10:09:40.118268: step: 2416/1541, loss: 0.5803418159484863 2023-01-23 10:09:40.983901: step: 2418/1541, loss: 0.11176493018865585 2023-01-23 10:09:41.844221: step: 2420/1541, loss: 0.20287825167179108 2023-01-23 10:09:42.701549: step: 2422/1541, loss: 0.09489720314741135 2023-01-23 10:09:43.649832: step: 2424/1541, loss: 0.5076581239700317 2023-01-23 10:09:44.639936: step: 2426/1541, loss: 0.08382488787174225 2023-01-23 10:09:45.511734: step: 2428/1541, loss: 0.08116928488016129 2023-01-23 10:09:46.385535: step: 2430/1541, loss: 0.3634673058986664 2023-01-23 10:09:47.268101: step: 2432/1541, loss: 0.12763705849647522 2023-01-23 10:09:48.177873: step: 2434/1541, loss: 0.24948450922966003 2023-01-23 10:09:49.039161: step: 2436/1541, loss: 0.5052886605262756 2023-01-23 10:09:49.945280: step: 2438/1541, loss: 0.14504456520080566 2023-01-23 10:09:50.845991: step: 2440/1541, loss: 0.0973585695028305 2023-01-23 10:09:51.813220: step: 2442/1541, loss: 0.2720952332019806 2023-01-23 10:09:52.707454: step: 2444/1541, loss: 0.11105157434940338 2023-01-23 10:09:53.587985: step: 2446/1541, loss: 0.08521769195795059 2023-01-23 10:09:54.446898: step: 2448/1541, loss: 0.22391101717948914 2023-01-23 10:09:55.351518: step: 2450/1541, loss: 0.10848104953765869 2023-01-23 10:09:56.287337: step: 2452/1541, loss: 0.12443847209215164 2023-01-23 10:09:57.197484: step: 2454/1541, loss: 6.8537702560424805 2023-01-23 10:09:58.080833: step: 2456/1541, loss: 0.0720854178071022 2023-01-23 10:09:58.954349: step: 2458/1541, loss: 0.09336688369512558 2023-01-23 10:09:59.870373: step: 2460/1541, loss: 0.5658198595046997 2023-01-23 10:10:00.808445: step: 2462/1541, loss: 0.13742485642433167 2023-01-23 10:10:01.695508: step: 2464/1541, loss: 0.6032230854034424 2023-01-23 10:10:02.594186: step: 2466/1541, loss: 0.5185580849647522 2023-01-23 10:10:03.522501: step: 2468/1541, loss: 0.00661793677136302 2023-01-23 10:10:04.392073: step: 2470/1541, loss: 0.8925098180770874 2023-01-23 10:10:05.336212: step: 2472/1541, loss: 0.16838665306568146 2023-01-23 10:10:06.243502: step: 2474/1541, loss: 1.017669439315796 2023-01-23 10:10:07.106998: step: 2476/1541, loss: 0.2405010610818863 2023-01-23 10:10:07.986426: step: 2478/1541, loss: 0.27425846457481384 2023-01-23 10:10:08.892964: step: 2480/1541, loss: 0.3316341042518616 2023-01-23 10:10:09.761513: step: 2482/1541, loss: 0.22223515808582306 2023-01-23 10:10:10.650305: step: 2484/1541, loss: 0.10793214291334152 2023-01-23 10:10:11.531189: step: 2486/1541, loss: 0.5485631227493286 2023-01-23 10:10:12.407138: step: 2488/1541, loss: 0.0760134682059288 2023-01-23 10:10:13.283995: step: 2490/1541, loss: 0.6706346273422241 2023-01-23 10:10:14.187199: step: 2492/1541, loss: 0.17385253310203552 2023-01-23 10:10:15.061299: step: 2494/1541, loss: 0.09731021523475647 2023-01-23 10:10:16.012804: step: 2496/1541, loss: 0.1662360578775406 2023-01-23 10:10:16.948320: step: 2498/1541, loss: 0.12463396787643433 2023-01-23 10:10:17.853525: step: 2500/1541, loss: 0.09887885302305222 2023-01-23 10:10:18.775478: step: 2502/1541, loss: 0.30696985125541687 2023-01-23 10:10:19.633581: step: 2504/1541, loss: 0.14133664965629578 2023-01-23 10:10:20.512993: step: 2506/1541, loss: 0.21513712406158447 2023-01-23 10:10:21.405818: step: 2508/1541, loss: 0.45811694860458374 2023-01-23 10:10:22.327589: step: 2510/1541, loss: 0.28087741136550903 2023-01-23 10:10:23.296199: step: 2512/1541, loss: 0.019363537430763245 2023-01-23 10:10:24.172216: step: 2514/1541, loss: 0.27291440963745117 2023-01-23 10:10:25.032768: step: 2516/1541, loss: 0.14518913626670837 2023-01-23 10:10:25.961776: step: 2518/1541, loss: 0.17460036277770996 2023-01-23 10:10:26.818655: step: 2520/1541, loss: 0.044933296740055084 2023-01-23 10:10:27.737300: step: 2522/1541, loss: 0.2949897348880768 2023-01-23 10:10:28.611784: step: 2524/1541, loss: 0.31596389412879944 2023-01-23 10:10:29.474799: step: 2526/1541, loss: 0.049179594963788986 2023-01-23 10:10:30.360516: step: 2528/1541, loss: 0.054962366819381714 2023-01-23 10:10:31.270296: step: 2530/1541, loss: 0.042414769530296326 2023-01-23 10:10:32.213436: step: 2532/1541, loss: 0.17695598304271698 2023-01-23 10:10:33.064809: step: 2534/1541, loss: 0.20116442441940308 2023-01-23 10:10:33.997408: step: 2536/1541, loss: 0.11236931383609772 2023-01-23 10:10:34.850366: step: 2538/1541, loss: 0.09608316421508789 2023-01-23 10:10:35.708569: step: 2540/1541, loss: 0.026486149057745934 2023-01-23 10:10:36.617997: step: 2542/1541, loss: 0.11872412264347076 2023-01-23 10:10:37.498337: step: 2544/1541, loss: 0.13977007567882538 2023-01-23 10:10:38.405813: step: 2546/1541, loss: 0.26302599906921387 2023-01-23 10:10:39.251779: step: 2548/1541, loss: 0.3215595781803131 2023-01-23 10:10:40.164201: step: 2550/1541, loss: 0.11892961710691452 2023-01-23 10:10:41.027526: step: 2552/1541, loss: 0.14198964834213257 2023-01-23 10:10:41.899141: step: 2554/1541, loss: 0.3734790086746216 2023-01-23 10:10:42.764493: step: 2556/1541, loss: 0.14694906771183014 2023-01-23 10:10:43.709559: step: 2558/1541, loss: 0.16643786430358887 2023-01-23 10:10:44.612721: step: 2560/1541, loss: 0.23431000113487244 2023-01-23 10:10:45.494256: step: 2562/1541, loss: 0.15287254750728607 2023-01-23 10:10:46.409489: step: 2564/1541, loss: 1.0824072360992432 2023-01-23 10:10:47.377785: step: 2566/1541, loss: 0.05614890530705452 2023-01-23 10:10:48.264572: step: 2568/1541, loss: 0.0888809785246849 2023-01-23 10:10:49.132096: step: 2570/1541, loss: 0.06550852209329605 2023-01-23 10:10:50.058578: step: 2572/1541, loss: 0.04819900542497635 2023-01-23 10:10:50.952555: step: 2574/1541, loss: 0.2254493236541748 2023-01-23 10:10:51.863086: step: 2576/1541, loss: 0.41184353828430176 2023-01-23 10:10:52.832987: step: 2578/1541, loss: 0.21855586767196655 2023-01-23 10:10:53.680471: step: 2580/1541, loss: 0.3805515766143799 2023-01-23 10:10:54.554775: step: 2582/1541, loss: 0.6051371097564697 2023-01-23 10:10:55.462387: step: 2584/1541, loss: 0.6534004211425781 2023-01-23 10:10:56.362787: step: 2586/1541, loss: 0.2351655811071396 2023-01-23 10:10:57.256137: step: 2588/1541, loss: 0.09470713138580322 2023-01-23 10:10:58.156106: step: 2590/1541, loss: 1.0770891904830933 2023-01-23 10:10:59.048570: step: 2592/1541, loss: 0.18026745319366455 2023-01-23 10:10:59.922869: step: 2594/1541, loss: 0.07306304574012756 2023-01-23 10:11:00.860620: step: 2596/1541, loss: 0.08283396810293198 2023-01-23 10:11:01.725925: step: 2598/1541, loss: 0.19631625711917877 2023-01-23 10:11:02.685093: step: 2600/1541, loss: 0.17251448333263397 2023-01-23 10:11:03.562894: step: 2602/1541, loss: 0.2094275951385498 2023-01-23 10:11:04.441656: step: 2604/1541, loss: 0.6572602987289429 2023-01-23 10:11:05.300338: step: 2606/1541, loss: 0.1661013811826706 2023-01-23 10:11:06.192469: step: 2608/1541, loss: 0.02741401270031929 2023-01-23 10:11:07.074055: step: 2610/1541, loss: 0.2653932571411133 2023-01-23 10:11:07.968903: step: 2612/1541, loss: 0.35123586654663086 2023-01-23 10:11:08.865068: step: 2614/1541, loss: 0.16563093662261963 2023-01-23 10:11:09.756002: step: 2616/1541, loss: 0.053927842527627945 2023-01-23 10:11:10.614697: step: 2618/1541, loss: 0.03133951872587204 2023-01-23 10:11:11.556648: step: 2620/1541, loss: 0.22086986899375916 2023-01-23 10:11:12.428161: step: 2622/1541, loss: 0.7249641418457031 2023-01-23 10:11:13.374196: step: 2624/1541, loss: 0.17772060632705688 2023-01-23 10:11:14.242332: step: 2626/1541, loss: 0.7809244990348816 2023-01-23 10:11:15.147054: step: 2628/1541, loss: 0.16479933261871338 2023-01-23 10:11:16.000045: step: 2630/1541, loss: 0.01864325813949108 2023-01-23 10:11:16.861741: step: 2632/1541, loss: 0.4603176414966583 2023-01-23 10:11:17.732612: step: 2634/1541, loss: 0.2757026255130768 2023-01-23 10:11:18.625681: step: 2636/1541, loss: 0.25816822052001953 2023-01-23 10:11:19.481462: step: 2638/1541, loss: 0.15194369852542877 2023-01-23 10:11:20.345191: step: 2640/1541, loss: 0.30556079745292664 2023-01-23 10:11:21.213888: step: 2642/1541, loss: 0.10092490911483765 2023-01-23 10:11:22.149975: step: 2644/1541, loss: 0.14507393538951874 2023-01-23 10:11:23.157669: step: 2646/1541, loss: 0.06833802908658981 2023-01-23 10:11:24.028610: step: 2648/1541, loss: 0.15820087492465973 2023-01-23 10:11:24.904157: step: 2650/1541, loss: 0.15197718143463135 2023-01-23 10:11:25.793435: step: 2652/1541, loss: 0.17880378663539886 2023-01-23 10:11:26.673183: step: 2654/1541, loss: 0.2499532401561737 2023-01-23 10:11:27.520162: step: 2656/1541, loss: 0.12246238440275192 2023-01-23 10:11:28.442211: step: 2658/1541, loss: 0.09106767177581787 2023-01-23 10:11:29.381207: step: 2660/1541, loss: 0.2738465368747711 2023-01-23 10:11:30.240969: step: 2662/1541, loss: 0.05027460306882858 2023-01-23 10:11:31.131184: step: 2664/1541, loss: 0.16698689758777618 2023-01-23 10:11:32.038698: step: 2666/1541, loss: 0.019778408110141754 2023-01-23 10:11:32.927102: step: 2668/1541, loss: 0.3657877445220947 2023-01-23 10:11:33.814948: step: 2670/1541, loss: 0.22608648240566254 2023-01-23 10:11:34.746553: step: 2672/1541, loss: 0.45893317461013794 2023-01-23 10:11:35.630723: step: 2674/1541, loss: 0.049723561853170395 2023-01-23 10:11:36.520529: step: 2676/1541, loss: 0.18595759570598602 2023-01-23 10:11:37.413895: step: 2678/1541, loss: 0.187515988945961 2023-01-23 10:11:38.261637: step: 2680/1541, loss: 0.2432844638824463 2023-01-23 10:11:39.146484: step: 2682/1541, loss: 0.20043393969535828 2023-01-23 10:11:40.018928: step: 2684/1541, loss: 0.22761142253875732 2023-01-23 10:11:40.869505: step: 2686/1541, loss: 0.0723414421081543 2023-01-23 10:11:41.787591: step: 2688/1541, loss: 0.4201914668083191 2023-01-23 10:11:42.738133: step: 2690/1541, loss: 0.15368960797786713 2023-01-23 10:11:43.604729: step: 2692/1541, loss: 0.6344630122184753 2023-01-23 10:11:44.442323: step: 2694/1541, loss: 0.05673738196492195 2023-01-23 10:11:45.335040: step: 2696/1541, loss: 0.028520718216896057 2023-01-23 10:11:46.339670: step: 2698/1541, loss: 0.11698056012392044 2023-01-23 10:11:47.262314: step: 2700/1541, loss: 0.5754138827323914 2023-01-23 10:11:48.152834: step: 2702/1541, loss: 3.2237634658813477 2023-01-23 10:11:49.042310: step: 2704/1541, loss: 0.09284034371376038 2023-01-23 10:11:49.920423: step: 2706/1541, loss: 0.04681868478655815 2023-01-23 10:11:50.788174: step: 2708/1541, loss: 0.10300564020872116 2023-01-23 10:11:51.650416: step: 2710/1541, loss: 0.014402221888303757 2023-01-23 10:11:52.522882: step: 2712/1541, loss: 0.12092840671539307 2023-01-23 10:11:53.421575: step: 2714/1541, loss: 0.04785105958580971 2023-01-23 10:11:54.292493: step: 2716/1541, loss: 0.05919170379638672 2023-01-23 10:11:55.257338: step: 2718/1541, loss: 0.11875342577695847 2023-01-23 10:11:56.140305: step: 2720/1541, loss: 0.24608387053012848 2023-01-23 10:11:57.033088: step: 2722/1541, loss: 0.1788409799337387 2023-01-23 10:11:57.917794: step: 2724/1541, loss: 0.2532130181789398 2023-01-23 10:11:58.838012: step: 2726/1541, loss: 0.14607399702072144 2023-01-23 10:11:59.680111: step: 2728/1541, loss: 0.1634272038936615 2023-01-23 10:12:00.566985: step: 2730/1541, loss: 0.03811071813106537 2023-01-23 10:12:01.439563: step: 2732/1541, loss: 0.10962023586034775 2023-01-23 10:12:02.363707: step: 2734/1541, loss: 0.6611946225166321 2023-01-23 10:12:03.269132: step: 2736/1541, loss: 0.3024672567844391 2023-01-23 10:12:04.147582: step: 2738/1541, loss: 0.15164968371391296 2023-01-23 10:12:05.041966: step: 2740/1541, loss: 1.01651132106781 2023-01-23 10:12:05.931112: step: 2742/1541, loss: 0.0688333511352539 2023-01-23 10:12:06.821942: step: 2744/1541, loss: 0.04520082101225853 2023-01-23 10:12:07.716807: step: 2746/1541, loss: 0.4028468728065491 2023-01-23 10:12:08.599791: step: 2748/1541, loss: 0.01488400250673294 2023-01-23 10:12:09.432101: step: 2750/1541, loss: 0.40952610969543457 2023-01-23 10:12:10.283278: step: 2752/1541, loss: 0.13124531507492065 2023-01-23 10:12:11.118969: step: 2754/1541, loss: 0.07360237836837769 2023-01-23 10:12:12.003237: step: 2756/1541, loss: 0.1945556253194809 2023-01-23 10:12:12.911685: step: 2758/1541, loss: 0.06383652985095978 2023-01-23 10:12:13.819555: step: 2760/1541, loss: 0.11079544574022293 2023-01-23 10:12:14.666864: step: 2762/1541, loss: 0.9016307592391968 2023-01-23 10:12:15.528647: step: 2764/1541, loss: 0.10533106327056885 2023-01-23 10:12:16.412556: step: 2766/1541, loss: 0.23001942038536072 2023-01-23 10:12:17.394438: step: 2768/1541, loss: 0.2947444021701813 2023-01-23 10:12:18.328943: step: 2770/1541, loss: 0.30245575308799744 2023-01-23 10:12:19.246653: step: 2772/1541, loss: 2.33168888092041 2023-01-23 10:12:20.173187: step: 2774/1541, loss: 0.30684083700180054 2023-01-23 10:12:21.059000: step: 2776/1541, loss: 0.15929776430130005 2023-01-23 10:12:21.950100: step: 2778/1541, loss: 0.13765409588813782 2023-01-23 10:12:22.846114: step: 2780/1541, loss: 0.2557592988014221 2023-01-23 10:12:23.734561: step: 2782/1541, loss: 0.22365419566631317 2023-01-23 10:12:24.641494: step: 2784/1541, loss: 0.1391599476337433 2023-01-23 10:12:25.512550: step: 2786/1541, loss: 0.055613692849874496 2023-01-23 10:12:26.450574: step: 2788/1541, loss: 0.2776593267917633 2023-01-23 10:12:27.326608: step: 2790/1541, loss: 1.2828950881958008 2023-01-23 10:12:28.194214: step: 2792/1541, loss: 0.2064485102891922 2023-01-23 10:12:29.052707: step: 2794/1541, loss: 0.1529894322156906 2023-01-23 10:12:29.929453: step: 2796/1541, loss: 2.5076777935028076 2023-01-23 10:12:30.844866: step: 2798/1541, loss: 0.17175152897834778 2023-01-23 10:12:31.729226: step: 2800/1541, loss: 0.147360160946846 2023-01-23 10:12:32.665940: step: 2802/1541, loss: 0.03647865355014801 2023-01-23 10:12:33.545472: step: 2804/1541, loss: 0.15975326299667358 2023-01-23 10:12:34.417534: step: 2806/1541, loss: 0.05710289627313614 2023-01-23 10:12:35.275386: step: 2808/1541, loss: 0.1312018483877182 2023-01-23 10:12:36.163411: step: 2810/1541, loss: 0.17349053919315338 2023-01-23 10:12:37.086229: step: 2812/1541, loss: 0.16597434878349304 2023-01-23 10:12:37.990937: step: 2814/1541, loss: 0.2706535756587982 2023-01-23 10:12:38.880977: step: 2816/1541, loss: 0.08533205837011337 2023-01-23 10:12:39.719841: step: 2818/1541, loss: 0.04902205616235733 2023-01-23 10:12:40.578114: step: 2820/1541, loss: 0.0929575115442276 2023-01-23 10:12:41.449529: step: 2822/1541, loss: 0.27915430068969727 2023-01-23 10:12:42.414635: step: 2824/1541, loss: 0.2617337107658386 2023-01-23 10:12:43.280938: step: 2826/1541, loss: 0.1939702183008194 2023-01-23 10:12:44.171714: step: 2828/1541, loss: 0.4070754945278168 2023-01-23 10:12:45.062204: step: 2830/1541, loss: 0.32368388772010803 2023-01-23 10:12:45.952485: step: 2832/1541, loss: 0.33675846457481384 2023-01-23 10:12:46.830707: step: 2834/1541, loss: 0.13385528326034546 2023-01-23 10:12:47.774058: step: 2836/1541, loss: 0.31600216031074524 2023-01-23 10:12:48.662749: step: 2838/1541, loss: 0.1519572138786316 2023-01-23 10:12:49.553722: step: 2840/1541, loss: 0.34179505705833435 2023-01-23 10:12:50.433008: step: 2842/1541, loss: 0.757990837097168 2023-01-23 10:12:51.329793: step: 2844/1541, loss: 0.43478676676750183 2023-01-23 10:12:52.251195: step: 2846/1541, loss: 0.07573794573545456 2023-01-23 10:12:53.137373: step: 2848/1541, loss: 0.3434188663959503 2023-01-23 10:12:54.073735: step: 2850/1541, loss: 0.25663116574287415 2023-01-23 10:12:54.947030: step: 2852/1541, loss: 0.11542245000600815 2023-01-23 10:12:55.907275: step: 2854/1541, loss: 0.35803651809692383 2023-01-23 10:12:56.862890: step: 2856/1541, loss: 0.11938852816820145 2023-01-23 10:12:57.757247: step: 2858/1541, loss: 0.18752816319465637 2023-01-23 10:12:58.659644: step: 2860/1541, loss: 0.30951783061027527 2023-01-23 10:12:59.594829: step: 2862/1541, loss: 0.058189306408166885 2023-01-23 10:13:00.458837: step: 2864/1541, loss: 0.2712905704975128 2023-01-23 10:13:01.305870: step: 2866/1541, loss: 0.044444359838962555 2023-01-23 10:13:02.234552: step: 2868/1541, loss: 0.07815201580524445 2023-01-23 10:13:03.143480: step: 2870/1541, loss: 0.1828152984380722 2023-01-23 10:13:04.045190: step: 2872/1541, loss: 0.049370620399713516 2023-01-23 10:13:04.926539: step: 2874/1541, loss: 0.2206227034330368 2023-01-23 10:13:05.814212: step: 2876/1541, loss: 0.19595089554786682 2023-01-23 10:13:06.708145: step: 2878/1541, loss: 1.0367509126663208 2023-01-23 10:13:07.570571: step: 2880/1541, loss: 0.09095415472984314 2023-01-23 10:13:08.426530: step: 2882/1541, loss: 0.24308548867702484 2023-01-23 10:13:09.334685: step: 2884/1541, loss: 0.041111718863248825 2023-01-23 10:13:10.298145: step: 2886/1541, loss: 0.019229918718338013 2023-01-23 10:13:11.184997: step: 2888/1541, loss: 0.18850862979888916 2023-01-23 10:13:12.077971: step: 2890/1541, loss: 0.07314188778400421 2023-01-23 10:13:12.949304: step: 2892/1541, loss: 0.12603704631328583 2023-01-23 10:13:13.836285: step: 2894/1541, loss: 0.26597750186920166 2023-01-23 10:13:14.748996: step: 2896/1541, loss: 0.25079742074012756 2023-01-23 10:13:15.676643: step: 2898/1541, loss: 0.16025523841381073 2023-01-23 10:13:16.564576: step: 2900/1541, loss: 0.08959300816059113 2023-01-23 10:13:17.479854: step: 2902/1541, loss: 0.04964866489171982 2023-01-23 10:13:18.496500: step: 2904/1541, loss: 0.17087094485759735 2023-01-23 10:13:19.383700: step: 2906/1541, loss: 0.26769885420799255 2023-01-23 10:13:20.224955: step: 2908/1541, loss: 0.0386015810072422 2023-01-23 10:13:21.098558: step: 2910/1541, loss: 0.16651543974876404 2023-01-23 10:13:21.953058: step: 2912/1541, loss: 0.04583032801747322 2023-01-23 10:13:22.846884: step: 2914/1541, loss: 0.4284862279891968 2023-01-23 10:13:23.765265: step: 2916/1541, loss: 0.06200511381030083 2023-01-23 10:13:24.655024: step: 2918/1541, loss: 0.39881932735443115 2023-01-23 10:13:25.503536: step: 2920/1541, loss: 0.27263566851615906 2023-01-23 10:13:26.380305: step: 2922/1541, loss: 0.3502596318721771 2023-01-23 10:13:27.269669: step: 2924/1541, loss: 0.6366008520126343 2023-01-23 10:13:28.181361: step: 2926/1541, loss: 0.13166089355945587 2023-01-23 10:13:29.071039: step: 2928/1541, loss: 0.17210625112056732 2023-01-23 10:13:29.955510: step: 2930/1541, loss: 0.05073566734790802 2023-01-23 10:13:30.852385: step: 2932/1541, loss: 0.09880444407463074 2023-01-23 10:13:31.723230: step: 2934/1541, loss: 0.11087120324373245 2023-01-23 10:13:32.602993: step: 2936/1541, loss: 0.524456262588501 2023-01-23 10:13:33.481362: step: 2938/1541, loss: 0.15489304065704346 2023-01-23 10:13:34.364139: step: 2940/1541, loss: 0.046448417007923126 2023-01-23 10:13:35.316859: step: 2942/1541, loss: 0.22389496862888336 2023-01-23 10:13:36.205283: step: 2944/1541, loss: 0.20583206415176392 2023-01-23 10:13:37.098860: step: 2946/1541, loss: 0.41971519589424133 2023-01-23 10:13:37.983186: step: 2948/1541, loss: 0.051076650619506836 2023-01-23 10:13:38.844837: step: 2950/1541, loss: 0.3170468807220459 2023-01-23 10:13:39.743114: step: 2952/1541, loss: 0.23957663774490356 2023-01-23 10:13:40.589601: step: 2954/1541, loss: 0.19465389847755432 2023-01-23 10:13:41.477289: step: 2956/1541, loss: 0.13842763006687164 2023-01-23 10:13:42.349274: step: 2958/1541, loss: 0.1171308308839798 2023-01-23 10:13:43.210634: step: 2960/1541, loss: 0.02415550872683525 2023-01-23 10:13:44.136247: step: 2962/1541, loss: 0.2577795088291168 2023-01-23 10:13:45.069246: step: 2964/1541, loss: 0.030495256185531616 2023-01-23 10:13:45.932797: step: 2966/1541, loss: 1.6318750381469727 2023-01-23 10:13:46.796936: step: 2968/1541, loss: 0.1225440576672554 2023-01-23 10:13:47.667380: step: 2970/1541, loss: 0.21410121023654938 2023-01-23 10:13:48.541057: step: 2972/1541, loss: 0.20584534108638763 2023-01-23 10:13:49.419803: step: 2974/1541, loss: 0.11787623912096024 2023-01-23 10:13:50.337084: step: 2976/1541, loss: 0.0871189758181572 2023-01-23 10:13:51.288751: step: 2978/1541, loss: 0.13777640461921692 2023-01-23 10:13:52.196407: step: 2980/1541, loss: 0.16869598627090454 2023-01-23 10:13:53.060919: step: 2982/1541, loss: 0.013870786875486374 2023-01-23 10:13:53.949837: step: 2984/1541, loss: 0.14982348680496216 2023-01-23 10:13:54.823829: step: 2986/1541, loss: 0.2793152630329132 2023-01-23 10:13:55.701923: step: 2988/1541, loss: 0.012298695743083954 2023-01-23 10:13:56.583115: step: 2990/1541, loss: 0.3036366105079651 2023-01-23 10:13:57.449806: step: 2992/1541, loss: 0.11173325031995773 2023-01-23 10:13:58.348350: step: 2994/1541, loss: 0.1957949995994568 2023-01-23 10:13:59.237714: step: 2996/1541, loss: 0.5591359734535217 2023-01-23 10:14:00.092648: step: 2998/1541, loss: 0.09066522121429443 2023-01-23 10:14:01.066365: step: 3000/1541, loss: 0.08310747146606445 2023-01-23 10:14:01.980179: step: 3002/1541, loss: 0.04298734292387962 2023-01-23 10:14:02.825082: step: 3004/1541, loss: 0.10897248983383179 2023-01-23 10:14:03.703226: step: 3006/1541, loss: 0.3953155279159546 2023-01-23 10:14:04.600029: step: 3008/1541, loss: 0.020053070038557053 2023-01-23 10:14:05.484465: step: 3010/1541, loss: 0.1030920073390007 2023-01-23 10:14:06.421253: step: 3012/1541, loss: 1.0568549633026123 2023-01-23 10:14:07.361558: step: 3014/1541, loss: 5.481478214263916 2023-01-23 10:14:08.230196: step: 3016/1541, loss: 0.0358293317258358 2023-01-23 10:14:09.115528: step: 3018/1541, loss: 0.37643754482269287 2023-01-23 10:14:09.993523: step: 3020/1541, loss: 0.019495612010359764 2023-01-23 10:14:10.858949: step: 3022/1541, loss: 0.16085925698280334 2023-01-23 10:14:11.745128: step: 3024/1541, loss: 0.42101800441741943 2023-01-23 10:14:12.649057: step: 3026/1541, loss: 0.18346905708312988 2023-01-23 10:14:13.522905: step: 3028/1541, loss: 1.804640293121338 2023-01-23 10:14:14.436407: step: 3030/1541, loss: 0.07244745641946793 2023-01-23 10:14:15.355566: step: 3032/1541, loss: 0.008720837533473969 2023-01-23 10:14:16.258281: step: 3034/1541, loss: 0.08329654484987259 2023-01-23 10:14:17.132465: step: 3036/1541, loss: 0.032096028327941895 2023-01-23 10:14:18.056612: step: 3038/1541, loss: 0.2052181363105774 2023-01-23 10:14:18.961970: step: 3040/1541, loss: 0.47688940167427063 2023-01-23 10:14:19.839444: step: 3042/1541, loss: 0.14262834191322327 2023-01-23 10:14:20.732880: step: 3044/1541, loss: 0.13714101910591125 2023-01-23 10:14:21.583278: step: 3046/1541, loss: 0.1893400102853775 2023-01-23 10:14:22.449719: step: 3048/1541, loss: 0.1370219886302948 2023-01-23 10:14:23.373121: step: 3050/1541, loss: 0.22943757474422455 2023-01-23 10:14:24.315589: step: 3052/1541, loss: 0.31404170393943787 2023-01-23 10:14:25.242779: step: 3054/1541, loss: 0.37609899044036865 2023-01-23 10:14:26.202096: step: 3056/1541, loss: 0.31237903237342834 2023-01-23 10:14:27.055758: step: 3058/1541, loss: 0.337245911359787 2023-01-23 10:14:27.979056: step: 3060/1541, loss: 0.24467208981513977 2023-01-23 10:14:28.945012: step: 3062/1541, loss: 0.5956370234489441 2023-01-23 10:14:29.842271: step: 3064/1541, loss: 0.09002216905355453 2023-01-23 10:14:30.741053: step: 3066/1541, loss: 0.11300903558731079 2023-01-23 10:14:31.675205: step: 3068/1541, loss: 0.07475613802671432 2023-01-23 10:14:32.592759: step: 3070/1541, loss: 0.5185861587524414 2023-01-23 10:14:33.470108: step: 3072/1541, loss: 0.4605749845504761 2023-01-23 10:14:34.428128: step: 3074/1541, loss: 0.12691351771354675 2023-01-23 10:14:35.400816: step: 3076/1541, loss: 0.28025832772254944 2023-01-23 10:14:36.239733: step: 3078/1541, loss: 0.3441110849380493 2023-01-23 10:14:37.121759: step: 3080/1541, loss: 0.5124251842498779 2023-01-23 10:14:38.055970: step: 3082/1541, loss: 0.04003768041729927 2023-01-23 10:14:38.674294: step: 3084/1541, loss: 0.014107637107372284 ================================================== Loss: 0.269 -------------------- Dev Chinese: {'event': {'p': 0.6164874551971327, 'r': 0.6870838881491345, 'f1': 0.649874055415617}, 'argument': {'p': 0.67583044823236, 'r': 0.6438973286799403, 'f1': 0.6594775496436357}, 'combined': 0.42857734964246347, 'epoch': 10} Test Chinese: {'event': {'p': 0.6093585699263933, 'r': 0.6944278010784901, 'f1': 0.6491178941472977}, 'argument': {'p': 0.7219347095899677, 'r': 0.7053898253760107, 'f1': 0.7135663771696976}, 'combined': 0.4631887040827105, 'epoch': 10} Dev Korean: {'event': {'p': 0.5696078431372549, 'r': 0.7736351531291611, 'f1': 0.6561264822134386}, 'argument': {'p': 0.6666608337589293, 'r': 0.6542356139182611, 'f1': 0.660389783947883}, 'combined': 0.43329922583141717, 'epoch': 10} Test Korean: {'event': {'p': 0.6096282173498571, 'r': 0.7663271420011983, 'f1': 0.6790549508893019}, 'argument': {'p': 0.7181686019868226, 'r': 0.7247272193565654, 'f1': 0.7214330047231264}, 'combined': 0.48989265359218415, 'epoch': 10} Dev Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6727483627478649, 'r': 0.6595770845615362, 'f1': 0.6660976184773536}, 'combined': 0.4593776679154163, 'epoch': 10} Test Russian: {'event': {'p': 0.6446886446886447, 'r': 0.738166566806471, 'f1': 0.6882681564245811}, 'argument': {'p': 0.7212116816452776, 'r': 0.7139607203935477, 'f1': 0.717567883903295}, 'combined': 0.4938791245636086, 'epoch': 10} Chinese: {'event': {'p': 0.6507936507936508, 'r': 0.7592592592592593, 'f1': 0.7008547008547008}, 'argument': {'p': 0.6484713203463203, 'r': 0.6427326360954679, 'f1': 0.6455892255892255}, 'combined': 0.4524642435753546, 'epoch': 10} Korean: {'event': {'p': 0.5625, 'r': 0.5714285714285714, 'f1': 0.5669291338582677}, 'argument': {'p': 0.6136009959539371, 'r': 0.40641104926819216, 'f1': 0.48896329365079366}, 'combined': 0.27720753655793023, 'epoch': 10} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5730551257490913, 'r': 0.6396894426966601, 'f1': 0.6045416711199205}, 'combined': 0.3109071451473877, 'epoch': 10} New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test for Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6727483627478649, 'r': 0.6595770845615362, 'f1': 0.6660976184773536}, 'combined': 0.4593776679154163, 'epoch': 10} Test for Russian: {'event': {'p': 0.6446886446886447, 'r': 0.738166566806471, 'f1': 0.6882681564245811}, 'argument': {'p': 0.7212116816452776, 'r': 0.7139607203935477, 'f1': 0.717567883903295}, 'combined': 0.4938791245636086, 'epoch': 10} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5730551257490913, 'r': 0.6396894426966601, 'f1': 0.6045416711199205}, 'combined': 0.3109071451473877, 'epoch': 10} ****************************** Epoch: 11 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 10:20:02.080015: step: 2/1541, loss: 0.09017379581928253 2023-01-23 10:20:03.011710: step: 4/1541, loss: 0.16566036641597748 2023-01-23 10:20:03.856528: step: 6/1541, loss: 0.4665556252002716 2023-01-23 10:20:04.750849: step: 8/1541, loss: 0.5072411298751831 2023-01-23 10:20:05.679484: step: 10/1541, loss: 0.06685210019350052 2023-01-23 10:20:06.549814: step: 12/1541, loss: 0.3711550831794739 2023-01-23 10:20:07.402901: step: 14/1541, loss: 0.3231619596481323 2023-01-23 10:20:08.341622: step: 16/1541, loss: 0.10419061034917831 2023-01-23 10:20:09.235996: step: 18/1541, loss: 0.20034252107143402 2023-01-23 10:20:10.100963: step: 20/1541, loss: 0.27352508902549744 2023-01-23 10:20:10.961724: step: 22/1541, loss: 0.056508876383304596 2023-01-23 10:20:11.822635: step: 24/1541, loss: 0.18903613090515137 2023-01-23 10:20:12.714452: step: 26/1541, loss: 0.16636322438716888 2023-01-23 10:20:13.596404: step: 28/1541, loss: 0.02408752404153347 2023-01-23 10:20:14.484162: step: 30/1541, loss: 0.0891093835234642 2023-01-23 10:20:15.367921: step: 32/1541, loss: 0.11893526464700699 2023-01-23 10:20:16.236930: step: 34/1541, loss: 0.07821672409772873 2023-01-23 10:20:17.126815: step: 36/1541, loss: 0.05724178999662399 2023-01-23 10:20:17.977775: step: 38/1541, loss: 0.07474569976329803 2023-01-23 10:20:18.849942: step: 40/1541, loss: 1.4973735809326172 2023-01-23 10:20:19.827642: step: 42/1541, loss: 0.7270180583000183 2023-01-23 10:20:20.727202: step: 44/1541, loss: 0.07470707595348358 2023-01-23 10:20:21.580924: step: 46/1541, loss: 0.083637535572052 2023-01-23 10:20:22.470578: step: 48/1541, loss: 0.2564467787742615 2023-01-23 10:20:23.331350: step: 50/1541, loss: 0.08169690519571304 2023-01-23 10:20:24.187242: step: 52/1541, loss: 0.10434196144342422 2023-01-23 10:20:25.091461: step: 54/1541, loss: 0.05369015410542488 2023-01-23 10:20:25.981968: step: 56/1541, loss: 0.05957628786563873 2023-01-23 10:20:26.896022: step: 58/1541, loss: 0.09483490884304047 2023-01-23 10:20:27.781864: step: 60/1541, loss: 0.015662042424082756 2023-01-23 10:20:28.716659: step: 62/1541, loss: 0.0935637429356575 2023-01-23 10:20:29.660871: step: 64/1541, loss: 0.07957729697227478 2023-01-23 10:20:30.568912: step: 66/1541, loss: 0.1497562825679779 2023-01-23 10:20:31.433844: step: 68/1541, loss: 0.20411086082458496 2023-01-23 10:20:32.307692: step: 70/1541, loss: 0.1900860220193863 2023-01-23 10:20:33.213563: step: 72/1541, loss: 0.3306039273738861 2023-01-23 10:20:34.114138: step: 74/1541, loss: 0.03612266108393669 2023-01-23 10:20:34.972075: step: 76/1541, loss: 0.1092112585902214 2023-01-23 10:20:35.886134: step: 78/1541, loss: 0.34030675888061523 2023-01-23 10:20:36.795808: step: 80/1541, loss: 0.15890608727931976 2023-01-23 10:20:37.673119: step: 82/1541, loss: 0.4668385982513428 2023-01-23 10:20:38.584821: step: 84/1541, loss: 0.006542955059558153 2023-01-23 10:20:39.443370: step: 86/1541, loss: 0.20529627799987793 2023-01-23 10:20:40.304672: step: 88/1541, loss: 0.517262876033783 2023-01-23 10:20:41.158980: step: 90/1541, loss: 0.07228750735521317 2023-01-23 10:20:42.062493: step: 92/1541, loss: 0.14008504152297974 2023-01-23 10:20:42.932004: step: 94/1541, loss: 0.05680648609995842 2023-01-23 10:20:43.822147: step: 96/1541, loss: 0.4944915771484375 2023-01-23 10:20:44.758359: step: 98/1541, loss: 0.1947016716003418 2023-01-23 10:20:45.666082: step: 100/1541, loss: 0.19945208728313446 2023-01-23 10:20:46.503346: step: 102/1541, loss: 0.18594829738140106 2023-01-23 10:20:47.366583: step: 104/1541, loss: 0.1264880746603012 2023-01-23 10:20:48.275444: step: 106/1541, loss: 0.03356241062283516 2023-01-23 10:20:49.128172: step: 108/1541, loss: 0.2081526815891266 2023-01-23 10:20:50.006280: step: 110/1541, loss: 0.009904208593070507 2023-01-23 10:20:50.879662: step: 112/1541, loss: 0.18303027749061584 2023-01-23 10:20:51.751212: step: 114/1541, loss: 0.00849162507802248 2023-01-23 10:20:52.662597: step: 116/1541, loss: 0.10726340115070343 2023-01-23 10:20:53.523669: step: 118/1541, loss: 0.366515576839447 2023-01-23 10:20:54.439527: step: 120/1541, loss: 0.3559934198856354 2023-01-23 10:20:55.341186: step: 122/1541, loss: 0.08227992057800293 2023-01-23 10:20:56.198533: step: 124/1541, loss: 0.07328810542821884 2023-01-23 10:20:57.060695: step: 126/1541, loss: 0.021189378574490547 2023-01-23 10:20:57.906592: step: 128/1541, loss: 0.25029581785202026 2023-01-23 10:20:58.807687: step: 130/1541, loss: 0.012509848922491074 2023-01-23 10:20:59.740225: step: 132/1541, loss: 0.2899138927459717 2023-01-23 10:21:00.634159: step: 134/1541, loss: 0.07041388005018234 2023-01-23 10:21:01.507565: step: 136/1541, loss: 0.0550868883728981 2023-01-23 10:21:02.423573: step: 138/1541, loss: 2.654728889465332 2023-01-23 10:21:03.288919: step: 140/1541, loss: 0.10829532891511917 2023-01-23 10:21:04.144966: step: 142/1541, loss: 0.014940816909074783 2023-01-23 10:21:05.022692: step: 144/1541, loss: 0.5395292639732361 2023-01-23 10:21:05.916169: step: 146/1541, loss: 0.05339612066745758 2023-01-23 10:21:06.776143: step: 148/1541, loss: 0.11646173894405365 2023-01-23 10:21:07.648704: step: 150/1541, loss: 0.2797074019908905 2023-01-23 10:21:08.559031: step: 152/1541, loss: 0.3384101986885071 2023-01-23 10:21:09.445018: step: 154/1541, loss: 0.067267507314682 2023-01-23 10:21:10.287979: step: 156/1541, loss: 0.2842180132865906 2023-01-23 10:21:11.161217: step: 158/1541, loss: 0.03052583336830139 2023-01-23 10:21:12.067399: step: 160/1541, loss: 0.04447431489825249 2023-01-23 10:21:12.929118: step: 162/1541, loss: 0.013401889242231846 2023-01-23 10:21:13.880391: step: 164/1541, loss: 0.13531340658664703 2023-01-23 10:21:14.763560: step: 166/1541, loss: 0.11942316591739655 2023-01-23 10:21:15.673500: step: 168/1541, loss: 0.2864352762699127 2023-01-23 10:21:16.557443: step: 170/1541, loss: 0.08914726227521896 2023-01-23 10:21:17.421655: step: 172/1541, loss: 0.12752074003219604 2023-01-23 10:21:18.288413: step: 174/1541, loss: 0.10495378077030182 2023-01-23 10:21:19.172609: step: 176/1541, loss: 0.23034563660621643 2023-01-23 10:21:20.035361: step: 178/1541, loss: 0.10189700126647949 2023-01-23 10:21:20.912162: step: 180/1541, loss: 0.09591629356145859 2023-01-23 10:21:21.837465: step: 182/1541, loss: 0.16725455224514008 2023-01-23 10:21:22.786587: step: 184/1541, loss: 0.04669857770204544 2023-01-23 10:21:23.675202: step: 186/1541, loss: 0.0715089738368988 2023-01-23 10:21:24.594577: step: 188/1541, loss: 0.6129305958747864 2023-01-23 10:21:25.502222: step: 190/1541, loss: 0.13866594433784485 2023-01-23 10:21:26.395799: step: 192/1541, loss: 0.34360939264297485 2023-01-23 10:21:27.275887: step: 194/1541, loss: 0.24699203670024872 2023-01-23 10:21:28.134275: step: 196/1541, loss: 0.20143379271030426 2023-01-23 10:21:29.050200: step: 198/1541, loss: 0.1673406958580017 2023-01-23 10:21:29.950517: step: 200/1541, loss: 0.2056070864200592 2023-01-23 10:21:30.887777: step: 202/1541, loss: 0.02207542210817337 2023-01-23 10:21:31.768861: step: 204/1541, loss: 0.14022956788539886 2023-01-23 10:21:32.672944: step: 206/1541, loss: 0.13748978078365326 2023-01-23 10:21:33.564443: step: 208/1541, loss: 0.15412046015262604 2023-01-23 10:21:34.427869: step: 210/1541, loss: 0.035491473972797394 2023-01-23 10:21:35.299020: step: 212/1541, loss: 0.07748469710350037 2023-01-23 10:21:36.201057: step: 214/1541, loss: 0.22419698536396027 2023-01-23 10:21:37.097491: step: 216/1541, loss: 0.07773473858833313 2023-01-23 10:21:38.041154: step: 218/1541, loss: 0.10911214351654053 2023-01-23 10:21:38.910768: step: 220/1541, loss: 0.3275148868560791 2023-01-23 10:21:39.813978: step: 222/1541, loss: 0.055319830775260925 2023-01-23 10:21:40.720903: step: 224/1541, loss: 0.026461299508810043 2023-01-23 10:21:41.604141: step: 226/1541, loss: 0.08316484838724136 2023-01-23 10:21:42.487284: step: 228/1541, loss: 0.056855395436286926 2023-01-23 10:21:43.371451: step: 230/1541, loss: 0.06482919305562973 2023-01-23 10:21:44.373903: step: 232/1541, loss: 0.1623365581035614 2023-01-23 10:21:45.295384: step: 234/1541, loss: 0.03998614847660065 2023-01-23 10:21:46.171408: step: 236/1541, loss: 0.41513678431510925 2023-01-23 10:21:47.069770: step: 238/1541, loss: 0.09208563715219498 2023-01-23 10:21:48.013932: step: 240/1541, loss: 0.0984877422451973 2023-01-23 10:21:48.903776: step: 242/1541, loss: 0.14512771368026733 2023-01-23 10:21:49.769086: step: 244/1541, loss: 2.0771687030792236 2023-01-23 10:21:50.659798: step: 246/1541, loss: 0.01338171400129795 2023-01-23 10:21:51.549808: step: 248/1541, loss: 0.48197171092033386 2023-01-23 10:21:52.451526: step: 250/1541, loss: 0.2782182991504669 2023-01-23 10:21:53.368874: step: 252/1541, loss: 0.06959912925958633 2023-01-23 10:21:54.280909: step: 254/1541, loss: 0.3423698842525482 2023-01-23 10:21:55.158451: step: 256/1541, loss: 0.059149861335754395 2023-01-23 10:21:56.032504: step: 258/1541, loss: 0.19070661067962646 2023-01-23 10:21:56.921190: step: 260/1541, loss: 0.00434080371633172 2023-01-23 10:21:57.892549: step: 262/1541, loss: 0.25336962938308716 2023-01-23 10:21:58.787312: step: 264/1541, loss: 0.054396454244852066 2023-01-23 10:21:59.675318: step: 266/1541, loss: 0.18086674809455872 2023-01-23 10:22:00.586409: step: 268/1541, loss: 0.09757653623819351 2023-01-23 10:22:01.525783: step: 270/1541, loss: 0.018267164006829262 2023-01-23 10:22:02.488122: step: 272/1541, loss: 0.07114218175411224 2023-01-23 10:22:03.463075: step: 274/1541, loss: 0.6212149858474731 2023-01-23 10:22:04.347580: step: 276/1541, loss: 0.16563035547733307 2023-01-23 10:22:05.234596: step: 278/1541, loss: 0.17793576419353485 2023-01-23 10:22:06.088876: step: 280/1541, loss: 0.03896021470427513 2023-01-23 10:22:07.042324: step: 282/1541, loss: 0.2779723107814789 2023-01-23 10:22:07.955209: step: 284/1541, loss: 0.22520719468593597 2023-01-23 10:22:08.855047: step: 286/1541, loss: 0.0921214148402214 2023-01-23 10:22:09.728236: step: 288/1541, loss: 0.5884298086166382 2023-01-23 10:22:10.599715: step: 290/1541, loss: 0.25209441781044006 2023-01-23 10:22:11.488020: step: 292/1541, loss: 0.041845474392175674 2023-01-23 10:22:12.353403: step: 294/1541, loss: 0.04027076065540314 2023-01-23 10:22:13.230054: step: 296/1541, loss: 0.01588246412575245 2023-01-23 10:22:14.105109: step: 298/1541, loss: 0.04305795952677727 2023-01-23 10:22:15.038430: step: 300/1541, loss: 0.03752481937408447 2023-01-23 10:22:15.923603: step: 302/1541, loss: 0.2880004644393921 2023-01-23 10:22:16.811843: step: 304/1541, loss: 0.5576452612876892 2023-01-23 10:22:17.709910: step: 306/1541, loss: 0.15818838775157928 2023-01-23 10:22:18.574989: step: 308/1541, loss: 0.27816322445869446 2023-01-23 10:22:19.496448: step: 310/1541, loss: 0.22717927396297455 2023-01-23 10:22:20.396863: step: 312/1541, loss: 0.1846771091222763 2023-01-23 10:22:21.291622: step: 314/1541, loss: 0.15033523738384247 2023-01-23 10:22:22.230426: step: 316/1541, loss: 0.04831845313310623 2023-01-23 10:22:23.075675: step: 318/1541, loss: 0.12314840406179428 2023-01-23 10:22:23.940493: step: 320/1541, loss: 0.0603628046810627 2023-01-23 10:22:24.792942: step: 322/1541, loss: 0.10752899944782257 2023-01-23 10:22:25.659734: step: 324/1541, loss: 0.011997542344033718 2023-01-23 10:22:26.577916: step: 326/1541, loss: 0.09226585924625397 2023-01-23 10:22:27.451083: step: 328/1541, loss: 0.034191474318504333 2023-01-23 10:22:28.326429: step: 330/1541, loss: 0.00523644732311368 2023-01-23 10:22:29.219011: step: 332/1541, loss: 0.3051168918609619 2023-01-23 10:22:30.163845: step: 334/1541, loss: 0.05641990900039673 2023-01-23 10:22:31.034187: step: 336/1541, loss: 0.017059065401554108 2023-01-23 10:22:31.944249: step: 338/1541, loss: 0.03010137379169464 2023-01-23 10:22:32.875286: step: 340/1541, loss: 0.030972931534051895 2023-01-23 10:22:33.764566: step: 342/1541, loss: 0.21785803139209747 2023-01-23 10:22:34.642723: step: 344/1541, loss: 0.14949287474155426 2023-01-23 10:22:35.466266: step: 346/1541, loss: 0.044521965086460114 2023-01-23 10:22:36.310889: step: 348/1541, loss: 0.0012455529067665339 2023-01-23 10:22:37.197277: step: 350/1541, loss: 0.003397200256586075 2023-01-23 10:22:38.074583: step: 352/1541, loss: 0.14424966275691986 2023-01-23 10:22:38.937040: step: 354/1541, loss: 0.43197333812713623 2023-01-23 10:22:39.831736: step: 356/1541, loss: 0.10246369242668152 2023-01-23 10:22:40.732835: step: 358/1541, loss: 0.02551509626209736 2023-01-23 10:22:41.702373: step: 360/1541, loss: 0.38244789838790894 2023-01-23 10:22:42.662786: step: 362/1541, loss: 0.18858623504638672 2023-01-23 10:22:43.525196: step: 364/1541, loss: 0.17382977902889252 2023-01-23 10:22:44.442111: step: 366/1541, loss: 0.0764092355966568 2023-01-23 10:22:45.323779: step: 368/1541, loss: 0.09356493502855301 2023-01-23 10:22:46.195898: step: 370/1541, loss: 0.08906075358390808 2023-01-23 10:22:47.117825: step: 372/1541, loss: 0.21185526251792908 2023-01-23 10:22:48.021021: step: 374/1541, loss: 0.022864585742354393 2023-01-23 10:22:48.905890: step: 376/1541, loss: 0.218095600605011 2023-01-23 10:22:49.784950: step: 378/1541, loss: 0.04075680673122406 2023-01-23 10:22:50.673387: step: 380/1541, loss: 0.31614452600479126 2023-01-23 10:22:51.563841: step: 382/1541, loss: 0.06891375780105591 2023-01-23 10:22:52.483968: step: 384/1541, loss: 0.14669707417488098 2023-01-23 10:22:53.313652: step: 386/1541, loss: 0.21702492237091064 2023-01-23 10:22:54.167422: step: 388/1541, loss: 0.2371894270181656 2023-01-23 10:22:55.054720: step: 390/1541, loss: 0.07791674137115479 2023-01-23 10:22:55.951564: step: 392/1541, loss: 0.05687431991100311 2023-01-23 10:22:56.817200: step: 394/1541, loss: 0.05118010565638542 2023-01-23 10:22:57.690165: step: 396/1541, loss: 0.10044101625680923 2023-01-23 10:22:58.596395: step: 398/1541, loss: 0.04857231676578522 2023-01-23 10:22:59.482397: step: 400/1541, loss: 0.051436301320791245 2023-01-23 10:23:00.377933: step: 402/1541, loss: 0.05280149728059769 2023-01-23 10:23:01.251276: step: 404/1541, loss: 0.14039160311222076 2023-01-23 10:23:02.173305: step: 406/1541, loss: 0.6874801516532898 2023-01-23 10:23:03.054344: step: 408/1541, loss: 0.060720812529325485 2023-01-23 10:23:03.946147: step: 410/1541, loss: 0.5150439143180847 2023-01-23 10:23:04.834488: step: 412/1541, loss: 0.07435344159603119 2023-01-23 10:23:05.794594: step: 414/1541, loss: 0.4745751619338989 2023-01-23 10:23:06.680198: step: 416/1541, loss: 1.0679643154144287 2023-01-23 10:23:07.527385: step: 418/1541, loss: 1.2089526653289795 2023-01-23 10:23:08.446606: step: 420/1541, loss: 0.08254075050354004 2023-01-23 10:23:09.343040: step: 422/1541, loss: 0.19299350678920746 2023-01-23 10:23:10.245253: step: 424/1541, loss: 0.019401781260967255 2023-01-23 10:23:11.115341: step: 426/1541, loss: 0.20493565499782562 2023-01-23 10:23:12.047120: step: 428/1541, loss: 0.1818379908800125 2023-01-23 10:23:12.903951: step: 430/1541, loss: 0.0771988108754158 2023-01-23 10:23:13.794868: step: 432/1541, loss: 0.23680295050144196 2023-01-23 10:23:14.712675: step: 434/1541, loss: 0.13679161667823792 2023-01-23 10:23:15.590497: step: 436/1541, loss: 0.4623459577560425 2023-01-23 10:23:16.463314: step: 438/1541, loss: 0.12457173317670822 2023-01-23 10:23:17.344077: step: 440/1541, loss: 0.1061481311917305 2023-01-23 10:23:18.250663: step: 442/1541, loss: 0.3004743754863739 2023-01-23 10:23:19.138000: step: 444/1541, loss: 0.09968983381986618 2023-01-23 10:23:20.021160: step: 446/1541, loss: 0.08238112181425095 2023-01-23 10:23:20.918371: step: 448/1541, loss: 0.2894080877304077 2023-01-23 10:23:21.809773: step: 450/1541, loss: 0.04010521620512009 2023-01-23 10:23:22.698504: step: 452/1541, loss: 0.032279230654239655 2023-01-23 10:23:23.612410: step: 454/1541, loss: 0.04548972100019455 2023-01-23 10:23:24.552723: step: 456/1541, loss: 0.15246251225471497 2023-01-23 10:23:25.448536: step: 458/1541, loss: 0.2905714511871338 2023-01-23 10:23:26.274772: step: 460/1541, loss: 0.06685551255941391 2023-01-23 10:23:27.203239: step: 462/1541, loss: 0.044038742780685425 2023-01-23 10:23:28.132471: step: 464/1541, loss: 0.31652122735977173 2023-01-23 10:23:29.044864: step: 466/1541, loss: 0.14035291969776154 2023-01-23 10:23:29.908701: step: 468/1541, loss: 0.036171164363622665 2023-01-23 10:23:30.801679: step: 470/1541, loss: 0.242538720369339 2023-01-23 10:23:31.671207: step: 472/1541, loss: 0.0631030723452568 2023-01-23 10:23:32.630681: step: 474/1541, loss: 0.10718542337417603 2023-01-23 10:23:33.509009: step: 476/1541, loss: 0.05203895643353462 2023-01-23 10:23:34.407225: step: 478/1541, loss: 0.5967968106269836 2023-01-23 10:23:35.292084: step: 480/1541, loss: 0.33845117688179016 2023-01-23 10:23:36.212231: step: 482/1541, loss: 0.7707451581954956 2023-01-23 10:23:37.068540: step: 484/1541, loss: 0.07505206763744354 2023-01-23 10:23:37.964058: step: 486/1541, loss: 0.08480788767337799 2023-01-23 10:23:38.853555: step: 488/1541, loss: 0.09754552692174911 2023-01-23 10:23:39.720142: step: 490/1541, loss: 0.1143505722284317 2023-01-23 10:23:40.587103: step: 492/1541, loss: 0.05646451190114021 2023-01-23 10:23:41.431852: step: 494/1541, loss: 0.7496365308761597 2023-01-23 10:23:42.307990: step: 496/1541, loss: 0.07376617193222046 2023-01-23 10:23:43.185068: step: 498/1541, loss: 0.03411487489938736 2023-01-23 10:23:44.085125: step: 500/1541, loss: 0.2545485198497772 2023-01-23 10:23:44.984445: step: 502/1541, loss: 0.02633332647383213 2023-01-23 10:23:45.893869: step: 504/1541, loss: 0.021680383011698723 2023-01-23 10:23:46.766015: step: 506/1541, loss: 0.327975332736969 2023-01-23 10:23:47.669604: step: 508/1541, loss: 0.16192547976970673 2023-01-23 10:23:48.559204: step: 510/1541, loss: 0.02118457481265068 2023-01-23 10:23:49.462270: step: 512/1541, loss: 0.027649033814668655 2023-01-23 10:23:50.362870: step: 514/1541, loss: 0.2744199335575104 2023-01-23 10:23:51.240254: step: 516/1541, loss: 0.08616501837968826 2023-01-23 10:23:52.100665: step: 518/1541, loss: 0.1189272403717041 2023-01-23 10:23:52.967273: step: 520/1541, loss: 0.8998648524284363 2023-01-23 10:23:53.801037: step: 522/1541, loss: 0.04100991040468216 2023-01-23 10:23:54.656916: step: 524/1541, loss: 0.14578762650489807 2023-01-23 10:23:55.533032: step: 526/1541, loss: 0.09286539256572723 2023-01-23 10:23:56.401976: step: 528/1541, loss: 0.02152900956571102 2023-01-23 10:23:57.266987: step: 530/1541, loss: 0.018421264365315437 2023-01-23 10:23:58.127792: step: 532/1541, loss: 0.37891924381256104 2023-01-23 10:23:59.014637: step: 534/1541, loss: 0.12260528653860092 2023-01-23 10:23:59.924605: step: 536/1541, loss: 0.07509930431842804 2023-01-23 10:24:00.862487: step: 538/1541, loss: 0.2878243625164032 2023-01-23 10:24:01.737748: step: 540/1541, loss: 0.015092739835381508 2023-01-23 10:24:02.658092: step: 542/1541, loss: 0.09980874508619308 2023-01-23 10:24:03.565446: step: 544/1541, loss: 0.2525864243507385 2023-01-23 10:24:04.422904: step: 546/1541, loss: 0.11617952585220337 2023-01-23 10:24:05.335839: step: 548/1541, loss: 0.023248769342899323 2023-01-23 10:24:06.198234: step: 550/1541, loss: 0.13817304372787476 2023-01-23 10:24:07.149134: step: 552/1541, loss: 0.17448049783706665 2023-01-23 10:24:08.008011: step: 554/1541, loss: 0.17945851385593414 2023-01-23 10:24:08.909868: step: 556/1541, loss: 0.3316504955291748 2023-01-23 10:24:09.787062: step: 558/1541, loss: 0.024365851655602455 2023-01-23 10:24:10.641796: step: 560/1541, loss: 0.0652187392115593 2023-01-23 10:24:11.549794: step: 562/1541, loss: 0.09797476977109909 2023-01-23 10:24:12.446379: step: 564/1541, loss: 0.40941011905670166 2023-01-23 10:24:13.367921: step: 566/1541, loss: 0.1289600282907486 2023-01-23 10:24:14.305350: step: 568/1541, loss: 0.14048725366592407 2023-01-23 10:24:15.293254: step: 570/1541, loss: 0.040136005729436874 2023-01-23 10:24:16.177380: step: 572/1541, loss: 0.1527218520641327 2023-01-23 10:24:17.083467: step: 574/1541, loss: 0.22996388375759125 2023-01-23 10:24:17.996448: step: 576/1541, loss: 0.10459211468696594 2023-01-23 10:24:18.898874: step: 578/1541, loss: 0.1604917049407959 2023-01-23 10:24:19.782761: step: 580/1541, loss: 0.02002616412937641 2023-01-23 10:24:20.624593: step: 582/1541, loss: 0.07155044376850128 2023-01-23 10:24:21.499697: step: 584/1541, loss: 0.07923689484596252 2023-01-23 10:24:22.416613: step: 586/1541, loss: 0.07493698596954346 2023-01-23 10:24:23.317665: step: 588/1541, loss: 0.026764821261167526 2023-01-23 10:24:24.170866: step: 590/1541, loss: 0.25559934973716736 2023-01-23 10:24:25.054333: step: 592/1541, loss: 0.009073223918676376 2023-01-23 10:24:25.933081: step: 594/1541, loss: 0.13741886615753174 2023-01-23 10:24:26.832035: step: 596/1541, loss: 0.023953158408403397 2023-01-23 10:24:27.678867: step: 598/1541, loss: 0.069493867456913 2023-01-23 10:24:28.542697: step: 600/1541, loss: 0.16579490900039673 2023-01-23 10:24:29.429867: step: 602/1541, loss: 0.22524890303611755 2023-01-23 10:24:30.261132: step: 604/1541, loss: 0.2293543964624405 2023-01-23 10:24:31.158894: step: 606/1541, loss: 0.018792128190398216 2023-01-23 10:24:32.044949: step: 608/1541, loss: 0.12152402102947235 2023-01-23 10:24:32.922821: step: 610/1541, loss: 0.08295916020870209 2023-01-23 10:24:33.795642: step: 612/1541, loss: 0.217905655503273 2023-01-23 10:24:34.731667: step: 614/1541, loss: 0.25113052129745483 2023-01-23 10:24:35.620510: step: 616/1541, loss: 0.10698767006397247 2023-01-23 10:24:36.458857: step: 618/1541, loss: 0.0474601648747921 2023-01-23 10:24:37.304809: step: 620/1541, loss: 0.08495486527681351 2023-01-23 10:24:38.213520: step: 622/1541, loss: 0.12824633717536926 2023-01-23 10:24:39.070761: step: 624/1541, loss: 0.17686711251735687 2023-01-23 10:24:39.922224: step: 626/1541, loss: 0.3246619999408722 2023-01-23 10:24:40.837934: step: 628/1541, loss: 0.2100869119167328 2023-01-23 10:24:41.770391: step: 630/1541, loss: 0.12014921009540558 2023-01-23 10:24:42.688399: step: 632/1541, loss: 0.08745823800563812 2023-01-23 10:24:43.578118: step: 634/1541, loss: 0.16206654906272888 2023-01-23 10:24:44.440751: step: 636/1541, loss: 0.048297345638275146 2023-01-23 10:24:45.324344: step: 638/1541, loss: 0.0799349769949913 2023-01-23 10:24:46.259675: step: 640/1541, loss: 0.4309615194797516 2023-01-23 10:24:47.118989: step: 642/1541, loss: 0.08798567950725555 2023-01-23 10:24:47.960269: step: 644/1541, loss: 0.05813656002283096 2023-01-23 10:24:48.869263: step: 646/1541, loss: 0.07475842535495758 2023-01-23 10:24:49.737010: step: 648/1541, loss: 0.023774858564138412 2023-01-23 10:24:50.594219: step: 650/1541, loss: 0.3709335923194885 2023-01-23 10:24:51.476078: step: 652/1541, loss: 0.35682234168052673 2023-01-23 10:24:52.380157: step: 654/1541, loss: 0.09219539165496826 2023-01-23 10:24:53.255273: step: 656/1541, loss: 0.10817499458789825 2023-01-23 10:24:54.137101: step: 658/1541, loss: 0.09730581194162369 2023-01-23 10:24:55.126123: step: 660/1541, loss: 0.03702357038855553 2023-01-23 10:24:56.038787: step: 662/1541, loss: 0.5016486644744873 2023-01-23 10:24:56.939716: step: 664/1541, loss: 0.1946871280670166 2023-01-23 10:24:57.890836: step: 666/1541, loss: 2.4726312160491943 2023-01-23 10:24:58.779043: step: 668/1541, loss: 0.1977391391992569 2023-01-23 10:24:59.679450: step: 670/1541, loss: 0.20406889915466309 2023-01-23 10:25:00.534290: step: 672/1541, loss: 0.20435354113578796 2023-01-23 10:25:01.464000: step: 674/1541, loss: 0.14706620573997498 2023-01-23 10:25:02.349449: step: 676/1541, loss: 0.03048664703965187 2023-01-23 10:25:03.293741: step: 678/1541, loss: 0.46571511030197144 2023-01-23 10:25:04.158594: step: 680/1541, loss: 0.08152467012405396 2023-01-23 10:25:05.037424: step: 682/1541, loss: 0.5380239486694336 2023-01-23 10:25:05.996936: step: 684/1541, loss: 0.01909308135509491 2023-01-23 10:25:06.867389: step: 686/1541, loss: 0.2451467663049698 2023-01-23 10:25:07.709423: step: 688/1541, loss: 0.019949251785874367 2023-01-23 10:25:08.612148: step: 690/1541, loss: 0.014801857993006706 2023-01-23 10:25:09.479052: step: 692/1541, loss: 0.20017944276332855 2023-01-23 10:25:10.357555: step: 694/1541, loss: 0.22736002504825592 2023-01-23 10:25:11.254938: step: 696/1541, loss: 0.003067635465413332 2023-01-23 10:25:12.168776: step: 698/1541, loss: 0.036852046847343445 2023-01-23 10:25:13.033074: step: 700/1541, loss: 0.034484345465898514 2023-01-23 10:25:14.031809: step: 702/1541, loss: 0.2597849369049072 2023-01-23 10:25:14.901409: step: 704/1541, loss: 0.030734965577721596 2023-01-23 10:25:15.750413: step: 706/1541, loss: 0.13762444257736206 2023-01-23 10:25:16.628472: step: 708/1541, loss: 0.28803855180740356 2023-01-23 10:25:17.517061: step: 710/1541, loss: 0.18978466093540192 2023-01-23 10:25:18.387052: step: 712/1541, loss: 0.40993183851242065 2023-01-23 10:25:19.289463: step: 714/1541, loss: 0.4054255783557892 2023-01-23 10:25:20.182959: step: 716/1541, loss: 0.03965717554092407 2023-01-23 10:25:21.100665: step: 718/1541, loss: 0.17821957170963287 2023-01-23 10:25:22.024801: step: 720/1541, loss: 0.13832375407218933 2023-01-23 10:25:22.891288: step: 722/1541, loss: 2.1936583518981934 2023-01-23 10:25:23.759528: step: 724/1541, loss: 0.08562491089105606 2023-01-23 10:25:24.648263: step: 726/1541, loss: 0.02921951562166214 2023-01-23 10:25:25.520589: step: 728/1541, loss: 0.1890244483947754 2023-01-23 10:25:26.407271: step: 730/1541, loss: 0.2819880247116089 2023-01-23 10:25:27.329657: step: 732/1541, loss: 0.29596251249313354 2023-01-23 10:25:28.195710: step: 734/1541, loss: 0.03373071923851967 2023-01-23 10:25:29.103642: step: 736/1541, loss: 0.16430184245109558 2023-01-23 10:25:30.085808: step: 738/1541, loss: 0.07852575927972794 2023-01-23 10:25:30.966211: step: 740/1541, loss: 0.18986336886882782 2023-01-23 10:25:31.859544: step: 742/1541, loss: 1.518639326095581 2023-01-23 10:25:32.751865: step: 744/1541, loss: 0.11052008718252182 2023-01-23 10:25:33.660876: step: 746/1541, loss: 0.0058857291005551815 2023-01-23 10:25:34.581026: step: 748/1541, loss: 0.23233547806739807 2023-01-23 10:25:35.435670: step: 750/1541, loss: 0.2377142459154129 2023-01-23 10:25:36.318896: step: 752/1541, loss: 0.34622183442115784 2023-01-23 10:25:37.234176: step: 754/1541, loss: 0.2931681275367737 2023-01-23 10:25:38.149047: step: 756/1541, loss: 0.2652182877063751 2023-01-23 10:25:38.996941: step: 758/1541, loss: 0.4458217918872833 2023-01-23 10:25:39.876528: step: 760/1541, loss: 0.1530395895242691 2023-01-23 10:25:40.788854: step: 762/1541, loss: 0.0711209625005722 2023-01-23 10:25:41.678119: step: 764/1541, loss: 0.1568879932165146 2023-01-23 10:25:42.561849: step: 766/1541, loss: 0.03206634148955345 2023-01-23 10:25:43.415414: step: 768/1541, loss: 0.3907594382762909 2023-01-23 10:25:44.320189: step: 770/1541, loss: 0.01447255164384842 2023-01-23 10:25:45.191725: step: 772/1541, loss: 0.1852094978094101 2023-01-23 10:25:46.095690: step: 774/1541, loss: 0.020221391692757607 2023-01-23 10:25:47.036595: step: 776/1541, loss: 0.17897379398345947 2023-01-23 10:25:47.922943: step: 778/1541, loss: 0.31398022174835205 2023-01-23 10:25:48.776688: step: 780/1541, loss: 0.1743018925189972 2023-01-23 10:25:49.634020: step: 782/1541, loss: 0.06192611902952194 2023-01-23 10:25:50.520335: step: 784/1541, loss: 0.013832957483828068 2023-01-23 10:25:51.385198: step: 786/1541, loss: 0.26536327600479126 2023-01-23 10:25:52.295324: step: 788/1541, loss: 1.578207015991211 2023-01-23 10:25:53.145764: step: 790/1541, loss: 0.14989978075027466 2023-01-23 10:25:54.010134: step: 792/1541, loss: 0.11992975324392319 2023-01-23 10:25:54.895881: step: 794/1541, loss: 0.08013349771499634 2023-01-23 10:25:55.764509: step: 796/1541, loss: 0.22303228080272675 2023-01-23 10:25:56.640624: step: 798/1541, loss: 0.03242185711860657 2023-01-23 10:25:57.538068: step: 800/1541, loss: 0.2727391719818115 2023-01-23 10:25:58.414373: step: 802/1541, loss: 0.4539547860622406 2023-01-23 10:25:59.312035: step: 804/1541, loss: 0.21222376823425293 2023-01-23 10:26:00.182584: step: 806/1541, loss: 0.036525048315525055 2023-01-23 10:26:01.046211: step: 808/1541, loss: 0.12726178765296936 2023-01-23 10:26:01.967800: step: 810/1541, loss: 0.2043916881084442 2023-01-23 10:26:02.864692: step: 812/1541, loss: 0.603480875492096 2023-01-23 10:26:03.817921: step: 814/1541, loss: 0.07082510739564896 2023-01-23 10:26:04.710097: step: 816/1541, loss: 0.019454896450042725 2023-01-23 10:26:05.623799: step: 818/1541, loss: 0.08674512803554535 2023-01-23 10:26:06.518635: step: 820/1541, loss: 0.4673219919204712 2023-01-23 10:26:07.396228: step: 822/1541, loss: 0.1231248751282692 2023-01-23 10:26:08.295161: step: 824/1541, loss: 0.2915087044239044 2023-01-23 10:26:09.250975: step: 826/1541, loss: 0.22309638559818268 2023-01-23 10:26:10.124116: step: 828/1541, loss: 0.062225792557001114 2023-01-23 10:26:11.002337: step: 830/1541, loss: 0.17680048942565918 2023-01-23 10:26:11.862763: step: 832/1541, loss: 1.2092747688293457 2023-01-23 10:26:12.740107: step: 834/1541, loss: 0.1599297970533371 2023-01-23 10:26:13.650453: step: 836/1541, loss: 0.04258193075656891 2023-01-23 10:26:14.526092: step: 838/1541, loss: 0.08865717798471451 2023-01-23 10:26:15.401653: step: 840/1541, loss: 0.13467846810817719 2023-01-23 10:26:16.281128: step: 842/1541, loss: 0.07156117260456085 2023-01-23 10:26:17.133586: step: 844/1541, loss: 0.198393315076828 2023-01-23 10:26:18.001360: step: 846/1541, loss: 0.033311642706394196 2023-01-23 10:26:18.879134: step: 848/1541, loss: 0.22726979851722717 2023-01-23 10:26:19.800364: step: 850/1541, loss: 0.28616592288017273 2023-01-23 10:26:20.683924: step: 852/1541, loss: 0.14036446809768677 2023-01-23 10:26:21.551917: step: 854/1541, loss: 0.09683216363191605 2023-01-23 10:26:22.445377: step: 856/1541, loss: 0.18846160173416138 2023-01-23 10:26:23.358761: step: 858/1541, loss: 0.1772298961877823 2023-01-23 10:26:24.221071: step: 860/1541, loss: 0.042533762753009796 2023-01-23 10:26:25.131663: step: 862/1541, loss: 0.2957211136817932 2023-01-23 10:26:26.037008: step: 864/1541, loss: 0.09645114839076996 2023-01-23 10:26:26.936222: step: 866/1541, loss: 0.08994557708501816 2023-01-23 10:26:27.829005: step: 868/1541, loss: 0.007463986985385418 2023-01-23 10:26:28.696139: step: 870/1541, loss: 0.0023049484007060528 2023-01-23 10:26:29.577349: step: 872/1541, loss: 0.04241134598851204 2023-01-23 10:26:30.461747: step: 874/1541, loss: 0.30492451786994934 2023-01-23 10:26:31.322404: step: 876/1541, loss: 0.36645200848579407 2023-01-23 10:26:32.172285: step: 878/1541, loss: 0.08870671689510345 2023-01-23 10:26:33.018865: step: 880/1541, loss: 0.04334888979792595 2023-01-23 10:26:33.899118: step: 882/1541, loss: 0.1389448344707489 2023-01-23 10:26:34.839851: step: 884/1541, loss: 0.16635218262672424 2023-01-23 10:26:35.749307: step: 886/1541, loss: 0.027545353397727013 2023-01-23 10:26:36.637783: step: 888/1541, loss: 0.04322892054915428 2023-01-23 10:26:37.524220: step: 890/1541, loss: 0.007190319709479809 2023-01-23 10:26:38.437080: step: 892/1541, loss: 1.1748244762420654 2023-01-23 10:26:39.342584: step: 894/1541, loss: 0.0819292739033699 2023-01-23 10:26:40.253412: step: 896/1541, loss: 1.4188971519470215 2023-01-23 10:26:41.165315: step: 898/1541, loss: 0.5560499429702759 2023-01-23 10:26:42.020677: step: 900/1541, loss: 0.0829339548945427 2023-01-23 10:26:42.929393: step: 902/1541, loss: 0.26201605796813965 2023-01-23 10:26:43.820968: step: 904/1541, loss: 0.005664180964231491 2023-01-23 10:26:44.775967: step: 906/1541, loss: 0.19393031299114227 2023-01-23 10:26:45.652276: step: 908/1541, loss: 0.08615335822105408 2023-01-23 10:26:46.538285: step: 910/1541, loss: 0.11862499266862869 2023-01-23 10:26:47.399852: step: 912/1541, loss: 0.29548195004463196 2023-01-23 10:26:48.264109: step: 914/1541, loss: 0.1660481095314026 2023-01-23 10:26:49.142075: step: 916/1541, loss: 0.030143460258841515 2023-01-23 10:26:50.070264: step: 918/1541, loss: 0.10185503959655762 2023-01-23 10:26:50.947135: step: 920/1541, loss: 0.0656929686665535 2023-01-23 10:26:51.810437: step: 922/1541, loss: 0.06236789748072624 2023-01-23 10:26:52.683164: step: 924/1541, loss: 0.023717742413282394 2023-01-23 10:26:53.575534: step: 926/1541, loss: 0.0070951455272734165 2023-01-23 10:26:54.509554: step: 928/1541, loss: 0.38139814138412476 2023-01-23 10:26:55.402050: step: 930/1541, loss: 0.23583029210567474 2023-01-23 10:26:56.284034: step: 932/1541, loss: 0.03836539387702942 2023-01-23 10:26:57.230862: step: 934/1541, loss: 0.7849975824356079 2023-01-23 10:26:58.111599: step: 936/1541, loss: 0.09586570411920547 2023-01-23 10:26:58.988764: step: 938/1541, loss: 0.15548637509346008 2023-01-23 10:26:59.882391: step: 940/1541, loss: 0.22466109693050385 2023-01-23 10:27:00.749055: step: 942/1541, loss: 0.08327692002058029 2023-01-23 10:27:01.629256: step: 944/1541, loss: 0.19247642159461975 2023-01-23 10:27:02.574057: step: 946/1541, loss: 0.5440042614936829 2023-01-23 10:27:03.473121: step: 948/1541, loss: 0.37389498949050903 2023-01-23 10:27:04.354761: step: 950/1541, loss: 0.33311405777931213 2023-01-23 10:27:05.259627: step: 952/1541, loss: 0.30567944049835205 2023-01-23 10:27:06.160139: step: 954/1541, loss: 0.3209693729877472 2023-01-23 10:27:07.097391: step: 956/1541, loss: 0.06249599531292915 2023-01-23 10:27:08.009562: step: 958/1541, loss: 0.22407841682434082 2023-01-23 10:27:08.922740: step: 960/1541, loss: 0.13848018646240234 2023-01-23 10:27:09.855944: step: 962/1541, loss: 0.17435020208358765 2023-01-23 10:27:10.768734: step: 964/1541, loss: 0.31932365894317627 2023-01-23 10:27:11.645701: step: 966/1541, loss: 0.3769531846046448 2023-01-23 10:27:12.553841: step: 968/1541, loss: 0.24220532178878784 2023-01-23 10:27:13.418458: step: 970/1541, loss: 0.2483653575181961 2023-01-23 10:27:14.312879: step: 972/1541, loss: 0.16435526311397552 2023-01-23 10:27:15.176428: step: 974/1541, loss: 0.12050015479326248 2023-01-23 10:27:16.047690: step: 976/1541, loss: 0.18437854945659637 2023-01-23 10:27:16.918353: step: 978/1541, loss: 0.11382130533456802 2023-01-23 10:27:17.800285: step: 980/1541, loss: 0.19943319261074066 2023-01-23 10:27:18.766627: step: 982/1541, loss: 0.016161803156137466 2023-01-23 10:27:19.664621: step: 984/1541, loss: 0.04287366941571236 2023-01-23 10:27:20.596382: step: 986/1541, loss: 0.18406175076961517 2023-01-23 10:27:21.440317: step: 988/1541, loss: 0.9117844700813293 2023-01-23 10:27:22.308669: step: 990/1541, loss: 0.10768326371908188 2023-01-23 10:27:23.155676: step: 992/1541, loss: 0.262521892786026 2023-01-23 10:27:24.045149: step: 994/1541, loss: 0.028555478900671005 2023-01-23 10:27:24.900769: step: 996/1541, loss: 0.11819688975811005 2023-01-23 10:27:25.870384: step: 998/1541, loss: 0.20282603800296783 2023-01-23 10:27:26.771185: step: 1000/1541, loss: 0.24656614661216736 2023-01-23 10:27:27.623554: step: 1002/1541, loss: 0.14541681110858917 2023-01-23 10:27:28.522961: step: 1004/1541, loss: 0.0245343130081892 2023-01-23 10:27:29.424560: step: 1006/1541, loss: 0.047107893973588943 2023-01-23 10:27:30.307751: step: 1008/1541, loss: 0.040835585445165634 2023-01-23 10:27:31.196022: step: 1010/1541, loss: 0.011893071234226227 2023-01-23 10:27:32.163972: step: 1012/1541, loss: 0.1289500594139099 2023-01-23 10:27:33.047642: step: 1014/1541, loss: 0.06470230221748352 2023-01-23 10:27:33.949228: step: 1016/1541, loss: 0.03079795092344284 2023-01-23 10:27:34.821287: step: 1018/1541, loss: 0.06757567822933197 2023-01-23 10:27:35.680419: step: 1020/1541, loss: 0.10150504857301712 2023-01-23 10:27:36.585541: step: 1022/1541, loss: 0.018982838839292526 2023-01-23 10:27:37.499021: step: 1024/1541, loss: 0.6861980557441711 2023-01-23 10:27:38.519337: step: 1026/1541, loss: 0.28631871938705444 2023-01-23 10:27:39.395720: step: 1028/1541, loss: 0.25972503423690796 2023-01-23 10:27:40.277395: step: 1030/1541, loss: 0.14492298662662506 2023-01-23 10:27:41.135123: step: 1032/1541, loss: 0.12633422017097473 2023-01-23 10:27:42.062044: step: 1034/1541, loss: 0.5508264303207397 2023-01-23 10:27:43.016201: step: 1036/1541, loss: 0.006713109090924263 2023-01-23 10:27:43.876468: step: 1038/1541, loss: 0.14977845549583435 2023-01-23 10:27:44.763453: step: 1040/1541, loss: 0.10552835464477539 2023-01-23 10:27:45.745286: step: 1042/1541, loss: 0.009135051630437374 2023-01-23 10:27:46.609772: step: 1044/1541, loss: 0.08970559388399124 2023-01-23 10:27:47.521564: step: 1046/1541, loss: 0.016732893884181976 2023-01-23 10:27:48.428952: step: 1048/1541, loss: 0.058450087904930115 2023-01-23 10:27:49.310068: step: 1050/1541, loss: 0.2579370439052582 2023-01-23 10:27:50.179100: step: 1052/1541, loss: 0.2687634825706482 2023-01-23 10:27:51.065006: step: 1054/1541, loss: 0.35054317116737366 2023-01-23 10:27:51.975192: step: 1056/1541, loss: 0.20351995527744293 2023-01-23 10:27:52.897794: step: 1058/1541, loss: 0.29854443669319153 2023-01-23 10:27:53.810309: step: 1060/1541, loss: 0.4391629993915558 2023-01-23 10:27:54.688459: step: 1062/1541, loss: 0.11023774743080139 2023-01-23 10:27:55.529114: step: 1064/1541, loss: 0.16451884806156158 2023-01-23 10:27:56.445242: step: 1066/1541, loss: 0.1692512035369873 2023-01-23 10:27:57.383060: step: 1068/1541, loss: 0.05291922390460968 2023-01-23 10:27:58.326132: step: 1070/1541, loss: 0.07481043040752411 2023-01-23 10:27:59.219941: step: 1072/1541, loss: 0.032957274466753006 2023-01-23 10:28:00.141538: step: 1074/1541, loss: 0.02909480407834053 2023-01-23 10:28:01.042623: step: 1076/1541, loss: 0.2589866816997528 2023-01-23 10:28:01.937711: step: 1078/1541, loss: 0.20916299521923065 2023-01-23 10:28:02.849843: step: 1080/1541, loss: 0.3736459016799927 2023-01-23 10:28:03.752606: step: 1082/1541, loss: 0.1589805632829666 2023-01-23 10:28:04.614908: step: 1084/1541, loss: 0.03411915898323059 2023-01-23 10:28:05.559331: step: 1086/1541, loss: 0.0281282439827919 2023-01-23 10:28:06.428972: step: 1088/1541, loss: 0.12391482293605804 2023-01-23 10:28:07.299219: step: 1090/1541, loss: 0.0690787211060524 2023-01-23 10:28:08.157798: step: 1092/1541, loss: 0.12250005453824997 2023-01-23 10:28:09.043583: step: 1094/1541, loss: 0.08345994353294373 2023-01-23 10:28:09.951113: step: 1096/1541, loss: 0.35577258467674255 2023-01-23 10:28:10.848451: step: 1098/1541, loss: 0.23313695192337036 2023-01-23 10:28:11.757613: step: 1100/1541, loss: 0.01838330179452896 2023-01-23 10:28:12.632765: step: 1102/1541, loss: 0.3179771602153778 2023-01-23 10:28:13.516528: step: 1104/1541, loss: 0.042009126394987106 2023-01-23 10:28:14.380286: step: 1106/1541, loss: 0.07259220629930496 2023-01-23 10:28:15.237987: step: 1108/1541, loss: 0.18996047973632812 2023-01-23 10:28:16.196155: step: 1110/1541, loss: 0.10122136771678925 2023-01-23 10:28:17.086041: step: 1112/1541, loss: 0.583800196647644 2023-01-23 10:28:17.922933: step: 1114/1541, loss: 0.019644688814878464 2023-01-23 10:28:18.812683: step: 1116/1541, loss: 0.2022344321012497 2023-01-23 10:28:19.683698: step: 1118/1541, loss: 0.029714317992329597 2023-01-23 10:28:20.559258: step: 1120/1541, loss: 0.1114569902420044 2023-01-23 10:28:21.424798: step: 1122/1541, loss: 0.07159441709518433 2023-01-23 10:28:22.281719: step: 1124/1541, loss: 0.13358105719089508 2023-01-23 10:28:23.170219: step: 1126/1541, loss: 0.20022085309028625 2023-01-23 10:28:24.041017: step: 1128/1541, loss: 0.13059929013252258 2023-01-23 10:28:24.935035: step: 1130/1541, loss: 0.03187417611479759 2023-01-23 10:28:25.814123: step: 1132/1541, loss: 0.1995895355939865 2023-01-23 10:28:26.718895: step: 1134/1541, loss: 0.03487657755613327 2023-01-23 10:28:27.680842: step: 1136/1541, loss: 0.6163919568061829 2023-01-23 10:28:28.737986: step: 1138/1541, loss: 0.06976333260536194 2023-01-23 10:28:29.657816: step: 1140/1541, loss: 0.02711539715528488 2023-01-23 10:28:30.538104: step: 1142/1541, loss: 0.01820821315050125 2023-01-23 10:28:31.455881: step: 1144/1541, loss: 0.18508774042129517 2023-01-23 10:28:32.334741: step: 1146/1541, loss: 0.02815510518848896 2023-01-23 10:28:33.230734: step: 1148/1541, loss: 0.4103888273239136 2023-01-23 10:28:34.069018: step: 1150/1541, loss: 0.14353510737419128 2023-01-23 10:28:34.973708: step: 1152/1541, loss: 0.25519686937332153 2023-01-23 10:28:35.872746: step: 1154/1541, loss: 0.1884785145521164 2023-01-23 10:28:36.744677: step: 1156/1541, loss: 0.05196187272667885 2023-01-23 10:28:37.613198: step: 1158/1541, loss: 0.06945130228996277 2023-01-23 10:28:38.529206: step: 1160/1541, loss: 0.17044447362422943 2023-01-23 10:28:39.401672: step: 1162/1541, loss: 0.1554514616727829 2023-01-23 10:28:40.374877: step: 1164/1541, loss: 0.25760653614997864 2023-01-23 10:28:41.272976: step: 1166/1541, loss: 0.22818812727928162 2023-01-23 10:28:42.159868: step: 1168/1541, loss: 0.09130606055259705 2023-01-23 10:28:43.016436: step: 1170/1541, loss: 0.01758653298020363 2023-01-23 10:28:43.881337: step: 1172/1541, loss: 0.8593234419822693 2023-01-23 10:28:44.810357: step: 1174/1541, loss: 0.06918112933635712 2023-01-23 10:28:45.689515: step: 1176/1541, loss: 1.2577528953552246 2023-01-23 10:28:46.562803: step: 1178/1541, loss: 0.10790503025054932 2023-01-23 10:28:47.446577: step: 1180/1541, loss: 0.11670823395252228 2023-01-23 10:28:48.296840: step: 1182/1541, loss: 0.23065395653247833 2023-01-23 10:28:49.219420: step: 1184/1541, loss: 0.08427488058805466 2023-01-23 10:28:50.095153: step: 1186/1541, loss: 0.6264159083366394 2023-01-23 10:28:50.946402: step: 1188/1541, loss: 0.2298872172832489 2023-01-23 10:28:51.828325: step: 1190/1541, loss: 0.05440961569547653 2023-01-23 10:28:52.739512: step: 1192/1541, loss: 0.1681968718767166 2023-01-23 10:28:53.610523: step: 1194/1541, loss: 0.07492861896753311 2023-01-23 10:28:54.509355: step: 1196/1541, loss: 0.06334839761257172 2023-01-23 10:28:55.426272: step: 1198/1541, loss: 0.1552882194519043 2023-01-23 10:28:56.343367: step: 1200/1541, loss: 0.4133347272872925 2023-01-23 10:28:57.266295: step: 1202/1541, loss: 0.03748966008424759 2023-01-23 10:28:58.247303: step: 1204/1541, loss: 0.2466922402381897 2023-01-23 10:28:59.101610: step: 1206/1541, loss: 0.0479394868016243 2023-01-23 10:29:00.076591: step: 1208/1541, loss: 0.05317329242825508 2023-01-23 10:29:00.991515: step: 1210/1541, loss: 0.05180210620164871 2023-01-23 10:29:01.934650: step: 1212/1541, loss: 0.08354678750038147 2023-01-23 10:29:02.869725: step: 1214/1541, loss: 0.10773255676031113 2023-01-23 10:29:03.808004: step: 1216/1541, loss: 0.2345687299966812 2023-01-23 10:29:04.689781: step: 1218/1541, loss: 0.06256063282489777 2023-01-23 10:29:05.530009: step: 1220/1541, loss: 0.20849888026714325 2023-01-23 10:29:06.414437: step: 1222/1541, loss: 0.32451131939888 2023-01-23 10:29:07.324891: step: 1224/1541, loss: 0.16121706366539001 2023-01-23 10:29:08.194766: step: 1226/1541, loss: 0.08832606673240662 2023-01-23 10:29:09.153928: step: 1228/1541, loss: 2.6330130100250244 2023-01-23 10:29:10.013946: step: 1230/1541, loss: 0.008271194994449615 2023-01-23 10:29:10.894657: step: 1232/1541, loss: 0.04873185232281685 2023-01-23 10:29:11.773445: step: 1234/1541, loss: 0.579222559928894 2023-01-23 10:29:12.717886: step: 1236/1541, loss: 0.3366144001483917 2023-01-23 10:29:13.611353: step: 1238/1541, loss: 0.16233791410923004 2023-01-23 10:29:14.529796: step: 1240/1541, loss: 0.08677423000335693 2023-01-23 10:29:15.483594: step: 1242/1541, loss: 0.8881179094314575 2023-01-23 10:29:16.384609: step: 1244/1541, loss: 0.052927665412425995 2023-01-23 10:29:17.300910: step: 1246/1541, loss: 0.15911951661109924 2023-01-23 10:29:18.153638: step: 1248/1541, loss: 0.020249243825674057 2023-01-23 10:29:19.031167: step: 1250/1541, loss: 0.32046476006507874 2023-01-23 10:29:19.934266: step: 1252/1541, loss: 2.256225109100342 2023-01-23 10:29:20.834689: step: 1254/1541, loss: 0.06879883259534836 2023-01-23 10:29:21.788605: step: 1256/1541, loss: 0.07963582873344421 2023-01-23 10:29:22.718825: step: 1258/1541, loss: 0.14111466705799103 2023-01-23 10:29:23.620996: step: 1260/1541, loss: 0.2621728181838989 2023-01-23 10:29:24.487390: step: 1262/1541, loss: 0.2512242794036865 2023-01-23 10:29:25.397162: step: 1264/1541, loss: 0.17010188102722168 2023-01-23 10:29:26.272307: step: 1266/1541, loss: 0.09736014902591705 2023-01-23 10:29:27.243968: step: 1268/1541, loss: 0.1236225962638855 2023-01-23 10:29:28.124421: step: 1270/1541, loss: 0.011911474168300629 2023-01-23 10:29:28.991132: step: 1272/1541, loss: 0.2239571362733841 2023-01-23 10:29:29.901016: step: 1274/1541, loss: 0.07729005068540573 2023-01-23 10:29:30.774847: step: 1276/1541, loss: 0.05967266112565994 2023-01-23 10:29:31.677854: step: 1278/1541, loss: 0.48240235447883606 2023-01-23 10:29:32.594750: step: 1280/1541, loss: 0.09398142993450165 2023-01-23 10:29:33.463879: step: 1282/1541, loss: 0.05666828528046608 2023-01-23 10:29:34.348690: step: 1284/1541, loss: 0.19524072110652924 2023-01-23 10:29:35.303534: step: 1286/1541, loss: 0.05756199732422829 2023-01-23 10:29:36.145621: step: 1288/1541, loss: 0.220774844288826 2023-01-23 10:29:37.046733: step: 1290/1541, loss: 0.020608926191926003 2023-01-23 10:29:37.954690: step: 1292/1541, loss: 0.18271423876285553 2023-01-23 10:29:38.872590: step: 1294/1541, loss: 0.09022761136293411 2023-01-23 10:29:39.719216: step: 1296/1541, loss: 0.008348757401108742 2023-01-23 10:29:40.613743: step: 1298/1541, loss: 0.09520021826028824 2023-01-23 10:29:41.468966: step: 1300/1541, loss: 0.11873800307512283 2023-01-23 10:29:42.340227: step: 1302/1541, loss: 0.4027450680732727 2023-01-23 10:29:43.240401: step: 1304/1541, loss: 0.24427098035812378 2023-01-23 10:29:44.224297: step: 1306/1541, loss: 0.11246512085199356 2023-01-23 10:29:45.133080: step: 1308/1541, loss: 0.542718231678009 2023-01-23 10:29:46.015652: step: 1310/1541, loss: 0.260122150182724 2023-01-23 10:29:46.928740: step: 1312/1541, loss: 0.12359218299388885 2023-01-23 10:29:47.781512: step: 1314/1541, loss: 0.18254169821739197 2023-01-23 10:29:48.653800: step: 1316/1541, loss: 0.0037288737948983908 2023-01-23 10:29:49.543288: step: 1318/1541, loss: 0.08771612495183945 2023-01-23 10:29:50.414443: step: 1320/1541, loss: 0.08834262192249298 2023-01-23 10:29:51.296706: step: 1322/1541, loss: 0.10237499326467514 2023-01-23 10:29:52.204662: step: 1324/1541, loss: 0.3039977252483368 2023-01-23 10:29:53.064152: step: 1326/1541, loss: 0.589273989200592 2023-01-23 10:29:53.973483: step: 1328/1541, loss: 0.18807269632816315 2023-01-23 10:29:54.852504: step: 1330/1541, loss: 0.32108691334724426 2023-01-23 10:29:55.763627: step: 1332/1541, loss: 0.24720360338687897 2023-01-23 10:29:56.653821: step: 1334/1541, loss: 0.4550822079181671 2023-01-23 10:29:57.528641: step: 1336/1541, loss: 0.17642049491405487 2023-01-23 10:29:58.409506: step: 1338/1541, loss: 0.04445411637425423 2023-01-23 10:29:59.328346: step: 1340/1541, loss: 0.21716094017028809 2023-01-23 10:30:00.209430: step: 1342/1541, loss: 0.08493166416883469 2023-01-23 10:30:01.075051: step: 1344/1541, loss: 0.6443010568618774 2023-01-23 10:30:01.957408: step: 1346/1541, loss: 0.0483509823679924 2023-01-23 10:30:02.854910: step: 1348/1541, loss: 0.06304288655519485 2023-01-23 10:30:03.770692: step: 1350/1541, loss: 0.3455091416835785 2023-01-23 10:30:04.653153: step: 1352/1541, loss: 0.02643459104001522 2023-01-23 10:30:05.550663: step: 1354/1541, loss: 0.37084758281707764 2023-01-23 10:30:06.436739: step: 1356/1541, loss: 0.2347051501274109 2023-01-23 10:30:07.310721: step: 1358/1541, loss: 0.12162460386753082 2023-01-23 10:30:08.197660: step: 1360/1541, loss: 0.2622501850128174 2023-01-23 10:30:09.051921: step: 1362/1541, loss: 0.13881759345531464 2023-01-23 10:30:10.020439: step: 1364/1541, loss: 0.07189592719078064 2023-01-23 10:30:10.970802: step: 1366/1541, loss: 0.06055107340216637 2023-01-23 10:30:11.893330: step: 1368/1541, loss: 0.36762624979019165 2023-01-23 10:30:12.773862: step: 1370/1541, loss: 0.18256552517414093 2023-01-23 10:30:13.639581: step: 1372/1541, loss: 0.06519514322280884 2023-01-23 10:30:14.540239: step: 1374/1541, loss: 0.29345643520355225 2023-01-23 10:30:15.416307: step: 1376/1541, loss: 0.6225574016571045 2023-01-23 10:30:16.264545: step: 1378/1541, loss: 0.09278743714094162 2023-01-23 10:30:17.121893: step: 1380/1541, loss: 0.19009646773338318 2023-01-23 10:30:17.972705: step: 1382/1541, loss: 0.035097636282444 2023-01-23 10:30:18.839920: step: 1384/1541, loss: 0.27805283665657043 2023-01-23 10:30:19.724165: step: 1386/1541, loss: 0.019220519810914993 2023-01-23 10:30:20.624182: step: 1388/1541, loss: 0.08846894651651382 2023-01-23 10:30:21.518201: step: 1390/1541, loss: 0.09388655424118042 2023-01-23 10:30:22.351377: step: 1392/1541, loss: 0.17419083416461945 2023-01-23 10:30:23.257252: step: 1394/1541, loss: 2.711385488510132 2023-01-23 10:30:24.123513: step: 1396/1541, loss: 0.19862356781959534 2023-01-23 10:30:25.019659: step: 1398/1541, loss: 0.036098070442676544 2023-01-23 10:30:25.986500: step: 1400/1541, loss: 0.8712528944015503 2023-01-23 10:30:26.877722: step: 1402/1541, loss: 0.15486501157283783 2023-01-23 10:30:27.767324: step: 1404/1541, loss: 0.16825763881206512 2023-01-23 10:30:28.722390: step: 1406/1541, loss: 0.10223603248596191 2023-01-23 10:30:29.627366: step: 1408/1541, loss: 0.1421528160572052 2023-01-23 10:30:30.526150: step: 1410/1541, loss: 0.38902950286865234 2023-01-23 10:30:31.417636: step: 1412/1541, loss: 0.04907439649105072 2023-01-23 10:30:32.296370: step: 1414/1541, loss: 0.007630124222487211 2023-01-23 10:30:33.124086: step: 1416/1541, loss: 0.09627074003219604 2023-01-23 10:30:33.997832: step: 1418/1541, loss: 0.42865434288978577 2023-01-23 10:30:35.009464: step: 1420/1541, loss: 0.3100264370441437 2023-01-23 10:30:35.911171: step: 1422/1541, loss: 0.10368989408016205 2023-01-23 10:30:36.787631: step: 1424/1541, loss: 0.2590923607349396 2023-01-23 10:30:37.731261: step: 1426/1541, loss: 0.5227692723274231 2023-01-23 10:30:38.671860: step: 1428/1541, loss: 0.04835556820034981 2023-01-23 10:30:39.590306: step: 1430/1541, loss: 0.2940661609172821 2023-01-23 10:30:40.527316: step: 1432/1541, loss: 0.07506614923477173 2023-01-23 10:30:41.403341: step: 1434/1541, loss: 0.18826232850551605 2023-01-23 10:30:42.269746: step: 1436/1541, loss: 0.10107620060443878 2023-01-23 10:30:43.157067: step: 1438/1541, loss: 0.13342103362083435 2023-01-23 10:30:44.044178: step: 1440/1541, loss: 0.17042870819568634 2023-01-23 10:30:44.983574: step: 1442/1541, loss: 0.15552489459514618 2023-01-23 10:30:45.830249: step: 1444/1541, loss: 0.10002627223730087 2023-01-23 10:30:46.710837: step: 1446/1541, loss: 0.41777926683425903 2023-01-23 10:30:47.625741: step: 1448/1541, loss: 0.05677998811006546 2023-01-23 10:30:48.519061: step: 1450/1541, loss: 0.5642271637916565 2023-01-23 10:30:49.462548: step: 1452/1541, loss: 0.16985686123371124 2023-01-23 10:30:50.315448: step: 1454/1541, loss: 0.2322392761707306 2023-01-23 10:30:51.258856: step: 1456/1541, loss: 0.20676833391189575 2023-01-23 10:30:52.153301: step: 1458/1541, loss: 0.3147340416908264 2023-01-23 10:30:53.013842: step: 1460/1541, loss: 0.21249879896640778 2023-01-23 10:30:53.951833: step: 1462/1541, loss: 0.20070436596870422 2023-01-23 10:30:54.853287: step: 1464/1541, loss: 0.09062612056732178 2023-01-23 10:30:55.723674: step: 1466/1541, loss: 0.25326889753341675 2023-01-23 10:30:56.617094: step: 1468/1541, loss: 0.2535042464733124 2023-01-23 10:30:57.493137: step: 1470/1541, loss: 0.28743815422058105 2023-01-23 10:30:58.420591: step: 1472/1541, loss: 0.09681081026792526 2023-01-23 10:30:59.285600: step: 1474/1541, loss: 0.05282067507505417 2023-01-23 10:31:00.190375: step: 1476/1541, loss: 0.19418489933013916 2023-01-23 10:31:01.069487: step: 1478/1541, loss: 0.14149707555770874 2023-01-23 10:31:01.969885: step: 1480/1541, loss: 0.3006240427494049 2023-01-23 10:31:02.887351: step: 1482/1541, loss: 0.3269377052783966 2023-01-23 10:31:03.797115: step: 1484/1541, loss: 0.29942336678504944 2023-01-23 10:31:04.653149: step: 1486/1541, loss: 0.05169253051280975 2023-01-23 10:31:05.530014: step: 1488/1541, loss: 0.1441398710012436 2023-01-23 10:31:06.436055: step: 1490/1541, loss: 0.08357186615467072 2023-01-23 10:31:07.345322: step: 1492/1541, loss: 0.47327303886413574 2023-01-23 10:31:08.251778: step: 1494/1541, loss: 0.15663455426692963 2023-01-23 10:31:09.131330: step: 1496/1541, loss: 0.14397969841957092 2023-01-23 10:31:10.053188: step: 1498/1541, loss: 0.07914845645427704 2023-01-23 10:31:10.906021: step: 1500/1541, loss: 0.01713707484304905 2023-01-23 10:31:11.774629: step: 1502/1541, loss: 0.16920337080955505 2023-01-23 10:31:12.689843: step: 1504/1541, loss: 0.13516810536384583 2023-01-23 10:31:13.627584: step: 1506/1541, loss: 0.26475369930267334 2023-01-23 10:31:14.538103: step: 1508/1541, loss: 0.1882440447807312 2023-01-23 10:31:15.412941: step: 1510/1541, loss: 0.17322738468647003 2023-01-23 10:31:16.387468: step: 1512/1541, loss: 0.4252968430519104 2023-01-23 10:31:17.239710: step: 1514/1541, loss: 0.053564202040433884 2023-01-23 10:31:18.121434: step: 1516/1541, loss: 0.06631220132112503 2023-01-23 10:31:19.042933: step: 1518/1541, loss: 0.6690426468849182 2023-01-23 10:31:19.971420: step: 1520/1541, loss: 0.06110912561416626 2023-01-23 10:31:20.885091: step: 1522/1541, loss: 0.24981315433979034 2023-01-23 10:31:21.756571: step: 1524/1541, loss: 0.024884283542633057 2023-01-23 10:31:22.606757: step: 1526/1541, loss: 0.033903200179338455 2023-01-23 10:31:23.494675: step: 1528/1541, loss: 0.2552068829536438 2023-01-23 10:31:24.380871: step: 1530/1541, loss: 0.26298075914382935 2023-01-23 10:31:25.287325: step: 1532/1541, loss: 0.34116968512535095 2023-01-23 10:31:26.156000: step: 1534/1541, loss: 0.2700466513633728 2023-01-23 10:31:27.066580: step: 1536/1541, loss: 0.024538546800613403 2023-01-23 10:31:27.997326: step: 1538/1541, loss: 0.1607624888420105 2023-01-23 10:31:28.895818: step: 1540/1541, loss: 0.7924625873565674 2023-01-23 10:31:29.797034: step: 1542/1541, loss: 0.09427458047866821 2023-01-23 10:31:30.680489: step: 1544/1541, loss: 0.06996536999940872 2023-01-23 10:31:31.570938: step: 1546/1541, loss: 0.2704039514064789 2023-01-23 10:31:32.442207: step: 1548/1541, loss: 0.21806804835796356 2023-01-23 10:31:33.328974: step: 1550/1541, loss: 0.28367120027542114 2023-01-23 10:31:34.241473: step: 1552/1541, loss: 0.031344328075647354 2023-01-23 10:31:35.084700: step: 1554/1541, loss: 0.038974400609731674 2023-01-23 10:31:35.972919: step: 1556/1541, loss: 0.09782549738883972 2023-01-23 10:31:36.878981: step: 1558/1541, loss: 0.2903163433074951 2023-01-23 10:31:37.765117: step: 1560/1541, loss: 0.23461566865444183 2023-01-23 10:31:38.629363: step: 1562/1541, loss: 0.3174018859863281 2023-01-23 10:31:39.566072: step: 1564/1541, loss: 2.0487215518951416 2023-01-23 10:31:40.469837: step: 1566/1541, loss: 0.15278387069702148 2023-01-23 10:31:41.374567: step: 1568/1541, loss: 0.34910863637924194 2023-01-23 10:31:42.277659: step: 1570/1541, loss: 0.059337861835956573 2023-01-23 10:31:43.161046: step: 1572/1541, loss: 0.09819545596837997 2023-01-23 10:31:44.032145: step: 1574/1541, loss: 0.5835196375846863 2023-01-23 10:31:44.898584: step: 1576/1541, loss: 0.05767235904932022 2023-01-23 10:31:45.747617: step: 1578/1541, loss: 0.08954896777868271 2023-01-23 10:31:46.648783: step: 1580/1541, loss: 0.015701550990343094 2023-01-23 10:31:47.565297: step: 1582/1541, loss: 0.5146732926368713 2023-01-23 10:31:48.426035: step: 1584/1541, loss: 0.03185788542032242 2023-01-23 10:31:49.311666: step: 1586/1541, loss: 0.048560138791799545 2023-01-23 10:31:50.230789: step: 1588/1541, loss: 0.04929432272911072 2023-01-23 10:31:51.110723: step: 1590/1541, loss: 0.00946519523859024 2023-01-23 10:31:52.005007: step: 1592/1541, loss: 0.1237887293100357 2023-01-23 10:31:52.907830: step: 1594/1541, loss: 0.18884657323360443 2023-01-23 10:31:53.787331: step: 1596/1541, loss: 0.05676424130797386 2023-01-23 10:31:54.675175: step: 1598/1541, loss: 0.15673492848873138 2023-01-23 10:31:55.565215: step: 1600/1541, loss: 0.029638025909662247 2023-01-23 10:31:56.460132: step: 1602/1541, loss: 0.043364740908145905 2023-01-23 10:31:57.310199: step: 1604/1541, loss: 0.12909665703773499 2023-01-23 10:31:58.151023: step: 1606/1541, loss: 0.020992817357182503 2023-01-23 10:31:59.060085: step: 1608/1541, loss: 0.3998163640499115 2023-01-23 10:31:59.936192: step: 1610/1541, loss: 0.6472717523574829 2023-01-23 10:32:00.838142: step: 1612/1541, loss: 0.3278971016407013 2023-01-23 10:32:01.711188: step: 1614/1541, loss: 0.08448003977537155 2023-01-23 10:32:02.600497: step: 1616/1541, loss: 0.03287851810455322 2023-01-23 10:32:03.515844: step: 1618/1541, loss: 0.31145063042640686 2023-01-23 10:32:04.481207: step: 1620/1541, loss: 0.1156843826174736 2023-01-23 10:32:05.354738: step: 1622/1541, loss: 0.3001474440097809 2023-01-23 10:32:06.216340: step: 1624/1541, loss: 0.32307514548301697 2023-01-23 10:32:07.095628: step: 1626/1541, loss: 0.1796121746301651 2023-01-23 10:32:07.983453: step: 1628/1541, loss: 0.8646010756492615 2023-01-23 10:32:08.959668: step: 1630/1541, loss: 0.18045002222061157 2023-01-23 10:32:09.886285: step: 1632/1541, loss: 0.14101271331310272 2023-01-23 10:32:10.796886: step: 1634/1541, loss: 0.09442520886659622 2023-01-23 10:32:11.697152: step: 1636/1541, loss: 0.25582483410835266 2023-01-23 10:32:12.654178: step: 1638/1541, loss: 0.17469792068004608 2023-01-23 10:32:13.570569: step: 1640/1541, loss: 0.2379375696182251 2023-01-23 10:32:14.472136: step: 1642/1541, loss: 0.08759026229381561 2023-01-23 10:32:15.341397: step: 1644/1541, loss: 0.028326183557510376 2023-01-23 10:32:16.237273: step: 1646/1541, loss: 0.1427995264530182 2023-01-23 10:32:17.177075: step: 1648/1541, loss: 0.0644705519080162 2023-01-23 10:32:18.064998: step: 1650/1541, loss: 0.21090646088123322 2023-01-23 10:32:18.966181: step: 1652/1541, loss: 0.1953275352716446 2023-01-23 10:32:19.888060: step: 1654/1541, loss: 0.07994583249092102 2023-01-23 10:32:20.768166: step: 1656/1541, loss: 0.27093735337257385 2023-01-23 10:32:21.629935: step: 1658/1541, loss: 0.03957191854715347 2023-01-23 10:32:22.515531: step: 1660/1541, loss: 3.8168282508850098 2023-01-23 10:32:23.406758: step: 1662/1541, loss: 0.14362230896949768 2023-01-23 10:32:24.289686: step: 1664/1541, loss: 0.2774057686328888 2023-01-23 10:32:25.224063: step: 1666/1541, loss: 0.28511953353881836 2023-01-23 10:32:26.091447: step: 1668/1541, loss: 0.040212180465459824 2023-01-23 10:32:26.969519: step: 1670/1541, loss: 0.16391880810260773 2023-01-23 10:32:27.815009: step: 1672/1541, loss: 0.33228087425231934 2023-01-23 10:32:28.655449: step: 1674/1541, loss: 0.07690548151731491 2023-01-23 10:32:29.568110: step: 1676/1541, loss: 0.0947771891951561 2023-01-23 10:32:30.445780: step: 1678/1541, loss: 0.05082401633262634 2023-01-23 10:32:31.301146: step: 1680/1541, loss: 0.23457062244415283 2023-01-23 10:32:32.207489: step: 1682/1541, loss: 0.13088758289813995 2023-01-23 10:32:33.091220: step: 1684/1541, loss: 0.08633921295404434 2023-01-23 10:32:33.958715: step: 1686/1541, loss: 0.22430241107940674 2023-01-23 10:32:34.854551: step: 1688/1541, loss: 0.10108505934476852 2023-01-23 10:32:35.706588: step: 1690/1541, loss: 0.3602122366428375 2023-01-23 10:32:36.639616: step: 1692/1541, loss: 0.5863536596298218 2023-01-23 10:32:37.582460: step: 1694/1541, loss: 0.008074907585978508 2023-01-23 10:32:38.494152: step: 1696/1541, loss: 0.029456738382577896 2023-01-23 10:32:39.378375: step: 1698/1541, loss: 0.4836869239807129 2023-01-23 10:32:40.280423: step: 1700/1541, loss: 0.13372984528541565 2023-01-23 10:32:41.148437: step: 1702/1541, loss: 0.030627857893705368 2023-01-23 10:32:42.062434: step: 1704/1541, loss: 0.0451931431889534 2023-01-23 10:32:42.934204: step: 1706/1541, loss: 0.9935059547424316 2023-01-23 10:32:43.844441: step: 1708/1541, loss: 0.10774924606084824 2023-01-23 10:32:44.772698: step: 1710/1541, loss: 0.10778547823429108 2023-01-23 10:32:45.654715: step: 1712/1541, loss: 0.10355938971042633 2023-01-23 10:32:46.532355: step: 1714/1541, loss: 0.12632600963115692 2023-01-23 10:32:47.477068: step: 1716/1541, loss: 0.0872010812163353 2023-01-23 10:32:48.362287: step: 1718/1541, loss: 0.13336163759231567 2023-01-23 10:32:49.237827: step: 1720/1541, loss: 0.09363280236721039 2023-01-23 10:32:50.122603: step: 1722/1541, loss: 0.026386229321360588 2023-01-23 10:32:50.984237: step: 1724/1541, loss: 0.22720225155353546 2023-01-23 10:32:51.868358: step: 1726/1541, loss: 0.01971159502863884 2023-01-23 10:32:52.782241: step: 1728/1541, loss: 0.3299814462661743 2023-01-23 10:32:53.713422: step: 1730/1541, loss: 0.014349807985126972 2023-01-23 10:32:54.646959: step: 1732/1541, loss: 0.059866614639759064 2023-01-23 10:32:55.502046: step: 1734/1541, loss: 0.09686598181724548 2023-01-23 10:32:56.395254: step: 1736/1541, loss: 0.044584859162569046 2023-01-23 10:32:57.259102: step: 1738/1541, loss: 0.016386045143008232 2023-01-23 10:32:58.172537: step: 1740/1541, loss: 0.1327511966228485 2023-01-23 10:32:59.080156: step: 1742/1541, loss: 0.08005210757255554 2023-01-23 10:32:59.973330: step: 1744/1541, loss: 0.18742261826992035 2023-01-23 10:33:00.825459: step: 1746/1541, loss: 0.05800739675760269 2023-01-23 10:33:01.683698: step: 1748/1541, loss: 0.023656215518712997 2023-01-23 10:33:02.654465: step: 1750/1541, loss: 0.28569531440734863 2023-01-23 10:33:03.676377: step: 1752/1541, loss: 0.2890245318412781 2023-01-23 10:33:04.563658: step: 1754/1541, loss: 0.38551709055900574 2023-01-23 10:33:05.470227: step: 1756/1541, loss: 0.042721107602119446 2023-01-23 10:33:06.365774: step: 1758/1541, loss: 0.010523466393351555 2023-01-23 10:33:07.251141: step: 1760/1541, loss: 0.007065827026963234 2023-01-23 10:33:08.093732: step: 1762/1541, loss: 0.24371999502182007 2023-01-23 10:33:08.964098: step: 1764/1541, loss: 0.09257182478904724 2023-01-23 10:33:09.871226: step: 1766/1541, loss: 0.21799080073833466 2023-01-23 10:33:10.747242: step: 1768/1541, loss: 0.12827712297439575 2023-01-23 10:33:11.787417: step: 1770/1541, loss: 0.32699739933013916 2023-01-23 10:33:12.654582: step: 1772/1541, loss: 0.25720497965812683 2023-01-23 10:33:13.533698: step: 1774/1541, loss: 0.5896942019462585 2023-01-23 10:33:14.440025: step: 1776/1541, loss: 0.12751826643943787 2023-01-23 10:33:15.366849: step: 1778/1541, loss: 0.1918625682592392 2023-01-23 10:33:16.208875: step: 1780/1541, loss: 0.03706405311822891 2023-01-23 10:33:17.107639: step: 1782/1541, loss: 0.1167624294757843 2023-01-23 10:33:17.974654: step: 1784/1541, loss: 0.053246915340423584 2023-01-23 10:33:18.868529: step: 1786/1541, loss: 0.18589477241039276 2023-01-23 10:33:19.782822: step: 1788/1541, loss: 0.054203663021326065 2023-01-23 10:33:20.652698: step: 1790/1541, loss: 0.1837528496980667 2023-01-23 10:33:21.540236: step: 1792/1541, loss: 0.09840534627437592 2023-01-23 10:33:22.490180: step: 1794/1541, loss: 0.15015584230422974 2023-01-23 10:33:23.383833: step: 1796/1541, loss: 0.0215507373213768 2023-01-23 10:33:24.282063: step: 1798/1541, loss: 0.16009022295475006 2023-01-23 10:33:25.133858: step: 1800/1541, loss: 0.027338160201907158 2023-01-23 10:33:26.022039: step: 1802/1541, loss: 0.6864137053489685 2023-01-23 10:33:27.016613: step: 1804/1541, loss: 0.1857345551252365 2023-01-23 10:33:27.876789: step: 1806/1541, loss: 0.0597144216299057 2023-01-23 10:33:28.733896: step: 1808/1541, loss: 0.045810457319021225 2023-01-23 10:33:29.583198: step: 1810/1541, loss: 0.36092591285705566 2023-01-23 10:33:30.461715: step: 1812/1541, loss: 0.18313060700893402 2023-01-23 10:33:31.404295: step: 1814/1541, loss: 0.02433173917233944 2023-01-23 10:33:32.327053: step: 1816/1541, loss: 0.12255202233791351 2023-01-23 10:33:33.270040: step: 1818/1541, loss: 0.7798082232475281 2023-01-23 10:33:34.159582: step: 1820/1541, loss: 0.2237403690814972 2023-01-23 10:33:35.042704: step: 1822/1541, loss: 0.05885908007621765 2023-01-23 10:33:35.916806: step: 1824/1541, loss: 0.10856591910123825 2023-01-23 10:33:36.832035: step: 1826/1541, loss: 0.12818895280361176 2023-01-23 10:33:37.746431: step: 1828/1541, loss: 0.10437163710594177 2023-01-23 10:33:38.617370: step: 1830/1541, loss: 0.06237795576453209 2023-01-23 10:33:39.456644: step: 1832/1541, loss: 0.1790468543767929 2023-01-23 10:33:40.368371: step: 1834/1541, loss: 0.0651276558637619 2023-01-23 10:33:41.240461: step: 1836/1541, loss: 0.06852492690086365 2023-01-23 10:33:42.123961: step: 1838/1541, loss: 0.248980313539505 2023-01-23 10:33:43.027563: step: 1840/1541, loss: 0.033822398632764816 2023-01-23 10:33:43.922757: step: 1842/1541, loss: 0.36542224884033203 2023-01-23 10:33:44.812346: step: 1844/1541, loss: 0.033804215490818024 2023-01-23 10:33:45.665229: step: 1846/1541, loss: 0.15720638632774353 2023-01-23 10:33:46.565655: step: 1848/1541, loss: 0.034060630947351456 2023-01-23 10:33:47.449457: step: 1850/1541, loss: 0.2273608148097992 2023-01-23 10:33:48.332851: step: 1852/1541, loss: 0.01840572990477085 2023-01-23 10:33:49.284846: step: 1854/1541, loss: 0.14514435827732086 2023-01-23 10:33:50.178805: step: 1856/1541, loss: 0.1835574060678482 2023-01-23 10:33:51.101621: step: 1858/1541, loss: 0.11551530659198761 2023-01-23 10:33:51.996792: step: 1860/1541, loss: 0.023958472535014153 2023-01-23 10:33:52.916966: step: 1862/1541, loss: 0.07831842452287674 2023-01-23 10:33:53.771435: step: 1864/1541, loss: 0.10508225858211517 2023-01-23 10:33:54.667770: step: 1866/1541, loss: 0.21301540732383728 2023-01-23 10:33:55.565409: step: 1868/1541, loss: 0.09070944786071777 2023-01-23 10:33:56.408896: step: 1870/1541, loss: 0.056368354707956314 2023-01-23 10:33:57.301399: step: 1872/1541, loss: 0.21682485938072205 2023-01-23 10:33:58.189249: step: 1874/1541, loss: 0.018194716423749924 2023-01-23 10:33:59.095927: step: 1876/1541, loss: 0.2898074984550476 2023-01-23 10:34:00.015909: step: 1878/1541, loss: 0.17972241342067719 2023-01-23 10:34:00.892425: step: 1880/1541, loss: 0.10274043679237366 2023-01-23 10:34:01.778668: step: 1882/1541, loss: 0.1999291181564331 2023-01-23 10:34:02.697911: step: 1884/1541, loss: 0.21457800269126892 2023-01-23 10:34:03.567264: step: 1886/1541, loss: 0.23905432224273682 2023-01-23 10:34:04.504401: step: 1888/1541, loss: 0.05300121381878853 2023-01-23 10:34:05.427101: step: 1890/1541, loss: 0.08843348920345306 2023-01-23 10:34:06.311254: step: 1892/1541, loss: 0.25357696413993835 2023-01-23 10:34:07.201240: step: 1894/1541, loss: 0.09537570923566818 2023-01-23 10:34:08.076119: step: 1896/1541, loss: 0.1543940305709839 2023-01-23 10:34:08.955864: step: 1898/1541, loss: 0.021764766424894333 2023-01-23 10:34:09.818660: step: 1900/1541, loss: 0.10704145580530167 2023-01-23 10:34:10.674546: step: 1902/1541, loss: 0.1731475293636322 2023-01-23 10:34:11.567909: step: 1904/1541, loss: 0.04512357711791992 2023-01-23 10:34:12.448907: step: 1906/1541, loss: 0.6794736385345459 2023-01-23 10:34:13.331465: step: 1908/1541, loss: 0.02782457321882248 2023-01-23 10:34:14.234999: step: 1910/1541, loss: 0.09561868757009506 2023-01-23 10:34:15.102238: step: 1912/1541, loss: 1.1692856550216675 2023-01-23 10:34:16.016882: step: 1914/1541, loss: 0.6289384961128235 2023-01-23 10:34:16.913509: step: 1916/1541, loss: 0.6783950328826904 2023-01-23 10:34:17.829733: step: 1918/1541, loss: 0.020750684663653374 2023-01-23 10:34:18.698122: step: 1920/1541, loss: 0.11627033352851868 2023-01-23 10:34:19.609297: step: 1922/1541, loss: 1.7380263805389404 2023-01-23 10:34:20.517792: step: 1924/1541, loss: 0.057931069284677505 2023-01-23 10:34:21.466100: step: 1926/1541, loss: 0.029674936085939407 2023-01-23 10:34:22.354493: step: 1928/1541, loss: 0.5407102108001709 2023-01-23 10:34:23.237747: step: 1930/1541, loss: 0.04512279853224754 2023-01-23 10:34:24.110502: step: 1932/1541, loss: 0.8454629182815552 2023-01-23 10:34:24.998341: step: 1934/1541, loss: 0.21461157500743866 2023-01-23 10:34:25.859033: step: 1936/1541, loss: 0.04850923642516136 2023-01-23 10:34:26.748861: step: 1938/1541, loss: 0.07997632771730423 2023-01-23 10:34:27.645607: step: 1940/1541, loss: 0.09280302375555038 2023-01-23 10:34:28.558383: step: 1942/1541, loss: 0.03174756094813347 2023-01-23 10:34:29.427209: step: 1944/1541, loss: 0.18100261688232422 2023-01-23 10:34:30.294667: step: 1946/1541, loss: 0.22390492260456085 2023-01-23 10:34:31.229059: step: 1948/1541, loss: 0.1186608150601387 2023-01-23 10:34:32.191980: step: 1950/1541, loss: 0.16017331182956696 2023-01-23 10:34:33.094099: step: 1952/1541, loss: 0.06751647591590881 2023-01-23 10:34:33.954500: step: 1954/1541, loss: 0.02190420590341091 2023-01-23 10:34:34.858644: step: 1956/1541, loss: 0.10950466245412827 2023-01-23 10:34:35.767104: step: 1958/1541, loss: 0.061707355082035065 2023-01-23 10:34:36.634918: step: 1960/1541, loss: 0.2969135642051697 2023-01-23 10:34:37.476224: step: 1962/1541, loss: 0.013981991447508335 2023-01-23 10:34:38.352421: step: 1964/1541, loss: 0.028514685109257698 2023-01-23 10:34:39.257021: step: 1966/1541, loss: 0.10268010199069977 2023-01-23 10:34:40.133584: step: 1968/1541, loss: 0.05180021747946739 2023-01-23 10:34:41.026731: step: 1970/1541, loss: 0.12167425453662872 2023-01-23 10:34:41.941657: step: 1972/1541, loss: 0.047857534140348434 2023-01-23 10:34:42.843836: step: 1974/1541, loss: 0.07057168334722519 2023-01-23 10:34:43.753326: step: 1976/1541, loss: 0.14307956397533417 2023-01-23 10:34:44.637284: step: 1978/1541, loss: 0.07973749935626984 2023-01-23 10:34:45.542910: step: 1980/1541, loss: 0.39232778549194336 2023-01-23 10:34:46.450673: step: 1982/1541, loss: 0.334076851606369 2023-01-23 10:34:47.313220: step: 1984/1541, loss: 0.30383604764938354 2023-01-23 10:34:48.212149: step: 1986/1541, loss: 0.1660516858100891 2023-01-23 10:34:49.071370: step: 1988/1541, loss: 0.09044912457466125 2023-01-23 10:34:49.922588: step: 1990/1541, loss: 0.24847723543643951 2023-01-23 10:34:50.823626: step: 1992/1541, loss: 0.13434572517871857 2023-01-23 10:34:51.697590: step: 1994/1541, loss: 0.3178328573703766 2023-01-23 10:34:52.563932: step: 1996/1541, loss: 0.1034507304430008 2023-01-23 10:34:53.450861: step: 1998/1541, loss: 0.2310989797115326 2023-01-23 10:34:54.419272: step: 2000/1541, loss: 0.5100858211517334 2023-01-23 10:34:55.276414: step: 2002/1541, loss: 0.20874811708927155 2023-01-23 10:34:56.264877: step: 2004/1541, loss: 4.591323375701904 2023-01-23 10:34:57.159602: step: 2006/1541, loss: 0.18812315165996552 2023-01-23 10:34:58.031199: step: 2008/1541, loss: 0.20369553565979004 2023-01-23 10:34:58.970340: step: 2010/1541, loss: 0.04757223278284073 2023-01-23 10:34:59.847989: step: 2012/1541, loss: 0.34593677520751953 2023-01-23 10:35:00.723488: step: 2014/1541, loss: 0.15419575572013855 2023-01-23 10:35:01.575771: step: 2016/1541, loss: 0.7226650714874268 2023-01-23 10:35:02.534347: step: 2018/1541, loss: 0.07582173496484756 2023-01-23 10:35:03.466334: step: 2020/1541, loss: 0.22720608115196228 2023-01-23 10:35:04.399654: step: 2022/1541, loss: 0.3159225583076477 2023-01-23 10:35:05.287759: step: 2024/1541, loss: 0.13455086946487427 2023-01-23 10:35:06.175420: step: 2026/1541, loss: 0.06649328768253326 2023-01-23 10:35:07.033025: step: 2028/1541, loss: 0.04098092392086983 2023-01-23 10:35:07.963067: step: 2030/1541, loss: 0.02037687972187996 2023-01-23 10:35:08.817565: step: 2032/1541, loss: 0.2834843397140503 2023-01-23 10:35:09.749938: step: 2034/1541, loss: 0.10044553875923157 2023-01-23 10:35:10.616735: step: 2036/1541, loss: 0.6874468326568604 2023-01-23 10:35:11.574592: step: 2038/1541, loss: 0.029678350314497948 2023-01-23 10:35:12.445410: step: 2040/1541, loss: 0.03935292735695839 2023-01-23 10:35:13.335596: step: 2042/1541, loss: 0.03938044235110283 2023-01-23 10:35:14.218100: step: 2044/1541, loss: 0.03032221645116806 2023-01-23 10:35:15.137913: step: 2046/1541, loss: 0.0063830153085291386 2023-01-23 10:35:16.040864: step: 2048/1541, loss: 0.24354249238967896 2023-01-23 10:35:16.895151: step: 2050/1541, loss: 0.008546628057956696 2023-01-23 10:35:17.741289: step: 2052/1541, loss: 0.04788954183459282 2023-01-23 10:35:18.635333: step: 2054/1541, loss: 0.14069928228855133 2023-01-23 10:35:19.551685: step: 2056/1541, loss: 0.15698346495628357 2023-01-23 10:35:20.447686: step: 2058/1541, loss: 0.12243714183568954 2023-01-23 10:35:21.424776: step: 2060/1541, loss: 0.0057556768879294395 2023-01-23 10:35:22.345982: step: 2062/1541, loss: 0.030207237228751183 2023-01-23 10:35:23.257356: step: 2064/1541, loss: 0.05232003703713417 2023-01-23 10:35:24.114147: step: 2066/1541, loss: 2.3932852745056152 2023-01-23 10:35:25.025811: step: 2068/1541, loss: 0.019977103918790817 2023-01-23 10:35:25.926582: step: 2070/1541, loss: 0.08603954315185547 2023-01-23 10:35:26.802462: step: 2072/1541, loss: 0.001842754427343607 2023-01-23 10:35:27.728076: step: 2074/1541, loss: 0.20129643380641937 2023-01-23 10:35:28.572648: step: 2076/1541, loss: 0.2898496389389038 2023-01-23 10:35:29.473057: step: 2078/1541, loss: 0.12811990082263947 2023-01-23 10:35:30.350465: step: 2080/1541, loss: 0.0038345034699887037 2023-01-23 10:35:31.264537: step: 2082/1541, loss: 0.3950558602809906 2023-01-23 10:35:32.173715: step: 2084/1541, loss: 0.7707599401473999 2023-01-23 10:35:33.043478: step: 2086/1541, loss: 0.2549678087234497 2023-01-23 10:35:34.053493: step: 2088/1541, loss: 0.10802653431892395 2023-01-23 10:35:34.960296: step: 2090/1541, loss: 0.6372095346450806 2023-01-23 10:35:35.884865: step: 2092/1541, loss: 0.17429517209529877 2023-01-23 10:35:36.738081: step: 2094/1541, loss: 0.3807503879070282 2023-01-23 10:35:37.651386: step: 2096/1541, loss: 1.0657784938812256 2023-01-23 10:35:38.549547: step: 2098/1541, loss: 0.3678096830844879 2023-01-23 10:35:39.418112: step: 2100/1541, loss: 0.7693429589271545 2023-01-23 10:35:40.340782: step: 2102/1541, loss: 0.05500901862978935 2023-01-23 10:35:41.230374: step: 2104/1541, loss: 0.6463882923126221 2023-01-23 10:35:42.231519: step: 2106/1541, loss: 0.1657414585351944 2023-01-23 10:35:43.121248: step: 2108/1541, loss: 0.045462097972631454 2023-01-23 10:35:44.056436: step: 2110/1541, loss: 0.389660120010376 2023-01-23 10:35:44.958733: step: 2112/1541, loss: 0.17090782523155212 2023-01-23 10:35:45.932382: step: 2114/1541, loss: 0.10288241505622864 2023-01-23 10:35:46.927641: step: 2116/1541, loss: 0.34396982192993164 2023-01-23 10:35:47.816633: step: 2118/1541, loss: 0.0312613844871521 2023-01-23 10:35:48.713451: step: 2120/1541, loss: 0.2414613515138626 2023-01-23 10:35:49.612943: step: 2122/1541, loss: 0.26414817571640015 2023-01-23 10:35:50.537359: step: 2124/1541, loss: 0.18485939502716064 2023-01-23 10:35:51.437751: step: 2126/1541, loss: 0.23327675461769104 2023-01-23 10:35:52.345238: step: 2128/1541, loss: 0.756580650806427 2023-01-23 10:35:53.239019: step: 2130/1541, loss: 0.10533550381660461 2023-01-23 10:35:54.101536: step: 2132/1541, loss: 0.6182387471199036 2023-01-23 10:35:54.965920: step: 2134/1541, loss: 0.13526283204555511 2023-01-23 10:35:55.845810: step: 2136/1541, loss: 0.3153495490550995 2023-01-23 10:35:56.814177: step: 2138/1541, loss: 0.32492196559906006 2023-01-23 10:35:57.725261: step: 2140/1541, loss: 0.154871866106987 2023-01-23 10:35:58.614442: step: 2142/1541, loss: 0.30859190225601196 2023-01-23 10:35:59.451955: step: 2144/1541, loss: 0.16343821585178375 2023-01-23 10:36:00.378699: step: 2146/1541, loss: 0.0933031290769577 2023-01-23 10:36:01.258360: step: 2148/1541, loss: 0.19948625564575195 2023-01-23 10:36:02.219076: step: 2150/1541, loss: 0.17007264494895935 2023-01-23 10:36:03.131621: step: 2152/1541, loss: 0.06925511360168457 2023-01-23 10:36:04.086321: step: 2154/1541, loss: 0.1473819613456726 2023-01-23 10:36:04.997366: step: 2156/1541, loss: 0.32080981135368347 2023-01-23 10:36:05.862796: step: 2158/1541, loss: 0.0767010971903801 2023-01-23 10:36:06.749980: step: 2160/1541, loss: 0.05634038895368576 2023-01-23 10:36:07.752243: step: 2162/1541, loss: 0.7358778119087219 2023-01-23 10:36:08.687420: step: 2164/1541, loss: 0.019199132919311523 2023-01-23 10:36:09.588652: step: 2166/1541, loss: 2.102672576904297 2023-01-23 10:36:10.524132: step: 2168/1541, loss: 0.07507351785898209 2023-01-23 10:36:11.407651: step: 2170/1541, loss: 0.1320861577987671 2023-01-23 10:36:12.281400: step: 2172/1541, loss: 0.33708563446998596 2023-01-23 10:36:13.200080: step: 2174/1541, loss: 0.0838889628648758 2023-01-23 10:36:14.086683: step: 2176/1541, loss: 0.08628273755311966 2023-01-23 10:36:14.963235: step: 2178/1541, loss: 0.2273932248353958 2023-01-23 10:36:15.871929: step: 2180/1541, loss: 0.42053285241127014 2023-01-23 10:36:16.760432: step: 2182/1541, loss: 0.11954290419816971 2023-01-23 10:36:17.626483: step: 2184/1541, loss: 0.18874584138393402 2023-01-23 10:36:18.522926: step: 2186/1541, loss: 0.30100560188293457 2023-01-23 10:36:19.431743: step: 2188/1541, loss: 0.14519602060317993 2023-01-23 10:36:20.354889: step: 2190/1541, loss: 0.6444390416145325 2023-01-23 10:36:21.277920: step: 2192/1541, loss: 0.07382877916097641 2023-01-23 10:36:22.136614: step: 2194/1541, loss: 0.346584677696228 2023-01-23 10:36:23.041037: step: 2196/1541, loss: 0.01740657165646553 2023-01-23 10:36:23.937966: step: 2198/1541, loss: 0.02579420804977417 2023-01-23 10:36:24.840202: step: 2200/1541, loss: 0.016413917765021324 2023-01-23 10:36:25.735464: step: 2202/1541, loss: 0.6516203284263611 2023-01-23 10:36:26.623850: step: 2204/1541, loss: 0.12141305953264236 2023-01-23 10:36:27.558130: step: 2206/1541, loss: 0.1321866810321808 2023-01-23 10:36:28.430564: step: 2208/1541, loss: 0.13365019857883453 2023-01-23 10:36:29.356730: step: 2210/1541, loss: 0.0435558520257473 2023-01-23 10:36:30.228227: step: 2212/1541, loss: 0.2795894145965576 2023-01-23 10:36:31.091518: step: 2214/1541, loss: 0.018072906881570816 2023-01-23 10:36:31.967880: step: 2216/1541, loss: 0.05606461688876152 2023-01-23 10:36:32.870579: step: 2218/1541, loss: 0.3696991801261902 2023-01-23 10:36:33.757797: step: 2220/1541, loss: 0.04530245438218117 2023-01-23 10:36:34.755273: step: 2222/1541, loss: 0.04361124709248543 2023-01-23 10:36:35.631865: step: 2224/1541, loss: 0.3746035099029541 2023-01-23 10:36:36.553180: step: 2226/1541, loss: 0.2067987322807312 2023-01-23 10:36:37.418525: step: 2228/1541, loss: 0.07568912953138351 2023-01-23 10:36:38.338314: step: 2230/1541, loss: 0.2552425265312195 2023-01-23 10:36:39.219219: step: 2232/1541, loss: 0.09740383177995682 2023-01-23 10:36:40.198522: step: 2234/1541, loss: 0.22707106173038483 2023-01-23 10:36:41.078129: step: 2236/1541, loss: 0.3632408380508423 2023-01-23 10:36:41.958658: step: 2238/1541, loss: 0.4435591697692871 2023-01-23 10:36:42.872927: step: 2240/1541, loss: 0.39585062861442566 2023-01-23 10:36:43.806200: step: 2242/1541, loss: 0.1463829129934311 2023-01-23 10:36:44.743942: step: 2244/1541, loss: 0.03184225410223007 2023-01-23 10:36:45.661868: step: 2246/1541, loss: 0.09908849000930786 2023-01-23 10:36:46.624676: step: 2248/1541, loss: 0.07526238262653351 2023-01-23 10:36:47.487642: step: 2250/1541, loss: 0.1208440512418747 2023-01-23 10:36:48.371623: step: 2252/1541, loss: 0.013575589284300804 2023-01-23 10:36:49.296964: step: 2254/1541, loss: 0.010385642759501934 2023-01-23 10:36:50.162993: step: 2256/1541, loss: 0.04046812281012535 2023-01-23 10:36:51.076878: step: 2258/1541, loss: 0.0626167505979538 2023-01-23 10:36:51.955316: step: 2260/1541, loss: 0.12084142118692398 2023-01-23 10:36:52.826704: step: 2262/1541, loss: 0.1028837040066719 2023-01-23 10:36:53.707116: step: 2264/1541, loss: 0.02749965898692608 2023-01-23 10:36:54.671564: step: 2266/1541, loss: 0.37227731943130493 2023-01-23 10:36:55.573072: step: 2268/1541, loss: 0.04810839518904686 2023-01-23 10:36:56.466938: step: 2270/1541, loss: 0.6485785841941833 2023-01-23 10:36:57.322834: step: 2272/1541, loss: 0.023562606424093246 2023-01-23 10:36:58.201912: step: 2274/1541, loss: 0.11073517799377441 2023-01-23 10:36:59.151069: step: 2276/1541, loss: 0.06254404038190842 2023-01-23 10:37:00.081608: step: 2278/1541, loss: 0.15359194576740265 2023-01-23 10:37:01.002876: step: 2280/1541, loss: 0.8700714111328125 2023-01-23 10:37:01.928863: step: 2282/1541, loss: 0.05263670161366463 2023-01-23 10:37:02.893306: step: 2284/1541, loss: 0.09472418576478958 2023-01-23 10:37:03.758648: step: 2286/1541, loss: 0.043037574738264084 2023-01-23 10:37:04.612671: step: 2288/1541, loss: 0.12844586372375488 2023-01-23 10:37:05.526675: step: 2290/1541, loss: 0.4915510416030884 2023-01-23 10:37:06.393285: step: 2292/1541, loss: 2.674637794494629 2023-01-23 10:37:07.275692: step: 2294/1541, loss: 0.12730026245117188 2023-01-23 10:37:08.151835: step: 2296/1541, loss: 0.10587403178215027 2023-01-23 10:37:09.057001: step: 2298/1541, loss: 0.03837665915489197 2023-01-23 10:37:09.924421: step: 2300/1541, loss: 0.2104591727256775 2023-01-23 10:37:10.810794: step: 2302/1541, loss: 0.48888370394706726 2023-01-23 10:37:11.670560: step: 2304/1541, loss: 0.24149593710899353 2023-01-23 10:37:12.537094: step: 2306/1541, loss: 0.16773898899555206 2023-01-23 10:37:13.416307: step: 2308/1541, loss: 0.02106625959277153 2023-01-23 10:37:14.316001: step: 2310/1541, loss: 0.14955508708953857 2023-01-23 10:37:15.245897: step: 2312/1541, loss: 0.3686622977256775 2023-01-23 10:37:16.182437: step: 2314/1541, loss: 0.14300577342510223 2023-01-23 10:37:17.072402: step: 2316/1541, loss: 0.016812175512313843 2023-01-23 10:37:17.949035: step: 2318/1541, loss: 0.06443284451961517 2023-01-23 10:37:18.809786: step: 2320/1541, loss: 0.6388365030288696 2023-01-23 10:37:19.758407: step: 2322/1541, loss: 0.09369058907032013 2023-01-23 10:37:20.673842: step: 2324/1541, loss: 0.0884099155664444 2023-01-23 10:37:21.579151: step: 2326/1541, loss: 0.3482989966869354 2023-01-23 10:37:22.510478: step: 2328/1541, loss: 0.0434500016272068 2023-01-23 10:37:23.418803: step: 2330/1541, loss: 0.17381985485553741 2023-01-23 10:37:24.384933: step: 2332/1541, loss: 0.03550637885928154 2023-01-23 10:37:25.315475: step: 2334/1541, loss: 0.15743571519851685 2023-01-23 10:37:26.245274: step: 2336/1541, loss: 0.23677940666675568 2023-01-23 10:37:27.182501: step: 2338/1541, loss: 0.08075953274965286 2023-01-23 10:37:28.034552: step: 2340/1541, loss: 0.3272678554058075 2023-01-23 10:37:28.900546: step: 2342/1541, loss: 0.2661368250846863 2023-01-23 10:37:29.785378: step: 2344/1541, loss: 0.3423738181591034 2023-01-23 10:37:30.712176: step: 2346/1541, loss: 0.09111402183771133 2023-01-23 10:37:31.616232: step: 2348/1541, loss: 0.3626060485839844 2023-01-23 10:37:32.559670: step: 2350/1541, loss: 0.04297005012631416 2023-01-23 10:37:33.435312: step: 2352/1541, loss: 0.047244876623153687 2023-01-23 10:37:34.340689: step: 2354/1541, loss: 0.11879436671733856 2023-01-23 10:37:35.195400: step: 2356/1541, loss: 0.11306638270616531 2023-01-23 10:37:36.136335: step: 2358/1541, loss: 0.09207607805728912 2023-01-23 10:37:37.092521: step: 2360/1541, loss: 0.065672867000103 2023-01-23 10:37:37.963749: step: 2362/1541, loss: 0.2260461002588272 2023-01-23 10:37:38.909479: step: 2364/1541, loss: 0.36671268939971924 2023-01-23 10:37:39.809256: step: 2366/1541, loss: 0.2569868266582489 2023-01-23 10:37:40.717949: step: 2368/1541, loss: 0.28892216086387634 2023-01-23 10:37:41.669139: step: 2370/1541, loss: 0.3403250277042389 2023-01-23 10:37:42.531722: step: 2372/1541, loss: 0.12048714607954025 2023-01-23 10:37:43.447303: step: 2374/1541, loss: 0.12853680551052094 2023-01-23 10:37:44.365411: step: 2376/1541, loss: 0.0841226875782013 2023-01-23 10:37:45.323942: step: 2378/1541, loss: 0.08271322399377823 2023-01-23 10:37:46.242606: step: 2380/1541, loss: 0.07384773343801498 2023-01-23 10:37:47.096216: step: 2382/1541, loss: 0.008419192396104336 2023-01-23 10:37:47.985634: step: 2384/1541, loss: 0.15971818566322327 2023-01-23 10:37:48.880670: step: 2386/1541, loss: 0.49582740664482117 2023-01-23 10:37:49.745868: step: 2388/1541, loss: 0.02493889443576336 2023-01-23 10:37:50.662798: step: 2390/1541, loss: 0.09065330773591995 2023-01-23 10:37:51.569072: step: 2392/1541, loss: 0.0854969173669815 2023-01-23 10:37:52.500950: step: 2394/1541, loss: 0.04339500889182091 2023-01-23 10:37:53.433204: step: 2396/1541, loss: 1.5828752517700195 2023-01-23 10:37:54.320017: step: 2398/1541, loss: 0.2111637145280838 2023-01-23 10:37:55.254451: step: 2400/1541, loss: 0.014248465187847614 2023-01-23 10:37:56.162347: step: 2402/1541, loss: 0.1051425114274025 2023-01-23 10:37:57.034364: step: 2404/1541, loss: 0.05756526440382004 2023-01-23 10:37:57.913699: step: 2406/1541, loss: 0.012513633817434311 2023-01-23 10:37:58.769307: step: 2408/1541, loss: 0.2844623923301697 2023-01-23 10:37:59.705837: step: 2410/1541, loss: 1.4038854837417603 2023-01-23 10:38:00.633135: step: 2412/1541, loss: 0.11833234876394272 2023-01-23 10:38:01.531781: step: 2414/1541, loss: 0.1543354094028473 2023-01-23 10:38:02.459646: step: 2416/1541, loss: 0.0501517690718174 2023-01-23 10:38:03.368419: step: 2418/1541, loss: 0.4236907362937927 2023-01-23 10:38:04.241519: step: 2420/1541, loss: 0.09826406836509705 2023-01-23 10:38:05.157193: step: 2422/1541, loss: 0.17086070775985718 2023-01-23 10:38:06.033446: step: 2424/1541, loss: 0.19636444747447968 2023-01-23 10:38:06.896418: step: 2426/1541, loss: 0.21491697430610657 2023-01-23 10:38:07.846003: step: 2428/1541, loss: 0.32982075214385986 2023-01-23 10:38:08.746518: step: 2430/1541, loss: 0.03573883697390556 2023-01-23 10:38:09.709055: step: 2432/1541, loss: 0.11050842702388763 2023-01-23 10:38:10.593151: step: 2434/1541, loss: 0.01666020229458809 2023-01-23 10:38:11.499687: step: 2436/1541, loss: 0.22522328794002533 2023-01-23 10:38:12.432002: step: 2438/1541, loss: 0.038905125111341476 2023-01-23 10:38:13.353182: step: 2440/1541, loss: 0.2304089218378067 2023-01-23 10:38:14.252882: step: 2442/1541, loss: 0.21999016404151917 2023-01-23 10:38:15.168054: step: 2444/1541, loss: 0.045235708355903625 2023-01-23 10:38:16.082559: step: 2446/1541, loss: 0.19293488562107086 2023-01-23 10:38:16.954068: step: 2448/1541, loss: 0.04636353626847267 2023-01-23 10:38:17.851439: step: 2450/1541, loss: 0.1768130362033844 2023-01-23 10:38:18.737119: step: 2452/1541, loss: 0.17173199355602264 2023-01-23 10:38:19.659889: step: 2454/1541, loss: 0.13150475919246674 2023-01-23 10:38:20.564855: step: 2456/1541, loss: 0.4424137473106384 2023-01-23 10:38:21.455972: step: 2458/1541, loss: 0.23313720524311066 2023-01-23 10:38:22.387198: step: 2460/1541, loss: 0.10447514802217484 2023-01-23 10:38:23.307556: step: 2462/1541, loss: 0.23069795966148376 2023-01-23 10:38:24.217437: step: 2464/1541, loss: 0.011395681649446487 2023-01-23 10:38:25.175632: step: 2466/1541, loss: 0.3265540301799774 2023-01-23 10:38:26.021646: step: 2468/1541, loss: 0.45940837264060974 2023-01-23 10:38:26.914998: step: 2470/1541, loss: 0.0662209615111351 2023-01-23 10:38:27.792961: step: 2472/1541, loss: 0.04606667906045914 2023-01-23 10:38:28.646836: step: 2474/1541, loss: 0.20328761637210846 2023-01-23 10:38:29.537485: step: 2476/1541, loss: 0.13175788521766663 2023-01-23 10:38:30.445141: step: 2478/1541, loss: 0.21888604760169983 2023-01-23 10:38:31.306486: step: 2480/1541, loss: 0.04188789427280426 2023-01-23 10:38:32.210127: step: 2482/1541, loss: 0.044141992926597595 2023-01-23 10:38:33.083468: step: 2484/1541, loss: 0.07682576030492783 2023-01-23 10:38:33.932966: step: 2486/1541, loss: 0.16714312136173248 2023-01-23 10:38:34.884350: step: 2488/1541, loss: 0.051498785614967346 2023-01-23 10:38:35.848393: step: 2490/1541, loss: 0.11193541437387466 2023-01-23 10:38:36.766978: step: 2492/1541, loss: 0.2472473382949829 2023-01-23 10:38:37.764255: step: 2494/1541, loss: 0.07895257323980331 2023-01-23 10:38:38.661444: step: 2496/1541, loss: 0.12856628000736237 2023-01-23 10:38:39.609502: step: 2498/1541, loss: 0.05411246418952942 2023-01-23 10:38:40.560392: step: 2500/1541, loss: 0.050949763506650925 2023-01-23 10:38:41.517847: step: 2502/1541, loss: 0.025782793760299683 2023-01-23 10:38:42.434400: step: 2504/1541, loss: 0.21146096289157867 2023-01-23 10:38:43.348524: step: 2506/1541, loss: 0.16675852239131927 2023-01-23 10:38:44.324383: step: 2508/1541, loss: 0.2182817906141281 2023-01-23 10:38:45.296902: step: 2510/1541, loss: 0.13399940729141235 2023-01-23 10:38:46.189502: step: 2512/1541, loss: 0.07191966474056244 2023-01-23 10:38:47.041188: step: 2514/1541, loss: 0.25691646337509155 2023-01-23 10:38:47.930900: step: 2516/1541, loss: 0.10418570786714554 2023-01-23 10:38:48.851324: step: 2518/1541, loss: 0.0889887884259224 2023-01-23 10:38:49.731754: step: 2520/1541, loss: 0.18381646275520325 2023-01-23 10:38:50.639209: step: 2522/1541, loss: 0.016541365534067154 2023-01-23 10:38:51.538922: step: 2524/1541, loss: 0.26749885082244873 2023-01-23 10:38:52.450370: step: 2526/1541, loss: 0.6154404878616333 2023-01-23 10:38:53.345013: step: 2528/1541, loss: 1.652976393699646 2023-01-23 10:38:54.297478: step: 2530/1541, loss: 0.10025474429130554 2023-01-23 10:38:55.264767: step: 2532/1541, loss: 0.05043798312544823 2023-01-23 10:38:56.172659: step: 2534/1541, loss: 0.13366712629795074 2023-01-23 10:38:57.085287: step: 2536/1541, loss: 0.3720283806324005 2023-01-23 10:38:57.974114: step: 2538/1541, loss: 0.016443602740764618 2023-01-23 10:38:58.888930: step: 2540/1541, loss: 0.0163528174161911 2023-01-23 10:38:59.826107: step: 2542/1541, loss: 0.2452394664287567 2023-01-23 10:39:00.795710: step: 2544/1541, loss: 0.13183382153511047 2023-01-23 10:39:01.700004: step: 2546/1541, loss: 0.03589244931936264 2023-01-23 10:39:02.625003: step: 2548/1541, loss: 0.1409449577331543 2023-01-23 10:39:03.513885: step: 2550/1541, loss: 0.11790212243795395 2023-01-23 10:39:04.437525: step: 2552/1541, loss: 0.03627083823084831 2023-01-23 10:39:05.375625: step: 2554/1541, loss: 0.17452161014080048 2023-01-23 10:39:06.240923: step: 2556/1541, loss: 0.0728529691696167 2023-01-23 10:39:07.131753: step: 2558/1541, loss: 0.02772931009531021 2023-01-23 10:39:08.048527: step: 2560/1541, loss: 0.09958522021770477 2023-01-23 10:39:08.897715: step: 2562/1541, loss: 0.03851613029837608 2023-01-23 10:39:09.892007: step: 2564/1541, loss: 0.0406024307012558 2023-01-23 10:39:10.828051: step: 2566/1541, loss: 0.012447052635252476 2023-01-23 10:39:11.711957: step: 2568/1541, loss: 0.2655370831489563 2023-01-23 10:39:12.580772: step: 2570/1541, loss: 0.08612513542175293 2023-01-23 10:39:13.469265: step: 2572/1541, loss: 0.03585223853588104 2023-01-23 10:39:14.382605: step: 2574/1541, loss: 0.21174129843711853 2023-01-23 10:39:15.248334: step: 2576/1541, loss: 0.20869779586791992 2023-01-23 10:39:16.143316: step: 2578/1541, loss: 0.10346987098455429 2023-01-23 10:39:17.047810: step: 2580/1541, loss: 0.20511886477470398 2023-01-23 10:39:17.983324: step: 2582/1541, loss: 0.4976383447647095 2023-01-23 10:39:18.891058: step: 2584/1541, loss: 0.21781933307647705 2023-01-23 10:39:19.758286: step: 2586/1541, loss: 0.3298124372959137 2023-01-23 10:39:20.634676: step: 2588/1541, loss: 0.37151142954826355 2023-01-23 10:39:21.527261: step: 2590/1541, loss: 0.06270250678062439 2023-01-23 10:39:22.465705: step: 2592/1541, loss: 0.1158028244972229 2023-01-23 10:39:23.348731: step: 2594/1541, loss: 0.06454154849052429 2023-01-23 10:39:24.267198: step: 2596/1541, loss: 0.6857270002365112 2023-01-23 10:39:25.142649: step: 2598/1541, loss: 0.06780349463224411 2023-01-23 10:39:26.207624: step: 2600/1541, loss: 0.11585768312215805 2023-01-23 10:39:27.087469: step: 2602/1541, loss: 0.24299630522727966 2023-01-23 10:39:27.980972: step: 2604/1541, loss: 0.044864773750305176 2023-01-23 10:39:28.852812: step: 2606/1541, loss: 0.26784443855285645 2023-01-23 10:39:29.802670: step: 2608/1541, loss: 0.29900529980659485 2023-01-23 10:39:30.712863: step: 2610/1541, loss: 0.04107898101210594 2023-01-23 10:39:31.576861: step: 2612/1541, loss: 0.09446080774068832 2023-01-23 10:39:32.461422: step: 2614/1541, loss: 0.38644522428512573 2023-01-23 10:39:33.345826: step: 2616/1541, loss: 0.45772379636764526 2023-01-23 10:39:34.249476: step: 2618/1541, loss: 0.10996567457914352 2023-01-23 10:39:35.180134: step: 2620/1541, loss: 0.10686217248439789 2023-01-23 10:39:36.067737: step: 2622/1541, loss: 0.0934700295329094 2023-01-23 10:39:36.974582: step: 2624/1541, loss: 0.12706194818019867 2023-01-23 10:39:37.912752: step: 2626/1541, loss: 0.042278267443180084 2023-01-23 10:39:38.784419: step: 2628/1541, loss: 0.03512642905116081 2023-01-23 10:39:39.675034: step: 2630/1541, loss: 0.022731762379407883 2023-01-23 10:39:40.583841: step: 2632/1541, loss: 0.09193282574415207 2023-01-23 10:39:41.473389: step: 2634/1541, loss: 0.04195697605609894 2023-01-23 10:39:42.411991: step: 2636/1541, loss: 0.03743822127580643 2023-01-23 10:39:43.304949: step: 2638/1541, loss: 0.18488232791423798 2023-01-23 10:39:44.235372: step: 2640/1541, loss: 0.12619180977344513 2023-01-23 10:39:45.143463: step: 2642/1541, loss: 0.2072412222623825 2023-01-23 10:39:46.058422: step: 2644/1541, loss: 0.22916722297668457 2023-01-23 10:39:46.941645: step: 2646/1541, loss: 0.014846155419945717 2023-01-23 10:39:47.824198: step: 2648/1541, loss: 0.47581151127815247 2023-01-23 10:39:48.748590: step: 2650/1541, loss: 0.1579306274652481 2023-01-23 10:39:49.672293: step: 2652/1541, loss: 0.22022873163223267 2023-01-23 10:39:50.602131: step: 2654/1541, loss: 0.3446131944656372 2023-01-23 10:39:51.492267: step: 2656/1541, loss: 0.00313621386885643 2023-01-23 10:39:52.399856: step: 2658/1541, loss: 0.01936180144548416 2023-01-23 10:39:53.259397: step: 2660/1541, loss: 0.05645596235990524 2023-01-23 10:39:54.183532: step: 2662/1541, loss: 0.012662317603826523 2023-01-23 10:39:55.097803: step: 2664/1541, loss: 0.6693516373634338 2023-01-23 10:39:55.972990: step: 2666/1541, loss: 0.9347275495529175 2023-01-23 10:39:56.860545: step: 2668/1541, loss: 0.016007496044039726 2023-01-23 10:39:57.761274: step: 2670/1541, loss: 0.010061847046017647 2023-01-23 10:39:58.629179: step: 2672/1541, loss: 0.14194032549858093 2023-01-23 10:39:59.522661: step: 2674/1541, loss: 0.062263648957014084 2023-01-23 10:40:00.403025: step: 2676/1541, loss: 0.16643944382667542 2023-01-23 10:40:01.383010: step: 2678/1541, loss: 0.046798646450042725 2023-01-23 10:40:02.302208: step: 2680/1541, loss: 0.28987565636634827 2023-01-23 10:40:03.194342: step: 2682/1541, loss: 0.24187663197517395 2023-01-23 10:40:04.145957: step: 2684/1541, loss: 0.026262398809194565 2023-01-23 10:40:05.042724: step: 2686/1541, loss: 2.041327953338623 2023-01-23 10:40:05.969947: step: 2688/1541, loss: 0.06213737651705742 2023-01-23 10:40:06.851109: step: 2690/1541, loss: 0.04383985698223114 2023-01-23 10:40:07.802532: step: 2692/1541, loss: 0.12335079908370972 2023-01-23 10:40:08.665489: step: 2694/1541, loss: 0.39521047472953796 2023-01-23 10:40:09.540021: step: 2696/1541, loss: 0.09788631647825241 2023-01-23 10:40:10.427084: step: 2698/1541, loss: 0.17984651029109955 2023-01-23 10:40:11.348716: step: 2700/1541, loss: 0.24953491985797882 2023-01-23 10:40:12.260538: step: 2702/1541, loss: 0.07718539983034134 2023-01-23 10:40:13.140546: step: 2704/1541, loss: 1.4691060781478882 2023-01-23 10:40:14.083100: step: 2706/1541, loss: 0.5555062890052795 2023-01-23 10:40:14.985512: step: 2708/1541, loss: 0.21319474279880524 2023-01-23 10:40:15.881309: step: 2710/1541, loss: 0.20147009193897247 2023-01-23 10:40:16.811035: step: 2712/1541, loss: 0.14731533825397491 2023-01-23 10:40:17.717773: step: 2714/1541, loss: 0.1752624213695526 2023-01-23 10:40:18.609209: step: 2716/1541, loss: 0.05264130234718323 2023-01-23 10:40:19.508784: step: 2718/1541, loss: 0.5228825211524963 2023-01-23 10:40:20.435774: step: 2720/1541, loss: 0.46286648511886597 2023-01-23 10:40:21.324202: step: 2722/1541, loss: 0.32472559809684753 2023-01-23 10:40:22.253154: step: 2724/1541, loss: 0.09450304508209229 2023-01-23 10:40:23.143039: step: 2726/1541, loss: 0.15858009457588196 2023-01-23 10:40:24.013962: step: 2728/1541, loss: 0.16362927854061127 2023-01-23 10:40:24.992034: step: 2730/1541, loss: 0.33416715264320374 2023-01-23 10:40:25.939209: step: 2732/1541, loss: 0.1958514302968979 2023-01-23 10:40:26.853331: step: 2734/1541, loss: 0.2509433925151825 2023-01-23 10:40:27.730897: step: 2736/1541, loss: 0.722019612789154 2023-01-23 10:40:28.620709: step: 2738/1541, loss: 0.6931430101394653 2023-01-23 10:40:29.535414: step: 2740/1541, loss: 1.8421101570129395 2023-01-23 10:40:30.427194: step: 2742/1541, loss: 0.2772493362426758 2023-01-23 10:40:31.335708: step: 2744/1541, loss: 0.2568669319152832 2023-01-23 10:40:32.239578: step: 2746/1541, loss: 0.1304989457130432 2023-01-23 10:40:33.141168: step: 2748/1541, loss: 0.10866190493106842 2023-01-23 10:40:34.032173: step: 2750/1541, loss: 0.10560481250286102 2023-01-23 10:40:34.934288: step: 2752/1541, loss: 0.4025968909263611 2023-01-23 10:40:35.876217: step: 2754/1541, loss: 0.07179494947195053 2023-01-23 10:40:36.795895: step: 2756/1541, loss: 0.20301204919815063 2023-01-23 10:40:37.715010: step: 2758/1541, loss: 0.02148205041885376 2023-01-23 10:40:38.618667: step: 2760/1541, loss: 0.16273871064186096 2023-01-23 10:40:39.487165: step: 2762/1541, loss: 0.06335137039422989 2023-01-23 10:40:40.373444: step: 2764/1541, loss: 2.476182699203491 2023-01-23 10:40:41.276719: step: 2766/1541, loss: 0.10690076649188995 2023-01-23 10:40:42.206740: step: 2768/1541, loss: 0.15413188934326172 2023-01-23 10:40:43.090285: step: 2770/1541, loss: 0.11736240983009338 2023-01-23 10:40:43.959230: step: 2772/1541, loss: 0.11694110929965973 2023-01-23 10:40:44.825236: step: 2774/1541, loss: 0.12992112338542938 2023-01-23 10:40:45.771353: step: 2776/1541, loss: 0.19318953156471252 2023-01-23 10:40:46.736911: step: 2778/1541, loss: 0.0454997681081295 2023-01-23 10:40:47.621240: step: 2780/1541, loss: 0.3428778648376465 2023-01-23 10:40:48.500682: step: 2782/1541, loss: 0.3934381902217865 2023-01-23 10:40:49.413322: step: 2784/1541, loss: 0.013335248455405235 2023-01-23 10:40:50.386422: step: 2786/1541, loss: 0.1996525526046753 2023-01-23 10:40:51.258530: step: 2788/1541, loss: 0.22385668754577637 2023-01-23 10:40:52.126309: step: 2790/1541, loss: 0.9546934366226196 2023-01-23 10:40:52.959537: step: 2792/1541, loss: 0.03952096402645111 2023-01-23 10:40:53.873094: step: 2794/1541, loss: 0.12626753747463226 2023-01-23 10:40:54.808734: step: 2796/1541, loss: 0.11230964958667755 2023-01-23 10:40:55.705200: step: 2798/1541, loss: 0.06195975840091705 2023-01-23 10:40:56.587437: step: 2800/1541, loss: 0.11057469993829727 2023-01-23 10:40:57.477720: step: 2802/1541, loss: 0.08529532700777054 2023-01-23 10:40:58.370657: step: 2804/1541, loss: 0.703787088394165 2023-01-23 10:40:59.258203: step: 2806/1541, loss: 0.015318581834435463 2023-01-23 10:41:00.141878: step: 2808/1541, loss: 0.09661495685577393 2023-01-23 10:41:01.010469: step: 2810/1541, loss: 0.3549388349056244 2023-01-23 10:41:01.887159: step: 2812/1541, loss: 1.7097687721252441 2023-01-23 10:41:02.773670: step: 2814/1541, loss: 0.02069520577788353 2023-01-23 10:41:03.659796: step: 2816/1541, loss: 0.1642676293849945 2023-01-23 10:41:04.578457: step: 2818/1541, loss: 0.2853754460811615 2023-01-23 10:41:05.467773: step: 2820/1541, loss: 0.05121564120054245 2023-01-23 10:41:06.403934: step: 2822/1541, loss: 0.11343833059072495 2023-01-23 10:41:07.279191: step: 2824/1541, loss: 0.17436359822750092 2023-01-23 10:41:08.184257: step: 2826/1541, loss: 0.16846145689487457 2023-01-23 10:41:09.064262: step: 2828/1541, loss: 1.214032769203186 2023-01-23 10:41:09.935284: step: 2830/1541, loss: 0.35413938760757446 2023-01-23 10:41:10.821335: step: 2832/1541, loss: 0.4637499153614044 2023-01-23 10:41:11.764785: step: 2834/1541, loss: 0.4847687780857086 2023-01-23 10:41:12.649899: step: 2836/1541, loss: 0.05850783362984657 2023-01-23 10:41:13.570334: step: 2838/1541, loss: 0.16663603484630585 2023-01-23 10:41:14.438942: step: 2840/1541, loss: 0.028603583574295044 2023-01-23 10:41:15.331896: step: 2842/1541, loss: 0.14672543108463287 2023-01-23 10:41:16.244066: step: 2844/1541, loss: 0.4939769506454468 2023-01-23 10:41:17.197461: step: 2846/1541, loss: 0.633854329586029 2023-01-23 10:41:18.104399: step: 2848/1541, loss: 0.36429649591445923 2023-01-23 10:41:19.015789: step: 2850/1541, loss: 0.2250470221042633 2023-01-23 10:41:19.927540: step: 2852/1541, loss: 0.3990769684314728 2023-01-23 10:41:20.841949: step: 2854/1541, loss: 0.37811028957366943 2023-01-23 10:41:21.732285: step: 2856/1541, loss: 0.1396334171295166 2023-01-23 10:41:22.609389: step: 2858/1541, loss: 0.019712626934051514 2023-01-23 10:41:23.454611: step: 2860/1541, loss: 0.057238154113292694 2023-01-23 10:41:24.350519: step: 2862/1541, loss: 0.1537182331085205 2023-01-23 10:41:25.263423: step: 2864/1541, loss: 0.18832199275493622 2023-01-23 10:41:26.176532: step: 2866/1541, loss: 0.2543492615222931 2023-01-23 10:41:27.089186: step: 2868/1541, loss: 0.052992478013038635 2023-01-23 10:41:28.029315: step: 2870/1541, loss: 0.1796945035457611 2023-01-23 10:41:28.997492: step: 2872/1541, loss: 0.14203138649463654 2023-01-23 10:41:29.891607: step: 2874/1541, loss: 0.20766772329807281 2023-01-23 10:41:30.775278: step: 2876/1541, loss: 3.1331048011779785 2023-01-23 10:41:31.642825: step: 2878/1541, loss: 0.17459897696971893 2023-01-23 10:41:32.585473: step: 2880/1541, loss: 0.23547987639904022 2023-01-23 10:41:33.457158: step: 2882/1541, loss: 0.15319417417049408 2023-01-23 10:41:34.358408: step: 2884/1541, loss: 0.07761432230472565 2023-01-23 10:41:35.237932: step: 2886/1541, loss: 0.15284240245819092 2023-01-23 10:41:36.140415: step: 2888/1541, loss: 0.28276172280311584 2023-01-23 10:41:37.036386: step: 2890/1541, loss: 0.12821462750434875 2023-01-23 10:41:37.970657: step: 2892/1541, loss: 0.053893957287073135 2023-01-23 10:41:38.920291: step: 2894/1541, loss: 0.17347492277622223 2023-01-23 10:41:39.866631: step: 2896/1541, loss: 0.1799631267786026 2023-01-23 10:41:40.914056: step: 2898/1541, loss: 0.12365413457155228 2023-01-23 10:41:41.781564: step: 2900/1541, loss: 0.324857622385025 2023-01-23 10:41:42.691140: step: 2902/1541, loss: 0.3283490240573883 2023-01-23 10:41:43.549393: step: 2904/1541, loss: 0.028699977323412895 2023-01-23 10:41:44.434919: step: 2906/1541, loss: 0.044454578310251236 2023-01-23 10:41:45.374367: step: 2908/1541, loss: 0.18627162277698517 2023-01-23 10:41:46.285337: step: 2910/1541, loss: 0.023251092061400414 2023-01-23 10:41:47.141231: step: 2912/1541, loss: 0.17165470123291016 2023-01-23 10:41:48.023306: step: 2914/1541, loss: 0.02846701443195343 2023-01-23 10:41:49.009724: step: 2916/1541, loss: 0.07649216055870056 2023-01-23 10:41:49.936382: step: 2918/1541, loss: 0.22371432185173035 2023-01-23 10:41:50.795700: step: 2920/1541, loss: 0.09048060327768326 2023-01-23 10:41:51.674301: step: 2922/1541, loss: 0.34771546721458435 2023-01-23 10:41:52.551229: step: 2924/1541, loss: 0.16119582951068878 2023-01-23 10:41:53.436551: step: 2926/1541, loss: 0.056110478937625885 2023-01-23 10:41:54.366183: step: 2928/1541, loss: 0.022275343537330627 2023-01-23 10:41:55.283576: step: 2930/1541, loss: 0.048506539314985275 2023-01-23 10:41:56.164628: step: 2932/1541, loss: 0.3406796157360077 2023-01-23 10:41:57.067485: step: 2934/1541, loss: 0.576016902923584 2023-01-23 10:41:57.936147: step: 2936/1541, loss: 0.12066955119371414 2023-01-23 10:41:58.817428: step: 2938/1541, loss: 0.16448119282722473 2023-01-23 10:41:59.720146: step: 2940/1541, loss: 0.15097510814666748 2023-01-23 10:42:00.678104: step: 2942/1541, loss: 0.5748886466026306 2023-01-23 10:42:01.647926: step: 2944/1541, loss: 0.13439589738845825 2023-01-23 10:42:02.545784: step: 2946/1541, loss: 0.036375436931848526 2023-01-23 10:42:03.526795: step: 2948/1541, loss: 0.2970890402793884 2023-01-23 10:42:04.440948: step: 2950/1541, loss: 0.32997825741767883 2023-01-23 10:42:05.329188: step: 2952/1541, loss: 0.1666489988565445 2023-01-23 10:42:06.195998: step: 2954/1541, loss: 0.318353533744812 2023-01-23 10:42:07.071641: step: 2956/1541, loss: 0.044166307896375656 2023-01-23 10:42:07.997794: step: 2958/1541, loss: 0.09019506722688675 2023-01-23 10:42:08.882353: step: 2960/1541, loss: 0.1654605269432068 2023-01-23 10:42:09.771481: step: 2962/1541, loss: 0.09212372452020645 2023-01-23 10:42:10.680834: step: 2964/1541, loss: 0.18833152949810028 2023-01-23 10:42:11.558351: step: 2966/1541, loss: 0.12611494958400726 2023-01-23 10:42:12.420302: step: 2968/1541, loss: 0.15589800477027893 2023-01-23 10:42:13.275388: step: 2970/1541, loss: 0.020345810800790787 2023-01-23 10:42:14.200106: step: 2972/1541, loss: 0.09018930792808533 2023-01-23 10:42:15.140153: step: 2974/1541, loss: 0.6041339635848999 2023-01-23 10:42:16.026107: step: 2976/1541, loss: 0.11072604358196259 2023-01-23 10:42:16.976509: step: 2978/1541, loss: 0.06611786037683487 2023-01-23 10:42:17.891998: step: 2980/1541, loss: 0.3288634419441223 2023-01-23 10:42:18.818169: step: 2982/1541, loss: 0.1791648417711258 2023-01-23 10:42:19.679126: step: 2984/1541, loss: 0.31855928897857666 2023-01-23 10:42:20.558396: step: 2986/1541, loss: 0.1794430911540985 2023-01-23 10:42:21.420416: step: 2988/1541, loss: 0.2151777744293213 2023-01-23 10:42:22.313869: step: 2990/1541, loss: 0.1854396015405655 2023-01-23 10:42:23.248942: step: 2992/1541, loss: 0.13364249467849731 2023-01-23 10:42:24.166002: step: 2994/1541, loss: 0.053389668464660645 2023-01-23 10:42:25.120380: step: 2996/1541, loss: 0.0640273466706276 2023-01-23 10:42:26.073599: step: 2998/1541, loss: 0.2276516705751419 2023-01-23 10:42:26.945078: step: 3000/1541, loss: 0.34819191694259644 2023-01-23 10:42:27.827394: step: 3002/1541, loss: 0.6735266447067261 2023-01-23 10:42:28.668912: step: 3004/1541, loss: 0.05114869028329849 2023-01-23 10:42:29.547133: step: 3006/1541, loss: 0.4752388894557953 2023-01-23 10:42:30.554911: step: 3008/1541, loss: 0.6706439256668091 2023-01-23 10:42:31.448336: step: 3010/1541, loss: 0.08368094265460968 2023-01-23 10:42:32.351758: step: 3012/1541, loss: 0.18208655714988708 2023-01-23 10:42:33.289830: step: 3014/1541, loss: 0.10254640877246857 2023-01-23 10:42:34.193609: step: 3016/1541, loss: 0.04053717479109764 2023-01-23 10:42:35.128867: step: 3018/1541, loss: 0.06644900143146515 2023-01-23 10:42:36.051896: step: 3020/1541, loss: 0.06219171732664108 2023-01-23 10:42:36.977979: step: 3022/1541, loss: 0.07564560323953629 2023-01-23 10:42:37.881860: step: 3024/1541, loss: 1.1941461563110352 2023-01-23 10:42:38.775232: step: 3026/1541, loss: 0.05160471051931381 2023-01-23 10:42:39.623615: step: 3028/1541, loss: 0.29519179463386536 2023-01-23 10:42:40.476831: step: 3030/1541, loss: 1.3946905136108398 2023-01-23 10:42:41.373094: step: 3032/1541, loss: 0.2538459599018097 2023-01-23 10:42:42.293021: step: 3034/1541, loss: 0.3388415277004242 2023-01-23 10:42:43.195256: step: 3036/1541, loss: 0.5035461783409119 2023-01-23 10:42:44.057880: step: 3038/1541, loss: 0.030686043202877045 2023-01-23 10:42:44.978790: step: 3040/1541, loss: 0.21024906635284424 2023-01-23 10:42:45.851030: step: 3042/1541, loss: 0.08168306946754456 2023-01-23 10:42:46.798622: step: 3044/1541, loss: 0.2085222452878952 2023-01-23 10:42:47.693763: step: 3046/1541, loss: 0.08895839005708694 2023-01-23 10:42:48.613732: step: 3048/1541, loss: 0.06218322739005089 2023-01-23 10:42:49.490679: step: 3050/1541, loss: 0.1689324975013733 2023-01-23 10:42:50.375381: step: 3052/1541, loss: 0.03529232367873192 2023-01-23 10:42:51.313209: step: 3054/1541, loss: 0.04951608553528786 2023-01-23 10:42:52.176471: step: 3056/1541, loss: 0.1367572695016861 2023-01-23 10:42:53.048952: step: 3058/1541, loss: 0.20144762098789215 2023-01-23 10:42:53.949029: step: 3060/1541, loss: 0.3265129327774048 2023-01-23 10:42:54.843991: step: 3062/1541, loss: 0.20587359368801117 2023-01-23 10:42:55.784520: step: 3064/1541, loss: 0.3793768882751465 2023-01-23 10:42:56.659013: step: 3066/1541, loss: 0.1176367923617363 2023-01-23 10:42:57.520577: step: 3068/1541, loss: 0.18818062543869019 2023-01-23 10:42:58.442642: step: 3070/1541, loss: 0.24182966351509094 2023-01-23 10:42:59.310489: step: 3072/1541, loss: 0.07601764798164368 2023-01-23 10:43:00.230499: step: 3074/1541, loss: 0.28450727462768555 2023-01-23 10:43:01.140512: step: 3076/1541, loss: 0.872484028339386 2023-01-23 10:43:02.024797: step: 3078/1541, loss: 0.11029744893312454 2023-01-23 10:43:02.949725: step: 3080/1541, loss: 0.07310435175895691 2023-01-23 10:43:03.848692: step: 3082/1541, loss: 0.06019618362188339 2023-01-23 10:43:04.493696: step: 3084/1541, loss: 0.27826330065727234 ================================================== Loss: 0.220 -------------------- Dev Chinese: {'event': {'p': 0.6248506571087217, 'r': 0.6964047936085219, 'f1': 0.6586901763224181}, 'argument': {'p': 0.6489144528347627, 'r': 0.6569943215081917, 'f1': 0.6529293914447226}, 'combined': 0.43007817597681347, 'epoch': 11} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7156565118366034, 'r': 0.7073076207032561, 'f1': 0.7114575737400191}, 'combined': 0.463413698269192, 'epoch': 11} Dev Korean: {'event': {'p': 0.5696078431372549, 'r': 0.7736351531291611, 'f1': 0.6561264822134386}, 'argument': {'p': 0.6482489491292114, 'r': 0.6659708628464128, 'f1': 0.6569904182799956}, 'combined': 0.4310688119939891, 'epoch': 11} Test Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.7129107352592366, 'r': 0.7181711447604467, 'f1': 0.7155312718320432}, 'combined': 0.4889242085733155, 'epoch': 11} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6526182691107695, 'r': 0.6683796159874069, 'f1': 0.6604049151255139}, 'combined': 0.4571078534787256, 'epoch': 11} Test Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7166244846941695, 'r': 0.713019474710081, 'f1': 0.7148174344806034}, 'combined': 0.49637881064770395, 'epoch': 11} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6380266075388026, 'r': 0.6944891391794046, 'f1': 0.6650616332819721}, 'combined': 0.4888487218995692, 'epoch': 11} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.544949494949495, 'r': 0.38924963924963923, 'f1': 0.4541245791245791}, 'combined': 0.2646080224820382, 'epoch': 11} Russian: {'event': {'p': 0.5588235294117647, 'r': 0.5277777777777778, 'f1': 0.5428571428571428}, 'argument': {'p': 0.5130588428665352, 'r': 0.6204432518386007, 'f1': 0.5616644174538913}, 'combined': 0.30490354090354094, 'epoch': 11} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test for Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6727483627478649, 'r': 0.6595770845615362, 'f1': 0.6660976184773536}, 'combined': 0.4593776679154163, 'epoch': 10} Test for Russian: {'event': {'p': 0.6446886446886447, 'r': 0.738166566806471, 'f1': 0.6882681564245811}, 'argument': {'p': 0.7212116816452776, 'r': 0.7139607203935477, 'f1': 0.717567883903295}, 'combined': 0.4938791245636086, 'epoch': 10} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5730551257490913, 'r': 0.6396894426966601, 'f1': 0.6045416711199205}, 'combined': 0.3109071451473877, 'epoch': 10} ****************************** Epoch: 12 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 10:48:29.044061: step: 2/1541, loss: 0.15065182745456696 2023-01-23 10:48:29.903686: step: 4/1541, loss: 0.04382067918777466 2023-01-23 10:48:30.835018: step: 6/1541, loss: 0.15076927840709686 2023-01-23 10:48:31.797487: step: 8/1541, loss: 0.07498997449874878 2023-01-23 10:48:32.686837: step: 10/1541, loss: 0.09017951041460037 2023-01-23 10:48:33.567678: step: 12/1541, loss: 0.01890479028224945 2023-01-23 10:48:34.439718: step: 14/1541, loss: 0.02525065280497074 2023-01-23 10:48:35.335056: step: 16/1541, loss: 0.12459991872310638 2023-01-23 10:48:36.240733: step: 18/1541, loss: 0.10628648102283478 2023-01-23 10:48:37.145583: step: 20/1541, loss: 0.10529087483882904 2023-01-23 10:48:38.072867: step: 22/1541, loss: 0.20245997607707977 2023-01-23 10:48:39.016791: step: 24/1541, loss: 0.024888083338737488 2023-01-23 10:48:40.006341: step: 26/1541, loss: 0.015108080580830574 2023-01-23 10:48:40.894297: step: 28/1541, loss: 0.03737048804759979 2023-01-23 10:48:41.759223: step: 30/1541, loss: 0.06052624061703682 2023-01-23 10:48:42.697284: step: 32/1541, loss: 0.021492881700396538 2023-01-23 10:48:43.574582: step: 34/1541, loss: 0.003268598113209009 2023-01-23 10:48:44.502033: step: 36/1541, loss: 0.09744513779878616 2023-01-23 10:48:45.391474: step: 38/1541, loss: 0.07638204842805862 2023-01-23 10:48:46.342625: step: 40/1541, loss: 0.018213503062725067 2023-01-23 10:48:47.241685: step: 42/1541, loss: 0.3670114278793335 2023-01-23 10:48:48.118393: step: 44/1541, loss: 0.01645488291978836 2023-01-23 10:48:48.975752: step: 46/1541, loss: 0.02279173955321312 2023-01-23 10:48:49.907265: step: 48/1541, loss: 0.16280490159988403 2023-01-23 10:48:50.861863: step: 50/1541, loss: 0.010593176819384098 2023-01-23 10:48:51.743183: step: 52/1541, loss: 0.07044793665409088 2023-01-23 10:48:52.636048: step: 54/1541, loss: 0.5709701776504517 2023-01-23 10:48:53.568886: step: 56/1541, loss: 0.10085704177618027 2023-01-23 10:48:54.439537: step: 58/1541, loss: 0.19741928577423096 2023-01-23 10:48:55.333416: step: 60/1541, loss: 0.0863085612654686 2023-01-23 10:48:56.300595: step: 62/1541, loss: 0.07834185659885406 2023-01-23 10:48:57.195719: step: 64/1541, loss: 0.04725930094718933 2023-01-23 10:48:58.067654: step: 66/1541, loss: 0.06666389107704163 2023-01-23 10:48:58.964593: step: 68/1541, loss: 0.195498526096344 2023-01-23 10:48:59.850348: step: 70/1541, loss: 0.1526222825050354 2023-01-23 10:49:00.729672: step: 72/1541, loss: 0.15034037828445435 2023-01-23 10:49:01.646188: step: 74/1541, loss: 0.9018265604972839 2023-01-23 10:49:02.556196: step: 76/1541, loss: 0.06696666032075882 2023-01-23 10:49:03.458872: step: 78/1541, loss: 0.08678669482469559 2023-01-23 10:49:04.365422: step: 80/1541, loss: 0.07430197298526764 2023-01-23 10:49:05.301408: step: 82/1541, loss: 0.31086403131484985 2023-01-23 10:49:06.186292: step: 84/1541, loss: 0.08164224773645401 2023-01-23 10:49:07.089333: step: 86/1541, loss: 0.12690140306949615 2023-01-23 10:49:07.981635: step: 88/1541, loss: 0.2700496315956116 2023-01-23 10:49:08.845214: step: 90/1541, loss: 0.23373737931251526 2023-01-23 10:49:09.718264: step: 92/1541, loss: 0.010352280922234058 2023-01-23 10:49:10.633063: step: 94/1541, loss: 0.0372135303914547 2023-01-23 10:49:11.506466: step: 96/1541, loss: 0.18752780556678772 2023-01-23 10:49:12.393915: step: 98/1541, loss: 0.22181972861289978 2023-01-23 10:49:13.284856: step: 100/1541, loss: 0.010268116369843483 2023-01-23 10:49:14.151651: step: 102/1541, loss: 0.040132179856300354 2023-01-23 10:49:15.023586: step: 104/1541, loss: 0.395503968000412 2023-01-23 10:49:15.959739: step: 106/1541, loss: 0.2070065438747406 2023-01-23 10:49:16.855915: step: 108/1541, loss: 0.26902997493743896 2023-01-23 10:49:17.766345: step: 110/1541, loss: 0.007261553313583136 2023-01-23 10:49:18.640333: step: 112/1541, loss: 0.6160948276519775 2023-01-23 10:49:19.581210: step: 114/1541, loss: 0.2558588683605194 2023-01-23 10:49:20.442154: step: 116/1541, loss: 0.19724899530410767 2023-01-23 10:49:21.318721: step: 118/1541, loss: 0.029200924560427666 2023-01-23 10:49:22.171421: step: 120/1541, loss: 0.008605249226093292 2023-01-23 10:49:23.059792: step: 122/1541, loss: 0.041528720408678055 2023-01-23 10:49:23.999369: step: 124/1541, loss: 0.02462008409202099 2023-01-23 10:49:24.888243: step: 126/1541, loss: 0.024628613144159317 2023-01-23 10:49:25.760287: step: 128/1541, loss: 0.08964076638221741 2023-01-23 10:49:26.688404: step: 130/1541, loss: 0.15926238894462585 2023-01-23 10:49:27.597379: step: 132/1541, loss: 0.2865748107433319 2023-01-23 10:49:28.498606: step: 134/1541, loss: 0.9734780788421631 2023-01-23 10:49:29.376982: step: 136/1541, loss: 0.016775574535131454 2023-01-23 10:49:30.263534: step: 138/1541, loss: 0.13844381272792816 2023-01-23 10:49:31.155707: step: 140/1541, loss: 0.9561323523521423 2023-01-23 10:49:32.028558: step: 142/1541, loss: 0.08465319126844406 2023-01-23 10:49:32.930201: step: 144/1541, loss: 0.07238496094942093 2023-01-23 10:49:33.874537: step: 146/1541, loss: 0.09987941384315491 2023-01-23 10:49:34.777017: step: 148/1541, loss: 0.26681917905807495 2023-01-23 10:49:35.654552: step: 150/1541, loss: 0.03938014805316925 2023-01-23 10:49:36.575216: step: 152/1541, loss: 0.18043799698352814 2023-01-23 10:49:37.454403: step: 154/1541, loss: 0.08257608860731125 2023-01-23 10:49:38.363890: step: 156/1541, loss: 0.013223489746451378 2023-01-23 10:49:39.277942: step: 158/1541, loss: 0.06823654472827911 2023-01-23 10:49:40.170410: step: 160/1541, loss: 0.12159869819879532 2023-01-23 10:49:41.093168: step: 162/1541, loss: 0.04383523762226105 2023-01-23 10:49:42.035559: step: 164/1541, loss: 0.06359213590621948 2023-01-23 10:49:42.917284: step: 166/1541, loss: 0.2576562166213989 2023-01-23 10:49:43.798417: step: 168/1541, loss: 0.038126807659864426 2023-01-23 10:49:44.671975: step: 170/1541, loss: 0.10541017353534698 2023-01-23 10:49:45.534949: step: 172/1541, loss: 0.011869655922055244 2023-01-23 10:49:46.448440: step: 174/1541, loss: 0.1293681561946869 2023-01-23 10:49:47.360494: step: 176/1541, loss: 0.14721037447452545 2023-01-23 10:49:48.289081: step: 178/1541, loss: 0.029080454260110855 2023-01-23 10:49:49.198162: step: 180/1541, loss: 0.6988165378570557 2023-01-23 10:49:50.078389: step: 182/1541, loss: 0.005068542901426554 2023-01-23 10:49:50.964022: step: 184/1541, loss: 0.009456931613385677 2023-01-23 10:49:51.867375: step: 186/1541, loss: 0.18500803411006927 2023-01-23 10:49:52.783848: step: 188/1541, loss: 0.21805402636528015 2023-01-23 10:49:53.672415: step: 190/1541, loss: 0.013308183290064335 2023-01-23 10:49:54.563488: step: 192/1541, loss: 0.08171455562114716 2023-01-23 10:49:55.441359: step: 194/1541, loss: 0.1394892930984497 2023-01-23 10:49:56.305990: step: 196/1541, loss: 0.06262443959712982 2023-01-23 10:49:57.202213: step: 198/1541, loss: 0.024190088734030724 2023-01-23 10:49:58.179036: step: 200/1541, loss: 0.11715487390756607 2023-01-23 10:49:59.117324: step: 202/1541, loss: 0.2550458610057831 2023-01-23 10:49:59.991848: step: 204/1541, loss: 0.30250370502471924 2023-01-23 10:50:00.913607: step: 206/1541, loss: 0.03195229917764664 2023-01-23 10:50:01.811394: step: 208/1541, loss: 0.2808311879634857 2023-01-23 10:50:02.734127: step: 210/1541, loss: 0.12176807224750519 2023-01-23 10:50:03.620485: step: 212/1541, loss: 0.1902812272310257 2023-01-23 10:50:04.557335: step: 214/1541, loss: 0.13772070407867432 2023-01-23 10:50:05.458477: step: 216/1541, loss: 0.184225931763649 2023-01-23 10:50:06.323599: step: 218/1541, loss: 0.3893049955368042 2023-01-23 10:50:07.185876: step: 220/1541, loss: 0.07007282972335815 2023-01-23 10:50:08.138612: step: 222/1541, loss: 0.1692684143781662 2023-01-23 10:50:09.015860: step: 224/1541, loss: 0.13737298548221588 2023-01-23 10:50:09.905359: step: 226/1541, loss: 0.005289469845592976 2023-01-23 10:50:10.820778: step: 228/1541, loss: 0.03068738430738449 2023-01-23 10:50:11.743332: step: 230/1541, loss: 2.778165817260742 2023-01-23 10:50:12.697541: step: 232/1541, loss: 0.03871290385723114 2023-01-23 10:50:13.577082: step: 234/1541, loss: 0.3157083988189697 2023-01-23 10:50:14.447606: step: 236/1541, loss: 0.09464024007320404 2023-01-23 10:50:15.378907: step: 238/1541, loss: 0.05526476725935936 2023-01-23 10:50:16.297524: step: 240/1541, loss: 0.2264605015516281 2023-01-23 10:50:17.208119: step: 242/1541, loss: 0.03941202163696289 2023-01-23 10:50:18.105867: step: 244/1541, loss: 0.006902446039021015 2023-01-23 10:50:19.019369: step: 246/1541, loss: 0.1489972174167633 2023-01-23 10:50:19.903007: step: 248/1541, loss: 0.009335796348750591 2023-01-23 10:50:20.832090: step: 250/1541, loss: 0.09338510781526566 2023-01-23 10:50:21.735256: step: 252/1541, loss: 0.06898855417966843 2023-01-23 10:50:22.648973: step: 254/1541, loss: 0.3443358838558197 2023-01-23 10:50:23.533168: step: 256/1541, loss: 0.361397385597229 2023-01-23 10:50:24.419107: step: 258/1541, loss: 0.35662388801574707 2023-01-23 10:50:25.268968: step: 260/1541, loss: 0.022383540868759155 2023-01-23 10:50:26.279029: step: 262/1541, loss: 1.0547776222229004 2023-01-23 10:50:27.255926: step: 264/1541, loss: 0.045674219727516174 2023-01-23 10:50:28.182039: step: 266/1541, loss: 0.1359458714723587 2023-01-23 10:50:29.094991: step: 268/1541, loss: 0.659726083278656 2023-01-23 10:50:29.983131: step: 270/1541, loss: 0.2542397081851959 2023-01-23 10:50:30.898869: step: 272/1541, loss: 0.1484426110982895 2023-01-23 10:50:31.798683: step: 274/1541, loss: 0.1094677746295929 2023-01-23 10:50:32.715519: step: 276/1541, loss: 0.00823646504431963 2023-01-23 10:50:33.630088: step: 278/1541, loss: 0.048113517463207245 2023-01-23 10:50:34.536968: step: 280/1541, loss: 0.15197740495204926 2023-01-23 10:50:35.432754: step: 282/1541, loss: 0.04252784326672554 2023-01-23 10:50:36.326197: step: 284/1541, loss: 0.7038304805755615 2023-01-23 10:50:37.239459: step: 286/1541, loss: 0.21765106916427612 2023-01-23 10:50:38.152390: step: 288/1541, loss: 0.1252509355545044 2023-01-23 10:50:39.020483: step: 290/1541, loss: 0.010140463709831238 2023-01-23 10:50:39.951554: step: 292/1541, loss: 0.0913480818271637 2023-01-23 10:50:40.855360: step: 294/1541, loss: 0.13923341035842896 2023-01-23 10:50:41.739350: step: 296/1541, loss: 0.00922553800046444 2023-01-23 10:50:42.654901: step: 298/1541, loss: 0.11425332725048065 2023-01-23 10:50:43.565454: step: 300/1541, loss: 0.346245139837265 2023-01-23 10:50:44.481044: step: 302/1541, loss: 0.03737778216600418 2023-01-23 10:50:45.404670: step: 304/1541, loss: 0.2057598978281021 2023-01-23 10:50:46.315023: step: 306/1541, loss: 0.15945398807525635 2023-01-23 10:50:47.216475: step: 308/1541, loss: 0.09704114496707916 2023-01-23 10:50:48.104298: step: 310/1541, loss: 0.1398831605911255 2023-01-23 10:50:49.000642: step: 312/1541, loss: 0.4768957197666168 2023-01-23 10:50:49.881677: step: 314/1541, loss: 0.18015499413013458 2023-01-23 10:50:50.772194: step: 316/1541, loss: 0.041717104613780975 2023-01-23 10:50:51.659509: step: 318/1541, loss: 0.054917193949222565 2023-01-23 10:50:52.584715: step: 320/1541, loss: 0.028988324105739594 2023-01-23 10:50:53.503904: step: 322/1541, loss: 0.6107415556907654 2023-01-23 10:50:54.407365: step: 324/1541, loss: 0.04105228930711746 2023-01-23 10:50:55.360292: step: 326/1541, loss: 0.40009331703186035 2023-01-23 10:50:56.235016: step: 328/1541, loss: 1.0549496412277222 2023-01-23 10:50:57.119533: step: 330/1541, loss: 0.1210419237613678 2023-01-23 10:50:58.032135: step: 332/1541, loss: 0.124063640832901 2023-01-23 10:50:58.955886: step: 334/1541, loss: 0.2693631947040558 2023-01-23 10:50:59.898447: step: 336/1541, loss: 0.4031192660331726 2023-01-23 10:51:00.779465: step: 338/1541, loss: 0.15167726576328278 2023-01-23 10:51:01.691371: step: 340/1541, loss: 0.12019933015108109 2023-01-23 10:51:02.645868: step: 342/1541, loss: 0.022477833554148674 2023-01-23 10:51:03.529518: step: 344/1541, loss: 0.27852264046669006 2023-01-23 10:51:04.427138: step: 346/1541, loss: 0.058668531477451324 2023-01-23 10:51:05.304546: step: 348/1541, loss: 0.13938137888908386 2023-01-23 10:51:06.171621: step: 350/1541, loss: 0.10439184308052063 2023-01-23 10:51:07.059358: step: 352/1541, loss: 0.27026456594467163 2023-01-23 10:51:07.958399: step: 354/1541, loss: 0.1020621582865715 2023-01-23 10:51:08.845372: step: 356/1541, loss: 0.3170846104621887 2023-01-23 10:51:09.735503: step: 358/1541, loss: 0.1578509509563446 2023-01-23 10:51:10.624151: step: 360/1541, loss: 0.01810762658715248 2023-01-23 10:51:11.479382: step: 362/1541, loss: 0.2456745058298111 2023-01-23 10:51:12.382451: step: 364/1541, loss: 0.3089207112789154 2023-01-23 10:51:13.277717: step: 366/1541, loss: 0.011866712011396885 2023-01-23 10:51:14.140519: step: 368/1541, loss: 0.07012419402599335 2023-01-23 10:51:15.042168: step: 370/1541, loss: 0.11713861674070358 2023-01-23 10:51:15.923882: step: 372/1541, loss: 0.008758815936744213 2023-01-23 10:51:16.820979: step: 374/1541, loss: 0.12860435247421265 2023-01-23 10:51:17.721364: step: 376/1541, loss: 0.00897594727575779 2023-01-23 10:51:18.633067: step: 378/1541, loss: 0.23118609189987183 2023-01-23 10:51:19.533304: step: 380/1541, loss: 0.0540463849902153 2023-01-23 10:51:20.414259: step: 382/1541, loss: 0.2218458503484726 2023-01-23 10:51:21.313332: step: 384/1541, loss: 0.005237102508544922 2023-01-23 10:51:22.237240: step: 386/1541, loss: 0.14349910616874695 2023-01-23 10:51:23.083263: step: 388/1541, loss: 0.16303378343582153 2023-01-23 10:51:23.950823: step: 390/1541, loss: 0.0027740385849028826 2023-01-23 10:51:24.827208: step: 392/1541, loss: 0.054483722895383835 2023-01-23 10:51:25.712517: step: 394/1541, loss: 0.024934979155659676 2023-01-23 10:51:26.604055: step: 396/1541, loss: 0.019893180578947067 2023-01-23 10:51:27.489233: step: 398/1541, loss: 0.22496356070041656 2023-01-23 10:51:28.394343: step: 400/1541, loss: 0.20192913711071014 2023-01-23 10:51:29.239542: step: 402/1541, loss: 0.010704074054956436 2023-01-23 10:51:30.091153: step: 404/1541, loss: 0.10963454097509384 2023-01-23 10:51:31.024571: step: 406/1541, loss: 0.15917466580867767 2023-01-23 10:51:31.964038: step: 408/1541, loss: 0.12206213921308517 2023-01-23 10:51:32.900489: step: 410/1541, loss: 0.13209931552410126 2023-01-23 10:51:33.855189: step: 412/1541, loss: 0.09905688464641571 2023-01-23 10:51:34.753435: step: 414/1541, loss: 0.16579696536064148 2023-01-23 10:51:35.625226: step: 416/1541, loss: 0.07742316275835037 2023-01-23 10:51:36.539599: step: 418/1541, loss: 0.31920772790908813 2023-01-23 10:51:37.428168: step: 420/1541, loss: 0.14649060368537903 2023-01-23 10:51:38.334955: step: 422/1541, loss: 0.159551739692688 2023-01-23 10:51:39.181010: step: 424/1541, loss: 0.05215613543987274 2023-01-23 10:51:40.083153: step: 426/1541, loss: 0.03519390523433685 2023-01-23 10:51:40.990774: step: 428/1541, loss: 0.13480593264102936 2023-01-23 10:51:41.890277: step: 430/1541, loss: 0.05880807712674141 2023-01-23 10:51:42.823942: step: 432/1541, loss: 0.02251080796122551 2023-01-23 10:51:43.745320: step: 434/1541, loss: 0.3308148980140686 2023-01-23 10:51:44.617186: step: 436/1541, loss: 0.13987132906913757 2023-01-23 10:51:45.475151: step: 438/1541, loss: 0.015392286702990532 2023-01-23 10:51:46.361016: step: 440/1541, loss: 0.11559735238552094 2023-01-23 10:51:47.274837: step: 442/1541, loss: 0.1243959441781044 2023-01-23 10:51:48.173764: step: 444/1541, loss: 0.26089534163475037 2023-01-23 10:51:49.085742: step: 446/1541, loss: 0.21769613027572632 2023-01-23 10:51:49.932799: step: 448/1541, loss: 0.3669629693031311 2023-01-23 10:51:50.834273: step: 450/1541, loss: 0.07330290228128433 2023-01-23 10:51:51.707223: step: 452/1541, loss: 0.20523673295974731 2023-01-23 10:51:52.626806: step: 454/1541, loss: 0.04103780910372734 2023-01-23 10:51:53.509920: step: 456/1541, loss: 0.023998640477657318 2023-01-23 10:51:54.390050: step: 458/1541, loss: 0.03028189204633236 2023-01-23 10:51:55.282548: step: 460/1541, loss: 0.16320474445819855 2023-01-23 10:51:56.179292: step: 462/1541, loss: 0.0627206414937973 2023-01-23 10:51:57.158317: step: 464/1541, loss: 0.021480079740285873 2023-01-23 10:51:58.052949: step: 466/1541, loss: 0.2815363109111786 2023-01-23 10:51:58.931743: step: 468/1541, loss: 0.10633967071771622 2023-01-23 10:51:59.797045: step: 470/1541, loss: 0.03670070320367813 2023-01-23 10:52:00.663314: step: 472/1541, loss: 0.26224204897880554 2023-01-23 10:52:01.563444: step: 474/1541, loss: 0.31739428639411926 2023-01-23 10:52:02.449000: step: 476/1541, loss: 0.016940807923674583 2023-01-23 10:52:03.332599: step: 478/1541, loss: 0.04252977669239044 2023-01-23 10:52:04.216479: step: 480/1541, loss: 0.08361634612083435 2023-01-23 10:52:05.099150: step: 482/1541, loss: 0.022567017003893852 2023-01-23 10:52:06.022563: step: 484/1541, loss: 0.12689240276813507 2023-01-23 10:52:06.891916: step: 486/1541, loss: 0.021899627521634102 2023-01-23 10:52:07.809780: step: 488/1541, loss: 0.23807647824287415 2023-01-23 10:52:08.707381: step: 490/1541, loss: 0.4949873983860016 2023-01-23 10:52:09.590644: step: 492/1541, loss: 0.17002227902412415 2023-01-23 10:52:10.447748: step: 494/1541, loss: 0.09099072962999344 2023-01-23 10:52:11.318484: step: 496/1541, loss: 0.38054871559143066 2023-01-23 10:52:12.206478: step: 498/1541, loss: 0.029964158311486244 2023-01-23 10:52:13.091935: step: 500/1541, loss: 0.07240623235702515 2023-01-23 10:52:14.018651: step: 502/1541, loss: 0.11495683342218399 2023-01-23 10:52:14.886360: step: 504/1541, loss: 0.06306733191013336 2023-01-23 10:52:15.778449: step: 506/1541, loss: 1.5757344961166382 2023-01-23 10:52:16.696731: step: 508/1541, loss: 0.4904182255268097 2023-01-23 10:52:17.589039: step: 510/1541, loss: 0.04572122544050217 2023-01-23 10:52:18.456910: step: 512/1541, loss: 0.008525095880031586 2023-01-23 10:52:19.314581: step: 514/1541, loss: 0.017176762223243713 2023-01-23 10:52:20.237872: step: 516/1541, loss: 0.023776492103934288 2023-01-23 10:52:21.094409: step: 518/1541, loss: 0.0508938767015934 2023-01-23 10:52:21.932342: step: 520/1541, loss: 0.039878442883491516 2023-01-23 10:52:22.858446: step: 522/1541, loss: 0.038452696055173874 2023-01-23 10:52:23.757975: step: 524/1541, loss: 0.14153191447257996 2023-01-23 10:52:24.664428: step: 526/1541, loss: 0.19181539118289948 2023-01-23 10:52:25.570417: step: 528/1541, loss: 0.29634177684783936 2023-01-23 10:52:26.476780: step: 530/1541, loss: 0.03934299945831299 2023-01-23 10:52:27.341090: step: 532/1541, loss: 0.0832786113023758 2023-01-23 10:52:28.255499: step: 534/1541, loss: 0.05633756145834923 2023-01-23 10:52:29.184129: step: 536/1541, loss: 0.043026551604270935 2023-01-23 10:52:30.089333: step: 538/1541, loss: 0.03379720076918602 2023-01-23 10:52:30.980635: step: 540/1541, loss: 0.03317954018712044 2023-01-23 10:52:31.947613: step: 542/1541, loss: 0.22905240952968597 2023-01-23 10:52:32.847823: step: 544/1541, loss: 0.5183790326118469 2023-01-23 10:52:33.690284: step: 546/1541, loss: 0.35606613755226135 2023-01-23 10:52:34.587974: step: 548/1541, loss: 0.28048640489578247 2023-01-23 10:52:35.507891: step: 550/1541, loss: 0.12129808962345123 2023-01-23 10:52:36.364615: step: 552/1541, loss: 0.38991302251815796 2023-01-23 10:52:37.279990: step: 554/1541, loss: 0.2779064476490021 2023-01-23 10:52:38.152847: step: 556/1541, loss: 0.02282700315117836 2023-01-23 10:52:39.132624: step: 558/1541, loss: 0.19626475870609283 2023-01-23 10:52:40.039430: step: 560/1541, loss: 0.08945582062005997 2023-01-23 10:52:40.948889: step: 562/1541, loss: 0.09754833579063416 2023-01-23 10:52:41.864519: step: 564/1541, loss: 0.3432943820953369 2023-01-23 10:52:42.755944: step: 566/1541, loss: 0.10354432463645935 2023-01-23 10:52:43.627792: step: 568/1541, loss: 0.0524820052087307 2023-01-23 10:52:44.520756: step: 570/1541, loss: 0.08843500167131424 2023-01-23 10:52:45.409849: step: 572/1541, loss: 0.15126565098762512 2023-01-23 10:52:46.317144: step: 574/1541, loss: 0.1006600633263588 2023-01-23 10:52:47.184759: step: 576/1541, loss: 0.06441549211740494 2023-01-23 10:52:48.098740: step: 578/1541, loss: 0.1824672669172287 2023-01-23 10:52:48.991735: step: 580/1541, loss: 0.17820540070533752 2023-01-23 10:52:49.903933: step: 582/1541, loss: 0.10894428938627243 2023-01-23 10:52:50.819682: step: 584/1541, loss: 0.33074432611465454 2023-01-23 10:52:51.758642: step: 586/1541, loss: 0.31747719645500183 2023-01-23 10:52:52.615302: step: 588/1541, loss: 0.03300880640745163 2023-01-23 10:52:53.533110: step: 590/1541, loss: 0.23903855681419373 2023-01-23 10:52:54.431835: step: 592/1541, loss: 0.07526511698961258 2023-01-23 10:52:55.358189: step: 594/1541, loss: 0.2273988127708435 2023-01-23 10:52:56.233404: step: 596/1541, loss: 0.10395794361829758 2023-01-23 10:52:57.128641: step: 598/1541, loss: 0.617633044719696 2023-01-23 10:52:58.006178: step: 600/1541, loss: 0.11484178155660629 2023-01-23 10:52:58.923309: step: 602/1541, loss: 0.1336301565170288 2023-01-23 10:52:59.846820: step: 604/1541, loss: 0.34270408749580383 2023-01-23 10:53:00.770605: step: 606/1541, loss: 0.04187392815947533 2023-01-23 10:53:01.636154: step: 608/1541, loss: 0.20687335729599 2023-01-23 10:53:02.574522: step: 610/1541, loss: 0.1394265741109848 2023-01-23 10:53:03.512276: step: 612/1541, loss: 0.3970886468887329 2023-01-23 10:53:04.409473: step: 614/1541, loss: 0.0895291343331337 2023-01-23 10:53:05.280441: step: 616/1541, loss: 0.023134469985961914 2023-01-23 10:53:06.171198: step: 618/1541, loss: 0.039719145745038986 2023-01-23 10:53:07.105787: step: 620/1541, loss: 0.07855283468961716 2023-01-23 10:53:08.019871: step: 622/1541, loss: 0.9619740843772888 2023-01-23 10:53:08.895469: step: 624/1541, loss: 0.07280128449201584 2023-01-23 10:53:09.924761: step: 626/1541, loss: 0.05478315055370331 2023-01-23 10:53:10.870889: step: 628/1541, loss: 0.012938931584358215 2023-01-23 10:53:11.898373: step: 630/1541, loss: 0.17247144877910614 2023-01-23 10:53:12.819029: step: 632/1541, loss: 0.16143029928207397 2023-01-23 10:53:13.717231: step: 634/1541, loss: 0.34732645750045776 2023-01-23 10:53:14.644004: step: 636/1541, loss: 0.012576092034578323 2023-01-23 10:53:15.538227: step: 638/1541, loss: 0.46238991618156433 2023-01-23 10:53:16.443080: step: 640/1541, loss: 0.18709221482276917 2023-01-23 10:53:17.339826: step: 642/1541, loss: 0.09661560505628586 2023-01-23 10:53:18.246672: step: 644/1541, loss: 0.08890821784734726 2023-01-23 10:53:19.134888: step: 646/1541, loss: 0.07228589057922363 2023-01-23 10:53:20.066253: step: 648/1541, loss: 0.0906028300523758 2023-01-23 10:53:20.917101: step: 650/1541, loss: 0.052288301289081573 2023-01-23 10:53:21.883279: step: 652/1541, loss: 0.01483806874603033 2023-01-23 10:53:22.738742: step: 654/1541, loss: 0.08195102214813232 2023-01-23 10:53:23.635042: step: 656/1541, loss: 0.06140241399407387 2023-01-23 10:53:24.505845: step: 658/1541, loss: 0.11062125861644745 2023-01-23 10:53:25.370823: step: 660/1541, loss: 0.25448721647262573 2023-01-23 10:53:26.230140: step: 662/1541, loss: 0.09907808154821396 2023-01-23 10:53:27.135215: step: 664/1541, loss: 0.32056307792663574 2023-01-23 10:53:28.018763: step: 666/1541, loss: 0.10671243071556091 2023-01-23 10:53:28.899801: step: 668/1541, loss: 0.00861575547605753 2023-01-23 10:53:29.824541: step: 670/1541, loss: 0.08510446548461914 2023-01-23 10:53:30.680700: step: 672/1541, loss: 0.26303693652153015 2023-01-23 10:53:31.579031: step: 674/1541, loss: 0.3996160328388214 2023-01-23 10:53:32.488046: step: 676/1541, loss: 0.07899381220340729 2023-01-23 10:53:33.363850: step: 678/1541, loss: 0.03623919188976288 2023-01-23 10:53:34.263640: step: 680/1541, loss: 0.037371888756752014 2023-01-23 10:53:35.220245: step: 682/1541, loss: 0.1367432326078415 2023-01-23 10:53:36.075142: step: 684/1541, loss: 0.3484998345375061 2023-01-23 10:53:36.961003: step: 686/1541, loss: 0.32192984223365784 2023-01-23 10:53:37.826307: step: 688/1541, loss: 0.06363144516944885 2023-01-23 10:53:38.688507: step: 690/1541, loss: 0.06850394606590271 2023-01-23 10:53:39.590985: step: 692/1541, loss: 0.23390786349773407 2023-01-23 10:53:40.568180: step: 694/1541, loss: 0.022284695878624916 2023-01-23 10:53:41.459067: step: 696/1541, loss: 0.7152247428894043 2023-01-23 10:53:42.376837: step: 698/1541, loss: 0.21398267149925232 2023-01-23 10:53:43.273355: step: 700/1541, loss: 0.09314452111721039 2023-01-23 10:53:44.243694: step: 702/1541, loss: 0.39144983887672424 2023-01-23 10:53:45.157132: step: 704/1541, loss: 0.14736010134220123 2023-01-23 10:53:46.104843: step: 706/1541, loss: 0.15300579369068146 2023-01-23 10:53:46.988058: step: 708/1541, loss: 0.16653528809547424 2023-01-23 10:53:47.907168: step: 710/1541, loss: 0.21242621541023254 2023-01-23 10:53:48.782309: step: 712/1541, loss: 0.05275854468345642 2023-01-23 10:53:49.683027: step: 714/1541, loss: 0.5559909343719482 2023-01-23 10:53:50.584606: step: 716/1541, loss: 0.10505814850330353 2023-01-23 10:53:51.541626: step: 718/1541, loss: 0.05167127400636673 2023-01-23 10:53:52.465174: step: 720/1541, loss: 0.4994061589241028 2023-01-23 10:53:53.361653: step: 722/1541, loss: 0.17931044101715088 2023-01-23 10:53:54.264520: step: 724/1541, loss: 0.3488003611564636 2023-01-23 10:53:55.177209: step: 726/1541, loss: 0.06533743441104889 2023-01-23 10:53:56.050451: step: 728/1541, loss: 0.08864837884902954 2023-01-23 10:53:56.967053: step: 730/1541, loss: 0.24373669922351837 2023-01-23 10:53:57.842109: step: 732/1541, loss: 0.3422672152519226 2023-01-23 10:53:58.714635: step: 734/1541, loss: 0.07085464894771576 2023-01-23 10:53:59.637373: step: 736/1541, loss: 0.08101682364940643 2023-01-23 10:54:00.517089: step: 738/1541, loss: 2.6071317195892334 2023-01-23 10:54:01.426497: step: 740/1541, loss: 0.13070239126682281 2023-01-23 10:54:02.440542: step: 742/1541, loss: 0.18866735696792603 2023-01-23 10:54:03.296981: step: 744/1541, loss: 0.03262780234217644 2023-01-23 10:54:04.187249: step: 746/1541, loss: 0.038281962275505066 2023-01-23 10:54:05.049302: step: 748/1541, loss: 0.18760325014591217 2023-01-23 10:54:05.934846: step: 750/1541, loss: 0.17111964523792267 2023-01-23 10:54:06.823425: step: 752/1541, loss: 0.030608851462602615 2023-01-23 10:54:07.737699: step: 754/1541, loss: 0.87012779712677 2023-01-23 10:54:08.608936: step: 756/1541, loss: 0.03794761747121811 2023-01-23 10:54:09.484256: step: 758/1541, loss: 0.009307030588388443 2023-01-23 10:54:10.399113: step: 760/1541, loss: 0.08939938247203827 2023-01-23 10:54:11.311105: step: 762/1541, loss: 0.5485588312149048 2023-01-23 10:54:12.183834: step: 764/1541, loss: 1.894363284111023 2023-01-23 10:54:13.100309: step: 766/1541, loss: 2.527460813522339 2023-01-23 10:54:13.984583: step: 768/1541, loss: 0.03232784941792488 2023-01-23 10:54:14.877160: step: 770/1541, loss: 0.08614025264978409 2023-01-23 10:54:15.861809: step: 772/1541, loss: 0.086118184030056 2023-01-23 10:54:16.737754: step: 774/1541, loss: 0.23036450147628784 2023-01-23 10:54:17.587376: step: 776/1541, loss: 0.28237318992614746 2023-01-23 10:54:18.510431: step: 778/1541, loss: 0.033234383910894394 2023-01-23 10:54:19.452823: step: 780/1541, loss: 0.042779553681612015 2023-01-23 10:54:20.353217: step: 782/1541, loss: 0.3851114511489868 2023-01-23 10:54:21.255560: step: 784/1541, loss: 0.023608101531863213 2023-01-23 10:54:22.100571: step: 786/1541, loss: 0.03896990418434143 2023-01-23 10:54:22.981384: step: 788/1541, loss: 0.07195651531219482 2023-01-23 10:54:23.893490: step: 790/1541, loss: 0.33579179644584656 2023-01-23 10:54:24.779296: step: 792/1541, loss: 1.2418262958526611 2023-01-23 10:54:25.700871: step: 794/1541, loss: 0.050163500010967255 2023-01-23 10:54:26.589121: step: 796/1541, loss: 0.2590738832950592 2023-01-23 10:54:27.483620: step: 798/1541, loss: 0.14752137660980225 2023-01-23 10:54:28.372949: step: 800/1541, loss: 0.30396729707717896 2023-01-23 10:54:29.264217: step: 802/1541, loss: 0.24727217853069305 2023-01-23 10:54:30.146422: step: 804/1541, loss: 0.3235030770301819 2023-01-23 10:54:31.032936: step: 806/1541, loss: 0.26407483220100403 2023-01-23 10:54:31.941471: step: 808/1541, loss: 0.09555736184120178 2023-01-23 10:54:32.821793: step: 810/1541, loss: 0.022702179849147797 2023-01-23 10:54:33.687618: step: 812/1541, loss: 0.5513862371444702 2023-01-23 10:54:34.565375: step: 814/1541, loss: 0.2558801472187042 2023-01-23 10:54:35.438104: step: 816/1541, loss: 0.1543164700269699 2023-01-23 10:54:36.358070: step: 818/1541, loss: 0.24261297285556793 2023-01-23 10:54:37.266143: step: 820/1541, loss: 0.10787568986415863 2023-01-23 10:54:38.208710: step: 822/1541, loss: 0.254700243473053 2023-01-23 10:54:39.054113: step: 824/1541, loss: 0.2400345802307129 2023-01-23 10:54:39.921468: step: 826/1541, loss: 1.016025185585022 2023-01-23 10:54:40.802787: step: 828/1541, loss: 0.04513707011938095 2023-01-23 10:54:41.678641: step: 830/1541, loss: 0.19636236131191254 2023-01-23 10:54:42.543751: step: 832/1541, loss: 0.0025083660148084164 2023-01-23 10:54:43.453397: step: 834/1541, loss: 0.05298234894871712 2023-01-23 10:54:44.382874: step: 836/1541, loss: 0.19620221853256226 2023-01-23 10:54:45.260615: step: 838/1541, loss: 0.05319647863507271 2023-01-23 10:54:46.138885: step: 840/1541, loss: 0.02903987094759941 2023-01-23 10:54:47.030106: step: 842/1541, loss: 0.04239409416913986 2023-01-23 10:54:47.949228: step: 844/1541, loss: 0.06445825099945068 2023-01-23 10:54:48.887298: step: 846/1541, loss: 0.13271823525428772 2023-01-23 10:54:49.760395: step: 848/1541, loss: 0.7158403396606445 2023-01-23 10:54:50.616090: step: 850/1541, loss: 0.00726162688806653 2023-01-23 10:54:51.499520: step: 852/1541, loss: 0.0585184209048748 2023-01-23 10:54:52.401329: step: 854/1541, loss: 0.2502136528491974 2023-01-23 10:54:53.318540: step: 856/1541, loss: 0.044091932475566864 2023-01-23 10:54:54.213841: step: 858/1541, loss: 0.036792855709791183 2023-01-23 10:54:55.095687: step: 860/1541, loss: 0.0072740837931632996 2023-01-23 10:54:56.024007: step: 862/1541, loss: 0.49083539843559265 2023-01-23 10:54:56.929923: step: 864/1541, loss: 0.13215388357639313 2023-01-23 10:54:57.833513: step: 866/1541, loss: 0.0272817425429821 2023-01-23 10:54:58.721851: step: 868/1541, loss: 0.023378318175673485 2023-01-23 10:54:59.586102: step: 870/1541, loss: 0.3752381503582001 2023-01-23 10:55:00.472821: step: 872/1541, loss: 0.06266254186630249 2023-01-23 10:55:01.407763: step: 874/1541, loss: 0.28372853994369507 2023-01-23 10:55:02.314650: step: 876/1541, loss: 0.07631760835647583 2023-01-23 10:55:03.299000: step: 878/1541, loss: 0.08428307622671127 2023-01-23 10:55:04.160189: step: 880/1541, loss: 0.07579637318849564 2023-01-23 10:55:05.026447: step: 882/1541, loss: 0.07966151833534241 2023-01-23 10:55:06.032457: step: 884/1541, loss: 0.06205206364393234 2023-01-23 10:55:06.922336: step: 886/1541, loss: 0.06798883527517319 2023-01-23 10:55:07.873222: step: 888/1541, loss: 0.07094410061836243 2023-01-23 10:55:08.774026: step: 890/1541, loss: 0.23872536420822144 2023-01-23 10:55:09.811769: step: 892/1541, loss: 0.24692322313785553 2023-01-23 10:55:10.743571: step: 894/1541, loss: 0.16966815292835236 2023-01-23 10:55:11.612387: step: 896/1541, loss: 1.8950544595718384 2023-01-23 10:55:12.480659: step: 898/1541, loss: 0.22469617426395416 2023-01-23 10:55:13.393449: step: 900/1541, loss: 0.08631163090467453 2023-01-23 10:55:14.291177: step: 902/1541, loss: 0.036188799887895584 2023-01-23 10:55:15.211639: step: 904/1541, loss: 0.10913647711277008 2023-01-23 10:55:16.133172: step: 906/1541, loss: 0.09180282056331635 2023-01-23 10:55:17.088517: step: 908/1541, loss: 0.031684935092926025 2023-01-23 10:55:17.954943: step: 910/1541, loss: 0.13092313706874847 2023-01-23 10:55:18.876184: step: 912/1541, loss: 0.23505648970603943 2023-01-23 10:55:19.796372: step: 914/1541, loss: 0.046225786209106445 2023-01-23 10:55:20.744001: step: 916/1541, loss: 0.13593344390392303 2023-01-23 10:55:21.661646: step: 918/1541, loss: 0.30126121640205383 2023-01-23 10:55:22.582787: step: 920/1541, loss: 0.10469453781843185 2023-01-23 10:55:23.447495: step: 922/1541, loss: 0.2048415243625641 2023-01-23 10:55:24.356905: step: 924/1541, loss: 0.15502819418907166 2023-01-23 10:55:25.269696: step: 926/1541, loss: 0.30314135551452637 2023-01-23 10:55:26.178162: step: 928/1541, loss: 0.04729524254798889 2023-01-23 10:55:27.025406: step: 930/1541, loss: 0.05412660911679268 2023-01-23 10:55:27.961363: step: 932/1541, loss: 0.07080312073230743 2023-01-23 10:55:28.857664: step: 934/1541, loss: 0.08874458819627762 2023-01-23 10:55:29.738299: step: 936/1541, loss: 0.337442010641098 2023-01-23 10:55:30.593219: step: 938/1541, loss: 0.11245795339345932 2023-01-23 10:55:31.485026: step: 940/1541, loss: 0.1337500512599945 2023-01-23 10:55:32.330760: step: 942/1541, loss: 0.06099781021475792 2023-01-23 10:55:33.266144: step: 944/1541, loss: 0.11059074848890305 2023-01-23 10:55:34.186017: step: 946/1541, loss: 0.15270960330963135 2023-01-23 10:55:35.085630: step: 948/1541, loss: 0.5302037000656128 2023-01-23 10:55:35.957275: step: 950/1541, loss: 0.056393399834632874 2023-01-23 10:55:36.846420: step: 952/1541, loss: 0.03400859981775284 2023-01-23 10:55:37.748979: step: 954/1541, loss: 0.13446010649204254 2023-01-23 10:55:38.635910: step: 956/1541, loss: 0.4697079658508301 2023-01-23 10:55:39.501309: step: 958/1541, loss: 0.28288084268569946 2023-01-23 10:55:40.402850: step: 960/1541, loss: 0.1642371416091919 2023-01-23 10:55:41.249301: step: 962/1541, loss: 0.033883705735206604 2023-01-23 10:55:42.196497: step: 964/1541, loss: 2.8868393898010254 2023-01-23 10:55:43.050152: step: 966/1541, loss: 0.04515470564365387 2023-01-23 10:55:43.930094: step: 968/1541, loss: 0.046961840242147446 2023-01-23 10:55:44.786021: step: 970/1541, loss: 0.14011138677597046 2023-01-23 10:55:45.663881: step: 972/1541, loss: 0.18183903396129608 2023-01-23 10:55:46.527093: step: 974/1541, loss: 0.023042753338813782 2023-01-23 10:55:47.464807: step: 976/1541, loss: 0.07242390513420105 2023-01-23 10:55:48.354027: step: 978/1541, loss: 0.08404750376939774 2023-01-23 10:55:49.209902: step: 980/1541, loss: 1.681152582168579 2023-01-23 10:55:50.117536: step: 982/1541, loss: 0.12187790870666504 2023-01-23 10:55:50.973714: step: 984/1541, loss: 0.18513137102127075 2023-01-23 10:55:51.868197: step: 986/1541, loss: 0.3205167353153229 2023-01-23 10:55:52.772355: step: 988/1541, loss: 0.08468274027109146 2023-01-23 10:55:53.720745: step: 990/1541, loss: 0.387555330991745 2023-01-23 10:55:54.670056: step: 992/1541, loss: 0.03329605236649513 2023-01-23 10:55:55.614398: step: 994/1541, loss: 0.0888928472995758 2023-01-23 10:55:56.495282: step: 996/1541, loss: 0.13254503905773163 2023-01-23 10:55:57.401422: step: 998/1541, loss: 0.03398989513516426 2023-01-23 10:55:58.291391: step: 1000/1541, loss: 0.13697633147239685 2023-01-23 10:55:59.268251: step: 1002/1541, loss: 0.28480029106140137 2023-01-23 10:56:00.200327: step: 1004/1541, loss: 0.016482580453157425 2023-01-23 10:56:01.144653: step: 1006/1541, loss: 0.03096509538590908 2023-01-23 10:56:02.075308: step: 1008/1541, loss: 0.07055621594190598 2023-01-23 10:56:02.945966: step: 1010/1541, loss: 0.13094405829906464 2023-01-23 10:56:03.825001: step: 1012/1541, loss: 0.05215431749820709 2023-01-23 10:56:04.800053: step: 1014/1541, loss: 0.025144560262560844 2023-01-23 10:56:05.678623: step: 1016/1541, loss: 0.147901251912117 2023-01-23 10:56:06.641628: step: 1018/1541, loss: 0.4712355136871338 2023-01-23 10:56:07.590738: step: 1020/1541, loss: 0.2720149755477905 2023-01-23 10:56:08.503197: step: 1022/1541, loss: 0.19286802411079407 2023-01-23 10:56:09.462420: step: 1024/1541, loss: 0.19152194261550903 2023-01-23 10:56:10.356261: step: 1026/1541, loss: 0.13987472653388977 2023-01-23 10:56:11.257959: step: 1028/1541, loss: 0.15639318525791168 2023-01-23 10:56:12.133714: step: 1030/1541, loss: 0.20476500689983368 2023-01-23 10:56:13.119760: step: 1032/1541, loss: 0.6381790637969971 2023-01-23 10:56:14.003000: step: 1034/1541, loss: 0.05278395861387253 2023-01-23 10:56:14.907269: step: 1036/1541, loss: 0.08236559480428696 2023-01-23 10:56:15.836313: step: 1038/1541, loss: 0.12119041383266449 2023-01-23 10:56:16.771300: step: 1040/1541, loss: 0.02181384153664112 2023-01-23 10:56:17.687339: step: 1042/1541, loss: 0.19135309755802155 2023-01-23 10:56:18.648295: step: 1044/1541, loss: 0.3452446460723877 2023-01-23 10:56:19.582583: step: 1046/1541, loss: 0.2885047197341919 2023-01-23 10:56:20.472973: step: 1048/1541, loss: 0.03778664022684097 2023-01-23 10:56:21.343339: step: 1050/1541, loss: 0.49839454889297485 2023-01-23 10:56:22.227125: step: 1052/1541, loss: 0.2936612367630005 2023-01-23 10:56:23.098915: step: 1054/1541, loss: 0.05432642623782158 2023-01-23 10:56:24.034853: step: 1056/1541, loss: 0.19738997519016266 2023-01-23 10:56:24.937568: step: 1058/1541, loss: 0.01165699865669012 2023-01-23 10:56:25.902869: step: 1060/1541, loss: 0.4063488841056824 2023-01-23 10:56:26.820948: step: 1062/1541, loss: 0.09927383065223694 2023-01-23 10:56:27.718103: step: 1064/1541, loss: 0.3002856373786926 2023-01-23 10:56:28.610675: step: 1066/1541, loss: 0.10225597023963928 2023-01-23 10:56:29.544994: step: 1068/1541, loss: 0.13777832686901093 2023-01-23 10:56:30.441327: step: 1070/1541, loss: 0.1617564558982849 2023-01-23 10:56:31.402829: step: 1072/1541, loss: 0.6273820400238037 2023-01-23 10:56:32.397757: step: 1074/1541, loss: 0.07021581381559372 2023-01-23 10:56:33.275999: step: 1076/1541, loss: 0.07487263530492783 2023-01-23 10:56:34.217021: step: 1078/1541, loss: 0.029028408229351044 2023-01-23 10:56:35.113723: step: 1080/1541, loss: 0.0393456369638443 2023-01-23 10:56:36.014781: step: 1082/1541, loss: 0.2900090217590332 2023-01-23 10:56:36.891020: step: 1084/1541, loss: 0.05625240132212639 2023-01-23 10:56:37.781411: step: 1086/1541, loss: 0.13846281170845032 2023-01-23 10:56:38.696571: step: 1088/1541, loss: 0.1250246912240982 2023-01-23 10:56:39.665745: step: 1090/1541, loss: 0.0999068021774292 2023-01-23 10:56:40.598778: step: 1092/1541, loss: 0.16834846138954163 2023-01-23 10:56:41.517718: step: 1094/1541, loss: 0.1484539955854416 2023-01-23 10:56:42.392804: step: 1096/1541, loss: 0.20036113262176514 2023-01-23 10:56:43.336598: step: 1098/1541, loss: 0.1815059632062912 2023-01-23 10:56:44.223019: step: 1100/1541, loss: 0.04000648856163025 2023-01-23 10:56:45.082168: step: 1102/1541, loss: 0.06698191910982132 2023-01-23 10:56:45.969756: step: 1104/1541, loss: 0.013536829501390457 2023-01-23 10:56:46.858152: step: 1106/1541, loss: 0.016742680221796036 2023-01-23 10:56:47.736239: step: 1108/1541, loss: 0.05096466839313507 2023-01-23 10:56:48.678974: step: 1110/1541, loss: 0.0685979425907135 2023-01-23 10:56:49.546491: step: 1112/1541, loss: 0.14331047236919403 2023-01-23 10:56:50.507570: step: 1114/1541, loss: 0.019680699333548546 2023-01-23 10:56:51.420004: step: 1116/1541, loss: 2.502317428588867 2023-01-23 10:56:52.322801: step: 1118/1541, loss: 0.42644810676574707 2023-01-23 10:56:53.190666: step: 1120/1541, loss: 0.01038927212357521 2023-01-23 10:56:54.180448: step: 1122/1541, loss: 0.00232538185082376 2023-01-23 10:56:55.153184: step: 1124/1541, loss: 0.12904049456119537 2023-01-23 10:56:56.060895: step: 1126/1541, loss: 0.09072377532720566 2023-01-23 10:56:56.918135: step: 1128/1541, loss: 0.08811960369348526 2023-01-23 10:56:57.849814: step: 1130/1541, loss: 0.21587473154067993 2023-01-23 10:56:58.721150: step: 1132/1541, loss: 0.17644792795181274 2023-01-23 10:56:59.594400: step: 1134/1541, loss: 0.19175666570663452 2023-01-23 10:57:00.532135: step: 1136/1541, loss: 0.17983409762382507 2023-01-23 10:57:01.432889: step: 1138/1541, loss: 0.030264783650636673 2023-01-23 10:57:02.322395: step: 1140/1541, loss: 0.04362252354621887 2023-01-23 10:57:03.225607: step: 1142/1541, loss: 0.3455270826816559 2023-01-23 10:57:04.094242: step: 1144/1541, loss: 0.18478044867515564 2023-01-23 10:57:04.965588: step: 1146/1541, loss: 0.34249812364578247 2023-01-23 10:57:05.916687: step: 1148/1541, loss: 0.154390349984169 2023-01-23 10:57:06.826683: step: 1150/1541, loss: 0.40086668729782104 2023-01-23 10:57:07.695863: step: 1152/1541, loss: 0.07477101683616638 2023-01-23 10:57:08.599195: step: 1154/1541, loss: 0.0773458257317543 2023-01-23 10:57:09.499556: step: 1156/1541, loss: 0.08668787777423859 2023-01-23 10:57:10.392370: step: 1158/1541, loss: 0.03869735822081566 2023-01-23 10:57:11.254181: step: 1160/1541, loss: 0.0683053582906723 2023-01-23 10:57:12.139891: step: 1162/1541, loss: 0.6294146776199341 2023-01-23 10:57:13.070562: step: 1164/1541, loss: 0.5549527406692505 2023-01-23 10:57:13.968794: step: 1166/1541, loss: 0.33959323167800903 2023-01-23 10:57:14.961546: step: 1168/1541, loss: 0.11804720759391785 2023-01-23 10:57:15.860234: step: 1170/1541, loss: 0.022615468129515648 2023-01-23 10:57:16.746160: step: 1172/1541, loss: 0.06012352183461189 2023-01-23 10:57:17.622818: step: 1174/1541, loss: 0.01607578620314598 2023-01-23 10:57:18.494941: step: 1176/1541, loss: 0.11535242199897766 2023-01-23 10:57:19.387932: step: 1178/1541, loss: 0.6218194961547852 2023-01-23 10:57:20.319791: step: 1180/1541, loss: 0.13626205921173096 2023-01-23 10:57:21.242937: step: 1182/1541, loss: 0.35821235179901123 2023-01-23 10:57:22.137016: step: 1184/1541, loss: 0.32255077362060547 2023-01-23 10:57:23.028109: step: 1186/1541, loss: 0.0327746644616127 2023-01-23 10:57:23.921699: step: 1188/1541, loss: 0.044421084225177765 2023-01-23 10:57:24.773243: step: 1190/1541, loss: 0.03736644983291626 2023-01-23 10:57:25.669232: step: 1192/1541, loss: 0.04253755882382393 2023-01-23 10:57:26.602867: step: 1194/1541, loss: 1.673529028892517 2023-01-23 10:57:27.521666: step: 1196/1541, loss: 0.010966502130031586 2023-01-23 10:57:28.395774: step: 1198/1541, loss: 0.04317736253142357 2023-01-23 10:57:29.293517: step: 1200/1541, loss: 0.04404230788350105 2023-01-23 10:57:30.233851: step: 1202/1541, loss: 0.12513047456741333 2023-01-23 10:57:31.093020: step: 1204/1541, loss: 0.11541498452425003 2023-01-23 10:57:31.985252: step: 1206/1541, loss: 0.14583809673786163 2023-01-23 10:57:32.886690: step: 1208/1541, loss: 0.21204182505607605 2023-01-23 10:57:33.886039: step: 1210/1541, loss: 0.2336588352918625 2023-01-23 10:57:34.766602: step: 1212/1541, loss: 0.0804692953824997 2023-01-23 10:57:35.670226: step: 1214/1541, loss: 0.23727163672447205 2023-01-23 10:57:36.588464: step: 1216/1541, loss: 0.6799241304397583 2023-01-23 10:57:37.466925: step: 1218/1541, loss: 1.5349706411361694 2023-01-23 10:57:38.352058: step: 1220/1541, loss: 0.07493157684803009 2023-01-23 10:57:39.230103: step: 1222/1541, loss: 0.023898573592305183 2023-01-23 10:57:40.213981: step: 1224/1541, loss: 0.09674189984798431 2023-01-23 10:57:41.110561: step: 1226/1541, loss: 0.2575100064277649 2023-01-23 10:57:41.978600: step: 1228/1541, loss: 0.16158932447433472 2023-01-23 10:57:42.875936: step: 1230/1541, loss: 3.616316318511963 2023-01-23 10:57:43.781944: step: 1232/1541, loss: 0.017359286546707153 2023-01-23 10:57:44.664345: step: 1234/1541, loss: 0.2155676633119583 2023-01-23 10:57:45.531024: step: 1236/1541, loss: 0.1545886993408203 2023-01-23 10:57:46.425023: step: 1238/1541, loss: 0.053794968873262405 2023-01-23 10:57:47.305947: step: 1240/1541, loss: 0.20068785548210144 2023-01-23 10:57:48.203447: step: 1242/1541, loss: 0.10305546969175339 2023-01-23 10:57:49.101034: step: 1244/1541, loss: 0.13785715401172638 2023-01-23 10:57:49.971612: step: 1246/1541, loss: 0.2236989438533783 2023-01-23 10:57:50.946406: step: 1248/1541, loss: 0.09431809186935425 2023-01-23 10:57:51.871860: step: 1250/1541, loss: 0.11167605221271515 2023-01-23 10:57:52.764233: step: 1252/1541, loss: 0.18666326999664307 2023-01-23 10:57:53.635113: step: 1254/1541, loss: 0.20307472348213196 2023-01-23 10:57:54.542711: step: 1256/1541, loss: 0.10893063247203827 2023-01-23 10:57:55.456929: step: 1258/1541, loss: 0.017402097582817078 2023-01-23 10:57:56.334185: step: 1260/1541, loss: 0.101799875497818 2023-01-23 10:57:57.214859: step: 1262/1541, loss: 0.08861365914344788 2023-01-23 10:57:58.097274: step: 1264/1541, loss: 0.009781714528799057 2023-01-23 10:57:59.002876: step: 1266/1541, loss: 0.06488344073295593 2023-01-23 10:57:59.928948: step: 1268/1541, loss: 0.2513866424560547 2023-01-23 10:58:00.791400: step: 1270/1541, loss: 0.056868165731430054 2023-01-23 10:58:01.633419: step: 1272/1541, loss: 0.11245332658290863 2023-01-23 10:58:02.544548: step: 1274/1541, loss: 0.1456209123134613 2023-01-23 10:58:03.450888: step: 1276/1541, loss: 0.07570720463991165 2023-01-23 10:58:04.369459: step: 1278/1541, loss: 0.010278988629579544 2023-01-23 10:58:05.331403: step: 1280/1541, loss: 0.19894246757030487 2023-01-23 10:58:06.224888: step: 1282/1541, loss: 0.3161975145339966 2023-01-23 10:58:07.140924: step: 1284/1541, loss: 0.022113710641860962 2023-01-23 10:58:08.029989: step: 1286/1541, loss: 0.022195037454366684 2023-01-23 10:58:08.941412: step: 1288/1541, loss: 0.17512643337249756 2023-01-23 10:58:09.834309: step: 1290/1541, loss: 0.226612851023674 2023-01-23 10:58:10.713275: step: 1292/1541, loss: 0.4430202543735504 2023-01-23 10:58:11.609209: step: 1294/1541, loss: 0.01132144033908844 2023-01-23 10:58:12.535872: step: 1296/1541, loss: 1.0921543836593628 2023-01-23 10:58:13.420182: step: 1298/1541, loss: 1.389824628829956 2023-01-23 10:58:14.308381: step: 1300/1541, loss: 0.028894193470478058 2023-01-23 10:58:15.219286: step: 1302/1541, loss: 0.09098110347986221 2023-01-23 10:58:16.118485: step: 1304/1541, loss: 0.22052572667598724 2023-01-23 10:58:17.092057: step: 1306/1541, loss: 0.04469166696071625 2023-01-23 10:58:17.994779: step: 1308/1541, loss: 0.014015440829098225 2023-01-23 10:58:18.835679: step: 1310/1541, loss: 0.9383053779602051 2023-01-23 10:58:19.747978: step: 1312/1541, loss: 0.07341410964727402 2023-01-23 10:58:20.662224: step: 1314/1541, loss: 0.028376711532473564 2023-01-23 10:58:21.546890: step: 1316/1541, loss: 0.31653738021850586 2023-01-23 10:58:22.425252: step: 1318/1541, loss: 0.01873159036040306 2023-01-23 10:58:23.305984: step: 1320/1541, loss: 0.0940139889717102 2023-01-23 10:58:24.193984: step: 1322/1541, loss: 0.3227732181549072 2023-01-23 10:58:25.188299: step: 1324/1541, loss: 0.026484675705432892 2023-01-23 10:58:26.060778: step: 1326/1541, loss: 0.12455873936414719 2023-01-23 10:58:26.968574: step: 1328/1541, loss: 0.7042444944381714 2023-01-23 10:58:27.856173: step: 1330/1541, loss: 0.12907126545906067 2023-01-23 10:58:28.772886: step: 1332/1541, loss: 0.02140720933675766 2023-01-23 10:58:29.665128: step: 1334/1541, loss: 0.06450311094522476 2023-01-23 10:58:30.514369: step: 1336/1541, loss: 0.3246726989746094 2023-01-23 10:58:31.397641: step: 1338/1541, loss: 0.07685776054859161 2023-01-23 10:58:32.285520: step: 1340/1541, loss: 0.10242496430873871 2023-01-23 10:58:33.214560: step: 1342/1541, loss: 0.05349709838628769 2023-01-23 10:58:34.235218: step: 1344/1541, loss: 0.04398048296570778 2023-01-23 10:58:35.104699: step: 1346/1541, loss: 0.1734524369239807 2023-01-23 10:58:36.008994: step: 1348/1541, loss: 1.3904181718826294 2023-01-23 10:58:36.901790: step: 1350/1541, loss: 0.12589362263679504 2023-01-23 10:58:37.791679: step: 1352/1541, loss: 0.16844266653060913 2023-01-23 10:58:38.689147: step: 1354/1541, loss: 0.19831737875938416 2023-01-23 10:58:39.608534: step: 1356/1541, loss: 0.07166433334350586 2023-01-23 10:58:40.490059: step: 1358/1541, loss: 0.08860771358013153 2023-01-23 10:58:41.420200: step: 1360/1541, loss: 0.3385556936264038 2023-01-23 10:58:42.278763: step: 1362/1541, loss: 0.012420637533068657 2023-01-23 10:58:43.168193: step: 1364/1541, loss: 0.04935769736766815 2023-01-23 10:58:44.068560: step: 1366/1541, loss: 0.014030968770384789 2023-01-23 10:58:45.025313: step: 1368/1541, loss: 0.5946865677833557 2023-01-23 10:58:45.973870: step: 1370/1541, loss: 0.18558959662914276 2023-01-23 10:58:46.893629: step: 1372/1541, loss: 0.2310851365327835 2023-01-23 10:58:47.772062: step: 1374/1541, loss: 0.04658379778265953 2023-01-23 10:58:48.688821: step: 1376/1541, loss: 0.08004864305257797 2023-01-23 10:58:49.614713: step: 1378/1541, loss: 0.2177584022283554 2023-01-23 10:58:50.508992: step: 1380/1541, loss: 0.23844407498836517 2023-01-23 10:58:51.407522: step: 1382/1541, loss: 0.34491321444511414 2023-01-23 10:58:52.331390: step: 1384/1541, loss: 0.00723786186426878 2023-01-23 10:58:53.299195: step: 1386/1541, loss: 0.046182140707969666 2023-01-23 10:58:54.220970: step: 1388/1541, loss: 0.2121431976556778 2023-01-23 10:58:55.086958: step: 1390/1541, loss: 0.5765627026557922 2023-01-23 10:58:55.971395: step: 1392/1541, loss: 0.15708599984645844 2023-01-23 10:58:56.856695: step: 1394/1541, loss: 0.06158341467380524 2023-01-23 10:58:57.772600: step: 1396/1541, loss: 0.11055897921323776 2023-01-23 10:58:58.728380: step: 1398/1541, loss: 0.18067923188209534 2023-01-23 10:58:59.694242: step: 1400/1541, loss: 0.17134366929531097 2023-01-23 10:59:00.564966: step: 1402/1541, loss: 0.19474543631076813 2023-01-23 10:59:01.489446: step: 1404/1541, loss: 0.04820549488067627 2023-01-23 10:59:02.389134: step: 1406/1541, loss: 0.028278546407818794 2023-01-23 10:59:03.325832: step: 1408/1541, loss: 0.6947258114814758 2023-01-23 10:59:04.219127: step: 1410/1541, loss: 0.15634360909461975 2023-01-23 10:59:05.094703: step: 1412/1541, loss: 0.05200529098510742 2023-01-23 10:59:05.982138: step: 1414/1541, loss: 0.04121391475200653 2023-01-23 10:59:06.863170: step: 1416/1541, loss: 0.13287870585918427 2023-01-23 10:59:07.742600: step: 1418/1541, loss: 0.39169347286224365 2023-01-23 10:59:08.659001: step: 1420/1541, loss: 0.1220693588256836 2023-01-23 10:59:09.600387: step: 1422/1541, loss: 0.07059584558010101 2023-01-23 10:59:10.541168: step: 1424/1541, loss: 0.4507361054420471 2023-01-23 10:59:11.450070: step: 1426/1541, loss: 0.3571646809577942 2023-01-23 10:59:12.388636: step: 1428/1541, loss: 0.14162859320640564 2023-01-23 10:59:13.299964: step: 1430/1541, loss: 0.12186150997877121 2023-01-23 10:59:14.246637: step: 1432/1541, loss: 0.2799062430858612 2023-01-23 10:59:15.130433: step: 1434/1541, loss: 0.13274827599525452 2023-01-23 10:59:16.021906: step: 1436/1541, loss: 0.004693401511758566 2023-01-23 10:59:16.918516: step: 1438/1541, loss: 0.18597732484340668 2023-01-23 10:59:17.820546: step: 1440/1541, loss: 0.0818631500005722 2023-01-23 10:59:18.699866: step: 1442/1541, loss: 0.09527057409286499 2023-01-23 10:59:19.610902: step: 1444/1541, loss: 0.07045122981071472 2023-01-23 10:59:20.506173: step: 1446/1541, loss: 0.1316308081150055 2023-01-23 10:59:21.505412: step: 1448/1541, loss: 0.34579312801361084 2023-01-23 10:59:22.390080: step: 1450/1541, loss: 0.09959918260574341 2023-01-23 10:59:23.293987: step: 1452/1541, loss: 0.2560153007507324 2023-01-23 10:59:24.165698: step: 1454/1541, loss: 0.0233655646443367 2023-01-23 10:59:25.072546: step: 1456/1541, loss: 0.20808526873588562 2023-01-23 10:59:25.971558: step: 1458/1541, loss: 0.06559573858976364 2023-01-23 10:59:26.835977: step: 1460/1541, loss: 0.13507749140262604 2023-01-23 10:59:27.721118: step: 1462/1541, loss: 0.2679741382598877 2023-01-23 10:59:28.625411: step: 1464/1541, loss: 0.06963441520929337 2023-01-23 10:59:29.498109: step: 1466/1541, loss: 0.12376166880130768 2023-01-23 10:59:30.397732: step: 1468/1541, loss: 0.34116876125335693 2023-01-23 10:59:31.283349: step: 1470/1541, loss: 0.15956726670265198 2023-01-23 10:59:32.226608: step: 1472/1541, loss: 0.06944524496793747 2023-01-23 10:59:33.146827: step: 1474/1541, loss: 0.11369882524013519 2023-01-23 10:59:34.038073: step: 1476/1541, loss: 0.40655040740966797 2023-01-23 10:59:34.924601: step: 1478/1541, loss: 0.02862984500825405 2023-01-23 10:59:35.847987: step: 1480/1541, loss: 0.048810794949531555 2023-01-23 10:59:36.781357: step: 1482/1541, loss: 0.07940181344747543 2023-01-23 10:59:37.672450: step: 1484/1541, loss: 0.06673071533441544 2023-01-23 10:59:38.585852: step: 1486/1541, loss: 0.10372414439916611 2023-01-23 10:59:39.465599: step: 1488/1541, loss: 1.7749446630477905 2023-01-23 10:59:40.364536: step: 1490/1541, loss: 0.16702227294445038 2023-01-23 10:59:41.234945: step: 1492/1541, loss: 5.861471652984619 2023-01-23 10:59:42.189575: step: 1494/1541, loss: 0.405781090259552 2023-01-23 10:59:43.066537: step: 1496/1541, loss: 0.3680087924003601 2023-01-23 10:59:43.941251: step: 1498/1541, loss: 0.10385690629482269 2023-01-23 10:59:44.831626: step: 1500/1541, loss: 0.035178206861019135 2023-01-23 10:59:45.740812: step: 1502/1541, loss: 0.01388395857065916 2023-01-23 10:59:46.669532: step: 1504/1541, loss: 0.08825299888849258 2023-01-23 10:59:47.579585: step: 1506/1541, loss: 0.2074362188577652 2023-01-23 10:59:48.546761: step: 1508/1541, loss: 0.12859538197517395 2023-01-23 10:59:49.430370: step: 1510/1541, loss: 0.1921752542257309 2023-01-23 10:59:50.384504: step: 1512/1541, loss: 0.03225284814834595 2023-01-23 10:59:51.251309: step: 1514/1541, loss: 0.01284201443195343 2023-01-23 10:59:52.168998: step: 1516/1541, loss: 0.4168817698955536 2023-01-23 10:59:53.133702: step: 1518/1541, loss: 0.0732756033539772 2023-01-23 10:59:54.060138: step: 1520/1541, loss: 0.09571841359138489 2023-01-23 10:59:54.946442: step: 1522/1541, loss: 0.305753618478775 2023-01-23 10:59:55.836507: step: 1524/1541, loss: 0.2659898102283478 2023-01-23 10:59:56.708883: step: 1526/1541, loss: 0.07718636095523834 2023-01-23 10:59:57.577843: step: 1528/1541, loss: 0.06019602715969086 2023-01-23 10:59:58.494540: step: 1530/1541, loss: 0.011411468498408794 2023-01-23 10:59:59.433454: step: 1532/1541, loss: 0.15709374845027924 2023-01-23 11:00:00.311515: step: 1534/1541, loss: 0.15581633150577545 2023-01-23 11:00:01.200467: step: 1536/1541, loss: 0.0348266065120697 2023-01-23 11:00:02.066386: step: 1538/1541, loss: 0.1601535677909851 2023-01-23 11:00:02.987134: step: 1540/1541, loss: 0.019094720482826233 2023-01-23 11:00:03.871000: step: 1542/1541, loss: 0.19914931058883667 2023-01-23 11:00:04.755117: step: 1544/1541, loss: 0.23040729761123657 2023-01-23 11:00:05.669940: step: 1546/1541, loss: 0.37933650612831116 2023-01-23 11:00:06.523167: step: 1548/1541, loss: 0.16485077142715454 2023-01-23 11:00:07.473164: step: 1550/1541, loss: 0.10149965435266495 2023-01-23 11:00:08.369722: step: 1552/1541, loss: 0.1695803999900818 2023-01-23 11:00:09.259192: step: 1554/1541, loss: 0.24405048787593842 2023-01-23 11:00:10.140569: step: 1556/1541, loss: 0.00396730238571763 2023-01-23 11:00:11.043656: step: 1558/1541, loss: 0.009988161735236645 2023-01-23 11:00:11.961722: step: 1560/1541, loss: 0.3173889219760895 2023-01-23 11:00:12.982022: step: 1562/1541, loss: 0.2949008047580719 2023-01-23 11:00:13.883162: step: 1564/1541, loss: 0.1295498162508011 2023-01-23 11:00:14.766597: step: 1566/1541, loss: 0.1322486698627472 2023-01-23 11:00:15.786900: step: 1568/1541, loss: 0.17689619958400726 2023-01-23 11:00:16.677578: step: 1570/1541, loss: 0.0516195110976696 2023-01-23 11:00:17.649670: step: 1572/1541, loss: 0.3952264189720154 2023-01-23 11:00:18.564079: step: 1574/1541, loss: 0.08468298614025116 2023-01-23 11:00:19.443702: step: 1576/1541, loss: 0.3077746033668518 2023-01-23 11:00:20.404664: step: 1578/1541, loss: 0.2831130027770996 2023-01-23 11:00:21.337680: step: 1580/1541, loss: 0.06568948179483414 2023-01-23 11:00:22.225342: step: 1582/1541, loss: 0.07591743767261505 2023-01-23 11:00:23.148664: step: 1584/1541, loss: 0.08174213767051697 2023-01-23 11:00:24.136554: step: 1586/1541, loss: 0.1998835951089859 2023-01-23 11:00:25.031803: step: 1588/1541, loss: 0.4375650882720947 2023-01-23 11:00:25.972248: step: 1590/1541, loss: 0.054411593824625015 2023-01-23 11:00:26.859801: step: 1592/1541, loss: 0.0544033981859684 2023-01-23 11:00:27.861133: step: 1594/1541, loss: 0.02075250633060932 2023-01-23 11:00:28.753007: step: 1596/1541, loss: 0.5242583155632019 2023-01-23 11:00:29.634568: step: 1598/1541, loss: 0.15119582414627075 2023-01-23 11:00:30.528355: step: 1600/1541, loss: 0.1011592447757721 2023-01-23 11:00:31.396552: step: 1602/1541, loss: 0.0675939992070198 2023-01-23 11:00:32.297052: step: 1604/1541, loss: 0.08243611454963684 2023-01-23 11:00:33.219568: step: 1606/1541, loss: 0.3155405819416046 2023-01-23 11:00:34.143070: step: 1608/1541, loss: 0.04124213755130768 2023-01-23 11:00:35.063662: step: 1610/1541, loss: 0.0588558055460453 2023-01-23 11:00:35.977104: step: 1612/1541, loss: 0.3842501640319824 2023-01-23 11:00:36.870163: step: 1614/1541, loss: 0.0417737253010273 2023-01-23 11:00:37.765930: step: 1616/1541, loss: 0.16343167424201965 2023-01-23 11:00:38.620823: step: 1618/1541, loss: 0.15732993185520172 2023-01-23 11:00:39.464448: step: 1620/1541, loss: 0.08604726195335388 2023-01-23 11:00:40.388560: step: 1622/1541, loss: 0.13300469517707825 2023-01-23 11:00:41.286084: step: 1624/1541, loss: 0.0689898207783699 2023-01-23 11:00:42.178305: step: 1626/1541, loss: 0.07118257135152817 2023-01-23 11:00:43.086818: step: 1628/1541, loss: 0.026104995980858803 2023-01-23 11:00:44.022465: step: 1630/1541, loss: 0.06847181171178818 2023-01-23 11:00:44.936161: step: 1632/1541, loss: 0.0898415595293045 2023-01-23 11:00:45.834845: step: 1634/1541, loss: 0.0914369747042656 2023-01-23 11:00:46.736926: step: 1636/1541, loss: 0.013612725771963596 2023-01-23 11:00:47.630493: step: 1638/1541, loss: 0.014569908380508423 2023-01-23 11:00:48.505068: step: 1640/1541, loss: 0.02112538181245327 2023-01-23 11:00:49.425597: step: 1642/1541, loss: 0.24206241965293884 2023-01-23 11:00:50.297156: step: 1644/1541, loss: 0.2451934665441513 2023-01-23 11:00:51.179869: step: 1646/1541, loss: 0.2621817886829376 2023-01-23 11:00:52.045352: step: 1648/1541, loss: 0.24191135168075562 2023-01-23 11:00:52.901172: step: 1650/1541, loss: 0.003166566137224436 2023-01-23 11:00:53.777647: step: 1652/1541, loss: 0.16761130094528198 2023-01-23 11:00:54.669725: step: 1654/1541, loss: 0.35816898941993713 2023-01-23 11:00:55.596718: step: 1656/1541, loss: 0.03243811056017876 2023-01-23 11:00:56.554754: step: 1658/1541, loss: 0.009745446965098381 2023-01-23 11:00:57.469061: step: 1660/1541, loss: 0.043314363807439804 2023-01-23 11:00:58.368409: step: 1662/1541, loss: 0.3646855056285858 2023-01-23 11:00:59.237866: step: 1664/1541, loss: 0.05533551424741745 2023-01-23 11:01:00.116990: step: 1666/1541, loss: 0.223558709025383 2023-01-23 11:01:01.012287: step: 1668/1541, loss: 0.10118541866540909 2023-01-23 11:01:01.961086: step: 1670/1541, loss: 0.2905830442905426 2023-01-23 11:01:02.929770: step: 1672/1541, loss: 0.1949380487203598 2023-01-23 11:01:03.825796: step: 1674/1541, loss: 0.015226575545966625 2023-01-23 11:01:04.760141: step: 1676/1541, loss: 0.13695639371871948 2023-01-23 11:01:05.744897: step: 1678/1541, loss: 0.08148501068353653 2023-01-23 11:01:06.658500: step: 1680/1541, loss: 0.06925436109304428 2023-01-23 11:01:07.560809: step: 1682/1541, loss: 0.6783674359321594 2023-01-23 11:01:08.439412: step: 1684/1541, loss: 0.035039741545915604 2023-01-23 11:01:09.344139: step: 1686/1541, loss: 0.34180954098701477 2023-01-23 11:01:10.340314: step: 1688/1541, loss: 0.013559047132730484 2023-01-23 11:01:11.202805: step: 1690/1541, loss: 0.10891636461019516 2023-01-23 11:01:12.083403: step: 1692/1541, loss: 0.08152865618467331 2023-01-23 11:01:13.003085: step: 1694/1541, loss: 0.06971461325883865 2023-01-23 11:01:13.879142: step: 1696/1541, loss: 0.8472903966903687 2023-01-23 11:01:14.790759: step: 1698/1541, loss: 0.23726671934127808 2023-01-23 11:01:15.670182: step: 1700/1541, loss: 0.01773732714354992 2023-01-23 11:01:16.659018: step: 1702/1541, loss: 0.1588200032711029 2023-01-23 11:01:17.595961: step: 1704/1541, loss: 0.3271508812904358 2023-01-23 11:01:18.565816: step: 1706/1541, loss: 0.10746900737285614 2023-01-23 11:01:19.437153: step: 1708/1541, loss: 0.11856704950332642 2023-01-23 11:01:20.336050: step: 1710/1541, loss: 0.04323194548487663 2023-01-23 11:01:21.248543: step: 1712/1541, loss: 0.016109194606542587 2023-01-23 11:01:22.114072: step: 1714/1541, loss: 0.26085442304611206 2023-01-23 11:01:23.002533: step: 1716/1541, loss: 0.3650035858154297 2023-01-23 11:01:23.987994: step: 1718/1541, loss: 0.005947379395365715 2023-01-23 11:01:24.904517: step: 1720/1541, loss: 0.09179896116256714 2023-01-23 11:01:25.887188: step: 1722/1541, loss: 0.023784738034009933 2023-01-23 11:01:26.746751: step: 1724/1541, loss: 0.2913505733013153 2023-01-23 11:01:27.666638: step: 1726/1541, loss: 0.1923646330833435 2023-01-23 11:01:28.565319: step: 1728/1541, loss: 0.15678094327449799 2023-01-23 11:01:29.447030: step: 1730/1541, loss: 0.0764213278889656 2023-01-23 11:01:30.375116: step: 1732/1541, loss: 0.013634396716952324 2023-01-23 11:01:31.285487: step: 1734/1541, loss: 0.44894781708717346 2023-01-23 11:01:32.214712: step: 1736/1541, loss: 0.13052599132061005 2023-01-23 11:01:33.144937: step: 1738/1541, loss: 0.11331914365291595 2023-01-23 11:01:34.022951: step: 1740/1541, loss: 0.1307620406150818 2023-01-23 11:01:34.914232: step: 1742/1541, loss: 0.250838041305542 2023-01-23 11:01:35.789942: step: 1744/1541, loss: 0.03137335181236267 2023-01-23 11:01:36.691514: step: 1746/1541, loss: 0.11094703525304794 2023-01-23 11:01:37.628618: step: 1748/1541, loss: 0.09324172139167786 2023-01-23 11:01:38.525525: step: 1750/1541, loss: 0.014481582678854465 2023-01-23 11:01:39.430030: step: 1752/1541, loss: 0.01727287843823433 2023-01-23 11:01:40.382289: step: 1754/1541, loss: 0.12303290516138077 2023-01-23 11:01:41.274098: step: 1756/1541, loss: 0.2682551443576813 2023-01-23 11:01:42.139382: step: 1758/1541, loss: 0.038000501692295074 2023-01-23 11:01:43.070289: step: 1760/1541, loss: 0.5132454633712769 2023-01-23 11:01:43.965573: step: 1762/1541, loss: 0.09941567480564117 2023-01-23 11:01:44.845150: step: 1764/1541, loss: 0.16811059415340424 2023-01-23 11:01:45.785678: step: 1766/1541, loss: 0.13575083017349243 2023-01-23 11:01:46.724432: step: 1768/1541, loss: 0.18924084305763245 2023-01-23 11:01:47.616326: step: 1770/1541, loss: 0.12518799304962158 2023-01-23 11:01:48.533361: step: 1772/1541, loss: 0.25413429737091064 2023-01-23 11:01:49.413218: step: 1774/1541, loss: 0.3271201550960541 2023-01-23 11:01:50.305834: step: 1776/1541, loss: 0.19082331657409668 2023-01-23 11:01:51.182385: step: 1778/1541, loss: 0.06273806840181351 2023-01-23 11:01:52.171604: step: 1780/1541, loss: 0.14259879291057587 2023-01-23 11:01:53.052638: step: 1782/1541, loss: 0.14382699131965637 2023-01-23 11:01:53.962712: step: 1784/1541, loss: 0.33593735098838806 2023-01-23 11:01:54.946187: step: 1786/1541, loss: 0.2548961043357849 2023-01-23 11:01:55.821670: step: 1788/1541, loss: 0.20432214438915253 2023-01-23 11:01:56.710716: step: 1790/1541, loss: 0.055709272623062134 2023-01-23 11:01:57.647804: step: 1792/1541, loss: 0.04250160604715347 2023-01-23 11:01:58.523453: step: 1794/1541, loss: 0.1416928619146347 2023-01-23 11:01:59.466692: step: 1796/1541, loss: 0.3451633155345917 2023-01-23 11:02:00.410357: step: 1798/1541, loss: 0.2113034576177597 2023-01-23 11:02:01.312819: step: 1800/1541, loss: 0.07934883236885071 2023-01-23 11:02:02.308956: step: 1802/1541, loss: 0.1893548220396042 2023-01-23 11:02:03.170685: step: 1804/1541, loss: 0.026928970590233803 2023-01-23 11:02:04.080555: step: 1806/1541, loss: 0.6213924884796143 2023-01-23 11:02:04.967218: step: 1808/1541, loss: 0.18789096176624298 2023-01-23 11:02:05.834809: step: 1810/1541, loss: 0.18088877201080322 2023-01-23 11:02:06.729589: step: 1812/1541, loss: 0.14714856445789337 2023-01-23 11:02:07.633869: step: 1814/1541, loss: 0.07001223415136337 2023-01-23 11:02:08.520190: step: 1816/1541, loss: 1.1074044704437256 2023-01-23 11:02:09.409255: step: 1818/1541, loss: 0.11079394817352295 2023-01-23 11:02:10.413536: step: 1820/1541, loss: 0.034513529390096664 2023-01-23 11:02:11.403100: step: 1822/1541, loss: 0.03653935715556145 2023-01-23 11:02:12.322738: step: 1824/1541, loss: 1.741828441619873 2023-01-23 11:02:13.276642: step: 1826/1541, loss: 0.2041463553905487 2023-01-23 11:02:14.168153: step: 1828/1541, loss: 0.2955499291419983 2023-01-23 11:02:15.082470: step: 1830/1541, loss: 0.016898352652788162 2023-01-23 11:02:16.030438: step: 1832/1541, loss: 0.10195890069007874 2023-01-23 11:02:16.908880: step: 1834/1541, loss: 1.2733376026153564 2023-01-23 11:02:17.830080: step: 1836/1541, loss: 0.16693569719791412 2023-01-23 11:02:18.744862: step: 1838/1541, loss: 0.08192678540945053 2023-01-23 11:02:19.644147: step: 1840/1541, loss: 0.11515285074710846 2023-01-23 11:02:20.505611: step: 1842/1541, loss: 0.04113754630088806 2023-01-23 11:02:21.454475: step: 1844/1541, loss: 0.07423967868089676 2023-01-23 11:02:22.364942: step: 1846/1541, loss: 0.04607606306672096 2023-01-23 11:02:23.262538: step: 1848/1541, loss: 0.8024094104766846 2023-01-23 11:02:24.185782: step: 1850/1541, loss: 0.3755791485309601 2023-01-23 11:02:25.067780: step: 1852/1541, loss: 0.21863120794296265 2023-01-23 11:02:25.996281: step: 1854/1541, loss: 0.8221717476844788 2023-01-23 11:02:26.906160: step: 1856/1541, loss: 1.28821861743927 2023-01-23 11:02:27.840777: step: 1858/1541, loss: 0.07603518664836884 2023-01-23 11:02:28.715788: step: 1860/1541, loss: 0.06296530365943909 2023-01-23 11:02:29.641279: step: 1862/1541, loss: 1.3419936895370483 2023-01-23 11:02:30.517043: step: 1864/1541, loss: 0.21623282134532928 2023-01-23 11:02:31.385609: step: 1866/1541, loss: 0.0852448120713234 2023-01-23 11:02:32.295106: step: 1868/1541, loss: 0.04902555048465729 2023-01-23 11:02:33.170789: step: 1870/1541, loss: 0.35602062940597534 2023-01-23 11:02:34.067323: step: 1872/1541, loss: 0.16429851949214935 2023-01-23 11:02:34.944464: step: 1874/1541, loss: 0.173233300447464 2023-01-23 11:02:35.832604: step: 1876/1541, loss: 0.187321275472641 2023-01-23 11:02:36.717837: step: 1878/1541, loss: 0.05109292268753052 2023-01-23 11:02:37.633737: step: 1880/1541, loss: 0.02327880449593067 2023-01-23 11:02:38.549257: step: 1882/1541, loss: 0.04679261893033981 2023-01-23 11:02:39.498204: step: 1884/1541, loss: 1.047922968864441 2023-01-23 11:02:40.386959: step: 1886/1541, loss: 0.19569699466228485 2023-01-23 11:02:41.237321: step: 1888/1541, loss: 0.023467618972063065 2023-01-23 11:02:42.117002: step: 1890/1541, loss: 0.15819066762924194 2023-01-23 11:02:43.030683: step: 1892/1541, loss: 0.2067972868680954 2023-01-23 11:02:43.898630: step: 1894/1541, loss: 0.4895990192890167 2023-01-23 11:02:44.789562: step: 1896/1541, loss: 0.6050904989242554 2023-01-23 11:02:45.673915: step: 1898/1541, loss: 0.29567545652389526 2023-01-23 11:02:46.586221: step: 1900/1541, loss: 0.01685485430061817 2023-01-23 11:02:47.505674: step: 1902/1541, loss: 0.030659059062600136 2023-01-23 11:02:48.455908: step: 1904/1541, loss: 0.1370440125465393 2023-01-23 11:02:49.376536: step: 1906/1541, loss: 0.3099104166030884 2023-01-23 11:02:50.237070: step: 1908/1541, loss: 0.07666443288326263 2023-01-23 11:02:51.157764: step: 1910/1541, loss: 0.07818277180194855 2023-01-23 11:02:52.090633: step: 1912/1541, loss: 0.051613472402095795 2023-01-23 11:02:52.976460: step: 1914/1541, loss: 0.12847654521465302 2023-01-23 11:02:53.960331: step: 1916/1541, loss: 0.11227167397737503 2023-01-23 11:02:54.831660: step: 1918/1541, loss: 0.3772585391998291 2023-01-23 11:02:55.700129: step: 1920/1541, loss: 0.34424251317977905 2023-01-23 11:02:56.569012: step: 1922/1541, loss: 0.14945785701274872 2023-01-23 11:02:57.423056: step: 1924/1541, loss: 0.09577111154794693 2023-01-23 11:02:58.386696: step: 1926/1541, loss: 0.11673615127801895 2023-01-23 11:02:59.271512: step: 1928/1541, loss: 0.5964233875274658 2023-01-23 11:03:00.163477: step: 1930/1541, loss: 0.01693265698850155 2023-01-23 11:03:01.047427: step: 1932/1541, loss: 0.34652993083000183 2023-01-23 11:03:01.968249: step: 1934/1541, loss: 0.018713487312197685 2023-01-23 11:03:02.905664: step: 1936/1541, loss: 0.2000000923871994 2023-01-23 11:03:03.822615: step: 1938/1541, loss: 0.00822274200618267 2023-01-23 11:03:04.736223: step: 1940/1541, loss: 0.6199573278427124 2023-01-23 11:03:05.680735: step: 1942/1541, loss: 0.027331141754984856 2023-01-23 11:03:06.571315: step: 1944/1541, loss: 0.3765419125556946 2023-01-23 11:03:07.489362: step: 1946/1541, loss: 0.33684346079826355 2023-01-23 11:03:08.464171: step: 1948/1541, loss: 0.0893535241484642 2023-01-23 11:03:09.341914: step: 1950/1541, loss: 0.07797451317310333 2023-01-23 11:03:10.199467: step: 1952/1541, loss: 0.1773868352174759 2023-01-23 11:03:11.096001: step: 1954/1541, loss: 1.1162925958633423 2023-01-23 11:03:11.999163: step: 1956/1541, loss: 0.08581340312957764 2023-01-23 11:03:12.903126: step: 1958/1541, loss: 0.6481606960296631 2023-01-23 11:03:13.816330: step: 1960/1541, loss: 0.08359938114881516 2023-01-23 11:03:14.677332: step: 1962/1541, loss: 0.05070237070322037 2023-01-23 11:03:15.599143: step: 1964/1541, loss: 0.24613559246063232 2023-01-23 11:03:16.602441: step: 1966/1541, loss: 0.6456906795501709 2023-01-23 11:03:17.471989: step: 1968/1541, loss: 0.3777046501636505 2023-01-23 11:03:18.340132: step: 1970/1541, loss: 0.296177476644516 2023-01-23 11:03:19.271121: step: 1972/1541, loss: 0.11486614495515823 2023-01-23 11:03:20.215436: step: 1974/1541, loss: 0.09014494717121124 2023-01-23 11:03:21.167649: step: 1976/1541, loss: 0.07121649384498596 2023-01-23 11:03:22.115185: step: 1978/1541, loss: 0.012099937535822392 2023-01-23 11:03:22.991781: step: 1980/1541, loss: 0.06488029658794403 2023-01-23 11:03:23.897991: step: 1982/1541, loss: 2.758394479751587 2023-01-23 11:03:24.812457: step: 1984/1541, loss: 0.5834335088729858 2023-01-23 11:03:25.679788: step: 1986/1541, loss: 0.31518206000328064 2023-01-23 11:03:26.588435: step: 1988/1541, loss: 0.2573487460613251 2023-01-23 11:03:27.501473: step: 1990/1541, loss: 0.25236520171165466 2023-01-23 11:03:28.393819: step: 1992/1541, loss: 0.0582737922668457 2023-01-23 11:03:29.290148: step: 1994/1541, loss: 0.11503811180591583 2023-01-23 11:03:30.187471: step: 1996/1541, loss: 0.04557742923498154 2023-01-23 11:03:31.083370: step: 1998/1541, loss: 0.13510285317897797 2023-01-23 11:03:31.973288: step: 2000/1541, loss: 0.02352667599916458 2023-01-23 11:03:32.864825: step: 2002/1541, loss: 0.09846852719783783 2023-01-23 11:03:33.728560: step: 2004/1541, loss: 0.16028860211372375 2023-01-23 11:03:34.650615: step: 2006/1541, loss: 0.3530903458595276 2023-01-23 11:03:35.672544: step: 2008/1541, loss: 0.3261334300041199 2023-01-23 11:03:36.566953: step: 2010/1541, loss: 0.012795766815543175 2023-01-23 11:03:37.434506: step: 2012/1541, loss: 0.25085562467575073 2023-01-23 11:03:38.336025: step: 2014/1541, loss: 0.08892220258712769 2023-01-23 11:03:39.270203: step: 2016/1541, loss: 0.15827767550945282 2023-01-23 11:03:40.201358: step: 2018/1541, loss: 0.12795288860797882 2023-01-23 11:03:41.091278: step: 2020/1541, loss: 0.009941202588379383 2023-01-23 11:03:41.976035: step: 2022/1541, loss: 0.049591053277254105 2023-01-23 11:03:42.902144: step: 2024/1541, loss: 0.19943922758102417 2023-01-23 11:03:43.776752: step: 2026/1541, loss: 0.16513237357139587 2023-01-23 11:03:44.656915: step: 2028/1541, loss: 0.05916508287191391 2023-01-23 11:03:45.560945: step: 2030/1541, loss: 0.4446491003036499 2023-01-23 11:03:46.441982: step: 2032/1541, loss: 1.4873250722885132 2023-01-23 11:03:47.353097: step: 2034/1541, loss: 0.09881212562322617 2023-01-23 11:03:48.254561: step: 2036/1541, loss: 1.3887858390808105 2023-01-23 11:03:49.184170: step: 2038/1541, loss: 0.08962878584861755 2023-01-23 11:03:50.078065: step: 2040/1541, loss: 0.33421358466148376 2023-01-23 11:03:51.021605: step: 2042/1541, loss: 0.09967255592346191 2023-01-23 11:03:51.944506: step: 2044/1541, loss: 0.1610529124736786 2023-01-23 11:03:52.836965: step: 2046/1541, loss: 0.15965424478054047 2023-01-23 11:03:53.739227: step: 2048/1541, loss: 0.054042935371398926 2023-01-23 11:03:54.630215: step: 2050/1541, loss: 0.22849994897842407 2023-01-23 11:03:55.522250: step: 2052/1541, loss: 0.034839559346437454 2023-01-23 11:03:56.421477: step: 2054/1541, loss: 0.08908480405807495 2023-01-23 11:03:57.347528: step: 2056/1541, loss: 0.15194907784461975 2023-01-23 11:03:58.265739: step: 2058/1541, loss: 0.1672554761171341 2023-01-23 11:03:59.160006: step: 2060/1541, loss: 0.07415025681257248 2023-01-23 11:04:00.138460: step: 2062/1541, loss: 0.18320225179195404 2023-01-23 11:04:01.031400: step: 2064/1541, loss: 0.250668466091156 2023-01-23 11:04:01.952816: step: 2066/1541, loss: 0.05023840442299843 2023-01-23 11:04:02.883935: step: 2068/1541, loss: 0.2293071448802948 2023-01-23 11:04:03.759082: step: 2070/1541, loss: 0.4193764328956604 2023-01-23 11:04:04.676653: step: 2072/1541, loss: 0.11747795343399048 2023-01-23 11:04:05.573547: step: 2074/1541, loss: 0.13108617067337036 2023-01-23 11:04:06.455456: step: 2076/1541, loss: 0.4022795259952545 2023-01-23 11:04:07.379208: step: 2078/1541, loss: 0.9340699911117554 2023-01-23 11:04:08.325160: step: 2080/1541, loss: 0.27531853318214417 2023-01-23 11:04:09.205800: step: 2082/1541, loss: 0.11266674846410751 2023-01-23 11:04:10.115866: step: 2084/1541, loss: 0.518937885761261 2023-01-23 11:04:10.988605: step: 2086/1541, loss: 0.3893969655036926 2023-01-23 11:04:11.924039: step: 2088/1541, loss: 0.30716991424560547 2023-01-23 11:04:12.844405: step: 2090/1541, loss: 0.04198158159852028 2023-01-23 11:04:13.774048: step: 2092/1541, loss: 0.3329947888851166 2023-01-23 11:04:14.678681: step: 2094/1541, loss: 0.1525053083896637 2023-01-23 11:04:15.578328: step: 2096/1541, loss: 0.11742618680000305 2023-01-23 11:04:16.442515: step: 2098/1541, loss: 0.2857320010662079 2023-01-23 11:04:17.394404: step: 2100/1541, loss: 0.10318553447723389 2023-01-23 11:04:18.267543: step: 2102/1541, loss: 0.08675916492938995 2023-01-23 11:04:19.127757: step: 2104/1541, loss: 0.08312838524580002 2023-01-23 11:04:20.024319: step: 2106/1541, loss: 0.3109945058822632 2023-01-23 11:04:20.922132: step: 2108/1541, loss: 0.26785871386528015 2023-01-23 11:04:21.773178: step: 2110/1541, loss: 0.22293606400489807 2023-01-23 11:04:22.634342: step: 2112/1541, loss: 0.018284514546394348 2023-01-23 11:04:23.514771: step: 2114/1541, loss: 0.015705369412899017 2023-01-23 11:04:24.388697: step: 2116/1541, loss: 0.01339444238692522 2023-01-23 11:04:25.257495: step: 2118/1541, loss: 0.10089622437953949 2023-01-23 11:04:26.128257: step: 2120/1541, loss: 0.6453135013580322 2023-01-23 11:04:27.035158: step: 2122/1541, loss: 0.4435846507549286 2023-01-23 11:04:27.967110: step: 2124/1541, loss: 0.04993734881281853 2023-01-23 11:04:28.931759: step: 2126/1541, loss: 0.1825350821018219 2023-01-23 11:04:29.803769: step: 2128/1541, loss: 0.04556751996278763 2023-01-23 11:04:30.685401: step: 2130/1541, loss: 0.10481772571802139 2023-01-23 11:04:31.559063: step: 2132/1541, loss: 0.13419775664806366 2023-01-23 11:04:32.459391: step: 2134/1541, loss: 0.13044647872447968 2023-01-23 11:04:33.371840: step: 2136/1541, loss: 0.03196109086275101 2023-01-23 11:04:34.289014: step: 2138/1541, loss: 0.36142390966415405 2023-01-23 11:04:35.168722: step: 2140/1541, loss: 0.2082597017288208 2023-01-23 11:04:36.039352: step: 2142/1541, loss: 0.08176276832818985 2023-01-23 11:04:36.941136: step: 2144/1541, loss: 0.11144772171974182 2023-01-23 11:04:37.800176: step: 2146/1541, loss: 0.5537014007568359 2023-01-23 11:04:38.710102: step: 2148/1541, loss: 0.13468140363693237 2023-01-23 11:04:39.568350: step: 2150/1541, loss: 0.0109969861805439 2023-01-23 11:04:40.493512: step: 2152/1541, loss: 0.09875823557376862 2023-01-23 11:04:41.430919: step: 2154/1541, loss: 0.09517363458871841 2023-01-23 11:04:42.393610: step: 2156/1541, loss: 0.04491374269127846 2023-01-23 11:04:43.284848: step: 2158/1541, loss: 0.4439712464809418 2023-01-23 11:04:44.151621: step: 2160/1541, loss: 0.06209130585193634 2023-01-23 11:04:45.050647: step: 2162/1541, loss: 0.08903606981039047 2023-01-23 11:04:45.930185: step: 2164/1541, loss: 0.03199990093708038 2023-01-23 11:04:46.878725: step: 2166/1541, loss: 0.24724845588207245 2023-01-23 11:04:47.752562: step: 2168/1541, loss: 0.296159952878952 2023-01-23 11:04:48.625292: step: 2170/1541, loss: 0.029221268370747566 2023-01-23 11:04:49.580186: step: 2172/1541, loss: 0.6098604202270508 2023-01-23 11:04:50.450643: step: 2174/1541, loss: 0.855502724647522 2023-01-23 11:04:51.410989: step: 2176/1541, loss: 0.004204849246889353 2023-01-23 11:04:52.303341: step: 2178/1541, loss: 0.06199491769075394 2023-01-23 11:04:53.205757: step: 2180/1541, loss: 0.12244786322116852 2023-01-23 11:04:54.093061: step: 2182/1541, loss: 0.14076213538646698 2023-01-23 11:04:55.082563: step: 2184/1541, loss: 0.11512894183397293 2023-01-23 11:04:55.955177: step: 2186/1541, loss: 0.10878084599971771 2023-01-23 11:04:56.810478: step: 2188/1541, loss: 0.0301212165504694 2023-01-23 11:04:57.756017: step: 2190/1541, loss: 0.24315889179706573 2023-01-23 11:04:58.662800: step: 2192/1541, loss: 0.3761451542377472 2023-01-23 11:04:59.619817: step: 2194/1541, loss: 0.17064091563224792 2023-01-23 11:05:00.549907: step: 2196/1541, loss: 0.2881818413734436 2023-01-23 11:05:01.403650: step: 2198/1541, loss: 0.9329718351364136 2023-01-23 11:05:02.245115: step: 2200/1541, loss: 0.011705071665346622 2023-01-23 11:05:03.192618: step: 2202/1541, loss: 0.1270173192024231 2023-01-23 11:05:04.141681: step: 2204/1541, loss: 0.15605562925338745 2023-01-23 11:05:05.079307: step: 2206/1541, loss: 0.0317501537501812 2023-01-23 11:05:05.969475: step: 2208/1541, loss: 0.8108746409416199 2023-01-23 11:05:06.882548: step: 2210/1541, loss: 0.16134190559387207 2023-01-23 11:05:07.782811: step: 2212/1541, loss: 0.004534564912319183 2023-01-23 11:05:08.689759: step: 2214/1541, loss: 0.08832483738660812 2023-01-23 11:05:09.678423: step: 2216/1541, loss: 0.09714509546756744 2023-01-23 11:05:10.564229: step: 2218/1541, loss: 0.043130408972501755 2023-01-23 11:05:11.458633: step: 2220/1541, loss: 0.13383544981479645 2023-01-23 11:05:12.339126: step: 2222/1541, loss: 0.43780282139778137 2023-01-23 11:05:13.260192: step: 2224/1541, loss: 0.3060995638370514 2023-01-23 11:05:14.134908: step: 2226/1541, loss: 0.016972053796052933 2023-01-23 11:05:15.000470: step: 2228/1541, loss: 0.046183861792087555 2023-01-23 11:05:15.845663: step: 2230/1541, loss: 2.5761125087738037 2023-01-23 11:05:16.729872: step: 2232/1541, loss: 0.24091430008411407 2023-01-23 11:05:17.612613: step: 2234/1541, loss: 0.09958507120609283 2023-01-23 11:05:18.524127: step: 2236/1541, loss: 0.14991013705730438 2023-01-23 11:05:19.438456: step: 2238/1541, loss: 0.16709299385547638 2023-01-23 11:05:20.329127: step: 2240/1541, loss: 0.04149411991238594 2023-01-23 11:05:21.221599: step: 2242/1541, loss: 0.08508624881505966 2023-01-23 11:05:22.129205: step: 2244/1541, loss: 0.047387734055519104 2023-01-23 11:05:23.032565: step: 2246/1541, loss: 0.0749465674161911 2023-01-23 11:05:23.936830: step: 2248/1541, loss: 0.030733197927474976 2023-01-23 11:05:24.873908: step: 2250/1541, loss: 0.030899494886398315 2023-01-23 11:05:25.854721: step: 2252/1541, loss: 0.042779870331287384 2023-01-23 11:05:26.783203: step: 2254/1541, loss: 0.029526103287935257 2023-01-23 11:05:27.649182: step: 2256/1541, loss: 0.2616056501865387 2023-01-23 11:05:28.534353: step: 2258/1541, loss: 2.5019495487213135 2023-01-23 11:05:29.386853: step: 2260/1541, loss: 0.041687335819005966 2023-01-23 11:05:30.265087: step: 2262/1541, loss: 0.08067622780799866 2023-01-23 11:05:31.187462: step: 2264/1541, loss: 0.016891807317733765 2023-01-23 11:05:32.080588: step: 2266/1541, loss: 0.040647219866514206 2023-01-23 11:05:32.979086: step: 2268/1541, loss: 0.20402134954929352 2023-01-23 11:05:33.924028: step: 2270/1541, loss: 0.09906711429357529 2023-01-23 11:05:34.872955: step: 2272/1541, loss: 0.1537947803735733 2023-01-23 11:05:35.762705: step: 2274/1541, loss: 0.1387595534324646 2023-01-23 11:05:36.653767: step: 2276/1541, loss: 0.05183370038866997 2023-01-23 11:05:37.552155: step: 2278/1541, loss: 0.03569239377975464 2023-01-23 11:05:38.436334: step: 2280/1541, loss: 0.029067223891615868 2023-01-23 11:05:39.361225: step: 2282/1541, loss: 0.12950024008750916 2023-01-23 11:05:40.287132: step: 2284/1541, loss: 0.0455692894756794 2023-01-23 11:05:41.231643: step: 2286/1541, loss: 0.02545769140124321 2023-01-23 11:05:42.101572: step: 2288/1541, loss: 0.22002823650836945 2023-01-23 11:05:42.987033: step: 2290/1541, loss: 0.26490509510040283 2023-01-23 11:05:43.949185: step: 2292/1541, loss: 0.42861926555633545 2023-01-23 11:05:44.899734: step: 2294/1541, loss: 0.31849730014801025 2023-01-23 11:05:45.783899: step: 2296/1541, loss: 0.10730482637882233 2023-01-23 11:05:46.680423: step: 2298/1541, loss: 0.8765898942947388 2023-01-23 11:05:47.571988: step: 2300/1541, loss: 0.04656997323036194 2023-01-23 11:05:48.452953: step: 2302/1541, loss: 0.057786691933870316 2023-01-23 11:05:49.411406: step: 2304/1541, loss: 0.007127499207854271 2023-01-23 11:05:50.281235: step: 2306/1541, loss: 0.26785242557525635 2023-01-23 11:05:51.165302: step: 2308/1541, loss: 0.09699708223342896 2023-01-23 11:05:52.052027: step: 2310/1541, loss: 0.06652519106864929 2023-01-23 11:05:52.929666: step: 2312/1541, loss: 0.13806000351905823 2023-01-23 11:05:53.785992: step: 2314/1541, loss: 0.1267741322517395 2023-01-23 11:05:54.684873: step: 2316/1541, loss: 0.02420300990343094 2023-01-23 11:05:55.669362: step: 2318/1541, loss: 0.10072243213653564 2023-01-23 11:05:56.712332: step: 2320/1541, loss: 0.3464707136154175 2023-01-23 11:05:57.585912: step: 2322/1541, loss: 0.13041430711746216 2023-01-23 11:05:58.484682: step: 2324/1541, loss: 0.218301922082901 2023-01-23 11:05:59.366218: step: 2326/1541, loss: 0.06180620193481445 2023-01-23 11:06:00.314433: step: 2328/1541, loss: 0.2496296763420105 2023-01-23 11:06:01.159597: step: 2330/1541, loss: 0.15076085925102234 2023-01-23 11:06:02.030648: step: 2332/1541, loss: 0.006984186824411154 2023-01-23 11:06:02.885126: step: 2334/1541, loss: 0.12182153016328812 2023-01-23 11:06:03.750983: step: 2336/1541, loss: 0.06851451843976974 2023-01-23 11:06:04.625113: step: 2338/1541, loss: 0.04755996912717819 2023-01-23 11:06:05.585580: step: 2340/1541, loss: 0.04375779256224632 2023-01-23 11:06:06.463710: step: 2342/1541, loss: 0.025170916691422462 2023-01-23 11:06:07.304363: step: 2344/1541, loss: 0.096440389752388 2023-01-23 11:06:08.213498: step: 2346/1541, loss: 0.1962190717458725 2023-01-23 11:06:09.073229: step: 2348/1541, loss: 0.15066009759902954 2023-01-23 11:06:09.934494: step: 2350/1541, loss: 0.5384611487388611 2023-01-23 11:06:10.856009: step: 2352/1541, loss: 0.07462149858474731 2023-01-23 11:06:11.770856: step: 2354/1541, loss: 0.032009024173021317 2023-01-23 11:06:12.650642: step: 2356/1541, loss: 0.02745186723768711 2023-01-23 11:06:13.534694: step: 2358/1541, loss: 0.0213660579174757 2023-01-23 11:06:14.465102: step: 2360/1541, loss: 0.09824349731206894 2023-01-23 11:06:15.353821: step: 2362/1541, loss: 0.07550110667943954 2023-01-23 11:06:16.221550: step: 2364/1541, loss: 0.5984411835670471 2023-01-23 11:06:17.100376: step: 2366/1541, loss: 0.0669301450252533 2023-01-23 11:06:17.999902: step: 2368/1541, loss: 0.0059512909501791 2023-01-23 11:06:18.898262: step: 2370/1541, loss: 0.06355179846286774 2023-01-23 11:06:19.757069: step: 2372/1541, loss: 0.05322245880961418 2023-01-23 11:06:20.675930: step: 2374/1541, loss: 0.14857974648475647 2023-01-23 11:06:21.558040: step: 2376/1541, loss: 0.15953360497951508 2023-01-23 11:06:22.427573: step: 2378/1541, loss: 0.04495316743850708 2023-01-23 11:06:23.290222: step: 2380/1541, loss: 0.4761045575141907 2023-01-23 11:06:24.174131: step: 2382/1541, loss: 0.3914478123188019 2023-01-23 11:06:25.073239: step: 2384/1541, loss: 1.0794223546981812 2023-01-23 11:06:25.968327: step: 2386/1541, loss: 0.07697747647762299 2023-01-23 11:06:26.862111: step: 2388/1541, loss: 0.19474585354328156 2023-01-23 11:06:27.740208: step: 2390/1541, loss: 0.07326005399227142 2023-01-23 11:06:28.648902: step: 2392/1541, loss: 0.18527166545391083 2023-01-23 11:06:29.630904: step: 2394/1541, loss: 2.2545289993286133 2023-01-23 11:06:30.535962: step: 2396/1541, loss: 0.3583097755908966 2023-01-23 11:06:31.404842: step: 2398/1541, loss: 0.025586146861314774 2023-01-23 11:06:32.326482: step: 2400/1541, loss: 0.06599605083465576 2023-01-23 11:06:33.252081: step: 2402/1541, loss: 0.10524258017539978 2023-01-23 11:06:34.202283: step: 2404/1541, loss: 0.06397300213575363 2023-01-23 11:06:35.092307: step: 2406/1541, loss: 0.21936902403831482 2023-01-23 11:06:35.972040: step: 2408/1541, loss: 0.0580906942486763 2023-01-23 11:06:36.885176: step: 2410/1541, loss: 0.060184597969055176 2023-01-23 11:06:37.779259: step: 2412/1541, loss: 0.13579948246479034 2023-01-23 11:06:38.678093: step: 2414/1541, loss: 0.18730328977108002 2023-01-23 11:06:39.626361: step: 2416/1541, loss: 0.3038286566734314 2023-01-23 11:06:40.544629: step: 2418/1541, loss: 0.29084357619285583 2023-01-23 11:06:41.428430: step: 2420/1541, loss: 0.05642341449856758 2023-01-23 11:06:42.363372: step: 2422/1541, loss: 0.26762035489082336 2023-01-23 11:06:43.249192: step: 2424/1541, loss: 0.027851426973938942 2023-01-23 11:06:44.118245: step: 2426/1541, loss: 0.0982842668890953 2023-01-23 11:06:45.006953: step: 2428/1541, loss: 0.3017136752605438 2023-01-23 11:06:45.957966: step: 2430/1541, loss: 0.021491898223757744 2023-01-23 11:06:46.928976: step: 2432/1541, loss: 0.2427920401096344 2023-01-23 11:06:47.809631: step: 2434/1541, loss: 0.015278402715921402 2023-01-23 11:06:48.700957: step: 2436/1541, loss: 0.010391056537628174 2023-01-23 11:06:49.584807: step: 2438/1541, loss: 1.1172412633895874 2023-01-23 11:06:50.483836: step: 2440/1541, loss: 0.04018842801451683 2023-01-23 11:06:51.439771: step: 2442/1541, loss: 0.07225695252418518 2023-01-23 11:06:52.354611: step: 2444/1541, loss: 0.11006682366132736 2023-01-23 11:06:53.242076: step: 2446/1541, loss: 0.32305121421813965 2023-01-23 11:06:54.171067: step: 2448/1541, loss: 0.021858062595129013 2023-01-23 11:06:55.074020: step: 2450/1541, loss: 0.1437443047761917 2023-01-23 11:06:55.944101: step: 2452/1541, loss: 0.27067819237709045 2023-01-23 11:06:56.821568: step: 2454/1541, loss: 0.06939520686864853 2023-01-23 11:06:57.753587: step: 2456/1541, loss: 0.055519603192806244 2023-01-23 11:06:58.640165: step: 2458/1541, loss: 0.020380932837724686 2023-01-23 11:06:59.538336: step: 2460/1541, loss: 0.03564048185944557 2023-01-23 11:07:00.507792: step: 2462/1541, loss: 0.08829308301210403 2023-01-23 11:07:01.403001: step: 2464/1541, loss: 0.213111013174057 2023-01-23 11:07:02.292445: step: 2466/1541, loss: 0.02198675274848938 2023-01-23 11:07:03.193207: step: 2468/1541, loss: 0.49724945425987244 2023-01-23 11:07:04.100819: step: 2470/1541, loss: 0.05827978625893593 2023-01-23 11:07:04.998771: step: 2472/1541, loss: 0.0557408407330513 2023-01-23 11:07:05.896414: step: 2474/1541, loss: 0.5011886954307556 2023-01-23 11:07:06.752417: step: 2476/1541, loss: 0.03844990208745003 2023-01-23 11:07:07.640332: step: 2478/1541, loss: 0.15810054540634155 2023-01-23 11:07:08.519537: step: 2480/1541, loss: 0.048010144382715225 2023-01-23 11:07:09.403316: step: 2482/1541, loss: 0.04888276010751724 2023-01-23 11:07:10.312052: step: 2484/1541, loss: 0.2911626100540161 2023-01-23 11:07:11.199146: step: 2486/1541, loss: 0.03822385519742966 2023-01-23 11:07:12.081343: step: 2488/1541, loss: 0.141260027885437 2023-01-23 11:07:12.973896: step: 2490/1541, loss: 0.23360927402973175 2023-01-23 11:07:13.983145: step: 2492/1541, loss: 0.36500170826911926 2023-01-23 11:07:14.890606: step: 2494/1541, loss: 0.13987460732460022 2023-01-23 11:07:15.748008: step: 2496/1541, loss: 0.08613267540931702 2023-01-23 11:07:16.669161: step: 2498/1541, loss: 0.026201728731393814 2023-01-23 11:07:17.542563: step: 2500/1541, loss: 0.13862591981887817 2023-01-23 11:07:18.448402: step: 2502/1541, loss: 1.4329801797866821 2023-01-23 11:07:19.323450: step: 2504/1541, loss: 0.22736713290214539 2023-01-23 11:07:20.229906: step: 2506/1541, loss: 0.055472273379564285 2023-01-23 11:07:21.210651: step: 2508/1541, loss: 0.13425852358341217 2023-01-23 11:07:22.089293: step: 2510/1541, loss: 0.02232218347489834 2023-01-23 11:07:22.977394: step: 2512/1541, loss: 0.09363342821598053 2023-01-23 11:07:23.851577: step: 2514/1541, loss: 0.11166932433843613 2023-01-23 11:07:24.751744: step: 2516/1541, loss: 0.1670263111591339 2023-01-23 11:07:25.619636: step: 2518/1541, loss: 0.23380154371261597 2023-01-23 11:07:26.505463: step: 2520/1541, loss: 0.0516558438539505 2023-01-23 11:07:27.418463: step: 2522/1541, loss: 0.27844175696372986 2023-01-23 11:07:28.343255: step: 2524/1541, loss: 0.5018842220306396 2023-01-23 11:07:29.222078: step: 2526/1541, loss: 0.05215984210371971 2023-01-23 11:07:30.082634: step: 2528/1541, loss: 0.15041527152061462 2023-01-23 11:07:30.948581: step: 2530/1541, loss: 0.0101605374366045 2023-01-23 11:07:31.820698: step: 2532/1541, loss: 0.11954188346862793 2023-01-23 11:07:32.736196: step: 2534/1541, loss: 0.26184460520744324 2023-01-23 11:07:33.639286: step: 2536/1541, loss: 0.11145457625389099 2023-01-23 11:07:34.514326: step: 2538/1541, loss: 0.1486329585313797 2023-01-23 11:07:35.432195: step: 2540/1541, loss: 0.037042852491140366 2023-01-23 11:07:36.319478: step: 2542/1541, loss: 0.5900550484657288 2023-01-23 11:07:37.259248: step: 2544/1541, loss: 0.14948494732379913 2023-01-23 11:07:38.135293: step: 2546/1541, loss: 0.15837575495243073 2023-01-23 11:07:39.027644: step: 2548/1541, loss: 0.11491257697343826 2023-01-23 11:07:39.938623: step: 2550/1541, loss: 0.08024503290653229 2023-01-23 11:07:40.806896: step: 2552/1541, loss: 0.13990621268749237 2023-01-23 11:07:41.714887: step: 2554/1541, loss: 0.13105328381061554 2023-01-23 11:07:42.638743: step: 2556/1541, loss: 0.09562376141548157 2023-01-23 11:07:43.542504: step: 2558/1541, loss: 0.040580712258815765 2023-01-23 11:07:44.409784: step: 2560/1541, loss: 0.1793852299451828 2023-01-23 11:07:45.326743: step: 2562/1541, loss: 0.1835908144712448 2023-01-23 11:07:46.215744: step: 2564/1541, loss: 0.06966163963079453 2023-01-23 11:07:47.098705: step: 2566/1541, loss: 0.27563464641571045 2023-01-23 11:07:48.028777: step: 2568/1541, loss: 0.054939135909080505 2023-01-23 11:07:48.959458: step: 2570/1541, loss: 0.11155836284160614 2023-01-23 11:07:49.875458: step: 2572/1541, loss: 0.3433961570262909 2023-01-23 11:07:50.740330: step: 2574/1541, loss: 0.0688556581735611 2023-01-23 11:07:51.626766: step: 2576/1541, loss: 0.1641780585050583 2023-01-23 11:07:52.623155: step: 2578/1541, loss: 0.2421393096446991 2023-01-23 11:07:53.507198: step: 2580/1541, loss: 0.47306913137435913 2023-01-23 11:07:54.380485: step: 2582/1541, loss: 0.17560307681560516 2023-01-23 11:07:55.255167: step: 2584/1541, loss: 0.08717413991689682 2023-01-23 11:07:56.202147: step: 2586/1541, loss: 0.13861313462257385 2023-01-23 11:07:57.074857: step: 2588/1541, loss: 0.06297850608825684 2023-01-23 11:07:57.940266: step: 2590/1541, loss: 0.24176125228405 2023-01-23 11:07:58.830850: step: 2592/1541, loss: 0.2611216902732849 2023-01-23 11:07:59.792901: step: 2594/1541, loss: 0.07136956602334976 2023-01-23 11:08:00.665686: step: 2596/1541, loss: 0.10866022109985352 2023-01-23 11:08:01.587438: step: 2598/1541, loss: 0.3939763605594635 2023-01-23 11:08:02.545818: step: 2600/1541, loss: 0.06747528910636902 2023-01-23 11:08:03.386553: step: 2602/1541, loss: 0.18591174483299255 2023-01-23 11:08:04.270392: step: 2604/1541, loss: 0.08201992511749268 2023-01-23 11:08:05.194018: step: 2606/1541, loss: 0.07356539368629456 2023-01-23 11:08:06.162293: step: 2608/1541, loss: 0.023127440363168716 2023-01-23 11:08:07.067374: step: 2610/1541, loss: 1.7836387157440186 2023-01-23 11:08:08.009312: step: 2612/1541, loss: 0.11608274281024933 2023-01-23 11:08:09.021382: step: 2614/1541, loss: 0.11975281685590744 2023-01-23 11:08:09.901858: step: 2616/1541, loss: 0.07452870905399323 2023-01-23 11:08:10.782484: step: 2618/1541, loss: 0.06423303484916687 2023-01-23 11:08:11.666374: step: 2620/1541, loss: 0.32987022399902344 2023-01-23 11:08:12.526733: step: 2622/1541, loss: 0.47374752163887024 2023-01-23 11:08:13.420289: step: 2624/1541, loss: 0.4278191924095154 2023-01-23 11:08:14.320665: step: 2626/1541, loss: 0.19417567551136017 2023-01-23 11:08:15.217938: step: 2628/1541, loss: 0.05385741591453552 2023-01-23 11:08:16.119133: step: 2630/1541, loss: 0.2386455088853836 2023-01-23 11:08:17.017598: step: 2632/1541, loss: 0.10771037638187408 2023-01-23 11:08:17.902340: step: 2634/1541, loss: 0.038073696196079254 2023-01-23 11:08:18.774777: step: 2636/1541, loss: 0.029644934460520744 2023-01-23 11:08:19.656377: step: 2638/1541, loss: 0.5125777721405029 2023-01-23 11:08:20.565522: step: 2640/1541, loss: 0.029410945251584053 2023-01-23 11:08:21.432633: step: 2642/1541, loss: 0.17639939486980438 2023-01-23 11:08:22.314244: step: 2644/1541, loss: 0.21275842189788818 2023-01-23 11:08:23.200860: step: 2646/1541, loss: 0.09145958721637726 2023-01-23 11:08:24.108873: step: 2648/1541, loss: 0.0905240923166275 2023-01-23 11:08:25.043057: step: 2650/1541, loss: 0.136661559343338 2023-01-23 11:08:25.959194: step: 2652/1541, loss: 0.08469682931900024 2023-01-23 11:08:26.859841: step: 2654/1541, loss: 0.07131733745336533 2023-01-23 11:08:27.804112: step: 2656/1541, loss: 0.04796261712908745 2023-01-23 11:08:28.697865: step: 2658/1541, loss: 0.3121054768562317 2023-01-23 11:08:29.611331: step: 2660/1541, loss: 0.04797462001442909 2023-01-23 11:08:30.494447: step: 2662/1541, loss: 0.15733325481414795 2023-01-23 11:08:31.465180: step: 2664/1541, loss: 0.14136478304862976 2023-01-23 11:08:32.363576: step: 2666/1541, loss: 0.048607777804136276 2023-01-23 11:08:33.250395: step: 2668/1541, loss: 0.17069454491138458 2023-01-23 11:08:34.127748: step: 2670/1541, loss: 0.2045099437236786 2023-01-23 11:08:35.081462: step: 2672/1541, loss: 0.2743965685367584 2023-01-23 11:08:36.002847: step: 2674/1541, loss: 0.03971172869205475 2023-01-23 11:08:36.908315: step: 2676/1541, loss: 0.01200022641569376 2023-01-23 11:08:37.787577: step: 2678/1541, loss: 0.05217214673757553 2023-01-23 11:08:38.649656: step: 2680/1541, loss: 0.05098145455121994 2023-01-23 11:08:39.506696: step: 2682/1541, loss: 0.018703408539295197 2023-01-23 11:08:40.412320: step: 2684/1541, loss: 0.3150632381439209 2023-01-23 11:08:41.294535: step: 2686/1541, loss: 0.043815769255161285 2023-01-23 11:08:42.172121: step: 2688/1541, loss: 0.07978139817714691 2023-01-23 11:08:43.084579: step: 2690/1541, loss: 0.14938338100910187 2023-01-23 11:08:43.989542: step: 2692/1541, loss: 0.18184618651866913 2023-01-23 11:08:44.896738: step: 2694/1541, loss: 0.09347331523895264 2023-01-23 11:08:45.790497: step: 2696/1541, loss: 0.20901978015899658 2023-01-23 11:08:46.688363: step: 2698/1541, loss: 0.0315992534160614 2023-01-23 11:08:47.564574: step: 2700/1541, loss: 0.014587841928005219 2023-01-23 11:08:48.527667: step: 2702/1541, loss: 0.3332769274711609 2023-01-23 11:08:49.427662: step: 2704/1541, loss: 0.02903798036277294 2023-01-23 11:08:50.354541: step: 2706/1541, loss: 0.07307648658752441 2023-01-23 11:08:51.238528: step: 2708/1541, loss: 0.012935517355799675 2023-01-23 11:08:52.155053: step: 2710/1541, loss: 0.03348632901906967 2023-01-23 11:08:53.141230: step: 2712/1541, loss: 0.21636487543582916 2023-01-23 11:08:54.112159: step: 2714/1541, loss: 0.07009215652942657 2023-01-23 11:08:55.045442: step: 2716/1541, loss: 0.10022993385791779 2023-01-23 11:08:55.880432: step: 2718/1541, loss: 0.12030115723609924 2023-01-23 11:08:56.770337: step: 2720/1541, loss: 0.1546846628189087 2023-01-23 11:08:57.640645: step: 2722/1541, loss: 0.26695841550827026 2023-01-23 11:08:58.496521: step: 2724/1541, loss: 0.014190342277288437 2023-01-23 11:08:59.413029: step: 2726/1541, loss: 0.3737087845802307 2023-01-23 11:09:00.294790: step: 2728/1541, loss: 0.6532816290855408 2023-01-23 11:09:01.192020: step: 2730/1541, loss: 0.08980380743741989 2023-01-23 11:09:02.147527: step: 2732/1541, loss: 0.04786490276455879 2023-01-23 11:09:03.057006: step: 2734/1541, loss: 0.02473190426826477 2023-01-23 11:09:03.949689: step: 2736/1541, loss: 0.40574243664741516 2023-01-23 11:09:04.962778: step: 2738/1541, loss: 0.11797230690717697 2023-01-23 11:09:05.852609: step: 2740/1541, loss: 0.27710530161857605 2023-01-23 11:09:06.775438: step: 2742/1541, loss: 0.23159264028072357 2023-01-23 11:09:07.637274: step: 2744/1541, loss: 0.09598314017057419 2023-01-23 11:09:08.587868: step: 2746/1541, loss: 0.15927785634994507 2023-01-23 11:09:09.497758: step: 2748/1541, loss: 0.1209833100438118 2023-01-23 11:09:10.431231: step: 2750/1541, loss: 0.05551287531852722 2023-01-23 11:09:11.288652: step: 2752/1541, loss: 0.1212279200553894 2023-01-23 11:09:12.159921: step: 2754/1541, loss: 0.07940760999917984 2023-01-23 11:09:13.048947: step: 2756/1541, loss: 0.06971939653158188 2023-01-23 11:09:13.929282: step: 2758/1541, loss: 0.08003591001033783 2023-01-23 11:09:14.807438: step: 2760/1541, loss: 0.05775000900030136 2023-01-23 11:09:15.749637: step: 2762/1541, loss: 0.05495601147413254 2023-01-23 11:09:16.796076: step: 2764/1541, loss: 0.18990465998649597 2023-01-23 11:09:17.663527: step: 2766/1541, loss: 0.05823913589119911 2023-01-23 11:09:18.606299: step: 2768/1541, loss: 0.10635975003242493 2023-01-23 11:09:19.529436: step: 2770/1541, loss: 0.2381718009710312 2023-01-23 11:09:20.475164: step: 2772/1541, loss: 0.09726070612668991 2023-01-23 11:09:21.341138: step: 2774/1541, loss: 0.08224543184041977 2023-01-23 11:09:22.274709: step: 2776/1541, loss: 0.09035901725292206 2023-01-23 11:09:23.150498: step: 2778/1541, loss: 0.025102147832512856 2023-01-23 11:09:24.116828: step: 2780/1541, loss: 0.04399208351969719 2023-01-23 11:09:25.030200: step: 2782/1541, loss: 0.09180182963609695 2023-01-23 11:09:25.915735: step: 2784/1541, loss: 0.14485816657543182 2023-01-23 11:09:26.793560: step: 2786/1541, loss: 0.11098986864089966 2023-01-23 11:09:27.670453: step: 2788/1541, loss: 0.7511041760444641 2023-01-23 11:09:28.565816: step: 2790/1541, loss: 0.40092459321022034 2023-01-23 11:09:29.414809: step: 2792/1541, loss: 0.19539308547973633 2023-01-23 11:09:30.292533: step: 2794/1541, loss: 0.06932687014341354 2023-01-23 11:09:31.220044: step: 2796/1541, loss: 0.16253328323364258 2023-01-23 11:09:32.182863: step: 2798/1541, loss: 0.01749325916171074 2023-01-23 11:09:33.139033: step: 2800/1541, loss: 0.096095971763134 2023-01-23 11:09:34.040373: step: 2802/1541, loss: 0.11841893196105957 2023-01-23 11:09:34.926233: step: 2804/1541, loss: 0.16979843378067017 2023-01-23 11:09:35.951559: step: 2806/1541, loss: 0.05338774621486664 2023-01-23 11:09:36.874817: step: 2808/1541, loss: 0.05301998183131218 2023-01-23 11:09:37.812027: step: 2810/1541, loss: 0.3348749577999115 2023-01-23 11:09:38.730756: step: 2812/1541, loss: 0.5543449521064758 2023-01-23 11:09:39.571577: step: 2814/1541, loss: 0.29889753460884094 2023-01-23 11:09:40.450587: step: 2816/1541, loss: 0.3269478380680084 2023-01-23 11:09:41.313698: step: 2818/1541, loss: 0.11295050382614136 2023-01-23 11:09:42.247405: step: 2820/1541, loss: 0.06735878437757492 2023-01-23 11:09:43.137980: step: 2822/1541, loss: 0.004864061251282692 2023-01-23 11:09:44.075712: step: 2824/1541, loss: 1.1249160766601562 2023-01-23 11:09:44.988560: step: 2826/1541, loss: 0.2634377181529999 2023-01-23 11:09:45.874194: step: 2828/1541, loss: 0.05142390355467796 2023-01-23 11:09:46.784236: step: 2830/1541, loss: 0.04347607120871544 2023-01-23 11:09:47.651583: step: 2832/1541, loss: 1.662239909172058 2023-01-23 11:09:48.531779: step: 2834/1541, loss: 0.014275351539254189 2023-01-23 11:09:49.455157: step: 2836/1541, loss: 0.014723028056323528 2023-01-23 11:09:50.385582: step: 2838/1541, loss: 0.10408998280763626 2023-01-23 11:09:51.321791: step: 2840/1541, loss: 0.03934718295931816 2023-01-23 11:09:52.192545: step: 2842/1541, loss: 0.17485885322093964 2023-01-23 11:09:53.050762: step: 2844/1541, loss: 0.06593652069568634 2023-01-23 11:09:54.004603: step: 2846/1541, loss: 0.07720793038606644 2023-01-23 11:09:54.897190: step: 2848/1541, loss: 0.1334277242422104 2023-01-23 11:09:55.809080: step: 2850/1541, loss: 0.056028641760349274 2023-01-23 11:09:56.697455: step: 2852/1541, loss: 0.3282397985458374 2023-01-23 11:09:57.611492: step: 2854/1541, loss: 0.21727433800697327 2023-01-23 11:09:58.497020: step: 2856/1541, loss: 0.13099732995033264 2023-01-23 11:09:59.359269: step: 2858/1541, loss: 0.05395054072141647 2023-01-23 11:10:00.292379: step: 2860/1541, loss: 0.41381341218948364 2023-01-23 11:10:01.197258: step: 2862/1541, loss: 0.5035332441329956 2023-01-23 11:10:02.092086: step: 2864/1541, loss: 0.0048980191349983215 2023-01-23 11:10:02.992516: step: 2866/1541, loss: 0.2925211191177368 2023-01-23 11:10:03.930571: step: 2868/1541, loss: 0.3033072352409363 2023-01-23 11:10:04.806788: step: 2870/1541, loss: 0.27147382497787476 2023-01-23 11:10:05.676045: step: 2872/1541, loss: 0.033689938485622406 2023-01-23 11:10:06.567850: step: 2874/1541, loss: 0.08621976524591446 2023-01-23 11:10:07.495159: step: 2876/1541, loss: 0.009228628128767014 2023-01-23 11:10:08.411607: step: 2878/1541, loss: 0.1634049117565155 2023-01-23 11:10:09.280153: step: 2880/1541, loss: 0.13028325140476227 2023-01-23 11:10:10.185900: step: 2882/1541, loss: 0.13639436662197113 2023-01-23 11:10:11.110955: step: 2884/1541, loss: 0.030504543334245682 2023-01-23 11:10:12.092315: step: 2886/1541, loss: 0.04603210836648941 2023-01-23 11:10:12.973129: step: 2888/1541, loss: 0.01804303377866745 2023-01-23 11:10:13.923471: step: 2890/1541, loss: 0.08625862002372742 2023-01-23 11:10:14.871256: step: 2892/1541, loss: 0.06155866011977196 2023-01-23 11:10:15.736147: step: 2894/1541, loss: 0.11613480001688004 2023-01-23 11:10:16.617234: step: 2896/1541, loss: 0.08335716277360916 2023-01-23 11:10:17.475395: step: 2898/1541, loss: 0.08187899738550186 2023-01-23 11:10:18.404035: step: 2900/1541, loss: 0.025506267324090004 2023-01-23 11:10:19.359585: step: 2902/1541, loss: 0.2022480070590973 2023-01-23 11:10:20.221953: step: 2904/1541, loss: 0.046982377767562866 2023-01-23 11:10:21.088122: step: 2906/1541, loss: 0.2451368123292923 2023-01-23 11:10:21.990662: step: 2908/1541, loss: 0.161002978682518 2023-01-23 11:10:22.862244: step: 2910/1541, loss: 0.07934209704399109 2023-01-23 11:10:23.748133: step: 2912/1541, loss: 0.06869519501924515 2023-01-23 11:10:24.625852: step: 2914/1541, loss: 0.08552278578281403 2023-01-23 11:10:25.549738: step: 2916/1541, loss: 0.1689458191394806 2023-01-23 11:10:26.470192: step: 2918/1541, loss: 0.1678653061389923 2023-01-23 11:10:27.379503: step: 2920/1541, loss: 0.05705339089035988 2023-01-23 11:10:28.225364: step: 2922/1541, loss: 0.4977562725543976 2023-01-23 11:10:29.172074: step: 2924/1541, loss: 0.017593469470739365 2023-01-23 11:10:30.045290: step: 2926/1541, loss: 0.07444893568754196 2023-01-23 11:10:30.904417: step: 2928/1541, loss: 0.12147696316242218 2023-01-23 11:10:31.800156: step: 2930/1541, loss: 0.7349756956100464 2023-01-23 11:10:32.699667: step: 2932/1541, loss: 0.12423070520162582 2023-01-23 11:10:33.593536: step: 2934/1541, loss: 0.06852922588586807 2023-01-23 11:10:34.480843: step: 2936/1541, loss: 0.3372209668159485 2023-01-23 11:10:35.393131: step: 2938/1541, loss: 0.43078139424324036 2023-01-23 11:10:36.286513: step: 2940/1541, loss: 0.07759422808885574 2023-01-23 11:10:37.237280: step: 2942/1541, loss: 0.06881958246231079 2023-01-23 11:10:38.198495: step: 2944/1541, loss: 1.6946667432785034 2023-01-23 11:10:39.081987: step: 2946/1541, loss: 0.13767360150814056 2023-01-23 11:10:39.940168: step: 2948/1541, loss: 0.15933109819889069 2023-01-23 11:10:40.876377: step: 2950/1541, loss: 0.16987626254558563 2023-01-23 11:10:41.758192: step: 2952/1541, loss: 0.13808028399944305 2023-01-23 11:10:42.694506: step: 2954/1541, loss: 0.10102548450231552 2023-01-23 11:10:43.614888: step: 2956/1541, loss: 0.06805040687322617 2023-01-23 11:10:44.481105: step: 2958/1541, loss: 0.0717669129371643 2023-01-23 11:10:45.393956: step: 2960/1541, loss: 0.05747063830494881 2023-01-23 11:10:46.309530: step: 2962/1541, loss: 0.15190652012825012 2023-01-23 11:10:47.234227: step: 2964/1541, loss: 0.39927253127098083 2023-01-23 11:10:48.109202: step: 2966/1541, loss: 0.5414514541625977 2023-01-23 11:10:48.990998: step: 2968/1541, loss: 0.06620098650455475 2023-01-23 11:10:49.964165: step: 2970/1541, loss: 0.2682337164878845 2023-01-23 11:10:50.860315: step: 2972/1541, loss: 0.14652076363563538 2023-01-23 11:10:51.752370: step: 2974/1541, loss: 0.23080481588840485 2023-01-23 11:10:52.770782: step: 2976/1541, loss: 1.362034559249878 2023-01-23 11:10:53.686716: step: 2978/1541, loss: 0.04519662261009216 2023-01-23 11:10:54.625068: step: 2980/1541, loss: 0.48221898078918457 2023-01-23 11:10:55.574138: step: 2982/1541, loss: 0.30078238248825073 2023-01-23 11:10:56.462184: step: 2984/1541, loss: 0.058635689318180084 2023-01-23 11:10:57.359724: step: 2986/1541, loss: 0.13254255056381226 2023-01-23 11:10:58.266035: step: 2988/1541, loss: 0.1401120275259018 2023-01-23 11:10:59.118195: step: 2990/1541, loss: 0.1259540468454361 2023-01-23 11:11:00.035795: step: 2992/1541, loss: 0.15630784630775452 2023-01-23 11:11:00.946730: step: 2994/1541, loss: 0.2152903825044632 2023-01-23 11:11:01.881020: step: 2996/1541, loss: 0.04444848746061325 2023-01-23 11:11:02.774908: step: 2998/1541, loss: 0.08531522750854492 2023-01-23 11:11:03.663863: step: 3000/1541, loss: 0.03967609629034996 2023-01-23 11:11:04.583735: step: 3002/1541, loss: 0.22141024470329285 2023-01-23 11:11:05.480646: step: 3004/1541, loss: 0.25323057174682617 2023-01-23 11:11:06.367215: step: 3006/1541, loss: 0.07451210170984268 2023-01-23 11:11:07.297071: step: 3008/1541, loss: 0.24111007153987885 2023-01-23 11:11:08.239485: step: 3010/1541, loss: 0.20668472349643707 2023-01-23 11:11:09.146032: step: 3012/1541, loss: 0.5085409879684448 2023-01-23 11:11:10.085679: step: 3014/1541, loss: 0.2564108669757843 2023-01-23 11:11:11.009149: step: 3016/1541, loss: 0.09051638841629028 2023-01-23 11:11:11.976555: step: 3018/1541, loss: 0.12603497505187988 2023-01-23 11:11:12.880211: step: 3020/1541, loss: 0.25847455859184265 2023-01-23 11:11:13.725119: step: 3022/1541, loss: 0.24942456185817719 2023-01-23 11:11:14.625919: step: 3024/1541, loss: 0.1115451231598854 2023-01-23 11:11:15.548726: step: 3026/1541, loss: 0.0226889681071043 2023-01-23 11:11:16.474872: step: 3028/1541, loss: 0.08917687088251114 2023-01-23 11:11:17.362580: step: 3030/1541, loss: 0.16430602967739105 2023-01-23 11:11:18.316666: step: 3032/1541, loss: 0.0817849412560463 2023-01-23 11:11:19.276847: step: 3034/1541, loss: 0.14376592636108398 2023-01-23 11:11:20.182600: step: 3036/1541, loss: 0.5012083053588867 2023-01-23 11:11:21.099200: step: 3038/1541, loss: 0.04243423789739609 2023-01-23 11:11:21.962105: step: 3040/1541, loss: 0.08033332228660583 2023-01-23 11:11:22.889091: step: 3042/1541, loss: 0.12417912483215332 2023-01-23 11:11:23.803257: step: 3044/1541, loss: 0.08676374703645706 2023-01-23 11:11:24.691779: step: 3046/1541, loss: 0.15853099524974823 2023-01-23 11:11:25.547110: step: 3048/1541, loss: 0.23693926632404327 2023-01-23 11:11:26.499155: step: 3050/1541, loss: 0.2844735085964203 2023-01-23 11:11:27.373957: step: 3052/1541, loss: 2.750671625137329 2023-01-23 11:11:28.337476: step: 3054/1541, loss: 0.038221195340156555 2023-01-23 11:11:29.174972: step: 3056/1541, loss: 0.050666794180870056 2023-01-23 11:11:30.090949: step: 3058/1541, loss: 0.22642798721790314 2023-01-23 11:11:30.968635: step: 3060/1541, loss: 0.05228833481669426 2023-01-23 11:11:31.892252: step: 3062/1541, loss: 0.14506611227989197 2023-01-23 11:11:32.790062: step: 3064/1541, loss: 0.5090827345848083 2023-01-23 11:11:33.642079: step: 3066/1541, loss: 0.18375588953495026 2023-01-23 11:11:34.557754: step: 3068/1541, loss: 0.30208835005760193 2023-01-23 11:11:35.424404: step: 3070/1541, loss: 0.12790778279304504 2023-01-23 11:11:36.360856: step: 3072/1541, loss: 0.36293599009513855 2023-01-23 11:11:37.247771: step: 3074/1541, loss: 0.024546440690755844 2023-01-23 11:11:38.117678: step: 3076/1541, loss: 0.07574616372585297 2023-01-23 11:11:39.046360: step: 3078/1541, loss: 0.4047372043132782 2023-01-23 11:11:40.011098: step: 3080/1541, loss: 0.19256722927093506 2023-01-23 11:11:40.905814: step: 3082/1541, loss: 0.741296112537384 2023-01-23 11:11:41.537332: step: 3084/1541, loss: 0.05450074374675751 ================================================== Loss: 0.208 -------------------- Dev Chinese: {'event': {'p': 0.6296296296296297, 'r': 0.7017310252996005, 'f1': 0.663727959697733}, 'argument': {'p': 0.6804853485228216, 'r': 0.6380544616312972, 'f1': 0.6585871876693117}, 'combined': 0.43712273035482024, 'epoch': 12} Test Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.7252073335731545, 'r': 0.6926880109229547, 'f1': 0.7085747581434514}, 'combined': 0.46193280228450156, 'epoch': 12} Dev Korean: {'event': {'p': 0.5774509803921568, 'r': 0.7842876165113183, 'f1': 0.6651609260304913}, 'argument': {'p': 0.6697063337039494, 'r': 0.6480718864557959, 'f1': 0.6587115197613453}, 'combined': 0.4381491644714087, 'epoch': 12} Test Korean: {'event': {'p': 0.6110581506196378, 'r': 0.7681246255242661, 'f1': 0.6806477302893549}, 'argument': {'p': 0.7174422676434833, 'r': 0.7071102925913643, 'f1': 0.7122388123147404}, 'combined': 0.48478373102601385, 'epoch': 12} Dev Russian: {'event': {'p': 0.6575829383886256, 'r': 0.7390146471371505, 'f1': 0.695924764890282}, 'argument': {'p': 0.6818843598661549, 'r': 0.6474301727537625, 'f1': 0.6642107613705911}, 'combined': 0.4622407179444238, 'epoch': 12} Test Russian: {'event': {'p': 0.652537938252224, 'r': 0.7471539844218095, 'f1': 0.6966480446927374}, 'argument': {'p': 0.7175449258880758, 'r': 0.6949126844980577, 'f1': 0.7060474835213332}, 'combined': 0.49186659885536455, 'epoch': 12} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6331168831168831, 'r': 0.6275140788415124, 'f1': 0.6303030303030303}, 'combined': 0.4525252525252525, 'epoch': 12} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.57687338501292, 'r': 0.3263888888888889, 'f1': 0.4169000933706817}, 'combined': 0.24948352044229768, 'epoch': 12} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5577695316551499, 'r': 0.5318267627409569, 'f1': 0.5444893047109796}, 'combined': 0.2800230709942181, 'epoch': 12} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test for Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6727483627478649, 'r': 0.6595770845615362, 'f1': 0.6660976184773536}, 'combined': 0.4593776679154163, 'epoch': 10} Test for Russian: {'event': {'p': 0.6446886446886447, 'r': 0.738166566806471, 'f1': 0.6882681564245811}, 'argument': {'p': 0.7212116816452776, 'r': 0.7139607203935477, 'f1': 0.717567883903295}, 'combined': 0.4938791245636086, 'epoch': 10} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5730551257490913, 'r': 0.6396894426966601, 'f1': 0.6045416711199205}, 'combined': 0.3109071451473877, 'epoch': 10} ****************************** Epoch: 13 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 11:17:02.816937: step: 2/1541, loss: 0.08895432204008102 2023-01-23 11:17:03.703269: step: 4/1541, loss: 0.23698896169662476 2023-01-23 11:17:04.651273: step: 6/1541, loss: 0.20004217326641083 2023-01-23 11:17:05.528817: step: 8/1541, loss: 0.12319144606590271 2023-01-23 11:17:06.433351: step: 10/1541, loss: 0.02877281978726387 2023-01-23 11:17:07.320045: step: 12/1541, loss: 0.05638044327497482 2023-01-23 11:17:08.222496: step: 14/1541, loss: 0.025257252156734467 2023-01-23 11:17:09.123773: step: 16/1541, loss: 0.0038039505016058683 2023-01-23 11:17:10.026987: step: 18/1541, loss: 0.06485405564308167 2023-01-23 11:17:10.949167: step: 20/1541, loss: 0.2161007970571518 2023-01-23 11:17:11.837107: step: 22/1541, loss: 0.20316052436828613 2023-01-23 11:17:12.735804: step: 24/1541, loss: 0.016562016680836678 2023-01-23 11:17:13.685527: step: 26/1541, loss: 0.0036129208747297525 2023-01-23 11:17:14.626897: step: 28/1541, loss: 0.140189066529274 2023-01-23 11:17:15.552064: step: 30/1541, loss: 0.08335771411657333 2023-01-23 11:17:16.461451: step: 32/1541, loss: 0.11652326583862305 2023-01-23 11:17:17.362160: step: 34/1541, loss: 0.03616369515657425 2023-01-23 11:17:18.279968: step: 36/1541, loss: 0.10129816830158234 2023-01-23 11:17:19.179491: step: 38/1541, loss: 0.007501545827835798 2023-01-23 11:17:20.115222: step: 40/1541, loss: 0.06990475952625275 2023-01-23 11:17:20.986939: step: 42/1541, loss: 0.1400817632675171 2023-01-23 11:17:21.887424: step: 44/1541, loss: 0.3315873146057129 2023-01-23 11:17:22.772643: step: 46/1541, loss: 0.2400333434343338 2023-01-23 11:17:23.650226: step: 48/1541, loss: 0.0298782791942358 2023-01-23 11:17:24.573937: step: 50/1541, loss: 0.10194452852010727 2023-01-23 11:17:25.466493: step: 52/1541, loss: 0.0009983073687180877 2023-01-23 11:17:26.444867: step: 54/1541, loss: 0.061052266508340836 2023-01-23 11:17:27.330409: step: 56/1541, loss: 0.20183607935905457 2023-01-23 11:17:28.223546: step: 58/1541, loss: 0.016183532774448395 2023-01-23 11:17:29.074103: step: 60/1541, loss: 0.014510581269860268 2023-01-23 11:17:29.947986: step: 62/1541, loss: 0.08440837264060974 2023-01-23 11:17:30.824607: step: 64/1541, loss: 0.18653760850429535 2023-01-23 11:17:31.684309: step: 66/1541, loss: 0.23043781518936157 2023-01-23 11:17:32.639395: step: 68/1541, loss: 0.02293945848941803 2023-01-23 11:17:33.503321: step: 70/1541, loss: 0.03270460665225983 2023-01-23 11:17:34.369918: step: 72/1541, loss: 0.09160023927688599 2023-01-23 11:17:35.273165: step: 74/1541, loss: 0.19647462666034698 2023-01-23 11:17:36.173011: step: 76/1541, loss: 0.1806938350200653 2023-01-23 11:17:37.061305: step: 78/1541, loss: 0.5492103695869446 2023-01-23 11:17:37.978139: step: 80/1541, loss: 2.456010103225708 2023-01-23 11:17:38.869250: step: 82/1541, loss: 0.6096706986427307 2023-01-23 11:17:39.745253: step: 84/1541, loss: 0.09613046795129776 2023-01-23 11:17:40.621170: step: 86/1541, loss: 0.048382215201854706 2023-01-23 11:17:41.564049: step: 88/1541, loss: 0.12024365365505219 2023-01-23 11:17:42.466437: step: 90/1541, loss: 1.1607218980789185 2023-01-23 11:17:43.323550: step: 92/1541, loss: 0.01519027166068554 2023-01-23 11:17:44.260775: step: 94/1541, loss: 0.28295600414276123 2023-01-23 11:17:45.210917: step: 96/1541, loss: 0.14626586437225342 2023-01-23 11:17:46.113674: step: 98/1541, loss: 0.19387997686862946 2023-01-23 11:17:47.006062: step: 100/1541, loss: 0.10971693694591522 2023-01-23 11:17:47.919918: step: 102/1541, loss: 0.0765812024474144 2023-01-23 11:17:48.900856: step: 104/1541, loss: 0.029261676594614983 2023-01-23 11:17:49.860048: step: 106/1541, loss: 0.09232769906520844 2023-01-23 11:17:50.782199: step: 108/1541, loss: 0.17021600902080536 2023-01-23 11:17:51.698724: step: 110/1541, loss: 0.08449646830558777 2023-01-23 11:17:52.703835: step: 112/1541, loss: 0.0491924025118351 2023-01-23 11:17:53.607961: step: 114/1541, loss: 0.017541497945785522 2023-01-23 11:17:54.562470: step: 116/1541, loss: 0.36928656697273254 2023-01-23 11:17:55.467298: step: 118/1541, loss: 0.07325224578380585 2023-01-23 11:17:56.335593: step: 120/1541, loss: 0.03648043051362038 2023-01-23 11:17:57.252471: step: 122/1541, loss: 0.18333713710308075 2023-01-23 11:17:58.132224: step: 124/1541, loss: 0.03193281218409538 2023-01-23 11:17:59.060144: step: 126/1541, loss: 0.07725208252668381 2023-01-23 11:17:59.951914: step: 128/1541, loss: 0.05634990334510803 2023-01-23 11:18:00.876120: step: 130/1541, loss: 0.12298884242773056 2023-01-23 11:18:01.785559: step: 132/1541, loss: 0.020584946498274803 2023-01-23 11:18:02.658008: step: 134/1541, loss: 0.12408389896154404 2023-01-23 11:18:03.523663: step: 136/1541, loss: 0.26787057518959045 2023-01-23 11:18:04.448154: step: 138/1541, loss: 0.03589794784784317 2023-01-23 11:18:05.349502: step: 140/1541, loss: 0.2013978511095047 2023-01-23 11:18:06.309306: step: 142/1541, loss: 0.10735389590263367 2023-01-23 11:18:07.165086: step: 144/1541, loss: 0.035653017461299896 2023-01-23 11:18:08.031708: step: 146/1541, loss: 0.07923198491334915 2023-01-23 11:18:08.891846: step: 148/1541, loss: 0.1110880896449089 2023-01-23 11:18:09.759559: step: 150/1541, loss: 0.4509876072406769 2023-01-23 11:18:10.661124: step: 152/1541, loss: 0.03620602563023567 2023-01-23 11:18:11.563630: step: 154/1541, loss: 0.07251117378473282 2023-01-23 11:18:12.456828: step: 156/1541, loss: 0.019673332571983337 2023-01-23 11:18:13.354144: step: 158/1541, loss: 0.2767210900783539 2023-01-23 11:18:14.306441: step: 160/1541, loss: 0.050838012248277664 2023-01-23 11:18:15.204347: step: 162/1541, loss: 0.2570236325263977 2023-01-23 11:18:16.137258: step: 164/1541, loss: 0.022590380162000656 2023-01-23 11:18:17.019189: step: 166/1541, loss: 0.005511914379894733 2023-01-23 11:18:17.993575: step: 168/1541, loss: 0.01664450205862522 2023-01-23 11:18:18.934094: step: 170/1541, loss: 0.14972548186779022 2023-01-23 11:18:19.803841: step: 172/1541, loss: 0.21863120794296265 2023-01-23 11:18:20.700127: step: 174/1541, loss: 0.03943130001425743 2023-01-23 11:18:21.652935: step: 176/1541, loss: 0.003807045053690672 2023-01-23 11:18:22.560768: step: 178/1541, loss: 0.4346446394920349 2023-01-23 11:18:23.445951: step: 180/1541, loss: 0.05350134149193764 2023-01-23 11:18:24.381699: step: 182/1541, loss: 0.015563597902655602 2023-01-23 11:18:25.316395: step: 184/1541, loss: 0.13697585463523865 2023-01-23 11:18:26.191069: step: 186/1541, loss: 0.002842263551428914 2023-01-23 11:18:27.091848: step: 188/1541, loss: 0.015239374712109566 2023-01-23 11:18:28.033199: step: 190/1541, loss: 0.20830929279327393 2023-01-23 11:18:28.918314: step: 192/1541, loss: 0.18871299922466278 2023-01-23 11:18:29.850541: step: 194/1541, loss: 0.03849199041724205 2023-01-23 11:18:30.704020: step: 196/1541, loss: 0.08352255821228027 2023-01-23 11:18:31.612449: step: 198/1541, loss: 5.543761253356934 2023-01-23 11:18:32.534285: step: 200/1541, loss: 0.056547146290540695 2023-01-23 11:18:33.485409: step: 202/1541, loss: 0.3331816792488098 2023-01-23 11:18:34.393050: step: 204/1541, loss: 0.05356016382575035 2023-01-23 11:18:35.315397: step: 206/1541, loss: 0.02403419278562069 2023-01-23 11:18:36.175754: step: 208/1541, loss: 0.06780949234962463 2023-01-23 11:18:37.077899: step: 210/1541, loss: 0.054585713893175125 2023-01-23 11:18:38.051935: step: 212/1541, loss: 0.030074629932641983 2023-01-23 11:18:38.948999: step: 214/1541, loss: 0.46480926871299744 2023-01-23 11:18:39.831486: step: 216/1541, loss: 0.01365241501480341 2023-01-23 11:18:40.776973: step: 218/1541, loss: 0.16260169446468353 2023-01-23 11:18:41.662979: step: 220/1541, loss: 0.15381589531898499 2023-01-23 11:18:42.583539: step: 222/1541, loss: 0.03210326284170151 2023-01-23 11:18:43.478576: step: 224/1541, loss: 0.011844818480312824 2023-01-23 11:18:44.403923: step: 226/1541, loss: 0.036159392446279526 2023-01-23 11:18:45.271587: step: 228/1541, loss: 0.19052188098430634 2023-01-23 11:18:46.155829: step: 230/1541, loss: 0.0008674765704199672 2023-01-23 11:18:46.988493: step: 232/1541, loss: 0.2261609584093094 2023-01-23 11:18:47.865113: step: 234/1541, loss: 0.7729452848434448 2023-01-23 11:18:48.741863: step: 236/1541, loss: 0.004885732661932707 2023-01-23 11:18:49.609604: step: 238/1541, loss: 0.04308485984802246 2023-01-23 11:18:50.489322: step: 240/1541, loss: 0.20574931800365448 2023-01-23 11:18:51.435493: step: 242/1541, loss: 0.014375550672411919 2023-01-23 11:18:52.439790: step: 244/1541, loss: 0.12527121603488922 2023-01-23 11:18:53.331635: step: 246/1541, loss: 0.10346010327339172 2023-01-23 11:18:54.238489: step: 248/1541, loss: 0.010226644575595856 2023-01-23 11:18:55.127530: step: 250/1541, loss: 0.06465450674295425 2023-01-23 11:18:56.008446: step: 252/1541, loss: 0.058543331921100616 2023-01-23 11:18:56.925106: step: 254/1541, loss: 0.3184067904949188 2023-01-23 11:18:57.815762: step: 256/1541, loss: 0.011197702027857304 2023-01-23 11:18:58.802528: step: 258/1541, loss: 0.08942911773920059 2023-01-23 11:18:59.701138: step: 260/1541, loss: 0.3360465466976166 2023-01-23 11:19:00.597240: step: 262/1541, loss: 0.007637513801455498 2023-01-23 11:19:01.465848: step: 264/1541, loss: 0.0425519160926342 2023-01-23 11:19:02.366155: step: 266/1541, loss: 0.010739197954535484 2023-01-23 11:19:03.254395: step: 268/1541, loss: 0.4328402876853943 2023-01-23 11:19:04.151862: step: 270/1541, loss: 0.36244654655456543 2023-01-23 11:19:05.042776: step: 272/1541, loss: 0.28261667490005493 2023-01-23 11:19:05.976750: step: 274/1541, loss: 0.15198329091072083 2023-01-23 11:19:06.864288: step: 276/1541, loss: 0.09144853055477142 2023-01-23 11:19:07.743719: step: 278/1541, loss: 0.07790078967809677 2023-01-23 11:19:08.687720: step: 280/1541, loss: 0.07421828806400299 2023-01-23 11:19:09.600976: step: 282/1541, loss: 0.17520849406719208 2023-01-23 11:19:10.493392: step: 284/1541, loss: 0.39892396330833435 2023-01-23 11:19:11.401316: step: 286/1541, loss: 0.0833853930234909 2023-01-23 11:19:12.287250: step: 288/1541, loss: 0.06077932193875313 2023-01-23 11:19:13.186923: step: 290/1541, loss: 1.221421480178833 2023-01-23 11:19:14.128075: step: 292/1541, loss: 0.24391679465770721 2023-01-23 11:19:14.999185: step: 294/1541, loss: 0.16520880162715912 2023-01-23 11:19:15.900044: step: 296/1541, loss: 0.0074324472807347775 2023-01-23 11:19:16.792759: step: 298/1541, loss: 0.03483924642205238 2023-01-23 11:19:17.716696: step: 300/1541, loss: 0.430961936712265 2023-01-23 11:19:18.634374: step: 302/1541, loss: 0.08014573156833649 2023-01-23 11:19:19.542817: step: 304/1541, loss: 0.05621134489774704 2023-01-23 11:19:20.518041: step: 306/1541, loss: 0.40939635038375854 2023-01-23 11:19:21.403062: step: 308/1541, loss: 0.07511640340089798 2023-01-23 11:19:22.346256: step: 310/1541, loss: 0.16316336393356323 2023-01-23 11:19:23.268731: step: 312/1541, loss: 0.28834784030914307 2023-01-23 11:19:24.230515: step: 314/1541, loss: 0.09213297814130783 2023-01-23 11:19:25.120543: step: 316/1541, loss: 0.09016310423612595 2023-01-23 11:19:26.007531: step: 318/1541, loss: 0.05129255726933479 2023-01-23 11:19:26.892775: step: 320/1541, loss: 1.6406705379486084 2023-01-23 11:19:27.846229: step: 322/1541, loss: 0.28852513432502747 2023-01-23 11:19:28.734959: step: 324/1541, loss: 0.08985734730958939 2023-01-23 11:19:29.664163: step: 326/1541, loss: 0.09323646128177643 2023-01-23 11:19:30.534747: step: 328/1541, loss: 0.13030345737934113 2023-01-23 11:19:31.510133: step: 330/1541, loss: 0.04758988320827484 2023-01-23 11:19:32.449495: step: 332/1541, loss: 0.21661552786827087 2023-01-23 11:19:33.382357: step: 334/1541, loss: 0.08836923539638519 2023-01-23 11:19:34.261269: step: 336/1541, loss: 0.10445104539394379 2023-01-23 11:19:35.144860: step: 338/1541, loss: 0.07641047239303589 2023-01-23 11:19:36.028234: step: 340/1541, loss: 0.017831413075327873 2023-01-23 11:19:36.919816: step: 342/1541, loss: 0.35727155208587646 2023-01-23 11:19:37.895558: step: 344/1541, loss: 0.007167052943259478 2023-01-23 11:19:38.754028: step: 346/1541, loss: 0.0015475268010050058 2023-01-23 11:19:39.642957: step: 348/1541, loss: 0.08382850885391235 2023-01-23 11:19:40.550812: step: 350/1541, loss: 0.1424015611410141 2023-01-23 11:19:41.436834: step: 352/1541, loss: 0.26795080304145813 2023-01-23 11:19:42.343141: step: 354/1541, loss: 0.18758893013000488 2023-01-23 11:19:43.234279: step: 356/1541, loss: 0.3906347453594208 2023-01-23 11:19:44.105685: step: 358/1541, loss: 0.0022305974271148443 2023-01-23 11:19:45.043088: step: 360/1541, loss: 0.36597898602485657 2023-01-23 11:19:45.976425: step: 362/1541, loss: 0.0954999029636383 2023-01-23 11:19:46.893988: step: 364/1541, loss: 0.011571954935789108 2023-01-23 11:19:47.846291: step: 366/1541, loss: 0.3119286894798279 2023-01-23 11:19:48.714465: step: 368/1541, loss: 0.5227856040000916 2023-01-23 11:19:49.638489: step: 370/1541, loss: 0.13012219965457916 2023-01-23 11:19:50.574221: step: 372/1541, loss: 0.01599101535975933 2023-01-23 11:19:51.463551: step: 374/1541, loss: 0.40686336159706116 2023-01-23 11:19:52.319683: step: 376/1541, loss: 0.36615681648254395 2023-01-23 11:19:53.214669: step: 378/1541, loss: 0.005524434149265289 2023-01-23 11:19:54.103322: step: 380/1541, loss: 0.11625252664089203 2023-01-23 11:19:55.052850: step: 382/1541, loss: 0.04177435487508774 2023-01-23 11:19:55.933359: step: 384/1541, loss: 0.32207855582237244 2023-01-23 11:19:56.804302: step: 386/1541, loss: 0.1823662370443344 2023-01-23 11:19:57.791376: step: 388/1541, loss: 0.02342626266181469 2023-01-23 11:19:58.669403: step: 390/1541, loss: 0.041125982999801636 2023-01-23 11:19:59.542144: step: 392/1541, loss: 0.03779388591647148 2023-01-23 11:20:00.437531: step: 394/1541, loss: 0.16883432865142822 2023-01-23 11:20:01.330884: step: 396/1541, loss: 0.13389407098293304 2023-01-23 11:20:02.287421: step: 398/1541, loss: 0.13911324739456177 2023-01-23 11:20:03.186679: step: 400/1541, loss: 0.14886893332004547 2023-01-23 11:20:04.086940: step: 402/1541, loss: 0.024962542578577995 2023-01-23 11:20:04.987969: step: 404/1541, loss: 0.2331853210926056 2023-01-23 11:20:05.923308: step: 406/1541, loss: 0.03808679059147835 2023-01-23 11:20:06.845774: step: 408/1541, loss: 0.08069610595703125 2023-01-23 11:20:07.764424: step: 410/1541, loss: 0.14918893575668335 2023-01-23 11:20:08.654136: step: 412/1541, loss: 0.32271653413772583 2023-01-23 11:20:09.518589: step: 414/1541, loss: 0.051495231688022614 2023-01-23 11:20:10.432958: step: 416/1541, loss: 2.772362232208252 2023-01-23 11:20:11.307431: step: 418/1541, loss: 0.1705651581287384 2023-01-23 11:20:12.193587: step: 420/1541, loss: 0.28381478786468506 2023-01-23 11:20:13.087155: step: 422/1541, loss: 0.012271685525774956 2023-01-23 11:20:13.993216: step: 424/1541, loss: 0.05505724251270294 2023-01-23 11:20:14.913123: step: 426/1541, loss: 0.06574033200740814 2023-01-23 11:20:15.808085: step: 428/1541, loss: 0.17430661618709564 2023-01-23 11:20:16.705456: step: 430/1541, loss: 0.12042824923992157 2023-01-23 11:20:17.621618: step: 432/1541, loss: 0.18345233798027039 2023-01-23 11:20:18.524910: step: 434/1541, loss: 0.0668507069349289 2023-01-23 11:20:19.467925: step: 436/1541, loss: 0.09523624926805496 2023-01-23 11:20:20.367643: step: 438/1541, loss: 0.07504560053348541 2023-01-23 11:20:21.245566: step: 440/1541, loss: 0.15586653351783752 2023-01-23 11:20:22.332020: step: 442/1541, loss: 0.050659772008657455 2023-01-23 11:20:23.246989: step: 444/1541, loss: 0.1514628827571869 2023-01-23 11:20:24.158943: step: 446/1541, loss: 0.03734216094017029 2023-01-23 11:20:25.052855: step: 448/1541, loss: 0.08253855258226395 2023-01-23 11:20:25.967240: step: 450/1541, loss: 0.20409171283245087 2023-01-23 11:20:26.915590: step: 452/1541, loss: 2.9583895206451416 2023-01-23 11:20:27.814247: step: 454/1541, loss: 0.014884819276630878 2023-01-23 11:20:28.737371: step: 456/1541, loss: 0.027544358745217323 2023-01-23 11:20:29.653884: step: 458/1541, loss: 0.5561017990112305 2023-01-23 11:20:30.564139: step: 460/1541, loss: 0.02307731658220291 2023-01-23 11:20:31.436962: step: 462/1541, loss: 0.005417968612164259 2023-01-23 11:20:32.332550: step: 464/1541, loss: 0.05780535936355591 2023-01-23 11:20:33.326487: step: 466/1541, loss: 0.24907530844211578 2023-01-23 11:20:34.219624: step: 468/1541, loss: 0.027733881026506424 2023-01-23 11:20:35.094941: step: 470/1541, loss: 0.14615210890769958 2023-01-23 11:20:35.978181: step: 472/1541, loss: 0.0524587407708168 2023-01-23 11:20:36.864116: step: 474/1541, loss: 0.13101781904697418 2023-01-23 11:20:37.772359: step: 476/1541, loss: 0.03802249953150749 2023-01-23 11:20:38.684342: step: 478/1541, loss: 0.10614607483148575 2023-01-23 11:20:39.558653: step: 480/1541, loss: 0.12570691108703613 2023-01-23 11:20:40.449386: step: 482/1541, loss: 0.23655597865581512 2023-01-23 11:20:41.319943: step: 484/1541, loss: 0.028574170544743538 2023-01-23 11:20:42.234600: step: 486/1541, loss: 0.14087152481079102 2023-01-23 11:20:43.173213: step: 488/1541, loss: 0.7951561808586121 2023-01-23 11:20:44.097859: step: 490/1541, loss: 0.20694980025291443 2023-01-23 11:20:45.030287: step: 492/1541, loss: 0.426440954208374 2023-01-23 11:20:45.933302: step: 494/1541, loss: 0.2548443675041199 2023-01-23 11:20:46.817445: step: 496/1541, loss: 0.14739499986171722 2023-01-23 11:20:47.724315: step: 498/1541, loss: 0.059594620019197464 2023-01-23 11:20:48.743597: step: 500/1541, loss: 0.4213975965976715 2023-01-23 11:20:49.628852: step: 502/1541, loss: 0.031032798811793327 2023-01-23 11:20:50.519220: step: 504/1541, loss: 0.015995647758245468 2023-01-23 11:20:51.419318: step: 506/1541, loss: 0.11734011024236679 2023-01-23 11:20:52.288935: step: 508/1541, loss: 0.04411523416638374 2023-01-23 11:20:53.208015: step: 510/1541, loss: 0.25411736965179443 2023-01-23 11:20:54.117456: step: 512/1541, loss: 0.2598198652267456 2023-01-23 11:20:54.994320: step: 514/1541, loss: 0.45403075218200684 2023-01-23 11:20:55.918818: step: 516/1541, loss: 0.07169981300830841 2023-01-23 11:20:56.811455: step: 518/1541, loss: 0.017328383401036263 2023-01-23 11:20:57.712967: step: 520/1541, loss: 0.1821524053812027 2023-01-23 11:20:58.586674: step: 522/1541, loss: 0.12853285670280457 2023-01-23 11:20:59.465556: step: 524/1541, loss: 0.06003900244832039 2023-01-23 11:21:00.352451: step: 526/1541, loss: 0.005993095692247152 2023-01-23 11:21:01.261796: step: 528/1541, loss: 0.0793399065732956 2023-01-23 11:21:02.153962: step: 530/1541, loss: 0.11350712925195694 2023-01-23 11:21:03.045483: step: 532/1541, loss: 0.0507783405482769 2023-01-23 11:21:04.016038: step: 534/1541, loss: 0.13850978016853333 2023-01-23 11:21:04.930707: step: 536/1541, loss: 0.13427908718585968 2023-01-23 11:21:05.811602: step: 538/1541, loss: 0.007052119355648756 2023-01-23 11:21:06.692720: step: 540/1541, loss: 0.016886811703443527 2023-01-23 11:21:07.634664: step: 542/1541, loss: 0.1835177093744278 2023-01-23 11:21:08.521495: step: 544/1541, loss: 0.012585699558258057 2023-01-23 11:21:09.406483: step: 546/1541, loss: 0.08431990444660187 2023-01-23 11:21:10.293126: step: 548/1541, loss: 0.007236145436763763 2023-01-23 11:21:11.248877: step: 550/1541, loss: 0.008560292422771454 2023-01-23 11:21:12.131297: step: 552/1541, loss: 0.05241640284657478 2023-01-23 11:21:13.046778: step: 554/1541, loss: 0.044268541038036346 2023-01-23 11:21:13.934117: step: 556/1541, loss: 0.007014994043856859 2023-01-23 11:21:14.818999: step: 558/1541, loss: 0.09700486063957214 2023-01-23 11:21:15.713953: step: 560/1541, loss: 0.01058073341846466 2023-01-23 11:21:16.605052: step: 562/1541, loss: 0.07885727286338806 2023-01-23 11:21:17.543996: step: 564/1541, loss: 0.4127909243106842 2023-01-23 11:21:18.483358: step: 566/1541, loss: 0.0284072645008564 2023-01-23 11:21:19.347543: step: 568/1541, loss: 0.37412336468696594 2023-01-23 11:21:20.241549: step: 570/1541, loss: 0.09657450020313263 2023-01-23 11:21:21.146713: step: 572/1541, loss: 0.019291307777166367 2023-01-23 11:21:22.045702: step: 574/1541, loss: 0.27042049169540405 2023-01-23 11:21:22.951273: step: 576/1541, loss: 0.07479176670312881 2023-01-23 11:21:23.863635: step: 578/1541, loss: 0.056847888976335526 2023-01-23 11:21:24.764770: step: 580/1541, loss: 0.2531483769416809 2023-01-23 11:21:25.670784: step: 582/1541, loss: 0.23445530235767365 2023-01-23 11:21:26.549255: step: 584/1541, loss: 0.2341143935918808 2023-01-23 11:21:27.393527: step: 586/1541, loss: 0.002173607237637043 2023-01-23 11:21:28.302010: step: 588/1541, loss: 0.358503133058548 2023-01-23 11:21:29.211174: step: 590/1541, loss: 0.22111395001411438 2023-01-23 11:21:30.135933: step: 592/1541, loss: 0.8346702456474304 2023-01-23 11:21:31.060515: step: 594/1541, loss: 0.24819393455982208 2023-01-23 11:21:31.949693: step: 596/1541, loss: 0.1268366128206253 2023-01-23 11:21:32.880873: step: 598/1541, loss: 0.07382490485906601 2023-01-23 11:21:33.769770: step: 600/1541, loss: 0.003870447864755988 2023-01-23 11:21:34.691737: step: 602/1541, loss: 0.12347486615180969 2023-01-23 11:21:35.598351: step: 604/1541, loss: 0.024377578869462013 2023-01-23 11:21:36.513575: step: 606/1541, loss: 0.07128682732582092 2023-01-23 11:21:37.388734: step: 608/1541, loss: 0.020932726562023163 2023-01-23 11:21:38.262699: step: 610/1541, loss: 0.06780478358268738 2023-01-23 11:21:39.173717: step: 612/1541, loss: 0.0509498193860054 2023-01-23 11:21:40.040232: step: 614/1541, loss: 0.03233930468559265 2023-01-23 11:21:40.940607: step: 616/1541, loss: 0.0568428635597229 2023-01-23 11:21:41.862218: step: 618/1541, loss: 0.04985390603542328 2023-01-23 11:21:42.773066: step: 620/1541, loss: 0.051830824464559555 2023-01-23 11:21:43.657302: step: 622/1541, loss: 0.41559290885925293 2023-01-23 11:21:44.548440: step: 624/1541, loss: 0.047369759529829025 2023-01-23 11:21:45.500125: step: 626/1541, loss: 0.024355541914701462 2023-01-23 11:21:46.426426: step: 628/1541, loss: 0.030948622152209282 2023-01-23 11:21:47.304058: step: 630/1541, loss: 0.19169144332408905 2023-01-23 11:21:48.240747: step: 632/1541, loss: 0.08422805368900299 2023-01-23 11:21:49.126119: step: 634/1541, loss: 0.18762272596359253 2023-01-23 11:21:50.050886: step: 636/1541, loss: 0.5880631804466248 2023-01-23 11:21:50.927638: step: 638/1541, loss: 0.03246375918388367 2023-01-23 11:21:51.802396: step: 640/1541, loss: 0.14207474887371063 2023-01-23 11:21:52.683328: step: 642/1541, loss: 0.021543147042393684 2023-01-23 11:21:53.594968: step: 644/1541, loss: 0.014327295124530792 2023-01-23 11:21:54.469952: step: 646/1541, loss: 0.2251201719045639 2023-01-23 11:21:55.361541: step: 648/1541, loss: 0.01875903643667698 2023-01-23 11:21:56.221265: step: 650/1541, loss: 0.04834948480129242 2023-01-23 11:21:57.107614: step: 652/1541, loss: 0.010484084486961365 2023-01-23 11:21:58.010837: step: 654/1541, loss: 0.12464575469493866 2023-01-23 11:21:58.958507: step: 656/1541, loss: 0.0212692990899086 2023-01-23 11:21:59.826006: step: 658/1541, loss: 0.15787342190742493 2023-01-23 11:22:00.689504: step: 660/1541, loss: 0.13940578699111938 2023-01-23 11:22:01.539472: step: 662/1541, loss: 0.14923612773418427 2023-01-23 11:22:02.421178: step: 664/1541, loss: 0.2388594001531601 2023-01-23 11:22:03.302330: step: 666/1541, loss: 0.038887135684490204 2023-01-23 11:22:04.162603: step: 668/1541, loss: 0.2740209698677063 2023-01-23 11:22:05.059621: step: 670/1541, loss: 0.013745979405939579 2023-01-23 11:22:06.001832: step: 672/1541, loss: 0.1241813525557518 2023-01-23 11:22:06.911527: step: 674/1541, loss: 0.4867747724056244 2023-01-23 11:22:07.813136: step: 676/1541, loss: 0.026798130944371223 2023-01-23 11:22:08.690971: step: 678/1541, loss: 0.13492968678474426 2023-01-23 11:22:09.665296: step: 680/1541, loss: 0.10995133221149445 2023-01-23 11:22:10.606915: step: 682/1541, loss: 0.06877252459526062 2023-01-23 11:22:11.490065: step: 684/1541, loss: 1.4781200885772705 2023-01-23 11:22:12.388415: step: 686/1541, loss: 0.26037901639938354 2023-01-23 11:22:13.327054: step: 688/1541, loss: 0.06993280351161957 2023-01-23 11:22:14.210384: step: 690/1541, loss: 0.04480289667844772 2023-01-23 11:22:15.122089: step: 692/1541, loss: 0.2133914679288864 2023-01-23 11:22:16.004568: step: 694/1541, loss: 0.21783043444156647 2023-01-23 11:22:16.938450: step: 696/1541, loss: 0.11751735210418701 2023-01-23 11:22:17.893791: step: 698/1541, loss: 0.19599542021751404 2023-01-23 11:22:18.796259: step: 700/1541, loss: 0.010988103225827217 2023-01-23 11:22:19.755792: step: 702/1541, loss: 0.15558288991451263 2023-01-23 11:22:20.662647: step: 704/1541, loss: 0.8204302787780762 2023-01-23 11:22:21.600414: step: 706/1541, loss: 0.05484428256750107 2023-01-23 11:22:22.457602: step: 708/1541, loss: 0.017036087810993195 2023-01-23 11:22:23.327852: step: 710/1541, loss: 0.061497762799263 2023-01-23 11:22:24.297477: step: 712/1541, loss: 0.13528890907764435 2023-01-23 11:22:25.188942: step: 714/1541, loss: 0.06183958053588867 2023-01-23 11:22:26.136649: step: 716/1541, loss: 0.01412094198167324 2023-01-23 11:22:27.091420: step: 718/1541, loss: 0.21667475998401642 2023-01-23 11:22:27.963658: step: 720/1541, loss: 0.2050422728061676 2023-01-23 11:22:28.842989: step: 722/1541, loss: 0.019592946395277977 2023-01-23 11:22:29.703889: step: 724/1541, loss: 0.1632014811038971 2023-01-23 11:22:30.581972: step: 726/1541, loss: 0.26939550042152405 2023-01-23 11:22:31.496455: step: 728/1541, loss: 0.03029789589345455 2023-01-23 11:22:32.371702: step: 730/1541, loss: 0.09714961051940918 2023-01-23 11:22:33.279387: step: 732/1541, loss: 0.018766507506370544 2023-01-23 11:22:34.155143: step: 734/1541, loss: 0.1912740170955658 2023-01-23 11:22:35.060544: step: 736/1541, loss: 0.038610607385635376 2023-01-23 11:22:35.945239: step: 738/1541, loss: 0.040436647832393646 2023-01-23 11:22:36.831478: step: 740/1541, loss: 0.21794241666793823 2023-01-23 11:22:37.706016: step: 742/1541, loss: 0.1437845081090927 2023-01-23 11:22:38.606143: step: 744/1541, loss: 0.06700699776411057 2023-01-23 11:22:39.514962: step: 746/1541, loss: 0.017622102051973343 2023-01-23 11:22:40.482571: step: 748/1541, loss: 0.18824365735054016 2023-01-23 11:22:41.368333: step: 750/1541, loss: 0.043952446430921555 2023-01-23 11:22:42.300698: step: 752/1541, loss: 0.032841794192790985 2023-01-23 11:22:43.172552: step: 754/1541, loss: 0.003128564218059182 2023-01-23 11:22:44.060337: step: 756/1541, loss: 0.18880626559257507 2023-01-23 11:22:44.943857: step: 758/1541, loss: 0.47806110978126526 2023-01-23 11:22:45.811950: step: 760/1541, loss: 0.008704790845513344 2023-01-23 11:22:46.710404: step: 762/1541, loss: 0.16287174820899963 2023-01-23 11:22:47.575929: step: 764/1541, loss: 0.038044337183237076 2023-01-23 11:22:48.482607: step: 766/1541, loss: 0.12139514088630676 2023-01-23 11:22:49.349914: step: 768/1541, loss: 0.0153114078566432 2023-01-23 11:22:50.197095: step: 770/1541, loss: 0.024107590317726135 2023-01-23 11:22:51.063756: step: 772/1541, loss: 0.054673805832862854 2023-01-23 11:22:52.053769: step: 774/1541, loss: 0.19401690363883972 2023-01-23 11:22:52.980805: step: 776/1541, loss: 0.0372779443860054 2023-01-23 11:22:53.877973: step: 778/1541, loss: 0.12050946056842804 2023-01-23 11:22:54.747761: step: 780/1541, loss: 0.06298215687274933 2023-01-23 11:22:55.703497: step: 782/1541, loss: 0.22432246804237366 2023-01-23 11:22:56.595741: step: 784/1541, loss: 0.005219710990786552 2023-01-23 11:22:57.491266: step: 786/1541, loss: 0.018951520323753357 2023-01-23 11:22:58.363097: step: 788/1541, loss: 0.12388404458761215 2023-01-23 11:22:59.328951: step: 790/1541, loss: 0.14184217154979706 2023-01-23 11:23:00.193791: step: 792/1541, loss: 0.03719285875558853 2023-01-23 11:23:01.083652: step: 794/1541, loss: 0.25711536407470703 2023-01-23 11:23:01.976289: step: 796/1541, loss: 0.12395112216472626 2023-01-23 11:23:02.875247: step: 798/1541, loss: 0.043800175189971924 2023-01-23 11:23:03.741949: step: 800/1541, loss: 0.009619597345590591 2023-01-23 11:23:04.611862: step: 802/1541, loss: 0.10930785536766052 2023-01-23 11:23:05.480544: step: 804/1541, loss: 0.0693071037530899 2023-01-23 11:23:06.406943: step: 806/1541, loss: 0.006001478061079979 2023-01-23 11:23:07.322055: step: 808/1541, loss: 0.0353296622633934 2023-01-23 11:23:08.227481: step: 810/1541, loss: 0.011082563549280167 2023-01-23 11:23:09.130926: step: 812/1541, loss: 0.0363728292286396 2023-01-23 11:23:10.034475: step: 814/1541, loss: 0.1864587366580963 2023-01-23 11:23:11.021226: step: 816/1541, loss: 0.3819454610347748 2023-01-23 11:23:11.941146: step: 818/1541, loss: 0.002075423952192068 2023-01-23 11:23:12.862436: step: 820/1541, loss: 0.14364513754844666 2023-01-23 11:23:13.749123: step: 822/1541, loss: 0.011360866948962212 2023-01-23 11:23:14.640672: step: 824/1541, loss: 0.21869006752967834 2023-01-23 11:23:15.632679: step: 826/1541, loss: 0.01820690929889679 2023-01-23 11:23:16.542712: step: 828/1541, loss: 0.05846665799617767 2023-01-23 11:23:17.453020: step: 830/1541, loss: 0.11709016561508179 2023-01-23 11:23:18.351347: step: 832/1541, loss: 0.306405246257782 2023-01-23 11:23:19.238283: step: 834/1541, loss: 0.3896979093551636 2023-01-23 11:23:20.122435: step: 836/1541, loss: 0.03457462042570114 2023-01-23 11:23:21.026057: step: 838/1541, loss: 0.11983492225408554 2023-01-23 11:23:22.012153: step: 840/1541, loss: 0.006587190553545952 2023-01-23 11:23:22.997149: step: 842/1541, loss: 0.030248645693063736 2023-01-23 11:23:23.868393: step: 844/1541, loss: 0.0043138605542480946 2023-01-23 11:23:24.784847: step: 846/1541, loss: 0.0860239639878273 2023-01-23 11:23:25.681860: step: 848/1541, loss: 0.25001659989356995 2023-01-23 11:23:26.603154: step: 850/1541, loss: 1.323241949081421 2023-01-23 11:23:27.468379: step: 852/1541, loss: 0.2684091329574585 2023-01-23 11:23:28.336019: step: 854/1541, loss: 0.0012262527598068118 2023-01-23 11:23:29.272529: step: 856/1541, loss: 0.0016788929933682084 2023-01-23 11:23:30.270969: step: 858/1541, loss: 0.004437070805579424 2023-01-23 11:23:31.152558: step: 860/1541, loss: 0.05413317680358887 2023-01-23 11:23:32.175427: step: 862/1541, loss: 0.003699235850945115 2023-01-23 11:23:33.112694: step: 864/1541, loss: 0.3020993173122406 2023-01-23 11:23:33.985099: step: 866/1541, loss: 0.08314864337444305 2023-01-23 11:23:34.865043: step: 868/1541, loss: 0.010990152135491371 2023-01-23 11:23:35.845375: step: 870/1541, loss: 0.2296188920736313 2023-01-23 11:23:36.785229: step: 872/1541, loss: 0.0358065664768219 2023-01-23 11:23:37.741977: step: 874/1541, loss: 0.27909380197525024 2023-01-23 11:23:38.647152: step: 876/1541, loss: 0.021706053987145424 2023-01-23 11:23:39.596513: step: 878/1541, loss: 0.209293395280838 2023-01-23 11:23:40.481027: step: 880/1541, loss: 0.14646129310131073 2023-01-23 11:23:41.400186: step: 882/1541, loss: 0.16864141821861267 2023-01-23 11:23:42.308417: step: 884/1541, loss: 0.21422378718852997 2023-01-23 11:23:43.231739: step: 886/1541, loss: 0.10072772949934006 2023-01-23 11:23:44.112083: step: 888/1541, loss: 1.0197840929031372 2023-01-23 11:23:44.980018: step: 890/1541, loss: 0.9048007130622864 2023-01-23 11:23:45.977999: step: 892/1541, loss: 0.04210460186004639 2023-01-23 11:23:46.951312: step: 894/1541, loss: 0.2211703509092331 2023-01-23 11:23:47.812323: step: 896/1541, loss: 0.020476438105106354 2023-01-23 11:23:48.689020: step: 898/1541, loss: 0.08692223578691483 2023-01-23 11:23:49.576356: step: 900/1541, loss: 0.06098426878452301 2023-01-23 11:23:50.532587: step: 902/1541, loss: 0.1660165786743164 2023-01-23 11:23:51.439848: step: 904/1541, loss: 0.04127439111471176 2023-01-23 11:23:52.347055: step: 906/1541, loss: 0.019273793324828148 2023-01-23 11:23:53.230081: step: 908/1541, loss: 0.11144057661294937 2023-01-23 11:23:54.106386: step: 910/1541, loss: 0.07701092958450317 2023-01-23 11:23:55.019759: step: 912/1541, loss: 0.10719592124223709 2023-01-23 11:23:55.948180: step: 914/1541, loss: 0.005437835119664669 2023-01-23 11:23:56.836047: step: 916/1541, loss: 0.12402309477329254 2023-01-23 11:23:57.717826: step: 918/1541, loss: 0.014288600534200668 2023-01-23 11:23:58.578735: step: 920/1541, loss: 0.12783373892307281 2023-01-23 11:23:59.480217: step: 922/1541, loss: 0.3028638958930969 2023-01-23 11:24:00.431903: step: 924/1541, loss: 1.3946356773376465 2023-01-23 11:24:01.335059: step: 926/1541, loss: 0.1547204554080963 2023-01-23 11:24:02.248894: step: 928/1541, loss: 0.15204088389873505 2023-01-23 11:24:03.134659: step: 930/1541, loss: 0.10807792842388153 2023-01-23 11:24:04.148186: step: 932/1541, loss: 0.05817187577486038 2023-01-23 11:24:05.031497: step: 934/1541, loss: 0.523530900478363 2023-01-23 11:24:05.903415: step: 936/1541, loss: 0.2785871922969818 2023-01-23 11:24:06.812840: step: 938/1541, loss: 0.19055119156837463 2023-01-23 11:24:07.708723: step: 940/1541, loss: 0.13469523191452026 2023-01-23 11:24:08.646006: step: 942/1541, loss: 0.01762516237795353 2023-01-23 11:24:09.519027: step: 944/1541, loss: 0.14374516904354095 2023-01-23 11:24:10.379594: step: 946/1541, loss: 0.05002021789550781 2023-01-23 11:24:11.359840: step: 948/1541, loss: 0.8664809465408325 2023-01-23 11:24:12.348039: step: 950/1541, loss: 0.019881809130311012 2023-01-23 11:24:13.263042: step: 952/1541, loss: 0.33266666531562805 2023-01-23 11:24:14.180696: step: 954/1541, loss: 0.12150844931602478 2023-01-23 11:24:15.080697: step: 956/1541, loss: 0.4752916395664215 2023-01-23 11:24:15.976452: step: 958/1541, loss: 0.7522746324539185 2023-01-23 11:24:16.869880: step: 960/1541, loss: 0.005655101500451565 2023-01-23 11:24:17.815124: step: 962/1541, loss: 0.20885036885738373 2023-01-23 11:24:18.685414: step: 964/1541, loss: 0.4845603406429291 2023-01-23 11:24:19.623546: step: 966/1541, loss: 0.023563187569379807 2023-01-23 11:24:20.510901: step: 968/1541, loss: 0.08029178529977798 2023-01-23 11:24:21.513460: step: 970/1541, loss: 0.018343763425946236 2023-01-23 11:24:22.420037: step: 972/1541, loss: 0.032735735177993774 2023-01-23 11:24:23.346347: step: 974/1541, loss: 0.12530411779880524 2023-01-23 11:24:24.266822: step: 976/1541, loss: 0.15204931795597076 2023-01-23 11:24:25.122901: step: 978/1541, loss: 0.13330626487731934 2023-01-23 11:24:26.080529: step: 980/1541, loss: 0.07481874525547028 2023-01-23 11:24:26.951740: step: 982/1541, loss: 0.3627530038356781 2023-01-23 11:24:27.884393: step: 984/1541, loss: 0.3120121657848358 2023-01-23 11:24:28.767300: step: 986/1541, loss: 0.09315769374370575 2023-01-23 11:24:29.693593: step: 988/1541, loss: 0.046266570687294006 2023-01-23 11:24:30.571140: step: 990/1541, loss: 0.024264492094516754 2023-01-23 11:24:31.451538: step: 992/1541, loss: 0.08842458575963974 2023-01-23 11:24:32.363702: step: 994/1541, loss: 0.17128819227218628 2023-01-23 11:24:33.257146: step: 996/1541, loss: 0.1091136559844017 2023-01-23 11:24:34.220000: step: 998/1541, loss: 0.22061696648597717 2023-01-23 11:24:35.107752: step: 1000/1541, loss: 2.5535621643066406 2023-01-23 11:24:35.981285: step: 1002/1541, loss: 0.04714610427618027 2023-01-23 11:24:36.931045: step: 1004/1541, loss: 0.08365772664546967 2023-01-23 11:24:37.934560: step: 1006/1541, loss: 0.01947442628443241 2023-01-23 11:24:38.885810: step: 1008/1541, loss: 0.41541787981987 2023-01-23 11:24:39.753990: step: 1010/1541, loss: 0.06587174534797668 2023-01-23 11:24:40.666085: step: 1012/1541, loss: 0.1850721836090088 2023-01-23 11:24:41.578866: step: 1014/1541, loss: 1.6883455514907837 2023-01-23 11:24:42.508499: step: 1016/1541, loss: 0.07852564007043839 2023-01-23 11:24:43.389088: step: 1018/1541, loss: 0.029341446235775948 2023-01-23 11:24:44.269243: step: 1020/1541, loss: 0.02814476564526558 2023-01-23 11:24:45.168840: step: 1022/1541, loss: 0.22505313158035278 2023-01-23 11:24:46.064676: step: 1024/1541, loss: 0.6252395510673523 2023-01-23 11:24:46.928315: step: 1026/1541, loss: 0.0954878181219101 2023-01-23 11:24:47.815554: step: 1028/1541, loss: 0.02550487406551838 2023-01-23 11:24:48.751208: step: 1030/1541, loss: 0.17339883744716644 2023-01-23 11:24:49.648946: step: 1032/1541, loss: 0.036638591438531876 2023-01-23 11:24:50.514849: step: 1034/1541, loss: 0.12855850160121918 2023-01-23 11:24:51.396421: step: 1036/1541, loss: 0.2677953243255615 2023-01-23 11:24:52.305413: step: 1038/1541, loss: 0.22103431820869446 2023-01-23 11:24:53.164080: step: 1040/1541, loss: 0.012074076570570469 2023-01-23 11:24:54.080749: step: 1042/1541, loss: 0.801826000213623 2023-01-23 11:24:55.039452: step: 1044/1541, loss: 0.13497740030288696 2023-01-23 11:24:56.038548: step: 1046/1541, loss: 0.14654786884784698 2023-01-23 11:24:56.966245: step: 1048/1541, loss: 0.18283620476722717 2023-01-23 11:24:57.869272: step: 1050/1541, loss: 0.4612809419631958 2023-01-23 11:24:58.775216: step: 1052/1541, loss: 0.1430204063653946 2023-01-23 11:24:59.680317: step: 1054/1541, loss: 0.038482338190078735 2023-01-23 11:25:00.569917: step: 1056/1541, loss: 0.47008058428764343 2023-01-23 11:25:01.482374: step: 1058/1541, loss: 0.18103903532028198 2023-01-23 11:25:02.410483: step: 1060/1541, loss: 0.5170518755912781 2023-01-23 11:25:03.299922: step: 1062/1541, loss: 0.05126442760229111 2023-01-23 11:25:04.227011: step: 1064/1541, loss: 0.1699373424053192 2023-01-23 11:25:05.169865: step: 1066/1541, loss: 0.014369796961545944 2023-01-23 11:25:06.129812: step: 1068/1541, loss: 0.1130654439330101 2023-01-23 11:25:07.019490: step: 1070/1541, loss: 0.04741756245493889 2023-01-23 11:25:07.932950: step: 1072/1541, loss: 0.18447162210941315 2023-01-23 11:25:08.874705: step: 1074/1541, loss: 0.12774397432804108 2023-01-23 11:25:09.808310: step: 1076/1541, loss: 0.3045876622200012 2023-01-23 11:25:10.706542: step: 1078/1541, loss: 0.140957772731781 2023-01-23 11:25:11.618867: step: 1080/1541, loss: 0.27344629168510437 2023-01-23 11:25:12.513008: step: 1082/1541, loss: 0.09657523781061172 2023-01-23 11:25:13.424508: step: 1084/1541, loss: 0.6184788942337036 2023-01-23 11:25:14.330220: step: 1086/1541, loss: 0.07858365774154663 2023-01-23 11:25:15.265850: step: 1088/1541, loss: 0.37664124369621277 2023-01-23 11:25:16.135644: step: 1090/1541, loss: 0.018400974571704865 2023-01-23 11:25:17.087222: step: 1092/1541, loss: 0.1979958713054657 2023-01-23 11:25:17.941806: step: 1094/1541, loss: 0.1339571177959442 2023-01-23 11:25:18.824981: step: 1096/1541, loss: 4.334758758544922 2023-01-23 11:25:19.691075: step: 1098/1541, loss: 0.21247412264347076 2023-01-23 11:25:20.625211: step: 1100/1541, loss: 0.0818544253706932 2023-01-23 11:25:21.528085: step: 1102/1541, loss: 0.2987462282180786 2023-01-23 11:25:22.394990: step: 1104/1541, loss: 0.17388293147087097 2023-01-23 11:25:23.278227: step: 1106/1541, loss: 0.03202049806714058 2023-01-23 11:25:24.201984: step: 1108/1541, loss: 0.11103161424398422 2023-01-23 11:25:25.075934: step: 1110/1541, loss: 0.01642845943570137 2023-01-23 11:25:26.004577: step: 1112/1541, loss: 1.5773601531982422 2023-01-23 11:25:26.936997: step: 1114/1541, loss: 0.06946337968111038 2023-01-23 11:25:27.854101: step: 1116/1541, loss: 0.06950987130403519 2023-01-23 11:25:28.754212: step: 1118/1541, loss: 0.026697425171732903 2023-01-23 11:25:29.646417: step: 1120/1541, loss: 0.19732451438903809 2023-01-23 11:25:30.520174: step: 1122/1541, loss: 0.22542186081409454 2023-01-23 11:25:31.406790: step: 1124/1541, loss: 0.010000377893447876 2023-01-23 11:25:32.304407: step: 1126/1541, loss: 0.3515513837337494 2023-01-23 11:25:33.169435: step: 1128/1541, loss: 0.1673852801322937 2023-01-23 11:25:34.145617: step: 1130/1541, loss: 0.01564805954694748 2023-01-23 11:25:35.031995: step: 1132/1541, loss: 0.022860035300254822 2023-01-23 11:25:35.927904: step: 1134/1541, loss: 0.1561005562543869 2023-01-23 11:25:36.784939: step: 1136/1541, loss: 0.03481217101216316 2023-01-23 11:25:37.648325: step: 1138/1541, loss: 0.14006249606609344 2023-01-23 11:25:38.499912: step: 1140/1541, loss: 0.22056972980499268 2023-01-23 11:25:39.403834: step: 1142/1541, loss: 0.02914394438266754 2023-01-23 11:25:40.273749: step: 1144/1541, loss: 0.1852840930223465 2023-01-23 11:25:41.126437: step: 1146/1541, loss: 0.08867320418357849 2023-01-23 11:25:42.025917: step: 1148/1541, loss: 0.36627885699272156 2023-01-23 11:25:42.905483: step: 1150/1541, loss: 0.010996206663548946 2023-01-23 11:25:43.812330: step: 1152/1541, loss: 0.044695742428302765 2023-01-23 11:25:44.731589: step: 1154/1541, loss: 0.019241465255618095 2023-01-23 11:25:45.619627: step: 1156/1541, loss: 0.22059041261672974 2023-01-23 11:25:46.559457: step: 1158/1541, loss: 0.9203548431396484 2023-01-23 11:25:47.465615: step: 1160/1541, loss: 0.020808735862374306 2023-01-23 11:25:48.372678: step: 1162/1541, loss: 0.38453197479248047 2023-01-23 11:25:49.264330: step: 1164/1541, loss: 0.05998583510518074 2023-01-23 11:25:50.155645: step: 1166/1541, loss: 0.10630304366350174 2023-01-23 11:25:51.035402: step: 1168/1541, loss: 6.207317352294922 2023-01-23 11:25:51.880119: step: 1170/1541, loss: 0.4769692122936249 2023-01-23 11:25:52.716602: step: 1172/1541, loss: 0.01700054109096527 2023-01-23 11:25:53.651631: step: 1174/1541, loss: 0.04651220515370369 2023-01-23 11:25:54.599373: step: 1176/1541, loss: 0.09505455195903778 2023-01-23 11:25:55.547189: step: 1178/1541, loss: 0.1069929450750351 2023-01-23 11:25:56.486636: step: 1180/1541, loss: 0.10199140757322311 2023-01-23 11:25:57.408688: step: 1182/1541, loss: 0.11905316263437271 2023-01-23 11:25:58.271256: step: 1184/1541, loss: 0.48079532384872437 2023-01-23 11:25:59.172460: step: 1186/1541, loss: 0.07884024828672409 2023-01-23 11:26:00.059309: step: 1188/1541, loss: 0.17273660004138947 2023-01-23 11:26:00.960100: step: 1190/1541, loss: 0.08974791318178177 2023-01-23 11:26:01.871913: step: 1192/1541, loss: 0.12699103355407715 2023-01-23 11:26:02.776794: step: 1194/1541, loss: 0.11000567674636841 2023-01-23 11:26:03.683029: step: 1196/1541, loss: 0.03850064054131508 2023-01-23 11:26:04.605318: step: 1198/1541, loss: 0.2200026959180832 2023-01-23 11:26:05.506089: step: 1200/1541, loss: 0.007476925849914551 2023-01-23 11:26:06.434230: step: 1202/1541, loss: 0.030488023534417152 2023-01-23 11:26:07.333470: step: 1204/1541, loss: 0.09970006346702576 2023-01-23 11:26:08.215028: step: 1206/1541, loss: 0.5932973027229309 2023-01-23 11:26:09.095247: step: 1208/1541, loss: 0.22723715007305145 2023-01-23 11:26:09.985756: step: 1210/1541, loss: 0.02578357420861721 2023-01-23 11:26:10.886941: step: 1212/1541, loss: 0.018359357491135597 2023-01-23 11:26:11.832153: step: 1214/1541, loss: 0.25420114398002625 2023-01-23 11:26:12.712981: step: 1216/1541, loss: 0.11976394802331924 2023-01-23 11:26:13.619925: step: 1218/1541, loss: 0.14710217714309692 2023-01-23 11:26:14.511888: step: 1220/1541, loss: 0.029660437256097794 2023-01-23 11:26:15.393841: step: 1222/1541, loss: 0.1059788390994072 2023-01-23 11:26:16.299978: step: 1224/1541, loss: 0.04488253965973854 2023-01-23 11:26:17.202912: step: 1226/1541, loss: 0.19688010215759277 2023-01-23 11:26:18.114443: step: 1228/1541, loss: 0.17270278930664062 2023-01-23 11:26:19.031420: step: 1230/1541, loss: 0.07902561128139496 2023-01-23 11:26:19.913638: step: 1232/1541, loss: 0.0869295597076416 2023-01-23 11:26:20.836716: step: 1234/1541, loss: 0.04524457827210426 2023-01-23 11:26:21.751016: step: 1236/1541, loss: 0.06769190728664398 2023-01-23 11:26:22.626836: step: 1238/1541, loss: 0.0279072318226099 2023-01-23 11:26:23.532565: step: 1240/1541, loss: 0.09849046170711517 2023-01-23 11:26:24.455660: step: 1242/1541, loss: 0.012115182355046272 2023-01-23 11:26:25.341887: step: 1244/1541, loss: 0.2209673672914505 2023-01-23 11:26:26.225833: step: 1246/1541, loss: 0.020001446828246117 2023-01-23 11:26:27.137603: step: 1248/1541, loss: 0.010278895497322083 2023-01-23 11:26:28.015476: step: 1250/1541, loss: 0.08629714697599411 2023-01-23 11:26:28.894779: step: 1252/1541, loss: 0.058594007045030594 2023-01-23 11:26:29.829173: step: 1254/1541, loss: 0.005913189612329006 2023-01-23 11:26:30.737000: step: 1256/1541, loss: 0.07767640799283981 2023-01-23 11:26:31.686936: step: 1258/1541, loss: 0.041819822043180466 2023-01-23 11:26:32.819879: step: 1260/1541, loss: 0.35994163155555725 2023-01-23 11:26:33.688345: step: 1262/1541, loss: 0.00302582373842597 2023-01-23 11:26:34.586256: step: 1264/1541, loss: 0.06073220819234848 2023-01-23 11:26:35.422128: step: 1266/1541, loss: 1.0276695489883423 2023-01-23 11:26:36.313916: step: 1268/1541, loss: 0.7755354642868042 2023-01-23 11:26:37.246392: step: 1270/1541, loss: 0.07673779875040054 2023-01-23 11:26:38.171879: step: 1272/1541, loss: 0.049155037850141525 2023-01-23 11:26:39.037507: step: 1274/1541, loss: 0.022049369290471077 2023-01-23 11:26:39.964334: step: 1276/1541, loss: 0.17652632296085358 2023-01-23 11:26:40.837720: step: 1278/1541, loss: 0.09820661693811417 2023-01-23 11:26:41.740987: step: 1280/1541, loss: 0.06982915103435516 2023-01-23 11:26:42.619713: step: 1282/1541, loss: 0.548762857913971 2023-01-23 11:26:43.511163: step: 1284/1541, loss: 0.018239889293909073 2023-01-23 11:26:44.485825: step: 1286/1541, loss: 1.1047114133834839 2023-01-23 11:26:45.406912: step: 1288/1541, loss: 0.4125983417034149 2023-01-23 11:26:46.324440: step: 1290/1541, loss: 0.235874205827713 2023-01-23 11:26:47.224710: step: 1292/1541, loss: 0.005265791900455952 2023-01-23 11:26:48.146313: step: 1294/1541, loss: 0.5335783362388611 2023-01-23 11:26:49.058889: step: 1296/1541, loss: 0.10777882486581802 2023-01-23 11:26:49.961503: step: 1298/1541, loss: 0.580350399017334 2023-01-23 11:26:50.801613: step: 1300/1541, loss: 0.045898448675870895 2023-01-23 11:26:51.693200: step: 1302/1541, loss: 0.09202850610017776 2023-01-23 11:26:52.583886: step: 1304/1541, loss: 0.05411418527364731 2023-01-23 11:26:53.500188: step: 1306/1541, loss: 0.015259746462106705 2023-01-23 11:26:54.381751: step: 1308/1541, loss: 0.011288641020655632 2023-01-23 11:26:55.281633: step: 1310/1541, loss: 0.0035586291924118996 2023-01-23 11:26:56.166067: step: 1312/1541, loss: 0.12248611450195312 2023-01-23 11:26:57.031917: step: 1314/1541, loss: 0.05418987572193146 2023-01-23 11:26:57.872937: step: 1316/1541, loss: 0.33461058139801025 2023-01-23 11:26:58.766778: step: 1318/1541, loss: 0.38669833540916443 2023-01-23 11:26:59.656546: step: 1320/1541, loss: 0.31124430894851685 2023-01-23 11:27:00.522764: step: 1322/1541, loss: 0.20145396888256073 2023-01-23 11:27:01.379412: step: 1324/1541, loss: 0.2016535997390747 2023-01-23 11:27:02.309688: step: 1326/1541, loss: 0.06533082574605942 2023-01-23 11:27:03.198860: step: 1328/1541, loss: 0.14481696486473083 2023-01-23 11:27:04.097692: step: 1330/1541, loss: 0.17235922813415527 2023-01-23 11:27:05.033809: step: 1332/1541, loss: 0.28952667117118835 2023-01-23 11:27:05.940352: step: 1334/1541, loss: 0.11233313381671906 2023-01-23 11:27:06.803663: step: 1336/1541, loss: 0.15474894642829895 2023-01-23 11:27:07.727909: step: 1338/1541, loss: 0.11641068011522293 2023-01-23 11:27:08.582799: step: 1340/1541, loss: 0.7777501940727234 2023-01-23 11:27:09.473294: step: 1342/1541, loss: 0.12436769902706146 2023-01-23 11:27:10.441757: step: 1344/1541, loss: 0.10018489509820938 2023-01-23 11:27:11.337065: step: 1346/1541, loss: 0.35666006803512573 2023-01-23 11:27:12.243942: step: 1348/1541, loss: 0.1643896847963333 2023-01-23 11:27:13.221407: step: 1350/1541, loss: 0.5003916621208191 2023-01-23 11:27:14.103065: step: 1352/1541, loss: 0.040318526327610016 2023-01-23 11:27:15.014754: step: 1354/1541, loss: 0.3008289635181427 2023-01-23 11:27:15.884989: step: 1356/1541, loss: 0.04588983207941055 2023-01-23 11:27:16.764953: step: 1358/1541, loss: 0.017675768584012985 2023-01-23 11:27:17.642565: step: 1360/1541, loss: 0.3887149393558502 2023-01-23 11:27:18.608991: step: 1362/1541, loss: 0.028542421758174896 2023-01-23 11:27:19.515646: step: 1364/1541, loss: 0.04933421313762665 2023-01-23 11:27:20.447244: step: 1366/1541, loss: 0.28630805015563965 2023-01-23 11:27:21.374811: step: 1368/1541, loss: 0.19787709414958954 2023-01-23 11:27:22.250423: step: 1370/1541, loss: 0.1010746881365776 2023-01-23 11:27:23.189397: step: 1372/1541, loss: 0.011080857366323471 2023-01-23 11:27:24.066472: step: 1374/1541, loss: 0.12951065599918365 2023-01-23 11:27:24.922015: step: 1376/1541, loss: 0.07513510435819626 2023-01-23 11:27:25.814371: step: 1378/1541, loss: 0.04874984174966812 2023-01-23 11:27:26.679997: step: 1380/1541, loss: 0.5124537944793701 2023-01-23 11:27:27.588441: step: 1382/1541, loss: 0.04713263735175133 2023-01-23 11:27:28.498943: step: 1384/1541, loss: 0.1557690054178238 2023-01-23 11:27:29.413092: step: 1386/1541, loss: 0.011422496289014816 2023-01-23 11:27:30.299318: step: 1388/1541, loss: 0.035729557275772095 2023-01-23 11:27:31.216003: step: 1390/1541, loss: 0.1319001168012619 2023-01-23 11:27:32.089648: step: 1392/1541, loss: 0.16261881589889526 2023-01-23 11:27:32.961201: step: 1394/1541, loss: 0.11641446501016617 2023-01-23 11:27:33.837263: step: 1396/1541, loss: 0.07931136339902878 2023-01-23 11:27:34.761565: step: 1398/1541, loss: 0.06947118043899536 2023-01-23 11:27:35.652000: step: 1400/1541, loss: 0.06812970340251923 2023-01-23 11:27:36.561091: step: 1402/1541, loss: 0.0467875674366951 2023-01-23 11:27:37.461119: step: 1404/1541, loss: 0.032044876366853714 2023-01-23 11:27:38.405844: step: 1406/1541, loss: 0.007277012336999178 2023-01-23 11:27:39.306480: step: 1408/1541, loss: 0.21576178073883057 2023-01-23 11:27:40.223091: step: 1410/1541, loss: 0.046777088195085526 2023-01-23 11:27:41.120870: step: 1412/1541, loss: 1.200054407119751 2023-01-23 11:27:41.994151: step: 1414/1541, loss: 0.13724441826343536 2023-01-23 11:27:42.852610: step: 1416/1541, loss: 1.3163418769836426 2023-01-23 11:27:43.758440: step: 1418/1541, loss: 0.08568943291902542 2023-01-23 11:27:44.643882: step: 1420/1541, loss: 0.04132668673992157 2023-01-23 11:27:45.508157: step: 1422/1541, loss: 0.030460458248853683 2023-01-23 11:27:46.368097: step: 1424/1541, loss: 0.02914373017847538 2023-01-23 11:27:47.373408: step: 1426/1541, loss: 0.07378309220075607 2023-01-23 11:27:48.252806: step: 1428/1541, loss: 0.379746675491333 2023-01-23 11:27:49.128676: step: 1430/1541, loss: 0.03951016440987587 2023-01-23 11:27:50.058748: step: 1432/1541, loss: 0.060640983283519745 2023-01-23 11:27:50.948681: step: 1434/1541, loss: 0.030003249645233154 2023-01-23 11:27:51.834462: step: 1436/1541, loss: 0.0443481020629406 2023-01-23 11:27:52.695644: step: 1438/1541, loss: 0.10429403930902481 2023-01-23 11:27:53.591035: step: 1440/1541, loss: 0.14850233495235443 2023-01-23 11:27:54.539358: step: 1442/1541, loss: 0.16663454473018646 2023-01-23 11:27:55.392278: step: 1444/1541, loss: 0.020804809406399727 2023-01-23 11:27:56.329160: step: 1446/1541, loss: 0.20727062225341797 2023-01-23 11:27:57.240714: step: 1448/1541, loss: 0.10392629355192184 2023-01-23 11:27:58.176950: step: 1450/1541, loss: 0.21574032306671143 2023-01-23 11:27:59.000848: step: 1452/1541, loss: 0.15132774412631989 2023-01-23 11:27:59.873742: step: 1454/1541, loss: 0.13207848370075226 2023-01-23 11:28:00.786404: step: 1456/1541, loss: 0.07506885379552841 2023-01-23 11:28:01.703896: step: 1458/1541, loss: 0.04481568560004234 2023-01-23 11:28:02.617290: step: 1460/1541, loss: 0.06733601540327072 2023-01-23 11:28:03.528298: step: 1462/1541, loss: 0.1526034027338028 2023-01-23 11:28:04.475299: step: 1464/1541, loss: 0.09948776662349701 2023-01-23 11:28:05.353483: step: 1466/1541, loss: 0.056336600333452225 2023-01-23 11:28:06.276719: step: 1468/1541, loss: 0.17507946491241455 2023-01-23 11:28:07.146495: step: 1470/1541, loss: 0.23616883158683777 2023-01-23 11:28:08.036637: step: 1472/1541, loss: 0.016175977885723114 2023-01-23 11:28:08.906097: step: 1474/1541, loss: 0.07375412434339523 2023-01-23 11:28:09.805488: step: 1476/1541, loss: 0.025645393878221512 2023-01-23 11:28:10.733884: step: 1478/1541, loss: 0.003009368199855089 2023-01-23 11:28:11.629063: step: 1480/1541, loss: 0.15373444557189941 2023-01-23 11:28:12.516125: step: 1482/1541, loss: 0.17388804256916046 2023-01-23 11:28:13.402406: step: 1484/1541, loss: 0.022092998027801514 2023-01-23 11:28:14.307751: step: 1486/1541, loss: 0.029000481590628624 2023-01-23 11:28:15.194881: step: 1488/1541, loss: 0.013052321970462799 2023-01-23 11:28:16.088056: step: 1490/1541, loss: 0.0859910324215889 2023-01-23 11:28:16.981643: step: 1492/1541, loss: 0.13832183182239532 2023-01-23 11:28:17.892748: step: 1494/1541, loss: 0.026819102466106415 2023-01-23 11:28:18.796100: step: 1496/1541, loss: 0.010098735801875591 2023-01-23 11:28:19.663320: step: 1498/1541, loss: 0.004591414239257574 2023-01-23 11:28:20.541325: step: 1500/1541, loss: 0.03228256478905678 2023-01-23 11:28:21.413045: step: 1502/1541, loss: 0.046213794499635696 2023-01-23 11:28:22.348309: step: 1504/1541, loss: 0.018834030255675316 2023-01-23 11:28:23.296132: step: 1506/1541, loss: 0.19068312644958496 2023-01-23 11:28:24.186518: step: 1508/1541, loss: 0.14435063302516937 2023-01-23 11:28:25.120056: step: 1510/1541, loss: 0.7740194797515869 2023-01-23 11:28:26.043922: step: 1512/1541, loss: 0.008333912119269371 2023-01-23 11:28:26.909492: step: 1514/1541, loss: 0.197796031832695 2023-01-23 11:28:27.817786: step: 1516/1541, loss: 0.08182873576879501 2023-01-23 11:28:28.735431: step: 1518/1541, loss: 0.18152853846549988 2023-01-23 11:28:29.633182: step: 1520/1541, loss: 0.062172628939151764 2023-01-23 11:28:30.540367: step: 1522/1541, loss: 0.6197700500488281 2023-01-23 11:28:31.496202: step: 1524/1541, loss: 0.1107693761587143 2023-01-23 11:28:32.361125: step: 1526/1541, loss: 0.23856376111507416 2023-01-23 11:28:33.262475: step: 1528/1541, loss: 0.010052417404949665 2023-01-23 11:28:34.186447: step: 1530/1541, loss: 0.09030185639858246 2023-01-23 11:28:35.092224: step: 1532/1541, loss: 0.04188487306237221 2023-01-23 11:28:35.983804: step: 1534/1541, loss: 0.05448955297470093 2023-01-23 11:28:36.891283: step: 1536/1541, loss: 0.23570312559604645 2023-01-23 11:28:37.794935: step: 1538/1541, loss: 0.04454491659998894 2023-01-23 11:28:38.651243: step: 1540/1541, loss: 0.051881641149520874 2023-01-23 11:28:39.533473: step: 1542/1541, loss: 0.21752667427062988 2023-01-23 11:28:40.441698: step: 1544/1541, loss: 0.10040756314992905 2023-01-23 11:28:41.360268: step: 1546/1541, loss: 0.02769000083208084 2023-01-23 11:28:42.286198: step: 1548/1541, loss: 0.5534182190895081 2023-01-23 11:28:43.186981: step: 1550/1541, loss: 0.21956560015678406 2023-01-23 11:28:44.063678: step: 1552/1541, loss: 0.805726945400238 2023-01-23 11:28:44.938148: step: 1554/1541, loss: 0.01878976635634899 2023-01-23 11:28:45.814953: step: 1556/1541, loss: 0.15675228834152222 2023-01-23 11:28:46.720778: step: 1558/1541, loss: 0.21988725662231445 2023-01-23 11:28:47.605663: step: 1560/1541, loss: 0.06440991163253784 2023-01-23 11:28:48.484256: step: 1562/1541, loss: 0.012370516546070576 2023-01-23 11:28:49.393801: step: 1564/1541, loss: 0.07030744105577469 2023-01-23 11:28:50.325065: step: 1566/1541, loss: 0.01438087597489357 2023-01-23 11:28:51.265137: step: 1568/1541, loss: 0.06506643444299698 2023-01-23 11:28:52.148493: step: 1570/1541, loss: 0.040266260504722595 2023-01-23 11:28:53.054453: step: 1572/1541, loss: 0.06501896679401398 2023-01-23 11:28:53.959081: step: 1574/1541, loss: 0.368649959564209 2023-01-23 11:28:54.841060: step: 1576/1541, loss: 0.20470030605793 2023-01-23 11:28:55.725240: step: 1578/1541, loss: 0.2497289925813675 2023-01-23 11:28:56.648277: step: 1580/1541, loss: 0.13358698785305023 2023-01-23 11:28:57.580846: step: 1582/1541, loss: 0.15814048051834106 2023-01-23 11:28:58.503420: step: 1584/1541, loss: 0.11875827610492706 2023-01-23 11:28:59.345762: step: 1586/1541, loss: 0.0629463940858841 2023-01-23 11:29:00.233426: step: 1588/1541, loss: 0.010304844006896019 2023-01-23 11:29:01.120197: step: 1590/1541, loss: 0.04191921278834343 2023-01-23 11:29:02.052835: step: 1592/1541, loss: 0.01605628803372383 2023-01-23 11:29:02.940854: step: 1594/1541, loss: 0.040691353380680084 2023-01-23 11:29:03.844308: step: 1596/1541, loss: 0.03666936978697777 2023-01-23 11:29:04.768515: step: 1598/1541, loss: 0.15367478132247925 2023-01-23 11:29:05.704784: step: 1600/1541, loss: 0.07092498987913132 2023-01-23 11:29:06.580722: step: 1602/1541, loss: 0.06473745405673981 2023-01-23 11:29:07.518297: step: 1604/1541, loss: 0.12075838446617126 2023-01-23 11:29:08.421937: step: 1606/1541, loss: 0.047565676271915436 2023-01-23 11:29:09.324028: step: 1608/1541, loss: 0.3857310712337494 2023-01-23 11:29:10.194777: step: 1610/1541, loss: 0.039693862199783325 2023-01-23 11:29:11.092895: step: 1612/1541, loss: 0.06821931898593903 2023-01-23 11:29:12.022652: step: 1614/1541, loss: 0.6618573665618896 2023-01-23 11:29:13.015660: step: 1616/1541, loss: 0.15908776223659515 2023-01-23 11:29:13.884923: step: 1618/1541, loss: 0.08099617063999176 2023-01-23 11:29:14.748960: step: 1620/1541, loss: 0.29128366708755493 2023-01-23 11:29:15.670734: step: 1622/1541, loss: 0.06397157907485962 2023-01-23 11:29:16.566250: step: 1624/1541, loss: 0.38646066188812256 2023-01-23 11:29:17.479085: step: 1626/1541, loss: 0.09340866655111313 2023-01-23 11:29:18.316989: step: 1628/1541, loss: 0.008994465693831444 2023-01-23 11:29:19.185667: step: 1630/1541, loss: 0.13173341751098633 2023-01-23 11:29:20.101461: step: 1632/1541, loss: 0.4985824525356293 2023-01-23 11:29:21.018074: step: 1634/1541, loss: 0.030624128878116608 2023-01-23 11:29:21.916467: step: 1636/1541, loss: 0.02720528282225132 2023-01-23 11:29:22.884007: step: 1638/1541, loss: 0.4675423204898834 2023-01-23 11:29:23.792550: step: 1640/1541, loss: 0.11226141452789307 2023-01-23 11:29:24.698211: step: 1642/1541, loss: 0.05369512736797333 2023-01-23 11:29:25.595962: step: 1644/1541, loss: 0.01193840242922306 2023-01-23 11:29:26.502648: step: 1646/1541, loss: 0.42508479952812195 2023-01-23 11:29:27.363011: step: 1648/1541, loss: 0.20862248539924622 2023-01-23 11:29:28.241079: step: 1650/1541, loss: 0.4978501796722412 2023-01-23 11:29:29.129774: step: 1652/1541, loss: 0.14967496693134308 2023-01-23 11:29:30.016682: step: 1654/1541, loss: 0.1550142467021942 2023-01-23 11:29:30.892640: step: 1656/1541, loss: 0.0492568165063858 2023-01-23 11:29:31.753067: step: 1658/1541, loss: 0.15164968371391296 2023-01-23 11:29:32.618271: step: 1660/1541, loss: 0.06761366873979568 2023-01-23 11:29:33.539300: step: 1662/1541, loss: 0.15395455062389374 2023-01-23 11:29:34.405836: step: 1664/1541, loss: 0.041736848652362823 2023-01-23 11:29:35.304687: step: 1666/1541, loss: 0.03494610637426376 2023-01-23 11:29:36.201670: step: 1668/1541, loss: 0.0262096356600523 2023-01-23 11:29:37.070036: step: 1670/1541, loss: 0.06499158591032028 2023-01-23 11:29:38.017190: step: 1672/1541, loss: 0.023166490718722343 2023-01-23 11:29:38.943972: step: 1674/1541, loss: 0.16012760996818542 2023-01-23 11:29:39.813988: step: 1676/1541, loss: 0.07961701601743698 2023-01-23 11:29:40.683874: step: 1678/1541, loss: 0.19168898463249207 2023-01-23 11:29:41.597113: step: 1680/1541, loss: 0.1766466349363327 2023-01-23 11:29:42.502137: step: 1682/1541, loss: 0.3412781059741974 2023-01-23 11:29:43.438420: step: 1684/1541, loss: 0.2846377491950989 2023-01-23 11:29:44.322922: step: 1686/1541, loss: 0.8189166188240051 2023-01-23 11:29:45.201081: step: 1688/1541, loss: 0.01239559892565012 2023-01-23 11:29:46.071613: step: 1690/1541, loss: 0.016074109822511673 2023-01-23 11:29:47.038730: step: 1692/1541, loss: 0.1860160231590271 2023-01-23 11:29:47.937610: step: 1694/1541, loss: 0.034479305148124695 2023-01-23 11:29:48.842227: step: 1696/1541, loss: 0.12060003727674484 2023-01-23 11:29:49.736245: step: 1698/1541, loss: 0.15999330580234528 2023-01-23 11:29:50.702086: step: 1700/1541, loss: 0.20376646518707275 2023-01-23 11:29:51.621501: step: 1702/1541, loss: 0.31837692856788635 2023-01-23 11:29:52.542169: step: 1704/1541, loss: 0.056962817907333374 2023-01-23 11:29:53.413871: step: 1706/1541, loss: 0.13015346229076385 2023-01-23 11:29:54.292053: step: 1708/1541, loss: 0.017358487471938133 2023-01-23 11:29:55.237384: step: 1710/1541, loss: 0.012774275615811348 2023-01-23 11:29:56.136349: step: 1712/1541, loss: 0.01684427820146084 2023-01-23 11:29:57.020608: step: 1714/1541, loss: 0.060114920139312744 2023-01-23 11:29:57.927201: step: 1716/1541, loss: 0.8350799083709717 2023-01-23 11:29:58.798894: step: 1718/1541, loss: 0.10209286212921143 2023-01-23 11:29:59.675656: step: 1720/1541, loss: 0.10635983943939209 2023-01-23 11:30:00.566650: step: 1722/1541, loss: 0.027666311711072922 2023-01-23 11:30:01.490782: step: 1724/1541, loss: 0.03445244953036308 2023-01-23 11:30:02.448509: step: 1726/1541, loss: 0.057437196373939514 2023-01-23 11:30:03.307083: step: 1728/1541, loss: 0.13039608299732208 2023-01-23 11:30:04.206600: step: 1730/1541, loss: 0.023331211879849434 2023-01-23 11:30:05.115461: step: 1732/1541, loss: 0.2270517498254776 2023-01-23 11:30:06.015742: step: 1734/1541, loss: 0.5757845640182495 2023-01-23 11:30:06.927162: step: 1736/1541, loss: 0.18472231924533844 2023-01-23 11:30:07.781982: step: 1738/1541, loss: 0.017508817836642265 2023-01-23 11:30:08.695414: step: 1740/1541, loss: 0.004944333806633949 2023-01-23 11:30:09.650666: step: 1742/1541, loss: 0.10220006853342056 2023-01-23 11:30:10.536109: step: 1744/1541, loss: 0.1719512790441513 2023-01-23 11:30:11.484386: step: 1746/1541, loss: 0.00824282132089138 2023-01-23 11:30:12.381879: step: 1748/1541, loss: 0.12341200560331345 2023-01-23 11:30:13.261684: step: 1750/1541, loss: 0.07477424293756485 2023-01-23 11:30:14.258099: step: 1752/1541, loss: 0.04935687035322189 2023-01-23 11:30:15.196280: step: 1754/1541, loss: 0.026681572198867798 2023-01-23 11:30:16.094116: step: 1756/1541, loss: 0.041641827672719955 2023-01-23 11:30:17.002538: step: 1758/1541, loss: 0.28539037704467773 2023-01-23 11:30:17.926867: step: 1760/1541, loss: 0.06960959732532501 2023-01-23 11:30:18.820529: step: 1762/1541, loss: 0.0839235782623291 2023-01-23 11:30:19.840829: step: 1764/1541, loss: 0.5498628616333008 2023-01-23 11:30:20.738625: step: 1766/1541, loss: 0.0822870209813118 2023-01-23 11:30:21.674523: step: 1768/1541, loss: 0.06634991616010666 2023-01-23 11:30:22.549240: step: 1770/1541, loss: 0.0027368408627808094 2023-01-23 11:30:23.485758: step: 1772/1541, loss: 0.07725048065185547 2023-01-23 11:30:24.361744: step: 1774/1541, loss: 0.012276582419872284 2023-01-23 11:30:25.217014: step: 1776/1541, loss: 0.03123147040605545 2023-01-23 11:30:26.077502: step: 1778/1541, loss: 0.016000807285308838 2023-01-23 11:30:26.999429: step: 1780/1541, loss: 0.004595949314534664 2023-01-23 11:30:27.840309: step: 1782/1541, loss: 0.03612002730369568 2023-01-23 11:30:28.762546: step: 1784/1541, loss: 0.20911607146263123 2023-01-23 11:30:29.657203: step: 1786/1541, loss: 0.30844026803970337 2023-01-23 11:30:30.572534: step: 1788/1541, loss: 0.00773126445710659 2023-01-23 11:30:31.442735: step: 1790/1541, loss: 0.015319227240979671 2023-01-23 11:30:32.345399: step: 1792/1541, loss: 0.004629236645996571 2023-01-23 11:30:33.268567: step: 1794/1541, loss: 0.11452486366033554 2023-01-23 11:30:34.201285: step: 1796/1541, loss: 0.05928951874375343 2023-01-23 11:30:35.149887: step: 1798/1541, loss: 0.10836081206798553 2023-01-23 11:30:36.038191: step: 1800/1541, loss: 0.1476447731256485 2023-01-23 11:30:36.971144: step: 1802/1541, loss: 0.007657576352357864 2023-01-23 11:30:37.915686: step: 1804/1541, loss: 0.08903269469738007 2023-01-23 11:30:38.832117: step: 1806/1541, loss: 0.05736815556883812 2023-01-23 11:30:39.713613: step: 1808/1541, loss: 0.7091104388237 2023-01-23 11:30:40.605959: step: 1810/1541, loss: 0.0110403411090374 2023-01-23 11:30:41.451446: step: 1812/1541, loss: 0.04817195236682892 2023-01-23 11:30:42.383171: step: 1814/1541, loss: 0.16941384971141815 2023-01-23 11:30:43.279510: step: 1816/1541, loss: 0.025701379403471947 2023-01-23 11:30:44.161878: step: 1818/1541, loss: 0.11957015842199326 2023-01-23 11:30:45.100788: step: 1820/1541, loss: 0.02683783322572708 2023-01-23 11:30:46.036310: step: 1822/1541, loss: 0.2532755732536316 2023-01-23 11:30:47.028464: step: 1824/1541, loss: 0.0016599418595433235 2023-01-23 11:30:47.939541: step: 1826/1541, loss: 0.0052885254845023155 2023-01-23 11:30:48.835185: step: 1828/1541, loss: 0.07124373316764832 2023-01-23 11:30:49.709039: step: 1830/1541, loss: 0.0685608983039856 2023-01-23 11:30:50.581173: step: 1832/1541, loss: 0.07810270041227341 2023-01-23 11:30:51.497263: step: 1834/1541, loss: 0.07246265560388565 2023-01-23 11:30:52.395674: step: 1836/1541, loss: 0.07051962614059448 2023-01-23 11:30:53.325335: step: 1838/1541, loss: 0.035234738141298294 2023-01-23 11:30:54.249351: step: 1840/1541, loss: 0.04080755263566971 2023-01-23 11:30:55.162177: step: 1842/1541, loss: 0.19669868052005768 2023-01-23 11:30:56.071576: step: 1844/1541, loss: 0.1296238899230957 2023-01-23 11:30:56.996689: step: 1846/1541, loss: 0.04059405252337456 2023-01-23 11:30:57.907167: step: 1848/1541, loss: 0.04345513880252838 2023-01-23 11:30:58.798548: step: 1850/1541, loss: 0.2837470471858978 2023-01-23 11:30:59.727992: step: 1852/1541, loss: 0.2700316905975342 2023-01-23 11:31:00.696292: step: 1854/1541, loss: 0.15891599655151367 2023-01-23 11:31:01.647537: step: 1856/1541, loss: 0.13955670595169067 2023-01-23 11:31:02.577199: step: 1858/1541, loss: 0.009391231462359428 2023-01-23 11:31:03.437244: step: 1860/1541, loss: 0.00949101708829403 2023-01-23 11:31:04.347852: step: 1862/1541, loss: 0.3552023768424988 2023-01-23 11:31:05.308076: step: 1864/1541, loss: 0.6598898768424988 2023-01-23 11:31:06.151659: step: 1866/1541, loss: 0.05348741635680199 2023-01-23 11:31:07.042203: step: 1868/1541, loss: 0.07394881546497345 2023-01-23 11:31:07.946983: step: 1870/1541, loss: 0.36046603322029114 2023-01-23 11:31:08.844295: step: 1872/1541, loss: 0.27239808440208435 2023-01-23 11:31:09.744113: step: 1874/1541, loss: 0.0709044337272644 2023-01-23 11:31:10.614875: step: 1876/1541, loss: 0.07959195971488953 2023-01-23 11:31:11.499122: step: 1878/1541, loss: 0.20707693696022034 2023-01-23 11:31:12.394989: step: 1880/1541, loss: 0.03845994919538498 2023-01-23 11:31:13.291968: step: 1882/1541, loss: 0.6976993083953857 2023-01-23 11:31:14.173132: step: 1884/1541, loss: 0.08190429955720901 2023-01-23 11:31:15.043080: step: 1886/1541, loss: 0.11597238481044769 2023-01-23 11:31:15.954933: step: 1888/1541, loss: 0.1932561844587326 2023-01-23 11:31:16.900992: step: 1890/1541, loss: 0.4200167953968048 2023-01-23 11:31:17.788723: step: 1892/1541, loss: 0.6850548982620239 2023-01-23 11:31:18.659957: step: 1894/1541, loss: 2.3448030948638916 2023-01-23 11:31:19.564599: step: 1896/1541, loss: 0.44520294666290283 2023-01-23 11:31:20.428920: step: 1898/1541, loss: 0.2121071219444275 2023-01-23 11:31:21.307898: step: 1900/1541, loss: 0.024510705843567848 2023-01-23 11:31:22.223835: step: 1902/1541, loss: 0.06012742593884468 2023-01-23 11:31:23.105791: step: 1904/1541, loss: 0.041975103318691254 2023-01-23 11:31:23.994746: step: 1906/1541, loss: 0.2553357183933258 2023-01-23 11:31:24.860857: step: 1908/1541, loss: 0.09519793093204498 2023-01-23 11:31:25.737775: step: 1910/1541, loss: 0.18880361318588257 2023-01-23 11:31:26.619857: step: 1912/1541, loss: 0.263629287481308 2023-01-23 11:31:27.544916: step: 1914/1541, loss: 0.1031312495470047 2023-01-23 11:31:28.414391: step: 1916/1541, loss: 0.093878373503685 2023-01-23 11:31:29.442846: step: 1918/1541, loss: 1.1740338802337646 2023-01-23 11:31:30.391587: step: 1920/1541, loss: 0.24244749546051025 2023-01-23 11:31:31.272475: step: 1922/1541, loss: 0.2632635533809662 2023-01-23 11:31:32.206519: step: 1924/1541, loss: 0.18531116843223572 2023-01-23 11:31:33.068628: step: 1926/1541, loss: 0.26783257722854614 2023-01-23 11:31:33.961772: step: 1928/1541, loss: 0.06442149728536606 2023-01-23 11:31:34.887683: step: 1930/1541, loss: 0.45655491948127747 2023-01-23 11:31:35.903404: step: 1932/1541, loss: 0.2932225465774536 2023-01-23 11:31:36.783956: step: 1934/1541, loss: 0.12099100649356842 2023-01-23 11:31:37.769630: step: 1936/1541, loss: 0.07594385743141174 2023-01-23 11:31:38.663363: step: 1938/1541, loss: 0.5671000480651855 2023-01-23 11:31:39.571978: step: 1940/1541, loss: 0.011348539032042027 2023-01-23 11:31:40.467112: step: 1942/1541, loss: 0.1769806146621704 2023-01-23 11:31:41.413793: step: 1944/1541, loss: 0.04024018719792366 2023-01-23 11:31:42.254104: step: 1946/1541, loss: 0.12414246052503586 2023-01-23 11:31:43.190800: step: 1948/1541, loss: 0.008293627761304379 2023-01-23 11:31:44.168007: step: 1950/1541, loss: 0.044566765427589417 2023-01-23 11:31:45.032306: step: 1952/1541, loss: 0.021668199449777603 2023-01-23 11:31:45.958724: step: 1954/1541, loss: 0.08599594980478287 2023-01-23 11:31:46.848181: step: 1956/1541, loss: 0.08278904855251312 2023-01-23 11:31:47.770000: step: 1958/1541, loss: 0.08120645582675934 2023-01-23 11:31:48.643175: step: 1960/1541, loss: 0.031467895954847336 2023-01-23 11:31:49.526146: step: 1962/1541, loss: 0.02185845747590065 2023-01-23 11:31:50.455875: step: 1964/1541, loss: 0.004004445392638445 2023-01-23 11:31:51.346730: step: 1966/1541, loss: 0.1213933527469635 2023-01-23 11:31:52.267424: step: 1968/1541, loss: 2.570669174194336 2023-01-23 11:31:53.147464: step: 1970/1541, loss: 0.06743604689836502 2023-01-23 11:31:54.047186: step: 1972/1541, loss: 0.2096899151802063 2023-01-23 11:31:54.971331: step: 1974/1541, loss: 0.029584025964140892 2023-01-23 11:31:55.892392: step: 1976/1541, loss: 0.06555704027414322 2023-01-23 11:31:56.743284: step: 1978/1541, loss: 0.03146801516413689 2023-01-23 11:31:57.636104: step: 1980/1541, loss: 0.06049183011054993 2023-01-23 11:31:58.577923: step: 1982/1541, loss: 0.17079931497573853 2023-01-23 11:31:59.456032: step: 1984/1541, loss: 0.05984499678015709 2023-01-23 11:32:00.409918: step: 1986/1541, loss: 0.1614760011434555 2023-01-23 11:32:01.283147: step: 1988/1541, loss: 0.03686691075563431 2023-01-23 11:32:02.178991: step: 1990/1541, loss: 0.0073898304253816605 2023-01-23 11:32:03.051704: step: 1992/1541, loss: 0.07116550207138062 2023-01-23 11:32:03.939878: step: 1994/1541, loss: 0.04787563160061836 2023-01-23 11:32:04.824029: step: 1996/1541, loss: 0.08918403834104538 2023-01-23 11:32:05.752077: step: 1998/1541, loss: 0.5566454529762268 2023-01-23 11:32:06.640715: step: 2000/1541, loss: 0.25114068388938904 2023-01-23 11:32:07.496462: step: 2002/1541, loss: 0.6099993586540222 2023-01-23 11:32:08.383256: step: 2004/1541, loss: 0.2797550857067108 2023-01-23 11:32:09.290002: step: 2006/1541, loss: 0.04973991587758064 2023-01-23 11:32:10.191326: step: 2008/1541, loss: 0.23518610000610352 2023-01-23 11:32:11.106598: step: 2010/1541, loss: 0.02996823377907276 2023-01-23 11:32:11.990406: step: 2012/1541, loss: 0.043001122772693634 2023-01-23 11:32:12.888487: step: 2014/1541, loss: 0.032081954181194305 2023-01-23 11:32:13.811955: step: 2016/1541, loss: 0.03240391984581947 2023-01-23 11:32:14.671035: step: 2018/1541, loss: 0.12980739772319794 2023-01-23 11:32:15.624642: step: 2020/1541, loss: 0.046676475554704666 2023-01-23 11:32:16.534977: step: 2022/1541, loss: 0.052797555923461914 2023-01-23 11:32:17.440422: step: 2024/1541, loss: 0.06007875129580498 2023-01-23 11:32:18.353603: step: 2026/1541, loss: 0.1748543679714203 2023-01-23 11:32:19.212864: step: 2028/1541, loss: 0.018446823582053185 2023-01-23 11:32:20.142642: step: 2030/1541, loss: 0.18950240314006805 2023-01-23 11:32:21.035325: step: 2032/1541, loss: 0.2883753180503845 2023-01-23 11:32:21.956773: step: 2034/1541, loss: 0.21878869831562042 2023-01-23 11:32:22.897005: step: 2036/1541, loss: 0.008580048568546772 2023-01-23 11:32:23.825857: step: 2038/1541, loss: 1.0696779489517212 2023-01-23 11:32:24.779985: step: 2040/1541, loss: 0.004492780193686485 2023-01-23 11:32:25.670382: step: 2042/1541, loss: 0.08442682027816772 2023-01-23 11:32:26.625993: step: 2044/1541, loss: 0.0719684511423111 2023-01-23 11:32:27.511160: step: 2046/1541, loss: 0.023368187248706818 2023-01-23 11:32:28.403373: step: 2048/1541, loss: 0.05571036785840988 2023-01-23 11:32:29.310011: step: 2050/1541, loss: 0.18620772659778595 2023-01-23 11:32:30.274243: step: 2052/1541, loss: 0.6986135840415955 2023-01-23 11:32:31.169217: step: 2054/1541, loss: 1.4346846342086792 2023-01-23 11:32:32.097783: step: 2056/1541, loss: 0.09303213655948639 2023-01-23 11:32:33.042835: step: 2058/1541, loss: 0.024297943338751793 2023-01-23 11:32:33.920647: step: 2060/1541, loss: 0.21867723762989044 2023-01-23 11:32:34.789196: step: 2062/1541, loss: 0.4828992486000061 2023-01-23 11:32:35.659689: step: 2064/1541, loss: 0.03450565040111542 2023-01-23 11:32:36.511530: step: 2066/1541, loss: 0.07459624111652374 2023-01-23 11:32:37.505733: step: 2068/1541, loss: 0.09926682710647583 2023-01-23 11:32:38.371639: step: 2070/1541, loss: 0.3472183048725128 2023-01-23 11:32:39.248200: step: 2072/1541, loss: 0.02868659980595112 2023-01-23 11:32:40.121765: step: 2074/1541, loss: 0.13654276728630066 2023-01-23 11:32:41.046718: step: 2076/1541, loss: 0.23977932333946228 2023-01-23 11:32:41.919668: step: 2078/1541, loss: 0.0997164249420166 2023-01-23 11:32:42.787584: step: 2080/1541, loss: 0.9712116122245789 2023-01-23 11:32:43.724774: step: 2082/1541, loss: 0.01960262842476368 2023-01-23 11:32:44.606792: step: 2084/1541, loss: 0.26048213243484497 2023-01-23 11:32:45.531539: step: 2086/1541, loss: 0.12610183656215668 2023-01-23 11:32:46.421609: step: 2088/1541, loss: 0.050930436700582504 2023-01-23 11:32:47.284210: step: 2090/1541, loss: 0.12022151052951813 2023-01-23 11:32:48.169325: step: 2092/1541, loss: 0.20939476788043976 2023-01-23 11:32:49.081024: step: 2094/1541, loss: 0.08685023337602615 2023-01-23 11:32:50.000335: step: 2096/1541, loss: 1.2972019910812378 2023-01-23 11:32:50.894136: step: 2098/1541, loss: 0.14530353248119354 2023-01-23 11:32:51.742286: step: 2100/1541, loss: 0.08671976625919342 2023-01-23 11:32:52.606081: step: 2102/1541, loss: 0.02251506969332695 2023-01-23 11:32:53.487138: step: 2104/1541, loss: 0.13370302319526672 2023-01-23 11:32:54.447879: step: 2106/1541, loss: 0.24628067016601562 2023-01-23 11:32:55.365810: step: 2108/1541, loss: 0.011168577708303928 2023-01-23 11:32:56.245453: step: 2110/1541, loss: 0.1538115292787552 2023-01-23 11:32:57.216625: step: 2112/1541, loss: 0.33072125911712646 2023-01-23 11:32:58.086930: step: 2114/1541, loss: 0.01806005835533142 2023-01-23 11:32:59.001280: step: 2116/1541, loss: 0.09343986213207245 2023-01-23 11:32:59.896923: step: 2118/1541, loss: 0.10641132295131683 2023-01-23 11:33:00.767203: step: 2120/1541, loss: 0.002810212317854166 2023-01-23 11:33:01.636829: step: 2122/1541, loss: 0.050713248550891876 2023-01-23 11:33:02.554296: step: 2124/1541, loss: 0.20982462167739868 2023-01-23 11:33:03.459051: step: 2126/1541, loss: 0.037072256207466125 2023-01-23 11:33:04.359310: step: 2128/1541, loss: 0.3370184302330017 2023-01-23 11:33:05.238732: step: 2130/1541, loss: 0.20970284938812256 2023-01-23 11:33:06.154296: step: 2132/1541, loss: 0.0722290426492691 2023-01-23 11:33:07.045726: step: 2134/1541, loss: 0.04932541400194168 2023-01-23 11:33:07.925054: step: 2136/1541, loss: 0.27415958046913147 2023-01-23 11:33:08.919657: step: 2138/1541, loss: 0.07550540566444397 2023-01-23 11:33:09.812526: step: 2140/1541, loss: 0.009115748107433319 2023-01-23 11:33:10.701966: step: 2142/1541, loss: 0.06873700022697449 2023-01-23 11:33:11.618836: step: 2144/1541, loss: 0.03752001374959946 2023-01-23 11:33:12.542704: step: 2146/1541, loss: 0.20425677299499512 2023-01-23 11:33:13.453271: step: 2148/1541, loss: 0.00704661151394248 2023-01-23 11:33:14.342593: step: 2150/1541, loss: 0.06588288396596909 2023-01-23 11:33:15.221196: step: 2152/1541, loss: 0.4374523460865021 2023-01-23 11:33:16.066563: step: 2154/1541, loss: 0.015521018765866756 2023-01-23 11:33:16.984600: step: 2156/1541, loss: 0.26488351821899414 2023-01-23 11:33:17.831150: step: 2158/1541, loss: 0.19497373700141907 2023-01-23 11:33:18.738072: step: 2160/1541, loss: 0.10386752337217331 2023-01-23 11:33:19.646985: step: 2162/1541, loss: 0.06035104766488075 2023-01-23 11:33:20.536309: step: 2164/1541, loss: 0.007865473628044128 2023-01-23 11:33:21.422199: step: 2166/1541, loss: 0.07806967198848724 2023-01-23 11:33:22.342874: step: 2168/1541, loss: 0.26000478863716125 2023-01-23 11:33:23.246278: step: 2170/1541, loss: 0.20239518582820892 2023-01-23 11:33:24.136633: step: 2172/1541, loss: 0.0023991623893380165 2023-01-23 11:33:25.029208: step: 2174/1541, loss: 0.04694639518857002 2023-01-23 11:33:25.955743: step: 2176/1541, loss: 0.20624622702598572 2023-01-23 11:33:26.854072: step: 2178/1541, loss: 0.11366432905197144 2023-01-23 11:33:27.704555: step: 2180/1541, loss: 0.01575280912220478 2023-01-23 11:33:28.617590: step: 2182/1541, loss: 0.042905572801828384 2023-01-23 11:33:29.487108: step: 2184/1541, loss: 0.01872638799250126 2023-01-23 11:33:30.397850: step: 2186/1541, loss: 0.017501741647720337 2023-01-23 11:33:31.306132: step: 2188/1541, loss: 0.07364249974489212 2023-01-23 11:33:32.186544: step: 2190/1541, loss: 0.478676438331604 2023-01-23 11:33:33.082149: step: 2192/1541, loss: 0.11948052048683167 2023-01-23 11:33:33.995216: step: 2194/1541, loss: 0.07954448461532593 2023-01-23 11:33:34.868058: step: 2196/1541, loss: 0.050284791737794876 2023-01-23 11:33:35.799821: step: 2198/1541, loss: 0.16928628087043762 2023-01-23 11:33:36.670236: step: 2200/1541, loss: 0.0057828049175441265 2023-01-23 11:33:37.533008: step: 2202/1541, loss: 0.028979480266571045 2023-01-23 11:33:38.428094: step: 2204/1541, loss: 0.04416550695896149 2023-01-23 11:33:39.313371: step: 2206/1541, loss: 0.05013269558548927 2023-01-23 11:33:40.175596: step: 2208/1541, loss: 0.21918326616287231 2023-01-23 11:33:41.091153: step: 2210/1541, loss: 0.40617942810058594 2023-01-23 11:33:42.001277: step: 2212/1541, loss: 0.022435741499066353 2023-01-23 11:33:42.884928: step: 2214/1541, loss: 0.36120089888572693 2023-01-23 11:33:43.782191: step: 2216/1541, loss: 0.22704853117465973 2023-01-23 11:33:44.678531: step: 2218/1541, loss: 0.004063382279127836 2023-01-23 11:33:45.613175: step: 2220/1541, loss: 0.4018087685108185 2023-01-23 11:33:46.512538: step: 2222/1541, loss: 0.1016872376203537 2023-01-23 11:33:47.399883: step: 2224/1541, loss: 0.017881132662296295 2023-01-23 11:33:48.291364: step: 2226/1541, loss: 0.5182378888130188 2023-01-23 11:33:49.215899: step: 2228/1541, loss: 0.12886106967926025 2023-01-23 11:33:50.117166: step: 2230/1541, loss: 0.24167297780513763 2023-01-23 11:33:50.998004: step: 2232/1541, loss: 0.6806938648223877 2023-01-23 11:33:51.869731: step: 2234/1541, loss: 0.009136760607361794 2023-01-23 11:33:52.786930: step: 2236/1541, loss: 0.21560163795948029 2023-01-23 11:33:53.700244: step: 2238/1541, loss: 0.004733535461127758 2023-01-23 11:33:54.591451: step: 2240/1541, loss: 0.037519343197345734 2023-01-23 11:33:55.498985: step: 2242/1541, loss: 0.009853295050561428 2023-01-23 11:33:56.423231: step: 2244/1541, loss: 0.021656567230820656 2023-01-23 11:33:57.325969: step: 2246/1541, loss: 0.26066404581069946 2023-01-23 11:33:58.204656: step: 2248/1541, loss: 0.16771355271339417 2023-01-23 11:33:59.104608: step: 2250/1541, loss: 0.020442143082618713 2023-01-23 11:33:59.979348: step: 2252/1541, loss: 0.057187266647815704 2023-01-23 11:34:00.924040: step: 2254/1541, loss: 0.05691935494542122 2023-01-23 11:34:01.879909: step: 2256/1541, loss: 0.8265860676765442 2023-01-23 11:34:02.740089: step: 2258/1541, loss: 0.17528332769870758 2023-01-23 11:34:03.678240: step: 2260/1541, loss: 0.13562597334384918 2023-01-23 11:34:04.600431: step: 2262/1541, loss: 0.1748066395521164 2023-01-23 11:34:05.497443: step: 2264/1541, loss: 0.12429675459861755 2023-01-23 11:34:06.374405: step: 2266/1541, loss: 0.030882518738508224 2023-01-23 11:34:07.241074: step: 2268/1541, loss: 0.25216761231422424 2023-01-23 11:34:08.195236: step: 2270/1541, loss: 0.005994490813463926 2023-01-23 11:34:09.145388: step: 2272/1541, loss: 1.0892362594604492 2023-01-23 11:34:10.054404: step: 2274/1541, loss: 0.21458736062049866 2023-01-23 11:34:10.960923: step: 2276/1541, loss: 0.008365095593035221 2023-01-23 11:34:11.834022: step: 2278/1541, loss: 0.02862369641661644 2023-01-23 11:34:12.736473: step: 2280/1541, loss: 0.028922507539391518 2023-01-23 11:34:13.668024: step: 2282/1541, loss: 0.4593231976032257 2023-01-23 11:34:14.541569: step: 2284/1541, loss: 0.17557206749916077 2023-01-23 11:34:15.428272: step: 2286/1541, loss: 0.01801036298274994 2023-01-23 11:34:16.367158: step: 2288/1541, loss: 0.05355557054281235 2023-01-23 11:34:17.318393: step: 2290/1541, loss: 0.04049030691385269 2023-01-23 11:34:18.165867: step: 2292/1541, loss: 0.0742291733622551 2023-01-23 11:34:19.103523: step: 2294/1541, loss: 0.34573838114738464 2023-01-23 11:34:19.997505: step: 2296/1541, loss: 0.28401920199394226 2023-01-23 11:34:20.909006: step: 2298/1541, loss: 0.14594346284866333 2023-01-23 11:34:21.802800: step: 2300/1541, loss: 0.016066698357462883 2023-01-23 11:34:22.661074: step: 2302/1541, loss: 0.14212509989738464 2023-01-23 11:34:23.555412: step: 2304/1541, loss: 0.07280077785253525 2023-01-23 11:34:24.488500: step: 2306/1541, loss: 0.01382734626531601 2023-01-23 11:34:25.438231: step: 2308/1541, loss: 0.08825158327817917 2023-01-23 11:34:26.307760: step: 2310/1541, loss: 0.3526081144809723 2023-01-23 11:34:27.186993: step: 2312/1541, loss: 0.004225259646773338 2023-01-23 11:34:28.063361: step: 2314/1541, loss: 0.11046448349952698 2023-01-23 11:34:28.928981: step: 2316/1541, loss: 0.10036294162273407 2023-01-23 11:34:29.828053: step: 2318/1541, loss: 0.013804025948047638 2023-01-23 11:34:30.736602: step: 2320/1541, loss: 0.11418575793504715 2023-01-23 11:34:31.577674: step: 2322/1541, loss: 0.2251463383436203 2023-01-23 11:34:32.470904: step: 2324/1541, loss: 0.004989235196262598 2023-01-23 11:34:33.328715: step: 2326/1541, loss: 0.0019348232308402658 2023-01-23 11:34:34.236500: step: 2328/1541, loss: 0.11820047348737717 2023-01-23 11:34:35.104814: step: 2330/1541, loss: 0.009243915788829327 2023-01-23 11:34:36.056508: step: 2332/1541, loss: 0.030018024146556854 2023-01-23 11:34:36.992245: step: 2334/1541, loss: 0.47863975167274475 2023-01-23 11:34:37.869201: step: 2336/1541, loss: 0.15439477562904358 2023-01-23 11:34:38.740917: step: 2338/1541, loss: 0.23377762734889984 2023-01-23 11:34:39.656217: step: 2340/1541, loss: 0.6696271896362305 2023-01-23 11:34:40.539607: step: 2342/1541, loss: 0.8003097772598267 2023-01-23 11:34:41.465257: step: 2344/1541, loss: 0.1808890551328659 2023-01-23 11:34:42.384402: step: 2346/1541, loss: 0.14432695508003235 2023-01-23 11:34:43.270687: step: 2348/1541, loss: 0.20267076790332794 2023-01-23 11:34:44.251380: step: 2350/1541, loss: 0.16510258615016937 2023-01-23 11:34:45.134972: step: 2352/1541, loss: 0.16507785022258759 2023-01-23 11:34:46.031991: step: 2354/1541, loss: 0.2354653775691986 2023-01-23 11:34:46.916638: step: 2356/1541, loss: 0.554267942905426 2023-01-23 11:34:47.844290: step: 2358/1541, loss: 0.18370243906974792 2023-01-23 11:34:48.756056: step: 2360/1541, loss: 0.11211465299129486 2023-01-23 11:34:49.631127: step: 2362/1541, loss: 0.26441943645477295 2023-01-23 11:34:50.508563: step: 2364/1541, loss: 0.10437969863414764 2023-01-23 11:34:51.377465: step: 2366/1541, loss: 0.18449455499649048 2023-01-23 11:34:52.233190: step: 2368/1541, loss: 0.05292963236570358 2023-01-23 11:34:53.107103: step: 2370/1541, loss: 0.08934586495161057 2023-01-23 11:34:54.005119: step: 2372/1541, loss: 0.2976383566856384 2023-01-23 11:34:54.899910: step: 2374/1541, loss: 0.26882055401802063 2023-01-23 11:34:55.750169: step: 2376/1541, loss: 0.09377484023571014 2023-01-23 11:34:56.666361: step: 2378/1541, loss: 0.20243395864963531 2023-01-23 11:34:57.568698: step: 2380/1541, loss: 0.028447236865758896 2023-01-23 11:34:58.514466: step: 2382/1541, loss: 0.0501677505671978 2023-01-23 11:34:59.435503: step: 2384/1541, loss: 0.024501368403434753 2023-01-23 11:35:00.333581: step: 2386/1541, loss: 0.09975787252187729 2023-01-23 11:35:01.220446: step: 2388/1541, loss: 0.027333011850714684 2023-01-23 11:35:02.089109: step: 2390/1541, loss: 0.1530616730451584 2023-01-23 11:35:03.004509: step: 2392/1541, loss: 0.06375551968812943 2023-01-23 11:35:03.919631: step: 2394/1541, loss: 0.0575222373008728 2023-01-23 11:35:04.807985: step: 2396/1541, loss: 0.21238283812999725 2023-01-23 11:35:05.742522: step: 2398/1541, loss: 0.13759538531303406 2023-01-23 11:35:06.668881: step: 2400/1541, loss: 0.1392236202955246 2023-01-23 11:35:07.561079: step: 2402/1541, loss: 0.10084495693445206 2023-01-23 11:35:08.457345: step: 2404/1541, loss: 0.06508480757474899 2023-01-23 11:35:09.354699: step: 2406/1541, loss: 0.3789641261100769 2023-01-23 11:35:10.272684: step: 2408/1541, loss: 0.08225540816783905 2023-01-23 11:35:11.196739: step: 2410/1541, loss: 0.23118413984775543 2023-01-23 11:35:12.118848: step: 2412/1541, loss: 0.08769562840461731 2023-01-23 11:35:12.967970: step: 2414/1541, loss: 0.027706686407327652 2023-01-23 11:35:13.883114: step: 2416/1541, loss: 0.03082558698952198 2023-01-23 11:35:14.756365: step: 2418/1541, loss: 0.014417937025427818 2023-01-23 11:35:15.598450: step: 2420/1541, loss: 0.07800771296024323 2023-01-23 11:35:16.481735: step: 2422/1541, loss: 0.0064838360995054245 2023-01-23 11:35:17.384917: step: 2424/1541, loss: 0.593387246131897 2023-01-23 11:35:18.299277: step: 2426/1541, loss: 0.07915643602609634 2023-01-23 11:35:19.206267: step: 2428/1541, loss: 0.07709163427352905 2023-01-23 11:35:20.104425: step: 2430/1541, loss: 0.31914088129997253 2023-01-23 11:35:20.993116: step: 2432/1541, loss: 0.0990348756313324 2023-01-23 11:35:21.905052: step: 2434/1541, loss: 0.013808275572955608 2023-01-23 11:35:22.867867: step: 2436/1541, loss: 0.01257202960550785 2023-01-23 11:35:23.788082: step: 2438/1541, loss: 0.014215352013707161 2023-01-23 11:35:24.766267: step: 2440/1541, loss: 0.018354715779423714 2023-01-23 11:35:25.660922: step: 2442/1541, loss: 0.023334544152021408 2023-01-23 11:35:26.558389: step: 2444/1541, loss: 0.03761690855026245 2023-01-23 11:35:27.455811: step: 2446/1541, loss: 0.05698951706290245 2023-01-23 11:35:28.310980: step: 2448/1541, loss: 0.04032937437295914 2023-01-23 11:35:29.210811: step: 2450/1541, loss: 0.012685036286711693 2023-01-23 11:35:30.129626: step: 2452/1541, loss: 0.030804120004177094 2023-01-23 11:35:31.020150: step: 2454/1541, loss: 0.11589612066745758 2023-01-23 11:35:31.953850: step: 2456/1541, loss: 0.2127329558134079 2023-01-23 11:35:32.809646: step: 2458/1541, loss: 0.14107325673103333 2023-01-23 11:35:33.711927: step: 2460/1541, loss: 0.005763455759733915 2023-01-23 11:35:34.601754: step: 2462/1541, loss: 0.07113393396139145 2023-01-23 11:35:35.515748: step: 2464/1541, loss: 0.11259970813989639 2023-01-23 11:35:36.398347: step: 2466/1541, loss: 0.4445929229259491 2023-01-23 11:35:37.287823: step: 2468/1541, loss: 0.20641544461250305 2023-01-23 11:35:38.200862: step: 2470/1541, loss: 0.0029468550346791744 2023-01-23 11:35:39.079142: step: 2472/1541, loss: 0.14900422096252441 2023-01-23 11:35:39.935126: step: 2474/1541, loss: 0.03562556952238083 2023-01-23 11:35:40.811440: step: 2476/1541, loss: 0.2511006295681 2023-01-23 11:35:41.697144: step: 2478/1541, loss: 0.20551680028438568 2023-01-23 11:35:42.576339: step: 2480/1541, loss: 0.34231501817703247 2023-01-23 11:35:43.526582: step: 2482/1541, loss: 0.17160896956920624 2023-01-23 11:35:44.384390: step: 2484/1541, loss: 0.13225167989730835 2023-01-23 11:35:45.245309: step: 2486/1541, loss: 0.36834511160850525 2023-01-23 11:35:46.156828: step: 2488/1541, loss: 0.26109156012535095 2023-01-23 11:35:47.053574: step: 2490/1541, loss: 1.7479276657104492 2023-01-23 11:35:47.930257: step: 2492/1541, loss: 1.6661555767059326 2023-01-23 11:35:48.798746: step: 2494/1541, loss: 0.04459596797823906 2023-01-23 11:35:49.692043: step: 2496/1541, loss: 0.12285660952329636 2023-01-23 11:35:50.603788: step: 2498/1541, loss: 0.49466845393180847 2023-01-23 11:35:51.527509: step: 2500/1541, loss: 0.25298112630844116 2023-01-23 11:35:52.497099: step: 2502/1541, loss: 0.27812057733535767 2023-01-23 11:35:53.410873: step: 2504/1541, loss: 0.04701291769742966 2023-01-23 11:35:54.302878: step: 2506/1541, loss: 0.3227766752243042 2023-01-23 11:35:55.174365: step: 2508/1541, loss: 0.18617166578769684 2023-01-23 11:35:56.057481: step: 2510/1541, loss: 0.1594512015581131 2023-01-23 11:35:56.921171: step: 2512/1541, loss: 0.02507997862994671 2023-01-23 11:35:57.833059: step: 2514/1541, loss: 2.1666836738586426 2023-01-23 11:35:58.749195: step: 2516/1541, loss: 0.10929574072360992 2023-01-23 11:35:59.638374: step: 2518/1541, loss: 0.15298473834991455 2023-01-23 11:36:00.513917: step: 2520/1541, loss: 0.10568028688430786 2023-01-23 11:36:01.375253: step: 2522/1541, loss: 0.02260458469390869 2023-01-23 11:36:02.273307: step: 2524/1541, loss: 0.20723497867584229 2023-01-23 11:36:03.226684: step: 2526/1541, loss: 0.2356993556022644 2023-01-23 11:36:04.119635: step: 2528/1541, loss: 0.008236709982156754 2023-01-23 11:36:05.012806: step: 2530/1541, loss: 0.082649365067482 2023-01-23 11:36:05.946293: step: 2532/1541, loss: 0.30110353231430054 2023-01-23 11:36:06.843944: step: 2534/1541, loss: 0.8072428107261658 2023-01-23 11:36:07.722140: step: 2536/1541, loss: 0.10913927853107452 2023-01-23 11:36:08.586121: step: 2538/1541, loss: 0.025437671691179276 2023-01-23 11:36:09.484551: step: 2540/1541, loss: 0.08997747302055359 2023-01-23 11:36:10.362340: step: 2542/1541, loss: 0.2438311129808426 2023-01-23 11:36:11.250677: step: 2544/1541, loss: 0.031549304723739624 2023-01-23 11:36:12.157532: step: 2546/1541, loss: 0.007832586765289307 2023-01-23 11:36:13.027543: step: 2548/1541, loss: 0.005012970883399248 2023-01-23 11:36:13.877116: step: 2550/1541, loss: 0.03613528236746788 2023-01-23 11:36:14.810045: step: 2552/1541, loss: 0.1991262435913086 2023-01-23 11:36:15.688702: step: 2554/1541, loss: 0.009424854069948196 2023-01-23 11:36:16.602822: step: 2556/1541, loss: 0.5824763178825378 2023-01-23 11:36:17.510519: step: 2558/1541, loss: 0.357384592294693 2023-01-23 11:36:18.412102: step: 2560/1541, loss: 0.875633716583252 2023-01-23 11:36:19.307927: step: 2562/1541, loss: 0.1477471888065338 2023-01-23 11:36:20.197838: step: 2564/1541, loss: 0.3170650601387024 2023-01-23 11:36:21.115394: step: 2566/1541, loss: 0.5567883849143982 2023-01-23 11:36:22.041445: step: 2568/1541, loss: 0.07891629636287689 2023-01-23 11:36:22.940199: step: 2570/1541, loss: 0.010021496564149857 2023-01-23 11:36:23.849543: step: 2572/1541, loss: 0.025620579719543457 2023-01-23 11:36:24.806084: step: 2574/1541, loss: 0.03829554468393326 2023-01-23 11:36:25.698645: step: 2576/1541, loss: 0.09235972166061401 2023-01-23 11:36:26.626201: step: 2578/1541, loss: 0.22861915826797485 2023-01-23 11:36:27.542162: step: 2580/1541, loss: 0.23739416897296906 2023-01-23 11:36:28.412240: step: 2582/1541, loss: 0.3262745440006256 2023-01-23 11:36:29.290574: step: 2584/1541, loss: 0.09362909197807312 2023-01-23 11:36:30.200264: step: 2586/1541, loss: 0.005042907316237688 2023-01-23 11:36:31.131591: step: 2588/1541, loss: 0.11365681886672974 2023-01-23 11:36:31.978582: step: 2590/1541, loss: 0.059092335402965546 2023-01-23 11:36:32.836757: step: 2592/1541, loss: 0.035894572734832764 2023-01-23 11:36:33.732119: step: 2594/1541, loss: 0.03661337494850159 2023-01-23 11:36:34.630615: step: 2596/1541, loss: 0.8679624795913696 2023-01-23 11:36:35.505424: step: 2598/1541, loss: 0.24779270589351654 2023-01-23 11:36:36.373762: step: 2600/1541, loss: 0.06456704437732697 2023-01-23 11:36:37.266638: step: 2602/1541, loss: 0.07946740090847015 2023-01-23 11:36:38.130787: step: 2604/1541, loss: 0.01567310281097889 2023-01-23 11:36:39.034568: step: 2606/1541, loss: 0.8372790217399597 2023-01-23 11:36:39.932805: step: 2608/1541, loss: 0.23960334062576294 2023-01-23 11:36:40.804800: step: 2610/1541, loss: 0.015771348029375076 2023-01-23 11:36:41.747854: step: 2612/1541, loss: 0.04094228520989418 2023-01-23 11:36:42.697159: step: 2614/1541, loss: 0.9485170841217041 2023-01-23 11:36:43.581730: step: 2616/1541, loss: 0.05198204144835472 2023-01-23 11:36:44.451136: step: 2618/1541, loss: 0.02491442672908306 2023-01-23 11:36:45.350104: step: 2620/1541, loss: 0.16426976025104523 2023-01-23 11:36:46.229828: step: 2622/1541, loss: 0.07302362471818924 2023-01-23 11:36:47.124273: step: 2624/1541, loss: 0.09068763256072998 2023-01-23 11:36:48.005233: step: 2626/1541, loss: 0.19919633865356445 2023-01-23 11:36:48.937420: step: 2628/1541, loss: 0.23910288512706757 2023-01-23 11:36:49.803434: step: 2630/1541, loss: 0.04746697470545769 2023-01-23 11:36:50.668173: step: 2632/1541, loss: 0.02383461967110634 2023-01-23 11:36:51.563511: step: 2634/1541, loss: 0.21739253401756287 2023-01-23 11:36:52.435245: step: 2636/1541, loss: 0.046969935297966 2023-01-23 11:36:53.345602: step: 2638/1541, loss: 0.13440044224262238 2023-01-23 11:36:54.234725: step: 2640/1541, loss: 0.09205654263496399 2023-01-23 11:36:55.140157: step: 2642/1541, loss: 0.042031556367874146 2023-01-23 11:36:56.013137: step: 2644/1541, loss: 0.07394807040691376 2023-01-23 11:36:56.897637: step: 2646/1541, loss: 0.03434804454445839 2023-01-23 11:36:57.851455: step: 2648/1541, loss: 0.04499046131968498 2023-01-23 11:36:58.714073: step: 2650/1541, loss: 0.46141356229782104 2023-01-23 11:36:59.581717: step: 2652/1541, loss: 0.02218281291425228 2023-01-23 11:37:00.452120: step: 2654/1541, loss: 0.22419115900993347 2023-01-23 11:37:01.327631: step: 2656/1541, loss: 0.05844039097428322 2023-01-23 11:37:02.243704: step: 2658/1541, loss: 0.16371183097362518 2023-01-23 11:37:03.161578: step: 2660/1541, loss: 0.22103102505207062 2023-01-23 11:37:04.117695: step: 2662/1541, loss: 0.21173161268234253 2023-01-23 11:37:05.039812: step: 2664/1541, loss: 0.1246035248041153 2023-01-23 11:37:05.957510: step: 2666/1541, loss: 0.06233112886548042 2023-01-23 11:37:06.828409: step: 2668/1541, loss: 0.02296309731900692 2023-01-23 11:37:07.718071: step: 2670/1541, loss: 0.014064552262425423 2023-01-23 11:37:08.598875: step: 2672/1541, loss: 0.042371563613414764 2023-01-23 11:37:09.468189: step: 2674/1541, loss: 3.2046656608581543 2023-01-23 11:37:10.424028: step: 2676/1541, loss: 0.6303646564483643 2023-01-23 11:37:11.273458: step: 2678/1541, loss: 0.14209412038326263 2023-01-23 11:37:12.237378: step: 2680/1541, loss: 0.03660982847213745 2023-01-23 11:37:13.153575: step: 2682/1541, loss: 0.04457782581448555 2023-01-23 11:37:14.042851: step: 2684/1541, loss: 0.5181131362915039 2023-01-23 11:37:14.994900: step: 2686/1541, loss: 0.070652075111866 2023-01-23 11:37:15.951901: step: 2688/1541, loss: 0.020981213077902794 2023-01-23 11:37:16.852969: step: 2690/1541, loss: 0.00637869443744421 2023-01-23 11:37:17.868351: step: 2692/1541, loss: 0.08341588824987411 2023-01-23 11:37:18.895208: step: 2694/1541, loss: 0.16866958141326904 2023-01-23 11:37:19.747216: step: 2696/1541, loss: 0.05216645076870918 2023-01-23 11:37:20.661255: step: 2698/1541, loss: 0.48024240136146545 2023-01-23 11:37:21.538665: step: 2700/1541, loss: 0.15890873968601227 2023-01-23 11:37:22.422051: step: 2702/1541, loss: 0.5056071281433105 2023-01-23 11:37:23.363626: step: 2704/1541, loss: 0.028828760609030724 2023-01-23 11:37:24.291528: step: 2706/1541, loss: 0.14611409604549408 2023-01-23 11:37:25.166537: step: 2708/1541, loss: 0.02882513217628002 2023-01-23 11:37:26.048675: step: 2710/1541, loss: 0.020809605717658997 2023-01-23 11:37:26.967119: step: 2712/1541, loss: 0.01762544922530651 2023-01-23 11:37:27.859568: step: 2714/1541, loss: 0.06130823493003845 2023-01-23 11:37:28.756956: step: 2716/1541, loss: 0.1186509057879448 2023-01-23 11:37:29.680223: step: 2718/1541, loss: 0.2795220613479614 2023-01-23 11:37:30.661655: step: 2720/1541, loss: 0.2672500014305115 2023-01-23 11:37:31.555639: step: 2722/1541, loss: 0.002243018476292491 2023-01-23 11:37:32.421716: step: 2724/1541, loss: 0.06315432488918304 2023-01-23 11:37:33.283159: step: 2726/1541, loss: 0.051169928163290024 2023-01-23 11:37:34.178580: step: 2728/1541, loss: 0.0706888809800148 2023-01-23 11:37:35.030679: step: 2730/1541, loss: 0.067075714468956 2023-01-23 11:37:35.965288: step: 2732/1541, loss: 0.0883985087275505 2023-01-23 11:37:36.899659: step: 2734/1541, loss: 0.024258749559521675 2023-01-23 11:37:37.790921: step: 2736/1541, loss: 0.8739057779312134 2023-01-23 11:37:38.836153: step: 2738/1541, loss: 0.1363229602575302 2023-01-23 11:37:39.708148: step: 2740/1541, loss: 0.09326792508363724 2023-01-23 11:37:40.609575: step: 2742/1541, loss: 0.09161478281021118 2023-01-23 11:37:41.517317: step: 2744/1541, loss: 0.2011553794145584 2023-01-23 11:37:42.425373: step: 2746/1541, loss: 0.06128900870680809 2023-01-23 11:37:43.352635: step: 2748/1541, loss: 0.31269171833992004 2023-01-23 11:37:44.222851: step: 2750/1541, loss: 0.7594977021217346 2023-01-23 11:37:45.122020: step: 2752/1541, loss: 0.13029560446739197 2023-01-23 11:37:46.004248: step: 2754/1541, loss: 0.020081816241145134 2023-01-23 11:37:46.862974: step: 2756/1541, loss: 0.05450548604130745 2023-01-23 11:37:47.773287: step: 2758/1541, loss: 0.4238777160644531 2023-01-23 11:37:48.668190: step: 2760/1541, loss: 0.17526313662528992 2023-01-23 11:37:49.573950: step: 2762/1541, loss: 0.19117148220539093 2023-01-23 11:37:50.521055: step: 2764/1541, loss: 0.956680953502655 2023-01-23 11:37:51.432591: step: 2766/1541, loss: 0.06216087564826012 2023-01-23 11:37:52.329444: step: 2768/1541, loss: 0.004775871057063341 2023-01-23 11:37:53.208654: step: 2770/1541, loss: 0.07335598766803741 2023-01-23 11:37:54.106260: step: 2772/1541, loss: 0.12434431910514832 2023-01-23 11:37:55.034398: step: 2774/1541, loss: 0.06540120393037796 2023-01-23 11:37:55.944122: step: 2776/1541, loss: 0.033686839044094086 2023-01-23 11:37:56.874401: step: 2778/1541, loss: 0.030248166993260384 2023-01-23 11:37:57.755295: step: 2780/1541, loss: 0.37871047854423523 2023-01-23 11:37:58.645779: step: 2782/1541, loss: 0.28186124563217163 2023-01-23 11:37:59.556967: step: 2784/1541, loss: 0.14599353075027466 2023-01-23 11:38:00.474985: step: 2786/1541, loss: 0.014561823569238186 2023-01-23 11:38:01.374695: step: 2788/1541, loss: 0.09840436279773712 2023-01-23 11:38:02.271310: step: 2790/1541, loss: 0.053766850382089615 2023-01-23 11:38:03.171173: step: 2792/1541, loss: 0.09633813053369522 2023-01-23 11:38:04.046391: step: 2794/1541, loss: 1.582826018333435 2023-01-23 11:38:04.929034: step: 2796/1541, loss: 0.04739665985107422 2023-01-23 11:38:05.819571: step: 2798/1541, loss: 0.16617076098918915 2023-01-23 11:38:06.735189: step: 2800/1541, loss: 0.03169289976358414 2023-01-23 11:38:07.621580: step: 2802/1541, loss: 0.0595494844019413 2023-01-23 11:38:08.519549: step: 2804/1541, loss: 0.0349736250936985 2023-01-23 11:38:09.399380: step: 2806/1541, loss: 0.2422534078359604 2023-01-23 11:38:10.307166: step: 2808/1541, loss: 0.02884247526526451 2023-01-23 11:38:11.232361: step: 2810/1541, loss: 0.06311638653278351 2023-01-23 11:38:12.189366: step: 2812/1541, loss: 0.030759328976273537 2023-01-23 11:38:13.078658: step: 2814/1541, loss: 0.0363200381398201 2023-01-23 11:38:13.929853: step: 2816/1541, loss: 0.03783240541815758 2023-01-23 11:38:14.792575: step: 2818/1541, loss: 0.08221829682588577 2023-01-23 11:38:15.661102: step: 2820/1541, loss: 0.01565798744559288 2023-01-23 11:38:16.614674: step: 2822/1541, loss: 0.24233803153038025 2023-01-23 11:38:17.529440: step: 2824/1541, loss: 0.2318052053451538 2023-01-23 11:38:18.487577: step: 2826/1541, loss: 0.7682631611824036 2023-01-23 11:38:19.388718: step: 2828/1541, loss: 0.05224963650107384 2023-01-23 11:38:20.268159: step: 2830/1541, loss: 0.3924812078475952 2023-01-23 11:38:21.137182: step: 2832/1541, loss: 0.09119682759046555 2023-01-23 11:38:22.001842: step: 2834/1541, loss: 0.08706507086753845 2023-01-23 11:38:22.926967: step: 2836/1541, loss: 0.17290732264518738 2023-01-23 11:38:23.804687: step: 2838/1541, loss: 0.021705418825149536 2023-01-23 11:38:24.698620: step: 2840/1541, loss: 0.021548930555582047 2023-01-23 11:38:25.587076: step: 2842/1541, loss: 0.9107738137245178 2023-01-23 11:38:26.470742: step: 2844/1541, loss: 0.03937433287501335 2023-01-23 11:38:27.357799: step: 2846/1541, loss: 0.04020410776138306 2023-01-23 11:38:28.257944: step: 2848/1541, loss: 2.5892891883850098 2023-01-23 11:38:29.181198: step: 2850/1541, loss: 0.541633665561676 2023-01-23 11:38:30.073533: step: 2852/1541, loss: 0.04344978928565979 2023-01-23 11:38:31.037797: step: 2854/1541, loss: 0.033260051161050797 2023-01-23 11:38:32.010555: step: 2856/1541, loss: 0.3403422236442566 2023-01-23 11:38:32.947943: step: 2858/1541, loss: 0.017294079065322876 2023-01-23 11:38:33.808344: step: 2860/1541, loss: 0.06775711476802826 2023-01-23 11:38:34.677060: step: 2862/1541, loss: 0.44008004665374756 2023-01-23 11:38:35.562572: step: 2864/1541, loss: 0.162650927901268 2023-01-23 11:38:36.447341: step: 2866/1541, loss: 0.014616139233112335 2023-01-23 11:38:37.325848: step: 2868/1541, loss: 0.10608730465173721 2023-01-23 11:38:38.220682: step: 2870/1541, loss: 0.06422507017850876 2023-01-23 11:38:39.146407: step: 2872/1541, loss: 0.0916912853717804 2023-01-23 11:38:40.030933: step: 2874/1541, loss: 0.3039783239364624 2023-01-23 11:38:40.926904: step: 2876/1541, loss: 0.031346697360277176 2023-01-23 11:38:41.827982: step: 2878/1541, loss: 0.010426538065075874 2023-01-23 11:38:42.705065: step: 2880/1541, loss: 0.4915004372596741 2023-01-23 11:38:43.616200: step: 2882/1541, loss: 0.03637097403407097 2023-01-23 11:38:44.546632: step: 2884/1541, loss: 0.1512916088104248 2023-01-23 11:38:45.440633: step: 2886/1541, loss: 0.08174282312393188 2023-01-23 11:38:46.330602: step: 2888/1541, loss: 0.09666713327169418 2023-01-23 11:38:47.234748: step: 2890/1541, loss: 1.0290430784225464 2023-01-23 11:38:48.124346: step: 2892/1541, loss: 0.12275087088346481 2023-01-23 11:38:49.017607: step: 2894/1541, loss: 0.07539547234773636 2023-01-23 11:38:49.925260: step: 2896/1541, loss: 0.2269413322210312 2023-01-23 11:38:50.803903: step: 2898/1541, loss: 0.0077399625442922115 2023-01-23 11:38:51.695875: step: 2900/1541, loss: 0.152678444981575 2023-01-23 11:38:52.621739: step: 2902/1541, loss: 0.22772859036922455 2023-01-23 11:38:53.603203: step: 2904/1541, loss: 0.04269054904580116 2023-01-23 11:38:54.500969: step: 2906/1541, loss: 0.006966077256947756 2023-01-23 11:38:55.361632: step: 2908/1541, loss: 0.22749873995780945 2023-01-23 11:38:56.242111: step: 2910/1541, loss: 0.12996362149715424 2023-01-23 11:38:57.152956: step: 2912/1541, loss: 0.05045245960354805 2023-01-23 11:38:58.042177: step: 2914/1541, loss: 0.2549847662448883 2023-01-23 11:38:58.953019: step: 2916/1541, loss: 0.09229620546102524 2023-01-23 11:38:59.876881: step: 2918/1541, loss: 0.03645305335521698 2023-01-23 11:39:00.778981: step: 2920/1541, loss: 0.18748360872268677 2023-01-23 11:39:01.681154: step: 2922/1541, loss: 0.007838007993996143 2023-01-23 11:39:02.607409: step: 2924/1541, loss: 0.11590820550918579 2023-01-23 11:39:03.487187: step: 2926/1541, loss: 0.07696296274662018 2023-01-23 11:39:04.381646: step: 2928/1541, loss: 0.014910401776432991 2023-01-23 11:39:05.313551: step: 2930/1541, loss: 0.0580567941069603 2023-01-23 11:39:06.231046: step: 2932/1541, loss: 0.24184468388557434 2023-01-23 11:39:07.128467: step: 2934/1541, loss: 0.05962969362735748 2023-01-23 11:39:08.049082: step: 2936/1541, loss: 0.4624490439891815 2023-01-23 11:39:08.930252: step: 2938/1541, loss: 0.1368478387594223 2023-01-23 11:39:09.836754: step: 2940/1541, loss: 0.11502488702535629 2023-01-23 11:39:10.776114: step: 2942/1541, loss: 0.0038996837101876736 2023-01-23 11:39:11.641122: step: 2944/1541, loss: 0.07323730736970901 2023-01-23 11:39:12.505918: step: 2946/1541, loss: 0.045724354684352875 2023-01-23 11:39:13.475890: step: 2948/1541, loss: 0.3469606637954712 2023-01-23 11:39:14.358011: step: 2950/1541, loss: 0.0019869632087647915 2023-01-23 11:39:15.292970: step: 2952/1541, loss: 0.011921189725399017 2023-01-23 11:39:16.202863: step: 2954/1541, loss: 0.020701704546809196 2023-01-23 11:39:17.097288: step: 2956/1541, loss: 0.39343124628067017 2023-01-23 11:39:18.102555: step: 2958/1541, loss: 0.28957951068878174 2023-01-23 11:39:18.987813: step: 2960/1541, loss: 1.2310994863510132 2023-01-23 11:39:19.921590: step: 2962/1541, loss: 1.4630564451217651 2023-01-23 11:39:20.884948: step: 2964/1541, loss: 0.11402853578329086 2023-01-23 11:39:21.755929: step: 2966/1541, loss: 0.0407387800514698 2023-01-23 11:39:22.634223: step: 2968/1541, loss: 0.24088704586029053 2023-01-23 11:39:23.550563: step: 2970/1541, loss: 0.05485342815518379 2023-01-23 11:39:24.525597: step: 2972/1541, loss: 0.2529009282588959 2023-01-23 11:39:25.441842: step: 2974/1541, loss: 0.013575393706560135 2023-01-23 11:39:26.311660: step: 2976/1541, loss: 0.14708268642425537 2023-01-23 11:39:27.231254: step: 2978/1541, loss: 0.512301504611969 2023-01-23 11:39:28.125397: step: 2980/1541, loss: 0.25191330909729004 2023-01-23 11:39:29.027196: step: 2982/1541, loss: 1.0555099248886108 2023-01-23 11:39:29.904073: step: 2984/1541, loss: 0.08422791957855225 2023-01-23 11:39:30.798644: step: 2986/1541, loss: 0.1564090996980667 2023-01-23 11:39:31.693570: step: 2988/1541, loss: 1.2763751745224 2023-01-23 11:39:32.647448: step: 2990/1541, loss: 0.18531917035579681 2023-01-23 11:39:33.489327: step: 2992/1541, loss: 0.05145352706313133 2023-01-23 11:39:34.419247: step: 2994/1541, loss: 0.04258367046713829 2023-01-23 11:39:35.347202: step: 2996/1541, loss: 0.09384145587682724 2023-01-23 11:39:36.282800: step: 2998/1541, loss: 0.058732300996780396 2023-01-23 11:39:37.188584: step: 3000/1541, loss: 0.043922360986471176 2023-01-23 11:39:38.106076: step: 3002/1541, loss: 0.29617393016815186 2023-01-23 11:39:39.057177: step: 3004/1541, loss: 0.020485498011112213 2023-01-23 11:39:39.960989: step: 3006/1541, loss: 0.37164306640625 2023-01-23 11:39:40.837115: step: 3008/1541, loss: 0.20892713963985443 2023-01-23 11:39:41.731042: step: 3010/1541, loss: 0.05579978600144386 2023-01-23 11:39:42.626362: step: 3012/1541, loss: 0.11677248775959015 2023-01-23 11:39:43.516512: step: 3014/1541, loss: 0.014915669336915016 2023-01-23 11:39:44.377542: step: 3016/1541, loss: 0.6218632459640503 2023-01-23 11:39:45.317595: step: 3018/1541, loss: 0.005356084555387497 2023-01-23 11:39:46.212494: step: 3020/1541, loss: 0.19652365148067474 2023-01-23 11:39:47.095172: step: 3022/1541, loss: 0.35351139307022095 2023-01-23 11:39:48.028424: step: 3024/1541, loss: 0.28678739070892334 2023-01-23 11:39:48.970004: step: 3026/1541, loss: 0.061893634498119354 2023-01-23 11:39:49.852317: step: 3028/1541, loss: 0.04391692578792572 2023-01-23 11:39:50.707966: step: 3030/1541, loss: 0.26246875524520874 2023-01-23 11:39:51.606841: step: 3032/1541, loss: 0.06453637778759003 2023-01-23 11:39:52.477979: step: 3034/1541, loss: 0.3579787015914917 2023-01-23 11:39:53.375486: step: 3036/1541, loss: 0.08787909150123596 2023-01-23 11:39:54.404153: step: 3038/1541, loss: 0.03778252750635147 2023-01-23 11:39:55.300063: step: 3040/1541, loss: 0.028092969208955765 2023-01-23 11:39:56.163437: step: 3042/1541, loss: 0.21057307720184326 2023-01-23 11:39:57.052459: step: 3044/1541, loss: 0.1422271877527237 2023-01-23 11:39:57.966116: step: 3046/1541, loss: 0.08575119823217392 2023-01-23 11:39:58.862809: step: 3048/1541, loss: 0.002431836910545826 2023-01-23 11:39:59.743579: step: 3050/1541, loss: 0.03800661116838455 2023-01-23 11:40:00.686541: step: 3052/1541, loss: 3.125236749649048 2023-01-23 11:40:01.596511: step: 3054/1541, loss: 0.17697960138320923 2023-01-23 11:40:02.498781: step: 3056/1541, loss: 0.19933928549289703 2023-01-23 11:40:03.395009: step: 3058/1541, loss: 0.09943161904811859 2023-01-23 11:40:04.292827: step: 3060/1541, loss: 0.3208312392234802 2023-01-23 11:40:05.186129: step: 3062/1541, loss: 0.029295969754457474 2023-01-23 11:40:06.090689: step: 3064/1541, loss: 0.13748188316822052 2023-01-23 11:40:06.950072: step: 3066/1541, loss: 0.015160530805587769 2023-01-23 11:40:07.849817: step: 3068/1541, loss: 0.12468627840280533 2023-01-23 11:40:08.705982: step: 3070/1541, loss: 0.042517900466918945 2023-01-23 11:40:09.598259: step: 3072/1541, loss: 0.05593306943774223 2023-01-23 11:40:10.463143: step: 3074/1541, loss: 0.12645913660526276 2023-01-23 11:40:11.360706: step: 3076/1541, loss: 0.01770763285458088 2023-01-23 11:40:12.325655: step: 3078/1541, loss: 0.11620675027370453 2023-01-23 11:40:13.209902: step: 3080/1541, loss: 0.15180730819702148 2023-01-23 11:40:14.068454: step: 3082/1541, loss: 0.04691915214061737 2023-01-23 11:40:14.695418: step: 3084/1541, loss: 0.005272303242236376 ================================================== Loss: 0.192 -------------------- Dev Chinese: {'event': {'p': 0.6248506571087217, 'r': 0.6964047936085219, 'f1': 0.6586901763224181}, 'argument': {'p': 0.6575667999224194, 'r': 0.6488944681934193, 'f1': 0.6532018504500723}, 'combined': 0.4302576420470879, 'epoch': 13} Test Chinese: {'event': {'p': 0.6098843322818086, 'r': 0.695026962252846, 'f1': 0.6496779613553626}, 'argument': {'p': 0.7141576827888098, 'r': 0.7024373707080306, 'f1': 0.708249042280353}, 'combined': 0.46013379392058773, 'epoch': 13} Dev Korean: {'event': {'p': 0.5656862745098039, 'r': 0.7683089214380826, 'f1': 0.6516092603049125}, 'argument': {'p': 0.6501253071310233, 'r': 0.6482611485442678, 'f1': 0.6491918896046328}, 'combined': 0.4230194469812232, 'epoch': 13} Test Korean: {'event': {'p': 0.611534795042898, 'r': 0.768723786698622, 'f1': 0.6811786567560393}, 'argument': {'p': 0.7080991336922304, 'r': 0.7143170748029806, 'f1': 0.7111945137136163}, 'combined': 0.4844505235437057, 'epoch': 13} Dev Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.653371185022781, 'r': 0.6528791886485169, 'f1': 0.6531250941809645}, 'combined': 0.4496121337998113, 'epoch': 13} Test Russian: {'event': {'p': 0.6478283621140764, 'r': 0.7417615338526063, 'f1': 0.6916201117318437}, 'argument': {'p': 0.7081306065307763, 'r': 0.708893952495416, 'f1': 0.7085120739072518}, 'combined': 0.49002119971909375, 'epoch': 13} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6373027798647631, 'r': 0.6824215607401447, 'f1': 0.659090909090909}, 'combined': 0.48445998445998434, 'epoch': 13} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5426245210727969, 'r': 0.4087301587301587, 'f1': 0.4662551440329218}, 'combined': 0.271676225656978, 'epoch': 13} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6150902184235518, 'r': 0.6436990657920891, 'f1': 0.6290695415695416}, 'combined': 0.3235214785214786, 'epoch': 13} New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5784313725490197, 'r': 0.7856191744340879, 'f1': 0.6662902315076229}, 'argument': {'p': 0.6601304922457741, 'r': 0.6658171870820622, 'f1': 0.6629616451789225}, 'combined': 0.44172486804693883, 'epoch': 9} Test for Korean: {'event': {'p': 0.6091515729265967, 'r': 0.7657279808268425, 'f1': 0.6785240244226175}, 'argument': {'p': 0.7048764243703549, 'r': 0.7189987288304991, 'f1': 0.71186754255186}, 'combined': 0.48301922982812695, 'epoch': 9} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6246043543144992, 'r': 0.3698315255809535, 'f1': 0.46458175114301603}, 'combined': 0.27801742588086, 'epoch': 9} -------------------- Dev for Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.653371185022781, 'r': 0.6528791886485169, 'f1': 0.6531250941809645}, 'combined': 0.4496121337998113, 'epoch': 13} Test for Russian: {'event': {'p': 0.6478283621140764, 'r': 0.7417615338526063, 'f1': 0.6916201117318437}, 'argument': {'p': 0.7081306065307763, 'r': 0.708893952495416, 'f1': 0.7085120739072518}, 'combined': 0.49002119971909375, 'epoch': 13} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6150902184235518, 'r': 0.6436990657920891, 'f1': 0.6290695415695416}, 'combined': 0.3235214785214786, 'epoch': 13} ****************************** Epoch: 14 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 11:45:46.867738: step: 2/1541, loss: 0.031666576862335205 2023-01-23 11:45:47.809754: step: 4/1541, loss: 0.016255654394626617 2023-01-23 11:45:48.698233: step: 6/1541, loss: 0.06648173928260803 2023-01-23 11:45:49.624538: step: 8/1541, loss: 0.19266971945762634 2023-01-23 11:45:50.494417: step: 10/1541, loss: 0.25800275802612305 2023-01-23 11:45:51.412274: step: 12/1541, loss: 0.00388694042339921 2023-01-23 11:45:52.287775: step: 14/1541, loss: 0.19767048954963684 2023-01-23 11:45:53.174453: step: 16/1541, loss: 0.12251540273427963 2023-01-23 11:45:54.083052: step: 18/1541, loss: 0.1953655630350113 2023-01-23 11:45:54.972312: step: 20/1541, loss: 0.005439768545329571 2023-01-23 11:45:55.862365: step: 22/1541, loss: 0.09752810746431351 2023-01-23 11:45:56.731796: step: 24/1541, loss: 0.18454065918922424 2023-01-23 11:45:57.636153: step: 26/1541, loss: 0.007444415241479874 2023-01-23 11:45:58.535381: step: 28/1541, loss: 0.02540609985589981 2023-01-23 11:45:59.419422: step: 30/1541, loss: 0.031926482915878296 2023-01-23 11:46:00.395129: step: 32/1541, loss: 0.18358325958251953 2023-01-23 11:46:01.265699: step: 34/1541, loss: 0.02085931785404682 2023-01-23 11:46:02.153036: step: 36/1541, loss: 0.0014522369019687176 2023-01-23 11:46:02.994796: step: 38/1541, loss: 0.014670002274215221 2023-01-23 11:46:03.874528: step: 40/1541, loss: 0.1665193885564804 2023-01-23 11:46:04.758240: step: 42/1541, loss: 0.0003275723138358444 2023-01-23 11:46:05.678359: step: 44/1541, loss: 0.1998453587293625 2023-01-23 11:46:06.621214: step: 46/1541, loss: 0.03521949052810669 2023-01-23 11:46:07.540702: step: 48/1541, loss: 0.03600703924894333 2023-01-23 11:46:08.463003: step: 50/1541, loss: 0.011603773571550846 2023-01-23 11:46:09.358521: step: 52/1541, loss: 0.009442606940865517 2023-01-23 11:46:10.255954: step: 54/1541, loss: 0.02562321349978447 2023-01-23 11:46:11.234951: step: 56/1541, loss: 0.32171961665153503 2023-01-23 11:46:12.126033: step: 58/1541, loss: 0.002445206046104431 2023-01-23 11:46:13.013286: step: 60/1541, loss: 0.07601181417703629 2023-01-23 11:46:13.898126: step: 62/1541, loss: 0.06156983599066734 2023-01-23 11:46:14.801056: step: 64/1541, loss: 0.18841920793056488 2023-01-23 11:46:15.682523: step: 66/1541, loss: 0.03826144337654114 2023-01-23 11:46:16.554229: step: 68/1541, loss: 0.24073994159698486 2023-01-23 11:46:17.456315: step: 70/1541, loss: 0.3498936891555786 2023-01-23 11:46:18.318321: step: 72/1541, loss: 0.14090627431869507 2023-01-23 11:46:19.226327: step: 74/1541, loss: 0.006326514296233654 2023-01-23 11:46:20.102100: step: 76/1541, loss: 0.11135128885507584 2023-01-23 11:46:20.982846: step: 78/1541, loss: 0.3254529535770416 2023-01-23 11:46:21.895737: step: 80/1541, loss: 0.9124009609222412 2023-01-23 11:46:22.903136: step: 82/1541, loss: 0.2594361901283264 2023-01-23 11:46:23.878578: step: 84/1541, loss: 0.1341596245765686 2023-01-23 11:46:24.753177: step: 86/1541, loss: 0.13584540784358978 2023-01-23 11:46:25.649669: step: 88/1541, loss: 0.09819640964269638 2023-01-23 11:46:26.530661: step: 90/1541, loss: 0.03772716596722603 2023-01-23 11:46:27.448390: step: 92/1541, loss: 0.06890884786844254 2023-01-23 11:46:28.290224: step: 94/1541, loss: 0.0492500439286232 2023-01-23 11:46:29.185953: step: 96/1541, loss: 0.04094153642654419 2023-01-23 11:46:30.062391: step: 98/1541, loss: 0.13687019050121307 2023-01-23 11:46:30.954609: step: 100/1541, loss: 0.009014492854475975 2023-01-23 11:46:31.817623: step: 102/1541, loss: 0.3313974440097809 2023-01-23 11:46:32.661478: step: 104/1541, loss: 0.007750899065285921 2023-01-23 11:46:33.602770: step: 106/1541, loss: 0.026899896562099457 2023-01-23 11:46:34.500028: step: 108/1541, loss: 0.05701465904712677 2023-01-23 11:46:35.392515: step: 110/1541, loss: 0.09451949596405029 2023-01-23 11:46:36.275603: step: 112/1541, loss: 0.09359536319971085 2023-01-23 11:46:37.189435: step: 114/1541, loss: 0.05252125859260559 2023-01-23 11:46:38.083858: step: 116/1541, loss: 0.04312901571393013 2023-01-23 11:46:38.997622: step: 118/1541, loss: 0.04828190058469772 2023-01-23 11:46:39.864270: step: 120/1541, loss: 0.14849728345870972 2023-01-23 11:46:40.745548: step: 122/1541, loss: 0.1597762405872345 2023-01-23 11:46:41.651447: step: 124/1541, loss: 0.05098623409867287 2023-01-23 11:46:42.592210: step: 126/1541, loss: 0.056048646569252014 2023-01-23 11:46:43.500137: step: 128/1541, loss: 0.017008012160658836 2023-01-23 11:46:44.434334: step: 130/1541, loss: 0.4902208149433136 2023-01-23 11:46:45.311583: step: 132/1541, loss: 0.04459291324019432 2023-01-23 11:46:46.250636: step: 134/1541, loss: 0.13057412207126617 2023-01-23 11:46:47.131179: step: 136/1541, loss: 0.02616041526198387 2023-01-23 11:46:48.042934: step: 138/1541, loss: 0.13026143610477448 2023-01-23 11:46:48.939951: step: 140/1541, loss: 0.2368236929178238 2023-01-23 11:46:49.811277: step: 142/1541, loss: 0.18800736963748932 2023-01-23 11:46:50.720542: step: 144/1541, loss: 0.14733976125717163 2023-01-23 11:46:51.633908: step: 146/1541, loss: 0.1808503121137619 2023-01-23 11:46:52.525346: step: 148/1541, loss: 0.011121666990220547 2023-01-23 11:46:53.439018: step: 150/1541, loss: 0.07193347811698914 2023-01-23 11:46:54.334465: step: 152/1541, loss: 0.0026456962805241346 2023-01-23 11:46:55.207218: step: 154/1541, loss: 0.11064060777425766 2023-01-23 11:46:56.126529: step: 156/1541, loss: 0.04986761510372162 2023-01-23 11:46:57.004562: step: 158/1541, loss: 0.27990439534187317 2023-01-23 11:46:57.919469: step: 160/1541, loss: 0.07564704120159149 2023-01-23 11:46:58.818120: step: 162/1541, loss: 0.3288930356502533 2023-01-23 11:46:59.747069: step: 164/1541, loss: 0.07648896425962448 2023-01-23 11:47:00.617264: step: 166/1541, loss: 0.0021344139240682125 2023-01-23 11:47:01.528837: step: 168/1541, loss: 0.7157396674156189 2023-01-23 11:47:02.436321: step: 170/1541, loss: 0.14119942486286163 2023-01-23 11:47:03.358340: step: 172/1541, loss: 0.049292486160993576 2023-01-23 11:47:04.294759: step: 174/1541, loss: 0.11623682081699371 2023-01-23 11:47:05.234789: step: 176/1541, loss: 0.16004811227321625 2023-01-23 11:47:06.120554: step: 178/1541, loss: 0.13750770688056946 2023-01-23 11:47:06.999489: step: 180/1541, loss: 0.01944231428205967 2023-01-23 11:47:07.875785: step: 182/1541, loss: 0.002498314715921879 2023-01-23 11:47:08.738823: step: 184/1541, loss: 0.03642755746841431 2023-01-23 11:47:09.632415: step: 186/1541, loss: 0.0697370395064354 2023-01-23 11:47:10.550871: step: 188/1541, loss: 0.012675119563937187 2023-01-23 11:47:11.433589: step: 190/1541, loss: 0.11202927678823471 2023-01-23 11:47:12.336360: step: 192/1541, loss: 0.012921581044793129 2023-01-23 11:47:13.235158: step: 194/1541, loss: 0.04892027750611305 2023-01-23 11:47:14.112050: step: 196/1541, loss: 0.13986456394195557 2023-01-23 11:47:15.007941: step: 198/1541, loss: 0.7273472547531128 2023-01-23 11:47:15.938238: step: 200/1541, loss: 0.28677377104759216 2023-01-23 11:47:16.827813: step: 202/1541, loss: 0.028930053114891052 2023-01-23 11:47:17.724054: step: 204/1541, loss: 0.10145895928144455 2023-01-23 11:47:18.589321: step: 206/1541, loss: 0.14202746748924255 2023-01-23 11:47:19.536495: step: 208/1541, loss: 0.1488836109638214 2023-01-23 11:47:20.419868: step: 210/1541, loss: 0.011389754712581635 2023-01-23 11:47:21.344566: step: 212/1541, loss: 2.3680646419525146 2023-01-23 11:47:22.271195: step: 214/1541, loss: 0.0935334786772728 2023-01-23 11:47:23.139430: step: 216/1541, loss: 0.29578083753585815 2023-01-23 11:47:24.091661: step: 218/1541, loss: 0.11006835848093033 2023-01-23 11:47:24.995827: step: 220/1541, loss: 0.03162531927227974 2023-01-23 11:47:25.852822: step: 222/1541, loss: 0.009163565002381802 2023-01-23 11:47:26.749613: step: 224/1541, loss: 0.1423085778951645 2023-01-23 11:47:27.610073: step: 226/1541, loss: 0.5642196536064148 2023-01-23 11:47:28.534795: step: 228/1541, loss: 0.27124929428100586 2023-01-23 11:47:29.424742: step: 230/1541, loss: 0.007055922411382198 2023-01-23 11:47:30.408141: step: 232/1541, loss: 0.08001900464296341 2023-01-23 11:47:31.291183: step: 234/1541, loss: 0.006974150892347097 2023-01-23 11:47:32.155509: step: 236/1541, loss: 0.4562218487262726 2023-01-23 11:47:33.024422: step: 238/1541, loss: 0.030893417075276375 2023-01-23 11:47:33.990372: step: 240/1541, loss: 0.14567293226718903 2023-01-23 11:47:34.914522: step: 242/1541, loss: 0.03248570114374161 2023-01-23 11:47:35.788532: step: 244/1541, loss: 0.11170586198568344 2023-01-23 11:47:36.718769: step: 246/1541, loss: 0.4251546561717987 2023-01-23 11:47:37.640202: step: 248/1541, loss: 0.12453651428222656 2023-01-23 11:47:38.499087: step: 250/1541, loss: 0.13984566926956177 2023-01-23 11:47:39.387190: step: 252/1541, loss: 0.031921327114105225 2023-01-23 11:47:40.309706: step: 254/1541, loss: 0.0964447632431984 2023-01-23 11:47:41.221435: step: 256/1541, loss: 0.26348885893821716 2023-01-23 11:47:42.121703: step: 258/1541, loss: 0.0501316599547863 2023-01-23 11:47:43.011216: step: 260/1541, loss: 0.04580335691571236 2023-01-23 11:47:43.896571: step: 262/1541, loss: 0.09637676179409027 2023-01-23 11:47:44.762930: step: 264/1541, loss: 0.014173055067658424 2023-01-23 11:47:45.665243: step: 266/1541, loss: 0.34493470191955566 2023-01-23 11:47:46.589755: step: 268/1541, loss: 0.14511114358901978 2023-01-23 11:47:47.488649: step: 270/1541, loss: 0.035118941217660904 2023-01-23 11:47:48.380466: step: 272/1541, loss: 0.06527074426412582 2023-01-23 11:47:49.267791: step: 274/1541, loss: 0.23188519477844238 2023-01-23 11:47:50.199050: step: 276/1541, loss: 0.10408170521259308 2023-01-23 11:47:51.122586: step: 278/1541, loss: 0.04266594722867012 2023-01-23 11:47:52.036324: step: 280/1541, loss: 0.032151877880096436 2023-01-23 11:47:52.916541: step: 282/1541, loss: 0.40008145570755005 2023-01-23 11:47:53.853734: step: 284/1541, loss: 0.42577677965164185 2023-01-23 11:47:54.866150: step: 286/1541, loss: 0.12911547720432281 2023-01-23 11:47:55.738715: step: 288/1541, loss: 0.12133845686912537 2023-01-23 11:47:56.652463: step: 290/1541, loss: 0.005115224048495293 2023-01-23 11:47:57.562567: step: 292/1541, loss: 0.29866400361061096 2023-01-23 11:47:58.459603: step: 294/1541, loss: 0.29568377137184143 2023-01-23 11:47:59.370803: step: 296/1541, loss: 0.20221209526062012 2023-01-23 11:48:00.246500: step: 298/1541, loss: 0.02694973722100258 2023-01-23 11:48:01.137788: step: 300/1541, loss: 0.04445300251245499 2023-01-23 11:48:02.015483: step: 302/1541, loss: 0.18138520419597626 2023-01-23 11:48:02.915501: step: 304/1541, loss: 0.0038429624401032925 2023-01-23 11:48:03.834525: step: 306/1541, loss: 0.2913567125797272 2023-01-23 11:48:04.731795: step: 308/1541, loss: 0.03434671461582184 2023-01-23 11:48:05.720430: step: 310/1541, loss: 0.06785701960325241 2023-01-23 11:48:06.589207: step: 312/1541, loss: 0.027782440185546875 2023-01-23 11:48:07.589898: step: 314/1541, loss: 0.027747301384806633 2023-01-23 11:48:08.531282: step: 316/1541, loss: 0.038492701947689056 2023-01-23 11:48:09.467201: step: 318/1541, loss: 0.07935434579849243 2023-01-23 11:48:10.387431: step: 320/1541, loss: 0.3921211063861847 2023-01-23 11:48:11.265589: step: 322/1541, loss: 0.09427058696746826 2023-01-23 11:48:12.144460: step: 324/1541, loss: 0.23531123995780945 2023-01-23 11:48:13.029840: step: 326/1541, loss: 0.07238192111253738 2023-01-23 11:48:13.942043: step: 328/1541, loss: 0.04417813569307327 2023-01-23 11:48:14.806200: step: 330/1541, loss: 0.12358657270669937 2023-01-23 11:48:15.741555: step: 332/1541, loss: 0.35113322734832764 2023-01-23 11:48:16.594921: step: 334/1541, loss: 0.013845749199390411 2023-01-23 11:48:17.446238: step: 336/1541, loss: 0.03381260856986046 2023-01-23 11:48:18.311951: step: 338/1541, loss: 0.007191651500761509 2023-01-23 11:48:19.206795: step: 340/1541, loss: 0.023450391367077827 2023-01-23 11:48:20.084464: step: 342/1541, loss: 0.09870918095111847 2023-01-23 11:48:21.048409: step: 344/1541, loss: 0.4870055913925171 2023-01-23 11:48:21.938981: step: 346/1541, loss: 0.0019569697324186563 2023-01-23 11:48:22.805209: step: 348/1541, loss: 0.016007188707590103 2023-01-23 11:48:23.697337: step: 350/1541, loss: 0.013926493003964424 2023-01-23 11:48:24.609327: step: 352/1541, loss: 0.1526831090450287 2023-01-23 11:48:25.448320: step: 354/1541, loss: 0.0017204543109983206 2023-01-23 11:48:26.340669: step: 356/1541, loss: 0.15991315245628357 2023-01-23 11:48:27.236589: step: 358/1541, loss: 0.03175666928291321 2023-01-23 11:48:28.136995: step: 360/1541, loss: 0.17838215827941895 2023-01-23 11:48:29.061826: step: 362/1541, loss: 0.011904161423444748 2023-01-23 11:48:29.913118: step: 364/1541, loss: 0.02346409484744072 2023-01-23 11:48:30.836651: step: 366/1541, loss: 0.1939854621887207 2023-01-23 11:48:31.768973: step: 368/1541, loss: 0.12585248053073883 2023-01-23 11:48:32.697498: step: 370/1541, loss: 0.1305983066558838 2023-01-23 11:48:33.584679: step: 372/1541, loss: 0.2877277731895447 2023-01-23 11:48:34.527851: step: 374/1541, loss: 0.08608070760965347 2023-01-23 11:48:35.475314: step: 376/1541, loss: 0.10127442330121994 2023-01-23 11:48:36.385451: step: 378/1541, loss: 0.0909426212310791 2023-01-23 11:48:37.242326: step: 380/1541, loss: 0.005984439980238676 2023-01-23 11:48:38.123547: step: 382/1541, loss: 0.5304558277130127 2023-01-23 11:48:39.025360: step: 384/1541, loss: 0.023212170228362083 2023-01-23 11:48:39.900065: step: 386/1541, loss: 0.19267502427101135 2023-01-23 11:48:40.780484: step: 388/1541, loss: 0.050350215286016464 2023-01-23 11:48:41.683881: step: 390/1541, loss: 0.30755922198295593 2023-01-23 11:48:42.613674: step: 392/1541, loss: 0.20817044377326965 2023-01-23 11:48:43.515546: step: 394/1541, loss: 0.025004519149661064 2023-01-23 11:48:44.417730: step: 396/1541, loss: 0.07108680158853531 2023-01-23 11:48:45.338215: step: 398/1541, loss: 0.5227971076965332 2023-01-23 11:48:46.259289: step: 400/1541, loss: 0.09266956895589828 2023-01-23 11:48:47.142403: step: 402/1541, loss: 0.032907240092754364 2023-01-23 11:48:48.067548: step: 404/1541, loss: 0.19445347785949707 2023-01-23 11:48:48.968500: step: 406/1541, loss: 0.00846538133919239 2023-01-23 11:48:49.881604: step: 408/1541, loss: 0.030208509415388107 2023-01-23 11:48:50.764111: step: 410/1541, loss: 0.1745818704366684 2023-01-23 11:48:51.652511: step: 412/1541, loss: 0.9908740520477295 2023-01-23 11:48:52.571890: step: 414/1541, loss: 0.13945618271827698 2023-01-23 11:48:53.442287: step: 416/1541, loss: 0.09944658726453781 2023-01-23 11:48:54.353637: step: 418/1541, loss: 0.003579668002203107 2023-01-23 11:48:55.242740: step: 420/1541, loss: 0.02662847563624382 2023-01-23 11:48:56.124821: step: 422/1541, loss: 0.18355415761470795 2023-01-23 11:48:57.095369: step: 424/1541, loss: 0.18178164958953857 2023-01-23 11:48:57.997657: step: 426/1541, loss: 0.012094762176275253 2023-01-23 11:48:58.855776: step: 428/1541, loss: 0.24856966733932495 2023-01-23 11:48:59.747959: step: 430/1541, loss: 0.22925375401973724 2023-01-23 11:49:00.588928: step: 432/1541, loss: 0.05462697893381119 2023-01-23 11:49:01.517002: step: 434/1541, loss: 0.038872767239809036 2023-01-23 11:49:02.412180: step: 436/1541, loss: 0.3058989942073822 2023-01-23 11:49:03.298129: step: 438/1541, loss: 0.044177740812301636 2023-01-23 11:49:04.224676: step: 440/1541, loss: 0.033388346433639526 2023-01-23 11:49:05.150597: step: 442/1541, loss: 0.006543638650327921 2023-01-23 11:49:06.063137: step: 444/1541, loss: 0.18339627981185913 2023-01-23 11:49:06.968630: step: 446/1541, loss: 0.32008683681488037 2023-01-23 11:49:07.859670: step: 448/1541, loss: 0.0018913449021056294 2023-01-23 11:49:08.761696: step: 450/1541, loss: 0.11341676861047745 2023-01-23 11:49:09.643480: step: 452/1541, loss: 8.684663772583008 2023-01-23 11:49:10.525447: step: 454/1541, loss: 0.00766428466886282 2023-01-23 11:49:11.413163: step: 456/1541, loss: 0.10152798891067505 2023-01-23 11:49:12.302270: step: 458/1541, loss: 0.28454121947288513 2023-01-23 11:49:13.222083: step: 460/1541, loss: 0.06432391703128815 2023-01-23 11:49:14.096660: step: 462/1541, loss: 0.16332240402698517 2023-01-23 11:49:15.017416: step: 464/1541, loss: 0.6611699461936951 2023-01-23 11:49:15.950357: step: 466/1541, loss: 0.056434515863657 2023-01-23 11:49:16.860054: step: 468/1541, loss: 0.012476441450417042 2023-01-23 11:49:17.733127: step: 470/1541, loss: 0.12998726963996887 2023-01-23 11:49:18.745565: step: 472/1541, loss: 0.10719074308872223 2023-01-23 11:49:19.652587: step: 474/1541, loss: 0.028164200484752655 2023-01-23 11:49:20.549385: step: 476/1541, loss: 0.01073535904288292 2023-01-23 11:49:21.434889: step: 478/1541, loss: 0.016269056126475334 2023-01-23 11:49:22.383305: step: 480/1541, loss: 0.08969372510910034 2023-01-23 11:49:23.293545: step: 482/1541, loss: 1.230133295059204 2023-01-23 11:49:24.156150: step: 484/1541, loss: 0.004244144074618816 2023-01-23 11:49:25.045419: step: 486/1541, loss: 0.1328079104423523 2023-01-23 11:49:25.940355: step: 488/1541, loss: 0.01166115514934063 2023-01-23 11:49:26.866180: step: 490/1541, loss: 0.027865523472428322 2023-01-23 11:49:27.755178: step: 492/1541, loss: 2.541501998901367 2023-01-23 11:49:28.653929: step: 494/1541, loss: 0.039740242063999176 2023-01-23 11:49:29.559430: step: 496/1541, loss: 0.09078335016965866 2023-01-23 11:49:30.494506: step: 498/1541, loss: 0.09586714208126068 2023-01-23 11:49:31.438891: step: 500/1541, loss: 0.06022334843873978 2023-01-23 11:49:32.434192: step: 502/1541, loss: 0.38211339712142944 2023-01-23 11:49:33.328117: step: 504/1541, loss: 0.09664562344551086 2023-01-23 11:49:34.259100: step: 506/1541, loss: 0.5069196820259094 2023-01-23 11:49:35.181188: step: 508/1541, loss: 0.1606425642967224 2023-01-23 11:49:36.222324: step: 510/1541, loss: 0.036573972553014755 2023-01-23 11:49:37.078316: step: 512/1541, loss: 0.0020281595643609762 2023-01-23 11:49:38.029934: step: 514/1541, loss: 0.033207859843969345 2023-01-23 11:49:38.907045: step: 516/1541, loss: 0.4815814197063446 2023-01-23 11:49:39.788083: step: 518/1541, loss: 0.02620786428451538 2023-01-23 11:49:40.700374: step: 520/1541, loss: 0.045171983540058136 2023-01-23 11:49:41.593947: step: 522/1541, loss: 0.10934020578861237 2023-01-23 11:49:42.465940: step: 524/1541, loss: 0.22943200170993805 2023-01-23 11:49:43.411373: step: 526/1541, loss: 0.01649601384997368 2023-01-23 11:49:44.303132: step: 528/1541, loss: 0.32926908135414124 2023-01-23 11:49:45.227751: step: 530/1541, loss: 0.05050654709339142 2023-01-23 11:49:46.098104: step: 532/1541, loss: 0.12785479426383972 2023-01-23 11:49:46.990465: step: 534/1541, loss: 0.2110719084739685 2023-01-23 11:49:47.949405: step: 536/1541, loss: 0.15089111030101776 2023-01-23 11:49:48.809963: step: 538/1541, loss: 0.012712469324469566 2023-01-23 11:49:49.715415: step: 540/1541, loss: 0.031031867489218712 2023-01-23 11:49:50.572526: step: 542/1541, loss: 0.12519294023513794 2023-01-23 11:49:51.490877: step: 544/1541, loss: 0.10353659838438034 2023-01-23 11:49:52.397556: step: 546/1541, loss: 0.006417770870029926 2023-01-23 11:49:53.281443: step: 548/1541, loss: 0.03930801898241043 2023-01-23 11:49:54.228081: step: 550/1541, loss: 0.06852038204669952 2023-01-23 11:49:55.096602: step: 552/1541, loss: 0.04986469820141792 2023-01-23 11:49:56.001499: step: 554/1541, loss: 0.13323460519313812 2023-01-23 11:49:56.904750: step: 556/1541, loss: 0.455850750207901 2023-01-23 11:49:57.779016: step: 558/1541, loss: 0.08231904357671738 2023-01-23 11:49:58.660248: step: 560/1541, loss: 0.06439851969480515 2023-01-23 11:49:59.561227: step: 562/1541, loss: 0.003698268672451377 2023-01-23 11:50:00.464013: step: 564/1541, loss: 0.27755871415138245 2023-01-23 11:50:01.355601: step: 566/1541, loss: 0.07257737964391708 2023-01-23 11:50:02.286103: step: 568/1541, loss: 0.11565890908241272 2023-01-23 11:50:03.155536: step: 570/1541, loss: 0.15560919046401978 2023-01-23 11:50:04.085336: step: 572/1541, loss: 0.18181633949279785 2023-01-23 11:50:04.966634: step: 574/1541, loss: 7.023135185241699 2023-01-23 11:50:05.917266: step: 576/1541, loss: 0.10176055878400803 2023-01-23 11:50:06.798388: step: 578/1541, loss: 0.02960319072008133 2023-01-23 11:50:07.652195: step: 580/1541, loss: 0.06709069758653641 2023-01-23 11:50:08.535697: step: 582/1541, loss: 0.01694110780954361 2023-01-23 11:50:09.493125: step: 584/1541, loss: 0.10530486702919006 2023-01-23 11:50:10.419532: step: 586/1541, loss: 0.0427742674946785 2023-01-23 11:50:11.291909: step: 588/1541, loss: 0.055254977196455 2023-01-23 11:50:12.200946: step: 590/1541, loss: 0.5744717717170715 2023-01-23 11:50:13.180276: step: 592/1541, loss: 0.33157145977020264 2023-01-23 11:50:14.069595: step: 594/1541, loss: 0.19093266129493713 2023-01-23 11:50:14.985667: step: 596/1541, loss: 0.13799113035202026 2023-01-23 11:50:15.912990: step: 598/1541, loss: 0.00839904509484768 2023-01-23 11:50:16.853260: step: 600/1541, loss: 0.044906534254550934 2023-01-23 11:50:17.777899: step: 602/1541, loss: 0.07538709044456482 2023-01-23 11:50:18.647123: step: 604/1541, loss: 0.20243577659130096 2023-01-23 11:50:19.578494: step: 606/1541, loss: 0.03769941255450249 2023-01-23 11:50:20.502547: step: 608/1541, loss: 0.023066556081175804 2023-01-23 11:50:21.390843: step: 610/1541, loss: 0.217838317155838 2023-01-23 11:50:22.282758: step: 612/1541, loss: 0.5382730960845947 2023-01-23 11:50:23.196102: step: 614/1541, loss: 0.2803942859172821 2023-01-23 11:50:24.111026: step: 616/1541, loss: 0.06715627014636993 2023-01-23 11:50:25.013719: step: 618/1541, loss: 0.017308376729488373 2023-01-23 11:50:25.873283: step: 620/1541, loss: 0.014438900165259838 2023-01-23 11:50:26.745998: step: 622/1541, loss: 0.10010573267936707 2023-01-23 11:50:27.627939: step: 624/1541, loss: 0.014646733179688454 2023-01-23 11:50:28.520424: step: 626/1541, loss: 0.005755532532930374 2023-01-23 11:50:29.417228: step: 628/1541, loss: 0.09500095993280411 2023-01-23 11:50:30.288897: step: 630/1541, loss: 0.1859283596277237 2023-01-23 11:50:31.176117: step: 632/1541, loss: 0.13091619312763214 2023-01-23 11:50:32.073547: step: 634/1541, loss: 0.2236030399799347 2023-01-23 11:50:32.942584: step: 636/1541, loss: 0.7533758282661438 2023-01-23 11:50:33.846876: step: 638/1541, loss: 0.28455883264541626 2023-01-23 11:50:34.761283: step: 640/1541, loss: 0.13362498581409454 2023-01-23 11:50:35.667862: step: 642/1541, loss: 0.048926569521427155 2023-01-23 11:50:36.587564: step: 644/1541, loss: 0.030203010886907578 2023-01-23 11:50:37.552966: step: 646/1541, loss: 0.02425207383930683 2023-01-23 11:50:38.467391: step: 648/1541, loss: 0.020470241084694862 2023-01-23 11:50:39.366794: step: 650/1541, loss: 0.008074442856013775 2023-01-23 11:50:40.277656: step: 652/1541, loss: 0.15367063879966736 2023-01-23 11:50:41.172418: step: 654/1541, loss: 0.12192515283823013 2023-01-23 11:50:42.043318: step: 656/1541, loss: 0.03492557257413864 2023-01-23 11:50:42.907674: step: 658/1541, loss: 0.033017128705978394 2023-01-23 11:50:43.799181: step: 660/1541, loss: 0.13752125203609467 2023-01-23 11:50:44.697958: step: 662/1541, loss: 0.014730260707437992 2023-01-23 11:50:45.548778: step: 664/1541, loss: 0.004490268882364035 2023-01-23 11:50:46.466281: step: 666/1541, loss: 0.03902365267276764 2023-01-23 11:50:47.383942: step: 668/1541, loss: 0.029933206737041473 2023-01-23 11:50:48.295775: step: 670/1541, loss: 0.2887967526912689 2023-01-23 11:50:49.188858: step: 672/1541, loss: 0.07608159631490707 2023-01-23 11:50:50.128231: step: 674/1541, loss: 0.2422683984041214 2023-01-23 11:50:51.028972: step: 676/1541, loss: 0.0707257017493248 2023-01-23 11:50:51.904475: step: 678/1541, loss: 0.03485904261469841 2023-01-23 11:50:52.792266: step: 680/1541, loss: 0.21780137717723846 2023-01-23 11:50:53.699545: step: 682/1541, loss: 0.36662787199020386 2023-01-23 11:50:54.656848: step: 684/1541, loss: 0.019843699410557747 2023-01-23 11:50:55.556522: step: 686/1541, loss: 0.1365511119365692 2023-01-23 11:50:56.453124: step: 688/1541, loss: 0.425354540348053 2023-01-23 11:50:57.322976: step: 690/1541, loss: 0.16794759035110474 2023-01-23 11:50:58.240365: step: 692/1541, loss: 0.4071732461452484 2023-01-23 11:50:59.153575: step: 694/1541, loss: 0.12061475217342377 2023-01-23 11:51:00.044786: step: 696/1541, loss: 0.262957900762558 2023-01-23 11:51:00.951943: step: 698/1541, loss: 0.23317596316337585 2023-01-23 11:51:01.858744: step: 700/1541, loss: 0.013017913326621056 2023-01-23 11:51:02.797643: step: 702/1541, loss: 0.27790722250938416 2023-01-23 11:51:03.666574: step: 704/1541, loss: 0.030160069465637207 2023-01-23 11:51:04.557750: step: 706/1541, loss: 0.09091481566429138 2023-01-23 11:51:05.473507: step: 708/1541, loss: 0.03601314127445221 2023-01-23 11:51:06.400670: step: 710/1541, loss: 0.04274101182818413 2023-01-23 11:51:07.297565: step: 712/1541, loss: 0.0897597223520279 2023-01-23 11:51:08.232337: step: 714/1541, loss: 0.08937804400920868 2023-01-23 11:51:09.203927: step: 716/1541, loss: 0.15246346592903137 2023-01-23 11:51:10.084062: step: 718/1541, loss: 0.0378357470035553 2023-01-23 11:51:10.979302: step: 720/1541, loss: 0.08438853919506073 2023-01-23 11:51:11.879557: step: 722/1541, loss: 0.015587848611176014 2023-01-23 11:51:12.754533: step: 724/1541, loss: 0.11556185781955719 2023-01-23 11:51:13.614275: step: 726/1541, loss: 0.09363541007041931 2023-01-23 11:51:14.504099: step: 728/1541, loss: 0.09623829275369644 2023-01-23 11:51:15.415468: step: 730/1541, loss: 0.047123972326517105 2023-01-23 11:51:16.332060: step: 732/1541, loss: 0.19747908413410187 2023-01-23 11:51:17.268905: step: 734/1541, loss: 0.02366304025053978 2023-01-23 11:51:18.161631: step: 736/1541, loss: 0.42323464155197144 2023-01-23 11:51:19.032683: step: 738/1541, loss: 0.06172638386487961 2023-01-23 11:51:19.928976: step: 740/1541, loss: 0.7785620093345642 2023-01-23 11:51:20.817861: step: 742/1541, loss: 0.29078543186187744 2023-01-23 11:51:21.699869: step: 744/1541, loss: 0.13192152976989746 2023-01-23 11:51:22.594038: step: 746/1541, loss: 0.0397116057574749 2023-01-23 11:51:23.518365: step: 748/1541, loss: 0.0773831456899643 2023-01-23 11:51:24.435009: step: 750/1541, loss: 0.1270792931318283 2023-01-23 11:51:25.338015: step: 752/1541, loss: 0.13453972339630127 2023-01-23 11:51:26.258908: step: 754/1541, loss: 0.021323950961232185 2023-01-23 11:51:27.149932: step: 756/1541, loss: 0.04834301024675369 2023-01-23 11:51:28.078285: step: 758/1541, loss: 0.09968139976263046 2023-01-23 11:51:28.990777: step: 760/1541, loss: 0.299378365278244 2023-01-23 11:51:29.879288: step: 762/1541, loss: 0.1947169452905655 2023-01-23 11:51:30.769924: step: 764/1541, loss: 0.1060631051659584 2023-01-23 11:51:31.640606: step: 766/1541, loss: 0.16798126697540283 2023-01-23 11:51:32.583491: step: 768/1541, loss: 0.08057677000761032 2023-01-23 11:51:33.492990: step: 770/1541, loss: 0.03864821046590805 2023-01-23 11:51:34.396351: step: 772/1541, loss: 0.23469816148281097 2023-01-23 11:51:35.278600: step: 774/1541, loss: 1.21866774559021 2023-01-23 11:51:36.170203: step: 776/1541, loss: 0.013097891584038734 2023-01-23 11:51:37.032761: step: 778/1541, loss: 0.009009207598865032 2023-01-23 11:51:37.942583: step: 780/1541, loss: 0.8470601439476013 2023-01-23 11:51:38.805940: step: 782/1541, loss: 0.10685563087463379 2023-01-23 11:51:39.753929: step: 784/1541, loss: 0.020629698410630226 2023-01-23 11:51:40.627646: step: 786/1541, loss: 0.032406408339738846 2023-01-23 11:51:41.572260: step: 788/1541, loss: 0.07675530761480331 2023-01-23 11:51:42.580436: step: 790/1541, loss: 0.22312501072883606 2023-01-23 11:51:43.503063: step: 792/1541, loss: 1.148276925086975 2023-01-23 11:51:44.373935: step: 794/1541, loss: 0.2364121973514557 2023-01-23 11:51:45.256464: step: 796/1541, loss: 0.02252083271741867 2023-01-23 11:51:46.128761: step: 798/1541, loss: 0.0899815633893013 2023-01-23 11:51:47.014460: step: 800/1541, loss: 0.11727787554264069 2023-01-23 11:51:47.923077: step: 802/1541, loss: 0.08012962341308594 2023-01-23 11:51:48.822579: step: 804/1541, loss: 0.01416700053960085 2023-01-23 11:51:49.722857: step: 806/1541, loss: 0.39860865473747253 2023-01-23 11:51:50.617951: step: 808/1541, loss: 0.05343712866306305 2023-01-23 11:51:51.559023: step: 810/1541, loss: 0.019445516169071198 2023-01-23 11:51:52.425045: step: 812/1541, loss: 0.14349238574504852 2023-01-23 11:51:53.330444: step: 814/1541, loss: 0.000625814835075289 2023-01-23 11:51:54.291176: step: 816/1541, loss: 0.01530514843761921 2023-01-23 11:51:55.190240: step: 818/1541, loss: 0.017373111099004745 2023-01-23 11:51:56.117550: step: 820/1541, loss: 0.07842501252889633 2023-01-23 11:51:57.093638: step: 822/1541, loss: 0.006824626587331295 2023-01-23 11:51:57.965506: step: 824/1541, loss: 0.029890626668930054 2023-01-23 11:51:58.869163: step: 826/1541, loss: 0.06577234715223312 2023-01-23 11:51:59.797822: step: 828/1541, loss: 0.5692539215087891 2023-01-23 11:52:00.751600: step: 830/1541, loss: 0.0045669302344322205 2023-01-23 11:52:01.642087: step: 832/1541, loss: 0.007477468345314264 2023-01-23 11:52:02.611269: step: 834/1541, loss: 0.04090901091694832 2023-01-23 11:52:03.507653: step: 836/1541, loss: 0.08053722977638245 2023-01-23 11:52:04.422663: step: 838/1541, loss: 0.03511793166399002 2023-01-23 11:52:05.319530: step: 840/1541, loss: 0.005788794718682766 2023-01-23 11:52:06.222555: step: 842/1541, loss: 0.04311392828822136 2023-01-23 11:52:07.137564: step: 844/1541, loss: 0.554997444152832 2023-01-23 11:52:08.007035: step: 846/1541, loss: 0.05297338590025902 2023-01-23 11:52:08.945570: step: 848/1541, loss: 0.032484859228134155 2023-01-23 11:52:09.876077: step: 850/1541, loss: 0.19810017943382263 2023-01-23 11:52:10.793601: step: 852/1541, loss: 0.004736205097287893 2023-01-23 11:52:11.704140: step: 854/1541, loss: 0.07125614583492279 2023-01-23 11:52:12.564594: step: 856/1541, loss: 0.1626203954219818 2023-01-23 11:52:13.424668: step: 858/1541, loss: 0.012491623871028423 2023-01-23 11:52:14.294045: step: 860/1541, loss: 0.43921834230422974 2023-01-23 11:52:15.171942: step: 862/1541, loss: 0.1755557358264923 2023-01-23 11:52:16.054127: step: 864/1541, loss: 0.030533326789736748 2023-01-23 11:52:16.979296: step: 866/1541, loss: 0.8660171031951904 2023-01-23 11:52:17.837074: step: 868/1541, loss: 0.2838675379753113 2023-01-23 11:52:18.722574: step: 870/1541, loss: 0.001997654791921377 2023-01-23 11:52:19.640617: step: 872/1541, loss: 0.03294575214385986 2023-01-23 11:52:20.545846: step: 874/1541, loss: 0.03858790919184685 2023-01-23 11:52:21.494078: step: 876/1541, loss: 0.01672336272895336 2023-01-23 11:52:22.438118: step: 878/1541, loss: 0.0011800454230979085 2023-01-23 11:52:23.338966: step: 880/1541, loss: 1.1558390855789185 2023-01-23 11:52:24.271770: step: 882/1541, loss: 0.510873556137085 2023-01-23 11:52:25.172409: step: 884/1541, loss: 0.02568931132555008 2023-01-23 11:52:26.089006: step: 886/1541, loss: 0.018223706632852554 2023-01-23 11:52:26.985115: step: 888/1541, loss: 0.2792662978172302 2023-01-23 11:52:27.886472: step: 890/1541, loss: 0.1026688739657402 2023-01-23 11:52:28.780895: step: 892/1541, loss: 0.11160627007484436 2023-01-23 11:52:29.652032: step: 894/1541, loss: 0.08889393508434296 2023-01-23 11:52:30.552597: step: 896/1541, loss: 0.008166885003447533 2023-01-23 11:52:31.439046: step: 898/1541, loss: 0.2076529711484909 2023-01-23 11:52:32.334096: step: 900/1541, loss: 0.02665853127837181 2023-01-23 11:52:33.286080: step: 902/1541, loss: 0.0027262484654784203 2023-01-23 11:52:34.224409: step: 904/1541, loss: 0.06133480370044708 2023-01-23 11:52:35.252452: step: 906/1541, loss: 0.06934890151023865 2023-01-23 11:52:36.154699: step: 908/1541, loss: 0.0401599295437336 2023-01-23 11:52:37.120825: step: 910/1541, loss: 0.08593754470348358 2023-01-23 11:52:38.033491: step: 912/1541, loss: 0.07202905416488647 2023-01-23 11:52:38.899912: step: 914/1541, loss: 0.10008642822504044 2023-01-23 11:52:39.742854: step: 916/1541, loss: 0.028437156230211258 2023-01-23 11:52:40.662361: step: 918/1541, loss: 0.12236686795949936 2023-01-23 11:52:41.550278: step: 920/1541, loss: 0.07165417820215225 2023-01-23 11:52:42.541284: step: 922/1541, loss: 0.05330215394496918 2023-01-23 11:52:43.399585: step: 924/1541, loss: 0.19169919192790985 2023-01-23 11:52:44.321906: step: 926/1541, loss: 0.04615946486592293 2023-01-23 11:52:45.197376: step: 928/1541, loss: 0.040327370166778564 2023-01-23 11:52:46.105512: step: 930/1541, loss: 0.09185163676738739 2023-01-23 11:52:47.009949: step: 932/1541, loss: 0.059094518423080444 2023-01-23 11:52:47.987138: step: 934/1541, loss: 0.058067046105861664 2023-01-23 11:52:48.941371: step: 936/1541, loss: 0.07190883159637451 2023-01-23 11:52:49.840058: step: 938/1541, loss: 0.014718986116349697 2023-01-23 11:52:50.766136: step: 940/1541, loss: 0.1552790105342865 2023-01-23 11:52:51.661384: step: 942/1541, loss: 0.020408829674124718 2023-01-23 11:52:52.539548: step: 944/1541, loss: 0.04431452602148056 2023-01-23 11:52:53.439080: step: 946/1541, loss: 0.244188129901886 2023-01-23 11:52:54.361392: step: 948/1541, loss: 0.1498764008283615 2023-01-23 11:52:55.233991: step: 950/1541, loss: 0.41229525208473206 2023-01-23 11:52:56.121629: step: 952/1541, loss: 0.0014734339201822877 2023-01-23 11:52:56.976130: step: 954/1541, loss: 0.07496301084756851 2023-01-23 11:52:57.862428: step: 956/1541, loss: 0.057720448821783066 2023-01-23 11:52:58.708043: step: 958/1541, loss: 0.00531450379639864 2023-01-23 11:52:59.563159: step: 960/1541, loss: 0.002995202550664544 2023-01-23 11:53:00.453660: step: 962/1541, loss: 0.033393457531929016 2023-01-23 11:53:01.362843: step: 964/1541, loss: 0.3503430485725403 2023-01-23 11:53:02.263157: step: 966/1541, loss: 0.10836584120988846 2023-01-23 11:53:03.142547: step: 968/1541, loss: 1.0923482179641724 2023-01-23 11:53:04.055148: step: 970/1541, loss: 0.012383240275084972 2023-01-23 11:53:04.937175: step: 972/1541, loss: 0.14469535648822784 2023-01-23 11:53:05.906497: step: 974/1541, loss: 0.11273539811372757 2023-01-23 11:53:06.779815: step: 976/1541, loss: 0.021994229406118393 2023-01-23 11:53:07.724616: step: 978/1541, loss: 0.2893867492675781 2023-01-23 11:53:08.644674: step: 980/1541, loss: 0.04498506709933281 2023-01-23 11:53:09.541196: step: 982/1541, loss: 0.04687219858169556 2023-01-23 11:53:10.485944: step: 984/1541, loss: 0.18527066707611084 2023-01-23 11:53:11.363051: step: 986/1541, loss: 0.2973688244819641 2023-01-23 11:53:12.229210: step: 988/1541, loss: 0.060743194073438644 2023-01-23 11:53:13.138064: step: 990/1541, loss: 0.015220018103718758 2023-01-23 11:53:14.088844: step: 992/1541, loss: 0.19862821698188782 2023-01-23 11:53:14.988346: step: 994/1541, loss: 0.02202650159597397 2023-01-23 11:53:15.885597: step: 996/1541, loss: 0.1516629010438919 2023-01-23 11:53:16.753156: step: 998/1541, loss: 0.7148510813713074 2023-01-23 11:53:17.671492: step: 1000/1541, loss: 0.0115775465965271 2023-01-23 11:53:18.582682: step: 1002/1541, loss: 0.09414799511432648 2023-01-23 11:53:19.469057: step: 1004/1541, loss: 0.3181488513946533 2023-01-23 11:53:20.407293: step: 1006/1541, loss: 0.06890654563903809 2023-01-23 11:53:21.318447: step: 1008/1541, loss: 0.10736365616321564 2023-01-23 11:53:22.252390: step: 1010/1541, loss: 0.055460672825574875 2023-01-23 11:53:23.140629: step: 1012/1541, loss: 0.03353944048285484 2023-01-23 11:53:24.022011: step: 1014/1541, loss: 0.19795182347297668 2023-01-23 11:53:24.900743: step: 1016/1541, loss: 0.008039968088269234 2023-01-23 11:53:25.787197: step: 1018/1541, loss: 0.0069231074303388596 2023-01-23 11:53:26.665430: step: 1020/1541, loss: 0.01003473810851574 2023-01-23 11:53:27.536611: step: 1022/1541, loss: 0.14392562210559845 2023-01-23 11:53:28.463592: step: 1024/1541, loss: 0.12379047274589539 2023-01-23 11:53:29.355777: step: 1026/1541, loss: 0.040571339428424835 2023-01-23 11:53:30.261241: step: 1028/1541, loss: 0.12169724702835083 2023-01-23 11:53:31.137313: step: 1030/1541, loss: 0.012250622734427452 2023-01-23 11:53:32.152315: step: 1032/1541, loss: 0.22919028997421265 2023-01-23 11:53:33.127769: step: 1034/1541, loss: 0.061488501727581024 2023-01-23 11:53:34.005139: step: 1036/1541, loss: 0.07262375205755234 2023-01-23 11:53:34.854851: step: 1038/1541, loss: 0.024285197257995605 2023-01-23 11:53:35.802721: step: 1040/1541, loss: 0.2652874290943146 2023-01-23 11:53:36.724201: step: 1042/1541, loss: 0.011212201789021492 2023-01-23 11:53:37.616905: step: 1044/1541, loss: 0.1903955489397049 2023-01-23 11:53:38.475181: step: 1046/1541, loss: 0.1919630914926529 2023-01-23 11:53:39.384455: step: 1048/1541, loss: 0.029754754155874252 2023-01-23 11:53:40.268526: step: 1050/1541, loss: 0.18298999965190887 2023-01-23 11:53:41.223774: step: 1052/1541, loss: 0.2073851227760315 2023-01-23 11:53:42.155174: step: 1054/1541, loss: 0.13894210755825043 2023-01-23 11:53:43.024331: step: 1056/1541, loss: 0.10231516510248184 2023-01-23 11:53:43.965071: step: 1058/1541, loss: 0.3909871280193329 2023-01-23 11:53:44.851571: step: 1060/1541, loss: 0.04292822629213333 2023-01-23 11:53:45.781051: step: 1062/1541, loss: 0.08791842311620712 2023-01-23 11:53:46.689753: step: 1064/1541, loss: 0.11125832796096802 2023-01-23 11:53:47.594058: step: 1066/1541, loss: 0.03789640590548515 2023-01-23 11:53:48.504500: step: 1068/1541, loss: 0.844463050365448 2023-01-23 11:53:49.425191: step: 1070/1541, loss: 0.30220097303390503 2023-01-23 11:53:50.311184: step: 1072/1541, loss: 0.0712190642952919 2023-01-23 11:53:51.170138: step: 1074/1541, loss: 3.0314745903015137 2023-01-23 11:53:52.035590: step: 1076/1541, loss: 0.013935140334069729 2023-01-23 11:53:52.958057: step: 1078/1541, loss: 0.08787105232477188 2023-01-23 11:53:53.822259: step: 1080/1541, loss: 0.17404741048812866 2023-01-23 11:53:54.778245: step: 1082/1541, loss: 0.07188589125871658 2023-01-23 11:53:55.682026: step: 1084/1541, loss: 0.04903603345155716 2023-01-23 11:53:56.580383: step: 1086/1541, loss: 0.12661753594875336 2023-01-23 11:53:57.522838: step: 1088/1541, loss: 0.07023294270038605 2023-01-23 11:53:58.413658: step: 1090/1541, loss: 0.2444765418767929 2023-01-23 11:53:59.270757: step: 1092/1541, loss: 0.005831921007484198 2023-01-23 11:54:00.145230: step: 1094/1541, loss: 0.20765697956085205 2023-01-23 11:54:01.026887: step: 1096/1541, loss: 0.03870366886258125 2023-01-23 11:54:01.935442: step: 1098/1541, loss: 0.016592595726251602 2023-01-23 11:54:02.834720: step: 1100/1541, loss: 0.2490687370300293 2023-01-23 11:54:03.748805: step: 1102/1541, loss: 0.3520933985710144 2023-01-23 11:54:04.630898: step: 1104/1541, loss: 0.05611676350235939 2023-01-23 11:54:05.516978: step: 1106/1541, loss: 0.17542628943920135 2023-01-23 11:54:06.534290: step: 1108/1541, loss: 0.33342042565345764 2023-01-23 11:54:07.398210: step: 1110/1541, loss: 0.3667638897895813 2023-01-23 11:54:08.263719: step: 1112/1541, loss: 0.03567889332771301 2023-01-23 11:54:09.136289: step: 1114/1541, loss: 0.19027262926101685 2023-01-23 11:54:10.023236: step: 1116/1541, loss: 0.2288288027048111 2023-01-23 11:54:10.887486: step: 1118/1541, loss: 0.06618856638669968 2023-01-23 11:54:11.797100: step: 1120/1541, loss: 0.32039427757263184 2023-01-23 11:54:12.665476: step: 1122/1541, loss: 0.037964917719364166 2023-01-23 11:54:13.571829: step: 1124/1541, loss: 0.05066267400979996 2023-01-23 11:54:14.470131: step: 1126/1541, loss: 0.2940013110637665 2023-01-23 11:54:15.353902: step: 1128/1541, loss: 0.06157400459051132 2023-01-23 11:54:16.206085: step: 1130/1541, loss: 0.9964472055435181 2023-01-23 11:54:17.094507: step: 1132/1541, loss: 0.48802998661994934 2023-01-23 11:54:17.988987: step: 1134/1541, loss: 0.4998731315135956 2023-01-23 11:54:18.883366: step: 1136/1541, loss: 0.01288153138011694 2023-01-23 11:54:19.818228: step: 1138/1541, loss: 0.02115303836762905 2023-01-23 11:54:20.718008: step: 1140/1541, loss: 0.0296027734875679 2023-01-23 11:54:21.608806: step: 1142/1541, loss: 0.04410051554441452 2023-01-23 11:54:22.483026: step: 1144/1541, loss: 0.036315932869911194 2023-01-23 11:54:23.350582: step: 1146/1541, loss: 0.0329771488904953 2023-01-23 11:54:24.317467: step: 1148/1541, loss: 1.2988816499710083 2023-01-23 11:54:25.198983: step: 1150/1541, loss: 0.09471631795167923 2023-01-23 11:54:26.137459: step: 1152/1541, loss: 0.24579359591007233 2023-01-23 11:54:27.063549: step: 1154/1541, loss: 0.05896524712443352 2023-01-23 11:54:27.918097: step: 1156/1541, loss: 0.056211475282907486 2023-01-23 11:54:28.794460: step: 1158/1541, loss: 0.032196495682001114 2023-01-23 11:54:29.737169: step: 1160/1541, loss: 0.21395671367645264 2023-01-23 11:54:30.625564: step: 1162/1541, loss: 0.1523798704147339 2023-01-23 11:54:31.525239: step: 1164/1541, loss: 0.2042427510023117 2023-01-23 11:54:32.401382: step: 1166/1541, loss: 0.042715124785900116 2023-01-23 11:54:33.346024: step: 1168/1541, loss: 0.003380304668098688 2023-01-23 11:54:34.282499: step: 1170/1541, loss: 0.10191260278224945 2023-01-23 11:54:35.171731: step: 1172/1541, loss: 0.11370740830898285 2023-01-23 11:54:36.087571: step: 1174/1541, loss: 0.12179945409297943 2023-01-23 11:54:36.978146: step: 1176/1541, loss: 0.014820230193436146 2023-01-23 11:54:37.945918: step: 1178/1541, loss: 0.029735784977674484 2023-01-23 11:54:38.830343: step: 1180/1541, loss: 0.07312603294849396 2023-01-23 11:54:39.709407: step: 1182/1541, loss: 0.1202128455042839 2023-01-23 11:54:40.589467: step: 1184/1541, loss: 0.08878955990076065 2023-01-23 11:54:41.569443: step: 1186/1541, loss: 0.02746589668095112 2023-01-23 11:54:42.486325: step: 1188/1541, loss: 0.033768072724342346 2023-01-23 11:54:43.428349: step: 1190/1541, loss: 0.14121697843074799 2023-01-23 11:54:44.343642: step: 1192/1541, loss: 0.02787938341498375 2023-01-23 11:54:45.197761: step: 1194/1541, loss: 0.0058261980302631855 2023-01-23 11:54:46.054250: step: 1196/1541, loss: 0.35693082213401794 2023-01-23 11:54:46.952536: step: 1198/1541, loss: 0.1275789588689804 2023-01-23 11:54:47.866342: step: 1200/1541, loss: 0.12296368181705475 2023-01-23 11:54:48.758735: step: 1202/1541, loss: 0.14270518720149994 2023-01-23 11:54:49.667812: step: 1204/1541, loss: 0.056646063923835754 2023-01-23 11:54:50.606973: step: 1206/1541, loss: 0.0906967744231224 2023-01-23 11:54:51.524473: step: 1208/1541, loss: 0.709784209728241 2023-01-23 11:54:52.405203: step: 1210/1541, loss: 0.10063636302947998 2023-01-23 11:54:53.292880: step: 1212/1541, loss: 0.04052039608359337 2023-01-23 11:54:54.242544: step: 1214/1541, loss: 0.1509343981742859 2023-01-23 11:54:55.146116: step: 1216/1541, loss: 0.04039076715707779 2023-01-23 11:54:56.048332: step: 1218/1541, loss: 0.36508908867836 2023-01-23 11:54:57.018164: step: 1220/1541, loss: 0.011675504967570305 2023-01-23 11:54:57.886477: step: 1222/1541, loss: 0.3844046890735626 2023-01-23 11:54:58.768059: step: 1224/1541, loss: 0.08137622475624084 2023-01-23 11:54:59.641750: step: 1226/1541, loss: 0.26737919449806213 2023-01-23 11:55:00.510672: step: 1228/1541, loss: 0.008100053295493126 2023-01-23 11:55:01.398374: step: 1230/1541, loss: 0.025393046438694 2023-01-23 11:55:02.338175: step: 1232/1541, loss: 0.0070480299182236195 2023-01-23 11:55:03.266154: step: 1234/1541, loss: 0.016845300793647766 2023-01-23 11:55:04.189441: step: 1236/1541, loss: 0.9145895838737488 2023-01-23 11:55:05.052103: step: 1238/1541, loss: 0.12976747751235962 2023-01-23 11:55:05.963242: step: 1240/1541, loss: 0.26657047867774963 2023-01-23 11:55:06.863949: step: 1242/1541, loss: 0.006629320792853832 2023-01-23 11:55:07.770425: step: 1244/1541, loss: 0.08973272144794464 2023-01-23 11:55:08.650654: step: 1246/1541, loss: 0.0018088684882968664 2023-01-23 11:55:09.483649: step: 1248/1541, loss: 0.2699248790740967 2023-01-23 11:55:10.377987: step: 1250/1541, loss: 0.1475883275270462 2023-01-23 11:55:11.284528: step: 1252/1541, loss: 0.027236944064497948 2023-01-23 11:55:12.194454: step: 1254/1541, loss: 0.11149835586547852 2023-01-23 11:55:13.119856: step: 1256/1541, loss: 0.32044312357902527 2023-01-23 11:55:14.009217: step: 1258/1541, loss: 0.11100222915410995 2023-01-23 11:55:14.937122: step: 1260/1541, loss: 0.39095520973205566 2023-01-23 11:55:15.860141: step: 1262/1541, loss: 0.14388522505760193 2023-01-23 11:55:16.800500: step: 1264/1541, loss: 0.013965051621198654 2023-01-23 11:55:17.769597: step: 1266/1541, loss: 0.021035995334386826 2023-01-23 11:55:18.640356: step: 1268/1541, loss: 0.13453446328639984 2023-01-23 11:55:19.496455: step: 1270/1541, loss: 0.06572511047124863 2023-01-23 11:55:20.431841: step: 1272/1541, loss: 0.016658717766404152 2023-01-23 11:55:21.295804: step: 1274/1541, loss: 0.05107719078660011 2023-01-23 11:55:22.153626: step: 1276/1541, loss: 0.19683942198753357 2023-01-23 11:55:23.065083: step: 1278/1541, loss: 0.15455979108810425 2023-01-23 11:55:23.955237: step: 1280/1541, loss: 0.043329037725925446 2023-01-23 11:55:24.845823: step: 1282/1541, loss: 0.03782458230853081 2023-01-23 11:55:25.748018: step: 1284/1541, loss: 0.0027148958761245012 2023-01-23 11:55:26.620929: step: 1286/1541, loss: 0.19500580430030823 2023-01-23 11:55:27.460881: step: 1288/1541, loss: 0.0061563365161418915 2023-01-23 11:55:28.344018: step: 1290/1541, loss: 0.10661947727203369 2023-01-23 11:55:29.223158: step: 1292/1541, loss: 0.16802839934825897 2023-01-23 11:55:30.096774: step: 1294/1541, loss: 0.036215879023075104 2023-01-23 11:55:30.979381: step: 1296/1541, loss: 0.06237103044986725 2023-01-23 11:55:31.950638: step: 1298/1541, loss: 0.07765983790159225 2023-01-23 11:55:32.840521: step: 1300/1541, loss: 0.0375850647687912 2023-01-23 11:55:33.712336: step: 1302/1541, loss: 0.0883847326040268 2023-01-23 11:55:34.620891: step: 1304/1541, loss: 0.054043568670749664 2023-01-23 11:55:35.514036: step: 1306/1541, loss: 0.13906629383563995 2023-01-23 11:55:36.363398: step: 1308/1541, loss: 0.16325952112674713 2023-01-23 11:55:37.263167: step: 1310/1541, loss: 0.09826646000146866 2023-01-23 11:55:38.129948: step: 1312/1541, loss: 0.21744221448898315 2023-01-23 11:55:39.027809: step: 1314/1541, loss: 0.2203635722398758 2023-01-23 11:55:39.946953: step: 1316/1541, loss: 0.029105961322784424 2023-01-23 11:55:40.845850: step: 1318/1541, loss: 0.09270593523979187 2023-01-23 11:55:41.717440: step: 1320/1541, loss: 0.42245054244995117 2023-01-23 11:55:42.584172: step: 1322/1541, loss: 0.046810828149318695 2023-01-23 11:55:43.510516: step: 1324/1541, loss: 0.16019387543201447 2023-01-23 11:55:44.419475: step: 1326/1541, loss: 0.06663056463003159 2023-01-23 11:55:45.277074: step: 1328/1541, loss: 0.09448933601379395 2023-01-23 11:55:46.167849: step: 1330/1541, loss: 0.12883837521076202 2023-01-23 11:55:47.029044: step: 1332/1541, loss: 0.022958112880587578 2023-01-23 11:55:47.912096: step: 1334/1541, loss: 0.4147368371486664 2023-01-23 11:55:48.834973: step: 1336/1541, loss: 0.03797737509012222 2023-01-23 11:55:49.713609: step: 1338/1541, loss: 0.07899047434329987 2023-01-23 11:55:50.606376: step: 1340/1541, loss: 0.08358243107795715 2023-01-23 11:55:51.504189: step: 1342/1541, loss: 0.10770541429519653 2023-01-23 11:55:52.390193: step: 1344/1541, loss: 0.010871434584259987 2023-01-23 11:55:53.298230: step: 1346/1541, loss: 0.1878027319908142 2023-01-23 11:55:54.312273: step: 1348/1541, loss: 0.013263617642223835 2023-01-23 11:55:55.219914: step: 1350/1541, loss: 0.03175558149814606 2023-01-23 11:55:56.084122: step: 1352/1541, loss: 0.0110686756670475 2023-01-23 11:55:57.019454: step: 1354/1541, loss: 0.039899393916130066 2023-01-23 11:55:57.927480: step: 1356/1541, loss: 0.6135589480400085 2023-01-23 11:55:58.869020: step: 1358/1541, loss: 0.03471938520669937 2023-01-23 11:55:59.850447: step: 1360/1541, loss: 0.0075060478411614895 2023-01-23 11:56:00.745331: step: 1362/1541, loss: 0.16617591679096222 2023-01-23 11:56:01.704141: step: 1364/1541, loss: 0.020567988976836205 2023-01-23 11:56:02.595074: step: 1366/1541, loss: 0.05044317990541458 2023-01-23 11:56:03.494961: step: 1368/1541, loss: 0.023315686732530594 2023-01-23 11:56:04.409088: step: 1370/1541, loss: 0.10866662114858627 2023-01-23 11:56:05.311917: step: 1372/1541, loss: 0.48281827569007874 2023-01-23 11:56:06.192971: step: 1374/1541, loss: 0.018724637106060982 2023-01-23 11:56:07.044309: step: 1376/1541, loss: 0.010266623459756374 2023-01-23 11:56:07.920482: step: 1378/1541, loss: 0.581248939037323 2023-01-23 11:56:08.883633: step: 1380/1541, loss: 0.07834754139184952 2023-01-23 11:56:09.793721: step: 1382/1541, loss: 0.05980846285820007 2023-01-23 11:56:10.725248: step: 1384/1541, loss: 0.851194441318512 2023-01-23 11:56:11.641217: step: 1386/1541, loss: 0.04160846397280693 2023-01-23 11:56:12.532004: step: 1388/1541, loss: 0.05246435105800629 2023-01-23 11:56:13.373176: step: 1390/1541, loss: 0.0023514677304774523 2023-01-23 11:56:14.231640: step: 1392/1541, loss: 0.03480292856693268 2023-01-23 11:56:15.141452: step: 1394/1541, loss: 0.07366949319839478 2023-01-23 11:56:16.042904: step: 1396/1541, loss: 0.08109945058822632 2023-01-23 11:56:16.903554: step: 1398/1541, loss: 0.07102231681346893 2023-01-23 11:56:17.795138: step: 1400/1541, loss: 0.14555341005325317 2023-01-23 11:56:18.698991: step: 1402/1541, loss: 0.22094686329364777 2023-01-23 11:56:19.616053: step: 1404/1541, loss: 0.07215718924999237 2023-01-23 11:56:20.496561: step: 1406/1541, loss: 0.11400105059146881 2023-01-23 11:56:21.369399: step: 1408/1541, loss: 0.15598760545253754 2023-01-23 11:56:22.280692: step: 1410/1541, loss: 0.4174787700176239 2023-01-23 11:56:23.183953: step: 1412/1541, loss: 0.37959304451942444 2023-01-23 11:56:24.045349: step: 1414/1541, loss: 0.1715773642063141 2023-01-23 11:56:25.002129: step: 1416/1541, loss: 0.3777301609516144 2023-01-23 11:56:25.950406: step: 1418/1541, loss: 0.026524212211370468 2023-01-23 11:56:26.839805: step: 1420/1541, loss: 0.3433566987514496 2023-01-23 11:56:27.734872: step: 1422/1541, loss: 0.10545901954174042 2023-01-23 11:56:28.671922: step: 1424/1541, loss: 0.03377740830183029 2023-01-23 11:56:29.554448: step: 1426/1541, loss: 0.039010532200336456 2023-01-23 11:56:30.503839: step: 1428/1541, loss: 0.03653330355882645 2023-01-23 11:56:31.388770: step: 1430/1541, loss: 0.18061131238937378 2023-01-23 11:56:32.276305: step: 1432/1541, loss: 0.21010194718837738 2023-01-23 11:56:33.206251: step: 1434/1541, loss: 0.005643784999847412 2023-01-23 11:56:34.119377: step: 1436/1541, loss: 0.03580952063202858 2023-01-23 11:56:35.024084: step: 1438/1541, loss: 0.03701722249388695 2023-01-23 11:56:35.919163: step: 1440/1541, loss: 0.02641618810594082 2023-01-23 11:56:36.818958: step: 1442/1541, loss: 0.22929468750953674 2023-01-23 11:56:37.727807: step: 1444/1541, loss: 0.08434735238552094 2023-01-23 11:56:38.611520: step: 1446/1541, loss: 0.03742353990674019 2023-01-23 11:56:39.559087: step: 1448/1541, loss: 0.20603418350219727 2023-01-23 11:56:40.438396: step: 1450/1541, loss: 0.03476475179195404 2023-01-23 11:56:41.355168: step: 1452/1541, loss: 0.014713268727064133 2023-01-23 11:56:42.241075: step: 1454/1541, loss: 0.17965346574783325 2023-01-23 11:56:43.148384: step: 1456/1541, loss: 0.051897402852773666 2023-01-23 11:56:44.041617: step: 1458/1541, loss: 0.11810135841369629 2023-01-23 11:56:44.942076: step: 1460/1541, loss: 0.02572145126760006 2023-01-23 11:56:45.831887: step: 1462/1541, loss: 0.0670563280582428 2023-01-23 11:56:46.715051: step: 1464/1541, loss: 0.04331470653414726 2023-01-23 11:56:47.650537: step: 1466/1541, loss: 0.1703374981880188 2023-01-23 11:56:48.583936: step: 1468/1541, loss: 0.19852697849273682 2023-01-23 11:56:49.486606: step: 1470/1541, loss: 0.240616112947464 2023-01-23 11:56:50.352152: step: 1472/1541, loss: 0.5464351773262024 2023-01-23 11:56:51.263285: step: 1474/1541, loss: 0.019927844405174255 2023-01-23 11:56:52.209681: step: 1476/1541, loss: 0.026330389082431793 2023-01-23 11:56:53.130011: step: 1478/1541, loss: 0.12303094565868378 2023-01-23 11:56:54.057055: step: 1480/1541, loss: 0.01429497729986906 2023-01-23 11:56:54.970108: step: 1482/1541, loss: 0.03675069659948349 2023-01-23 11:56:55.912022: step: 1484/1541, loss: 0.1797703206539154 2023-01-23 11:56:56.782582: step: 1486/1541, loss: 0.06055309250950813 2023-01-23 11:56:57.722543: step: 1488/1541, loss: 0.06812254339456558 2023-01-23 11:56:58.674056: step: 1490/1541, loss: 0.04813367500901222 2023-01-23 11:56:59.550576: step: 1492/1541, loss: 0.2441977858543396 2023-01-23 11:57:00.497466: step: 1494/1541, loss: 0.17240214347839355 2023-01-23 11:57:01.392135: step: 1496/1541, loss: 0.10923471301794052 2023-01-23 11:57:02.279362: step: 1498/1541, loss: 0.04400259256362915 2023-01-23 11:57:03.179952: step: 1500/1541, loss: 0.05007609352469444 2023-01-23 11:57:04.094041: step: 1502/1541, loss: 0.15604281425476074 2023-01-23 11:57:05.023195: step: 1504/1541, loss: 0.3724099397659302 2023-01-23 11:57:05.874302: step: 1506/1541, loss: 0.007673637941479683 2023-01-23 11:57:06.792207: step: 1508/1541, loss: 0.4824645519256592 2023-01-23 11:57:07.705638: step: 1510/1541, loss: 0.0038275504484772682 2023-01-23 11:57:08.622495: step: 1512/1541, loss: 0.6850258111953735 2023-01-23 11:57:09.520104: step: 1514/1541, loss: 0.2277555763721466 2023-01-23 11:57:10.533795: step: 1516/1541, loss: 0.019278688356280327 2023-01-23 11:57:11.455389: step: 1518/1541, loss: 0.056487031280994415 2023-01-23 11:57:12.391902: step: 1520/1541, loss: 0.09760112315416336 2023-01-23 11:57:13.295404: step: 1522/1541, loss: 0.13011817634105682 2023-01-23 11:57:14.158523: step: 1524/1541, loss: 0.1506921350955963 2023-01-23 11:57:15.081386: step: 1526/1541, loss: 0.06952016055583954 2023-01-23 11:57:15.955948: step: 1528/1541, loss: 0.215957373380661 2023-01-23 11:57:16.883255: step: 1530/1541, loss: 0.16863836348056793 2023-01-23 11:57:17.775543: step: 1532/1541, loss: 0.010523354634642601 2023-01-23 11:57:18.589904: step: 1534/1541, loss: 0.05778353288769722 2023-01-23 11:57:19.541837: step: 1536/1541, loss: 0.0286610908806324 2023-01-23 11:57:20.460990: step: 1538/1541, loss: 0.032936811447143555 2023-01-23 11:57:21.372195: step: 1540/1541, loss: 0.14695830643177032 2023-01-23 11:57:22.257916: step: 1542/1541, loss: 0.09686138480901718 2023-01-23 11:57:23.160646: step: 1544/1541, loss: 0.016242727637290955 2023-01-23 11:57:24.061254: step: 1546/1541, loss: 0.25374269485473633 2023-01-23 11:57:24.941761: step: 1548/1541, loss: 0.1515974998474121 2023-01-23 11:57:25.848598: step: 1550/1541, loss: 0.020527416840195656 2023-01-23 11:57:26.729485: step: 1552/1541, loss: 0.4551461935043335 2023-01-23 11:57:27.615501: step: 1554/1541, loss: 0.047397345304489136 2023-01-23 11:57:28.487083: step: 1556/1541, loss: 0.19239218533039093 2023-01-23 11:57:29.415200: step: 1558/1541, loss: 0.04461223632097244 2023-01-23 11:57:30.289780: step: 1560/1541, loss: 0.08816258609294891 2023-01-23 11:57:31.173692: step: 1562/1541, loss: 0.17234177887439728 2023-01-23 11:57:32.100512: step: 1564/1541, loss: 0.11464367806911469 2023-01-23 11:57:33.007066: step: 1566/1541, loss: 0.13016781210899353 2023-01-23 11:57:33.923789: step: 1568/1541, loss: 0.1847931146621704 2023-01-23 11:57:34.850321: step: 1570/1541, loss: 0.4748336374759674 2023-01-23 11:57:35.739935: step: 1572/1541, loss: 0.04561488330364227 2023-01-23 11:57:36.633048: step: 1574/1541, loss: 0.010003527626395226 2023-01-23 11:57:37.599930: step: 1576/1541, loss: 0.10449284315109253 2023-01-23 11:57:38.475349: step: 1578/1541, loss: 0.6636273264884949 2023-01-23 11:57:39.368757: step: 1580/1541, loss: 0.003520019818097353 2023-01-23 11:57:40.237814: step: 1582/1541, loss: 0.011206545867025852 2023-01-23 11:57:41.127563: step: 1584/1541, loss: 0.00907503068447113 2023-01-23 11:57:42.015873: step: 1586/1541, loss: 0.008454113267362118 2023-01-23 11:57:42.879352: step: 1588/1541, loss: 0.08168566972017288 2023-01-23 11:57:43.745153: step: 1590/1541, loss: 0.40985727310180664 2023-01-23 11:57:44.645228: step: 1592/1541, loss: 0.024840380996465683 2023-01-23 11:57:45.579712: step: 1594/1541, loss: 0.5287293195724487 2023-01-23 11:57:46.529880: step: 1596/1541, loss: 0.12128729373216629 2023-01-23 11:57:47.449997: step: 1598/1541, loss: 0.044812873005867004 2023-01-23 11:57:48.340651: step: 1600/1541, loss: 0.05503957346081734 2023-01-23 11:57:49.181911: step: 1602/1541, loss: 0.07949269562959671 2023-01-23 11:57:50.117074: step: 1604/1541, loss: 0.048983022570610046 2023-01-23 11:57:51.081669: step: 1606/1541, loss: 0.021388521417975426 2023-01-23 11:57:51.964879: step: 1608/1541, loss: 0.17024731636047363 2023-01-23 11:57:52.841702: step: 1610/1541, loss: 0.17997287213802338 2023-01-23 11:57:53.809837: step: 1612/1541, loss: 0.02698984183371067 2023-01-23 11:57:54.779419: step: 1614/1541, loss: 0.09063898772001266 2023-01-23 11:57:55.665681: step: 1616/1541, loss: 0.11669853329658508 2023-01-23 11:57:56.546914: step: 1618/1541, loss: 0.11224672198295593 2023-01-23 11:57:57.426590: step: 1620/1541, loss: 0.13193975389003754 2023-01-23 11:57:58.306239: step: 1622/1541, loss: 0.01091798860579729 2023-01-23 11:57:59.168774: step: 1624/1541, loss: 0.2488652467727661 2023-01-23 11:58:00.061290: step: 1626/1541, loss: 0.3590756058692932 2023-01-23 11:58:01.006825: step: 1628/1541, loss: 0.0054200864396989346 2023-01-23 11:58:01.944915: step: 1630/1541, loss: 0.11056213080883026 2023-01-23 11:58:02.860493: step: 1632/1541, loss: 0.22276514768600464 2023-01-23 11:58:03.694155: step: 1634/1541, loss: 0.07364872097969055 2023-01-23 11:58:04.562766: step: 1636/1541, loss: 0.011295810341835022 2023-01-23 11:58:05.522753: step: 1638/1541, loss: 0.0538230761885643 2023-01-23 11:58:06.458443: step: 1640/1541, loss: 0.045888885855674744 2023-01-23 11:58:07.324260: step: 1642/1541, loss: 0.03803671523928642 2023-01-23 11:58:08.306389: step: 1644/1541, loss: 0.03825972601771355 2023-01-23 11:58:09.200634: step: 1646/1541, loss: 0.025145361199975014 2023-01-23 11:58:10.120402: step: 1648/1541, loss: 0.012167254462838173 2023-01-23 11:58:11.040924: step: 1650/1541, loss: 0.14652760326862335 2023-01-23 11:58:11.963950: step: 1652/1541, loss: 0.012645842507481575 2023-01-23 11:58:12.890588: step: 1654/1541, loss: 0.07935967296361923 2023-01-23 11:58:13.841456: step: 1656/1541, loss: 0.05172600969672203 2023-01-23 11:58:14.718412: step: 1658/1541, loss: 0.041305311024188995 2023-01-23 11:58:15.594415: step: 1660/1541, loss: 0.2271450161933899 2023-01-23 11:58:16.488850: step: 1662/1541, loss: 0.11574476212263107 2023-01-23 11:58:17.361413: step: 1664/1541, loss: 0.16577723622322083 2023-01-23 11:58:18.277088: step: 1666/1541, loss: 0.024252096191048622 2023-01-23 11:58:19.229470: step: 1668/1541, loss: 0.013899402692914009 2023-01-23 11:58:20.102494: step: 1670/1541, loss: 0.037113454192876816 2023-01-23 11:58:21.025351: step: 1672/1541, loss: 0.013371937908232212 2023-01-23 11:58:21.946242: step: 1674/1541, loss: 0.009596157819032669 2023-01-23 11:58:22.824689: step: 1676/1541, loss: 0.08612383902072906 2023-01-23 11:58:23.713688: step: 1678/1541, loss: 0.027307750657200813 2023-01-23 11:58:24.590644: step: 1680/1541, loss: 0.0031884824857115746 2023-01-23 11:58:25.519319: step: 1682/1541, loss: 0.506837785243988 2023-01-23 11:58:26.402692: step: 1684/1541, loss: 0.10735546052455902 2023-01-23 11:58:27.320987: step: 1686/1541, loss: 0.034872278571128845 2023-01-23 11:58:28.177138: step: 1688/1541, loss: 0.19745701551437378 2023-01-23 11:58:29.090192: step: 1690/1541, loss: 0.017671454697847366 2023-01-23 11:58:30.091514: step: 1692/1541, loss: 0.111813023686409 2023-01-23 11:58:30.969256: step: 1694/1541, loss: 0.02131194993853569 2023-01-23 11:58:31.865536: step: 1696/1541, loss: 0.002829186385497451 2023-01-23 11:58:32.766629: step: 1698/1541, loss: 0.0019188349833711982 2023-01-23 11:58:33.647647: step: 1700/1541, loss: 0.007312712725251913 2023-01-23 11:58:34.511029: step: 1702/1541, loss: 0.06458999961614609 2023-01-23 11:58:35.464379: step: 1704/1541, loss: 0.12782494723796844 2023-01-23 11:58:36.347906: step: 1706/1541, loss: 0.3020605444908142 2023-01-23 11:58:37.231843: step: 1708/1541, loss: 0.0963180661201477 2023-01-23 11:58:38.091360: step: 1710/1541, loss: 0.01214284636080265 2023-01-23 11:58:39.010107: step: 1712/1541, loss: 0.09100920706987381 2023-01-23 11:58:39.906332: step: 1714/1541, loss: 0.049852751195430756 2023-01-23 11:58:40.850655: step: 1716/1541, loss: 0.13224121928215027 2023-01-23 11:58:41.728823: step: 1718/1541, loss: 0.06215439736843109 2023-01-23 11:58:42.691868: step: 1720/1541, loss: 0.02902773581445217 2023-01-23 11:58:43.597245: step: 1722/1541, loss: 0.08324913680553436 2023-01-23 11:58:44.510521: step: 1724/1541, loss: 0.08119214326143265 2023-01-23 11:58:45.514037: step: 1726/1541, loss: 0.4501889944076538 2023-01-23 11:58:46.481082: step: 1728/1541, loss: 0.016756601631641388 2023-01-23 11:58:47.378555: step: 1730/1541, loss: 0.073664091527462 2023-01-23 11:58:48.277481: step: 1732/1541, loss: 0.8909053206443787 2023-01-23 11:58:49.181298: step: 1734/1541, loss: 0.07367309927940369 2023-01-23 11:58:50.100045: step: 1736/1541, loss: 0.11949583888053894 2023-01-23 11:58:51.002668: step: 1738/1541, loss: 0.04631811007857323 2023-01-23 11:58:51.883758: step: 1740/1541, loss: 0.3776511549949646 2023-01-23 11:58:52.805689: step: 1742/1541, loss: 0.03230562433600426 2023-01-23 11:58:53.702159: step: 1744/1541, loss: 0.08831760287284851 2023-01-23 11:58:54.632831: step: 1746/1541, loss: 0.018824944272637367 2023-01-23 11:58:55.545680: step: 1748/1541, loss: 0.4464626908302307 2023-01-23 11:58:56.471041: step: 1750/1541, loss: 0.20873293280601501 2023-01-23 11:58:57.398435: step: 1752/1541, loss: 0.02818438969552517 2023-01-23 11:58:58.253779: step: 1754/1541, loss: 0.3214849829673767 2023-01-23 11:58:59.143173: step: 1756/1541, loss: 0.3483171761035919 2023-01-23 11:59:00.038784: step: 1758/1541, loss: 0.1930607706308365 2023-01-23 11:59:00.922976: step: 1760/1541, loss: 0.046002522110939026 2023-01-23 11:59:01.869124: step: 1762/1541, loss: 0.030694901943206787 2023-01-23 11:59:02.789884: step: 1764/1541, loss: 0.4699627161026001 2023-01-23 11:59:03.666444: step: 1766/1541, loss: 0.024944981560111046 2023-01-23 11:59:04.590411: step: 1768/1541, loss: 0.03947727009654045 2023-01-23 11:59:05.456132: step: 1770/1541, loss: 0.060509294271469116 2023-01-23 11:59:06.359024: step: 1772/1541, loss: 0.14715595543384552 2023-01-23 11:59:07.307252: step: 1774/1541, loss: 0.1714000701904297 2023-01-23 11:59:08.242283: step: 1776/1541, loss: 0.10109300911426544 2023-01-23 11:59:09.120255: step: 1778/1541, loss: 0.15757298469543457 2023-01-23 11:59:10.004722: step: 1780/1541, loss: 0.06679096817970276 2023-01-23 11:59:10.863059: step: 1782/1541, loss: 0.17529404163360596 2023-01-23 11:59:11.751226: step: 1784/1541, loss: 0.012502029538154602 2023-01-23 11:59:12.670069: step: 1786/1541, loss: 0.14055708050727844 2023-01-23 11:59:13.566173: step: 1788/1541, loss: 0.21080546081066132 2023-01-23 11:59:14.445638: step: 1790/1541, loss: 0.029769495129585266 2023-01-23 11:59:15.365233: step: 1792/1541, loss: 0.28399595618247986 2023-01-23 11:59:16.278962: step: 1794/1541, loss: 0.039303094148635864 2023-01-23 11:59:17.155225: step: 1796/1541, loss: 0.05996910482645035 2023-01-23 11:59:18.069947: step: 1798/1541, loss: 0.15938083827495575 2023-01-23 11:59:18.970266: step: 1800/1541, loss: 0.023682326078414917 2023-01-23 11:59:19.899185: step: 1802/1541, loss: 0.1333131045103073 2023-01-23 11:59:20.814232: step: 1804/1541, loss: 0.1207260712981224 2023-01-23 11:59:21.693597: step: 1806/1541, loss: 0.03370305150747299 2023-01-23 11:59:22.650579: step: 1808/1541, loss: 0.13162629306316376 2023-01-23 11:59:23.525363: step: 1810/1541, loss: 0.10639514774084091 2023-01-23 11:59:24.413890: step: 1812/1541, loss: 0.31794652342796326 2023-01-23 11:59:25.302564: step: 1814/1541, loss: 0.22978149354457855 2023-01-23 11:59:26.215648: step: 1816/1541, loss: 0.2089189887046814 2023-01-23 11:59:27.084795: step: 1818/1541, loss: 0.0016471087001264095 2023-01-23 11:59:27.950911: step: 1820/1541, loss: 0.33575642108917236 2023-01-23 11:59:28.897980: step: 1822/1541, loss: 0.17391446232795715 2023-01-23 11:59:29.791220: step: 1824/1541, loss: 0.06584089994430542 2023-01-23 11:59:30.747903: step: 1826/1541, loss: 0.031229697167873383 2023-01-23 11:59:31.676765: step: 1828/1541, loss: 0.042669329792261124 2023-01-23 11:59:32.627751: step: 1830/1541, loss: 0.063878633081913 2023-01-23 11:59:33.549291: step: 1832/1541, loss: 0.03471077233552933 2023-01-23 11:59:34.515241: step: 1834/1541, loss: 0.05474235117435455 2023-01-23 11:59:35.350244: step: 1836/1541, loss: 0.07405540347099304 2023-01-23 11:59:36.198066: step: 1838/1541, loss: 0.010372478514909744 2023-01-23 11:59:37.085875: step: 1840/1541, loss: 0.1388321965932846 2023-01-23 11:59:37.942803: step: 1842/1541, loss: 0.02017500065267086 2023-01-23 11:59:38.808939: step: 1844/1541, loss: 0.46990227699279785 2023-01-23 11:59:39.677990: step: 1846/1541, loss: 0.32610297203063965 2023-01-23 11:59:40.562885: step: 1848/1541, loss: 0.04350075125694275 2023-01-23 11:59:41.468594: step: 1850/1541, loss: 0.0784759670495987 2023-01-23 11:59:42.335387: step: 1852/1541, loss: 0.026288392022252083 2023-01-23 11:59:43.238501: step: 1854/1541, loss: 0.059249259531497955 2023-01-23 11:59:44.155501: step: 1856/1541, loss: 0.3086262345314026 2023-01-23 11:59:45.017376: step: 1858/1541, loss: 0.005438880529254675 2023-01-23 11:59:46.009395: step: 1860/1541, loss: 0.01875491440296173 2023-01-23 11:59:46.921080: step: 1862/1541, loss: 1.5425775051116943 2023-01-23 11:59:47.790307: step: 1864/1541, loss: 0.02553054876625538 2023-01-23 11:59:48.693193: step: 1866/1541, loss: 0.5290684103965759 2023-01-23 11:59:49.615299: step: 1868/1541, loss: 0.010336549952626228 2023-01-23 11:59:50.495392: step: 1870/1541, loss: 0.12127823382616043 2023-01-23 11:59:51.406354: step: 1872/1541, loss: 0.2541543245315552 2023-01-23 11:59:52.345572: step: 1874/1541, loss: 0.03275036811828613 2023-01-23 11:59:53.258314: step: 1876/1541, loss: 0.13952001929283142 2023-01-23 11:59:54.192133: step: 1878/1541, loss: 1.041398525238037 2023-01-23 11:59:55.066223: step: 1880/1541, loss: 0.037000562995672226 2023-01-23 11:59:55.945384: step: 1882/1541, loss: 0.14224912226200104 2023-01-23 11:59:56.848619: step: 1884/1541, loss: 0.9937924146652222 2023-01-23 11:59:57.796109: step: 1886/1541, loss: 0.011235173791646957 2023-01-23 11:59:58.750013: step: 1888/1541, loss: 0.06404720991849899 2023-01-23 11:59:59.698210: step: 1890/1541, loss: 0.01754673197865486 2023-01-23 12:00:00.601029: step: 1892/1541, loss: 0.02583581954240799 2023-01-23 12:00:01.515554: step: 1894/1541, loss: 0.12673330307006836 2023-01-23 12:00:02.424253: step: 1896/1541, loss: 0.009300686419010162 2023-01-23 12:00:03.306336: step: 1898/1541, loss: 0.01748601347208023 2023-01-23 12:00:04.229758: step: 1900/1541, loss: 0.034041550010442734 2023-01-23 12:00:05.117264: step: 1902/1541, loss: 0.10128306597471237 2023-01-23 12:00:05.991038: step: 1904/1541, loss: 0.55990070104599 2023-01-23 12:00:06.931783: step: 1906/1541, loss: 0.24913717806339264 2023-01-23 12:00:07.790787: step: 1908/1541, loss: 0.010041802190244198 2023-01-23 12:00:08.716009: step: 1910/1541, loss: 0.03110647201538086 2023-01-23 12:00:09.609902: step: 1912/1541, loss: 0.020602097734808922 2023-01-23 12:00:10.487641: step: 1914/1541, loss: 0.12822756171226501 2023-01-23 12:00:11.377031: step: 1916/1541, loss: 0.31408244371414185 2023-01-23 12:00:12.274701: step: 1918/1541, loss: 0.01970924250781536 2023-01-23 12:00:13.153676: step: 1920/1541, loss: 0.21286924183368683 2023-01-23 12:00:14.066430: step: 1922/1541, loss: 0.14274601638317108 2023-01-23 12:00:14.934101: step: 1924/1541, loss: 0.06918949633836746 2023-01-23 12:00:15.806419: step: 1926/1541, loss: 0.08188322931528091 2023-01-23 12:00:16.713740: step: 1928/1541, loss: 0.05207543075084686 2023-01-23 12:00:17.691920: step: 1930/1541, loss: 0.08320742100477219 2023-01-23 12:00:18.592686: step: 1932/1541, loss: 0.03382580727338791 2023-01-23 12:00:19.503061: step: 1934/1541, loss: 0.10160308331251144 2023-01-23 12:00:20.390000: step: 1936/1541, loss: 0.01228929590433836 2023-01-23 12:00:21.342666: step: 1938/1541, loss: 0.02789164148271084 2023-01-23 12:00:22.292626: step: 1940/1541, loss: 0.14636875689029694 2023-01-23 12:00:23.215792: step: 1942/1541, loss: 0.024953270331025124 2023-01-23 12:00:24.121139: step: 1944/1541, loss: 0.5750260353088379 2023-01-23 12:00:25.047647: step: 1946/1541, loss: 0.006244228687137365 2023-01-23 12:00:25.991185: step: 1948/1541, loss: 0.1283167004585266 2023-01-23 12:00:26.893425: step: 1950/1541, loss: 0.008772356435656548 2023-01-23 12:00:27.762654: step: 1952/1541, loss: 0.08220914006233215 2023-01-23 12:00:28.662253: step: 1954/1541, loss: 0.03892015293240547 2023-01-23 12:00:29.567885: step: 1956/1541, loss: 0.04342474788427353 2023-01-23 12:00:30.466861: step: 1958/1541, loss: 0.099350705742836 2023-01-23 12:00:31.368330: step: 1960/1541, loss: 0.08647065609693527 2023-01-23 12:00:32.258249: step: 1962/1541, loss: 0.38426369428634644 2023-01-23 12:00:33.150187: step: 1964/1541, loss: 0.028273189440369606 2023-01-23 12:00:34.045277: step: 1966/1541, loss: 0.47877952456474304 2023-01-23 12:00:35.038132: step: 1968/1541, loss: 0.10315965861082077 2023-01-23 12:00:35.928263: step: 1970/1541, loss: 0.12765038013458252 2023-01-23 12:00:36.844808: step: 1972/1541, loss: 0.11378844082355499 2023-01-23 12:00:37.750386: step: 1974/1541, loss: 0.1760164201259613 2023-01-23 12:00:38.610751: step: 1976/1541, loss: 0.07235028594732285 2023-01-23 12:00:39.518487: step: 1978/1541, loss: 0.00335374241694808 2023-01-23 12:00:40.405505: step: 1980/1541, loss: 0.06640607863664627 2023-01-23 12:00:41.339418: step: 1982/1541, loss: 0.1440710723400116 2023-01-23 12:00:42.243918: step: 1984/1541, loss: 0.02548724040389061 2023-01-23 12:00:43.119295: step: 1986/1541, loss: 0.27843940258026123 2023-01-23 12:00:44.079605: step: 1988/1541, loss: 0.3141710162162781 2023-01-23 12:00:45.027685: step: 1990/1541, loss: 0.09582332521677017 2023-01-23 12:00:46.026366: step: 1992/1541, loss: 0.17495492100715637 2023-01-23 12:00:46.869343: step: 1994/1541, loss: 0.2912653982639313 2023-01-23 12:00:47.748562: step: 1996/1541, loss: 0.40870872139930725 2023-01-23 12:00:48.660734: step: 1998/1541, loss: 0.007179433945566416 2023-01-23 12:00:49.593703: step: 2000/1541, loss: 0.1317211240530014 2023-01-23 12:00:50.528799: step: 2002/1541, loss: 0.019593078643083572 2023-01-23 12:00:51.428979: step: 2004/1541, loss: 0.044888101518154144 2023-01-23 12:00:52.398840: step: 2006/1541, loss: 0.16483119130134583 2023-01-23 12:00:53.259128: step: 2008/1541, loss: 0.023262904956936836 2023-01-23 12:00:54.188373: step: 2010/1541, loss: 0.4012124836444855 2023-01-23 12:00:55.098192: step: 2012/1541, loss: 0.01316719688475132 2023-01-23 12:00:56.026726: step: 2014/1541, loss: 0.02647569216787815 2023-01-23 12:00:56.949866: step: 2016/1541, loss: 0.07819744199514389 2023-01-23 12:00:57.875464: step: 2018/1541, loss: 0.023482121527194977 2023-01-23 12:00:58.768989: step: 2020/1541, loss: 0.016960691660642624 2023-01-23 12:00:59.670966: step: 2022/1541, loss: 2.4758763313293457 2023-01-23 12:01:00.544617: step: 2024/1541, loss: 0.02671113610267639 2023-01-23 12:01:01.562957: step: 2026/1541, loss: 0.291235089302063 2023-01-23 12:01:02.473935: step: 2028/1541, loss: 0.0035439643543213606 2023-01-23 12:01:03.440234: step: 2030/1541, loss: 0.10628371685743332 2023-01-23 12:01:04.337194: step: 2032/1541, loss: 0.0208634864538908 2023-01-23 12:01:05.314362: step: 2034/1541, loss: 0.2699022591114044 2023-01-23 12:01:06.205775: step: 2036/1541, loss: 0.09345415979623795 2023-01-23 12:01:07.127513: step: 2038/1541, loss: 0.024850910529494286 2023-01-23 12:01:08.036803: step: 2040/1541, loss: 0.14667291939258575 2023-01-23 12:01:08.908553: step: 2042/1541, loss: 0.009425940923392773 2023-01-23 12:01:09.821409: step: 2044/1541, loss: 0.03228779509663582 2023-01-23 12:01:10.740988: step: 2046/1541, loss: 0.10464580357074738 2023-01-23 12:01:11.652864: step: 2048/1541, loss: 0.10464094579219818 2023-01-23 12:01:12.520322: step: 2050/1541, loss: 0.024019109085202217 2023-01-23 12:01:13.485184: step: 2052/1541, loss: 0.2682187855243683 2023-01-23 12:01:14.375471: step: 2054/1541, loss: 0.020164694637060165 2023-01-23 12:01:15.273376: step: 2056/1541, loss: 0.007639762945473194 2023-01-23 12:01:16.170111: step: 2058/1541, loss: 0.03724660724401474 2023-01-23 12:01:17.104901: step: 2060/1541, loss: 0.004365375731140375 2023-01-23 12:01:17.967124: step: 2062/1541, loss: 0.038315024226903915 2023-01-23 12:01:18.826965: step: 2064/1541, loss: 0.005355610512197018 2023-01-23 12:01:19.717088: step: 2066/1541, loss: 0.23892441391944885 2023-01-23 12:01:20.591229: step: 2068/1541, loss: 0.09276523441076279 2023-01-23 12:01:21.532804: step: 2070/1541, loss: 0.0674266666173935 2023-01-23 12:01:22.467874: step: 2072/1541, loss: 0.05178743973374367 2023-01-23 12:01:23.392713: step: 2074/1541, loss: 0.01716175302863121 2023-01-23 12:01:24.280736: step: 2076/1541, loss: 0.010516300797462463 2023-01-23 12:01:25.178152: step: 2078/1541, loss: 0.15256986021995544 2023-01-23 12:01:26.060107: step: 2080/1541, loss: 0.009281717240810394 2023-01-23 12:01:26.959765: step: 2082/1541, loss: 0.03161761164665222 2023-01-23 12:01:27.921602: step: 2084/1541, loss: 0.08924653381109238 2023-01-23 12:01:28.862251: step: 2086/1541, loss: 0.031036922708153725 2023-01-23 12:01:29.738056: step: 2088/1541, loss: 0.1734703779220581 2023-01-23 12:01:30.629432: step: 2090/1541, loss: 0.004187465645372868 2023-01-23 12:01:31.495416: step: 2092/1541, loss: 0.002827201271429658 2023-01-23 12:01:32.461724: step: 2094/1541, loss: 1.3176069259643555 2023-01-23 12:01:33.363768: step: 2096/1541, loss: 0.11819557100534439 2023-01-23 12:01:34.297521: step: 2098/1541, loss: 0.10942115634679794 2023-01-23 12:01:35.231736: step: 2100/1541, loss: 0.2099107950925827 2023-01-23 12:01:36.133125: step: 2102/1541, loss: 0.01963997818529606 2023-01-23 12:01:37.014911: step: 2104/1541, loss: 0.07016116380691528 2023-01-23 12:01:37.965051: step: 2106/1541, loss: 0.03433134779334068 2023-01-23 12:01:38.856886: step: 2108/1541, loss: 0.42170172929763794 2023-01-23 12:01:39.728996: step: 2110/1541, loss: 0.04611865431070328 2023-01-23 12:01:40.661430: step: 2112/1541, loss: 0.5584936738014221 2023-01-23 12:01:41.559268: step: 2114/1541, loss: 0.15325702726840973 2023-01-23 12:01:42.492384: step: 2116/1541, loss: 0.12109146267175674 2023-01-23 12:01:43.384693: step: 2118/1541, loss: 0.13812197744846344 2023-01-23 12:01:44.350742: step: 2120/1541, loss: 0.03452243655920029 2023-01-23 12:01:45.243879: step: 2122/1541, loss: 0.049146659672260284 2023-01-23 12:01:46.135855: step: 2124/1541, loss: 0.07125338912010193 2023-01-23 12:01:47.016130: step: 2126/1541, loss: 0.02268911711871624 2023-01-23 12:01:47.969226: step: 2128/1541, loss: 0.16138197481632233 2023-01-23 12:01:48.830994: step: 2130/1541, loss: 0.010923665016889572 2023-01-23 12:01:49.715699: step: 2132/1541, loss: 0.002817576052621007 2023-01-23 12:01:50.621552: step: 2134/1541, loss: 0.019930880516767502 2023-01-23 12:01:51.499637: step: 2136/1541, loss: 0.023296181112527847 2023-01-23 12:01:52.391292: step: 2138/1541, loss: 0.11749321222305298 2023-01-23 12:01:53.287715: step: 2140/1541, loss: 0.22905069589614868 2023-01-23 12:01:54.145050: step: 2142/1541, loss: 0.13084140419960022 2023-01-23 12:01:55.035933: step: 2144/1541, loss: 0.052895210683345795 2023-01-23 12:01:55.971104: step: 2146/1541, loss: 0.02069435454905033 2023-01-23 12:01:56.833737: step: 2148/1541, loss: 0.37524908781051636 2023-01-23 12:01:57.717117: step: 2150/1541, loss: 0.40474629402160645 2023-01-23 12:01:58.636056: step: 2152/1541, loss: 0.043742164969444275 2023-01-23 12:01:59.535068: step: 2154/1541, loss: 0.012030637823045254 2023-01-23 12:02:00.424522: step: 2156/1541, loss: 0.0881660133600235 2023-01-23 12:02:01.325320: step: 2158/1541, loss: 0.13641205430030823 2023-01-23 12:02:02.293176: step: 2160/1541, loss: 0.19955819845199585 2023-01-23 12:02:03.170217: step: 2162/1541, loss: 0.12661781907081604 2023-01-23 12:02:04.065512: step: 2164/1541, loss: 0.09647047519683838 2023-01-23 12:02:04.948993: step: 2166/1541, loss: 0.0561404824256897 2023-01-23 12:02:05.815628: step: 2168/1541, loss: 0.15388724207878113 2023-01-23 12:02:06.695740: step: 2170/1541, loss: 0.5708940029144287 2023-01-23 12:02:07.643308: step: 2172/1541, loss: 0.06382504850625992 2023-01-23 12:02:08.567097: step: 2174/1541, loss: 0.050611335784196854 2023-01-23 12:02:09.499011: step: 2176/1541, loss: 0.005400754977017641 2023-01-23 12:02:10.355782: step: 2178/1541, loss: 0.5671573877334595 2023-01-23 12:02:11.251264: step: 2180/1541, loss: 0.36807599663734436 2023-01-23 12:02:12.123126: step: 2182/1541, loss: 0.021916912868618965 2023-01-23 12:02:13.083258: step: 2184/1541, loss: 0.10334897041320801 2023-01-23 12:02:13.984481: step: 2186/1541, loss: 0.001932365819811821 2023-01-23 12:02:14.963749: step: 2188/1541, loss: 0.020948413759469986 2023-01-23 12:02:15.860427: step: 2190/1541, loss: 0.7393520474433899 2023-01-23 12:02:16.775391: step: 2192/1541, loss: 0.04591042548418045 2023-01-23 12:02:17.647356: step: 2194/1541, loss: 0.18572306632995605 2023-01-23 12:02:18.557436: step: 2196/1541, loss: 0.011851108632981777 2023-01-23 12:02:19.452089: step: 2198/1541, loss: 0.05952928215265274 2023-01-23 12:02:20.355044: step: 2200/1541, loss: 0.08315674215555191 2023-01-23 12:02:21.250954: step: 2202/1541, loss: 0.055982448160648346 2023-01-23 12:02:22.165898: step: 2204/1541, loss: 0.06942706555128098 2023-01-23 12:02:23.064806: step: 2206/1541, loss: 0.2859356999397278 2023-01-23 12:02:23.953300: step: 2208/1541, loss: 0.08626246452331543 2023-01-23 12:02:24.851564: step: 2210/1541, loss: 0.006992998532950878 2023-01-23 12:02:25.729049: step: 2212/1541, loss: 0.13928981125354767 2023-01-23 12:02:26.579790: step: 2214/1541, loss: 0.009718650951981544 2023-01-23 12:02:27.559190: step: 2216/1541, loss: 0.001795978401787579 2023-01-23 12:02:28.445717: step: 2218/1541, loss: 0.08365505933761597 2023-01-23 12:02:29.382885: step: 2220/1541, loss: 0.691299557685852 2023-01-23 12:02:30.256264: step: 2222/1541, loss: 0.036450404673814774 2023-01-23 12:02:31.138180: step: 2224/1541, loss: 0.0022572146262973547 2023-01-23 12:02:32.027367: step: 2226/1541, loss: 0.006940731778740883 2023-01-23 12:02:32.941981: step: 2228/1541, loss: 0.08970603346824646 2023-01-23 12:02:33.831409: step: 2230/1541, loss: 0.0017619194695726037 2023-01-23 12:02:34.717411: step: 2232/1541, loss: 0.12041757255792618 2023-01-23 12:02:35.595695: step: 2234/1541, loss: 0.09535911679267883 2023-01-23 12:02:36.539168: step: 2236/1541, loss: 0.17643937468528748 2023-01-23 12:02:37.424808: step: 2238/1541, loss: 0.15465648472309113 2023-01-23 12:02:38.357241: step: 2240/1541, loss: 0.5845192074775696 2023-01-23 12:02:39.221814: step: 2242/1541, loss: 0.00850777979940176 2023-01-23 12:02:40.079960: step: 2244/1541, loss: 0.08483689278364182 2023-01-23 12:02:40.947684: step: 2246/1541, loss: 0.27138984203338623 2023-01-23 12:02:41.836190: step: 2248/1541, loss: 0.022148724645376205 2023-01-23 12:02:42.715568: step: 2250/1541, loss: 0.023492969572544098 2023-01-23 12:02:43.606319: step: 2252/1541, loss: 0.03556077927350998 2023-01-23 12:02:44.494183: step: 2254/1541, loss: 0.19630004465579987 2023-01-23 12:02:45.406283: step: 2256/1541, loss: 0.05790349841117859 2023-01-23 12:02:46.301952: step: 2258/1541, loss: 0.08737814426422119 2023-01-23 12:02:47.227031: step: 2260/1541, loss: 0.00925775058567524 2023-01-23 12:02:48.134947: step: 2262/1541, loss: 0.18471239507198334 2023-01-23 12:02:49.013535: step: 2264/1541, loss: 0.05917886644601822 2023-01-23 12:02:49.900839: step: 2266/1541, loss: 0.26977241039276123 2023-01-23 12:02:50.823856: step: 2268/1541, loss: 0.20534484088420868 2023-01-23 12:02:51.736210: step: 2270/1541, loss: 0.012702619656920433 2023-01-23 12:02:52.636182: step: 2272/1541, loss: 0.3364814519882202 2023-01-23 12:02:53.534191: step: 2274/1541, loss: 0.08270353823900223 2023-01-23 12:02:54.544839: step: 2276/1541, loss: 1.520639181137085 2023-01-23 12:02:55.398147: step: 2278/1541, loss: 0.0019612028263509274 2023-01-23 12:02:56.292227: step: 2280/1541, loss: 0.13033723831176758 2023-01-23 12:02:57.179338: step: 2282/1541, loss: 0.061570096760988235 2023-01-23 12:02:58.119636: step: 2284/1541, loss: 0.4849884808063507 2023-01-23 12:02:58.994823: step: 2286/1541, loss: 0.10531792044639587 2023-01-23 12:02:59.879908: step: 2288/1541, loss: 0.21303334832191467 2023-01-23 12:03:00.763044: step: 2290/1541, loss: 0.012577706947922707 2023-01-23 12:03:01.625840: step: 2292/1541, loss: 0.1513131856918335 2023-01-23 12:03:02.538300: step: 2294/1541, loss: 0.9865948557853699 2023-01-23 12:03:03.517909: step: 2296/1541, loss: 0.005723548587411642 2023-01-23 12:03:04.424502: step: 2298/1541, loss: 0.09040974825620651 2023-01-23 12:03:05.310653: step: 2300/1541, loss: 0.07973994314670563 2023-01-23 12:03:06.180789: step: 2302/1541, loss: 0.18529559671878815 2023-01-23 12:03:07.076183: step: 2304/1541, loss: 0.06538482010364532 2023-01-23 12:03:07.983450: step: 2306/1541, loss: 0.01493779756128788 2023-01-23 12:03:08.866412: step: 2308/1541, loss: 0.0044568199664354324 2023-01-23 12:03:09.750529: step: 2310/1541, loss: 0.03172073885798454 2023-01-23 12:03:10.684404: step: 2312/1541, loss: 0.11282863467931747 2023-01-23 12:03:11.611748: step: 2314/1541, loss: 0.22220845520496368 2023-01-23 12:03:12.509302: step: 2316/1541, loss: 0.014538913033902645 2023-01-23 12:03:13.492534: step: 2318/1541, loss: 0.21089673042297363 2023-01-23 12:03:14.382901: step: 2320/1541, loss: 0.04897138848900795 2023-01-23 12:03:15.323469: step: 2322/1541, loss: 0.10565987229347229 2023-01-23 12:03:16.194646: step: 2324/1541, loss: 0.1763264536857605 2023-01-23 12:03:17.100756: step: 2326/1541, loss: 0.04901006072759628 2023-01-23 12:03:18.002755: step: 2328/1541, loss: 0.10490376502275467 2023-01-23 12:03:18.861933: step: 2330/1541, loss: 0.15505436062812805 2023-01-23 12:03:19.832442: step: 2332/1541, loss: 0.2782129943370819 2023-01-23 12:03:20.730969: step: 2334/1541, loss: 0.3005744516849518 2023-01-23 12:03:21.635104: step: 2336/1541, loss: 0.017246268689632416 2023-01-23 12:03:22.499151: step: 2338/1541, loss: 0.015488365665078163 2023-01-23 12:03:23.361746: step: 2340/1541, loss: 0.12572553753852844 2023-01-23 12:03:24.222765: step: 2342/1541, loss: 0.07613480091094971 2023-01-23 12:03:25.106434: step: 2344/1541, loss: 0.2745489180088043 2023-01-23 12:03:25.965321: step: 2346/1541, loss: 0.12758077681064606 2023-01-23 12:03:26.839735: step: 2348/1541, loss: 0.0656246542930603 2023-01-23 12:03:27.786930: step: 2350/1541, loss: 0.3565385341644287 2023-01-23 12:03:28.646021: step: 2352/1541, loss: 0.06661813706159592 2023-01-23 12:03:29.691482: step: 2354/1541, loss: 0.05561681464314461 2023-01-23 12:03:30.602053: step: 2356/1541, loss: 0.26988455653190613 2023-01-23 12:03:31.542586: step: 2358/1541, loss: 0.044269122183322906 2023-01-23 12:03:32.536052: step: 2360/1541, loss: 0.21023307740688324 2023-01-23 12:03:33.411043: step: 2362/1541, loss: 0.11974865198135376 2023-01-23 12:03:34.330324: step: 2364/1541, loss: 0.13114768266677856 2023-01-23 12:03:35.238989: step: 2366/1541, loss: 0.06878376752138138 2023-01-23 12:03:36.129030: step: 2368/1541, loss: 0.03738733008503914 2023-01-23 12:03:37.042810: step: 2370/1541, loss: 0.07551084458827972 2023-01-23 12:03:37.951537: step: 2372/1541, loss: 0.021635187789797783 2023-01-23 12:03:38.891928: step: 2374/1541, loss: 0.035501375794410706 2023-01-23 12:03:39.798036: step: 2376/1541, loss: 0.4852328598499298 2023-01-23 12:03:40.687259: step: 2378/1541, loss: 0.12661738693714142 2023-01-23 12:03:41.594836: step: 2380/1541, loss: 0.11264916509389877 2023-01-23 12:03:42.515904: step: 2382/1541, loss: 0.10712683945894241 2023-01-23 12:03:43.406345: step: 2384/1541, loss: 0.08209527283906937 2023-01-23 12:03:44.310939: step: 2386/1541, loss: 0.1620466262102127 2023-01-23 12:03:45.267256: step: 2388/1541, loss: 0.2627256512641907 2023-01-23 12:03:46.164798: step: 2390/1541, loss: 0.15739302337169647 2023-01-23 12:03:47.093577: step: 2392/1541, loss: 0.02562168426811695 2023-01-23 12:03:47.977193: step: 2394/1541, loss: 0.009691375307738781 2023-01-23 12:03:48.880870: step: 2396/1541, loss: 0.12838877737522125 2023-01-23 12:03:49.727378: step: 2398/1541, loss: 0.04429345577955246 2023-01-23 12:03:50.633492: step: 2400/1541, loss: 0.031104527413845062 2023-01-23 12:03:51.493272: step: 2402/1541, loss: 0.019970189779996872 2023-01-23 12:03:52.351218: step: 2404/1541, loss: 0.3110894560813904 2023-01-23 12:03:53.273245: step: 2406/1541, loss: 0.015956170856952667 2023-01-23 12:03:54.142901: step: 2408/1541, loss: 0.018242286518216133 2023-01-23 12:03:55.029367: step: 2410/1541, loss: 0.24778392910957336 2023-01-23 12:03:55.943522: step: 2412/1541, loss: 0.060399964451789856 2023-01-23 12:03:56.891399: step: 2414/1541, loss: 0.05779462680220604 2023-01-23 12:03:57.781937: step: 2416/1541, loss: 0.09151080995798111 2023-01-23 12:03:58.683897: step: 2418/1541, loss: 0.005770559422671795 2023-01-23 12:03:59.579034: step: 2420/1541, loss: 0.1599704623222351 2023-01-23 12:04:00.436636: step: 2422/1541, loss: 0.27950841188430786 2023-01-23 12:04:01.345087: step: 2424/1541, loss: 0.11787085980176926 2023-01-23 12:04:02.295725: step: 2426/1541, loss: 0.059227585792541504 2023-01-23 12:04:03.224017: step: 2428/1541, loss: 0.4585672914981842 2023-01-23 12:04:04.116836: step: 2430/1541, loss: 0.21520791947841644 2023-01-23 12:04:05.010020: step: 2432/1541, loss: 0.049998361617326736 2023-01-23 12:04:05.895796: step: 2434/1541, loss: 0.05185048654675484 2023-01-23 12:04:06.804222: step: 2436/1541, loss: 0.02340824156999588 2023-01-23 12:04:07.692080: step: 2438/1541, loss: 0.11334141343832016 2023-01-23 12:04:08.630229: step: 2440/1541, loss: 0.13485866785049438 2023-01-23 12:04:09.518451: step: 2442/1541, loss: 0.1440737396478653 2023-01-23 12:04:10.406509: step: 2444/1541, loss: 0.16599595546722412 2023-01-23 12:04:11.275935: step: 2446/1541, loss: 0.08887434750795364 2023-01-23 12:04:12.146398: step: 2448/1541, loss: 0.07471323758363724 2023-01-23 12:04:13.087045: step: 2450/1541, loss: 0.021178562194108963 2023-01-23 12:04:14.022595: step: 2452/1541, loss: 0.10486745834350586 2023-01-23 12:04:14.958035: step: 2454/1541, loss: 0.1142020970582962 2023-01-23 12:04:15.858858: step: 2456/1541, loss: 0.010038899257779121 2023-01-23 12:04:16.757318: step: 2458/1541, loss: 0.012474187649786472 2023-01-23 12:04:17.702957: step: 2460/1541, loss: 0.06199609860777855 2023-01-23 12:04:18.638770: step: 2462/1541, loss: 0.4148786962032318 2023-01-23 12:04:19.506979: step: 2464/1541, loss: 0.0199104156345129 2023-01-23 12:04:20.456604: step: 2466/1541, loss: 0.049666740000247955 2023-01-23 12:04:21.362732: step: 2468/1541, loss: 0.08405421674251556 2023-01-23 12:04:22.233567: step: 2470/1541, loss: 0.27945664525032043 2023-01-23 12:04:23.158490: step: 2472/1541, loss: 0.0670793429017067 2023-01-23 12:04:24.032765: step: 2474/1541, loss: 0.064456045627594 2023-01-23 12:04:24.938659: step: 2476/1541, loss: 0.09822428226470947 2023-01-23 12:04:25.880743: step: 2478/1541, loss: 0.1439501792192459 2023-01-23 12:04:26.740588: step: 2480/1541, loss: 0.3475129306316376 2023-01-23 12:04:27.651639: step: 2482/1541, loss: 0.11059410870075226 2023-01-23 12:04:28.511837: step: 2484/1541, loss: 0.13727827370166779 2023-01-23 12:04:29.415860: step: 2486/1541, loss: 0.06049419566988945 2023-01-23 12:04:30.310013: step: 2488/1541, loss: 0.08317592740058899 2023-01-23 12:04:31.222635: step: 2490/1541, loss: 0.08990170061588287 2023-01-23 12:04:32.086599: step: 2492/1541, loss: 0.13098950684070587 2023-01-23 12:04:33.042164: step: 2494/1541, loss: 0.42957839369773865 2023-01-23 12:04:33.977517: step: 2496/1541, loss: 0.12540635466575623 2023-01-23 12:04:34.878719: step: 2498/1541, loss: 0.24119293689727783 2023-01-23 12:04:35.790626: step: 2500/1541, loss: 0.04840554669499397 2023-01-23 12:04:36.684465: step: 2502/1541, loss: 0.021085208281874657 2023-01-23 12:04:37.562503: step: 2504/1541, loss: 0.011247174814343452 2023-01-23 12:04:38.472202: step: 2506/1541, loss: 0.01728341542184353 2023-01-23 12:04:39.385550: step: 2508/1541, loss: 0.17871145904064178 2023-01-23 12:04:40.302236: step: 2510/1541, loss: 0.016442731022834778 2023-01-23 12:04:41.206475: step: 2512/1541, loss: 0.09324923157691956 2023-01-23 12:04:42.081066: step: 2514/1541, loss: 0.020553074777126312 2023-01-23 12:04:43.042442: step: 2516/1541, loss: 0.07970597594976425 2023-01-23 12:04:43.936818: step: 2518/1541, loss: 0.03767699748277664 2023-01-23 12:04:44.822700: step: 2520/1541, loss: 0.01794586516916752 2023-01-23 12:04:45.702035: step: 2522/1541, loss: 0.09157893806695938 2023-01-23 12:04:46.581741: step: 2524/1541, loss: 0.05729422718286514 2023-01-23 12:04:47.447818: step: 2526/1541, loss: 0.010438635013997555 2023-01-23 12:04:48.354469: step: 2528/1541, loss: 0.08710533380508423 2023-01-23 12:04:49.241182: step: 2530/1541, loss: 0.11596661806106567 2023-01-23 12:04:50.107871: step: 2532/1541, loss: 0.23727092146873474 2023-01-23 12:04:51.024486: step: 2534/1541, loss: 0.08823160082101822 2023-01-23 12:04:51.955951: step: 2536/1541, loss: 0.07567261159420013 2023-01-23 12:04:52.835584: step: 2538/1541, loss: 0.07817392796278 2023-01-23 12:04:53.790049: step: 2540/1541, loss: 0.042139578610658646 2023-01-23 12:04:54.711004: step: 2542/1541, loss: 0.010561853647232056 2023-01-23 12:04:55.617062: step: 2544/1541, loss: 0.22712258994579315 2023-01-23 12:04:56.507655: step: 2546/1541, loss: 0.26353684067726135 2023-01-23 12:04:57.426181: step: 2548/1541, loss: 0.04387572407722473 2023-01-23 12:04:58.312566: step: 2550/1541, loss: 0.5621964335441589 2023-01-23 12:04:59.162162: step: 2552/1541, loss: 0.4853934645652771 2023-01-23 12:05:00.095396: step: 2554/1541, loss: 0.02656835876405239 2023-01-23 12:05:01.008955: step: 2556/1541, loss: 0.07976377010345459 2023-01-23 12:05:01.957145: step: 2558/1541, loss: 0.126473069190979 2023-01-23 12:05:02.846990: step: 2560/1541, loss: 0.09166957437992096 2023-01-23 12:05:03.743066: step: 2562/1541, loss: 0.04972298815846443 2023-01-23 12:05:04.631068: step: 2564/1541, loss: 0.008494789712131023 2023-01-23 12:05:05.512883: step: 2566/1541, loss: 0.04356156289577484 2023-01-23 12:05:06.394229: step: 2568/1541, loss: 0.14626969397068024 2023-01-23 12:05:07.258751: step: 2570/1541, loss: 0.07000089436769485 2023-01-23 12:05:08.184067: step: 2572/1541, loss: 0.018559657037258148 2023-01-23 12:05:09.093777: step: 2574/1541, loss: 0.015468243509531021 2023-01-23 12:05:09.976161: step: 2576/1541, loss: 0.023623650893568993 2023-01-23 12:05:10.927421: step: 2578/1541, loss: 0.0916057601571083 2023-01-23 12:05:11.870758: step: 2580/1541, loss: 0.2420779913663864 2023-01-23 12:05:12.764132: step: 2582/1541, loss: 0.09295205026865005 2023-01-23 12:05:13.693394: step: 2584/1541, loss: 0.38680315017700195 2023-01-23 12:05:14.618326: step: 2586/1541, loss: 0.09023603796958923 2023-01-23 12:05:15.528842: step: 2588/1541, loss: 0.014799116179347038 2023-01-23 12:05:16.451033: step: 2590/1541, loss: 0.03010929562151432 2023-01-23 12:05:17.375136: step: 2592/1541, loss: 0.005603988189250231 2023-01-23 12:05:18.272432: step: 2594/1541, loss: 0.05059821531176567 2023-01-23 12:05:19.152837: step: 2596/1541, loss: 0.027062255889177322 2023-01-23 12:05:20.015337: step: 2598/1541, loss: 0.01360902190208435 2023-01-23 12:05:20.906793: step: 2600/1541, loss: 0.05394749715924263 2023-01-23 12:05:21.772914: step: 2602/1541, loss: 0.201163649559021 2023-01-23 12:05:22.632859: step: 2604/1541, loss: 0.023244647309184074 2023-01-23 12:05:23.494612: step: 2606/1541, loss: 0.002218524692580104 2023-01-23 12:05:24.381262: step: 2608/1541, loss: 0.03599932789802551 2023-01-23 12:05:25.267921: step: 2610/1541, loss: 0.1021365374326706 2023-01-23 12:05:26.139536: step: 2612/1541, loss: 0.13866904377937317 2023-01-23 12:05:27.066107: step: 2614/1541, loss: 0.013632526621222496 2023-01-23 12:05:28.024256: step: 2616/1541, loss: 0.27469387650489807 2023-01-23 12:05:28.946715: step: 2618/1541, loss: 0.161576047539711 2023-01-23 12:05:29.895559: step: 2620/1541, loss: 0.027328457683324814 2023-01-23 12:05:30.827230: step: 2622/1541, loss: 0.2278553694486618 2023-01-23 12:05:31.669784: step: 2624/1541, loss: 0.005104795563966036 2023-01-23 12:05:32.591820: step: 2626/1541, loss: 0.3831213414669037 2023-01-23 12:05:33.491997: step: 2628/1541, loss: 0.1769002377986908 2023-01-23 12:05:34.448806: step: 2630/1541, loss: 0.22228233516216278 2023-01-23 12:05:35.348550: step: 2632/1541, loss: 0.19987422227859497 2023-01-23 12:05:36.312603: step: 2634/1541, loss: 0.04830208048224449 2023-01-23 12:05:37.218781: step: 2636/1541, loss: 0.905610978603363 2023-01-23 12:05:38.182500: step: 2638/1541, loss: 0.03566622734069824 2023-01-23 12:05:39.062070: step: 2640/1541, loss: 0.010778472758829594 2023-01-23 12:05:39.973143: step: 2642/1541, loss: 0.14784230291843414 2023-01-23 12:05:40.884679: step: 2644/1541, loss: 0.0723516047000885 2023-01-23 12:05:41.760241: step: 2646/1541, loss: 0.48528197407722473 2023-01-23 12:05:42.742323: step: 2648/1541, loss: 0.007067437749356031 2023-01-23 12:05:43.661264: step: 2650/1541, loss: 0.08988397568464279 2023-01-23 12:05:44.544042: step: 2652/1541, loss: 0.03463256359100342 2023-01-23 12:05:45.436119: step: 2654/1541, loss: 0.2481924593448639 2023-01-23 12:05:46.311034: step: 2656/1541, loss: 0.0911882147192955 2023-01-23 12:05:47.171522: step: 2658/1541, loss: 0.36611878871917725 2023-01-23 12:05:48.057960: step: 2660/1541, loss: 1.0480751991271973 2023-01-23 12:05:48.938249: step: 2662/1541, loss: 0.12947577238082886 2023-01-23 12:05:49.854439: step: 2664/1541, loss: 0.12195230275392532 2023-01-23 12:05:50.741609: step: 2666/1541, loss: 0.2938801050186157 2023-01-23 12:05:51.660098: step: 2668/1541, loss: 1.6989508867263794 2023-01-23 12:05:52.560895: step: 2670/1541, loss: 0.03710160031914711 2023-01-23 12:05:53.416696: step: 2672/1541, loss: 0.1296013444662094 2023-01-23 12:05:54.342487: step: 2674/1541, loss: 0.21460804343223572 2023-01-23 12:05:55.253737: step: 2676/1541, loss: 0.2656828761100769 2023-01-23 12:05:56.269825: step: 2678/1541, loss: 1.0215758085250854 2023-01-23 12:05:57.279120: step: 2680/1541, loss: 0.581135630607605 2023-01-23 12:05:58.148344: step: 2682/1541, loss: 0.10365764051675797 2023-01-23 12:05:59.025829: step: 2684/1541, loss: 0.527000367641449 2023-01-23 12:05:59.898466: step: 2686/1541, loss: 0.05387770012021065 2023-01-23 12:06:00.827865: step: 2688/1541, loss: 0.09320622682571411 2023-01-23 12:06:01.720628: step: 2690/1541, loss: 0.03420691192150116 2023-01-23 12:06:02.641904: step: 2692/1541, loss: 0.04867972433567047 2023-01-23 12:06:03.526736: step: 2694/1541, loss: 0.2728825807571411 2023-01-23 12:06:04.387937: step: 2696/1541, loss: 0.11510562896728516 2023-01-23 12:06:05.296492: step: 2698/1541, loss: 0.015484726056456566 2023-01-23 12:06:06.250572: step: 2700/1541, loss: 0.1257568597793579 2023-01-23 12:06:07.178210: step: 2702/1541, loss: 0.2468070536851883 2023-01-23 12:06:08.086530: step: 2704/1541, loss: 0.537996768951416 2023-01-23 12:06:08.957784: step: 2706/1541, loss: 0.06395864486694336 2023-01-23 12:06:09.821691: step: 2708/1541, loss: 0.04544796422123909 2023-01-23 12:06:10.751313: step: 2710/1541, loss: 0.20088745653629303 2023-01-23 12:06:11.631733: step: 2712/1541, loss: 0.01573304273188114 2023-01-23 12:06:12.545238: step: 2714/1541, loss: 0.3252909183502197 2023-01-23 12:06:13.446448: step: 2716/1541, loss: 0.16854727268218994 2023-01-23 12:06:14.334888: step: 2718/1541, loss: 0.2181726098060608 2023-01-23 12:06:15.204564: step: 2720/1541, loss: 0.021871432662010193 2023-01-23 12:06:16.083580: step: 2722/1541, loss: 0.23768572509288788 2023-01-23 12:06:16.987857: step: 2724/1541, loss: 0.05723194032907486 2023-01-23 12:06:17.893953: step: 2726/1541, loss: 0.03693098947405815 2023-01-23 12:06:18.790945: step: 2728/1541, loss: 0.6316779255867004 2023-01-23 12:06:19.661205: step: 2730/1541, loss: 0.1369178742170334 2023-01-23 12:06:20.663638: step: 2732/1541, loss: 0.03741960972547531 2023-01-23 12:06:21.642921: step: 2734/1541, loss: 0.26195114850997925 2023-01-23 12:06:22.533003: step: 2736/1541, loss: 0.24582986533641815 2023-01-23 12:06:23.417031: step: 2738/1541, loss: 0.17319104075431824 2023-01-23 12:06:24.330746: step: 2740/1541, loss: 0.0961846336722374 2023-01-23 12:06:25.209297: step: 2742/1541, loss: 0.15694300830364227 2023-01-23 12:06:26.126929: step: 2744/1541, loss: 0.02073586732149124 2023-01-23 12:06:27.029491: step: 2746/1541, loss: 0.1813327968120575 2023-01-23 12:06:27.878014: step: 2748/1541, loss: 0.056621260941028595 2023-01-23 12:06:28.770732: step: 2750/1541, loss: 0.4214448928833008 2023-01-23 12:06:29.657514: step: 2752/1541, loss: 0.16832879185676575 2023-01-23 12:06:30.565333: step: 2754/1541, loss: 0.03068852610886097 2023-01-23 12:06:31.434326: step: 2756/1541, loss: 0.09300405532121658 2023-01-23 12:06:32.344177: step: 2758/1541, loss: 0.20874762535095215 2023-01-23 12:06:33.257168: step: 2760/1541, loss: 0.052824247628450394 2023-01-23 12:06:34.176327: step: 2762/1541, loss: 0.12717404961585999 2023-01-23 12:06:35.068978: step: 2764/1541, loss: 0.04760178178548813 2023-01-23 12:06:36.005805: step: 2766/1541, loss: 0.21058747172355652 2023-01-23 12:06:36.849159: step: 2768/1541, loss: 0.2630040645599365 2023-01-23 12:06:37.749573: step: 2770/1541, loss: 0.6912291049957275 2023-01-23 12:06:38.652632: step: 2772/1541, loss: 0.247049942612648 2023-01-23 12:06:39.606803: step: 2774/1541, loss: 0.5440865159034729 2023-01-23 12:06:40.454294: step: 2776/1541, loss: 0.022970013320446014 2023-01-23 12:06:41.374462: step: 2778/1541, loss: 0.11322925984859467 2023-01-23 12:06:42.288249: step: 2780/1541, loss: 0.12119120359420776 2023-01-23 12:06:43.243595: step: 2782/1541, loss: 0.011212754063308239 2023-01-23 12:06:44.096022: step: 2784/1541, loss: 0.07564745098352432 2023-01-23 12:06:44.934405: step: 2786/1541, loss: 0.28474143147468567 2023-01-23 12:06:45.894283: step: 2788/1541, loss: 0.05257272347807884 2023-01-23 12:06:46.767913: step: 2790/1541, loss: 0.35125619173049927 2023-01-23 12:06:47.669261: step: 2792/1541, loss: 0.5399719476699829 2023-01-23 12:06:48.511803: step: 2794/1541, loss: 0.021732166409492493 2023-01-23 12:06:49.395446: step: 2796/1541, loss: 0.13878238201141357 2023-01-23 12:06:50.268548: step: 2798/1541, loss: 0.026605647057294846 2023-01-23 12:06:51.217767: step: 2800/1541, loss: 0.046408992260694504 2023-01-23 12:06:52.145552: step: 2802/1541, loss: 0.12578800320625305 2023-01-23 12:06:53.007751: step: 2804/1541, loss: 0.030430380254983902 2023-01-23 12:06:53.879473: step: 2806/1541, loss: 1.0934439897537231 2023-01-23 12:06:54.776950: step: 2808/1541, loss: 0.08925390988588333 2023-01-23 12:06:55.680128: step: 2810/1541, loss: 0.06779981404542923 2023-01-23 12:06:56.523674: step: 2812/1541, loss: 0.03454230725765228 2023-01-23 12:06:57.451752: step: 2814/1541, loss: 0.3269170820713043 2023-01-23 12:06:58.342164: step: 2816/1541, loss: 0.35199984908103943 2023-01-23 12:06:59.214417: step: 2818/1541, loss: 0.006365895736962557 2023-01-23 12:07:00.144150: step: 2820/1541, loss: 0.13017027080059052 2023-01-23 12:07:01.045228: step: 2822/1541, loss: 0.07927341759204865 2023-01-23 12:07:01.922353: step: 2824/1541, loss: 0.020901618525385857 2023-01-23 12:07:02.822820: step: 2826/1541, loss: 0.11883427202701569 2023-01-23 12:07:03.807018: step: 2828/1541, loss: 0.07734061777591705 2023-01-23 12:07:04.712840: step: 2830/1541, loss: 0.08020929992198944 2023-01-23 12:07:05.556811: step: 2832/1541, loss: 0.7721551060676575 2023-01-23 12:07:06.420919: step: 2834/1541, loss: 0.1296740174293518 2023-01-23 12:07:07.385017: step: 2836/1541, loss: 0.21210436522960663 2023-01-23 12:07:08.281308: step: 2838/1541, loss: 0.0238115843385458 2023-01-23 12:07:09.190455: step: 2840/1541, loss: 0.2206343412399292 2023-01-23 12:07:10.095402: step: 2842/1541, loss: 0.022284291684627533 2023-01-23 12:07:10.995459: step: 2844/1541, loss: 0.014227368868887424 2023-01-23 12:07:11.897649: step: 2846/1541, loss: 0.12686854600906372 2023-01-23 12:07:12.765814: step: 2848/1541, loss: 0.052797745913267136 2023-01-23 12:07:13.672617: step: 2850/1541, loss: 0.010827369056642056 2023-01-23 12:07:14.545797: step: 2852/1541, loss: 0.41396015882492065 2023-01-23 12:07:15.403642: step: 2854/1541, loss: 0.0527007170021534 2023-01-23 12:07:16.261765: step: 2856/1541, loss: 0.0794353038072586 2023-01-23 12:07:17.224682: step: 2858/1541, loss: 0.015352902933955193 2023-01-23 12:07:18.129160: step: 2860/1541, loss: 0.004947802983224392 2023-01-23 12:07:19.088454: step: 2862/1541, loss: 0.01979389414191246 2023-01-23 12:07:19.980112: step: 2864/1541, loss: 0.13338208198547363 2023-01-23 12:07:20.876106: step: 2866/1541, loss: 0.09400129318237305 2023-01-23 12:07:21.806050: step: 2868/1541, loss: 1.1074849367141724 2023-01-23 12:07:22.670939: step: 2870/1541, loss: 0.0707223042845726 2023-01-23 12:07:23.583957: step: 2872/1541, loss: 0.03405968099832535 2023-01-23 12:07:24.520503: step: 2874/1541, loss: 0.08503363281488419 2023-01-23 12:07:25.448167: step: 2876/1541, loss: 0.05992406606674194 2023-01-23 12:07:26.366712: step: 2878/1541, loss: 0.12176819890737534 2023-01-23 12:07:27.263260: step: 2880/1541, loss: 0.03192483261227608 2023-01-23 12:07:28.171037: step: 2882/1541, loss: 0.5990725755691528 2023-01-23 12:07:29.039471: step: 2884/1541, loss: 0.1022069901227951 2023-01-23 12:07:29.925153: step: 2886/1541, loss: 0.05009637773036957 2023-01-23 12:07:30.859697: step: 2888/1541, loss: 0.27236407995224 2023-01-23 12:07:31.756853: step: 2890/1541, loss: 0.09951487928628922 2023-01-23 12:07:32.623789: step: 2892/1541, loss: 0.033675163984298706 2023-01-23 12:07:33.540159: step: 2894/1541, loss: 0.10635956376791 2023-01-23 12:07:34.434873: step: 2896/1541, loss: 0.03235364332795143 2023-01-23 12:07:35.363112: step: 2898/1541, loss: 0.0361311100423336 2023-01-23 12:07:36.299709: step: 2900/1541, loss: 0.17947286367416382 2023-01-23 12:07:37.219024: step: 2902/1541, loss: 0.043251682072877884 2023-01-23 12:07:38.091540: step: 2904/1541, loss: 0.21681657433509827 2023-01-23 12:07:39.042051: step: 2906/1541, loss: 0.14775872230529785 2023-01-23 12:07:39.941321: step: 2908/1541, loss: 0.020805910229682922 2023-01-23 12:07:40.821915: step: 2910/1541, loss: 0.019423160701990128 2023-01-23 12:07:41.727253: step: 2912/1541, loss: 0.05910489335656166 2023-01-23 12:07:42.607610: step: 2914/1541, loss: 0.08718772232532501 2023-01-23 12:07:43.511125: step: 2916/1541, loss: 0.03176022693514824 2023-01-23 12:07:44.445716: step: 2918/1541, loss: 0.03124554082751274 2023-01-23 12:07:45.380561: step: 2920/1541, loss: 0.4299192726612091 2023-01-23 12:07:46.289893: step: 2922/1541, loss: 0.0371151901781559 2023-01-23 12:07:47.180274: step: 2924/1541, loss: 0.03375839442014694 2023-01-23 12:07:48.122687: step: 2926/1541, loss: 0.07979563623666763 2023-01-23 12:07:49.029390: step: 2928/1541, loss: 0.019702838733792305 2023-01-23 12:07:49.943569: step: 2930/1541, loss: 0.06253411620855331 2023-01-23 12:07:50.859890: step: 2932/1541, loss: 0.024401068687438965 2023-01-23 12:07:51.723649: step: 2934/1541, loss: 0.07475457340478897 2023-01-23 12:07:52.622165: step: 2936/1541, loss: 0.02805565670132637 2023-01-23 12:07:53.532500: step: 2938/1541, loss: 0.06975866854190826 2023-01-23 12:07:54.423719: step: 2940/1541, loss: 0.02899869717657566 2023-01-23 12:07:55.315110: step: 2942/1541, loss: 0.02726111188530922 2023-01-23 12:07:56.207044: step: 2944/1541, loss: 0.011790474876761436 2023-01-23 12:07:57.093675: step: 2946/1541, loss: 0.10858668386936188 2023-01-23 12:07:58.000617: step: 2948/1541, loss: 0.2090202420949936 2023-01-23 12:07:58.918813: step: 2950/1541, loss: 0.08393290638923645 2023-01-23 12:07:59.817510: step: 2952/1541, loss: 0.18179023265838623 2023-01-23 12:08:00.733510: step: 2954/1541, loss: 0.0056620510295033455 2023-01-23 12:08:01.616136: step: 2956/1541, loss: 0.22212664783000946 2023-01-23 12:08:02.571243: step: 2958/1541, loss: 0.010751430876553059 2023-01-23 12:08:03.483634: step: 2960/1541, loss: 0.21391870081424713 2023-01-23 12:08:04.360284: step: 2962/1541, loss: 0.38340312242507935 2023-01-23 12:08:05.248954: step: 2964/1541, loss: 0.14395056664943695 2023-01-23 12:08:06.145246: step: 2966/1541, loss: 0.041438985615968704 2023-01-23 12:08:07.162123: step: 2968/1541, loss: 0.14636972546577454 2023-01-23 12:08:08.093382: step: 2970/1541, loss: 0.31845325231552124 2023-01-23 12:08:08.967669: step: 2972/1541, loss: 0.11547759920358658 2023-01-23 12:08:09.829931: step: 2974/1541, loss: 0.04961874336004257 2023-01-23 12:08:10.742558: step: 2976/1541, loss: 0.1321425586938858 2023-01-23 12:08:11.687119: step: 2978/1541, loss: 0.04523526877164841 2023-01-23 12:08:12.570210: step: 2980/1541, loss: 0.2070092260837555 2023-01-23 12:08:13.440588: step: 2982/1541, loss: 0.07539255172014236 2023-01-23 12:08:14.324159: step: 2984/1541, loss: 0.22200952470302582 2023-01-23 12:08:15.297371: step: 2986/1541, loss: 0.023260192945599556 2023-01-23 12:08:16.166041: step: 2988/1541, loss: 0.14246949553489685 2023-01-23 12:08:17.067493: step: 2990/1541, loss: 0.35868382453918457 2023-01-23 12:08:17.998900: step: 2992/1541, loss: 0.09906183928251266 2023-01-23 12:08:18.930862: step: 2994/1541, loss: 0.31545209884643555 2023-01-23 12:08:19.833781: step: 2996/1541, loss: 0.01933179795742035 2023-01-23 12:08:20.738286: step: 2998/1541, loss: 0.2051047682762146 2023-01-23 12:08:21.629754: step: 3000/1541, loss: 0.011605973355472088 2023-01-23 12:08:22.523774: step: 3002/1541, loss: 0.0658191591501236 2023-01-23 12:08:23.422617: step: 3004/1541, loss: 0.0562937967479229 2023-01-23 12:08:24.301868: step: 3006/1541, loss: 0.10175832360982895 2023-01-23 12:08:25.188323: step: 3008/1541, loss: 0.09325538575649261 2023-01-23 12:08:26.102590: step: 3010/1541, loss: 0.006991176400333643 2023-01-23 12:08:27.016177: step: 3012/1541, loss: 0.27860334515571594 2023-01-23 12:08:27.934443: step: 3014/1541, loss: 0.2682133615016937 2023-01-23 12:08:28.813462: step: 3016/1541, loss: 0.3697226047515869 2023-01-23 12:08:29.706778: step: 3018/1541, loss: 0.27764180302619934 2023-01-23 12:08:30.666167: step: 3020/1541, loss: 0.17269693315029144 2023-01-23 12:08:31.563195: step: 3022/1541, loss: 0.0038727205246686935 2023-01-23 12:08:32.446768: step: 3024/1541, loss: 0.03954947367310524 2023-01-23 12:08:33.307622: step: 3026/1541, loss: 0.10584580153226852 2023-01-23 12:08:34.229092: step: 3028/1541, loss: 0.14054617285728455 2023-01-23 12:08:35.115296: step: 3030/1541, loss: 0.020366471260786057 2023-01-23 12:08:35.964803: step: 3032/1541, loss: 0.3220168650150299 2023-01-23 12:08:36.857180: step: 3034/1541, loss: 0.09240230917930603 2023-01-23 12:08:37.731194: step: 3036/1541, loss: 0.11955010145902634 2023-01-23 12:08:38.585865: step: 3038/1541, loss: 0.007454953156411648 2023-01-23 12:08:39.500288: step: 3040/1541, loss: 0.008027126081287861 2023-01-23 12:08:40.388535: step: 3042/1541, loss: 0.008207247592508793 2023-01-23 12:08:41.273243: step: 3044/1541, loss: 2.8115291595458984 2023-01-23 12:08:42.140068: step: 3046/1541, loss: 0.044054869562387466 2023-01-23 12:08:43.097912: step: 3048/1541, loss: 0.11175432801246643 2023-01-23 12:08:43.984444: step: 3050/1541, loss: 0.016577623784542084 2023-01-23 12:08:44.938848: step: 3052/1541, loss: 0.18633101880550385 2023-01-23 12:08:45.830780: step: 3054/1541, loss: 0.023931998759508133 2023-01-23 12:08:46.786876: step: 3056/1541, loss: 2.2796993255615234 2023-01-23 12:08:47.696519: step: 3058/1541, loss: 0.023066021502017975 2023-01-23 12:08:48.627912: step: 3060/1541, loss: 0.16293813288211823 2023-01-23 12:08:49.557080: step: 3062/1541, loss: 0.09852012991905212 2023-01-23 12:08:50.487667: step: 3064/1541, loss: 0.13454769551753998 2023-01-23 12:08:51.332715: step: 3066/1541, loss: 0.08816786855459213 2023-01-23 12:08:52.210255: step: 3068/1541, loss: 0.2697616517543793 2023-01-23 12:08:53.096603: step: 3070/1541, loss: 0.2053859829902649 2023-01-23 12:08:53.990657: step: 3072/1541, loss: 0.01465174276381731 2023-01-23 12:08:54.909744: step: 3074/1541, loss: 1.3873205184936523 2023-01-23 12:08:55.777435: step: 3076/1541, loss: 0.37549254298210144 2023-01-23 12:08:56.672820: step: 3078/1541, loss: 0.3581085801124573 2023-01-23 12:08:57.585895: step: 3080/1541, loss: 0.00799867045134306 2023-01-23 12:08:58.471371: step: 3082/1541, loss: 0.0031672550830990076 2023-01-23 12:08:59.056258: step: 3084/1541, loss: 0.002729891799390316 ================================================== Loss: 0.162 -------------------- Dev Chinese: {'event': {'p': 0.6296296296296297, 'r': 0.7017310252996005, 'f1': 0.663727959697733}, 'argument': {'p': 0.6748177089933514, 'r': 0.6370949799191888, 'f1': 0.6554140049008907}, 'combined': 0.4350166002301882, 'epoch': 14} Test Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7171542618958446, 'r': 0.6977929283021589, 'f1': 0.7073411303536308}, 'combined': 0.46152473641107805, 'epoch': 14} Dev Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6654555335459805, 'r': 0.6519157449960623, 'f1': 0.6586160590862389}, 'combined': 0.4566955243569782, 'epoch': 14} Test Russian: {'event': {'p': 0.652014652014652, 'r': 0.7465548232474536, 'f1': 0.6960893854748604}, 'argument': {'p': 0.7071902650999489, 'r': 0.702871933739784, 'f1': 0.7050244869506607}, 'combined': 0.4907600618662141, 'epoch': 14} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6410109498344793, 'r': 0.6750469294717082, 'f1': 0.6575888192267503}, 'combined': 0.48335588421795317, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5810623534386653, 'r': 0.6216015873995024, 'f1': 0.60064872490289}, 'combined': 0.2917436663814037, 'epoch': 14} New best korean model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.653371185022781, 'r': 0.6528791886485169, 'f1': 0.6531250941809645}, 'combined': 0.4496121337998113, 'epoch': 13} Test for Russian: {'event': {'p': 0.6478283621140764, 'r': 0.7417615338526063, 'f1': 0.6916201117318437}, 'argument': {'p': 0.7081306065307763, 'r': 0.708893952495416, 'f1': 0.7085120739072518}, 'combined': 0.49002119971909375, 'epoch': 13} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6150902184235518, 'r': 0.6436990657920891, 'f1': 0.6290695415695416}, 'combined': 0.3235214785214786, 'epoch': 13} ****************************** Epoch: 15 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 12:14:27.730038: step: 2/1541, loss: 0.4067099988460541 2023-01-23 12:14:28.633357: step: 4/1541, loss: 0.06780609488487244 2023-01-23 12:14:29.606074: step: 6/1541, loss: 0.10900995880365372 2023-01-23 12:14:30.537811: step: 8/1541, loss: 0.045695219188928604 2023-01-23 12:14:31.444493: step: 10/1541, loss: 0.0009667086997069418 2023-01-23 12:14:32.358234: step: 12/1541, loss: 0.016366921365261078 2023-01-23 12:14:33.335604: step: 14/1541, loss: 0.0978393629193306 2023-01-23 12:14:34.226034: step: 16/1541, loss: 0.1785341054201126 2023-01-23 12:14:35.137563: step: 18/1541, loss: 0.006470227148383856 2023-01-23 12:14:36.085640: step: 20/1541, loss: 0.935967743396759 2023-01-23 12:14:36.974862: step: 22/1541, loss: 0.05645192787051201 2023-01-23 12:14:37.915005: step: 24/1541, loss: 0.004886425565928221 2023-01-23 12:14:38.794204: step: 26/1541, loss: 0.6234488487243652 2023-01-23 12:14:39.670058: step: 28/1541, loss: 0.3010755181312561 2023-01-23 12:14:40.585189: step: 30/1541, loss: 0.18158940970897675 2023-01-23 12:14:41.495537: step: 32/1541, loss: 0.2149730622768402 2023-01-23 12:14:42.396194: step: 34/1541, loss: 0.32225754857063293 2023-01-23 12:14:43.276273: step: 36/1541, loss: 0.006557898595929146 2023-01-23 12:14:44.164699: step: 38/1541, loss: 0.058468714356422424 2023-01-23 12:14:45.076593: step: 40/1541, loss: 0.017200855538249016 2023-01-23 12:14:45.955315: step: 42/1541, loss: 0.00702578853815794 2023-01-23 12:14:46.868549: step: 44/1541, loss: 0.11467967182397842 2023-01-23 12:14:47.758462: step: 46/1541, loss: 0.01837102323770523 2023-01-23 12:14:48.630034: step: 48/1541, loss: 0.01396170910447836 2023-01-23 12:14:49.535164: step: 50/1541, loss: 0.18154475092887878 2023-01-23 12:14:50.447052: step: 52/1541, loss: 0.23091551661491394 2023-01-23 12:14:51.350640: step: 54/1541, loss: 0.9546921849250793 2023-01-23 12:14:52.237893: step: 56/1541, loss: 0.19628280401229858 2023-01-23 12:14:53.130969: step: 58/1541, loss: 0.01622302085161209 2023-01-23 12:14:54.058688: step: 60/1541, loss: 0.03839785233139992 2023-01-23 12:14:54.973931: step: 62/1541, loss: 0.013722328469157219 2023-01-23 12:14:55.902373: step: 64/1541, loss: 0.006860438734292984 2023-01-23 12:14:56.816622: step: 66/1541, loss: 0.0031874149572104216 2023-01-23 12:14:57.698120: step: 68/1541, loss: 1.3236500024795532 2023-01-23 12:14:58.554476: step: 70/1541, loss: 0.08855244517326355 2023-01-23 12:14:59.494235: step: 72/1541, loss: 0.09116602689027786 2023-01-23 12:15:00.411660: step: 74/1541, loss: 0.029062649235129356 2023-01-23 12:15:01.304828: step: 76/1541, loss: 0.18593525886535645 2023-01-23 12:15:02.238275: step: 78/1541, loss: 0.5472089052200317 2023-01-23 12:15:03.196297: step: 80/1541, loss: 0.14668041467666626 2023-01-23 12:15:04.046515: step: 82/1541, loss: 0.0016447899397462606 2023-01-23 12:15:04.942972: step: 84/1541, loss: 0.017707835882902145 2023-01-23 12:15:05.816698: step: 86/1541, loss: 0.027886221185326576 2023-01-23 12:15:06.682731: step: 88/1541, loss: 0.0563303641974926 2023-01-23 12:15:07.575001: step: 90/1541, loss: 0.1371728479862213 2023-01-23 12:15:08.459293: step: 92/1541, loss: 0.02924475446343422 2023-01-23 12:15:09.327872: step: 94/1541, loss: 0.010221661068499088 2023-01-23 12:15:10.232041: step: 96/1541, loss: 0.08338901400566101 2023-01-23 12:15:11.087403: step: 98/1541, loss: 0.0002034613280557096 2023-01-23 12:15:11.980309: step: 100/1541, loss: 0.0034483294002711773 2023-01-23 12:15:12.874538: step: 102/1541, loss: 0.3192348778247833 2023-01-23 12:15:13.774193: step: 104/1541, loss: 0.26259225606918335 2023-01-23 12:15:14.628041: step: 106/1541, loss: 0.06085563078522682 2023-01-23 12:15:15.523783: step: 108/1541, loss: 0.03426782414317131 2023-01-23 12:15:16.426063: step: 110/1541, loss: 0.18717265129089355 2023-01-23 12:15:17.444327: step: 112/1541, loss: 0.0056410594843328 2023-01-23 12:15:18.329520: step: 114/1541, loss: 0.015378494746983051 2023-01-23 12:15:19.259505: step: 116/1541, loss: 0.37391936779022217 2023-01-23 12:15:20.163530: step: 118/1541, loss: 0.03903043642640114 2023-01-23 12:15:21.065487: step: 120/1541, loss: 0.16377195715904236 2023-01-23 12:15:22.023471: step: 122/1541, loss: 0.026448164135217667 2023-01-23 12:15:22.997565: step: 124/1541, loss: 0.11562903970479965 2023-01-23 12:15:23.854360: step: 126/1541, loss: 0.04641823098063469 2023-01-23 12:15:24.734734: step: 128/1541, loss: 0.02527760900557041 2023-01-23 12:15:25.606318: step: 130/1541, loss: 0.004057181999087334 2023-01-23 12:15:26.483743: step: 132/1541, loss: 0.017190583050251007 2023-01-23 12:15:27.374891: step: 134/1541, loss: 0.006769916508346796 2023-01-23 12:15:28.326084: step: 136/1541, loss: 0.012929564341902733 2023-01-23 12:15:29.208293: step: 138/1541, loss: 0.016155114397406578 2023-01-23 12:15:30.104253: step: 140/1541, loss: 0.15451423823833466 2023-01-23 12:15:31.013282: step: 142/1541, loss: 0.013279041275382042 2023-01-23 12:15:31.928087: step: 144/1541, loss: 0.0166152473539114 2023-01-23 12:15:32.809831: step: 146/1541, loss: 0.006635860539972782 2023-01-23 12:15:33.716149: step: 148/1541, loss: 0.030763912945985794 2023-01-23 12:15:34.630113: step: 150/1541, loss: 0.061766643077135086 2023-01-23 12:15:35.522497: step: 152/1541, loss: 0.14302273094654083 2023-01-23 12:15:36.415211: step: 154/1541, loss: 0.013868939131498337 2023-01-23 12:15:37.305333: step: 156/1541, loss: 0.024335535243153572 2023-01-23 12:15:38.228886: step: 158/1541, loss: 0.04011725261807442 2023-01-23 12:15:39.120929: step: 160/1541, loss: 0.001873594243079424 2023-01-23 12:15:39.988540: step: 162/1541, loss: 0.012849214486777782 2023-01-23 12:15:40.874634: step: 164/1541, loss: 0.021740306168794632 2023-01-23 12:15:41.789869: step: 166/1541, loss: 0.14386160671710968 2023-01-23 12:15:42.661754: step: 168/1541, loss: 0.03663760796189308 2023-01-23 12:15:43.565322: step: 170/1541, loss: 0.17220783233642578 2023-01-23 12:15:44.435369: step: 172/1541, loss: 0.005404041148722172 2023-01-23 12:15:45.323046: step: 174/1541, loss: 0.14415805041790009 2023-01-23 12:15:46.243569: step: 176/1541, loss: 0.006252594757825136 2023-01-23 12:15:47.195171: step: 178/1541, loss: 0.016664402559399605 2023-01-23 12:15:48.113137: step: 180/1541, loss: 0.05869397521018982 2023-01-23 12:15:48.988639: step: 182/1541, loss: 0.008371117524802685 2023-01-23 12:15:49.863136: step: 184/1541, loss: 0.2165045589208603 2023-01-23 12:15:50.771302: step: 186/1541, loss: 0.13814392685890198 2023-01-23 12:15:51.705850: step: 188/1541, loss: 0.3042333722114563 2023-01-23 12:15:52.587156: step: 190/1541, loss: 0.06469045579433441 2023-01-23 12:15:53.456833: step: 192/1541, loss: 0.20177625119686127 2023-01-23 12:15:54.415120: step: 194/1541, loss: 0.032818861305713654 2023-01-23 12:15:55.332132: step: 196/1541, loss: 0.008627619594335556 2023-01-23 12:15:56.206520: step: 198/1541, loss: 0.028901731595396996 2023-01-23 12:15:57.136045: step: 200/1541, loss: 0.031067635864019394 2023-01-23 12:15:58.002017: step: 202/1541, loss: 0.006355793680995703 2023-01-23 12:15:58.883439: step: 204/1541, loss: 0.024225672706961632 2023-01-23 12:15:59.752442: step: 206/1541, loss: 0.027342934161424637 2023-01-23 12:16:00.699672: step: 208/1541, loss: 0.037153616547584534 2023-01-23 12:16:01.629059: step: 210/1541, loss: 0.189172625541687 2023-01-23 12:16:02.555867: step: 212/1541, loss: 0.05348421260714531 2023-01-23 12:16:03.499411: step: 214/1541, loss: 0.14361000061035156 2023-01-23 12:16:04.381454: step: 216/1541, loss: 0.09663659334182739 2023-01-23 12:16:05.285207: step: 218/1541, loss: 0.02651454135775566 2023-01-23 12:16:06.185604: step: 220/1541, loss: 0.06554917246103287 2023-01-23 12:16:07.043588: step: 222/1541, loss: 0.002179620321840048 2023-01-23 12:16:07.881994: step: 224/1541, loss: 0.25530046224594116 2023-01-23 12:16:08.804667: step: 226/1541, loss: 0.008815038949251175 2023-01-23 12:16:09.688932: step: 228/1541, loss: 0.08316577970981598 2023-01-23 12:16:10.559302: step: 230/1541, loss: 0.047608382999897 2023-01-23 12:16:11.453088: step: 232/1541, loss: 0.003930551931262016 2023-01-23 12:16:12.367790: step: 234/1541, loss: 0.09380553662776947 2023-01-23 12:16:13.258687: step: 236/1541, loss: 0.07757551223039627 2023-01-23 12:16:14.139177: step: 238/1541, loss: 0.0023360587656497955 2023-01-23 12:16:15.005890: step: 240/1541, loss: 0.040971677750349045 2023-01-23 12:16:15.875038: step: 242/1541, loss: 0.05194815993309021 2023-01-23 12:16:16.751754: step: 244/1541, loss: 0.004455762915313244 2023-01-23 12:16:17.632330: step: 246/1541, loss: 0.19874835014343262 2023-01-23 12:16:18.489408: step: 248/1541, loss: 0.0293506421148777 2023-01-23 12:16:19.372117: step: 250/1541, loss: 0.20989744365215302 2023-01-23 12:16:20.275157: step: 252/1541, loss: 0.5318740606307983 2023-01-23 12:16:21.306055: step: 254/1541, loss: 0.4401799440383911 2023-01-23 12:16:22.211707: step: 256/1541, loss: 0.002695339499041438 2023-01-23 12:16:23.104701: step: 258/1541, loss: 0.7168267369270325 2023-01-23 12:16:24.006567: step: 260/1541, loss: 0.20209559798240662 2023-01-23 12:16:24.968164: step: 262/1541, loss: 0.05904941260814667 2023-01-23 12:16:25.887584: step: 264/1541, loss: 0.24848821759223938 2023-01-23 12:16:26.761265: step: 266/1541, loss: 0.08335084468126297 2023-01-23 12:16:27.726086: step: 268/1541, loss: 0.08275201916694641 2023-01-23 12:16:28.646682: step: 270/1541, loss: 0.014996551908552647 2023-01-23 12:16:29.542428: step: 272/1541, loss: 0.14223510026931763 2023-01-23 12:16:30.426433: step: 274/1541, loss: 0.023268327116966248 2023-01-23 12:16:31.357942: step: 276/1541, loss: 0.4293275773525238 2023-01-23 12:16:32.255474: step: 278/1541, loss: 0.18427959084510803 2023-01-23 12:16:33.147962: step: 280/1541, loss: 0.2844092547893524 2023-01-23 12:16:34.035379: step: 282/1541, loss: 0.033464156091213226 2023-01-23 12:16:34.900106: step: 284/1541, loss: 0.07468093931674957 2023-01-23 12:16:35.821923: step: 286/1541, loss: 0.06074933335185051 2023-01-23 12:16:36.699180: step: 288/1541, loss: 0.05671871826052666 2023-01-23 12:16:37.587029: step: 290/1541, loss: 0.0580018125474453 2023-01-23 12:16:38.484687: step: 292/1541, loss: 0.02755161188542843 2023-01-23 12:16:39.381359: step: 294/1541, loss: 0.05817568302154541 2023-01-23 12:16:40.279256: step: 296/1541, loss: 0.3320251703262329 2023-01-23 12:16:41.214302: step: 298/1541, loss: 0.019390571862459183 2023-01-23 12:16:42.159554: step: 300/1541, loss: 0.3217492997646332 2023-01-23 12:16:43.086704: step: 302/1541, loss: 0.009363885037600994 2023-01-23 12:16:43.988947: step: 304/1541, loss: 0.3125971257686615 2023-01-23 12:16:44.863122: step: 306/1541, loss: 0.30947306752204895 2023-01-23 12:16:45.744768: step: 308/1541, loss: 0.10481208562850952 2023-01-23 12:16:46.642131: step: 310/1541, loss: 0.04618526250123978 2023-01-23 12:16:47.536792: step: 312/1541, loss: 0.01694180816411972 2023-01-23 12:16:48.488207: step: 314/1541, loss: 0.20500393211841583 2023-01-23 12:16:49.404423: step: 316/1541, loss: 0.10117392241954803 2023-01-23 12:16:50.279982: step: 318/1541, loss: 0.041384220123291016 2023-01-23 12:16:51.180044: step: 320/1541, loss: 0.12290305644273758 2023-01-23 12:16:52.065824: step: 322/1541, loss: 0.013825376518070698 2023-01-23 12:16:52.971504: step: 324/1541, loss: 0.03944411873817444 2023-01-23 12:16:53.855178: step: 326/1541, loss: 0.2644805312156677 2023-01-23 12:16:54.737384: step: 328/1541, loss: 0.023412594571709633 2023-01-23 12:16:55.618546: step: 330/1541, loss: 0.047519560903310776 2023-01-23 12:16:56.518134: step: 332/1541, loss: 0.2610672116279602 2023-01-23 12:16:57.417985: step: 334/1541, loss: 0.003112342208623886 2023-01-23 12:16:58.355356: step: 336/1541, loss: 0.030232222750782967 2023-01-23 12:16:59.244537: step: 338/1541, loss: 0.3649350702762604 2023-01-23 12:17:00.132335: step: 340/1541, loss: 0.041667189449071884 2023-01-23 12:17:01.025211: step: 342/1541, loss: 1.1461290121078491 2023-01-23 12:17:01.911537: step: 344/1541, loss: 0.008074592798948288 2023-01-23 12:17:02.844056: step: 346/1541, loss: 0.1188100278377533 2023-01-23 12:17:03.773453: step: 348/1541, loss: 0.04563954845070839 2023-01-23 12:17:04.676510: step: 350/1541, loss: 0.00924527458846569 2023-01-23 12:17:05.548587: step: 352/1541, loss: 1.3934298753738403 2023-01-23 12:17:06.464018: step: 354/1541, loss: 0.020917588844895363 2023-01-23 12:17:07.373014: step: 356/1541, loss: 0.10955531150102615 2023-01-23 12:17:08.252209: step: 358/1541, loss: 0.08737689256668091 2023-01-23 12:17:09.189284: step: 360/1541, loss: 0.32222262024879456 2023-01-23 12:17:10.078224: step: 362/1541, loss: 0.017832428216934204 2023-01-23 12:17:10.996314: step: 364/1541, loss: 0.3604421019554138 2023-01-23 12:17:11.862095: step: 366/1541, loss: 0.05891122668981552 2023-01-23 12:17:12.729682: step: 368/1541, loss: 0.12857778370380402 2023-01-23 12:17:13.632591: step: 370/1541, loss: 0.008624634705483913 2023-01-23 12:17:14.557803: step: 372/1541, loss: 0.08463335037231445 2023-01-23 12:17:15.531762: step: 374/1541, loss: 0.036678850650787354 2023-01-23 12:17:16.443060: step: 376/1541, loss: 0.010421927087008953 2023-01-23 12:17:17.355847: step: 378/1541, loss: 0.3606750965118408 2023-01-23 12:17:18.277271: step: 380/1541, loss: 0.02483459562063217 2023-01-23 12:17:19.134140: step: 382/1541, loss: 0.04801655188202858 2023-01-23 12:17:20.024413: step: 384/1541, loss: 0.005288888234645128 2023-01-23 12:17:20.955515: step: 386/1541, loss: 0.019136201590299606 2023-01-23 12:17:21.845024: step: 388/1541, loss: 0.25061315298080444 2023-01-23 12:17:22.741281: step: 390/1541, loss: 0.08464079350233078 2023-01-23 12:17:23.649371: step: 392/1541, loss: 0.005044234916567802 2023-01-23 12:17:24.572944: step: 394/1541, loss: 0.00802662130445242 2023-01-23 12:17:25.494691: step: 396/1541, loss: 0.20319858193397522 2023-01-23 12:17:26.377901: step: 398/1541, loss: 0.010283492505550385 2023-01-23 12:17:27.285378: step: 400/1541, loss: 0.008685746230185032 2023-01-23 12:17:28.170876: step: 402/1541, loss: 0.10849893093109131 2023-01-23 12:17:29.076901: step: 404/1541, loss: 0.2903962731361389 2023-01-23 12:17:29.958657: step: 406/1541, loss: 0.03639773279428482 2023-01-23 12:17:30.832806: step: 408/1541, loss: 0.03456498309969902 2023-01-23 12:17:31.726955: step: 410/1541, loss: 0.05606146156787872 2023-01-23 12:17:32.638089: step: 412/1541, loss: 0.04898795858025551 2023-01-23 12:17:33.560110: step: 414/1541, loss: 0.06031615287065506 2023-01-23 12:17:34.477104: step: 416/1541, loss: 0.14575672149658203 2023-01-23 12:17:35.342166: step: 418/1541, loss: 0.005725711118429899 2023-01-23 12:17:36.244252: step: 420/1541, loss: 0.24408283829689026 2023-01-23 12:17:37.134233: step: 422/1541, loss: 0.04790831729769707 2023-01-23 12:17:38.038023: step: 424/1541, loss: 0.07065438479185104 2023-01-23 12:17:38.891954: step: 426/1541, loss: 0.001846138620749116 2023-01-23 12:17:39.817635: step: 428/1541, loss: 0.0897681787610054 2023-01-23 12:17:40.675757: step: 430/1541, loss: 0.033434562385082245 2023-01-23 12:17:41.557987: step: 432/1541, loss: 0.01572410948574543 2023-01-23 12:17:42.433815: step: 434/1541, loss: 0.6571817994117737 2023-01-23 12:17:43.308346: step: 436/1541, loss: 0.001101227360777557 2023-01-23 12:17:44.197720: step: 438/1541, loss: 0.39041197299957275 2023-01-23 12:17:45.083044: step: 440/1541, loss: 0.13712851703166962 2023-01-23 12:17:45.959156: step: 442/1541, loss: 0.2595762014389038 2023-01-23 12:17:46.853050: step: 444/1541, loss: 0.09874233603477478 2023-01-23 12:17:47.749490: step: 446/1541, loss: 0.005135589744895697 2023-01-23 12:17:48.659381: step: 448/1541, loss: 0.020761489868164062 2023-01-23 12:17:49.586276: step: 450/1541, loss: 0.0626884251832962 2023-01-23 12:17:50.477488: step: 452/1541, loss: 0.22988261282444 2023-01-23 12:17:51.358657: step: 454/1541, loss: 0.01080105360597372 2023-01-23 12:17:52.236200: step: 456/1541, loss: 0.048781175166368484 2023-01-23 12:17:53.168317: step: 458/1541, loss: 0.04573604837059975 2023-01-23 12:17:54.082215: step: 460/1541, loss: 0.31420475244522095 2023-01-23 12:17:54.996262: step: 462/1541, loss: 0.04439940303564072 2023-01-23 12:17:55.880899: step: 464/1541, loss: 0.07847367972135544 2023-01-23 12:17:56.798602: step: 466/1541, loss: 0.20555436611175537 2023-01-23 12:17:57.698661: step: 468/1541, loss: 0.04673485457897186 2023-01-23 12:17:58.600732: step: 470/1541, loss: 0.003925798926502466 2023-01-23 12:17:59.528123: step: 472/1541, loss: 0.10323560237884521 2023-01-23 12:18:00.423558: step: 474/1541, loss: 0.07357529550790787 2023-01-23 12:18:01.312737: step: 476/1541, loss: 0.013855962082743645 2023-01-23 12:18:02.253683: step: 478/1541, loss: 1.0853297710418701 2023-01-23 12:18:03.136003: step: 480/1541, loss: 0.035179927945137024 2023-01-23 12:18:04.048659: step: 482/1541, loss: 0.12279050052165985 2023-01-23 12:18:04.942636: step: 484/1541, loss: 0.004957666154950857 2023-01-23 12:18:05.828655: step: 486/1541, loss: 0.06413131207227707 2023-01-23 12:18:06.694040: step: 488/1541, loss: 0.18930256366729736 2023-01-23 12:18:07.584165: step: 490/1541, loss: 0.3098352253437042 2023-01-23 12:18:08.494417: step: 492/1541, loss: 0.1995859295129776 2023-01-23 12:18:09.399909: step: 494/1541, loss: 0.055178202688694 2023-01-23 12:18:10.292566: step: 496/1541, loss: 0.024316180497407913 2023-01-23 12:18:11.352918: step: 498/1541, loss: 0.0025849435478448868 2023-01-23 12:18:12.241041: step: 500/1541, loss: 0.010883730836212635 2023-01-23 12:18:13.163777: step: 502/1541, loss: 0.41042646765708923 2023-01-23 12:18:14.061922: step: 504/1541, loss: 0.020577717572450638 2023-01-23 12:18:14.941529: step: 506/1541, loss: 0.021816011518239975 2023-01-23 12:18:15.816440: step: 508/1541, loss: 0.005582851357758045 2023-01-23 12:18:16.677242: step: 510/1541, loss: 0.03377614915370941 2023-01-23 12:18:17.560309: step: 512/1541, loss: 0.13582676649093628 2023-01-23 12:18:18.453988: step: 514/1541, loss: 0.005816690158098936 2023-01-23 12:18:19.359014: step: 516/1541, loss: 0.10815032571554184 2023-01-23 12:18:20.256881: step: 518/1541, loss: 0.14162790775299072 2023-01-23 12:18:21.215591: step: 520/1541, loss: 0.014749046415090561 2023-01-23 12:18:22.078984: step: 522/1541, loss: 0.10314329713582993 2023-01-23 12:18:22.957879: step: 524/1541, loss: 0.007904594764113426 2023-01-23 12:18:23.859431: step: 526/1541, loss: 0.00724190566688776 2023-01-23 12:18:24.673813: step: 528/1541, loss: 0.20755325257778168 2023-01-23 12:18:25.576121: step: 530/1541, loss: 0.08097235858440399 2023-01-23 12:18:26.468215: step: 532/1541, loss: 0.11153706163167953 2023-01-23 12:18:27.355572: step: 534/1541, loss: 0.0071891606785357 2023-01-23 12:18:28.269276: step: 536/1541, loss: 0.25288864970207214 2023-01-23 12:18:29.155491: step: 538/1541, loss: 0.04290465638041496 2023-01-23 12:18:30.058512: step: 540/1541, loss: 0.006449795328080654 2023-01-23 12:18:30.979786: step: 542/1541, loss: 0.08103392273187637 2023-01-23 12:18:31.902899: step: 544/1541, loss: 0.02580043114721775 2023-01-23 12:18:32.850646: step: 546/1541, loss: 0.11043974757194519 2023-01-23 12:18:33.723409: step: 548/1541, loss: 0.004413043148815632 2023-01-23 12:18:34.707244: step: 550/1541, loss: 0.004341694992035627 2023-01-23 12:18:35.584419: step: 552/1541, loss: 0.010722370818257332 2023-01-23 12:18:36.489309: step: 554/1541, loss: 0.0008737329626455903 2023-01-23 12:18:37.414828: step: 556/1541, loss: 0.13630230724811554 2023-01-23 12:18:38.388415: step: 558/1541, loss: 1.1597892045974731 2023-01-23 12:18:39.263425: step: 560/1541, loss: 0.130357563495636 2023-01-23 12:18:40.110117: step: 562/1541, loss: 0.23500432074069977 2023-01-23 12:18:41.008590: step: 564/1541, loss: 0.1650581955909729 2023-01-23 12:18:41.878358: step: 566/1541, loss: 0.002473449567332864 2023-01-23 12:18:42.731658: step: 568/1541, loss: 0.06218365579843521 2023-01-23 12:18:43.623907: step: 570/1541, loss: 0.05996173620223999 2023-01-23 12:18:44.517271: step: 572/1541, loss: 0.017064519226551056 2023-01-23 12:18:45.430575: step: 574/1541, loss: 0.0965208038687706 2023-01-23 12:18:46.307177: step: 576/1541, loss: 1.1071611642837524 2023-01-23 12:18:47.169414: step: 578/1541, loss: 0.03853858262300491 2023-01-23 12:18:48.093334: step: 580/1541, loss: 0.05052978917956352 2023-01-23 12:18:49.006328: step: 582/1541, loss: 0.10002465546131134 2023-01-23 12:18:49.925802: step: 584/1541, loss: 0.09748438000679016 2023-01-23 12:18:50.842006: step: 586/1541, loss: 0.21693521738052368 2023-01-23 12:18:51.751613: step: 588/1541, loss: 0.6222359538078308 2023-01-23 12:18:52.737325: step: 590/1541, loss: 0.011220755986869335 2023-01-23 12:18:53.628593: step: 592/1541, loss: 0.016758928075432777 2023-01-23 12:18:54.582135: step: 594/1541, loss: 0.07908864319324493 2023-01-23 12:18:55.557228: step: 596/1541, loss: 0.06432012468576431 2023-01-23 12:18:56.409337: step: 598/1541, loss: 0.003743100445717573 2023-01-23 12:18:57.281004: step: 600/1541, loss: 0.31664079427719116 2023-01-23 12:18:58.197384: step: 602/1541, loss: 0.02529330551624298 2023-01-23 12:18:59.086620: step: 604/1541, loss: 0.09198378026485443 2023-01-23 12:18:59.946835: step: 606/1541, loss: 0.015681812539696693 2023-01-23 12:19:00.829903: step: 608/1541, loss: 0.008552473038434982 2023-01-23 12:19:01.708560: step: 610/1541, loss: 0.3057028651237488 2023-01-23 12:19:02.580024: step: 612/1541, loss: 0.13444052636623383 2023-01-23 12:19:03.461866: step: 614/1541, loss: 0.25224384665489197 2023-01-23 12:19:04.402938: step: 616/1541, loss: 0.004250320605933666 2023-01-23 12:19:05.300412: step: 618/1541, loss: 0.03171277791261673 2023-01-23 12:19:06.202434: step: 620/1541, loss: 0.010000470094382763 2023-01-23 12:19:07.144918: step: 622/1541, loss: 0.028362879529595375 2023-01-23 12:19:08.015705: step: 624/1541, loss: 0.016509830951690674 2023-01-23 12:19:08.868996: step: 626/1541, loss: 0.015479701571166515 2023-01-23 12:19:09.763896: step: 628/1541, loss: 0.1312105804681778 2023-01-23 12:19:10.627041: step: 630/1541, loss: 0.16929954290390015 2023-01-23 12:19:11.533146: step: 632/1541, loss: 0.049934834241867065 2023-01-23 12:19:12.430167: step: 634/1541, loss: 0.1230403333902359 2023-01-23 12:19:13.302025: step: 636/1541, loss: 0.0163526963442564 2023-01-23 12:19:14.235332: step: 638/1541, loss: 0.018938157707452774 2023-01-23 12:19:15.160821: step: 640/1541, loss: 0.04860479384660721 2023-01-23 12:19:16.069667: step: 642/1541, loss: 0.007731952238827944 2023-01-23 12:19:17.023835: step: 644/1541, loss: 0.021832536906003952 2023-01-23 12:19:17.935614: step: 646/1541, loss: 0.08569221198558807 2023-01-23 12:19:18.806064: step: 648/1541, loss: 0.2944330871105194 2023-01-23 12:19:19.703961: step: 650/1541, loss: 0.0032574248034507036 2023-01-23 12:19:20.579349: step: 652/1541, loss: 0.12160337716341019 2023-01-23 12:19:21.478634: step: 654/1541, loss: 0.03944031149148941 2023-01-23 12:19:22.391175: step: 656/1541, loss: 0.11156072467565536 2023-01-23 12:19:23.289102: step: 658/1541, loss: 1.2666593790054321 2023-01-23 12:19:24.193165: step: 660/1541, loss: 0.551501989364624 2023-01-23 12:19:25.075584: step: 662/1541, loss: 0.061333492398262024 2023-01-23 12:19:25.946954: step: 664/1541, loss: 0.06821636110544205 2023-01-23 12:19:26.848693: step: 666/1541, loss: 1.4634082317352295 2023-01-23 12:19:27.734143: step: 668/1541, loss: 0.0006456610281020403 2023-01-23 12:19:28.689156: step: 670/1541, loss: 0.2376357614994049 2023-01-23 12:19:29.622059: step: 672/1541, loss: 0.033606067299842834 2023-01-23 12:19:30.548112: step: 674/1541, loss: 0.05026794224977493 2023-01-23 12:19:31.421673: step: 676/1541, loss: 0.11132895201444626 2023-01-23 12:19:32.290849: step: 678/1541, loss: 0.1622699499130249 2023-01-23 12:19:33.208288: step: 680/1541, loss: 0.005262251477688551 2023-01-23 12:19:34.147166: step: 682/1541, loss: 0.035114631056785583 2023-01-23 12:19:35.020577: step: 684/1541, loss: 0.1290803700685501 2023-01-23 12:19:35.926355: step: 686/1541, loss: 0.04123770445585251 2023-01-23 12:19:36.803753: step: 688/1541, loss: 0.18110282719135284 2023-01-23 12:19:37.714613: step: 690/1541, loss: 0.5118951797485352 2023-01-23 12:19:38.594608: step: 692/1541, loss: 0.005201089661568403 2023-01-23 12:19:39.516949: step: 694/1541, loss: 0.36927375197410583 2023-01-23 12:19:40.410753: step: 696/1541, loss: 0.1280173808336258 2023-01-23 12:19:41.267681: step: 698/1541, loss: 0.008035251870751381 2023-01-23 12:19:42.143197: step: 700/1541, loss: 0.007821801118552685 2023-01-23 12:19:43.040608: step: 702/1541, loss: 0.01647956296801567 2023-01-23 12:19:43.964761: step: 704/1541, loss: 0.31388017535209656 2023-01-23 12:19:44.837448: step: 706/1541, loss: 0.028450926765799522 2023-01-23 12:19:45.749865: step: 708/1541, loss: 0.19882021844387054 2023-01-23 12:19:46.680492: step: 710/1541, loss: 0.20398300886154175 2023-01-23 12:19:47.602887: step: 712/1541, loss: 0.23040170967578888 2023-01-23 12:19:48.473721: step: 714/1541, loss: 0.015168161131441593 2023-01-23 12:19:49.376798: step: 716/1541, loss: 0.5740026235580444 2023-01-23 12:19:50.268201: step: 718/1541, loss: 0.23000170290470123 2023-01-23 12:19:51.189009: step: 720/1541, loss: 0.047825489193201065 2023-01-23 12:19:52.158261: step: 722/1541, loss: 0.06869639456272125 2023-01-23 12:19:53.094999: step: 724/1541, loss: 0.08699088543653488 2023-01-23 12:19:53.990247: step: 726/1541, loss: 0.18729476630687714 2023-01-23 12:19:54.904129: step: 728/1541, loss: 0.2933919131755829 2023-01-23 12:19:55.765736: step: 730/1541, loss: 0.4944111406803131 2023-01-23 12:19:56.697517: step: 732/1541, loss: 0.012121575884521008 2023-01-23 12:19:57.604499: step: 734/1541, loss: 0.03122948296368122 2023-01-23 12:19:58.516732: step: 736/1541, loss: 0.13241787254810333 2023-01-23 12:19:59.426730: step: 738/1541, loss: 0.08937380462884903 2023-01-23 12:20:00.322732: step: 740/1541, loss: 0.35023078322410583 2023-01-23 12:20:01.200439: step: 742/1541, loss: 0.03728384152054787 2023-01-23 12:20:02.196230: step: 744/1541, loss: 0.029992196708917618 2023-01-23 12:20:03.111680: step: 746/1541, loss: 0.04021493345499039 2023-01-23 12:20:04.007843: step: 748/1541, loss: 0.016754314303398132 2023-01-23 12:20:04.887009: step: 750/1541, loss: 0.11557214707136154 2023-01-23 12:20:05.739815: step: 752/1541, loss: 0.09522692859172821 2023-01-23 12:20:06.613606: step: 754/1541, loss: 0.02998659759759903 2023-01-23 12:20:07.529631: step: 756/1541, loss: 0.3899754285812378 2023-01-23 12:20:08.360818: step: 758/1541, loss: 0.06641635298728943 2023-01-23 12:20:09.266050: step: 760/1541, loss: 0.5062195658683777 2023-01-23 12:20:10.193646: step: 762/1541, loss: 0.015614783391356468 2023-01-23 12:20:11.160196: step: 764/1541, loss: 0.30524516105651855 2023-01-23 12:20:12.028784: step: 766/1541, loss: 0.31508681178092957 2023-01-23 12:20:12.908250: step: 768/1541, loss: 0.060892075300216675 2023-01-23 12:20:13.893052: step: 770/1541, loss: 0.10797780752182007 2023-01-23 12:20:14.767341: step: 772/1541, loss: 0.03405417874455452 2023-01-23 12:20:15.701779: step: 774/1541, loss: 0.009963138960301876 2023-01-23 12:20:16.591557: step: 776/1541, loss: 0.5031338334083557 2023-01-23 12:20:17.493731: step: 778/1541, loss: 0.4589686989784241 2023-01-23 12:20:18.432574: step: 780/1541, loss: 0.04872330278158188 2023-01-23 12:20:19.318314: step: 782/1541, loss: 0.12772825360298157 2023-01-23 12:20:20.183208: step: 784/1541, loss: 0.00619846535846591 2023-01-23 12:20:21.110759: step: 786/1541, loss: 0.014462785795331001 2023-01-23 12:20:22.019411: step: 788/1541, loss: 0.17951202392578125 2023-01-23 12:20:22.965611: step: 790/1541, loss: 0.0024814673233777285 2023-01-23 12:20:23.860744: step: 792/1541, loss: 0.17538785934448242 2023-01-23 12:20:24.780802: step: 794/1541, loss: 0.042496226727962494 2023-01-23 12:20:25.730310: step: 796/1541, loss: 0.3512438237667084 2023-01-23 12:20:26.662200: step: 798/1541, loss: 0.05613245069980621 2023-01-23 12:20:27.539437: step: 800/1541, loss: 0.0030774713959544897 2023-01-23 12:20:28.474394: step: 802/1541, loss: 0.04816320165991783 2023-01-23 12:20:29.375811: step: 804/1541, loss: 0.009304355829954147 2023-01-23 12:20:30.277515: step: 806/1541, loss: 0.11235206574201584 2023-01-23 12:20:31.144824: step: 808/1541, loss: 0.05752961337566376 2023-01-23 12:20:32.011096: step: 810/1541, loss: 0.24540802836418152 2023-01-23 12:20:32.881287: step: 812/1541, loss: 0.03215090557932854 2023-01-23 12:20:33.774389: step: 814/1541, loss: 0.12097320705652237 2023-01-23 12:20:34.649348: step: 816/1541, loss: 0.04700126498937607 2023-01-23 12:20:35.534958: step: 818/1541, loss: 0.03956938907504082 2023-01-23 12:20:36.483589: step: 820/1541, loss: 0.11057056486606598 2023-01-23 12:20:37.365438: step: 822/1541, loss: 0.005733026657253504 2023-01-23 12:20:38.341620: step: 824/1541, loss: 0.12531812489032745 2023-01-23 12:20:39.290602: step: 826/1541, loss: 1.1267905235290527 2023-01-23 12:20:40.166025: step: 828/1541, loss: 0.04285607486963272 2023-01-23 12:20:41.083579: step: 830/1541, loss: 0.3819854259490967 2023-01-23 12:20:41.962931: step: 832/1541, loss: 0.08584155887365341 2023-01-23 12:20:42.876884: step: 834/1541, loss: 0.08415927737951279 2023-01-23 12:20:43.778856: step: 836/1541, loss: 0.6750777959823608 2023-01-23 12:20:44.631302: step: 838/1541, loss: 0.5812424421310425 2023-01-23 12:20:45.580949: step: 840/1541, loss: 0.046554017812013626 2023-01-23 12:20:46.467064: step: 842/1541, loss: 0.06625492125749588 2023-01-23 12:20:47.339089: step: 844/1541, loss: 0.009984346106648445 2023-01-23 12:20:48.235084: step: 846/1541, loss: 0.05605258792638779 2023-01-23 12:20:49.114554: step: 848/1541, loss: 0.0953342393040657 2023-01-23 12:20:49.991055: step: 850/1541, loss: 0.03645393252372742 2023-01-23 12:20:50.919780: step: 852/1541, loss: 0.29214128851890564 2023-01-23 12:20:51.798698: step: 854/1541, loss: 0.5878388285636902 2023-01-23 12:20:52.735711: step: 856/1541, loss: 0.07939299941062927 2023-01-23 12:20:53.626843: step: 858/1541, loss: 0.22199608385562897 2023-01-23 12:20:54.511632: step: 860/1541, loss: 0.011402169242501259 2023-01-23 12:20:55.399655: step: 862/1541, loss: 0.049998216331005096 2023-01-23 12:20:56.374220: step: 864/1541, loss: 0.44870057702064514 2023-01-23 12:20:57.251009: step: 866/1541, loss: 0.2434786856174469 2023-01-23 12:20:58.152972: step: 868/1541, loss: 0.23285998404026031 2023-01-23 12:20:59.058670: step: 870/1541, loss: 0.1773359179496765 2023-01-23 12:20:59.932841: step: 872/1541, loss: 0.007484900765120983 2023-01-23 12:21:00.839226: step: 874/1541, loss: 0.10335783660411835 2023-01-23 12:21:01.773250: step: 876/1541, loss: 0.1928868442773819 2023-01-23 12:21:02.630270: step: 878/1541, loss: 0.26422226428985596 2023-01-23 12:21:03.566200: step: 880/1541, loss: 0.17720302939414978 2023-01-23 12:21:04.466655: step: 882/1541, loss: 0.5039218664169312 2023-01-23 12:21:05.395722: step: 884/1541, loss: 0.02136816643178463 2023-01-23 12:21:06.355121: step: 886/1541, loss: 0.13132241368293762 2023-01-23 12:21:07.258378: step: 888/1541, loss: 0.036689192056655884 2023-01-23 12:21:08.141874: step: 890/1541, loss: 0.10360404849052429 2023-01-23 12:21:09.070749: step: 892/1541, loss: 0.023020390421152115 2023-01-23 12:21:09.951995: step: 894/1541, loss: 0.07347550988197327 2023-01-23 12:21:10.853029: step: 896/1541, loss: 0.010742387734353542 2023-01-23 12:21:11.743801: step: 898/1541, loss: 0.03603934496641159 2023-01-23 12:21:12.678691: step: 900/1541, loss: 0.011365529149770737 2023-01-23 12:21:13.557253: step: 902/1541, loss: 0.0372081995010376 2023-01-23 12:21:14.400935: step: 904/1541, loss: 0.12297080457210541 2023-01-23 12:21:15.299498: step: 906/1541, loss: 0.3247906267642975 2023-01-23 12:21:16.201028: step: 908/1541, loss: 0.14520201086997986 2023-01-23 12:21:17.043652: step: 910/1541, loss: 0.10387277603149414 2023-01-23 12:21:17.908409: step: 912/1541, loss: 0.020180650055408478 2023-01-23 12:21:18.762389: step: 914/1541, loss: 0.07768954336643219 2023-01-23 12:21:19.612869: step: 916/1541, loss: 0.014360473491251469 2023-01-23 12:21:20.489966: step: 918/1541, loss: 0.510814905166626 2023-01-23 12:21:21.395117: step: 920/1541, loss: 0.06473717838525772 2023-01-23 12:21:22.315760: step: 922/1541, loss: 0.017404183745384216 2023-01-23 12:21:23.183461: step: 924/1541, loss: 0.20124658942222595 2023-01-23 12:21:24.127304: step: 926/1541, loss: 0.12125884741544724 2023-01-23 12:21:25.047420: step: 928/1541, loss: 0.03518446162343025 2023-01-23 12:21:25.924371: step: 930/1541, loss: 0.0906638503074646 2023-01-23 12:21:26.888610: step: 932/1541, loss: 0.0030425433069467545 2023-01-23 12:21:27.810661: step: 934/1541, loss: 0.05116158723831177 2023-01-23 12:21:28.704233: step: 936/1541, loss: 0.022401470690965652 2023-01-23 12:21:29.621643: step: 938/1541, loss: 0.042548540979623795 2023-01-23 12:21:30.530075: step: 940/1541, loss: 0.006630975753068924 2023-01-23 12:21:31.432433: step: 942/1541, loss: 0.17754767835140228 2023-01-23 12:21:32.349272: step: 944/1541, loss: 0.22219984233379364 2023-01-23 12:21:33.211698: step: 946/1541, loss: 0.0262735728174448 2023-01-23 12:21:34.107886: step: 948/1541, loss: 0.008901470340788364 2023-01-23 12:21:34.981310: step: 950/1541, loss: 0.02477225661277771 2023-01-23 12:21:35.882555: step: 952/1541, loss: 0.18926826119422913 2023-01-23 12:21:36.761906: step: 954/1541, loss: 0.032699305564165115 2023-01-23 12:21:37.660934: step: 956/1541, loss: 0.14678478240966797 2023-01-23 12:21:38.601293: step: 958/1541, loss: 0.024352557957172394 2023-01-23 12:21:39.511224: step: 960/1541, loss: 0.1439683437347412 2023-01-23 12:21:40.468620: step: 962/1541, loss: 0.030054764822125435 2023-01-23 12:21:41.352911: step: 964/1541, loss: 0.030008167028427124 2023-01-23 12:21:42.228274: step: 966/1541, loss: 0.014248802326619625 2023-01-23 12:21:43.139469: step: 968/1541, loss: 0.003603618359193206 2023-01-23 12:21:44.016228: step: 970/1541, loss: 0.027299530804157257 2023-01-23 12:21:44.920324: step: 972/1541, loss: 0.09538746625185013 2023-01-23 12:21:45.805625: step: 974/1541, loss: 0.3464047312736511 2023-01-23 12:21:46.706358: step: 976/1541, loss: 0.04860107973217964 2023-01-23 12:21:47.630121: step: 978/1541, loss: 0.017860472202301025 2023-01-23 12:21:48.495315: step: 980/1541, loss: 0.06671252846717834 2023-01-23 12:21:49.408307: step: 982/1541, loss: 0.27944445610046387 2023-01-23 12:21:50.295088: step: 984/1541, loss: 0.37613189220428467 2023-01-23 12:21:51.180186: step: 986/1541, loss: 0.562585175037384 2023-01-23 12:21:52.094400: step: 988/1541, loss: 0.35655462741851807 2023-01-23 12:21:53.023854: step: 990/1541, loss: 0.19460268318653107 2023-01-23 12:21:53.899828: step: 992/1541, loss: 0.06440407037734985 2023-01-23 12:21:54.774448: step: 994/1541, loss: 0.005390047561377287 2023-01-23 12:21:55.658381: step: 996/1541, loss: 0.030556943267583847 2023-01-23 12:21:56.608197: step: 998/1541, loss: 0.39935433864593506 2023-01-23 12:21:57.541445: step: 1000/1541, loss: 0.062184982001781464 2023-01-23 12:21:58.432836: step: 1002/1541, loss: 0.26581189036369324 2023-01-23 12:21:59.298581: step: 1004/1541, loss: 0.0049825734458863735 2023-01-23 12:22:00.223147: step: 1006/1541, loss: 0.03780402988195419 2023-01-23 12:22:01.130655: step: 1008/1541, loss: 0.03746917098760605 2023-01-23 12:22:02.045399: step: 1010/1541, loss: 0.19715413451194763 2023-01-23 12:22:02.887286: step: 1012/1541, loss: 0.01753527671098709 2023-01-23 12:22:03.792052: step: 1014/1541, loss: 0.2768494486808777 2023-01-23 12:22:04.724970: step: 1016/1541, loss: 0.19653402268886566 2023-01-23 12:22:05.561040: step: 1018/1541, loss: 0.05328771844506264 2023-01-23 12:22:06.412035: step: 1020/1541, loss: 0.19399456679821014 2023-01-23 12:22:07.275300: step: 1022/1541, loss: 0.0594821460545063 2023-01-23 12:22:08.190141: step: 1024/1541, loss: 0.02632882632315159 2023-01-23 12:22:09.075673: step: 1026/1541, loss: 0.002997542032971978 2023-01-23 12:22:09.962280: step: 1028/1541, loss: 0.00361145893111825 2023-01-23 12:22:10.853708: step: 1030/1541, loss: 0.03138960152864456 2023-01-23 12:22:11.740769: step: 1032/1541, loss: 0.17487530410289764 2023-01-23 12:22:12.658313: step: 1034/1541, loss: 0.05677339434623718 2023-01-23 12:22:13.574237: step: 1036/1541, loss: 0.23964732885360718 2023-01-23 12:22:14.458673: step: 1038/1541, loss: 0.050506554543972015 2023-01-23 12:22:15.362943: step: 1040/1541, loss: 0.06296286731958389 2023-01-23 12:22:16.268238: step: 1042/1541, loss: 0.3026930093765259 2023-01-23 12:22:17.146105: step: 1044/1541, loss: 0.04282674193382263 2023-01-23 12:22:18.039119: step: 1046/1541, loss: 0.0653853490948677 2023-01-23 12:22:18.985418: step: 1048/1541, loss: 0.005411029793322086 2023-01-23 12:22:19.864097: step: 1050/1541, loss: 0.02570343390107155 2023-01-23 12:22:20.753196: step: 1052/1541, loss: 0.012209447100758553 2023-01-23 12:22:21.628026: step: 1054/1541, loss: 0.0048933858051896095 2023-01-23 12:22:22.466019: step: 1056/1541, loss: 0.005448037292808294 2023-01-23 12:22:23.359813: step: 1058/1541, loss: 0.37260469794273376 2023-01-23 12:22:24.226959: step: 1060/1541, loss: 0.001958518521860242 2023-01-23 12:22:25.097015: step: 1062/1541, loss: 0.003798214253038168 2023-01-23 12:22:25.968244: step: 1064/1541, loss: 0.16216063499450684 2023-01-23 12:22:26.926741: step: 1066/1541, loss: 0.04228448495268822 2023-01-23 12:22:27.904810: step: 1068/1541, loss: 0.06302454322576523 2023-01-23 12:22:28.794602: step: 1070/1541, loss: 0.016167597845196724 2023-01-23 12:22:29.715006: step: 1072/1541, loss: 0.09643065929412842 2023-01-23 12:22:30.687439: step: 1074/1541, loss: 0.03818436339497566 2023-01-23 12:22:31.551496: step: 1076/1541, loss: 0.9132183790206909 2023-01-23 12:22:32.509230: step: 1078/1541, loss: 0.06370316445827484 2023-01-23 12:22:33.399559: step: 1080/1541, loss: 0.02953401580452919 2023-01-23 12:22:34.353780: step: 1082/1541, loss: 0.015059160068631172 2023-01-23 12:22:35.233128: step: 1084/1541, loss: 0.02338787354528904 2023-01-23 12:22:36.102084: step: 1086/1541, loss: 0.056568704545497894 2023-01-23 12:22:37.108007: step: 1088/1541, loss: 0.020358292385935783 2023-01-23 12:22:37.989520: step: 1090/1541, loss: 0.012165535241365433 2023-01-23 12:22:38.859551: step: 1092/1541, loss: 0.1114230826497078 2023-01-23 12:22:39.804346: step: 1094/1541, loss: 0.053229764103889465 2023-01-23 12:22:40.704393: step: 1096/1541, loss: 0.00564451701939106 2023-01-23 12:22:41.641689: step: 1098/1541, loss: 0.13928821682929993 2023-01-23 12:22:42.512816: step: 1100/1541, loss: 0.057440854609012604 2023-01-23 12:22:43.400141: step: 1102/1541, loss: 0.08632699400186539 2023-01-23 12:22:44.257105: step: 1104/1541, loss: 0.13633929193019867 2023-01-23 12:22:45.128188: step: 1106/1541, loss: 0.022658025845885277 2023-01-23 12:22:46.038965: step: 1108/1541, loss: 0.14611166715621948 2023-01-23 12:22:46.939032: step: 1110/1541, loss: 0.07523344457149506 2023-01-23 12:22:47.828573: step: 1112/1541, loss: 0.031192950904369354 2023-01-23 12:22:48.677435: step: 1114/1541, loss: 0.07711824029684067 2023-01-23 12:22:49.553046: step: 1116/1541, loss: 0.01930995099246502 2023-01-23 12:22:50.422005: step: 1118/1541, loss: 2.1649274826049805 2023-01-23 12:22:51.358732: step: 1120/1541, loss: 0.09399145841598511 2023-01-23 12:22:52.337133: step: 1122/1541, loss: 0.13023199141025543 2023-01-23 12:22:53.211286: step: 1124/1541, loss: 0.27824801206588745 2023-01-23 12:22:54.121703: step: 1126/1541, loss: 0.016991691663861275 2023-01-23 12:22:55.004244: step: 1128/1541, loss: 0.3958165943622589 2023-01-23 12:22:55.906007: step: 1130/1541, loss: 0.06585425138473511 2023-01-23 12:22:56.912431: step: 1132/1541, loss: 0.21279644966125488 2023-01-23 12:22:57.860848: step: 1134/1541, loss: 0.022008290514349937 2023-01-23 12:22:58.766972: step: 1136/1541, loss: 0.03574465960264206 2023-01-23 12:22:59.647528: step: 1138/1541, loss: 0.015515055507421494 2023-01-23 12:23:00.515661: step: 1140/1541, loss: 0.11110100150108337 2023-01-23 12:23:01.406324: step: 1142/1541, loss: 0.034225597977638245 2023-01-23 12:23:02.308330: step: 1144/1541, loss: 0.14620302617549896 2023-01-23 12:23:03.241432: step: 1146/1541, loss: 0.08825882524251938 2023-01-23 12:23:04.100895: step: 1148/1541, loss: 0.004178558010607958 2023-01-23 12:23:05.008872: step: 1150/1541, loss: 0.0473274327814579 2023-01-23 12:23:05.890529: step: 1152/1541, loss: 0.21824122965335846 2023-01-23 12:23:06.743974: step: 1154/1541, loss: 0.024382954463362694 2023-01-23 12:23:07.722539: step: 1156/1541, loss: 0.06836492568254471 2023-01-23 12:23:08.607902: step: 1158/1541, loss: 0.05286375433206558 2023-01-23 12:23:09.480181: step: 1160/1541, loss: 0.21204957365989685 2023-01-23 12:23:10.450719: step: 1162/1541, loss: 0.0013121053343638778 2023-01-23 12:23:11.322555: step: 1164/1541, loss: 0.09757303446531296 2023-01-23 12:23:12.221095: step: 1166/1541, loss: 0.02968578040599823 2023-01-23 12:23:13.086039: step: 1168/1541, loss: 0.025383278727531433 2023-01-23 12:23:14.047506: step: 1170/1541, loss: 0.009652221575379372 2023-01-23 12:23:15.005151: step: 1172/1541, loss: 0.098670594394207 2023-01-23 12:23:15.920593: step: 1174/1541, loss: 0.06952695548534393 2023-01-23 12:23:16.813111: step: 1176/1541, loss: 0.004398268181830645 2023-01-23 12:23:17.657150: step: 1178/1541, loss: 0.003972180187702179 2023-01-23 12:23:18.552885: step: 1180/1541, loss: 0.10191850364208221 2023-01-23 12:23:19.566273: step: 1182/1541, loss: 0.28667810559272766 2023-01-23 12:23:20.486867: step: 1184/1541, loss: 0.11399544030427933 2023-01-23 12:23:21.406787: step: 1186/1541, loss: 0.05571867898106575 2023-01-23 12:23:22.270039: step: 1188/1541, loss: 0.10570015758275986 2023-01-23 12:23:23.180527: step: 1190/1541, loss: 0.0034837934654206038 2023-01-23 12:23:24.063737: step: 1192/1541, loss: 0.31980100274086 2023-01-23 12:23:24.923525: step: 1194/1541, loss: 0.011518276296555996 2023-01-23 12:23:25.791800: step: 1196/1541, loss: 0.12963399291038513 2023-01-23 12:23:26.677302: step: 1198/1541, loss: 0.2690485417842865 2023-01-23 12:23:27.615705: step: 1200/1541, loss: 0.013781832531094551 2023-01-23 12:23:28.512133: step: 1202/1541, loss: 0.13441786170005798 2023-01-23 12:23:29.386774: step: 1204/1541, loss: 0.006486084777861834 2023-01-23 12:23:30.301566: step: 1206/1541, loss: 0.015432910993695259 2023-01-23 12:23:31.182564: step: 1208/1541, loss: 0.040530167520046234 2023-01-23 12:23:32.113738: step: 1210/1541, loss: 0.18092475831508636 2023-01-23 12:23:33.013238: step: 1212/1541, loss: 0.08696281909942627 2023-01-23 12:23:33.895830: step: 1214/1541, loss: 0.3480096161365509 2023-01-23 12:23:34.823106: step: 1216/1541, loss: 0.3307022750377655 2023-01-23 12:23:35.729803: step: 1218/1541, loss: 0.21077212691307068 2023-01-23 12:23:36.648547: step: 1220/1541, loss: 0.053238093852996826 2023-01-23 12:23:37.538994: step: 1222/1541, loss: 0.14818601310253143 2023-01-23 12:23:38.413762: step: 1224/1541, loss: 0.01008789986371994 2023-01-23 12:23:39.388334: step: 1226/1541, loss: 0.014471916481852531 2023-01-23 12:23:40.251448: step: 1228/1541, loss: 0.005274545401334763 2023-01-23 12:23:41.151325: step: 1230/1541, loss: 0.15626344084739685 2023-01-23 12:23:42.074309: step: 1232/1541, loss: 0.36842358112335205 2023-01-23 12:23:42.986629: step: 1234/1541, loss: 0.08084002882242203 2023-01-23 12:23:43.891944: step: 1236/1541, loss: 0.06627455353736877 2023-01-23 12:23:44.784130: step: 1238/1541, loss: 0.035069093108177185 2023-01-23 12:23:45.668915: step: 1240/1541, loss: 0.09245655685663223 2023-01-23 12:23:46.627298: step: 1242/1541, loss: 0.08872610330581665 2023-01-23 12:23:47.565294: step: 1244/1541, loss: 0.15518881380558014 2023-01-23 12:23:48.521116: step: 1246/1541, loss: 0.1910715401172638 2023-01-23 12:23:49.388810: step: 1248/1541, loss: 0.15453404188156128 2023-01-23 12:23:50.278248: step: 1250/1541, loss: 0.05009943246841431 2023-01-23 12:23:51.195831: step: 1252/1541, loss: 0.03782614320516586 2023-01-23 12:23:52.056098: step: 1254/1541, loss: 0.05413992702960968 2023-01-23 12:23:52.981197: step: 1256/1541, loss: 0.26207244396209717 2023-01-23 12:23:53.951302: step: 1258/1541, loss: 0.030188560485839844 2023-01-23 12:23:54.832571: step: 1260/1541, loss: 0.019591886550188065 2023-01-23 12:23:55.688829: step: 1262/1541, loss: 0.0883016437292099 2023-01-23 12:23:56.595579: step: 1264/1541, loss: 0.0028957980684936047 2023-01-23 12:23:57.528888: step: 1266/1541, loss: 0.011610234156250954 2023-01-23 12:23:58.440386: step: 1268/1541, loss: 0.02826382778584957 2023-01-23 12:23:59.305945: step: 1270/1541, loss: 0.012247627601027489 2023-01-23 12:24:00.258875: step: 1272/1541, loss: 0.048852842301130295 2023-01-23 12:24:01.188514: step: 1274/1541, loss: 0.05525306612253189 2023-01-23 12:24:02.090365: step: 1276/1541, loss: 0.08486683666706085 2023-01-23 12:24:02.962907: step: 1278/1541, loss: 0.09486042708158493 2023-01-23 12:24:03.841783: step: 1280/1541, loss: 0.06479842960834503 2023-01-23 12:24:04.737434: step: 1282/1541, loss: 0.2666528820991516 2023-01-23 12:24:05.621563: step: 1284/1541, loss: 0.05059148743748665 2023-01-23 12:24:06.492301: step: 1286/1541, loss: 0.03213532269001007 2023-01-23 12:24:07.413425: step: 1288/1541, loss: 0.08148051053285599 2023-01-23 12:24:08.281186: step: 1290/1541, loss: 0.0014101974666118622 2023-01-23 12:24:09.191879: step: 1292/1541, loss: 0.062432948499917984 2023-01-23 12:24:10.120454: step: 1294/1541, loss: 0.06479673087596893 2023-01-23 12:24:11.059450: step: 1296/1541, loss: 0.36482712626457214 2023-01-23 12:24:11.948316: step: 1298/1541, loss: 0.0013269777409732342 2023-01-23 12:24:12.880225: step: 1300/1541, loss: 1.1187676191329956 2023-01-23 12:24:13.768588: step: 1302/1541, loss: 0.0133112957701087 2023-01-23 12:24:14.635314: step: 1304/1541, loss: 0.17886854708194733 2023-01-23 12:24:15.510967: step: 1306/1541, loss: 0.022173499688506126 2023-01-23 12:24:16.448358: step: 1308/1541, loss: 0.244638130068779 2023-01-23 12:24:17.357350: step: 1310/1541, loss: 0.01204416062682867 2023-01-23 12:24:18.328363: step: 1312/1541, loss: 0.236839160323143 2023-01-23 12:24:19.197954: step: 1314/1541, loss: 0.0448889434337616 2023-01-23 12:24:20.122901: step: 1316/1541, loss: 0.46605613827705383 2023-01-23 12:24:21.063172: step: 1318/1541, loss: 0.13211511075496674 2023-01-23 12:24:21.929066: step: 1320/1541, loss: 0.002231219317764044 2023-01-23 12:24:22.886897: step: 1322/1541, loss: 0.15386536717414856 2023-01-23 12:24:23.786686: step: 1324/1541, loss: 0.06379909813404083 2023-01-23 12:24:24.716637: step: 1326/1541, loss: 0.026522386819124222 2023-01-23 12:24:25.672688: step: 1328/1541, loss: 0.09927960485219955 2023-01-23 12:24:26.555162: step: 1330/1541, loss: 0.1853967159986496 2023-01-23 12:24:27.549579: step: 1332/1541, loss: 0.06689920276403427 2023-01-23 12:24:28.421050: step: 1334/1541, loss: 0.013151755556464195 2023-01-23 12:24:29.327137: step: 1336/1541, loss: 0.042977433651685715 2023-01-23 12:24:30.170891: step: 1338/1541, loss: 0.04146670550107956 2023-01-23 12:24:31.126666: step: 1340/1541, loss: 0.19206155836582184 2023-01-23 12:24:32.017320: step: 1342/1541, loss: 0.04338456317782402 2023-01-23 12:24:32.860851: step: 1344/1541, loss: 0.02623971551656723 2023-01-23 12:24:33.751287: step: 1346/1541, loss: 0.021166035905480385 2023-01-23 12:24:34.644984: step: 1348/1541, loss: 0.06502369791269302 2023-01-23 12:24:35.532982: step: 1350/1541, loss: 0.01339706964790821 2023-01-23 12:24:36.460723: step: 1352/1541, loss: 0.2239721566438675 2023-01-23 12:24:37.351541: step: 1354/1541, loss: 0.008032428100705147 2023-01-23 12:24:38.225268: step: 1356/1541, loss: 0.0037073343992233276 2023-01-23 12:24:39.115213: step: 1358/1541, loss: 0.28766924142837524 2023-01-23 12:24:40.074928: step: 1360/1541, loss: 0.30091261863708496 2023-01-23 12:24:40.949796: step: 1362/1541, loss: 0.08732225000858307 2023-01-23 12:24:41.871933: step: 1364/1541, loss: 0.005116126965731382 2023-01-23 12:24:42.776320: step: 1366/1541, loss: 0.00970345176756382 2023-01-23 12:24:43.612675: step: 1368/1541, loss: 0.0022447023075073957 2023-01-23 12:24:44.508607: step: 1370/1541, loss: 0.16836589574813843 2023-01-23 12:24:45.402754: step: 1372/1541, loss: 0.1384894996881485 2023-01-23 12:24:46.325845: step: 1374/1541, loss: 0.0065490189008414745 2023-01-23 12:24:47.219793: step: 1376/1541, loss: 0.927980363368988 2023-01-23 12:24:48.109283: step: 1378/1541, loss: 0.0696093961596489 2023-01-23 12:24:48.990218: step: 1380/1541, loss: 0.13405969738960266 2023-01-23 12:24:49.862171: step: 1382/1541, loss: 0.004751421045511961 2023-01-23 12:24:50.867892: step: 1384/1541, loss: 0.1801052838563919 2023-01-23 12:24:51.806939: step: 1386/1541, loss: 0.005123482085764408 2023-01-23 12:24:52.678897: step: 1388/1541, loss: 0.03953227400779724 2023-01-23 12:24:53.545737: step: 1390/1541, loss: 0.19950848817825317 2023-01-23 12:24:54.492460: step: 1392/1541, loss: 0.015815118327736855 2023-01-23 12:24:55.407932: step: 1394/1541, loss: 0.3475731611251831 2023-01-23 12:24:56.319869: step: 1396/1541, loss: 0.001445733942091465 2023-01-23 12:24:57.252217: step: 1398/1541, loss: 0.25630682706832886 2023-01-23 12:24:58.164363: step: 1400/1541, loss: 0.165665403008461 2023-01-23 12:24:59.040487: step: 1402/1541, loss: 0.07079266756772995 2023-01-23 12:24:59.902130: step: 1404/1541, loss: 0.2622164785861969 2023-01-23 12:25:00.775628: step: 1406/1541, loss: 0.23081539571285248 2023-01-23 12:25:01.665215: step: 1408/1541, loss: 0.12773464620113373 2023-01-23 12:25:02.595577: step: 1410/1541, loss: 0.05323303863406181 2023-01-23 12:25:03.481414: step: 1412/1541, loss: 0.10560065507888794 2023-01-23 12:25:04.526578: step: 1414/1541, loss: 0.011692766100168228 2023-01-23 12:25:05.416305: step: 1416/1541, loss: 0.05036662891507149 2023-01-23 12:25:06.288254: step: 1418/1541, loss: 0.019512634724378586 2023-01-23 12:25:07.185960: step: 1420/1541, loss: 0.057287611067295074 2023-01-23 12:25:08.052158: step: 1422/1541, loss: 0.23020373284816742 2023-01-23 12:25:08.906996: step: 1424/1541, loss: 0.07873715460300446 2023-01-23 12:25:09.772239: step: 1426/1541, loss: 0.005061282776296139 2023-01-23 12:25:10.681460: step: 1428/1541, loss: 0.07998548448085785 2023-01-23 12:25:11.619242: step: 1430/1541, loss: 0.1549142599105835 2023-01-23 12:25:12.530122: step: 1432/1541, loss: 0.1629413515329361 2023-01-23 12:25:13.401354: step: 1434/1541, loss: 0.026314502581954002 2023-01-23 12:25:14.296110: step: 1436/1541, loss: 0.08710683137178421 2023-01-23 12:25:15.205537: step: 1438/1541, loss: 0.02982488088309765 2023-01-23 12:25:16.100123: step: 1440/1541, loss: 0.02066745050251484 2023-01-23 12:25:16.978257: step: 1442/1541, loss: 0.013486879877746105 2023-01-23 12:25:17.857304: step: 1444/1541, loss: 0.46352824568748474 2023-01-23 12:25:18.762978: step: 1446/1541, loss: 0.08666578680276871 2023-01-23 12:25:19.676896: step: 1448/1541, loss: 0.02121368795633316 2023-01-23 12:25:20.600888: step: 1450/1541, loss: 0.09464829415082932 2023-01-23 12:25:21.535684: step: 1452/1541, loss: 0.3238312005996704 2023-01-23 12:25:22.436517: step: 1454/1541, loss: 0.08406972885131836 2023-01-23 12:25:23.336445: step: 1456/1541, loss: 0.012473775073885918 2023-01-23 12:25:24.285958: step: 1458/1541, loss: 0.14196373522281647 2023-01-23 12:25:25.173817: step: 1460/1541, loss: 0.006177019793540239 2023-01-23 12:25:26.151950: step: 1462/1541, loss: 0.21225281059741974 2023-01-23 12:25:27.032944: step: 1464/1541, loss: 0.025884412229061127 2023-01-23 12:25:27.914973: step: 1466/1541, loss: 0.1805267035961151 2023-01-23 12:25:28.806160: step: 1468/1541, loss: 0.10788341611623764 2023-01-23 12:25:29.670938: step: 1470/1541, loss: 0.134514719247818 2023-01-23 12:25:30.585847: step: 1472/1541, loss: 0.15622548758983612 2023-01-23 12:25:31.500795: step: 1474/1541, loss: 0.045702286064624786 2023-01-23 12:25:32.413199: step: 1476/1541, loss: 0.15560133755207062 2023-01-23 12:25:33.367370: step: 1478/1541, loss: 0.10306990891695023 2023-01-23 12:25:34.245399: step: 1480/1541, loss: 0.08220187574625015 2023-01-23 12:25:35.149272: step: 1482/1541, loss: 0.1410474181175232 2023-01-23 12:25:36.089921: step: 1484/1541, loss: 0.19671569764614105 2023-01-23 12:25:36.956008: step: 1486/1541, loss: 0.6485517621040344 2023-01-23 12:25:37.847591: step: 1488/1541, loss: 0.10544095933437347 2023-01-23 12:25:38.727327: step: 1490/1541, loss: 2.1700191497802734 2023-01-23 12:25:39.579700: step: 1492/1541, loss: 0.042050495743751526 2023-01-23 12:25:40.495442: step: 1494/1541, loss: 0.011101284995675087 2023-01-23 12:25:41.405779: step: 1496/1541, loss: 0.017762530595064163 2023-01-23 12:25:42.274513: step: 1498/1541, loss: 0.04840799793601036 2023-01-23 12:25:43.135660: step: 1500/1541, loss: 0.23644740879535675 2023-01-23 12:25:44.020958: step: 1502/1541, loss: 0.2915313243865967 2023-01-23 12:25:44.897468: step: 1504/1541, loss: 0.021890364587306976 2023-01-23 12:25:45.891880: step: 1506/1541, loss: 0.004295239690691233 2023-01-23 12:25:46.797217: step: 1508/1541, loss: 0.13237304985523224 2023-01-23 12:25:47.770648: step: 1510/1541, loss: 0.07026165723800659 2023-01-23 12:25:48.631697: step: 1512/1541, loss: 0.05863840878009796 2023-01-23 12:25:49.578514: step: 1514/1541, loss: 0.3174375295639038 2023-01-23 12:25:50.460446: step: 1516/1541, loss: 0.03497273847460747 2023-01-23 12:25:51.332023: step: 1518/1541, loss: 0.21927744150161743 2023-01-23 12:25:52.266833: step: 1520/1541, loss: 0.05767485126852989 2023-01-23 12:25:53.152229: step: 1522/1541, loss: 0.00922110490500927 2023-01-23 12:25:54.176797: step: 1524/1541, loss: 0.3358911871910095 2023-01-23 12:25:55.082946: step: 1526/1541, loss: 0.02151709608733654 2023-01-23 12:25:56.027567: step: 1528/1541, loss: 0.12274321913719177 2023-01-23 12:25:56.920548: step: 1530/1541, loss: 0.13681605458259583 2023-01-23 12:25:57.812896: step: 1532/1541, loss: 0.01693209819495678 2023-01-23 12:25:58.752522: step: 1534/1541, loss: 0.0049302545376122 2023-01-23 12:25:59.646403: step: 1536/1541, loss: 0.08734680712223053 2023-01-23 12:26:00.516256: step: 1538/1541, loss: 0.010930501855909824 2023-01-23 12:26:01.396489: step: 1540/1541, loss: 0.0017174971289932728 2023-01-23 12:26:02.260846: step: 1542/1541, loss: 0.005116797052323818 2023-01-23 12:26:03.158732: step: 1544/1541, loss: 0.024186454713344574 2023-01-23 12:26:04.130801: step: 1546/1541, loss: 0.02787025459110737 2023-01-23 12:26:05.039345: step: 1548/1541, loss: 0.008637314662337303 2023-01-23 12:26:05.960169: step: 1550/1541, loss: 0.02005905471742153 2023-01-23 12:26:06.843395: step: 1552/1541, loss: 0.0728549137711525 2023-01-23 12:26:07.761848: step: 1554/1541, loss: 0.07897374778985977 2023-01-23 12:26:08.748442: step: 1556/1541, loss: 0.3468884825706482 2023-01-23 12:26:09.660525: step: 1558/1541, loss: 0.004893683362752199 2023-01-23 12:26:10.532601: step: 1560/1541, loss: 0.05803830549120903 2023-01-23 12:26:11.445402: step: 1562/1541, loss: 0.5359712243080139 2023-01-23 12:26:12.341687: step: 1564/1541, loss: 0.07947492599487305 2023-01-23 12:26:13.235929: step: 1566/1541, loss: 0.013636346906423569 2023-01-23 12:26:14.128982: step: 1568/1541, loss: 0.010009421035647392 2023-01-23 12:26:15.035144: step: 1570/1541, loss: 0.22640308737754822 2023-01-23 12:26:15.931784: step: 1572/1541, loss: 0.07321908324956894 2023-01-23 12:26:16.857731: step: 1574/1541, loss: 0.013735473155975342 2023-01-23 12:26:17.759084: step: 1576/1541, loss: 0.005645673722028732 2023-01-23 12:26:18.786393: step: 1578/1541, loss: 0.06631253659725189 2023-01-23 12:26:19.727829: step: 1580/1541, loss: 0.258613258600235 2023-01-23 12:26:20.614670: step: 1582/1541, loss: 0.04029170051217079 2023-01-23 12:26:21.506354: step: 1584/1541, loss: 0.03825920447707176 2023-01-23 12:26:22.402063: step: 1586/1541, loss: 0.0021252152509987354 2023-01-23 12:26:23.349779: step: 1588/1541, loss: 0.01214627269655466 2023-01-23 12:26:24.240051: step: 1590/1541, loss: 0.10896096378564835 2023-01-23 12:26:25.222535: step: 1592/1541, loss: 0.36894282698631287 2023-01-23 12:26:26.121313: step: 1594/1541, loss: 0.0014844542602077127 2023-01-23 12:26:26.993659: step: 1596/1541, loss: 0.8404668569564819 2023-01-23 12:26:27.919814: step: 1598/1541, loss: 0.03354848548769951 2023-01-23 12:26:28.813431: step: 1600/1541, loss: 0.2828894257545471 2023-01-23 12:26:29.690133: step: 1602/1541, loss: 0.005599297117441893 2023-01-23 12:26:30.555663: step: 1604/1541, loss: 0.0869644358754158 2023-01-23 12:26:31.471741: step: 1606/1541, loss: 0.8488907217979431 2023-01-23 12:26:32.377556: step: 1608/1541, loss: 0.009124533273279667 2023-01-23 12:26:33.261029: step: 1610/1541, loss: 0.06921197474002838 2023-01-23 12:26:34.129537: step: 1612/1541, loss: 0.32260382175445557 2023-01-23 12:26:35.027712: step: 1614/1541, loss: 0.7066970467567444 2023-01-23 12:26:35.930184: step: 1616/1541, loss: 0.021857311949133873 2023-01-23 12:26:36.792388: step: 1618/1541, loss: 0.218643918633461 2023-01-23 12:26:37.721866: step: 1620/1541, loss: 0.1327867954969406 2023-01-23 12:26:38.616564: step: 1622/1541, loss: 0.07172999531030655 2023-01-23 12:26:39.539353: step: 1624/1541, loss: 0.025516243651509285 2023-01-23 12:26:40.439605: step: 1626/1541, loss: 0.007215303834527731 2023-01-23 12:26:41.296027: step: 1628/1541, loss: 0.013485627248883247 2023-01-23 12:26:42.159622: step: 1630/1541, loss: 0.1640479862689972 2023-01-23 12:26:43.038795: step: 1632/1541, loss: 0.12028858065605164 2023-01-23 12:26:43.974496: step: 1634/1541, loss: 0.08657946437597275 2023-01-23 12:26:44.923181: step: 1636/1541, loss: 0.25949010252952576 2023-01-23 12:26:45.812146: step: 1638/1541, loss: 0.12249908596277237 2023-01-23 12:26:46.738198: step: 1640/1541, loss: 0.2500786781311035 2023-01-23 12:26:47.644568: step: 1642/1541, loss: 0.06509881466627121 2023-01-23 12:26:48.540705: step: 1644/1541, loss: 0.17096590995788574 2023-01-23 12:26:49.462920: step: 1646/1541, loss: 0.07605469226837158 2023-01-23 12:26:50.370076: step: 1648/1541, loss: 0.209955632686615 2023-01-23 12:26:51.285534: step: 1650/1541, loss: 0.036734286695718765 2023-01-23 12:26:52.159543: step: 1652/1541, loss: 0.015415051020681858 2023-01-23 12:26:53.066454: step: 1654/1541, loss: 0.024451470002532005 2023-01-23 12:26:53.940529: step: 1656/1541, loss: 0.004425848834216595 2023-01-23 12:26:54.834490: step: 1658/1541, loss: 0.0808446854352951 2023-01-23 12:26:55.767480: step: 1660/1541, loss: 0.22285906970500946 2023-01-23 12:26:56.797883: step: 1662/1541, loss: 0.6712421774864197 2023-01-23 12:26:57.694837: step: 1664/1541, loss: 0.04858541861176491 2023-01-23 12:26:58.567019: step: 1666/1541, loss: 0.005691291764378548 2023-01-23 12:26:59.511558: step: 1668/1541, loss: 0.05719343200325966 2023-01-23 12:27:00.381841: step: 1670/1541, loss: 0.2047186642885208 2023-01-23 12:27:01.315739: step: 1672/1541, loss: 0.0991177037358284 2023-01-23 12:27:02.216080: step: 1674/1541, loss: 0.04962170124053955 2023-01-23 12:27:03.156410: step: 1676/1541, loss: 0.033040717244148254 2023-01-23 12:27:04.046214: step: 1678/1541, loss: 0.07816872745752335 2023-01-23 12:27:04.927819: step: 1680/1541, loss: 0.06239880993962288 2023-01-23 12:27:05.832345: step: 1682/1541, loss: 0.029569916427135468 2023-01-23 12:27:06.725263: step: 1684/1541, loss: 0.1533263921737671 2023-01-23 12:27:07.739833: step: 1686/1541, loss: 0.16240710020065308 2023-01-23 12:27:08.627306: step: 1688/1541, loss: 0.020452866330742836 2023-01-23 12:27:09.523539: step: 1690/1541, loss: 0.21534530818462372 2023-01-23 12:27:10.424839: step: 1692/1541, loss: 0.007003274280577898 2023-01-23 12:27:11.357303: step: 1694/1541, loss: 0.013833352364599705 2023-01-23 12:27:12.227969: step: 1696/1541, loss: 0.28511667251586914 2023-01-23 12:27:13.127281: step: 1698/1541, loss: 0.1681157648563385 2023-01-23 12:27:14.007693: step: 1700/1541, loss: 0.13720400631427765 2023-01-23 12:27:14.898291: step: 1702/1541, loss: 0.04106442630290985 2023-01-23 12:27:15.757011: step: 1704/1541, loss: 0.051175061613321304 2023-01-23 12:27:16.690657: step: 1706/1541, loss: 0.11823053658008575 2023-01-23 12:27:17.585770: step: 1708/1541, loss: 0.052353762090206146 2023-01-23 12:27:18.473959: step: 1710/1541, loss: 0.02558225393295288 2023-01-23 12:27:19.429062: step: 1712/1541, loss: 0.7815349698066711 2023-01-23 12:27:20.336409: step: 1714/1541, loss: 0.21403878927230835 2023-01-23 12:27:21.246108: step: 1716/1541, loss: 0.11885490268468857 2023-01-23 12:27:22.162818: step: 1718/1541, loss: 0.1391572654247284 2023-01-23 12:27:23.005771: step: 1720/1541, loss: 0.0575190931558609 2023-01-23 12:27:23.876084: step: 1722/1541, loss: 0.011801248416304588 2023-01-23 12:27:24.761768: step: 1724/1541, loss: 0.18509869277477264 2023-01-23 12:27:25.674385: step: 1726/1541, loss: 0.15428724884986877 2023-01-23 12:27:26.548823: step: 1728/1541, loss: 0.017284253612160683 2023-01-23 12:27:27.439234: step: 1730/1541, loss: 0.2295965701341629 2023-01-23 12:27:28.363196: step: 1732/1541, loss: 0.01976446993649006 2023-01-23 12:27:29.293915: step: 1734/1541, loss: 0.23427903652191162 2023-01-23 12:27:30.219628: step: 1736/1541, loss: 0.018379172310233116 2023-01-23 12:27:31.105377: step: 1738/1541, loss: 0.017666257917881012 2023-01-23 12:27:32.012302: step: 1740/1541, loss: 0.21852366626262665 2023-01-23 12:27:32.922496: step: 1742/1541, loss: 0.016650214791297913 2023-01-23 12:27:33.826389: step: 1744/1541, loss: 0.07837174087762833 2023-01-23 12:27:34.727846: step: 1746/1541, loss: 0.10511161386966705 2023-01-23 12:27:35.617056: step: 1748/1541, loss: 0.012902913615107536 2023-01-23 12:27:36.471754: step: 1750/1541, loss: 0.10323866456747055 2023-01-23 12:27:37.347807: step: 1752/1541, loss: 0.03952514007687569 2023-01-23 12:27:38.301939: step: 1754/1541, loss: 0.11962093412876129 2023-01-23 12:27:39.188346: step: 1756/1541, loss: 0.034932561218738556 2023-01-23 12:27:40.155156: step: 1758/1541, loss: 0.18569403886795044 2023-01-23 12:27:41.077328: step: 1760/1541, loss: 0.16812266409397125 2023-01-23 12:27:41.965029: step: 1762/1541, loss: 0.23802529275417328 2023-01-23 12:27:42.867939: step: 1764/1541, loss: 0.048042114824056625 2023-01-23 12:27:43.766817: step: 1766/1541, loss: 0.7789487838745117 2023-01-23 12:27:44.643956: step: 1768/1541, loss: 0.4309166967868805 2023-01-23 12:27:45.555249: step: 1770/1541, loss: 0.27183279395103455 2023-01-23 12:27:46.443374: step: 1772/1541, loss: 0.09689205139875412 2023-01-23 12:27:47.328646: step: 1774/1541, loss: 0.10616559535264969 2023-01-23 12:27:48.247088: step: 1776/1541, loss: 0.057664912194013596 2023-01-23 12:27:49.204936: step: 1778/1541, loss: 0.12951017916202545 2023-01-23 12:27:50.155618: step: 1780/1541, loss: 0.098833829164505 2023-01-23 12:27:51.010094: step: 1782/1541, loss: 0.20208880305290222 2023-01-23 12:27:51.865202: step: 1784/1541, loss: 0.019334862008690834 2023-01-23 12:27:52.794480: step: 1786/1541, loss: 0.30621615052223206 2023-01-23 12:27:53.663961: step: 1788/1541, loss: 0.11257334798574448 2023-01-23 12:27:54.538613: step: 1790/1541, loss: 0.019886961206793785 2023-01-23 12:27:55.462092: step: 1792/1541, loss: 1.087652564048767 2023-01-23 12:27:56.363987: step: 1794/1541, loss: 0.0932227373123169 2023-01-23 12:27:57.254844: step: 1796/1541, loss: 0.030279455706477165 2023-01-23 12:27:58.223075: step: 1798/1541, loss: 0.17689646780490875 2023-01-23 12:27:59.110714: step: 1800/1541, loss: 0.2710043489933014 2023-01-23 12:27:59.989859: step: 1802/1541, loss: 0.10125324130058289 2023-01-23 12:28:00.884022: step: 1804/1541, loss: 0.06098519638180733 2023-01-23 12:28:01.766612: step: 1806/1541, loss: 0.06295540928840637 2023-01-23 12:28:02.706549: step: 1808/1541, loss: 0.07159898430109024 2023-01-23 12:28:03.570189: step: 1810/1541, loss: 0.03822444751858711 2023-01-23 12:28:04.435938: step: 1812/1541, loss: 0.30378755927085876 2023-01-23 12:28:05.366348: step: 1814/1541, loss: 0.20409969985485077 2023-01-23 12:28:06.244262: step: 1816/1541, loss: 0.12349346280097961 2023-01-23 12:28:07.097098: step: 1818/1541, loss: 0.059159234166145325 2023-01-23 12:28:07.984650: step: 1820/1541, loss: 0.08148153126239777 2023-01-23 12:28:08.883816: step: 1822/1541, loss: 0.13433010876178741 2023-01-23 12:28:09.787924: step: 1824/1541, loss: 0.3276512026786804 2023-01-23 12:28:10.690937: step: 1826/1541, loss: 0.0265662744641304 2023-01-23 12:28:11.621608: step: 1828/1541, loss: 0.19423557817935944 2023-01-23 12:28:12.505667: step: 1830/1541, loss: 0.04843945428729057 2023-01-23 12:28:13.377979: step: 1832/1541, loss: 0.11701148003339767 2023-01-23 12:28:14.252619: step: 1834/1541, loss: 0.11851757764816284 2023-01-23 12:28:15.161457: step: 1836/1541, loss: 0.5786503553390503 2023-01-23 12:28:16.103613: step: 1838/1541, loss: 0.027230150997638702 2023-01-23 12:28:16.987887: step: 1840/1541, loss: 0.14643900096416473 2023-01-23 12:28:17.880139: step: 1842/1541, loss: 0.04577215760946274 2023-01-23 12:28:18.774307: step: 1844/1541, loss: 0.06938089430332184 2023-01-23 12:28:19.801140: step: 1846/1541, loss: 0.07569759339094162 2023-01-23 12:28:20.729885: step: 1848/1541, loss: 0.06675612181425095 2023-01-23 12:28:21.634192: step: 1850/1541, loss: 0.07826977223157883 2023-01-23 12:28:22.555369: step: 1852/1541, loss: 0.05066428333520889 2023-01-23 12:28:23.511844: step: 1854/1541, loss: 0.0678812563419342 2023-01-23 12:28:24.407353: step: 1856/1541, loss: 0.039288002997636795 2023-01-23 12:28:25.286313: step: 1858/1541, loss: 0.0575370229780674 2023-01-23 12:28:26.149772: step: 1860/1541, loss: 0.31783759593963623 2023-01-23 12:28:27.092294: step: 1862/1541, loss: 0.08434776216745377 2023-01-23 12:28:28.080861: step: 1864/1541, loss: 0.20690275728702545 2023-01-23 12:28:28.994513: step: 1866/1541, loss: 0.034825053066015244 2023-01-23 12:28:29.893877: step: 1868/1541, loss: 0.3521793484687805 2023-01-23 12:28:30.812053: step: 1870/1541, loss: 0.23951756954193115 2023-01-23 12:28:31.730359: step: 1872/1541, loss: 0.01701902411878109 2023-01-23 12:28:32.603504: step: 1874/1541, loss: 0.8965387344360352 2023-01-23 12:28:33.542325: step: 1876/1541, loss: 0.03957081213593483 2023-01-23 12:28:34.413474: step: 1878/1541, loss: 0.03000400774180889 2023-01-23 12:28:35.316437: step: 1880/1541, loss: 0.07325729727745056 2023-01-23 12:28:36.211364: step: 1882/1541, loss: 0.07124800980091095 2023-01-23 12:28:37.079482: step: 1884/1541, loss: 0.00849496852606535 2023-01-23 12:28:37.961122: step: 1886/1541, loss: 0.009440735913813114 2023-01-23 12:28:38.823053: step: 1888/1541, loss: 0.02897210419178009 2023-01-23 12:28:39.732286: step: 1890/1541, loss: 0.016289900988340378 2023-01-23 12:28:40.639953: step: 1892/1541, loss: 0.008583221584558487 2023-01-23 12:28:41.557115: step: 1894/1541, loss: 0.01250504981726408 2023-01-23 12:28:42.419004: step: 1896/1541, loss: 0.2970149517059326 2023-01-23 12:28:43.301844: step: 1898/1541, loss: 0.08184446394443512 2023-01-23 12:28:44.223614: step: 1900/1541, loss: 0.1829090267419815 2023-01-23 12:28:45.110826: step: 1902/1541, loss: 0.04623909294605255 2023-01-23 12:28:46.042763: step: 1904/1541, loss: 0.13487157225608826 2023-01-23 12:28:46.959080: step: 1906/1541, loss: 0.05611436441540718 2023-01-23 12:28:47.926127: step: 1908/1541, loss: 0.10933244228363037 2023-01-23 12:28:48.832075: step: 1910/1541, loss: 0.12137977033853531 2023-01-23 12:28:49.761846: step: 1912/1541, loss: 0.22001653909683228 2023-01-23 12:28:50.610785: step: 1914/1541, loss: 0.03317508101463318 2023-01-23 12:28:51.660509: step: 1916/1541, loss: 0.038204729557037354 2023-01-23 12:28:52.575339: step: 1918/1541, loss: 0.2152019590139389 2023-01-23 12:28:53.525635: step: 1920/1541, loss: 0.08458178490400314 2023-01-23 12:28:54.429758: step: 1922/1541, loss: 0.08970504999160767 2023-01-23 12:28:55.307915: step: 1924/1541, loss: 0.03630514442920685 2023-01-23 12:28:56.216832: step: 1926/1541, loss: 0.008083013817667961 2023-01-23 12:28:57.095905: step: 1928/1541, loss: 0.05955272167921066 2023-01-23 12:28:58.023319: step: 1930/1541, loss: 0.08465693891048431 2023-01-23 12:28:58.900547: step: 1932/1541, loss: 0.060640014708042145 2023-01-23 12:28:59.764878: step: 1934/1541, loss: 0.11236763000488281 2023-01-23 12:29:00.701575: step: 1936/1541, loss: 0.16265352070331573 2023-01-23 12:29:01.610872: step: 1938/1541, loss: 0.01598764955997467 2023-01-23 12:29:02.537014: step: 1940/1541, loss: 0.15249693393707275 2023-01-23 12:29:03.440392: step: 1942/1541, loss: 0.19197288155555725 2023-01-23 12:29:04.384826: step: 1944/1541, loss: 1.3260648250579834 2023-01-23 12:29:05.287529: step: 1946/1541, loss: 0.24772204458713531 2023-01-23 12:29:06.173566: step: 1948/1541, loss: 0.007884248159825802 2023-01-23 12:29:07.069211: step: 1950/1541, loss: 0.22248892486095428 2023-01-23 12:29:07.995669: step: 1952/1541, loss: 0.3108612596988678 2023-01-23 12:29:08.887393: step: 1954/1541, loss: 0.02802818827331066 2023-01-23 12:29:09.792652: step: 1956/1541, loss: 0.015705596655607224 2023-01-23 12:29:10.707849: step: 1958/1541, loss: 0.0711275041103363 2023-01-23 12:29:11.664213: step: 1960/1541, loss: 0.10159614682197571 2023-01-23 12:29:12.559069: step: 1962/1541, loss: 0.1390264630317688 2023-01-23 12:29:13.526357: step: 1964/1541, loss: 0.07786591351032257 2023-01-23 12:29:14.429068: step: 1966/1541, loss: 0.005207784473896027 2023-01-23 12:29:15.340954: step: 1968/1541, loss: 0.2739327847957611 2023-01-23 12:29:16.240423: step: 1970/1541, loss: 0.23184829950332642 2023-01-23 12:29:17.114309: step: 1972/1541, loss: 0.020764397457242012 2023-01-23 12:29:18.038924: step: 1974/1541, loss: 0.01559411734342575 2023-01-23 12:29:18.936377: step: 1976/1541, loss: 0.014862454496324062 2023-01-23 12:29:19.806166: step: 1978/1541, loss: 0.02583685703575611 2023-01-23 12:29:20.699371: step: 1980/1541, loss: 0.03652186319231987 2023-01-23 12:29:21.612855: step: 1982/1541, loss: 0.21305516362190247 2023-01-23 12:29:22.511613: step: 1984/1541, loss: 0.014463057741522789 2023-01-23 12:29:23.447292: step: 1986/1541, loss: 0.1475120186805725 2023-01-23 12:29:24.376934: step: 1988/1541, loss: 0.14335164427757263 2023-01-23 12:29:25.236909: step: 1990/1541, loss: 0.015109464526176453 2023-01-23 12:29:26.123415: step: 1992/1541, loss: 0.011218358762562275 2023-01-23 12:29:27.016325: step: 1994/1541, loss: 0.013156261295080185 2023-01-23 12:29:28.015198: step: 1996/1541, loss: 0.15872906148433685 2023-01-23 12:29:28.932021: step: 1998/1541, loss: 0.003573363646864891 2023-01-23 12:29:29.835414: step: 2000/1541, loss: 0.5825691223144531 2023-01-23 12:29:30.737897: step: 2002/1541, loss: 0.05914817750453949 2023-01-23 12:29:31.647798: step: 2004/1541, loss: 0.07019980996847153 2023-01-23 12:29:32.638235: step: 2006/1541, loss: 0.2002788931131363 2023-01-23 12:29:33.612011: step: 2008/1541, loss: 0.05304823815822601 2023-01-23 12:29:34.625611: step: 2010/1541, loss: 0.13866794109344482 2023-01-23 12:29:35.538420: step: 2012/1541, loss: 0.018854279071092606 2023-01-23 12:29:36.455652: step: 2014/1541, loss: 0.06039100140333176 2023-01-23 12:29:37.321124: step: 2016/1541, loss: 0.39202064275741577 2023-01-23 12:29:38.201948: step: 2018/1541, loss: 0.1672629863023758 2023-01-23 12:29:39.109958: step: 2020/1541, loss: 0.012276675552129745 2023-01-23 12:29:40.020007: step: 2022/1541, loss: 0.05670721083879471 2023-01-23 12:29:40.936444: step: 2024/1541, loss: 0.05877874046564102 2023-01-23 12:29:41.845095: step: 2026/1541, loss: 0.002329899463802576 2023-01-23 12:29:42.710368: step: 2028/1541, loss: 3.2844698429107666 2023-01-23 12:29:43.671367: step: 2030/1541, loss: 0.09185227751731873 2023-01-23 12:29:44.586859: step: 2032/1541, loss: 0.13238930702209473 2023-01-23 12:29:45.452282: step: 2034/1541, loss: 0.0020452444441616535 2023-01-23 12:29:46.317279: step: 2036/1541, loss: 0.10912314057350159 2023-01-23 12:29:47.179230: step: 2038/1541, loss: 0.06988765299320221 2023-01-23 12:29:48.107287: step: 2040/1541, loss: 0.15654803812503815 2023-01-23 12:29:48.979158: step: 2042/1541, loss: 0.3009239733219147 2023-01-23 12:29:49.881530: step: 2044/1541, loss: 0.15211521089076996 2023-01-23 12:29:50.744794: step: 2046/1541, loss: 0.14013785123825073 2023-01-23 12:29:51.630691: step: 2048/1541, loss: 0.11475728452205658 2023-01-23 12:29:52.522000: step: 2050/1541, loss: 0.05518459901213646 2023-01-23 12:29:53.435837: step: 2052/1541, loss: 0.33101627230644226 2023-01-23 12:29:54.326976: step: 2054/1541, loss: 0.40225884318351746 2023-01-23 12:29:55.212777: step: 2056/1541, loss: 0.4764920473098755 2023-01-23 12:29:56.109125: step: 2058/1541, loss: 0.03987084701657295 2023-01-23 12:29:56.995699: step: 2060/1541, loss: 0.7609792947769165 2023-01-23 12:29:57.883803: step: 2062/1541, loss: 0.024072108790278435 2023-01-23 12:29:58.735950: step: 2064/1541, loss: 0.006493483670055866 2023-01-23 12:29:59.665921: step: 2066/1541, loss: 0.21880708634853363 2023-01-23 12:30:00.561876: step: 2068/1541, loss: 0.14670798182487488 2023-01-23 12:30:01.502253: step: 2070/1541, loss: 0.11260728538036346 2023-01-23 12:30:02.453946: step: 2072/1541, loss: 0.09540862590074539 2023-01-23 12:30:03.319329: step: 2074/1541, loss: 0.025848327204585075 2023-01-23 12:30:04.194443: step: 2076/1541, loss: 0.13149192929267883 2023-01-23 12:30:05.115927: step: 2078/1541, loss: 0.02158258482813835 2023-01-23 12:30:05.946323: step: 2080/1541, loss: 0.033148955553770065 2023-01-23 12:30:06.869628: step: 2082/1541, loss: 0.014545421116054058 2023-01-23 12:30:07.761973: step: 2084/1541, loss: 1.1601377725601196 2023-01-23 12:30:08.652449: step: 2086/1541, loss: 0.22669605910778046 2023-01-23 12:30:09.565953: step: 2088/1541, loss: 0.013596056029200554 2023-01-23 12:30:10.429001: step: 2090/1541, loss: 0.039001792669296265 2023-01-23 12:30:11.276434: step: 2092/1541, loss: 0.766584575176239 2023-01-23 12:30:12.187396: step: 2094/1541, loss: 0.025588396936655045 2023-01-23 12:30:13.098321: step: 2096/1541, loss: 0.05331654101610184 2023-01-23 12:30:14.017953: step: 2098/1541, loss: 0.1116505041718483 2023-01-23 12:30:14.924385: step: 2100/1541, loss: 0.0754842460155487 2023-01-23 12:30:15.890006: step: 2102/1541, loss: 0.0060253506526350975 2023-01-23 12:30:16.789894: step: 2104/1541, loss: 0.015247531235218048 2023-01-23 12:30:17.688873: step: 2106/1541, loss: 0.03451257944107056 2023-01-23 12:30:18.610700: step: 2108/1541, loss: 0.027700994163751602 2023-01-23 12:30:19.495247: step: 2110/1541, loss: 0.21062670648097992 2023-01-23 12:30:20.359369: step: 2112/1541, loss: 0.02369949221611023 2023-01-23 12:30:21.245094: step: 2114/1541, loss: 0.1409928798675537 2023-01-23 12:30:22.143351: step: 2116/1541, loss: 0.4671686887741089 2023-01-23 12:30:23.017346: step: 2118/1541, loss: 0.014587531797587872 2023-01-23 12:30:23.919290: step: 2120/1541, loss: 0.17009378969669342 2023-01-23 12:30:24.801851: step: 2122/1541, loss: 0.026341678574681282 2023-01-23 12:30:25.721073: step: 2124/1541, loss: 0.0031869057565927505 2023-01-23 12:30:26.630443: step: 2126/1541, loss: 0.04423340782523155 2023-01-23 12:30:27.589333: step: 2128/1541, loss: 0.07206705212593079 2023-01-23 12:30:28.522181: step: 2130/1541, loss: 0.4358288645744324 2023-01-23 12:30:29.398210: step: 2132/1541, loss: 1.0068930387496948 2023-01-23 12:30:30.300089: step: 2134/1541, loss: 0.21053044497966766 2023-01-23 12:30:31.209717: step: 2136/1541, loss: 0.03624286502599716 2023-01-23 12:30:32.218766: step: 2138/1541, loss: 0.41819506883621216 2023-01-23 12:30:33.107744: step: 2140/1541, loss: 0.09010125696659088 2023-01-23 12:30:34.000829: step: 2142/1541, loss: 0.03511147201061249 2023-01-23 12:30:34.924187: step: 2144/1541, loss: 0.576751172542572 2023-01-23 12:30:35.872326: step: 2146/1541, loss: 0.2721478044986725 2023-01-23 12:30:36.770833: step: 2148/1541, loss: 0.019599132239818573 2023-01-23 12:30:37.784613: step: 2150/1541, loss: 0.0622742623090744 2023-01-23 12:30:38.692821: step: 2152/1541, loss: 0.034931983798742294 2023-01-23 12:30:39.554718: step: 2154/1541, loss: 0.07854252308607101 2023-01-23 12:30:40.430467: step: 2156/1541, loss: 0.005304573569446802 2023-01-23 12:30:41.336888: step: 2158/1541, loss: 0.21126727759838104 2023-01-23 12:30:42.296750: step: 2160/1541, loss: 0.003021918935701251 2023-01-23 12:30:43.236121: step: 2162/1541, loss: 0.3749352991580963 2023-01-23 12:30:44.114554: step: 2164/1541, loss: 0.06737679243087769 2023-01-23 12:30:45.008944: step: 2166/1541, loss: 0.004723746329545975 2023-01-23 12:30:45.892556: step: 2168/1541, loss: 0.1157737597823143 2023-01-23 12:30:46.756697: step: 2170/1541, loss: 0.13838927447795868 2023-01-23 12:30:47.656787: step: 2172/1541, loss: 0.05091153085231781 2023-01-23 12:30:48.534827: step: 2174/1541, loss: 0.6167060732841492 2023-01-23 12:30:49.431045: step: 2176/1541, loss: 0.47606584429740906 2023-01-23 12:30:50.325153: step: 2178/1541, loss: 0.10965947806835175 2023-01-23 12:30:51.227499: step: 2180/1541, loss: 0.0864923745393753 2023-01-23 12:30:52.153464: step: 2182/1541, loss: 0.02031133696436882 2023-01-23 12:30:53.057466: step: 2184/1541, loss: 0.041503388434648514 2023-01-23 12:30:53.927813: step: 2186/1541, loss: 0.01736896112561226 2023-01-23 12:30:54.841940: step: 2188/1541, loss: 0.019945036619901657 2023-01-23 12:30:55.711907: step: 2190/1541, loss: 0.12560497224330902 2023-01-23 12:30:56.684965: step: 2192/1541, loss: 0.05524826422333717 2023-01-23 12:30:57.543857: step: 2194/1541, loss: 0.22767193615436554 2023-01-23 12:30:58.488366: step: 2196/1541, loss: 0.04867284372448921 2023-01-23 12:30:59.367784: step: 2198/1541, loss: 0.15259653329849243 2023-01-23 12:31:00.254242: step: 2200/1541, loss: 0.1543745994567871 2023-01-23 12:31:01.213799: step: 2202/1541, loss: 0.04746248945593834 2023-01-23 12:31:02.074127: step: 2204/1541, loss: 0.09268403053283691 2023-01-23 12:31:03.055919: step: 2206/1541, loss: 0.1699354648590088 2023-01-23 12:31:04.017413: step: 2208/1541, loss: 1.1270169019699097 2023-01-23 12:31:04.861694: step: 2210/1541, loss: 0.12459250539541245 2023-01-23 12:31:05.725711: step: 2212/1541, loss: 0.11283964663743973 2023-01-23 12:31:06.623284: step: 2214/1541, loss: 0.00868627242743969 2023-01-23 12:31:07.514812: step: 2216/1541, loss: 0.13987913727760315 2023-01-23 12:31:08.388332: step: 2218/1541, loss: 0.34032243490219116 2023-01-23 12:31:09.299610: step: 2220/1541, loss: 0.25140294432640076 2023-01-23 12:31:10.152544: step: 2222/1541, loss: 0.079492948949337 2023-01-23 12:31:11.026908: step: 2224/1541, loss: 0.05588265880942345 2023-01-23 12:31:11.932310: step: 2226/1541, loss: 0.39704054594039917 2023-01-23 12:31:12.827989: step: 2228/1541, loss: 0.0021342700347304344 2023-01-23 12:31:13.706066: step: 2230/1541, loss: 0.04470738023519516 2023-01-23 12:31:14.658618: step: 2232/1541, loss: 0.07028640061616898 2023-01-23 12:31:15.553204: step: 2234/1541, loss: 0.00508405826985836 2023-01-23 12:31:16.450960: step: 2236/1541, loss: 0.6397021412849426 2023-01-23 12:31:17.370835: step: 2238/1541, loss: 0.012999179773032665 2023-01-23 12:31:18.239655: step: 2240/1541, loss: 0.006998000200837851 2023-01-23 12:31:19.109152: step: 2242/1541, loss: 0.2180861532688141 2023-01-23 12:31:20.046925: step: 2244/1541, loss: 0.08550728112459183 2023-01-23 12:31:20.909423: step: 2246/1541, loss: 0.035655997693538666 2023-01-23 12:31:21.804346: step: 2248/1541, loss: 0.038232576102018356 2023-01-23 12:31:22.752082: step: 2250/1541, loss: 0.01869918592274189 2023-01-23 12:31:23.604280: step: 2252/1541, loss: 0.18385040760040283 2023-01-23 12:31:24.514380: step: 2254/1541, loss: 0.21289819478988647 2023-01-23 12:31:25.428951: step: 2256/1541, loss: 0.21259555220603943 2023-01-23 12:31:26.357075: step: 2258/1541, loss: 0.08582721650600433 2023-01-23 12:31:27.264454: step: 2260/1541, loss: 0.552303671836853 2023-01-23 12:31:28.146351: step: 2262/1541, loss: 0.0385344959795475 2023-01-23 12:31:29.046108: step: 2264/1541, loss: 0.04049859941005707 2023-01-23 12:31:29.930664: step: 2266/1541, loss: 0.1637614667415619 2023-01-23 12:31:30.806869: step: 2268/1541, loss: 0.00494881346821785 2023-01-23 12:31:31.690946: step: 2270/1541, loss: 0.07249116152524948 2023-01-23 12:31:32.589859: step: 2272/1541, loss: 0.020362488925457 2023-01-23 12:31:33.498102: step: 2274/1541, loss: 0.08286383748054504 2023-01-23 12:31:34.412764: step: 2276/1541, loss: 0.5795413255691528 2023-01-23 12:31:35.317126: step: 2278/1541, loss: 0.003481984371319413 2023-01-23 12:31:36.250750: step: 2280/1541, loss: 0.25433653593063354 2023-01-23 12:31:37.200405: step: 2282/1541, loss: 0.0066746072843670845 2023-01-23 12:31:38.100607: step: 2284/1541, loss: 0.017411671578884125 2023-01-23 12:31:38.978365: step: 2286/1541, loss: 0.037071775645017624 2023-01-23 12:31:39.839289: step: 2288/1541, loss: 0.04684513807296753 2023-01-23 12:31:40.727435: step: 2290/1541, loss: 0.0024079452268779278 2023-01-23 12:31:41.592379: step: 2292/1541, loss: 0.061114951968193054 2023-01-23 12:31:42.522391: step: 2294/1541, loss: 0.2080199271440506 2023-01-23 12:31:43.435392: step: 2296/1541, loss: 0.027639083564281464 2023-01-23 12:31:44.390890: step: 2298/1541, loss: 0.008934145793318748 2023-01-23 12:31:45.347308: step: 2300/1541, loss: 0.34857234358787537 2023-01-23 12:31:46.263744: step: 2302/1541, loss: 0.07540392875671387 2023-01-23 12:31:47.209993: step: 2304/1541, loss: 0.18330536782741547 2023-01-23 12:31:48.147300: step: 2306/1541, loss: 0.014032822102308273 2023-01-23 12:31:49.016152: step: 2308/1541, loss: 0.053097449243068695 2023-01-23 12:31:49.935578: step: 2310/1541, loss: 1.8890533447265625 2023-01-23 12:31:50.831251: step: 2312/1541, loss: 0.0055419206619262695 2023-01-23 12:31:51.716878: step: 2314/1541, loss: 0.1301662176847458 2023-01-23 12:31:52.616847: step: 2316/1541, loss: 0.18965506553649902 2023-01-23 12:31:53.479912: step: 2318/1541, loss: 0.012924890965223312 2023-01-23 12:31:54.367052: step: 2320/1541, loss: 0.05280700698494911 2023-01-23 12:31:55.225086: step: 2322/1541, loss: 0.04222112521529198 2023-01-23 12:31:56.082093: step: 2324/1541, loss: 0.4608892798423767 2023-01-23 12:31:56.994452: step: 2326/1541, loss: 0.01025965716689825 2023-01-23 12:31:57.946763: step: 2328/1541, loss: 0.056807659566402435 2023-01-23 12:31:58.838332: step: 2330/1541, loss: 0.30824941396713257 2023-01-23 12:31:59.719118: step: 2332/1541, loss: 0.0462627187371254 2023-01-23 12:32:00.644500: step: 2334/1541, loss: 0.009293588809669018 2023-01-23 12:32:01.562647: step: 2336/1541, loss: 0.08656096458435059 2023-01-23 12:32:02.439705: step: 2338/1541, loss: 0.18139921128749847 2023-01-23 12:32:03.337775: step: 2340/1541, loss: 0.08112409710884094 2023-01-23 12:32:04.258261: step: 2342/1541, loss: 0.28468745946884155 2023-01-23 12:32:05.172692: step: 2344/1541, loss: 5.446339130401611 2023-01-23 12:32:06.056576: step: 2346/1541, loss: 1.0048332214355469 2023-01-23 12:32:06.919552: step: 2348/1541, loss: 0.4007885158061981 2023-01-23 12:32:07.840002: step: 2350/1541, loss: 0.06901548057794571 2023-01-23 12:32:08.735233: step: 2352/1541, loss: 0.011031687259674072 2023-01-23 12:32:09.604679: step: 2354/1541, loss: 0.05122764781117439 2023-01-23 12:32:10.491202: step: 2356/1541, loss: 0.09611912816762924 2023-01-23 12:32:11.336358: step: 2358/1541, loss: 0.31338924169540405 2023-01-23 12:32:12.224424: step: 2360/1541, loss: 0.11332530528306961 2023-01-23 12:32:13.179363: step: 2362/1541, loss: 0.11883392930030823 2023-01-23 12:32:14.051376: step: 2364/1541, loss: 1.2544710636138916 2023-01-23 12:32:15.000557: step: 2366/1541, loss: 0.4969238340854645 2023-01-23 12:32:15.898351: step: 2368/1541, loss: 0.5786892771720886 2023-01-23 12:32:16.792013: step: 2370/1541, loss: 0.09712295234203339 2023-01-23 12:32:17.681422: step: 2372/1541, loss: 0.002954674419015646 2023-01-23 12:32:18.551706: step: 2374/1541, loss: 0.0669102817773819 2023-01-23 12:32:19.431192: step: 2376/1541, loss: 0.1670268326997757 2023-01-23 12:32:20.283617: step: 2378/1541, loss: 0.03496754914522171 2023-01-23 12:32:21.143663: step: 2380/1541, loss: 0.2237381935119629 2023-01-23 12:32:22.049622: step: 2382/1541, loss: 0.10303898900747299 2023-01-23 12:32:22.997629: step: 2384/1541, loss: 0.11998553574085236 2023-01-23 12:32:23.873562: step: 2386/1541, loss: 0.011977944523096085 2023-01-23 12:32:24.778855: step: 2388/1541, loss: 0.005616259761154652 2023-01-23 12:32:25.735250: step: 2390/1541, loss: 0.043724022805690765 2023-01-23 12:32:26.584263: step: 2392/1541, loss: 0.06310482323169708 2023-01-23 12:32:27.469052: step: 2394/1541, loss: 0.056148428469896317 2023-01-23 12:32:28.349729: step: 2396/1541, loss: 0.03752695024013519 2023-01-23 12:32:29.281707: step: 2398/1541, loss: 0.09219352155923843 2023-01-23 12:32:30.235422: step: 2400/1541, loss: 0.34390515089035034 2023-01-23 12:32:31.130843: step: 2402/1541, loss: 0.010763529688119888 2023-01-23 12:32:32.053723: step: 2404/1541, loss: 0.06098013371229172 2023-01-23 12:32:32.907097: step: 2406/1541, loss: 0.35742706060409546 2023-01-23 12:32:33.786080: step: 2408/1541, loss: 0.012909173965454102 2023-01-23 12:32:34.667588: step: 2410/1541, loss: 0.04658494517207146 2023-01-23 12:32:35.531204: step: 2412/1541, loss: 0.015860548242926598 2023-01-23 12:32:36.404858: step: 2414/1541, loss: 0.09847991168498993 2023-01-23 12:32:37.397055: step: 2416/1541, loss: 0.010077746585011482 2023-01-23 12:32:38.313724: step: 2418/1541, loss: 0.012798735871911049 2023-01-23 12:32:39.182795: step: 2420/1541, loss: 0.23020224273204803 2023-01-23 12:32:40.086374: step: 2422/1541, loss: 0.028584465384483337 2023-01-23 12:32:40.993199: step: 2424/1541, loss: 0.008956581354141235 2023-01-23 12:32:41.852853: step: 2426/1541, loss: 0.03720313310623169 2023-01-23 12:32:42.775646: step: 2428/1541, loss: 0.07685024291276932 2023-01-23 12:32:43.725586: step: 2430/1541, loss: 0.0646616742014885 2023-01-23 12:32:44.606840: step: 2432/1541, loss: 0.14471061527729034 2023-01-23 12:32:45.550948: step: 2434/1541, loss: 0.015796350315213203 2023-01-23 12:32:46.456349: step: 2436/1541, loss: 0.7711708545684814 2023-01-23 12:32:47.393066: step: 2438/1541, loss: 0.08292678743600845 2023-01-23 12:32:48.294304: step: 2440/1541, loss: 0.18761450052261353 2023-01-23 12:32:49.164979: step: 2442/1541, loss: 0.2854921221733093 2023-01-23 12:32:50.086861: step: 2444/1541, loss: 0.009301014244556427 2023-01-23 12:32:50.976889: step: 2446/1541, loss: 0.003088220488280058 2023-01-23 12:32:51.851092: step: 2448/1541, loss: 0.028507689014077187 2023-01-23 12:32:52.740900: step: 2450/1541, loss: 0.007902377285063267 2023-01-23 12:32:53.641697: step: 2452/1541, loss: 0.0423688106238842 2023-01-23 12:32:54.511044: step: 2454/1541, loss: 0.02455521747469902 2023-01-23 12:32:55.381225: step: 2456/1541, loss: 0.02624962478876114 2023-01-23 12:32:56.255346: step: 2458/1541, loss: 0.10336019098758698 2023-01-23 12:32:57.247229: step: 2460/1541, loss: 0.0366738922894001 2023-01-23 12:32:58.142432: step: 2462/1541, loss: 0.1817064732313156 2023-01-23 12:32:59.060278: step: 2464/1541, loss: 0.0006517125875689089 2023-01-23 12:32:59.970876: step: 2466/1541, loss: 0.04982319846749306 2023-01-23 12:33:00.862146: step: 2468/1541, loss: 0.5164476037025452 2023-01-23 12:33:01.765676: step: 2470/1541, loss: 0.17115898430347443 2023-01-23 12:33:02.657179: step: 2472/1541, loss: 0.17743046581745148 2023-01-23 12:33:03.566766: step: 2474/1541, loss: 0.28806719183921814 2023-01-23 12:33:04.438192: step: 2476/1541, loss: 0.03459771350026131 2023-01-23 12:33:05.310108: step: 2478/1541, loss: 0.017490772530436516 2023-01-23 12:33:06.234942: step: 2480/1541, loss: 0.03164055570960045 2023-01-23 12:33:07.104154: step: 2482/1541, loss: 0.18944963812828064 2023-01-23 12:33:07.990535: step: 2484/1541, loss: 0.03512324020266533 2023-01-23 12:33:08.888789: step: 2486/1541, loss: 0.2523486316204071 2023-01-23 12:33:09.798468: step: 2488/1541, loss: 1.0991227626800537 2023-01-23 12:33:10.672651: step: 2490/1541, loss: 0.27752068638801575 2023-01-23 12:33:11.554372: step: 2492/1541, loss: 0.03281470015645027 2023-01-23 12:33:12.504190: step: 2494/1541, loss: 0.0058271633461117744 2023-01-23 12:33:13.395767: step: 2496/1541, loss: 0.17048020660877228 2023-01-23 12:33:14.344161: step: 2498/1541, loss: 0.24364164471626282 2023-01-23 12:33:15.236988: step: 2500/1541, loss: 0.23765480518341064 2023-01-23 12:33:16.148130: step: 2502/1541, loss: 0.009489988908171654 2023-01-23 12:33:17.062999: step: 2504/1541, loss: 0.4682339131832123 2023-01-23 12:33:17.937494: step: 2506/1541, loss: 0.17500799894332886 2023-01-23 12:33:18.835021: step: 2508/1541, loss: 0.02139933966100216 2023-01-23 12:33:19.743286: step: 2510/1541, loss: 0.16655611991882324 2023-01-23 12:33:20.629101: step: 2512/1541, loss: 0.17890119552612305 2023-01-23 12:33:21.494627: step: 2514/1541, loss: 0.018240071833133698 2023-01-23 12:33:22.325019: step: 2516/1541, loss: 0.02652296982705593 2023-01-23 12:33:23.230762: step: 2518/1541, loss: 0.16893276572227478 2023-01-23 12:33:24.135200: step: 2520/1541, loss: 0.029514191672205925 2023-01-23 12:33:25.071071: step: 2522/1541, loss: 0.20839239656925201 2023-01-23 12:33:25.957626: step: 2524/1541, loss: 0.09072870761156082 2023-01-23 12:33:26.850463: step: 2526/1541, loss: 0.16822342574596405 2023-01-23 12:33:27.712400: step: 2528/1541, loss: 0.2192280888557434 2023-01-23 12:33:28.655976: step: 2530/1541, loss: 0.043392494320869446 2023-01-23 12:33:29.533113: step: 2532/1541, loss: 0.044857557862997055 2023-01-23 12:33:30.425911: step: 2534/1541, loss: 0.13045057654380798 2023-01-23 12:33:31.329170: step: 2536/1541, loss: 0.13843919336795807 2023-01-23 12:33:32.240011: step: 2538/1541, loss: 0.10332716256380081 2023-01-23 12:33:33.159441: step: 2540/1541, loss: 0.5384786128997803 2023-01-23 12:33:34.021426: step: 2542/1541, loss: 0.1133793294429779 2023-01-23 12:33:34.991535: step: 2544/1541, loss: 0.0027301169466227293 2023-01-23 12:33:35.974863: step: 2546/1541, loss: 0.042293209582567215 2023-01-23 12:33:36.900649: step: 2548/1541, loss: 0.03500259295105934 2023-01-23 12:33:37.823633: step: 2550/1541, loss: 0.051101718097925186 2023-01-23 12:33:38.735996: step: 2552/1541, loss: 0.04746999591588974 2023-01-23 12:33:39.746298: step: 2554/1541, loss: 0.055825695395469666 2023-01-23 12:33:40.671352: step: 2556/1541, loss: 0.17529040575027466 2023-01-23 12:33:41.551269: step: 2558/1541, loss: 0.4500895142555237 2023-01-23 12:33:42.470565: step: 2560/1541, loss: 0.03711312264204025 2023-01-23 12:33:43.351951: step: 2562/1541, loss: 0.1591656357049942 2023-01-23 12:33:44.248381: step: 2564/1541, loss: 0.16571035981178284 2023-01-23 12:33:45.185779: step: 2566/1541, loss: 0.1229952797293663 2023-01-23 12:33:46.056422: step: 2568/1541, loss: 0.047997601330280304 2023-01-23 12:33:46.962395: step: 2570/1541, loss: 0.058264195919036865 2023-01-23 12:33:47.863892: step: 2572/1541, loss: 0.08911561965942383 2023-01-23 12:33:48.739505: step: 2574/1541, loss: 0.01259087398648262 2023-01-23 12:33:49.700865: step: 2576/1541, loss: 0.10598181188106537 2023-01-23 12:33:50.665178: step: 2578/1541, loss: 0.12945443391799927 2023-01-23 12:33:51.556931: step: 2580/1541, loss: 0.12446435540914536 2023-01-23 12:33:52.513871: step: 2582/1541, loss: 0.0900784358382225 2023-01-23 12:33:53.476690: step: 2584/1541, loss: 0.19816884398460388 2023-01-23 12:33:54.319269: step: 2586/1541, loss: 0.07400494068861008 2023-01-23 12:33:55.270378: step: 2588/1541, loss: 0.19040150940418243 2023-01-23 12:33:56.190297: step: 2590/1541, loss: 0.0035116970539093018 2023-01-23 12:33:57.093618: step: 2592/1541, loss: 0.05511682108044624 2023-01-23 12:33:57.973547: step: 2594/1541, loss: 0.11235391348600388 2023-01-23 12:33:58.880229: step: 2596/1541, loss: 0.006207244470715523 2023-01-23 12:33:59.772309: step: 2598/1541, loss: 0.27171406149864197 2023-01-23 12:34:00.649578: step: 2600/1541, loss: 0.028453268110752106 2023-01-23 12:34:01.591043: step: 2602/1541, loss: 0.005279879551380873 2023-01-23 12:34:02.521675: step: 2604/1541, loss: 0.04045771434903145 2023-01-23 12:34:03.462867: step: 2606/1541, loss: 0.12375756353139877 2023-01-23 12:34:04.319271: step: 2608/1541, loss: 0.17497532069683075 2023-01-23 12:34:05.235022: step: 2610/1541, loss: 0.10664669424295425 2023-01-23 12:34:06.142529: step: 2612/1541, loss: 0.04238265007734299 2023-01-23 12:34:07.050593: step: 2614/1541, loss: 0.2994988262653351 2023-01-23 12:34:07.985084: step: 2616/1541, loss: 0.05994782596826553 2023-01-23 12:34:08.907138: step: 2618/1541, loss: 0.02153419330716133 2023-01-23 12:34:09.875010: step: 2620/1541, loss: 0.031568825244903564 2023-01-23 12:34:10.843676: step: 2622/1541, loss: 2.0754714012145996 2023-01-23 12:34:11.694552: step: 2624/1541, loss: 0.1217939555644989 2023-01-23 12:34:12.592657: step: 2626/1541, loss: 0.012549982406198978 2023-01-23 12:34:13.481663: step: 2628/1541, loss: 0.04965157061815262 2023-01-23 12:34:14.350438: step: 2630/1541, loss: 0.013098344206809998 2023-01-23 12:34:15.221402: step: 2632/1541, loss: 0.016785886138677597 2023-01-23 12:34:16.101082: step: 2634/1541, loss: 0.1939343810081482 2023-01-23 12:34:17.003594: step: 2636/1541, loss: 0.0032682218588888645 2023-01-23 12:34:17.892255: step: 2638/1541, loss: 0.39441293478012085 2023-01-23 12:34:18.826936: step: 2640/1541, loss: 0.11441846936941147 2023-01-23 12:34:19.705033: step: 2642/1541, loss: 0.03902164474129677 2023-01-23 12:34:20.621922: step: 2644/1541, loss: 0.0068587446585297585 2023-01-23 12:34:21.515304: step: 2646/1541, loss: 0.07500458508729935 2023-01-23 12:34:22.444368: step: 2648/1541, loss: 0.22858422994613647 2023-01-23 12:34:23.353814: step: 2650/1541, loss: 0.11516527831554413 2023-01-23 12:34:24.255260: step: 2652/1541, loss: 0.014335096813738346 2023-01-23 12:34:25.176943: step: 2654/1541, loss: 0.02752714790403843 2023-01-23 12:34:26.088019: step: 2656/1541, loss: 0.12882545590400696 2023-01-23 12:34:27.015329: step: 2658/1541, loss: 0.10108859837055206 2023-01-23 12:34:27.924985: step: 2660/1541, loss: 0.0396517813205719 2023-01-23 12:34:28.881459: step: 2662/1541, loss: 0.04466237127780914 2023-01-23 12:34:29.767807: step: 2664/1541, loss: 0.16654753684997559 2023-01-23 12:34:30.705407: step: 2666/1541, loss: 0.16503867506980896 2023-01-23 12:34:31.563737: step: 2668/1541, loss: 0.0071597108617424965 2023-01-23 12:34:32.434928: step: 2670/1541, loss: 0.18455329537391663 2023-01-23 12:34:33.437025: step: 2672/1541, loss: 0.3674471378326416 2023-01-23 12:34:34.324049: step: 2674/1541, loss: 0.06170015409588814 2023-01-23 12:34:35.211388: step: 2676/1541, loss: 0.3207728862762451 2023-01-23 12:34:36.070740: step: 2678/1541, loss: 0.09829927980899811 2023-01-23 12:34:37.002562: step: 2680/1541, loss: 0.020034095272421837 2023-01-23 12:34:37.897067: step: 2682/1541, loss: 0.03846992179751396 2023-01-23 12:34:38.777616: step: 2684/1541, loss: 0.13024574518203735 2023-01-23 12:34:39.715948: step: 2686/1541, loss: 0.2791222333908081 2023-01-23 12:34:40.710894: step: 2688/1541, loss: 0.07167911529541016 2023-01-23 12:34:41.612928: step: 2690/1541, loss: 0.3408375084400177 2023-01-23 12:34:42.520962: step: 2692/1541, loss: 0.05685824155807495 2023-01-23 12:34:43.451925: step: 2694/1541, loss: 0.2065342217683792 2023-01-23 12:34:44.318373: step: 2696/1541, loss: 0.1292852759361267 2023-01-23 12:34:45.211536: step: 2698/1541, loss: 0.1905210018157959 2023-01-23 12:34:46.087374: step: 2700/1541, loss: 0.4294896721839905 2023-01-23 12:34:47.043384: step: 2702/1541, loss: 0.03719039633870125 2023-01-23 12:34:47.957512: step: 2704/1541, loss: 0.3107651174068451 2023-01-23 12:34:48.808317: step: 2706/1541, loss: 0.01846674270927906 2023-01-23 12:34:49.700137: step: 2708/1541, loss: 0.046305734664201736 2023-01-23 12:34:50.566844: step: 2710/1541, loss: 0.031666722148656845 2023-01-23 12:34:51.443054: step: 2712/1541, loss: 0.0259409062564373 2023-01-23 12:34:52.348723: step: 2714/1541, loss: 0.6224770545959473 2023-01-23 12:34:53.215959: step: 2716/1541, loss: 0.041873328387737274 2023-01-23 12:34:54.124153: step: 2718/1541, loss: 0.061044804751873016 2023-01-23 12:34:54.991692: step: 2720/1541, loss: 0.09008996933698654 2023-01-23 12:34:55.886402: step: 2722/1541, loss: 0.24861155450344086 2023-01-23 12:34:56.802941: step: 2724/1541, loss: 0.31161022186279297 2023-01-23 12:34:57.684254: step: 2726/1541, loss: 0.009243832901120186 2023-01-23 12:34:58.560959: step: 2728/1541, loss: 0.025354087352752686 2023-01-23 12:34:59.483300: step: 2730/1541, loss: 0.4919911324977875 2023-01-23 12:35:00.362783: step: 2732/1541, loss: 0.09097196161746979 2023-01-23 12:35:01.206837: step: 2734/1541, loss: 0.003016688860952854 2023-01-23 12:35:02.062175: step: 2736/1541, loss: 0.522607684135437 2023-01-23 12:35:03.004201: step: 2738/1541, loss: 0.20197826623916626 2023-01-23 12:35:03.914825: step: 2740/1541, loss: 0.02682710438966751 2023-01-23 12:35:04.850841: step: 2742/1541, loss: 0.005969717632979155 2023-01-23 12:35:05.750600: step: 2744/1541, loss: 0.3534577190876007 2023-01-23 12:35:06.629215: step: 2746/1541, loss: 0.007516293786466122 2023-01-23 12:35:07.542018: step: 2748/1541, loss: 0.03468315303325653 2023-01-23 12:35:08.424149: step: 2750/1541, loss: 0.22298899292945862 2023-01-23 12:35:09.396260: step: 2752/1541, loss: 0.18615752458572388 2023-01-23 12:35:10.285558: step: 2754/1541, loss: 0.08129167556762695 2023-01-23 12:35:11.301516: step: 2756/1541, loss: 0.08866824954748154 2023-01-23 12:35:12.191276: step: 2758/1541, loss: 0.16268810629844666 2023-01-23 12:35:13.083943: step: 2760/1541, loss: 0.014879006892442703 2023-01-23 12:35:14.015992: step: 2762/1541, loss: 0.004542415961623192 2023-01-23 12:35:14.882811: step: 2764/1541, loss: 0.15227413177490234 2023-01-23 12:35:15.810145: step: 2766/1541, loss: 0.004767466802150011 2023-01-23 12:35:16.680768: step: 2768/1541, loss: 0.07499847561120987 2023-01-23 12:35:17.574068: step: 2770/1541, loss: 0.09721529483795166 2023-01-23 12:35:18.458692: step: 2772/1541, loss: 0.02984756976366043 2023-01-23 12:35:19.340136: step: 2774/1541, loss: 0.06223940849304199 2023-01-23 12:35:20.234848: step: 2776/1541, loss: 0.6198899149894714 2023-01-23 12:35:21.172980: step: 2778/1541, loss: 0.05431987717747688 2023-01-23 12:35:22.076500: step: 2780/1541, loss: 0.07857105880975723 2023-01-23 12:35:22.980960: step: 2782/1541, loss: 0.17872756719589233 2023-01-23 12:35:23.917542: step: 2784/1541, loss: 0.4122846722602844 2023-01-23 12:35:24.781114: step: 2786/1541, loss: 0.2428559809923172 2023-01-23 12:35:25.682791: step: 2788/1541, loss: 0.06253668665885925 2023-01-23 12:35:26.583337: step: 2790/1541, loss: 0.08301383256912231 2023-01-23 12:35:27.462195: step: 2792/1541, loss: 0.26500535011291504 2023-01-23 12:35:28.315963: step: 2794/1541, loss: 0.14481623470783234 2023-01-23 12:35:29.281361: step: 2796/1541, loss: 0.02010619454085827 2023-01-23 12:35:30.208051: step: 2798/1541, loss: 0.022336885333061218 2023-01-23 12:35:31.082113: step: 2800/1541, loss: 0.9269806146621704 2023-01-23 12:35:31.976769: step: 2802/1541, loss: 0.6508173942565918 2023-01-23 12:35:32.951345: step: 2804/1541, loss: 0.14485524594783783 2023-01-23 12:35:33.856044: step: 2806/1541, loss: 0.1424359381198883 2023-01-23 12:35:34.746934: step: 2808/1541, loss: 0.3441653549671173 2023-01-23 12:35:35.632481: step: 2810/1541, loss: 0.4111829102039337 2023-01-23 12:35:36.593697: step: 2812/1541, loss: 0.09936600178480148 2023-01-23 12:35:37.451200: step: 2814/1541, loss: 0.04523858800530434 2023-01-23 12:35:38.397944: step: 2816/1541, loss: 0.1239456981420517 2023-01-23 12:35:39.312125: step: 2818/1541, loss: 0.3422819674015045 2023-01-23 12:35:40.181916: step: 2820/1541, loss: 0.2571322023868561 2023-01-23 12:35:41.066003: step: 2822/1541, loss: 0.9558186531066895 2023-01-23 12:35:41.972692: step: 2824/1541, loss: 0.014806854538619518 2023-01-23 12:35:42.881814: step: 2826/1541, loss: 0.115105040371418 2023-01-23 12:35:43.744256: step: 2828/1541, loss: 0.08528408408164978 2023-01-23 12:35:44.677198: step: 2830/1541, loss: 0.14086975157260895 2023-01-23 12:35:45.546978: step: 2832/1541, loss: 0.05272335931658745 2023-01-23 12:35:46.455187: step: 2834/1541, loss: 0.22865943610668182 2023-01-23 12:35:47.355797: step: 2836/1541, loss: 0.12513363361358643 2023-01-23 12:35:48.265998: step: 2838/1541, loss: 0.04588817432522774 2023-01-23 12:35:49.109233: step: 2840/1541, loss: 0.05620709806680679 2023-01-23 12:35:50.039624: step: 2842/1541, loss: 0.09916722774505615 2023-01-23 12:35:50.942511: step: 2844/1541, loss: 0.05620114132761955 2023-01-23 12:35:51.860410: step: 2846/1541, loss: 0.015938211232423782 2023-01-23 12:35:52.792975: step: 2848/1541, loss: 0.3143911361694336 2023-01-23 12:35:53.737559: step: 2850/1541, loss: 0.015407169237732887 2023-01-23 12:35:54.704000: step: 2852/1541, loss: 0.11920614540576935 2023-01-23 12:35:55.610114: step: 2854/1541, loss: 0.47557446360588074 2023-01-23 12:35:56.512567: step: 2856/1541, loss: 0.022069459781050682 2023-01-23 12:35:57.392586: step: 2858/1541, loss: 0.05086839199066162 2023-01-23 12:35:58.322266: step: 2860/1541, loss: 0.08128531277179718 2023-01-23 12:35:59.236935: step: 2862/1541, loss: 0.3133484423160553 2023-01-23 12:36:00.140916: step: 2864/1541, loss: 0.10363736003637314 2023-01-23 12:36:01.030745: step: 2866/1541, loss: 0.383748322725296 2023-01-23 12:36:01.964735: step: 2868/1541, loss: 0.03602882847189903 2023-01-23 12:36:02.917562: step: 2870/1541, loss: 0.07062389701604843 2023-01-23 12:36:03.790858: step: 2872/1541, loss: 0.08041663467884064 2023-01-23 12:36:04.701996: step: 2874/1541, loss: 0.020152727141976357 2023-01-23 12:36:05.598248: step: 2876/1541, loss: 0.6972501873970032 2023-01-23 12:36:06.481421: step: 2878/1541, loss: 0.03644750267267227 2023-01-23 12:36:07.346172: step: 2880/1541, loss: 0.020847100764513016 2023-01-23 12:36:08.225385: step: 2882/1541, loss: 0.00965674128383398 2023-01-23 12:36:09.099735: step: 2884/1541, loss: 0.06351014226675034 2023-01-23 12:36:10.020116: step: 2886/1541, loss: 0.10321706533432007 2023-01-23 12:36:10.896129: step: 2888/1541, loss: 0.23134447634220123 2023-01-23 12:36:11.752906: step: 2890/1541, loss: 0.018717188388109207 2023-01-23 12:36:12.665109: step: 2892/1541, loss: 0.07413409650325775 2023-01-23 12:36:13.563346: step: 2894/1541, loss: 0.07000089436769485 2023-01-23 12:36:14.525686: step: 2896/1541, loss: 0.01386209949851036 2023-01-23 12:36:15.434174: step: 2898/1541, loss: 0.03209194540977478 2023-01-23 12:36:16.345472: step: 2900/1541, loss: 0.12161367386579514 2023-01-23 12:36:17.236356: step: 2902/1541, loss: 0.08265917748212814 2023-01-23 12:36:18.130393: step: 2904/1541, loss: 0.017919450998306274 2023-01-23 12:36:19.067158: step: 2906/1541, loss: 0.44077205657958984 2023-01-23 12:36:20.021685: step: 2908/1541, loss: 0.0869775116443634 2023-01-23 12:36:20.986714: step: 2910/1541, loss: 0.7532469630241394 2023-01-23 12:36:21.848149: step: 2912/1541, loss: 0.45201605558395386 2023-01-23 12:36:22.729556: step: 2914/1541, loss: 0.04300621524453163 2023-01-23 12:36:23.623926: step: 2916/1541, loss: 1.0531960725784302 2023-01-23 12:36:24.494856: step: 2918/1541, loss: 0.011706982739269733 2023-01-23 12:36:25.415925: step: 2920/1541, loss: 0.01005922257900238 2023-01-23 12:36:26.288447: step: 2922/1541, loss: 0.025357672944664955 2023-01-23 12:36:27.200237: step: 2924/1541, loss: 0.11210297793149948 2023-01-23 12:36:28.127363: step: 2926/1541, loss: 0.037014298141002655 2023-01-23 12:36:29.042568: step: 2928/1541, loss: 0.02613816224038601 2023-01-23 12:36:29.908596: step: 2930/1541, loss: 0.08868645876646042 2023-01-23 12:36:30.777713: step: 2932/1541, loss: 0.23838216066360474 2023-01-23 12:36:31.668793: step: 2934/1541, loss: 1.2258524894714355 2023-01-23 12:36:32.626827: step: 2936/1541, loss: 0.5298982858657837 2023-01-23 12:36:33.552266: step: 2938/1541, loss: 0.14620742201805115 2023-01-23 12:36:34.438909: step: 2940/1541, loss: 0.03950062021613121 2023-01-23 12:36:35.317118: step: 2942/1541, loss: 0.0010826059151440859 2023-01-23 12:36:36.198852: step: 2944/1541, loss: 0.08987817913293839 2023-01-23 12:36:37.109369: step: 2946/1541, loss: 0.11358671635389328 2023-01-23 12:36:38.003856: step: 2948/1541, loss: 0.6141034364700317 2023-01-23 12:36:38.894989: step: 2950/1541, loss: 0.01878192275762558 2023-01-23 12:36:39.824284: step: 2952/1541, loss: 0.08224999159574509 2023-01-23 12:36:40.755756: step: 2954/1541, loss: 0.16879525780677795 2023-01-23 12:36:41.620969: step: 2956/1541, loss: 0.08662989735603333 2023-01-23 12:36:42.562868: step: 2958/1541, loss: 0.10898777842521667 2023-01-23 12:36:43.481247: step: 2960/1541, loss: 0.04708476364612579 2023-01-23 12:36:44.421952: step: 2962/1541, loss: 0.20245391130447388 2023-01-23 12:36:45.316411: step: 2964/1541, loss: 0.03318117558956146 2023-01-23 12:36:46.197066: step: 2966/1541, loss: 0.00968098547309637 2023-01-23 12:36:47.103008: step: 2968/1541, loss: 0.0666779950261116 2023-01-23 12:36:47.997609: step: 2970/1541, loss: 0.21045483648777008 2023-01-23 12:36:48.897898: step: 2972/1541, loss: 0.28764623403549194 2023-01-23 12:36:49.783425: step: 2974/1541, loss: 0.04982233792543411 2023-01-23 12:36:50.653967: step: 2976/1541, loss: 0.020760614424943924 2023-01-23 12:36:51.623959: step: 2978/1541, loss: 0.08731896430253983 2023-01-23 12:36:52.532054: step: 2980/1541, loss: 0.006120753474533558 2023-01-23 12:36:53.418511: step: 2982/1541, loss: 0.04284466430544853 2023-01-23 12:36:54.354194: step: 2984/1541, loss: 2.734532356262207 2023-01-23 12:36:55.291064: step: 2986/1541, loss: 0.0842384472489357 2023-01-23 12:36:56.172669: step: 2988/1541, loss: 0.06284460425376892 2023-01-23 12:36:57.076258: step: 2990/1541, loss: 0.6492648720741272 2023-01-23 12:36:57.947196: step: 2992/1541, loss: 0.017050478607416153 2023-01-23 12:36:58.861926: step: 2994/1541, loss: 0.03254832327365875 2023-01-23 12:36:59.780393: step: 2996/1541, loss: 0.22660957276821136 2023-01-23 12:37:00.644889: step: 2998/1541, loss: 0.19332946836948395 2023-01-23 12:37:01.542826: step: 3000/1541, loss: 0.09878627210855484 2023-01-23 12:37:02.450271: step: 3002/1541, loss: 0.11255527287721634 2023-01-23 12:37:03.377688: step: 3004/1541, loss: 0.7917173504829407 2023-01-23 12:37:04.278280: step: 3006/1541, loss: 0.027012154459953308 2023-01-23 12:37:05.144536: step: 3008/1541, loss: 0.07131410390138626 2023-01-23 12:37:06.100738: step: 3010/1541, loss: 0.007434388622641563 2023-01-23 12:37:06.985768: step: 3012/1541, loss: 0.3121560513973236 2023-01-23 12:37:07.912599: step: 3014/1541, loss: 0.014277113601565361 2023-01-23 12:37:08.810833: step: 3016/1541, loss: 0.4033387005329132 2023-01-23 12:37:09.737132: step: 3018/1541, loss: 0.00772778270766139 2023-01-23 12:37:10.640961: step: 3020/1541, loss: 0.11199319362640381 2023-01-23 12:37:11.534740: step: 3022/1541, loss: 0.06758496910333633 2023-01-23 12:37:12.480805: step: 3024/1541, loss: 0.008210153318941593 2023-01-23 12:37:13.358911: step: 3026/1541, loss: 0.1152881532907486 2023-01-23 12:37:14.290545: step: 3028/1541, loss: 0.058808330446481705 2023-01-23 12:37:15.175246: step: 3030/1541, loss: 0.027851520106196404 2023-01-23 12:37:16.077118: step: 3032/1541, loss: 0.060161832720041275 2023-01-23 12:37:16.961703: step: 3034/1541, loss: 0.019514238461852074 2023-01-23 12:37:17.865919: step: 3036/1541, loss: 0.046913016587495804 2023-01-23 12:37:18.750881: step: 3038/1541, loss: 0.42527642846107483 2023-01-23 12:37:19.667445: step: 3040/1541, loss: 0.05105166509747505 2023-01-23 12:37:20.648452: step: 3042/1541, loss: 0.00516106141731143 2023-01-23 12:37:21.520221: step: 3044/1541, loss: 0.13393937051296234 2023-01-23 12:37:22.429090: step: 3046/1541, loss: 5.875114917755127 2023-01-23 12:37:23.315212: step: 3048/1541, loss: 0.22393739223480225 2023-01-23 12:37:24.208428: step: 3050/1541, loss: 0.29276955127716064 2023-01-23 12:37:25.111939: step: 3052/1541, loss: 0.0386449433863163 2023-01-23 12:37:26.035784: step: 3054/1541, loss: 0.1778755635023117 2023-01-23 12:37:26.931169: step: 3056/1541, loss: 0.006699716672301292 2023-01-23 12:37:27.838219: step: 3058/1541, loss: 0.0013522516237571836 2023-01-23 12:37:28.711036: step: 3060/1541, loss: 0.3704323470592499 2023-01-23 12:37:29.639983: step: 3062/1541, loss: 0.047669120132923126 2023-01-23 12:37:30.528128: step: 3064/1541, loss: 0.03351971507072449 2023-01-23 12:37:31.418567: step: 3066/1541, loss: 0.010381707921624184 2023-01-23 12:37:32.264097: step: 3068/1541, loss: 0.0021802703849971294 2023-01-23 12:37:33.179383: step: 3070/1541, loss: 0.021947002038359642 2023-01-23 12:37:34.076129: step: 3072/1541, loss: 0.3257683515548706 2023-01-23 12:37:34.994646: step: 3074/1541, loss: 0.04045868664979935 2023-01-23 12:37:35.877296: step: 3076/1541, loss: 0.10937626659870148 2023-01-23 12:37:36.762209: step: 3078/1541, loss: 0.01301384437829256 2023-01-23 12:37:37.644340: step: 3080/1541, loss: 0.04368371516466141 2023-01-23 12:37:38.569231: step: 3082/1541, loss: 0.005028023384511471 2023-01-23 12:37:39.208619: step: 3084/1541, loss: 0.0014749257825314999 ================================================== Loss: 0.152 -------------------- Dev Chinese: {'event': {'p': 0.6308243727598566, 'r': 0.7030625832223701, 'f1': 0.6649874055415615}, 'argument': {'p': 0.6718781990356308, 'r': 0.6520557055858218, 'f1': 0.6618185569696915}, 'combined': 0.4401010051385353, 'epoch': 15} Test Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7131147922606513, 'r': 0.6951633436411749, 'f1': 0.7040246535822872}, 'combined': 0.45936080729395945, 'epoch': 15} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6635301955532534, 'r': 0.6606722047612078, 'f1': 0.6620981159945053}, 'combined': 0.43591552978520226, 'epoch': 15} Test Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.7133767687394067, 'r': 0.7146291858193565, 'f1': 0.7140024280699973}, 'combined': 0.4878795460186284, 'epoch': 15} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6707048979170669, 'r': 0.6636342137522785, 'f1': 0.6671508220007766}, 'combined': 0.4626136734375291, 'epoch': 15} Test Russian: {'event': {'p': 0.6541077969649398, 'r': 0.7489514679448772, 'f1': 0.6983240223463687}, 'argument': {'p': 0.709326223967639, 'r': 0.7032135383750991, 'f1': 0.7062566550215115}, 'combined': 0.49319598814351356, 'epoch': 15} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6254208754208753, 'r': 0.5977473853580048, 'f1': 0.611271081859317}, 'combined': 0.4493103678624039, 'epoch': 15} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5306484295845998, 'r': 0.3239022881880025, 'f1': 0.40226574500768053}, 'combined': 0.23439106402022328, 'epoch': 15} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5357032452434751, 'r': 0.5606196752547995, 'f1': 0.5478783189990086}, 'combined': 0.2817659926280616, 'epoch': 15} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.653371185022781, 'r': 0.6528791886485169, 'f1': 0.6531250941809645}, 'combined': 0.4496121337998113, 'epoch': 13} Test for Russian: {'event': {'p': 0.6478283621140764, 'r': 0.7417615338526063, 'f1': 0.6916201117318437}, 'argument': {'p': 0.7081306065307763, 'r': 0.708893952495416, 'f1': 0.7085120739072518}, 'combined': 0.49002119971909375, 'epoch': 13} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6150902184235518, 'r': 0.6436990657920891, 'f1': 0.6290695415695416}, 'combined': 0.3235214785214786, 'epoch': 13} ****************************** Epoch: 16 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 12:43:01.494366: step: 2/1541, loss: 0.10078703612089157 2023-01-23 12:43:02.381149: step: 4/1541, loss: 0.023505568504333496 2023-01-23 12:43:03.271879: step: 6/1541, loss: 0.01596911996603012 2023-01-23 12:43:04.133327: step: 8/1541, loss: 0.04408178851008415 2023-01-23 12:43:05.029978: step: 10/1541, loss: 0.02434989996254444 2023-01-23 12:43:05.927819: step: 12/1541, loss: 0.05180354043841362 2023-01-23 12:43:06.867021: step: 14/1541, loss: 0.43353891372680664 2023-01-23 12:43:07.796924: step: 16/1541, loss: 0.10116872936487198 2023-01-23 12:43:08.695077: step: 18/1541, loss: 0.02437145821750164 2023-01-23 12:43:09.570542: step: 20/1541, loss: 0.0010851046536117792 2023-01-23 12:43:10.470418: step: 22/1541, loss: 0.03708914294838905 2023-01-23 12:43:11.387984: step: 24/1541, loss: 0.0049264095723629 2023-01-23 12:43:12.272685: step: 26/1541, loss: 0.0027221939526498318 2023-01-23 12:43:13.148341: step: 28/1541, loss: 0.03965354710817337 2023-01-23 12:43:14.025723: step: 30/1541, loss: 0.029657691717147827 2023-01-23 12:43:14.928263: step: 32/1541, loss: 0.05946590006351471 2023-01-23 12:43:15.816935: step: 34/1541, loss: 0.17278437316417694 2023-01-23 12:43:16.746999: step: 36/1541, loss: 0.15533052384853363 2023-01-23 12:43:17.599961: step: 38/1541, loss: 0.0024545297492295504 2023-01-23 12:43:18.487576: step: 40/1541, loss: 0.010637001134455204 2023-01-23 12:43:19.369960: step: 42/1541, loss: 0.09563866257667542 2023-01-23 12:43:20.264830: step: 44/1541, loss: 0.018438640981912613 2023-01-23 12:43:21.178295: step: 46/1541, loss: 0.012865358963608742 2023-01-23 12:43:22.070449: step: 48/1541, loss: 0.016380557790398598 2023-01-23 12:43:22.953020: step: 50/1541, loss: 0.02856365591287613 2023-01-23 12:43:23.830444: step: 52/1541, loss: 0.019520025700330734 2023-01-23 12:43:24.714936: step: 54/1541, loss: 0.22766375541687012 2023-01-23 12:43:25.586545: step: 56/1541, loss: 0.1773194819688797 2023-01-23 12:43:26.514494: step: 58/1541, loss: 0.21502013504505157 2023-01-23 12:43:27.415869: step: 60/1541, loss: 0.043334536254405975 2023-01-23 12:43:28.306563: step: 62/1541, loss: 0.03681979328393936 2023-01-23 12:43:29.220016: step: 64/1541, loss: 0.24695441126823425 2023-01-23 12:43:30.113840: step: 66/1541, loss: 0.04893432557582855 2023-01-23 12:43:31.065180: step: 68/1541, loss: 0.018907414749264717 2023-01-23 12:43:31.966210: step: 70/1541, loss: 0.0024534561671316624 2023-01-23 12:43:32.901655: step: 72/1541, loss: 0.2102050483226776 2023-01-23 12:43:33.762415: step: 74/1541, loss: 0.33814650774002075 2023-01-23 12:43:34.637340: step: 76/1541, loss: 0.032180678099393845 2023-01-23 12:43:35.512310: step: 78/1541, loss: 0.10840396583080292 2023-01-23 12:43:36.398940: step: 80/1541, loss: 0.46265730261802673 2023-01-23 12:43:37.336684: step: 82/1541, loss: 0.023489946499466896 2023-01-23 12:43:38.238951: step: 84/1541, loss: 0.08192133903503418 2023-01-23 12:43:39.111916: step: 86/1541, loss: 0.03446851298213005 2023-01-23 12:43:39.984658: step: 88/1541, loss: 1.5857172012329102 2023-01-23 12:43:40.853210: step: 90/1541, loss: 0.08369944989681244 2023-01-23 12:43:41.810392: step: 92/1541, loss: 0.051798660308122635 2023-01-23 12:43:42.761290: step: 94/1541, loss: 0.03200814872980118 2023-01-23 12:43:43.698174: step: 96/1541, loss: 0.0070434329099953175 2023-01-23 12:43:44.587570: step: 98/1541, loss: 0.022567735984921455 2023-01-23 12:43:45.539759: step: 100/1541, loss: 0.004104143008589745 2023-01-23 12:43:46.383474: step: 102/1541, loss: 0.03004922717809677 2023-01-23 12:43:47.289372: step: 104/1541, loss: 0.03441547602415085 2023-01-23 12:43:48.157967: step: 106/1541, loss: 0.004098449368029833 2023-01-23 12:43:49.034469: step: 108/1541, loss: 0.5151780843734741 2023-01-23 12:43:49.882986: step: 110/1541, loss: 0.033794861286878586 2023-01-23 12:43:50.797530: step: 112/1541, loss: 0.0008440729579888284 2023-01-23 12:43:51.708202: step: 114/1541, loss: 0.1895473152399063 2023-01-23 12:43:52.600933: step: 116/1541, loss: 0.009262080304324627 2023-01-23 12:43:53.514048: step: 118/1541, loss: 0.054881684482097626 2023-01-23 12:43:54.424014: step: 120/1541, loss: 0.24250996112823486 2023-01-23 12:43:55.333069: step: 122/1541, loss: 0.0045790583826601505 2023-01-23 12:43:56.207222: step: 124/1541, loss: 0.26888325810432434 2023-01-23 12:43:57.084770: step: 126/1541, loss: 0.005255766212940216 2023-01-23 12:43:57.975665: step: 128/1541, loss: 0.004287770949304104 2023-01-23 12:43:58.901095: step: 130/1541, loss: 0.16782309114933014 2023-01-23 12:43:59.841205: step: 132/1541, loss: 0.14675430953502655 2023-01-23 12:44:00.740027: step: 134/1541, loss: 0.29728221893310547 2023-01-23 12:44:01.620456: step: 136/1541, loss: 0.08232780545949936 2023-01-23 12:44:02.592723: step: 138/1541, loss: 0.15787096321582794 2023-01-23 12:44:03.505998: step: 140/1541, loss: 0.0723835825920105 2023-01-23 12:44:04.403227: step: 142/1541, loss: 0.11736726760864258 2023-01-23 12:44:05.285001: step: 144/1541, loss: 0.0030356505885720253 2023-01-23 12:44:06.151216: step: 146/1541, loss: 0.019244441762566566 2023-01-23 12:44:07.061637: step: 148/1541, loss: 0.026312019675970078 2023-01-23 12:44:07.935747: step: 150/1541, loss: 0.07547467201948166 2023-01-23 12:44:08.838756: step: 152/1541, loss: 0.05894985795021057 2023-01-23 12:44:09.744811: step: 154/1541, loss: 0.02198703959584236 2023-01-23 12:44:10.648639: step: 156/1541, loss: 0.10065297037363052 2023-01-23 12:44:11.528068: step: 158/1541, loss: 0.005796192213892937 2023-01-23 12:44:12.405475: step: 160/1541, loss: 0.002139331540092826 2023-01-23 12:44:13.319007: step: 162/1541, loss: 0.009594537317752838 2023-01-23 12:44:14.267384: step: 164/1541, loss: 0.02553756907582283 2023-01-23 12:44:15.176984: step: 166/1541, loss: 0.0019628098234534264 2023-01-23 12:44:16.056243: step: 168/1541, loss: 0.0029928041622042656 2023-01-23 12:44:16.958091: step: 170/1541, loss: 0.012586849741637707 2023-01-23 12:44:17.784182: step: 172/1541, loss: 0.12034914642572403 2023-01-23 12:44:18.683066: step: 174/1541, loss: 0.01279736589640379 2023-01-23 12:44:19.564105: step: 176/1541, loss: 0.05872819945216179 2023-01-23 12:44:20.427335: step: 178/1541, loss: 0.0864357054233551 2023-01-23 12:44:21.315498: step: 180/1541, loss: 0.039667531847953796 2023-01-23 12:44:22.217987: step: 182/1541, loss: 0.08867406100034714 2023-01-23 12:44:23.090522: step: 184/1541, loss: 0.09174931049346924 2023-01-23 12:44:23.983071: step: 186/1541, loss: 0.046484146267175674 2023-01-23 12:44:24.895302: step: 188/1541, loss: 0.12998762726783752 2023-01-23 12:44:25.791119: step: 190/1541, loss: 0.11635337769985199 2023-01-23 12:44:26.628125: step: 192/1541, loss: 0.0007435822626575828 2023-01-23 12:44:27.546153: step: 194/1541, loss: 0.14317384362220764 2023-01-23 12:44:28.417039: step: 196/1541, loss: 0.09167701005935669 2023-01-23 12:44:29.308146: step: 198/1541, loss: 0.09385265409946442 2023-01-23 12:44:30.208613: step: 200/1541, loss: 0.16858473420143127 2023-01-23 12:44:31.099334: step: 202/1541, loss: 0.24745845794677734 2023-01-23 12:44:32.166225: step: 204/1541, loss: 0.004902943968772888 2023-01-23 12:44:33.039853: step: 206/1541, loss: 0.046193577349185944 2023-01-23 12:44:33.904400: step: 208/1541, loss: 0.0640600323677063 2023-01-23 12:44:34.818682: step: 210/1541, loss: 0.0029981592670083046 2023-01-23 12:44:35.719160: step: 212/1541, loss: 0.02455374039709568 2023-01-23 12:44:36.624493: step: 214/1541, loss: 0.0010287101613357663 2023-01-23 12:44:37.529127: step: 216/1541, loss: 0.005839936900883913 2023-01-23 12:44:38.422981: step: 218/1541, loss: 0.0034638349898159504 2023-01-23 12:44:39.348858: step: 220/1541, loss: 0.43341001868247986 2023-01-23 12:44:40.274268: step: 222/1541, loss: 0.30292949080467224 2023-01-23 12:44:41.155020: step: 224/1541, loss: 0.01396216731518507 2023-01-23 12:44:41.999333: step: 226/1541, loss: 0.0020441317465156317 2023-01-23 12:44:42.933707: step: 228/1541, loss: 0.0052698589861392975 2023-01-23 12:44:43.809390: step: 230/1541, loss: 0.007369804196059704 2023-01-23 12:44:44.737530: step: 232/1541, loss: 0.17316675186157227 2023-01-23 12:44:45.655766: step: 234/1541, loss: 0.04053246229887009 2023-01-23 12:44:46.524964: step: 236/1541, loss: 0.007749876007437706 2023-01-23 12:44:47.467179: step: 238/1541, loss: 0.20420056581497192 2023-01-23 12:44:48.361891: step: 240/1541, loss: 0.06261292845010757 2023-01-23 12:44:49.279453: step: 242/1541, loss: 0.00910213403403759 2023-01-23 12:44:50.179547: step: 244/1541, loss: 0.014807121828198433 2023-01-23 12:44:51.046331: step: 246/1541, loss: 0.008705359883606434 2023-01-23 12:44:51.995130: step: 248/1541, loss: 0.0043565137311816216 2023-01-23 12:44:52.975628: step: 250/1541, loss: 0.06183421611785889 2023-01-23 12:44:53.882648: step: 252/1541, loss: 0.13326488435268402 2023-01-23 12:44:54.873554: step: 254/1541, loss: 0.00754773523658514 2023-01-23 12:44:55.757978: step: 256/1541, loss: 0.3812662661075592 2023-01-23 12:44:56.681883: step: 258/1541, loss: 0.09807596355676651 2023-01-23 12:44:57.568294: step: 260/1541, loss: 0.14732159674167633 2023-01-23 12:44:58.473806: step: 262/1541, loss: 0.3854771852493286 2023-01-23 12:44:59.360607: step: 264/1541, loss: 0.01287788338959217 2023-01-23 12:45:00.289654: step: 266/1541, loss: 0.020332584157586098 2023-01-23 12:45:01.240815: step: 268/1541, loss: 0.006360411643981934 2023-01-23 12:45:02.182505: step: 270/1541, loss: 0.020138457417488098 2023-01-23 12:45:03.069339: step: 272/1541, loss: 0.14750871062278748 2023-01-23 12:45:03.920846: step: 274/1541, loss: 0.09003916382789612 2023-01-23 12:45:04.826300: step: 276/1541, loss: 0.13875964283943176 2023-01-23 12:45:05.743096: step: 278/1541, loss: 0.19630172848701477 2023-01-23 12:45:06.632555: step: 280/1541, loss: 0.0035009924322366714 2023-01-23 12:45:07.534482: step: 282/1541, loss: 0.032578181475400925 2023-01-23 12:45:08.431691: step: 284/1541, loss: 0.0026607043109834194 2023-01-23 12:45:09.329756: step: 286/1541, loss: 0.16339561343193054 2023-01-23 12:45:10.262378: step: 288/1541, loss: 0.024770423769950867 2023-01-23 12:45:11.159047: step: 290/1541, loss: 0.08672768622636795 2023-01-23 12:45:12.045896: step: 292/1541, loss: 0.01941266469657421 2023-01-23 12:45:12.994440: step: 294/1541, loss: 0.002868655836209655 2023-01-23 12:45:13.986594: step: 296/1541, loss: 0.06189948692917824 2023-01-23 12:45:14.996822: step: 298/1541, loss: 0.00963553600013256 2023-01-23 12:45:15.905195: step: 300/1541, loss: 0.02028900571167469 2023-01-23 12:45:16.777403: step: 302/1541, loss: 0.08548074215650558 2023-01-23 12:45:17.647616: step: 304/1541, loss: 0.007669057697057724 2023-01-23 12:45:18.555970: step: 306/1541, loss: 0.013232944533228874 2023-01-23 12:45:19.434948: step: 308/1541, loss: 0.2768191695213318 2023-01-23 12:45:20.334952: step: 310/1541, loss: 0.11530668288469315 2023-01-23 12:45:21.266415: step: 312/1541, loss: 0.03617751598358154 2023-01-23 12:45:22.168430: step: 314/1541, loss: 0.004546839278191328 2023-01-23 12:45:23.053186: step: 316/1541, loss: 0.18474999070167542 2023-01-23 12:45:23.929588: step: 318/1541, loss: 0.18176552653312683 2023-01-23 12:45:24.804316: step: 320/1541, loss: 0.11128833144903183 2023-01-23 12:45:25.687614: step: 322/1541, loss: 0.011984067969024181 2023-01-23 12:45:26.582034: step: 324/1541, loss: 0.06029896438121796 2023-01-23 12:45:27.479353: step: 326/1541, loss: 0.2647508978843689 2023-01-23 12:45:28.343440: step: 328/1541, loss: 0.003577041206881404 2023-01-23 12:45:29.294310: step: 330/1541, loss: 0.00114407017827034 2023-01-23 12:45:30.223376: step: 332/1541, loss: 0.2678869366645813 2023-01-23 12:45:31.134003: step: 334/1541, loss: 0.02103549800813198 2023-01-23 12:45:32.036244: step: 336/1541, loss: 0.16344721615314484 2023-01-23 12:45:32.940165: step: 338/1541, loss: 0.2543943226337433 2023-01-23 12:45:33.815110: step: 340/1541, loss: 0.08315401524305344 2023-01-23 12:45:34.708026: step: 342/1541, loss: 0.011622907593846321 2023-01-23 12:45:35.597914: step: 344/1541, loss: 0.018241818994283676 2023-01-23 12:45:36.578414: step: 346/1541, loss: 0.02402408979833126 2023-01-23 12:45:37.549462: step: 348/1541, loss: 0.003675138344988227 2023-01-23 12:45:38.487129: step: 350/1541, loss: 0.01484118215739727 2023-01-23 12:45:39.433373: step: 352/1541, loss: 0.26445648074150085 2023-01-23 12:45:40.320712: step: 354/1541, loss: 0.003459898754954338 2023-01-23 12:45:41.223847: step: 356/1541, loss: 0.10321309417486191 2023-01-23 12:45:42.156565: step: 358/1541, loss: 0.12159716337919235 2023-01-23 12:45:43.063513: step: 360/1541, loss: 0.17405842244625092 2023-01-23 12:45:43.969248: step: 362/1541, loss: 0.15804614126682281 2023-01-23 12:45:44.880945: step: 364/1541, loss: 0.03918787091970444 2023-01-23 12:45:45.802782: step: 366/1541, loss: 0.1865253448486328 2023-01-23 12:45:46.773367: step: 368/1541, loss: 0.4573092758655548 2023-01-23 12:45:47.673342: step: 370/1541, loss: 0.048628661781549454 2023-01-23 12:45:48.551955: step: 372/1541, loss: 0.03760534152388573 2023-01-23 12:45:49.487985: step: 374/1541, loss: 0.19499361515045166 2023-01-23 12:45:50.374255: step: 376/1541, loss: 0.011234564706683159 2023-01-23 12:45:51.290477: step: 378/1541, loss: 0.1523740589618683 2023-01-23 12:45:52.163780: step: 380/1541, loss: 0.07840796560049057 2023-01-23 12:45:53.068936: step: 382/1541, loss: 0.19388367235660553 2023-01-23 12:45:53.968611: step: 384/1541, loss: 0.06624379754066467 2023-01-23 12:45:54.850538: step: 386/1541, loss: 0.058625396341085434 2023-01-23 12:45:55.741383: step: 388/1541, loss: 0.004571536090224981 2023-01-23 12:45:56.607135: step: 390/1541, loss: 0.447056382894516 2023-01-23 12:45:57.505661: step: 392/1541, loss: 0.0013334680115804076 2023-01-23 12:45:58.403744: step: 394/1541, loss: 0.5699134469032288 2023-01-23 12:45:59.345818: step: 396/1541, loss: 0.7624063491821289 2023-01-23 12:46:00.201514: step: 398/1541, loss: 0.18639221787452698 2023-01-23 12:46:01.089395: step: 400/1541, loss: 0.007220536004751921 2023-01-23 12:46:01.999808: step: 402/1541, loss: 0.2654464840888977 2023-01-23 12:46:02.920354: step: 404/1541, loss: 0.037189364433288574 2023-01-23 12:46:03.784303: step: 406/1541, loss: 0.010235311463475227 2023-01-23 12:46:04.687798: step: 408/1541, loss: 0.10351701825857162 2023-01-23 12:46:05.551549: step: 410/1541, loss: 0.11415107548236847 2023-01-23 12:46:06.516050: step: 412/1541, loss: 0.013862241059541702 2023-01-23 12:46:07.416354: step: 414/1541, loss: 0.19575557112693787 2023-01-23 12:46:08.329041: step: 416/1541, loss: 0.030312923714518547 2023-01-23 12:46:09.230264: step: 418/1541, loss: 0.0547315776348114 2023-01-23 12:46:10.122434: step: 420/1541, loss: 0.13599485158920288 2023-01-23 12:46:11.010580: step: 422/1541, loss: 0.04544278606772423 2023-01-23 12:46:11.878538: step: 424/1541, loss: 0.028960175812244415 2023-01-23 12:46:12.846143: step: 426/1541, loss: 0.01864204742014408 2023-01-23 12:46:13.738686: step: 428/1541, loss: 0.33549702167510986 2023-01-23 12:46:14.679269: step: 430/1541, loss: 0.03323754295706749 2023-01-23 12:46:15.671834: step: 432/1541, loss: 0.03430287167429924 2023-01-23 12:46:16.549446: step: 434/1541, loss: 0.034902963787317276 2023-01-23 12:46:17.415397: step: 436/1541, loss: 0.08591540157794952 2023-01-23 12:46:18.317013: step: 438/1541, loss: 0.147801011800766 2023-01-23 12:46:19.163799: step: 440/1541, loss: 0.011224287562072277 2023-01-23 12:46:20.061093: step: 442/1541, loss: 0.01041927095502615 2023-01-23 12:46:20.976353: step: 444/1541, loss: 0.03865811973810196 2023-01-23 12:46:21.879995: step: 446/1541, loss: 0.049321066588163376 2023-01-23 12:46:22.780668: step: 448/1541, loss: 0.002989076543599367 2023-01-23 12:46:23.737615: step: 450/1541, loss: 0.05113562196493149 2023-01-23 12:46:24.641972: step: 452/1541, loss: 0.0034721544943749905 2023-01-23 12:46:25.598616: step: 454/1541, loss: 0.42883408069610596 2023-01-23 12:46:26.491854: step: 456/1541, loss: 0.015612994320690632 2023-01-23 12:46:27.358067: step: 458/1541, loss: 0.028616633266210556 2023-01-23 12:46:28.239300: step: 460/1541, loss: 0.04376991465687752 2023-01-23 12:46:29.198521: step: 462/1541, loss: 0.0016493467846885324 2023-01-23 12:46:30.119318: step: 464/1541, loss: 0.045067910104990005 2023-01-23 12:46:30.995606: step: 466/1541, loss: 0.017294935882091522 2023-01-23 12:46:31.969903: step: 468/1541, loss: 0.0017003959510475397 2023-01-23 12:46:32.897906: step: 470/1541, loss: 0.004294888582080603 2023-01-23 12:46:33.853228: step: 472/1541, loss: 0.016476698219776154 2023-01-23 12:46:34.730169: step: 474/1541, loss: 0.005175080616027117 2023-01-23 12:46:35.646582: step: 476/1541, loss: 0.0021195788867771626 2023-01-23 12:46:36.536291: step: 478/1541, loss: 0.028677351772785187 2023-01-23 12:46:37.460974: step: 480/1541, loss: 0.2349473088979721 2023-01-23 12:46:38.352461: step: 482/1541, loss: 0.06761004775762558 2023-01-23 12:46:39.220987: step: 484/1541, loss: 0.3530767858028412 2023-01-23 12:46:40.132865: step: 486/1541, loss: 0.4300079643726349 2023-01-23 12:46:41.007979: step: 488/1541, loss: 0.21546083688735962 2023-01-23 12:46:41.906934: step: 490/1541, loss: 0.19701124727725983 2023-01-23 12:46:42.766115: step: 492/1541, loss: 0.06698118895292282 2023-01-23 12:46:43.649805: step: 494/1541, loss: 0.3154909014701843 2023-01-23 12:46:44.652527: step: 496/1541, loss: 0.007682012394070625 2023-01-23 12:46:45.506702: step: 498/1541, loss: 0.03864194080233574 2023-01-23 12:46:46.400746: step: 500/1541, loss: 0.12753115594387054 2023-01-23 12:46:47.328336: step: 502/1541, loss: 0.004623153246939182 2023-01-23 12:46:48.202320: step: 504/1541, loss: 0.047490984201431274 2023-01-23 12:46:49.085144: step: 506/1541, loss: 0.11895725876092911 2023-01-23 12:46:50.099852: step: 508/1541, loss: 0.23286139965057373 2023-01-23 12:46:50.985872: step: 510/1541, loss: 0.05626167356967926 2023-01-23 12:46:51.919587: step: 512/1541, loss: 0.06384002417325974 2023-01-23 12:46:52.786512: step: 514/1541, loss: 0.044367607682943344 2023-01-23 12:46:53.675918: step: 516/1541, loss: 0.275208979845047 2023-01-23 12:46:54.531609: step: 518/1541, loss: 0.2540561258792877 2023-01-23 12:46:55.414301: step: 520/1541, loss: 0.18131257593631744 2023-01-23 12:46:56.269655: step: 522/1541, loss: 0.12267902493476868 2023-01-23 12:46:57.168527: step: 524/1541, loss: 0.2667689025402069 2023-01-23 12:46:58.055454: step: 526/1541, loss: 0.16143998503684998 2023-01-23 12:46:58.958000: step: 528/1541, loss: 0.2137916088104248 2023-01-23 12:46:59.809912: step: 530/1541, loss: 0.01848616451025009 2023-01-23 12:47:00.696086: step: 532/1541, loss: 0.010398047044873238 2023-01-23 12:47:01.549294: step: 534/1541, loss: 0.04270049184560776 2023-01-23 12:47:02.444331: step: 536/1541, loss: 0.04224410653114319 2023-01-23 12:47:03.344420: step: 538/1541, loss: 0.02688087522983551 2023-01-23 12:47:04.265719: step: 540/1541, loss: 0.00942639634013176 2023-01-23 12:47:05.155395: step: 542/1541, loss: 0.022531423717737198 2023-01-23 12:47:06.048533: step: 544/1541, loss: 0.0926789715886116 2023-01-23 12:47:06.963450: step: 546/1541, loss: 0.16588161885738373 2023-01-23 12:47:07.860555: step: 548/1541, loss: 0.009777948260307312 2023-01-23 12:47:08.765900: step: 550/1541, loss: 0.26101022958755493 2023-01-23 12:47:09.690620: step: 552/1541, loss: 0.06960573047399521 2023-01-23 12:47:10.581194: step: 554/1541, loss: 0.07700692117214203 2023-01-23 12:47:11.468054: step: 556/1541, loss: 0.1451648473739624 2023-01-23 12:47:12.361650: step: 558/1541, loss: 0.040642254054546356 2023-01-23 12:47:13.260599: step: 560/1541, loss: 0.07802489399909973 2023-01-23 12:47:14.112011: step: 562/1541, loss: 0.14143432676792145 2023-01-23 12:47:15.001027: step: 564/1541, loss: 0.024002540856599808 2023-01-23 12:47:15.911161: step: 566/1541, loss: 0.03061336651444435 2023-01-23 12:47:16.760977: step: 568/1541, loss: 0.020550468936562538 2023-01-23 12:47:17.661678: step: 570/1541, loss: 0.36147651076316833 2023-01-23 12:47:18.540975: step: 572/1541, loss: 0.6881090402603149 2023-01-23 12:47:19.454623: step: 574/1541, loss: 0.07490189373493195 2023-01-23 12:47:20.324382: step: 576/1541, loss: 0.013391071930527687 2023-01-23 12:47:21.228075: step: 578/1541, loss: 0.01268272940069437 2023-01-23 12:47:22.105010: step: 580/1541, loss: 0.02999984659254551 2023-01-23 12:47:23.053928: step: 582/1541, loss: 0.05546999350190163 2023-01-23 12:47:23.973133: step: 584/1541, loss: 0.0998665988445282 2023-01-23 12:47:24.855507: step: 586/1541, loss: 0.0908573567867279 2023-01-23 12:47:25.756218: step: 588/1541, loss: 0.01536950096487999 2023-01-23 12:47:26.630280: step: 590/1541, loss: 0.15700796246528625 2023-01-23 12:47:27.518070: step: 592/1541, loss: 0.03812483325600624 2023-01-23 12:47:28.401383: step: 594/1541, loss: 2.002042293548584 2023-01-23 12:47:29.329401: step: 596/1541, loss: 0.015303514897823334 2023-01-23 12:47:30.208570: step: 598/1541, loss: 0.17047865688800812 2023-01-23 12:47:31.071457: step: 600/1541, loss: 0.125444695353508 2023-01-23 12:47:31.959415: step: 602/1541, loss: 0.31482088565826416 2023-01-23 12:47:32.879042: step: 604/1541, loss: 0.19262489676475525 2023-01-23 12:47:33.770010: step: 606/1541, loss: 0.024957086890935898 2023-01-23 12:47:34.648426: step: 608/1541, loss: 0.011723082512617111 2023-01-23 12:47:35.535860: step: 610/1541, loss: 0.03552941232919693 2023-01-23 12:47:36.430252: step: 612/1541, loss: 0.024216998368501663 2023-01-23 12:47:37.325875: step: 614/1541, loss: 0.01787400245666504 2023-01-23 12:47:38.265803: step: 616/1541, loss: 0.0025932651478797197 2023-01-23 12:47:39.124381: step: 618/1541, loss: 0.0645614042878151 2023-01-23 12:47:40.052135: step: 620/1541, loss: 0.021146226674318314 2023-01-23 12:47:41.056694: step: 622/1541, loss: 0.0419485941529274 2023-01-23 12:47:41.960480: step: 624/1541, loss: 0.14951133728027344 2023-01-23 12:47:42.857151: step: 626/1541, loss: 0.0076879761181771755 2023-01-23 12:47:43.748037: step: 628/1541, loss: 0.04537595808506012 2023-01-23 12:47:44.622773: step: 630/1541, loss: 0.009298632852733135 2023-01-23 12:47:45.505213: step: 632/1541, loss: 0.2143295705318451 2023-01-23 12:47:46.440900: step: 634/1541, loss: 0.7394535541534424 2023-01-23 12:47:47.355548: step: 636/1541, loss: 0.10615573078393936 2023-01-23 12:47:48.236638: step: 638/1541, loss: 0.15574590861797333 2023-01-23 12:47:49.139449: step: 640/1541, loss: 0.02871125377714634 2023-01-23 12:47:50.080264: step: 642/1541, loss: 0.10513844341039658 2023-01-23 12:47:50.948063: step: 644/1541, loss: 0.17481055855751038 2023-01-23 12:47:51.816426: step: 646/1541, loss: 0.02976984716951847 2023-01-23 12:47:52.734422: step: 648/1541, loss: 0.07042120397090912 2023-01-23 12:47:53.611272: step: 650/1541, loss: 0.17723535001277924 2023-01-23 12:47:54.445224: step: 652/1541, loss: 0.03637285903096199 2023-01-23 12:47:55.321485: step: 654/1541, loss: 0.006605230271816254 2023-01-23 12:47:56.255005: step: 656/1541, loss: 0.10969411581754684 2023-01-23 12:47:57.149828: step: 658/1541, loss: 0.0044292304664850235 2023-01-23 12:47:58.033762: step: 660/1541, loss: 0.01752563752233982 2023-01-23 12:47:58.972508: step: 662/1541, loss: 0.01961301825940609 2023-01-23 12:47:59.879323: step: 664/1541, loss: 0.11634010076522827 2023-01-23 12:48:00.771483: step: 666/1541, loss: 0.10863765329122543 2023-01-23 12:48:01.626839: step: 668/1541, loss: 0.009627507999539375 2023-01-23 12:48:02.564577: step: 670/1541, loss: 0.05181455612182617 2023-01-23 12:48:03.510282: step: 672/1541, loss: 0.013524804264307022 2023-01-23 12:48:04.354667: step: 674/1541, loss: 0.06880662590265274 2023-01-23 12:48:05.261857: step: 676/1541, loss: 0.0020591062493622303 2023-01-23 12:48:06.154959: step: 678/1541, loss: 0.1053084135055542 2023-01-23 12:48:07.083585: step: 680/1541, loss: 0.02070516347885132 2023-01-23 12:48:07.955866: step: 682/1541, loss: 0.010158749297261238 2023-01-23 12:48:08.808623: step: 684/1541, loss: 0.011836132034659386 2023-01-23 12:48:09.694024: step: 686/1541, loss: 0.044979918748140335 2023-01-23 12:48:10.584345: step: 688/1541, loss: 0.12759225070476532 2023-01-23 12:48:11.468592: step: 690/1541, loss: 0.18873095512390137 2023-01-23 12:48:12.390800: step: 692/1541, loss: 0.001765867113135755 2023-01-23 12:48:13.308786: step: 694/1541, loss: 0.40814557671546936 2023-01-23 12:48:14.197456: step: 696/1541, loss: 0.05199079215526581 2023-01-23 12:48:15.103935: step: 698/1541, loss: 0.25732165575027466 2023-01-23 12:48:16.036638: step: 700/1541, loss: 0.09946392476558685 2023-01-23 12:48:16.897239: step: 702/1541, loss: 0.07200857251882553 2023-01-23 12:48:17.768002: step: 704/1541, loss: 0.2614858150482178 2023-01-23 12:48:18.687101: step: 706/1541, loss: 0.7656309604644775 2023-01-23 12:48:19.546118: step: 708/1541, loss: 0.4504137933254242 2023-01-23 12:48:20.451200: step: 710/1541, loss: 0.0028957058675587177 2023-01-23 12:48:21.355720: step: 712/1541, loss: 0.3285368084907532 2023-01-23 12:48:22.226052: step: 714/1541, loss: 0.011332651600241661 2023-01-23 12:48:23.139504: step: 716/1541, loss: 0.13374686241149902 2023-01-23 12:48:24.069107: step: 718/1541, loss: 0.02711470238864422 2023-01-23 12:48:24.964316: step: 720/1541, loss: 0.07446981221437454 2023-01-23 12:48:25.847705: step: 722/1541, loss: 0.07481230050325394 2023-01-23 12:48:26.791314: step: 724/1541, loss: 1.5464324951171875 2023-01-23 12:48:27.678500: step: 726/1541, loss: 0.35456591844558716 2023-01-23 12:48:28.600523: step: 728/1541, loss: 0.02980067767202854 2023-01-23 12:48:29.503765: step: 730/1541, loss: 0.06717536598443985 2023-01-23 12:48:30.428076: step: 732/1541, loss: 0.015119175426661968 2023-01-23 12:48:31.330408: step: 734/1541, loss: 0.010442757979035378 2023-01-23 12:48:32.225333: step: 736/1541, loss: 0.2712106704711914 2023-01-23 12:48:33.087311: step: 738/1541, loss: 0.02666393853724003 2023-01-23 12:48:33.969313: step: 740/1541, loss: 0.03347392752766609 2023-01-23 12:48:34.939458: step: 742/1541, loss: 0.0032456344924867153 2023-01-23 12:48:35.836020: step: 744/1541, loss: 0.02647988311946392 2023-01-23 12:48:36.706263: step: 746/1541, loss: 0.13293269276618958 2023-01-23 12:48:37.650799: step: 748/1541, loss: 0.0629475936293602 2023-01-23 12:48:38.537661: step: 750/1541, loss: 0.06455090641975403 2023-01-23 12:48:39.420665: step: 752/1541, loss: 0.06175047159194946 2023-01-23 12:48:40.297304: step: 754/1541, loss: 0.0646761804819107 2023-01-23 12:48:41.178785: step: 756/1541, loss: 0.0328083299100399 2023-01-23 12:48:42.183108: step: 758/1541, loss: 0.06436438858509064 2023-01-23 12:48:43.109896: step: 760/1541, loss: 0.2197449952363968 2023-01-23 12:48:43.990319: step: 762/1541, loss: 0.01094872783869505 2023-01-23 12:48:44.868037: step: 764/1541, loss: 0.20073789358139038 2023-01-23 12:48:45.775770: step: 766/1541, loss: 0.03195254132151604 2023-01-23 12:48:46.671504: step: 768/1541, loss: 0.05981036648154259 2023-01-23 12:48:47.569394: step: 770/1541, loss: 0.49423423409461975 2023-01-23 12:48:48.446046: step: 772/1541, loss: 0.20068319141864777 2023-01-23 12:48:49.355969: step: 774/1541, loss: 1.0012446641921997 2023-01-23 12:48:50.293933: step: 776/1541, loss: 0.027816878631711006 2023-01-23 12:48:51.172552: step: 778/1541, loss: 0.2662307918071747 2023-01-23 12:48:52.084859: step: 780/1541, loss: 0.21963171660900116 2023-01-23 12:48:52.998731: step: 782/1541, loss: 0.0006049031508155167 2023-01-23 12:48:53.863803: step: 784/1541, loss: 0.029759448021650314 2023-01-23 12:48:54.800173: step: 786/1541, loss: 0.02815239690244198 2023-01-23 12:48:55.716963: step: 788/1541, loss: 0.16464954614639282 2023-01-23 12:48:56.624151: step: 790/1541, loss: 0.10351795703172684 2023-01-23 12:48:57.530556: step: 792/1541, loss: 0.004317734390497208 2023-01-23 12:48:58.423612: step: 794/1541, loss: 0.23948821425437927 2023-01-23 12:48:59.316273: step: 796/1541, loss: 0.004932440351694822 2023-01-23 12:49:00.257909: step: 798/1541, loss: 0.000988866901025176 2023-01-23 12:49:01.153361: step: 800/1541, loss: 0.034669190645217896 2023-01-23 12:49:02.088020: step: 802/1541, loss: 0.2054949700832367 2023-01-23 12:49:02.983546: step: 804/1541, loss: 0.0014158457051962614 2023-01-23 12:49:03.858458: step: 806/1541, loss: 0.07178507000207901 2023-01-23 12:49:04.727992: step: 808/1541, loss: 0.00038797830347903073 2023-01-23 12:49:05.710761: step: 810/1541, loss: 0.37294435501098633 2023-01-23 12:49:06.569563: step: 812/1541, loss: 0.019991204142570496 2023-01-23 12:49:07.534859: step: 814/1541, loss: 0.15221726894378662 2023-01-23 12:49:08.437587: step: 816/1541, loss: 0.0333060547709465 2023-01-23 12:49:09.329961: step: 818/1541, loss: 0.17509450018405914 2023-01-23 12:49:10.243990: step: 820/1541, loss: 0.0026901820674538612 2023-01-23 12:49:11.220235: step: 822/1541, loss: 0.00920487754046917 2023-01-23 12:49:12.120998: step: 824/1541, loss: 0.005617271177470684 2023-01-23 12:49:13.020644: step: 826/1541, loss: 0.204498291015625 2023-01-23 12:49:13.947771: step: 828/1541, loss: 0.04866069555282593 2023-01-23 12:49:14.851853: step: 830/1541, loss: 0.059752948582172394 2023-01-23 12:49:15.745195: step: 832/1541, loss: 0.09746405482292175 2023-01-23 12:49:16.631012: step: 834/1541, loss: 0.0017791276331990957 2023-01-23 12:49:17.577547: step: 836/1541, loss: 0.023377692326903343 2023-01-23 12:49:18.438608: step: 838/1541, loss: 0.2032848745584488 2023-01-23 12:49:19.319357: step: 840/1541, loss: 0.11858617514371872 2023-01-23 12:49:20.220366: step: 842/1541, loss: 0.33331558108329773 2023-01-23 12:49:21.129355: step: 844/1541, loss: 0.03143952414393425 2023-01-23 12:49:22.035023: step: 846/1541, loss: 0.0030033166985958815 2023-01-23 12:49:22.958483: step: 848/1541, loss: 0.046849578619003296 2023-01-23 12:49:23.847464: step: 850/1541, loss: 0.23055694997310638 2023-01-23 12:49:24.748800: step: 852/1541, loss: 0.03847142308950424 2023-01-23 12:49:25.693148: step: 854/1541, loss: 0.18083642423152924 2023-01-23 12:49:26.549765: step: 856/1541, loss: 0.0008116972749121487 2023-01-23 12:49:27.464859: step: 858/1541, loss: 0.0014120228588581085 2023-01-23 12:49:28.351915: step: 860/1541, loss: 0.39770427346229553 2023-01-23 12:49:29.252120: step: 862/1541, loss: 0.01162104681134224 2023-01-23 12:49:30.116158: step: 864/1541, loss: 0.0473431758582592 2023-01-23 12:49:31.001244: step: 866/1541, loss: 0.03943867236375809 2023-01-23 12:49:31.925906: step: 868/1541, loss: 1.3636654615402222 2023-01-23 12:49:32.832553: step: 870/1541, loss: 0.03600294515490532 2023-01-23 12:49:33.747160: step: 872/1541, loss: 0.07732564210891724 2023-01-23 12:49:34.640263: step: 874/1541, loss: 0.07368495315313339 2023-01-23 12:49:35.551509: step: 876/1541, loss: 0.5795609951019287 2023-01-23 12:49:36.448933: step: 878/1541, loss: 0.09592999517917633 2023-01-23 12:49:37.325062: step: 880/1541, loss: 0.3886106014251709 2023-01-23 12:49:38.215277: step: 882/1541, loss: 0.035312049090862274 2023-01-23 12:49:39.146054: step: 884/1541, loss: 0.04247191548347473 2023-01-23 12:49:40.040692: step: 886/1541, loss: 0.011164596304297447 2023-01-23 12:49:40.980808: step: 888/1541, loss: 0.04276084154844284 2023-01-23 12:49:41.934160: step: 890/1541, loss: 0.4805251955986023 2023-01-23 12:49:42.831306: step: 892/1541, loss: 0.06020132079720497 2023-01-23 12:49:43.741049: step: 894/1541, loss: 0.1306965947151184 2023-01-23 12:49:44.621807: step: 896/1541, loss: 0.0015394100919365883 2023-01-23 12:49:45.492721: step: 898/1541, loss: 0.028942005708813667 2023-01-23 12:49:46.381890: step: 900/1541, loss: 0.009555214084684849 2023-01-23 12:49:47.282478: step: 902/1541, loss: 0.04632541909813881 2023-01-23 12:49:48.169671: step: 904/1541, loss: 0.02536400780081749 2023-01-23 12:49:49.065521: step: 906/1541, loss: 0.007035941816866398 2023-01-23 12:49:49.976455: step: 908/1541, loss: 0.03495832532644272 2023-01-23 12:49:50.845692: step: 910/1541, loss: 0.07221534103155136 2023-01-23 12:49:51.744479: step: 912/1541, loss: 0.0077326493337750435 2023-01-23 12:49:52.634862: step: 914/1541, loss: 0.030465740710496902 2023-01-23 12:49:53.525903: step: 916/1541, loss: 0.02140074037015438 2023-01-23 12:49:54.380638: step: 918/1541, loss: 0.06284932047128677 2023-01-23 12:49:55.269013: step: 920/1541, loss: 0.025295676663517952 2023-01-23 12:49:56.137579: step: 922/1541, loss: 0.44474413990974426 2023-01-23 12:49:57.077858: step: 924/1541, loss: 0.23059894144535065 2023-01-23 12:49:57.982972: step: 926/1541, loss: 0.17223544418811798 2023-01-23 12:49:58.921073: step: 928/1541, loss: 0.19423389434814453 2023-01-23 12:49:59.825597: step: 930/1541, loss: 0.02314552292227745 2023-01-23 12:50:00.760909: step: 932/1541, loss: 0.014319348149001598 2023-01-23 12:50:01.662580: step: 934/1541, loss: 0.0034465761855244637 2023-01-23 12:50:02.574043: step: 936/1541, loss: 0.017057104036211967 2023-01-23 12:50:03.460881: step: 938/1541, loss: 0.0016430009854957461 2023-01-23 12:50:04.389053: step: 940/1541, loss: 0.08121365308761597 2023-01-23 12:50:05.239204: step: 942/1541, loss: 0.05976342037320137 2023-01-23 12:50:06.144694: step: 944/1541, loss: 0.0022269822657108307 2023-01-23 12:50:07.049571: step: 946/1541, loss: 0.14134088158607483 2023-01-23 12:50:07.956858: step: 948/1541, loss: 0.10172650963068008 2023-01-23 12:50:08.846664: step: 950/1541, loss: 0.23716524243354797 2023-01-23 12:50:09.748620: step: 952/1541, loss: 0.11816380172967911 2023-01-23 12:50:10.647841: step: 954/1541, loss: 0.1791573315858841 2023-01-23 12:50:11.525705: step: 956/1541, loss: 0.23260800540447235 2023-01-23 12:50:12.428108: step: 958/1541, loss: 0.07590910792350769 2023-01-23 12:50:13.310597: step: 960/1541, loss: 0.049749523401260376 2023-01-23 12:50:14.218362: step: 962/1541, loss: 0.006542001385241747 2023-01-23 12:50:15.124181: step: 964/1541, loss: 0.04079059511423111 2023-01-23 12:50:16.010476: step: 966/1541, loss: 0.05926583707332611 2023-01-23 12:50:16.904266: step: 968/1541, loss: 0.012880828231573105 2023-01-23 12:50:17.769771: step: 970/1541, loss: 0.06071431189775467 2023-01-23 12:50:18.621483: step: 972/1541, loss: 0.09415486454963684 2023-01-23 12:50:19.519186: step: 974/1541, loss: 0.27857184410095215 2023-01-23 12:50:20.436123: step: 976/1541, loss: 0.05583702027797699 2023-01-23 12:50:21.410961: step: 978/1541, loss: 0.12036775052547455 2023-01-23 12:50:22.322007: step: 980/1541, loss: 0.0899353101849556 2023-01-23 12:50:23.175805: step: 982/1541, loss: 0.05731847509741783 2023-01-23 12:50:24.055103: step: 984/1541, loss: 0.007843151688575745 2023-01-23 12:50:24.924014: step: 986/1541, loss: 0.09568564593791962 2023-01-23 12:50:26.045173: step: 988/1541, loss: 0.27190303802490234 2023-01-23 12:50:26.968470: step: 990/1541, loss: 0.06347699463367462 2023-01-23 12:50:27.870195: step: 992/1541, loss: 0.03960920125246048 2023-01-23 12:50:28.771949: step: 994/1541, loss: 0.040497787296772 2023-01-23 12:50:29.716743: step: 996/1541, loss: 0.0035135254729539156 2023-01-23 12:50:30.598806: step: 998/1541, loss: 0.00353897362947464 2023-01-23 12:50:31.453496: step: 1000/1541, loss: 0.0031202808022499084 2023-01-23 12:50:32.358361: step: 1002/1541, loss: 0.03513331711292267 2023-01-23 12:50:33.271656: step: 1004/1541, loss: 0.09003648161888123 2023-01-23 12:50:34.168081: step: 1006/1541, loss: 0.10079528391361237 2023-01-23 12:50:35.053348: step: 1008/1541, loss: 0.43870750069618225 2023-01-23 12:50:35.948830: step: 1010/1541, loss: 0.035482533276081085 2023-01-23 12:50:36.813190: step: 1012/1541, loss: 0.3000198304653168 2023-01-23 12:50:37.703793: step: 1014/1541, loss: 0.5103501677513123 2023-01-23 12:50:38.602319: step: 1016/1541, loss: 0.05386679619550705 2023-01-23 12:50:39.479679: step: 1018/1541, loss: 0.15065789222717285 2023-01-23 12:50:40.459361: step: 1020/1541, loss: 0.017649756744503975 2023-01-23 12:50:41.337016: step: 1022/1541, loss: 0.006573161110281944 2023-01-23 12:50:42.268817: step: 1024/1541, loss: 0.013129416853189468 2023-01-23 12:50:43.162924: step: 1026/1541, loss: 0.10598810017108917 2023-01-23 12:50:44.054528: step: 1028/1541, loss: 0.8423148393630981 2023-01-23 12:50:44.946761: step: 1030/1541, loss: 0.32306164503097534 2023-01-23 12:50:45.831325: step: 1032/1541, loss: 0.467055082321167 2023-01-23 12:50:46.765773: step: 1034/1541, loss: 0.03336166962981224 2023-01-23 12:50:47.636331: step: 1036/1541, loss: 0.17949454486370087 2023-01-23 12:50:48.502647: step: 1038/1541, loss: 0.0009385913144797087 2023-01-23 12:50:49.358985: step: 1040/1541, loss: 0.35975509881973267 2023-01-23 12:50:50.278869: step: 1042/1541, loss: 0.010586279444396496 2023-01-23 12:50:51.191425: step: 1044/1541, loss: 0.30346858501434326 2023-01-23 12:50:52.144266: step: 1046/1541, loss: 0.009400714188814163 2023-01-23 12:50:53.020060: step: 1048/1541, loss: 0.09746433049440384 2023-01-23 12:50:53.939406: step: 1050/1541, loss: 0.019314076751470566 2023-01-23 12:50:54.819182: step: 1052/1541, loss: 0.03730872645974159 2023-01-23 12:50:55.680295: step: 1054/1541, loss: 1.3483706712722778 2023-01-23 12:50:56.594544: step: 1056/1541, loss: 0.018644236028194427 2023-01-23 12:50:57.469614: step: 1058/1541, loss: 0.04774804785847664 2023-01-23 12:50:58.333708: step: 1060/1541, loss: 0.017827996984124184 2023-01-23 12:50:59.218515: step: 1062/1541, loss: 0.03235948085784912 2023-01-23 12:51:00.102356: step: 1064/1541, loss: 0.11444660276174545 2023-01-23 12:51:00.975710: step: 1066/1541, loss: 0.6415947079658508 2023-01-23 12:51:01.948010: step: 1068/1541, loss: 0.0429130494594574 2023-01-23 12:51:02.872100: step: 1070/1541, loss: 0.05484459921717644 2023-01-23 12:51:03.856426: step: 1072/1541, loss: 0.23256874084472656 2023-01-23 12:51:04.744394: step: 1074/1541, loss: 0.11588820070028305 2023-01-23 12:51:05.659414: step: 1076/1541, loss: 0.05778570473194122 2023-01-23 12:51:06.551804: step: 1078/1541, loss: 0.11761099845170975 2023-01-23 12:51:07.461869: step: 1080/1541, loss: 0.009142271243035793 2023-01-23 12:51:08.383420: step: 1082/1541, loss: 0.00760574359446764 2023-01-23 12:51:09.273606: step: 1084/1541, loss: 0.522825300693512 2023-01-23 12:51:10.220955: step: 1086/1541, loss: 0.010159425437450409 2023-01-23 12:51:11.073235: step: 1088/1541, loss: 0.06176315248012543 2023-01-23 12:51:11.941790: step: 1090/1541, loss: 0.1208227127790451 2023-01-23 12:51:12.785752: step: 1092/1541, loss: 0.021113570779561996 2023-01-23 12:51:13.730917: step: 1094/1541, loss: 0.03201061487197876 2023-01-23 12:51:14.722764: step: 1096/1541, loss: 0.07020621746778488 2023-01-23 12:51:15.683298: step: 1098/1541, loss: 0.0468885712325573 2023-01-23 12:51:16.591059: step: 1100/1541, loss: 0.05179181694984436 2023-01-23 12:51:17.495168: step: 1102/1541, loss: 0.005426083225756884 2023-01-23 12:51:18.408180: step: 1104/1541, loss: 0.12243672460317612 2023-01-23 12:51:19.256683: step: 1106/1541, loss: 0.013900136575102806 2023-01-23 12:51:20.235683: step: 1108/1541, loss: 1.6865520477294922 2023-01-23 12:51:21.181201: step: 1110/1541, loss: 0.03218291327357292 2023-01-23 12:51:22.088593: step: 1112/1541, loss: 0.02000621147453785 2023-01-23 12:51:23.026904: step: 1114/1541, loss: 0.2664611041545868 2023-01-23 12:51:23.932228: step: 1116/1541, loss: 0.07273639738559723 2023-01-23 12:51:24.893702: step: 1118/1541, loss: 0.7908318042755127 2023-01-23 12:51:25.772180: step: 1120/1541, loss: 0.006764700170606375 2023-01-23 12:51:26.641180: step: 1122/1541, loss: 0.024324646219611168 2023-01-23 12:51:27.529691: step: 1124/1541, loss: 0.09624366462230682 2023-01-23 12:51:28.421604: step: 1126/1541, loss: 0.005639452487230301 2023-01-23 12:51:29.314476: step: 1128/1541, loss: 0.1754089742898941 2023-01-23 12:51:30.235574: step: 1130/1541, loss: 0.07722976058721542 2023-01-23 12:51:31.120964: step: 1132/1541, loss: 0.048987094312906265 2023-01-23 12:51:32.084153: step: 1134/1541, loss: 0.016711167991161346 2023-01-23 12:51:33.005200: step: 1136/1541, loss: 0.15874333679676056 2023-01-23 12:51:33.877570: step: 1138/1541, loss: 0.38920915126800537 2023-01-23 12:51:34.789053: step: 1140/1541, loss: 0.05100462958216667 2023-01-23 12:51:35.694078: step: 1142/1541, loss: 0.13686883449554443 2023-01-23 12:51:36.596792: step: 1144/1541, loss: 0.10640565305948257 2023-01-23 12:51:37.499506: step: 1146/1541, loss: 0.20918944478034973 2023-01-23 12:51:38.493173: step: 1148/1541, loss: 0.03764379024505615 2023-01-23 12:51:39.396371: step: 1150/1541, loss: 0.1006685420870781 2023-01-23 12:51:40.270680: step: 1152/1541, loss: 0.03052978217601776 2023-01-23 12:51:41.167600: step: 1154/1541, loss: 0.04036421701312065 2023-01-23 12:51:42.065783: step: 1156/1541, loss: 0.06299964338541031 2023-01-23 12:51:42.929329: step: 1158/1541, loss: 0.05907610058784485 2023-01-23 12:51:43.860188: step: 1160/1541, loss: 0.03845176845788956 2023-01-23 12:51:44.774170: step: 1162/1541, loss: 0.06761578470468521 2023-01-23 12:51:45.618157: step: 1164/1541, loss: 0.1345089077949524 2023-01-23 12:51:46.494263: step: 1166/1541, loss: 0.008211951702833176 2023-01-23 12:51:47.376229: step: 1168/1541, loss: 0.04494099318981171 2023-01-23 12:51:48.281047: step: 1170/1541, loss: 0.007126500830054283 2023-01-23 12:51:49.187308: step: 1172/1541, loss: 0.6555222868919373 2023-01-23 12:51:50.096405: step: 1174/1541, loss: 0.3673449158668518 2023-01-23 12:51:50.962692: step: 1176/1541, loss: 0.015882987529039383 2023-01-23 12:51:51.867069: step: 1178/1541, loss: 0.06289404630661011 2023-01-23 12:51:52.748179: step: 1180/1541, loss: 0.027030611410737038 2023-01-23 12:51:53.629018: step: 1182/1541, loss: 0.30931785702705383 2023-01-23 12:51:54.525813: step: 1184/1541, loss: 0.02098671719431877 2023-01-23 12:51:55.464051: step: 1186/1541, loss: 0.05953443795442581 2023-01-23 12:51:56.419940: step: 1188/1541, loss: 0.04943438991904259 2023-01-23 12:51:57.303794: step: 1190/1541, loss: 0.0031065954826772213 2023-01-23 12:51:58.192766: step: 1192/1541, loss: 0.016004109755158424 2023-01-23 12:51:59.110908: step: 1194/1541, loss: 0.03878467157483101 2023-01-23 12:52:00.073314: step: 1196/1541, loss: 0.1468789428472519 2023-01-23 12:52:00.972243: step: 1198/1541, loss: 0.14286093413829803 2023-01-23 12:52:01.878171: step: 1200/1541, loss: 0.13198503851890564 2023-01-23 12:52:02.750168: step: 1202/1541, loss: 0.17924706637859344 2023-01-23 12:52:03.661404: step: 1204/1541, loss: 0.23551836609840393 2023-01-23 12:52:04.623150: step: 1206/1541, loss: 0.2042665332555771 2023-01-23 12:52:05.568431: step: 1208/1541, loss: 0.22628509998321533 2023-01-23 12:52:06.419083: step: 1210/1541, loss: 0.0038000433705747128 2023-01-23 12:52:07.341317: step: 1212/1541, loss: 0.021833613514900208 2023-01-23 12:52:08.229244: step: 1214/1541, loss: 0.001912547741085291 2023-01-23 12:52:09.100990: step: 1216/1541, loss: 0.23090767860412598 2023-01-23 12:52:10.033792: step: 1218/1541, loss: 0.1034044623374939 2023-01-23 12:52:10.951746: step: 1220/1541, loss: 0.12965907156467438 2023-01-23 12:52:11.829892: step: 1222/1541, loss: 0.31827569007873535 2023-01-23 12:52:12.736310: step: 1224/1541, loss: 0.12783770263195038 2023-01-23 12:52:13.647135: step: 1226/1541, loss: 0.025910625234246254 2023-01-23 12:52:14.547086: step: 1228/1541, loss: 0.2199186384677887 2023-01-23 12:52:15.427491: step: 1230/1541, loss: 0.0035982877016067505 2023-01-23 12:52:16.280528: step: 1232/1541, loss: 0.18334351480007172 2023-01-23 12:52:17.184309: step: 1234/1541, loss: 0.20339952409267426 2023-01-23 12:52:18.065169: step: 1236/1541, loss: 0.024894878268241882 2023-01-23 12:52:18.956787: step: 1238/1541, loss: 0.32575175166130066 2023-01-23 12:52:19.813064: step: 1240/1541, loss: 0.126214399933815 2023-01-23 12:52:20.711007: step: 1242/1541, loss: 0.23041653633117676 2023-01-23 12:52:21.615233: step: 1244/1541, loss: 0.1334242820739746 2023-01-23 12:52:22.515549: step: 1246/1541, loss: 0.1919204294681549 2023-01-23 12:52:23.397779: step: 1248/1541, loss: 0.15809696912765503 2023-01-23 12:52:24.296282: step: 1250/1541, loss: 0.06501799821853638 2023-01-23 12:52:25.270892: step: 1252/1541, loss: 0.09828589856624603 2023-01-23 12:52:26.185081: step: 1254/1541, loss: 0.17477641999721527 2023-01-23 12:52:27.109273: step: 1256/1541, loss: 0.015215124003589153 2023-01-23 12:52:28.002868: step: 1258/1541, loss: 0.2187768667936325 2023-01-23 12:52:28.918845: step: 1260/1541, loss: 0.7064292430877686 2023-01-23 12:52:29.792645: step: 1262/1541, loss: 0.008954095654189587 2023-01-23 12:52:30.691319: step: 1264/1541, loss: 0.5501185655593872 2023-01-23 12:52:31.566574: step: 1266/1541, loss: 0.09792737662792206 2023-01-23 12:52:32.437957: step: 1268/1541, loss: 0.02466781623661518 2023-01-23 12:52:33.345497: step: 1270/1541, loss: 0.07366533577442169 2023-01-23 12:52:34.254718: step: 1272/1541, loss: 0.1279362142086029 2023-01-23 12:52:35.251608: step: 1274/1541, loss: 0.00584210641682148 2023-01-23 12:52:36.124769: step: 1276/1541, loss: 0.07712718099355698 2023-01-23 12:52:37.020696: step: 1278/1541, loss: 0.006384974345564842 2023-01-23 12:52:37.906537: step: 1280/1541, loss: 0.043107595294713974 2023-01-23 12:52:38.793686: step: 1282/1541, loss: 0.006748927291482687 2023-01-23 12:52:39.801866: step: 1284/1541, loss: 0.03570539131760597 2023-01-23 12:52:40.689240: step: 1286/1541, loss: 0.023991316556930542 2023-01-23 12:52:41.589875: step: 1288/1541, loss: 0.042638882994651794 2023-01-23 12:52:42.450754: step: 1290/1541, loss: 0.013605983927845955 2023-01-23 12:52:43.338878: step: 1292/1541, loss: 0.08715461939573288 2023-01-23 12:52:44.223868: step: 1294/1541, loss: 0.004933366551995277 2023-01-23 12:52:45.129073: step: 1296/1541, loss: 0.03170119971036911 2023-01-23 12:52:46.009757: step: 1298/1541, loss: 0.005606648046523333 2023-01-23 12:52:46.892695: step: 1300/1541, loss: 0.034287285059690475 2023-01-23 12:52:47.821856: step: 1302/1541, loss: 0.36250415444374084 2023-01-23 12:52:48.709250: step: 1304/1541, loss: 0.16371826827526093 2023-01-23 12:52:49.579571: step: 1306/1541, loss: 0.0030719672795385122 2023-01-23 12:52:50.441993: step: 1308/1541, loss: 0.03140875697135925 2023-01-23 12:52:51.316449: step: 1310/1541, loss: 0.002327181398868561 2023-01-23 12:52:52.265644: step: 1312/1541, loss: 0.4491935968399048 2023-01-23 12:52:53.146291: step: 1314/1541, loss: 0.16486355662345886 2023-01-23 12:52:54.020825: step: 1316/1541, loss: 0.3412688970565796 2023-01-23 12:52:54.909696: step: 1318/1541, loss: 0.6356620788574219 2023-01-23 12:52:55.816004: step: 1320/1541, loss: 0.0595397874712944 2023-01-23 12:52:56.703371: step: 1322/1541, loss: 0.11344572901725769 2023-01-23 12:52:57.580060: step: 1324/1541, loss: 0.12016611546278 2023-01-23 12:52:58.482762: step: 1326/1541, loss: 0.16477367281913757 2023-01-23 12:52:59.419323: step: 1328/1541, loss: 0.003528811037540436 2023-01-23 12:53:00.395320: step: 1330/1541, loss: 0.0599648542702198 2023-01-23 12:53:01.285135: step: 1332/1541, loss: 0.003984540700912476 2023-01-23 12:53:02.208584: step: 1334/1541, loss: 0.1820220947265625 2023-01-23 12:53:03.192877: step: 1336/1541, loss: 0.02955222874879837 2023-01-23 12:53:04.096759: step: 1338/1541, loss: 0.21656063199043274 2023-01-23 12:53:04.977683: step: 1340/1541, loss: 0.002754833083599806 2023-01-23 12:53:05.850957: step: 1342/1541, loss: 0.20056241750717163 2023-01-23 12:53:06.725502: step: 1344/1541, loss: 0.007660957984626293 2023-01-23 12:53:07.597173: step: 1346/1541, loss: 0.037668149918317795 2023-01-23 12:53:08.537239: step: 1348/1541, loss: 0.052274055778980255 2023-01-23 12:53:09.513092: step: 1350/1541, loss: 0.0013242517597973347 2023-01-23 12:53:10.421137: step: 1352/1541, loss: 0.0236805509775877 2023-01-23 12:53:11.364242: step: 1354/1541, loss: 0.04092559590935707 2023-01-23 12:53:12.289758: step: 1356/1541, loss: 0.1494445651769638 2023-01-23 12:53:13.197540: step: 1358/1541, loss: 0.02859235182404518 2023-01-23 12:53:14.125856: step: 1360/1541, loss: 0.20894759893417358 2023-01-23 12:53:15.010519: step: 1362/1541, loss: 0.46499279141426086 2023-01-23 12:53:16.009973: step: 1364/1541, loss: 0.02892809361219406 2023-01-23 12:53:16.905814: step: 1366/1541, loss: 0.5786864757537842 2023-01-23 12:53:17.811324: step: 1368/1541, loss: 0.28485289216041565 2023-01-23 12:53:18.729911: step: 1370/1541, loss: 0.7440303564071655 2023-01-23 12:53:19.622306: step: 1372/1541, loss: 0.009194309823215008 2023-01-23 12:53:20.521495: step: 1374/1541, loss: 0.05088617652654648 2023-01-23 12:53:21.494506: step: 1376/1541, loss: 0.7350720167160034 2023-01-23 12:53:22.392860: step: 1378/1541, loss: 0.12443521618843079 2023-01-23 12:53:23.286834: step: 1380/1541, loss: 0.352203369140625 2023-01-23 12:53:24.187016: step: 1382/1541, loss: 0.15740488469600677 2023-01-23 12:53:25.111468: step: 1384/1541, loss: 0.11565406620502472 2023-01-23 12:53:26.059488: step: 1386/1541, loss: 0.01169333141297102 2023-01-23 12:53:27.035144: step: 1388/1541, loss: 0.005605082027614117 2023-01-23 12:53:27.947530: step: 1390/1541, loss: 0.054194528609514236 2023-01-23 12:53:28.812402: step: 1392/1541, loss: 0.11530670523643494 2023-01-23 12:53:29.760345: step: 1394/1541, loss: 0.042424872517585754 2023-01-23 12:53:30.655497: step: 1396/1541, loss: 0.024001220241189003 2023-01-23 12:53:31.544730: step: 1398/1541, loss: 0.12757906317710876 2023-01-23 12:53:32.550708: step: 1400/1541, loss: 0.045893024653196335 2023-01-23 12:53:33.430149: step: 1402/1541, loss: 0.15520019829273224 2023-01-23 12:53:34.305989: step: 1404/1541, loss: 0.2822684645652771 2023-01-23 12:53:35.211513: step: 1406/1541, loss: 0.053170960396528244 2023-01-23 12:53:36.146337: step: 1408/1541, loss: 0.08608352392911911 2023-01-23 12:53:37.056184: step: 1410/1541, loss: 0.0067674992606043816 2023-01-23 12:53:37.993114: step: 1412/1541, loss: 0.04927699267864227 2023-01-23 12:53:38.911373: step: 1414/1541, loss: 0.12852336466312408 2023-01-23 12:53:39.804261: step: 1416/1541, loss: 0.01925213634967804 2023-01-23 12:53:40.685664: step: 1418/1541, loss: 0.009967314079403877 2023-01-23 12:53:41.622399: step: 1420/1541, loss: 0.0005297799361869693 2023-01-23 12:53:42.514882: step: 1422/1541, loss: 0.39949384331703186 2023-01-23 12:53:43.374912: step: 1424/1541, loss: 0.9105861186981201 2023-01-23 12:53:44.248360: step: 1426/1541, loss: 0.00219051749445498 2023-01-23 12:53:45.120330: step: 1428/1541, loss: 0.09066759794950485 2023-01-23 12:53:46.011434: step: 1430/1541, loss: 0.31356000900268555 2023-01-23 12:53:46.920764: step: 1432/1541, loss: 0.16106151044368744 2023-01-23 12:53:47.835481: step: 1434/1541, loss: 0.007002724334597588 2023-01-23 12:53:48.723057: step: 1436/1541, loss: 0.0039705862291157246 2023-01-23 12:53:49.702686: step: 1438/1541, loss: 0.06087087467312813 2023-01-23 12:53:50.551222: step: 1440/1541, loss: 0.009585563093423843 2023-01-23 12:53:51.417473: step: 1442/1541, loss: 0.015561610460281372 2023-01-23 12:53:52.403658: step: 1444/1541, loss: 0.2583433985710144 2023-01-23 12:53:53.306887: step: 1446/1541, loss: 0.17954827845096588 2023-01-23 12:53:54.201250: step: 1448/1541, loss: 0.027625489979982376 2023-01-23 12:53:55.076736: step: 1450/1541, loss: 0.07536563277244568 2023-01-23 12:53:55.964355: step: 1452/1541, loss: 0.12104413658380508 2023-01-23 12:53:56.836873: step: 1454/1541, loss: 0.019113745540380478 2023-01-23 12:53:57.809704: step: 1456/1541, loss: 0.007871067151427269 2023-01-23 12:53:58.675004: step: 1458/1541, loss: 0.01112340483814478 2023-01-23 12:53:59.545289: step: 1460/1541, loss: 0.14581668376922607 2023-01-23 12:54:00.483609: step: 1462/1541, loss: 0.001577988383360207 2023-01-23 12:54:01.349995: step: 1464/1541, loss: 0.08763513714075089 2023-01-23 12:54:02.234540: step: 1466/1541, loss: 0.10008267313241959 2023-01-23 12:54:03.128848: step: 1468/1541, loss: 0.11129443347454071 2023-01-23 12:54:04.099326: step: 1470/1541, loss: 0.009679893031716347 2023-01-23 12:54:05.007234: step: 1472/1541, loss: 0.25289487838745117 2023-01-23 12:54:05.925844: step: 1474/1541, loss: 0.018299546092748642 2023-01-23 12:54:06.822060: step: 1476/1541, loss: 0.09753911942243576 2023-01-23 12:54:07.745499: step: 1478/1541, loss: 0.1796395629644394 2023-01-23 12:54:08.683453: step: 1480/1541, loss: 0.01824740320444107 2023-01-23 12:54:09.562093: step: 1482/1541, loss: 0.05900292843580246 2023-01-23 12:54:10.417205: step: 1484/1541, loss: 0.005723288748413324 2023-01-23 12:54:11.311065: step: 1486/1541, loss: 0.4077696204185486 2023-01-23 12:54:12.261943: step: 1488/1541, loss: 0.007069111336022615 2023-01-23 12:54:13.177803: step: 1490/1541, loss: 0.14280064404010773 2023-01-23 12:54:14.127187: step: 1492/1541, loss: 0.07333075255155563 2023-01-23 12:54:15.017368: step: 1494/1541, loss: 0.11749714612960815 2023-01-23 12:54:16.000930: step: 1496/1541, loss: 0.000863337074406445 2023-01-23 12:54:16.853719: step: 1498/1541, loss: 0.029329124838113785 2023-01-23 12:54:17.799811: step: 1500/1541, loss: 0.07832527160644531 2023-01-23 12:54:18.662278: step: 1502/1541, loss: 0.006287143100053072 2023-01-23 12:54:19.549971: step: 1504/1541, loss: 0.03470728173851967 2023-01-23 12:54:20.480763: step: 1506/1541, loss: 0.4327303171157837 2023-01-23 12:54:21.376897: step: 1508/1541, loss: 0.029438428580760956 2023-01-23 12:54:22.252066: step: 1510/1541, loss: 0.04534035548567772 2023-01-23 12:54:23.184146: step: 1512/1541, loss: 0.0023632103111594915 2023-01-23 12:54:24.069277: step: 1514/1541, loss: 0.04071962088346481 2023-01-23 12:54:24.946980: step: 1516/1541, loss: 0.007451372686773539 2023-01-23 12:54:25.810865: step: 1518/1541, loss: 0.06933377683162689 2023-01-23 12:54:26.711589: step: 1520/1541, loss: 0.10922345519065857 2023-01-23 12:54:27.576132: step: 1522/1541, loss: 0.007707946002483368 2023-01-23 12:54:28.466753: step: 1524/1541, loss: 0.33247727155685425 2023-01-23 12:54:29.390294: step: 1526/1541, loss: 0.06455137580633163 2023-01-23 12:54:30.286310: step: 1528/1541, loss: 0.145322784781456 2023-01-23 12:54:31.200173: step: 1530/1541, loss: 1.4395917654037476 2023-01-23 12:54:32.091573: step: 1532/1541, loss: 0.31197240948677063 2023-01-23 12:54:32.996899: step: 1534/1541, loss: 2.1468138694763184 2023-01-23 12:54:33.942544: step: 1536/1541, loss: 0.07058969140052795 2023-01-23 12:54:34.826481: step: 1538/1541, loss: 0.06151663139462471 2023-01-23 12:54:35.720289: step: 1540/1541, loss: 0.19676388800144196 2023-01-23 12:54:36.588458: step: 1542/1541, loss: 0.5592328310012817 2023-01-23 12:54:37.473399: step: 1544/1541, loss: 0.054624952375888824 2023-01-23 12:54:38.345075: step: 1546/1541, loss: 0.12897497415542603 2023-01-23 12:54:39.223335: step: 1548/1541, loss: 0.20506098866462708 2023-01-23 12:54:40.101274: step: 1550/1541, loss: 0.10845275968313217 2023-01-23 12:54:40.986155: step: 1552/1541, loss: 0.45648112893104553 2023-01-23 12:54:41.952720: step: 1554/1541, loss: 0.5025690197944641 2023-01-23 12:54:42.933545: step: 1556/1541, loss: 0.0166596882045269 2023-01-23 12:54:43.831961: step: 1558/1541, loss: 0.1656765341758728 2023-01-23 12:54:44.722063: step: 1560/1541, loss: 0.020856430754065514 2023-01-23 12:54:45.628752: step: 1562/1541, loss: 0.21734727919101715 2023-01-23 12:54:46.565650: step: 1564/1541, loss: 0.1612607091665268 2023-01-23 12:54:47.443402: step: 1566/1541, loss: 0.01901012845337391 2023-01-23 12:54:48.329943: step: 1568/1541, loss: 0.024103529751300812 2023-01-23 12:54:49.172944: step: 1570/1541, loss: 0.008389162831008434 2023-01-23 12:54:50.042163: step: 1572/1541, loss: 0.7027661800384521 2023-01-23 12:54:50.949201: step: 1574/1541, loss: 0.22942934930324554 2023-01-23 12:54:51.837091: step: 1576/1541, loss: 0.034440021961927414 2023-01-23 12:54:52.778244: step: 1578/1541, loss: 0.3864327371120453 2023-01-23 12:54:53.675083: step: 1580/1541, loss: 0.3901802897453308 2023-01-23 12:54:54.640113: step: 1582/1541, loss: 0.2922973036766052 2023-01-23 12:54:55.525015: step: 1584/1541, loss: 0.063906230032444 2023-01-23 12:54:56.453892: step: 1586/1541, loss: 0.02304961159825325 2023-01-23 12:54:57.387129: step: 1588/1541, loss: 0.025916989892721176 2023-01-23 12:54:58.270175: step: 1590/1541, loss: 0.5922805070877075 2023-01-23 12:54:59.153760: step: 1592/1541, loss: 0.04522135853767395 2023-01-23 12:55:00.044135: step: 1594/1541, loss: 0.08167757838964462 2023-01-23 12:55:00.952881: step: 1596/1541, loss: 0.0061268312856554985 2023-01-23 12:55:01.868248: step: 1598/1541, loss: 0.042459603399038315 2023-01-23 12:55:02.783877: step: 1600/1541, loss: 0.0063098156824707985 2023-01-23 12:55:03.655082: step: 1602/1541, loss: 0.057872168719768524 2023-01-23 12:55:04.568088: step: 1604/1541, loss: 0.030331913381814957 2023-01-23 12:55:05.429345: step: 1606/1541, loss: 0.0036284015513956547 2023-01-23 12:55:06.325220: step: 1608/1541, loss: 0.033677197992801666 2023-01-23 12:55:07.208352: step: 1610/1541, loss: 0.07038899511098862 2023-01-23 12:55:08.099088: step: 1612/1541, loss: 0.08253790438175201 2023-01-23 12:55:08.978205: step: 1614/1541, loss: 0.03619177266955376 2023-01-23 12:55:09.841268: step: 1616/1541, loss: 0.12121398746967316 2023-01-23 12:55:10.764472: step: 1618/1541, loss: 0.26652655005455017 2023-01-23 12:55:11.653219: step: 1620/1541, loss: 0.08210580796003342 2023-01-23 12:55:12.574181: step: 1622/1541, loss: 0.3978864848613739 2023-01-23 12:55:13.489960: step: 1624/1541, loss: 0.00974244438111782 2023-01-23 12:55:14.401614: step: 1626/1541, loss: 0.08150899410247803 2023-01-23 12:55:15.284650: step: 1628/1541, loss: 0.168132022023201 2023-01-23 12:55:16.182810: step: 1630/1541, loss: 0.047349389642477036 2023-01-23 12:55:17.048351: step: 1632/1541, loss: 0.25493180751800537 2023-01-23 12:55:17.961851: step: 1634/1541, loss: 0.03573444113135338 2023-01-23 12:55:18.847965: step: 1636/1541, loss: 0.049497418105602264 2023-01-23 12:55:19.732152: step: 1638/1541, loss: 0.017816875129938126 2023-01-23 12:55:20.623045: step: 1640/1541, loss: 0.06828157603740692 2023-01-23 12:55:21.524194: step: 1642/1541, loss: 0.08549545705318451 2023-01-23 12:55:22.401801: step: 1644/1541, loss: 0.007842116989195347 2023-01-23 12:55:23.292546: step: 1646/1541, loss: 0.1726798266172409 2023-01-23 12:55:24.129230: step: 1648/1541, loss: 0.009375616908073425 2023-01-23 12:55:25.045030: step: 1650/1541, loss: 0.32068967819213867 2023-01-23 12:55:25.947506: step: 1652/1541, loss: 0.10586469620466232 2023-01-23 12:55:26.903871: step: 1654/1541, loss: 0.22991307079792023 2023-01-23 12:55:27.775309: step: 1656/1541, loss: 0.1692398488521576 2023-01-23 12:55:28.652330: step: 1658/1541, loss: 0.014659429900348186 2023-01-23 12:55:29.522485: step: 1660/1541, loss: 0.08740386366844177 2023-01-23 12:55:30.377386: step: 1662/1541, loss: 0.06571219861507416 2023-01-23 12:55:31.237909: step: 1664/1541, loss: 0.11176062375307083 2023-01-23 12:55:32.170241: step: 1666/1541, loss: 0.010002178139984608 2023-01-23 12:55:33.035440: step: 1668/1541, loss: 0.00492779491469264 2023-01-23 12:55:33.944987: step: 1670/1541, loss: 0.006079231854528189 2023-01-23 12:55:34.903359: step: 1672/1541, loss: 0.010285906493663788 2023-01-23 12:55:35.805526: step: 1674/1541, loss: 0.010917849838733673 2023-01-23 12:55:36.676292: step: 1676/1541, loss: 0.05849570035934448 2023-01-23 12:55:37.559922: step: 1678/1541, loss: 0.053842660039663315 2023-01-23 12:55:38.513747: step: 1680/1541, loss: 0.08095306158065796 2023-01-23 12:55:39.392366: step: 1682/1541, loss: 0.19110065698623657 2023-01-23 12:55:40.313491: step: 1684/1541, loss: 0.011786618269979954 2023-01-23 12:55:41.251446: step: 1686/1541, loss: 0.06690333038568497 2023-01-23 12:55:42.148819: step: 1688/1541, loss: 0.09807458519935608 2023-01-23 12:55:43.117654: step: 1690/1541, loss: 0.012774148024618626 2023-01-23 12:55:44.010363: step: 1692/1541, loss: 0.12610314786434174 2023-01-23 12:55:44.918532: step: 1694/1541, loss: 0.13478729128837585 2023-01-23 12:55:45.801887: step: 1696/1541, loss: 0.12162354588508606 2023-01-23 12:55:46.710805: step: 1698/1541, loss: 0.008829799480736256 2023-01-23 12:55:47.625537: step: 1700/1541, loss: 0.293697714805603 2023-01-23 12:55:48.507432: step: 1702/1541, loss: 0.7271438837051392 2023-01-23 12:55:49.386952: step: 1704/1541, loss: 0.3991200029850006 2023-01-23 12:55:50.262249: step: 1706/1541, loss: 0.176533043384552 2023-01-23 12:55:51.203636: step: 1708/1541, loss: 0.008181796409189701 2023-01-23 12:55:52.124465: step: 1710/1541, loss: 0.09205195307731628 2023-01-23 12:55:53.091810: step: 1712/1541, loss: 0.2753978967666626 2023-01-23 12:55:53.999475: step: 1714/1541, loss: 0.34104517102241516 2023-01-23 12:55:54.905191: step: 1716/1541, loss: 0.007356834597885609 2023-01-23 12:55:55.801388: step: 1718/1541, loss: 0.10810244083404541 2023-01-23 12:55:56.679404: step: 1720/1541, loss: 0.03480812907218933 2023-01-23 12:55:57.576469: step: 1722/1541, loss: 0.0874468982219696 2023-01-23 12:55:58.438833: step: 1724/1541, loss: 0.02053234726190567 2023-01-23 12:55:59.350398: step: 1726/1541, loss: 0.0049074748530983925 2023-01-23 12:56:00.248134: step: 1728/1541, loss: 0.4148097336292267 2023-01-23 12:56:01.125415: step: 1730/1541, loss: 0.1017925888299942 2023-01-23 12:56:02.002538: step: 1732/1541, loss: 0.003519192337989807 2023-01-23 12:56:02.917047: step: 1734/1541, loss: 0.011386540718376637 2023-01-23 12:56:03.829649: step: 1736/1541, loss: 0.04803088307380676 2023-01-23 12:56:04.718348: step: 1738/1541, loss: 0.016581187024712563 2023-01-23 12:56:05.616622: step: 1740/1541, loss: 0.124090276658535 2023-01-23 12:56:06.489685: step: 1742/1541, loss: 0.014583052136003971 2023-01-23 12:56:07.363492: step: 1744/1541, loss: 0.136732280254364 2023-01-23 12:56:08.234676: step: 1746/1541, loss: 0.043801791965961456 2023-01-23 12:56:09.141973: step: 1748/1541, loss: 0.0673188790678978 2023-01-23 12:56:10.025544: step: 1750/1541, loss: 0.08686139434576035 2023-01-23 12:56:10.922876: step: 1752/1541, loss: 0.0845082700252533 2023-01-23 12:56:11.793537: step: 1754/1541, loss: 0.06605594605207443 2023-01-23 12:56:12.724932: step: 1756/1541, loss: 0.13206227123737335 2023-01-23 12:56:13.647082: step: 1758/1541, loss: 0.27464669942855835 2023-01-23 12:56:14.554996: step: 1760/1541, loss: 0.26705726981163025 2023-01-23 12:56:15.455191: step: 1762/1541, loss: 0.07115618884563446 2023-01-23 12:56:16.436465: step: 1764/1541, loss: 0.00257203821092844 2023-01-23 12:56:17.345780: step: 1766/1541, loss: 0.032002657651901245 2023-01-23 12:56:18.227797: step: 1768/1541, loss: 0.021506138145923615 2023-01-23 12:56:19.158042: step: 1770/1541, loss: 0.02605585567653179 2023-01-23 12:56:20.095038: step: 1772/1541, loss: 0.10061691701412201 2023-01-23 12:56:20.967041: step: 1774/1541, loss: 0.46636250615119934 2023-01-23 12:56:21.852239: step: 1776/1541, loss: 0.12253526598215103 2023-01-23 12:56:22.742075: step: 1778/1541, loss: 0.005849625915288925 2023-01-23 12:56:23.627274: step: 1780/1541, loss: 0.02254248969256878 2023-01-23 12:56:24.547388: step: 1782/1541, loss: 0.07925979793071747 2023-01-23 12:56:25.409804: step: 1784/1541, loss: 0.13538408279418945 2023-01-23 12:56:26.285639: step: 1786/1541, loss: 0.0450315959751606 2023-01-23 12:56:27.177159: step: 1788/1541, loss: 0.09497229754924774 2023-01-23 12:56:28.062634: step: 1790/1541, loss: 0.10443467646837234 2023-01-23 12:56:28.974689: step: 1792/1541, loss: 0.1315089762210846 2023-01-23 12:56:29.929679: step: 1794/1541, loss: 0.20343665778636932 2023-01-23 12:56:30.782566: step: 1796/1541, loss: 0.14867733418941498 2023-01-23 12:56:31.689741: step: 1798/1541, loss: 0.13702262938022614 2023-01-23 12:56:32.633453: step: 1800/1541, loss: 0.14384174346923828 2023-01-23 12:56:33.535216: step: 1802/1541, loss: 0.13467495143413544 2023-01-23 12:56:34.443338: step: 1804/1541, loss: 0.05710771307349205 2023-01-23 12:56:35.391027: step: 1806/1541, loss: 0.06736401468515396 2023-01-23 12:56:36.256129: step: 1808/1541, loss: 0.019365720450878143 2023-01-23 12:56:37.171495: step: 1810/1541, loss: 0.027493733912706375 2023-01-23 12:56:38.043020: step: 1812/1541, loss: 0.02106170542538166 2023-01-23 12:56:38.930614: step: 1814/1541, loss: 0.09894923865795135 2023-01-23 12:56:39.891440: step: 1816/1541, loss: 0.11217647045850754 2023-01-23 12:56:40.784979: step: 1818/1541, loss: 0.10750433802604675 2023-01-23 12:56:41.666147: step: 1820/1541, loss: 0.011364408768713474 2023-01-23 12:56:42.573494: step: 1822/1541, loss: 0.04253300651907921 2023-01-23 12:56:43.523365: step: 1824/1541, loss: 0.010080946609377861 2023-01-23 12:56:44.456456: step: 1826/1541, loss: 0.08084692806005478 2023-01-23 12:56:45.349513: step: 1828/1541, loss: 0.3454258441925049 2023-01-23 12:56:46.265064: step: 1830/1541, loss: 0.01874316670000553 2023-01-23 12:56:47.209481: step: 1832/1541, loss: 0.08384151756763458 2023-01-23 12:56:48.158653: step: 1834/1541, loss: 0.007591239642351866 2023-01-23 12:56:49.054142: step: 1836/1541, loss: 0.0382501445710659 2023-01-23 12:56:49.945862: step: 1838/1541, loss: 0.08561550825834274 2023-01-23 12:56:50.822617: step: 1840/1541, loss: 0.0065818168222904205 2023-01-23 12:56:51.715887: step: 1842/1541, loss: 0.10441164672374725 2023-01-23 12:56:52.609075: step: 1844/1541, loss: 0.007864536717534065 2023-01-23 12:56:53.471409: step: 1846/1541, loss: 0.013943090103566647 2023-01-23 12:56:54.407012: step: 1848/1541, loss: 0.2980644106864929 2023-01-23 12:56:55.357096: step: 1850/1541, loss: 0.31636250019073486 2023-01-23 12:56:56.278855: step: 1852/1541, loss: 0.15991456806659698 2023-01-23 12:56:57.190644: step: 1854/1541, loss: 0.016029002144932747 2023-01-23 12:56:58.116983: step: 1856/1541, loss: 0.3595508933067322 2023-01-23 12:56:59.040795: step: 1858/1541, loss: 0.022066064178943634 2023-01-23 12:56:59.937737: step: 1860/1541, loss: 0.6893834471702576 2023-01-23 12:57:00.823802: step: 1862/1541, loss: 0.0038783159106969833 2023-01-23 12:57:01.727821: step: 1864/1541, loss: 0.0777617022395134 2023-01-23 12:57:02.651986: step: 1866/1541, loss: 0.3769124448299408 2023-01-23 12:57:03.572291: step: 1868/1541, loss: 0.19450367987155914 2023-01-23 12:57:04.418772: step: 1870/1541, loss: 0.023740999400615692 2023-01-23 12:57:05.313415: step: 1872/1541, loss: 0.09348791092634201 2023-01-23 12:57:06.180998: step: 1874/1541, loss: 0.0024319870863109827 2023-01-23 12:57:07.040110: step: 1876/1541, loss: 2.187044143676758 2023-01-23 12:57:07.938905: step: 1878/1541, loss: 0.11882015317678452 2023-01-23 12:57:08.828998: step: 1880/1541, loss: 0.15778258442878723 2023-01-23 12:57:09.748309: step: 1882/1541, loss: 0.009337947703897953 2023-01-23 12:57:10.626651: step: 1884/1541, loss: 0.12447194755077362 2023-01-23 12:57:11.520085: step: 1886/1541, loss: 0.021738391369581223 2023-01-23 12:57:12.444326: step: 1888/1541, loss: 0.0891120582818985 2023-01-23 12:57:13.321184: step: 1890/1541, loss: 0.030402857810258865 2023-01-23 12:57:14.180010: step: 1892/1541, loss: 2.479677677154541 2023-01-23 12:57:15.112979: step: 1894/1541, loss: 0.13938789069652557 2023-01-23 12:57:16.018769: step: 1896/1541, loss: 0.015132814645767212 2023-01-23 12:57:16.915087: step: 1898/1541, loss: 0.1189451515674591 2023-01-23 12:57:17.804463: step: 1900/1541, loss: 0.07587211579084396 2023-01-23 12:57:18.698435: step: 1902/1541, loss: 0.08386467397212982 2023-01-23 12:57:19.567651: step: 1904/1541, loss: 0.3752747178077698 2023-01-23 12:57:20.445509: step: 1906/1541, loss: 0.06685665249824524 2023-01-23 12:57:21.308793: step: 1908/1541, loss: 0.4297540783882141 2023-01-23 12:57:22.177446: step: 1910/1541, loss: 0.1541537046432495 2023-01-23 12:57:23.100334: step: 1912/1541, loss: 0.10071375966072083 2023-01-23 12:57:23.972163: step: 1914/1541, loss: 0.001554253394715488 2023-01-23 12:57:24.888195: step: 1916/1541, loss: 0.2561701834201813 2023-01-23 12:57:25.792731: step: 1918/1541, loss: 0.026128511875867844 2023-01-23 12:57:26.709811: step: 1920/1541, loss: 0.06313100457191467 2023-01-23 12:57:27.628316: step: 1922/1541, loss: 0.021763458847999573 2023-01-23 12:57:28.515880: step: 1924/1541, loss: 0.016255205497145653 2023-01-23 12:57:29.372178: step: 1926/1541, loss: 0.0071945651434361935 2023-01-23 12:57:30.318735: step: 1928/1541, loss: 0.200147345662117 2023-01-23 12:57:31.279223: step: 1930/1541, loss: 0.0493394210934639 2023-01-23 12:57:32.260710: step: 1932/1541, loss: 0.10225234925746918 2023-01-23 12:57:33.224774: step: 1934/1541, loss: 0.027800433337688446 2023-01-23 12:57:34.118359: step: 1936/1541, loss: 0.409865140914917 2023-01-23 12:57:35.038897: step: 1938/1541, loss: 0.12001041322946548 2023-01-23 12:57:35.972264: step: 1940/1541, loss: 0.18029431998729706 2023-01-23 12:57:36.859769: step: 1942/1541, loss: 0.168724924325943 2023-01-23 12:57:37.751057: step: 1944/1541, loss: 0.09951720386743546 2023-01-23 12:57:38.624515: step: 1946/1541, loss: 0.17860230803489685 2023-01-23 12:57:39.526784: step: 1948/1541, loss: 0.03241148963570595 2023-01-23 12:57:40.420965: step: 1950/1541, loss: 0.06400874257087708 2023-01-23 12:57:41.343667: step: 1952/1541, loss: 0.07956408709287643 2023-01-23 12:57:42.242356: step: 1954/1541, loss: 0.13846062123775482 2023-01-23 12:57:43.134593: step: 1956/1541, loss: 0.04684952646493912 2023-01-23 12:57:44.053301: step: 1958/1541, loss: 0.09552155435085297 2023-01-23 12:57:44.956293: step: 1960/1541, loss: 0.009594676084816456 2023-01-23 12:57:45.868953: step: 1962/1541, loss: 0.02814318798482418 2023-01-23 12:57:46.767270: step: 1964/1541, loss: 0.07798069715499878 2023-01-23 12:57:47.667723: step: 1966/1541, loss: 0.060509465634822845 2023-01-23 12:57:48.562175: step: 1968/1541, loss: 0.008789172396063805 2023-01-23 12:57:49.448002: step: 1970/1541, loss: 0.14006152749061584 2023-01-23 12:57:50.339929: step: 1972/1541, loss: 0.016398724168539047 2023-01-23 12:57:51.241709: step: 1974/1541, loss: 0.13325932621955872 2023-01-23 12:57:52.140769: step: 1976/1541, loss: 0.030899863690137863 2023-01-23 12:57:53.050226: step: 1978/1541, loss: 0.08355781435966492 2023-01-23 12:57:53.957584: step: 1980/1541, loss: 0.049336355179548264 2023-01-23 12:57:54.843059: step: 1982/1541, loss: 0.08830936253070831 2023-01-23 12:57:55.740439: step: 1984/1541, loss: 0.007591149304062128 2023-01-23 12:57:56.662553: step: 1986/1541, loss: 0.011551238596439362 2023-01-23 12:57:57.603504: step: 1988/1541, loss: 0.008406098000705242 2023-01-23 12:57:58.506589: step: 1990/1541, loss: 0.3854954242706299 2023-01-23 12:57:59.473707: step: 1992/1541, loss: 0.1057441234588623 2023-01-23 12:58:00.362103: step: 1994/1541, loss: 0.06442061066627502 2023-01-23 12:58:01.259357: step: 1996/1541, loss: 0.36336398124694824 2023-01-23 12:58:02.241587: step: 1998/1541, loss: 0.26170068979263306 2023-01-23 12:58:03.157776: step: 2000/1541, loss: 0.03491884469985962 2023-01-23 12:58:04.063086: step: 2002/1541, loss: 0.19158704578876495 2023-01-23 12:58:04.956803: step: 2004/1541, loss: 1.3368067741394043 2023-01-23 12:58:05.950774: step: 2006/1541, loss: 0.031520016491413116 2023-01-23 12:58:06.890772: step: 2008/1541, loss: 0.004709070548415184 2023-01-23 12:58:07.858417: step: 2010/1541, loss: 0.15639495849609375 2023-01-23 12:58:08.752775: step: 2012/1541, loss: 0.15241177380084991 2023-01-23 12:58:09.646846: step: 2014/1541, loss: 0.13893574476242065 2023-01-23 12:58:10.517692: step: 2016/1541, loss: 0.01058842334896326 2023-01-23 12:58:11.415832: step: 2018/1541, loss: 0.0759548619389534 2023-01-23 12:58:12.300833: step: 2020/1541, loss: 0.1410951018333435 2023-01-23 12:58:13.190844: step: 2022/1541, loss: 0.06297465413808823 2023-01-23 12:58:14.095307: step: 2024/1541, loss: 0.013777739368379116 2023-01-23 12:58:15.011447: step: 2026/1541, loss: 0.07562338560819626 2023-01-23 12:58:15.908102: step: 2028/1541, loss: 0.09286772459745407 2023-01-23 12:58:16.796974: step: 2030/1541, loss: 0.027087580412626266 2023-01-23 12:58:17.648614: step: 2032/1541, loss: 0.031497299671173096 2023-01-23 12:58:18.523070: step: 2034/1541, loss: 0.01578589715063572 2023-01-23 12:58:19.473319: step: 2036/1541, loss: 0.22163380682468414 2023-01-23 12:58:20.371546: step: 2038/1541, loss: 0.10477067530155182 2023-01-23 12:58:21.256482: step: 2040/1541, loss: 0.22463876008987427 2023-01-23 12:58:22.146017: step: 2042/1541, loss: 0.37639471888542175 2023-01-23 12:58:23.005200: step: 2044/1541, loss: 0.12370400130748749 2023-01-23 12:58:23.905877: step: 2046/1541, loss: 0.00561459781602025 2023-01-23 12:58:24.811163: step: 2048/1541, loss: 0.007368101272732019 2023-01-23 12:58:25.761234: step: 2050/1541, loss: 0.03937966004014015 2023-01-23 12:58:26.639085: step: 2052/1541, loss: 0.041852355003356934 2023-01-23 12:58:27.514097: step: 2054/1541, loss: 0.011543077416718006 2023-01-23 12:58:28.422149: step: 2056/1541, loss: 0.005639967508614063 2023-01-23 12:58:29.277927: step: 2058/1541, loss: 0.011549637652933598 2023-01-23 12:58:30.148991: step: 2060/1541, loss: 0.0016211819602176547 2023-01-23 12:58:31.029542: step: 2062/1541, loss: 0.0958634614944458 2023-01-23 12:58:31.944982: step: 2064/1541, loss: 0.0888603925704956 2023-01-23 12:58:32.852667: step: 2066/1541, loss: 0.009408091194927692 2023-01-23 12:58:33.761507: step: 2068/1541, loss: 0.03033219277858734 2023-01-23 12:58:34.663363: step: 2070/1541, loss: 0.03389215096831322 2023-01-23 12:58:35.543827: step: 2072/1541, loss: 0.006609183270484209 2023-01-23 12:58:36.458710: step: 2074/1541, loss: 1.5295629501342773 2023-01-23 12:58:37.327209: step: 2076/1541, loss: 0.018371785059571266 2023-01-23 12:58:38.240789: step: 2078/1541, loss: 0.05464239791035652 2023-01-23 12:58:39.098799: step: 2080/1541, loss: 0.31594139337539673 2023-01-23 12:58:39.930462: step: 2082/1541, loss: 0.32951197028160095 2023-01-23 12:58:40.833559: step: 2084/1541, loss: 0.1776280254125595 2023-01-23 12:58:41.740002: step: 2086/1541, loss: 0.2466161996126175 2023-01-23 12:58:42.624382: step: 2088/1541, loss: 0.15766693651676178 2023-01-23 12:58:43.546594: step: 2090/1541, loss: 0.014547256752848625 2023-01-23 12:58:44.440815: step: 2092/1541, loss: 0.1008329764008522 2023-01-23 12:58:45.352228: step: 2094/1541, loss: 0.032097168266773224 2023-01-23 12:58:46.284342: step: 2096/1541, loss: 0.09274788945913315 2023-01-23 12:58:47.180997: step: 2098/1541, loss: 0.10325542092323303 2023-01-23 12:58:48.128813: step: 2100/1541, loss: 0.4889413118362427 2023-01-23 12:58:49.049490: step: 2102/1541, loss: 0.10883127897977829 2023-01-23 12:58:49.942931: step: 2104/1541, loss: 0.014758926816284657 2023-01-23 12:58:50.874570: step: 2106/1541, loss: 0.3744484782218933 2023-01-23 12:58:51.837567: step: 2108/1541, loss: 0.019766299054026604 2023-01-23 12:58:52.741442: step: 2110/1541, loss: 0.07760705798864365 2023-01-23 12:58:53.615183: step: 2112/1541, loss: 0.0027974643744528294 2023-01-23 12:58:54.511650: step: 2114/1541, loss: 0.19833983480930328 2023-01-23 12:58:55.406382: step: 2116/1541, loss: 0.07919784635305405 2023-01-23 12:58:56.290910: step: 2118/1541, loss: 0.014926813542842865 2023-01-23 12:58:57.200287: step: 2120/1541, loss: 0.017078161239624023 2023-01-23 12:58:58.099616: step: 2122/1541, loss: 0.006696589291095734 2023-01-23 12:58:58.966958: step: 2124/1541, loss: 0.002559026936069131 2023-01-23 12:58:59.873780: step: 2126/1541, loss: 0.3442333936691284 2023-01-23 12:59:00.793993: step: 2128/1541, loss: 0.21494321525096893 2023-01-23 12:59:01.684759: step: 2130/1541, loss: 0.299698144197464 2023-01-23 12:59:02.627490: step: 2132/1541, loss: 0.23251873254776 2023-01-23 12:59:03.518795: step: 2134/1541, loss: 0.45481014251708984 2023-01-23 12:59:04.434615: step: 2136/1541, loss: 0.26523950695991516 2023-01-23 12:59:05.388533: step: 2138/1541, loss: 0.07508014142513275 2023-01-23 12:59:06.343616: step: 2140/1541, loss: 0.1399865448474884 2023-01-23 12:59:07.266111: step: 2142/1541, loss: 0.0033331753220409155 2023-01-23 12:59:08.129290: step: 2144/1541, loss: 0.16283956170082092 2023-01-23 12:59:09.045905: step: 2146/1541, loss: 0.00861344113945961 2023-01-23 12:59:09.958812: step: 2148/1541, loss: 0.006784089375287294 2023-01-23 12:59:10.962327: step: 2150/1541, loss: 0.1260848194360733 2023-01-23 12:59:11.897080: step: 2152/1541, loss: 0.22732898592948914 2023-01-23 12:59:12.802923: step: 2154/1541, loss: 0.10624582320451736 2023-01-23 12:59:13.731652: step: 2156/1541, loss: 0.11635585129261017 2023-01-23 12:59:14.635970: step: 2158/1541, loss: 0.07558129727840424 2023-01-23 12:59:15.502114: step: 2160/1541, loss: 0.013156963512301445 2023-01-23 12:59:16.396579: step: 2162/1541, loss: 0.0005865939310751855 2023-01-23 12:59:17.281490: step: 2164/1541, loss: 0.03327760100364685 2023-01-23 12:59:18.179564: step: 2166/1541, loss: 0.259010374546051 2023-01-23 12:59:19.034645: step: 2168/1541, loss: 0.11733857542276382 2023-01-23 12:59:19.944283: step: 2170/1541, loss: 0.02298678457736969 2023-01-23 12:59:20.843242: step: 2172/1541, loss: 0.0050622751004993916 2023-01-23 12:59:21.754213: step: 2174/1541, loss: 0.1355680525302887 2023-01-23 12:59:22.629970: step: 2176/1541, loss: 0.018788399174809456 2023-01-23 12:59:23.552539: step: 2178/1541, loss: 0.0040474990382790565 2023-01-23 12:59:24.456563: step: 2180/1541, loss: 0.02404801733791828 2023-01-23 12:59:25.352908: step: 2182/1541, loss: 0.07120765745639801 2023-01-23 12:59:26.227205: step: 2184/1541, loss: 0.06367888301610947 2023-01-23 12:59:27.083955: step: 2186/1541, loss: 0.16422516107559204 2023-01-23 12:59:27.938621: step: 2188/1541, loss: 0.22099339962005615 2023-01-23 12:59:28.806207: step: 2190/1541, loss: 0.06760194897651672 2023-01-23 12:59:29.760587: step: 2192/1541, loss: 1.1942309141159058 2023-01-23 12:59:30.654233: step: 2194/1541, loss: 0.1717127114534378 2023-01-23 12:59:31.545377: step: 2196/1541, loss: 0.406244695186615 2023-01-23 12:59:32.442585: step: 2198/1541, loss: 0.022480668500065804 2023-01-23 12:59:33.338264: step: 2200/1541, loss: 0.23079678416252136 2023-01-23 12:59:34.233771: step: 2202/1541, loss: 0.11434385180473328 2023-01-23 12:59:35.181470: step: 2204/1541, loss: 0.25196707248687744 2023-01-23 12:59:36.089884: step: 2206/1541, loss: 0.07635027170181274 2023-01-23 12:59:36.952920: step: 2208/1541, loss: 0.09277758002281189 2023-01-23 12:59:37.942126: step: 2210/1541, loss: 0.20282156765460968 2023-01-23 12:59:38.858779: step: 2212/1541, loss: 0.10353663563728333 2023-01-23 12:59:39.744713: step: 2214/1541, loss: 0.3707633912563324 2023-01-23 12:59:40.638363: step: 2216/1541, loss: 0.13798104226589203 2023-01-23 12:59:41.530478: step: 2218/1541, loss: 0.1538337916135788 2023-01-23 12:59:42.413658: step: 2220/1541, loss: 0.02787799760699272 2023-01-23 12:59:43.286334: step: 2222/1541, loss: 0.004077278543263674 2023-01-23 12:59:44.170905: step: 2224/1541, loss: 0.051554348319768906 2023-01-23 12:59:45.045423: step: 2226/1541, loss: 0.2806086242198944 2023-01-23 12:59:45.973695: step: 2228/1541, loss: 0.0616505891084671 2023-01-23 12:59:46.867763: step: 2230/1541, loss: 0.027316903695464134 2023-01-23 12:59:47.737887: step: 2232/1541, loss: 0.01982332020998001 2023-01-23 12:59:48.641563: step: 2234/1541, loss: 0.050297919660806656 2023-01-23 12:59:49.532764: step: 2236/1541, loss: 0.035232555121183395 2023-01-23 12:59:50.419924: step: 2238/1541, loss: 0.019439049065113068 2023-01-23 12:59:51.300248: step: 2240/1541, loss: 0.0782044529914856 2023-01-23 12:59:52.206679: step: 2242/1541, loss: 0.03433433175086975 2023-01-23 12:59:53.103168: step: 2244/1541, loss: 0.017080673947930336 2023-01-23 12:59:53.997361: step: 2246/1541, loss: 0.02821042202413082 2023-01-23 12:59:54.910546: step: 2248/1541, loss: 0.13120043277740479 2023-01-23 12:59:55.742838: step: 2250/1541, loss: 0.035993169993162155 2023-01-23 12:59:56.611469: step: 2252/1541, loss: 0.017674826085567474 2023-01-23 12:59:57.500078: step: 2254/1541, loss: 0.5013899207115173 2023-01-23 12:59:58.424184: step: 2256/1541, loss: 0.05091734230518341 2023-01-23 12:59:59.356218: step: 2258/1541, loss: 0.10638706386089325 2023-01-23 13:00:00.266150: step: 2260/1541, loss: 0.1608436405658722 2023-01-23 13:00:01.165735: step: 2262/1541, loss: 0.0292905755341053 2023-01-23 13:00:02.084999: step: 2264/1541, loss: 0.1417776644229889 2023-01-23 13:00:02.966503: step: 2266/1541, loss: 0.2806359529495239 2023-01-23 13:00:03.832361: step: 2268/1541, loss: 0.053172193467617035 2023-01-23 13:00:04.725088: step: 2270/1541, loss: 0.04239146038889885 2023-01-23 13:00:05.608387: step: 2272/1541, loss: 0.011904869228601456 2023-01-23 13:00:06.488838: step: 2274/1541, loss: 0.4766696095466614 2023-01-23 13:00:07.396376: step: 2276/1541, loss: 0.07077986001968384 2023-01-23 13:00:08.299293: step: 2278/1541, loss: 0.4210030734539032 2023-01-23 13:00:09.234343: step: 2280/1541, loss: 0.0011364948004484177 2023-01-23 13:00:10.209580: step: 2282/1541, loss: 0.015886420384049416 2023-01-23 13:00:11.123646: step: 2284/1541, loss: 0.0010013834107667208 2023-01-23 13:00:12.028345: step: 2286/1541, loss: 0.2587890625 2023-01-23 13:00:12.891124: step: 2288/1541, loss: 0.009875791147351265 2023-01-23 13:00:13.788052: step: 2290/1541, loss: 0.09660400450229645 2023-01-23 13:00:14.655952: step: 2292/1541, loss: 0.4924570322036743 2023-01-23 13:00:15.509281: step: 2294/1541, loss: 0.0860024094581604 2023-01-23 13:00:16.413384: step: 2296/1541, loss: 0.25387051701545715 2023-01-23 13:00:17.344386: step: 2298/1541, loss: 0.35645005106925964 2023-01-23 13:00:18.224423: step: 2300/1541, loss: 0.008159389719367027 2023-01-23 13:00:19.096621: step: 2302/1541, loss: 0.32917866110801697 2023-01-23 13:00:20.023255: step: 2304/1541, loss: 0.04358308017253876 2023-01-23 13:00:20.935746: step: 2306/1541, loss: 0.5742539167404175 2023-01-23 13:00:21.828963: step: 2308/1541, loss: 0.06532134115695953 2023-01-23 13:00:22.712472: step: 2310/1541, loss: 0.13394054770469666 2023-01-23 13:00:23.637317: step: 2312/1541, loss: 0.009318860247731209 2023-01-23 13:00:24.544890: step: 2314/1541, loss: 0.02385442703962326 2023-01-23 13:00:25.448075: step: 2316/1541, loss: 0.007928131148219109 2023-01-23 13:00:26.340400: step: 2318/1541, loss: 0.4998524785041809 2023-01-23 13:00:27.260137: step: 2320/1541, loss: 0.15546320378780365 2023-01-23 13:00:28.148293: step: 2322/1541, loss: 0.21497943997383118 2023-01-23 13:00:29.028921: step: 2324/1541, loss: 0.007865853607654572 2023-01-23 13:00:29.986258: step: 2326/1541, loss: 0.11024979501962662 2023-01-23 13:00:30.872558: step: 2328/1541, loss: 0.037068892270326614 2023-01-23 13:00:31.732678: step: 2330/1541, loss: 0.010136247612535954 2023-01-23 13:00:32.656286: step: 2332/1541, loss: 0.04824729263782501 2023-01-23 13:00:33.561317: step: 2334/1541, loss: 0.06357043981552124 2023-01-23 13:00:34.429947: step: 2336/1541, loss: 0.33712154626846313 2023-01-23 13:00:35.295089: step: 2338/1541, loss: 0.386414498090744 2023-01-23 13:00:36.213634: step: 2340/1541, loss: 0.0038938168436288834 2023-01-23 13:00:37.174029: step: 2342/1541, loss: 0.05627516657114029 2023-01-23 13:00:38.110485: step: 2344/1541, loss: 0.007038681302219629 2023-01-23 13:00:39.025644: step: 2346/1541, loss: 0.037852101027965546 2023-01-23 13:00:39.958861: step: 2348/1541, loss: 0.05219053477048874 2023-01-23 13:00:40.889023: step: 2350/1541, loss: 0.23393075168132782 2023-01-23 13:00:41.801574: step: 2352/1541, loss: 0.01695871353149414 2023-01-23 13:00:42.665066: step: 2354/1541, loss: 0.006097038742154837 2023-01-23 13:00:43.572249: step: 2356/1541, loss: 0.21576707065105438 2023-01-23 13:00:44.493649: step: 2358/1541, loss: 0.29910925030708313 2023-01-23 13:00:45.381266: step: 2360/1541, loss: 0.2250332236289978 2023-01-23 13:00:46.300833: step: 2362/1541, loss: 0.2922086715698242 2023-01-23 13:00:47.188489: step: 2364/1541, loss: 0.0896415263414383 2023-01-23 13:00:48.086955: step: 2366/1541, loss: 0.10223188996315002 2023-01-23 13:00:48.919456: step: 2368/1541, loss: 0.022870507091283798 2023-01-23 13:00:49.812957: step: 2370/1541, loss: 0.013162178918719292 2023-01-23 13:00:50.689081: step: 2372/1541, loss: 0.0035824128426611423 2023-01-23 13:00:51.610296: step: 2374/1541, loss: 0.0004172127810306847 2023-01-23 13:00:52.489931: step: 2376/1541, loss: 0.10964279621839523 2023-01-23 13:00:53.382449: step: 2378/1541, loss: 1.070623517036438 2023-01-23 13:00:54.348742: step: 2380/1541, loss: 0.020779011771082878 2023-01-23 13:00:55.224260: step: 2382/1541, loss: 0.005308646243065596 2023-01-23 13:00:56.128157: step: 2384/1541, loss: 0.11234055459499359 2023-01-23 13:00:57.013639: step: 2386/1541, loss: 0.025297801941633224 2023-01-23 13:00:57.933635: step: 2388/1541, loss: 0.01603803224861622 2023-01-23 13:00:58.815640: step: 2390/1541, loss: 0.023678891360759735 2023-01-23 13:00:59.679955: step: 2392/1541, loss: 0.12165612727403641 2023-01-23 13:01:00.608333: step: 2394/1541, loss: 0.02114851400256157 2023-01-23 13:01:01.529802: step: 2396/1541, loss: 0.16228432953357697 2023-01-23 13:01:02.421826: step: 2398/1541, loss: 0.006947637069970369 2023-01-23 13:01:03.322541: step: 2400/1541, loss: 0.045516226440668106 2023-01-23 13:01:04.244025: step: 2402/1541, loss: 0.1792355328798294 2023-01-23 13:01:05.231199: step: 2404/1541, loss: 0.00721691083163023 2023-01-23 13:01:06.134640: step: 2406/1541, loss: 0.08229441195726395 2023-01-23 13:01:06.975450: step: 2408/1541, loss: 0.17589035630226135 2023-01-23 13:01:07.936318: step: 2410/1541, loss: 0.6672582030296326 2023-01-23 13:01:08.839147: step: 2412/1541, loss: 0.022248106077313423 2023-01-23 13:01:09.717863: step: 2414/1541, loss: 0.025357196107506752 2023-01-23 13:01:10.632522: step: 2416/1541, loss: 0.36169517040252686 2023-01-23 13:01:11.535846: step: 2418/1541, loss: 0.002772754291072488 2023-01-23 13:01:12.447177: step: 2420/1541, loss: 0.04804131016135216 2023-01-23 13:01:13.307000: step: 2422/1541, loss: 0.015106532722711563 2023-01-23 13:01:14.193249: step: 2424/1541, loss: 0.2669373154640198 2023-01-23 13:01:15.070189: step: 2426/1541, loss: 0.3706118166446686 2023-01-23 13:01:15.956378: step: 2428/1541, loss: 0.12659333646297455 2023-01-23 13:01:16.837756: step: 2430/1541, loss: 0.10837392508983612 2023-01-23 13:01:17.731354: step: 2432/1541, loss: 0.0392378605902195 2023-01-23 13:01:18.609081: step: 2434/1541, loss: 0.08859145641326904 2023-01-23 13:01:19.543596: step: 2436/1541, loss: 0.0068802544847130775 2023-01-23 13:01:20.441019: step: 2438/1541, loss: 0.055082641541957855 2023-01-23 13:01:21.319119: step: 2440/1541, loss: 0.16209086775779724 2023-01-23 13:01:22.220725: step: 2442/1541, loss: 0.052702147513628006 2023-01-23 13:01:23.150264: step: 2444/1541, loss: 0.08017724007368088 2023-01-23 13:01:24.052646: step: 2446/1541, loss: 0.14658311009407043 2023-01-23 13:01:24.947971: step: 2448/1541, loss: 0.22468990087509155 2023-01-23 13:01:25.977147: step: 2450/1541, loss: 0.04545004293322563 2023-01-23 13:01:26.847996: step: 2452/1541, loss: 0.0237265657633543 2023-01-23 13:01:27.755742: step: 2454/1541, loss: 0.40971648693084717 2023-01-23 13:01:28.631675: step: 2456/1541, loss: 0.15450723469257355 2023-01-23 13:01:29.529619: step: 2458/1541, loss: 0.013188459910452366 2023-01-23 13:01:30.519272: step: 2460/1541, loss: 0.4484611749649048 2023-01-23 13:01:31.428788: step: 2462/1541, loss: 0.3011724650859833 2023-01-23 13:01:32.368983: step: 2464/1541, loss: 0.1282537430524826 2023-01-23 13:01:33.243269: step: 2466/1541, loss: 0.124588742852211 2023-01-23 13:01:34.141921: step: 2468/1541, loss: 0.13207367062568665 2023-01-23 13:01:35.047443: step: 2470/1541, loss: 0.032694362103939056 2023-01-23 13:01:36.038935: step: 2472/1541, loss: 0.13365565240383148 2023-01-23 13:01:36.918682: step: 2474/1541, loss: 0.20038481056690216 2023-01-23 13:01:37.785790: step: 2476/1541, loss: 0.21627019345760345 2023-01-23 13:01:38.662867: step: 2478/1541, loss: 0.1904972940683365 2023-01-23 13:01:39.568721: step: 2480/1541, loss: 0.22112303972244263 2023-01-23 13:01:40.497938: step: 2482/1541, loss: 0.015430991537868977 2023-01-23 13:01:41.463576: step: 2484/1541, loss: 0.051351696252822876 2023-01-23 13:01:42.353978: step: 2486/1541, loss: 0.07728742808103561 2023-01-23 13:01:43.236927: step: 2488/1541, loss: 0.19970199465751648 2023-01-23 13:01:44.090464: step: 2490/1541, loss: 0.004750348627567291 2023-01-23 13:01:44.990428: step: 2492/1541, loss: 0.46021270751953125 2023-01-23 13:01:45.912599: step: 2494/1541, loss: 0.012769588269293308 2023-01-23 13:01:46.787235: step: 2496/1541, loss: 0.06115797907114029 2023-01-23 13:01:47.713491: step: 2498/1541, loss: 0.010010472498834133 2023-01-23 13:01:48.623289: step: 2500/1541, loss: 0.041274044662714005 2023-01-23 13:01:49.605573: step: 2502/1541, loss: 0.09631749987602234 2023-01-23 13:01:50.517155: step: 2504/1541, loss: 0.04170272871851921 2023-01-23 13:01:51.411240: step: 2506/1541, loss: 0.03808686137199402 2023-01-23 13:01:52.280970: step: 2508/1541, loss: 0.26114094257354736 2023-01-23 13:01:53.198308: step: 2510/1541, loss: 0.045009929686784744 2023-01-23 13:01:54.088785: step: 2512/1541, loss: 0.1837494969367981 2023-01-23 13:01:55.022556: step: 2514/1541, loss: 0.14500361680984497 2023-01-23 13:01:55.878273: step: 2516/1541, loss: 0.01275578048080206 2023-01-23 13:01:56.773012: step: 2518/1541, loss: 0.015403013676404953 2023-01-23 13:01:57.718532: step: 2520/1541, loss: 0.15062013268470764 2023-01-23 13:01:58.614098: step: 2522/1541, loss: 0.1655309796333313 2023-01-23 13:01:59.510854: step: 2524/1541, loss: 0.5095075368881226 2023-01-23 13:02:00.372334: step: 2526/1541, loss: 0.05370286479592323 2023-01-23 13:02:01.261211: step: 2528/1541, loss: 0.010916263796389103 2023-01-23 13:02:02.226393: step: 2530/1541, loss: 0.14842292666435242 2023-01-23 13:02:03.109894: step: 2532/1541, loss: 0.48956748843193054 2023-01-23 13:02:04.041988: step: 2534/1541, loss: 0.0035380572080612183 2023-01-23 13:02:04.920767: step: 2536/1541, loss: 0.18711619079113007 2023-01-23 13:02:05.806815: step: 2538/1541, loss: 0.06169520691037178 2023-01-23 13:02:06.678466: step: 2540/1541, loss: 0.016950426623225212 2023-01-23 13:02:07.591335: step: 2542/1541, loss: 0.018202736973762512 2023-01-23 13:02:08.496920: step: 2544/1541, loss: 0.3118707835674286 2023-01-23 13:02:09.511474: step: 2546/1541, loss: 0.12324333190917969 2023-01-23 13:02:10.397801: step: 2548/1541, loss: 0.4739491045475006 2023-01-23 13:02:11.267390: step: 2550/1541, loss: 0.10338111221790314 2023-01-23 13:02:12.189377: step: 2552/1541, loss: 0.09832007437944412 2023-01-23 13:02:13.058491: step: 2554/1541, loss: 0.02455100230872631 2023-01-23 13:02:13.934077: step: 2556/1541, loss: 0.08985194563865662 2023-01-23 13:02:14.867574: step: 2558/1541, loss: 0.04606384038925171 2023-01-23 13:02:15.755101: step: 2560/1541, loss: 0.08112071454524994 2023-01-23 13:02:16.717159: step: 2562/1541, loss: 0.24094025790691376 2023-01-23 13:02:17.634506: step: 2564/1541, loss: 0.0028001521714031696 2023-01-23 13:02:18.562693: step: 2566/1541, loss: 0.030794261023402214 2023-01-23 13:02:19.437256: step: 2568/1541, loss: 0.35744383931159973 2023-01-23 13:02:20.315151: step: 2570/1541, loss: 0.09482239931821823 2023-01-23 13:02:21.242118: step: 2572/1541, loss: 0.2456659972667694 2023-01-23 13:02:22.252479: step: 2574/1541, loss: 0.3000737726688385 2023-01-23 13:02:23.094333: step: 2576/1541, loss: 0.010539567098021507 2023-01-23 13:02:23.988039: step: 2578/1541, loss: 0.8795064091682434 2023-01-23 13:02:24.883830: step: 2580/1541, loss: 0.09382301568984985 2023-01-23 13:02:25.791110: step: 2582/1541, loss: 0.13361695408821106 2023-01-23 13:02:26.721665: step: 2584/1541, loss: 0.11072303354740143 2023-01-23 13:02:27.588341: step: 2586/1541, loss: 0.012122494168579578 2023-01-23 13:02:28.508170: step: 2588/1541, loss: 0.32877346873283386 2023-01-23 13:02:29.449706: step: 2590/1541, loss: 0.00401960127055645 2023-01-23 13:02:30.364198: step: 2592/1541, loss: 0.03175202012062073 2023-01-23 13:02:31.318882: step: 2594/1541, loss: 0.140004962682724 2023-01-23 13:02:32.229931: step: 2596/1541, loss: 0.08881635963916779 2023-01-23 13:02:33.120311: step: 2598/1541, loss: 0.014672392047941685 2023-01-23 13:02:34.002589: step: 2600/1541, loss: 0.34381192922592163 2023-01-23 13:02:34.970969: step: 2602/1541, loss: 0.13098697364330292 2023-01-23 13:02:35.823992: step: 2604/1541, loss: 0.0788983702659607 2023-01-23 13:02:36.720642: step: 2606/1541, loss: 0.11131703108549118 2023-01-23 13:02:37.631784: step: 2608/1541, loss: 0.13383087515830994 2023-01-23 13:02:38.484108: step: 2610/1541, loss: 0.003805413842201233 2023-01-23 13:02:39.370918: step: 2612/1541, loss: 0.026667721569538116 2023-01-23 13:02:40.229359: step: 2614/1541, loss: 0.10578204691410065 2023-01-23 13:02:41.135137: step: 2616/1541, loss: 0.09512489289045334 2023-01-23 13:02:42.039674: step: 2618/1541, loss: 0.04446876794099808 2023-01-23 13:02:42.897553: step: 2620/1541, loss: 0.0699077770113945 2023-01-23 13:02:43.822387: step: 2622/1541, loss: 0.2863940894603729 2023-01-23 13:02:44.705574: step: 2624/1541, loss: 0.04123019427061081 2023-01-23 13:02:45.597213: step: 2626/1541, loss: 0.10946665704250336 2023-01-23 13:02:46.433532: step: 2628/1541, loss: 0.0023023702669888735 2023-01-23 13:02:47.283053: step: 2630/1541, loss: 0.052206482738256454 2023-01-23 13:02:48.236547: step: 2632/1541, loss: 0.008139020763337612 2023-01-23 13:02:49.110130: step: 2634/1541, loss: 0.1966855227947235 2023-01-23 13:02:50.012333: step: 2636/1541, loss: 0.02467426471412182 2023-01-23 13:02:50.903633: step: 2638/1541, loss: 0.06375492364168167 2023-01-23 13:02:51.809588: step: 2640/1541, loss: 0.06994792819023132 2023-01-23 13:02:52.714167: step: 2642/1541, loss: 0.027409294620156288 2023-01-23 13:02:53.645307: step: 2644/1541, loss: 0.05799352377653122 2023-01-23 13:02:54.557723: step: 2646/1541, loss: 0.014405917376279831 2023-01-23 13:02:55.440748: step: 2648/1541, loss: 0.3217315375804901 2023-01-23 13:02:56.363600: step: 2650/1541, loss: 0.03535277023911476 2023-01-23 13:02:57.234825: step: 2652/1541, loss: 0.02838827110826969 2023-01-23 13:02:58.260413: step: 2654/1541, loss: 0.04279239475727081 2023-01-23 13:02:59.171063: step: 2656/1541, loss: 0.03578317165374756 2023-01-23 13:03:00.062997: step: 2658/1541, loss: 0.019405163824558258 2023-01-23 13:03:00.993972: step: 2660/1541, loss: 0.5666196346282959 2023-01-23 13:03:01.907393: step: 2662/1541, loss: 0.27521419525146484 2023-01-23 13:03:02.762174: step: 2664/1541, loss: 0.12478148937225342 2023-01-23 13:03:03.698278: step: 2666/1541, loss: 0.045916587114334106 2023-01-23 13:03:04.531472: step: 2668/1541, loss: 0.21405446529388428 2023-01-23 13:03:05.417705: step: 2670/1541, loss: 2.5378143787384033 2023-01-23 13:03:06.273581: step: 2672/1541, loss: 0.057204313576221466 2023-01-23 13:03:07.157959: step: 2674/1541, loss: 0.32764771580696106 2023-01-23 13:03:08.117620: step: 2676/1541, loss: 0.1624036282300949 2023-01-23 13:03:09.039607: step: 2678/1541, loss: 0.09063319861888885 2023-01-23 13:03:09.920935: step: 2680/1541, loss: 0.2863345146179199 2023-01-23 13:03:10.842517: step: 2682/1541, loss: 0.3795023560523987 2023-01-23 13:03:11.744079: step: 2684/1541, loss: 0.01091268565505743 2023-01-23 13:03:12.648518: step: 2686/1541, loss: 0.08149717003107071 2023-01-23 13:03:13.629915: step: 2688/1541, loss: 0.0872349962592125 2023-01-23 13:03:14.598070: step: 2690/1541, loss: 0.21977342665195465 2023-01-23 13:03:15.503631: step: 2692/1541, loss: 0.005545108579099178 2023-01-23 13:03:16.371765: step: 2694/1541, loss: 0.18825772404670715 2023-01-23 13:03:17.281403: step: 2696/1541, loss: 0.036693863570690155 2023-01-23 13:03:18.190767: step: 2698/1541, loss: 0.15427720546722412 2023-01-23 13:03:19.085761: step: 2700/1541, loss: 0.6822912693023682 2023-01-23 13:03:19.982341: step: 2702/1541, loss: 0.07476639002561569 2023-01-23 13:03:20.863914: step: 2704/1541, loss: 0.08282135426998138 2023-01-23 13:03:21.789591: step: 2706/1541, loss: 0.007506342139095068 2023-01-23 13:03:22.680332: step: 2708/1541, loss: 0.00804726593196392 2023-01-23 13:03:23.597801: step: 2710/1541, loss: 0.021957065910100937 2023-01-23 13:03:24.461105: step: 2712/1541, loss: 0.005740923807024956 2023-01-23 13:03:25.410299: step: 2714/1541, loss: 0.05825607478618622 2023-01-23 13:03:26.289965: step: 2716/1541, loss: 0.18303073942661285 2023-01-23 13:03:27.189532: step: 2718/1541, loss: 0.019053490832448006 2023-01-23 13:03:28.139204: step: 2720/1541, loss: 0.005624241661280394 2023-01-23 13:03:29.048910: step: 2722/1541, loss: 0.18563251197338104 2023-01-23 13:03:29.947433: step: 2724/1541, loss: 0.0230804905295372 2023-01-23 13:03:30.845444: step: 2726/1541, loss: 0.12422823160886765 2023-01-23 13:03:31.719735: step: 2728/1541, loss: 0.34424617886543274 2023-01-23 13:03:32.614718: step: 2730/1541, loss: 0.0037701986730098724 2023-01-23 13:03:33.518307: step: 2732/1541, loss: 0.006015249993652105 2023-01-23 13:03:34.410015: step: 2734/1541, loss: 0.09446972608566284 2023-01-23 13:03:35.300072: step: 2736/1541, loss: 0.11826588213443756 2023-01-23 13:03:36.191365: step: 2738/1541, loss: 0.02782285213470459 2023-01-23 13:03:37.119844: step: 2740/1541, loss: 0.009570125490427017 2023-01-23 13:03:37.981661: step: 2742/1541, loss: 0.008672123774886131 2023-01-23 13:03:38.897249: step: 2744/1541, loss: 0.007097133435308933 2023-01-23 13:03:39.756664: step: 2746/1541, loss: 0.20415817201137543 2023-01-23 13:03:40.651760: step: 2748/1541, loss: 0.08684301376342773 2023-01-23 13:03:41.541021: step: 2750/1541, loss: 0.10111495852470398 2023-01-23 13:03:42.422149: step: 2752/1541, loss: 0.14414535462856293 2023-01-23 13:03:43.344383: step: 2754/1541, loss: 0.32935044169425964 2023-01-23 13:03:44.213705: step: 2756/1541, loss: 0.0075659966096282005 2023-01-23 13:03:45.119102: step: 2758/1541, loss: 0.011474961414933205 2023-01-23 13:03:46.021868: step: 2760/1541, loss: 0.029108509421348572 2023-01-23 13:03:46.890953: step: 2762/1541, loss: 0.1284327656030655 2023-01-23 13:03:47.803936: step: 2764/1541, loss: 0.018882600590586662 2023-01-23 13:03:48.733224: step: 2766/1541, loss: 0.00958002544939518 2023-01-23 13:03:49.641445: step: 2768/1541, loss: 0.0269071813672781 2023-01-23 13:03:50.578309: step: 2770/1541, loss: 0.05403922498226166 2023-01-23 13:03:51.490368: step: 2772/1541, loss: 0.005758846178650856 2023-01-23 13:03:52.396256: step: 2774/1541, loss: 0.036489732563495636 2023-01-23 13:03:53.313757: step: 2776/1541, loss: 0.08749420195817947 2023-01-23 13:03:54.188801: step: 2778/1541, loss: 0.004746339749544859 2023-01-23 13:03:55.196723: step: 2780/1541, loss: 0.10621418058872223 2023-01-23 13:03:56.099070: step: 2782/1541, loss: 0.8097884058952332 2023-01-23 13:03:57.024266: step: 2784/1541, loss: 0.15229026973247528 2023-01-23 13:03:57.903285: step: 2786/1541, loss: 0.060808561742305756 2023-01-23 13:03:58.785126: step: 2788/1541, loss: 0.00790266040712595 2023-01-23 13:03:59.686974: step: 2790/1541, loss: 0.008491511456668377 2023-01-23 13:04:00.654012: step: 2792/1541, loss: 0.22626464068889618 2023-01-23 13:04:01.574089: step: 2794/1541, loss: 0.054696932435035706 2023-01-23 13:04:02.550445: step: 2796/1541, loss: 0.010769408196210861 2023-01-23 13:04:03.411702: step: 2798/1541, loss: 0.31920021772384644 2023-01-23 13:04:04.337129: step: 2800/1541, loss: 0.21865001320838928 2023-01-23 13:04:05.211124: step: 2802/1541, loss: 0.02073201723396778 2023-01-23 13:04:06.077924: step: 2804/1541, loss: 0.039081767201423645 2023-01-23 13:04:06.945295: step: 2806/1541, loss: 0.019388481974601746 2023-01-23 13:04:07.910971: step: 2808/1541, loss: 0.02710493840277195 2023-01-23 13:04:08.855283: step: 2810/1541, loss: 0.46050649881362915 2023-01-23 13:04:09.776617: step: 2812/1541, loss: 0.005249560344964266 2023-01-23 13:04:10.734843: step: 2814/1541, loss: 0.017099246382713318 2023-01-23 13:04:11.629501: step: 2816/1541, loss: 0.013505993410944939 2023-01-23 13:04:12.538611: step: 2818/1541, loss: 0.013081508688628674 2023-01-23 13:04:13.429147: step: 2820/1541, loss: 0.02692202664911747 2023-01-23 13:04:14.302316: step: 2822/1541, loss: 0.003762774169445038 2023-01-23 13:04:15.209783: step: 2824/1541, loss: 0.022130070254206657 2023-01-23 13:04:16.112227: step: 2826/1541, loss: 0.03380878269672394 2023-01-23 13:04:17.107582: step: 2828/1541, loss: 0.1195497214794159 2023-01-23 13:04:18.006432: step: 2830/1541, loss: 0.1773412972688675 2023-01-23 13:04:18.877605: step: 2832/1541, loss: 0.3437703847885132 2023-01-23 13:04:19.779610: step: 2834/1541, loss: 0.008960937149822712 2023-01-23 13:04:20.634010: step: 2836/1541, loss: 0.5260209441184998 2023-01-23 13:04:21.509702: step: 2838/1541, loss: 0.14782968163490295 2023-01-23 13:04:22.475444: step: 2840/1541, loss: 0.017315983772277832 2023-01-23 13:04:23.361237: step: 2842/1541, loss: 0.0620012991130352 2023-01-23 13:04:24.198472: step: 2844/1541, loss: 0.04787329584360123 2023-01-23 13:04:25.114057: step: 2846/1541, loss: 0.09181936085224152 2023-01-23 13:04:26.060568: step: 2848/1541, loss: 0.03579877316951752 2023-01-23 13:04:26.930595: step: 2850/1541, loss: 1.2013111114501953 2023-01-23 13:04:27.808245: step: 2852/1541, loss: 0.03414672240614891 2023-01-23 13:04:28.680890: step: 2854/1541, loss: 0.07286129146814346 2023-01-23 13:04:29.606109: step: 2856/1541, loss: 0.23068633675575256 2023-01-23 13:04:30.523824: step: 2858/1541, loss: 0.05845417082309723 2023-01-23 13:04:31.411930: step: 2860/1541, loss: 0.01691143587231636 2023-01-23 13:04:32.340756: step: 2862/1541, loss: 0.1769234538078308 2023-01-23 13:04:33.296122: step: 2864/1541, loss: 0.15332461893558502 2023-01-23 13:04:34.181885: step: 2866/1541, loss: 0.13714352250099182 2023-01-23 13:04:35.069185: step: 2868/1541, loss: 0.2726218104362488 2023-01-23 13:04:36.013356: step: 2870/1541, loss: 0.06773468106985092 2023-01-23 13:04:36.933396: step: 2872/1541, loss: 0.027487117797136307 2023-01-23 13:04:37.859856: step: 2874/1541, loss: 0.015290157869458199 2023-01-23 13:04:38.746628: step: 2876/1541, loss: 0.16714037954807281 2023-01-23 13:04:39.664011: step: 2878/1541, loss: 0.3069893717765808 2023-01-23 13:04:40.571017: step: 2880/1541, loss: 0.6264067888259888 2023-01-23 13:04:41.423234: step: 2882/1541, loss: 0.20621907711029053 2023-01-23 13:04:42.298754: step: 2884/1541, loss: 0.062196262180805206 2023-01-23 13:04:43.194508: step: 2886/1541, loss: 0.03657510504126549 2023-01-23 13:04:44.083994: step: 2888/1541, loss: 0.03096708469092846 2023-01-23 13:04:45.010057: step: 2890/1541, loss: 0.1328442543745041 2023-01-23 13:04:45.928566: step: 2892/1541, loss: 0.2432091385126114 2023-01-23 13:04:46.855502: step: 2894/1541, loss: 0.0019497910980135202 2023-01-23 13:04:47.720182: step: 2896/1541, loss: 0.0021350374445319176 2023-01-23 13:04:48.625010: step: 2898/1541, loss: 0.047009192407131195 2023-01-23 13:04:49.510921: step: 2900/1541, loss: 0.04003332555294037 2023-01-23 13:04:50.430404: step: 2902/1541, loss: 0.23281611502170563 2023-01-23 13:04:51.336342: step: 2904/1541, loss: 0.061531685292720795 2023-01-23 13:04:52.244183: step: 2906/1541, loss: 0.16971372067928314 2023-01-23 13:04:53.189010: step: 2908/1541, loss: 0.011602441780269146 2023-01-23 13:04:54.210467: step: 2910/1541, loss: 0.14021067321300507 2023-01-23 13:04:55.130968: step: 2912/1541, loss: 0.40942680835723877 2023-01-23 13:04:55.998270: step: 2914/1541, loss: 0.2862105667591095 2023-01-23 13:04:56.879497: step: 2916/1541, loss: 0.012123200111091137 2023-01-23 13:04:57.788397: step: 2918/1541, loss: 0.1079990565776825 2023-01-23 13:04:58.672722: step: 2920/1541, loss: 0.0034606903791427612 2023-01-23 13:04:59.545101: step: 2922/1541, loss: 0.04021282121539116 2023-01-23 13:05:00.442528: step: 2924/1541, loss: 0.009952399879693985 2023-01-23 13:05:01.360754: step: 2926/1541, loss: 0.19188182055950165 2023-01-23 13:05:02.276105: step: 2928/1541, loss: 0.050486475229263306 2023-01-23 13:05:03.243510: step: 2930/1541, loss: 0.3049682676792145 2023-01-23 13:05:04.132609: step: 2932/1541, loss: 0.020739268511533737 2023-01-23 13:05:05.070388: step: 2934/1541, loss: 0.03642189875245094 2023-01-23 13:05:06.030693: step: 2936/1541, loss: 0.06756045669317245 2023-01-23 13:05:06.960078: step: 2938/1541, loss: 0.036249805241823196 2023-01-23 13:05:07.949680: step: 2940/1541, loss: 0.008029496297240257 2023-01-23 13:05:08.907668: step: 2942/1541, loss: 0.05850042402744293 2023-01-23 13:05:09.823535: step: 2944/1541, loss: 0.007878270000219345 2023-01-23 13:05:10.724307: step: 2946/1541, loss: 0.5172780156135559 2023-01-23 13:05:11.741788: step: 2948/1541, loss: 0.1998274028301239 2023-01-23 13:05:12.643413: step: 2950/1541, loss: 0.12197411060333252 2023-01-23 13:05:13.639507: step: 2952/1541, loss: 0.12074074149131775 2023-01-23 13:05:14.566919: step: 2954/1541, loss: 0.0030269892886281013 2023-01-23 13:05:15.464567: step: 2956/1541, loss: 0.01329775806516409 2023-01-23 13:05:16.333279: step: 2958/1541, loss: 0.020714538171887398 2023-01-23 13:05:17.204185: step: 2960/1541, loss: 0.07378868758678436 2023-01-23 13:05:18.075374: step: 2962/1541, loss: 0.02434561215341091 2023-01-23 13:05:18.957474: step: 2964/1541, loss: 0.17709048092365265 2023-01-23 13:05:19.879497: step: 2966/1541, loss: 0.13124606013298035 2023-01-23 13:05:20.736620: step: 2968/1541, loss: 0.17251348495483398 2023-01-23 13:05:21.649023: step: 2970/1541, loss: 0.4738338887691498 2023-01-23 13:05:22.576166: step: 2972/1541, loss: 0.16258201003074646 2023-01-23 13:05:23.445416: step: 2974/1541, loss: 0.09811357408761978 2023-01-23 13:05:24.324375: step: 2976/1541, loss: 0.1166197657585144 2023-01-23 13:05:25.215184: step: 2978/1541, loss: 0.07055705785751343 2023-01-23 13:05:26.108045: step: 2980/1541, loss: 0.03618785738945007 2023-01-23 13:05:27.021136: step: 2982/1541, loss: 0.18849214911460876 2023-01-23 13:05:28.040476: step: 2984/1541, loss: 0.022463690489530563 2023-01-23 13:05:28.954917: step: 2986/1541, loss: 0.10348910838365555 2023-01-23 13:05:29.833521: step: 2988/1541, loss: 0.03355032578110695 2023-01-23 13:05:30.725766: step: 2990/1541, loss: 0.0812939926981926 2023-01-23 13:05:31.636997: step: 2992/1541, loss: 0.003236036980524659 2023-01-23 13:05:32.578012: step: 2994/1541, loss: 0.026026975363492966 2023-01-23 13:05:33.518247: step: 2996/1541, loss: 0.02743743360042572 2023-01-23 13:05:34.398379: step: 2998/1541, loss: 0.012691427022218704 2023-01-23 13:05:35.302165: step: 3000/1541, loss: 0.0216755922883749 2023-01-23 13:05:36.200963: step: 3002/1541, loss: 0.33292174339294434 2023-01-23 13:05:37.126029: step: 3004/1541, loss: 1.7537288665771484 2023-01-23 13:05:38.041832: step: 3006/1541, loss: 0.12281259894371033 2023-01-23 13:05:38.921836: step: 3008/1541, loss: 0.015722284093499184 2023-01-23 13:05:39.792758: step: 3010/1541, loss: 0.09532134979963303 2023-01-23 13:05:40.659883: step: 3012/1541, loss: 1.8838707208633423 2023-01-23 13:05:41.545466: step: 3014/1541, loss: 0.002446899889037013 2023-01-23 13:05:42.464211: step: 3016/1541, loss: 0.020603209733963013 2023-01-23 13:05:43.343000: step: 3018/1541, loss: 0.03295421600341797 2023-01-23 13:05:44.291546: step: 3020/1541, loss: 0.11306258291006088 2023-01-23 13:05:45.227060: step: 3022/1541, loss: 0.01147913746535778 2023-01-23 13:05:46.096613: step: 3024/1541, loss: 0.19876310229301453 2023-01-23 13:05:47.021361: step: 3026/1541, loss: 0.07613828033208847 2023-01-23 13:05:47.886395: step: 3028/1541, loss: 0.07360949367284775 2023-01-23 13:05:48.758100: step: 3030/1541, loss: 0.04515676200389862 2023-01-23 13:05:49.624795: step: 3032/1541, loss: 0.0419900007545948 2023-01-23 13:05:50.479646: step: 3034/1541, loss: 0.016191042959690094 2023-01-23 13:05:51.325220: step: 3036/1541, loss: 0.07982748746871948 2023-01-23 13:05:52.184502: step: 3038/1541, loss: 0.01776954159140587 2023-01-23 13:05:53.034272: step: 3040/1541, loss: 0.03498777747154236 2023-01-23 13:05:53.938229: step: 3042/1541, loss: 0.00728599401190877 2023-01-23 13:05:54.877463: step: 3044/1541, loss: 0.3123028874397278 2023-01-23 13:05:55.809170: step: 3046/1541, loss: 0.055575817823410034 2023-01-23 13:05:56.745225: step: 3048/1541, loss: 0.3590783476829529 2023-01-23 13:05:57.755815: step: 3050/1541, loss: 0.03500468283891678 2023-01-23 13:05:58.631778: step: 3052/1541, loss: 0.11951565742492676 2023-01-23 13:05:59.523167: step: 3054/1541, loss: 0.11825946718454361 2023-01-23 13:06:00.409589: step: 3056/1541, loss: 0.39476481080055237 2023-01-23 13:06:01.292023: step: 3058/1541, loss: 0.028844134882092476 2023-01-23 13:06:02.234612: step: 3060/1541, loss: 0.006992535199970007 2023-01-23 13:06:03.150479: step: 3062/1541, loss: 0.15554681420326233 2023-01-23 13:06:04.074294: step: 3064/1541, loss: 0.15827377140522003 2023-01-23 13:06:04.936961: step: 3066/1541, loss: 0.017387595027685165 2023-01-23 13:06:05.809062: step: 3068/1541, loss: 0.13585563004016876 2023-01-23 13:06:06.713965: step: 3070/1541, loss: 0.012244484387338161 2023-01-23 13:06:07.563942: step: 3072/1541, loss: 0.13750220835208893 2023-01-23 13:06:08.471317: step: 3074/1541, loss: 0.029581161215901375 2023-01-23 13:06:09.354387: step: 3076/1541, loss: 0.2288571149110794 2023-01-23 13:06:10.264362: step: 3078/1541, loss: 0.14284059405326843 2023-01-23 13:06:11.140802: step: 3080/1541, loss: 0.09503206610679626 2023-01-23 13:06:12.046523: step: 3082/1541, loss: 0.015126914717257023 2023-01-23 13:06:12.653789: step: 3084/1541, loss: 0.07346644252538681 ================================================== Loss: 0.132 -------------------- Dev Chinese: {'event': {'p': 0.6236559139784946, 'r': 0.6950732356857523, 'f1': 0.6574307304785894}, 'argument': {'p': 0.6669470437176143, 'r': 0.6529660109688754, 'f1': 0.6598824810732926}, 'combined': 0.4338270215620387, 'epoch': 16} Test Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.721787349796131, 'r': 0.6978330650290092, 'f1': 0.7096081084033167}, 'combined': 0.46260646215707685, 'epoch': 16} Dev Korean: {'event': {'p': 0.5676470588235294, 'r': 0.7709720372836218, 'f1': 0.6538678712591757}, 'argument': {'p': 0.6663945785118833, 'r': 0.6601755336298629, 'f1': 0.6632704784323108}, 'combined': 0.43369125580159007, 'epoch': 16} Test Korean: {'event': {'p': 0.611534795042898, 'r': 0.768723786698622, 'f1': 0.6811786567560393}, 'argument': {'p': 0.7250903249811195, 'r': 0.712101351825504, 'f1': 0.7185371428840014}, 'combined': 0.4894521658190463, 'epoch': 16} Dev Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Chinese: {'event': {'p': 0.6349206349206349, 'r': 0.7407407407407407, 'f1': 0.6837606837606837}, 'argument': {'p': 0.6703869047619048, 'r': 0.6644542772861356, 'f1': 0.6674074074074073}, 'combined': 0.456346945235834, 'epoch': 16} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5656436817151103, 'r': 0.35995507018234285, 'f1': 0.43994508577841906}, 'combined': 0.2563459554929371, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} New best russian model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 17 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 13:11:43.255450: step: 2/1541, loss: 0.042268943041563034 2023-01-23 13:11:44.125322: step: 4/1541, loss: 0.030808961018919945 2023-01-23 13:11:45.069558: step: 6/1541, loss: 0.011460784822702408 2023-01-23 13:11:45.979233: step: 8/1541, loss: 0.003678889013826847 2023-01-23 13:11:46.898800: step: 10/1541, loss: 0.007852029986679554 2023-01-23 13:11:47.841457: step: 12/1541, loss: 0.1625223159790039 2023-01-23 13:11:48.758470: step: 14/1541, loss: 0.10108053684234619 2023-01-23 13:11:49.684404: step: 16/1541, loss: 0.0525890477001667 2023-01-23 13:11:50.621139: step: 18/1541, loss: 0.0008247439400292933 2023-01-23 13:11:51.520279: step: 20/1541, loss: 0.0001810133981052786 2023-01-23 13:11:52.381303: step: 22/1541, loss: 0.07197894901037216 2023-01-23 13:11:53.233797: step: 24/1541, loss: 0.0026010547298938036 2023-01-23 13:11:54.096063: step: 26/1541, loss: 0.002608254784718156 2023-01-23 13:11:54.981060: step: 28/1541, loss: 0.1366163194179535 2023-01-23 13:11:55.938528: step: 30/1541, loss: 0.04756665974855423 2023-01-23 13:11:56.864508: step: 32/1541, loss: 0.011029500514268875 2023-01-23 13:11:57.781475: step: 34/1541, loss: 0.1657838374376297 2023-01-23 13:11:58.708157: step: 36/1541, loss: 0.15857253968715668 2023-01-23 13:11:59.639865: step: 38/1541, loss: 0.009784740395843983 2023-01-23 13:12:00.533994: step: 40/1541, loss: 0.026202065870165825 2023-01-23 13:12:01.420876: step: 42/1541, loss: 0.02773352339863777 2023-01-23 13:12:02.350820: step: 44/1541, loss: 0.0011478682281449437 2023-01-23 13:12:03.223606: step: 46/1541, loss: 0.019256550818681717 2023-01-23 13:12:04.098863: step: 48/1541, loss: 0.17065244913101196 2023-01-23 13:12:05.005419: step: 50/1541, loss: 0.015415569767355919 2023-01-23 13:12:05.891126: step: 52/1541, loss: 0.022760514169931412 2023-01-23 13:12:06.777481: step: 54/1541, loss: 0.016345860436558723 2023-01-23 13:12:07.725910: step: 56/1541, loss: 0.023766037076711655 2023-01-23 13:12:08.698136: step: 58/1541, loss: 0.02957374043762684 2023-01-23 13:12:09.560003: step: 60/1541, loss: 0.05567308887839317 2023-01-23 13:12:10.428946: step: 62/1541, loss: 0.0008154523675329983 2023-01-23 13:12:11.305170: step: 64/1541, loss: 0.9258949160575867 2023-01-23 13:12:12.194780: step: 66/1541, loss: 0.04630435258150101 2023-01-23 13:12:13.054547: step: 68/1541, loss: 0.21959596872329712 2023-01-23 13:12:13.964869: step: 70/1541, loss: 0.0168949943035841 2023-01-23 13:12:14.839543: step: 72/1541, loss: 0.05669567734003067 2023-01-23 13:12:15.749290: step: 74/1541, loss: 0.1999906599521637 2023-01-23 13:12:16.647544: step: 76/1541, loss: 0.012969614937901497 2023-01-23 13:12:17.559833: step: 78/1541, loss: 0.03458376228809357 2023-01-23 13:12:18.478715: step: 80/1541, loss: 0.07563360035419464 2023-01-23 13:12:19.401258: step: 82/1541, loss: 0.0030612207483500242 2023-01-23 13:12:20.294835: step: 84/1541, loss: 0.33734241127967834 2023-01-23 13:12:21.155570: step: 86/1541, loss: 0.003572280751541257 2023-01-23 13:12:22.032475: step: 88/1541, loss: 0.005578272044658661 2023-01-23 13:12:22.912094: step: 90/1541, loss: 0.023630628362298012 2023-01-23 13:12:23.887749: step: 92/1541, loss: 0.00314378272742033 2023-01-23 13:12:24.796279: step: 94/1541, loss: 0.013499312102794647 2023-01-23 13:12:25.759959: step: 96/1541, loss: 0.02096383459866047 2023-01-23 13:12:26.638173: step: 98/1541, loss: 0.0033399323001503944 2023-01-23 13:12:27.510156: step: 100/1541, loss: 0.06927762180566788 2023-01-23 13:12:28.399213: step: 102/1541, loss: 0.25334465503692627 2023-01-23 13:12:29.278146: step: 104/1541, loss: 0.06652157753705978 2023-01-23 13:12:30.233431: step: 106/1541, loss: 0.05637587606906891 2023-01-23 13:12:31.121548: step: 108/1541, loss: 0.0030704897362738848 2023-01-23 13:12:32.038201: step: 110/1541, loss: 2.112053871154785 2023-01-23 13:12:32.952138: step: 112/1541, loss: 0.47320494055747986 2023-01-23 13:12:33.844713: step: 114/1541, loss: 0.007173737045377493 2023-01-23 13:12:34.709834: step: 116/1541, loss: 0.007974528707563877 2023-01-23 13:12:35.590611: step: 118/1541, loss: 0.0018116928404197097 2023-01-23 13:12:36.518677: step: 120/1541, loss: 0.23702621459960938 2023-01-23 13:12:37.365080: step: 122/1541, loss: 0.12903669476509094 2023-01-23 13:12:38.239638: step: 124/1541, loss: 0.11931020766496658 2023-01-23 13:12:39.127636: step: 126/1541, loss: 0.06060180813074112 2023-01-23 13:12:40.027639: step: 128/1541, loss: 0.09945014119148254 2023-01-23 13:12:40.908343: step: 130/1541, loss: 0.2033626288175583 2023-01-23 13:12:41.838920: step: 132/1541, loss: 1.072324514389038 2023-01-23 13:12:42.765026: step: 134/1541, loss: 0.07817517220973969 2023-01-23 13:12:43.678112: step: 136/1541, loss: 0.7306015491485596 2023-01-23 13:12:44.587853: step: 138/1541, loss: 0.26177239418029785 2023-01-23 13:12:45.487667: step: 140/1541, loss: 0.3256319761276245 2023-01-23 13:12:46.383205: step: 142/1541, loss: 0.12779538333415985 2023-01-23 13:12:47.260325: step: 144/1541, loss: 0.2594977021217346 2023-01-23 13:12:48.128211: step: 146/1541, loss: 0.0011339926859363914 2023-01-23 13:12:49.034497: step: 148/1541, loss: 0.4269261062145233 2023-01-23 13:12:49.948690: step: 150/1541, loss: 0.02986721321940422 2023-01-23 13:12:50.918745: step: 152/1541, loss: 0.0010480547789484262 2023-01-23 13:12:51.818083: step: 154/1541, loss: 0.01587754301726818 2023-01-23 13:12:52.679849: step: 156/1541, loss: 0.00832216627895832 2023-01-23 13:12:53.572198: step: 158/1541, loss: 0.023384539410471916 2023-01-23 13:12:54.456049: step: 160/1541, loss: 0.033046744763851166 2023-01-23 13:12:55.403057: step: 162/1541, loss: 0.18231667578220367 2023-01-23 13:12:56.317287: step: 164/1541, loss: 0.008537225425243378 2023-01-23 13:12:57.243654: step: 166/1541, loss: 0.03885462135076523 2023-01-23 13:12:58.148111: step: 168/1541, loss: 0.04367733374238014 2023-01-23 13:12:59.027699: step: 170/1541, loss: 0.01726510562002659 2023-01-23 13:12:59.898253: step: 172/1541, loss: 0.003043003845959902 2023-01-23 13:13:00.778266: step: 174/1541, loss: 0.0656018853187561 2023-01-23 13:13:01.717506: step: 176/1541, loss: 0.05245737358927727 2023-01-23 13:13:02.592782: step: 178/1541, loss: 0.0032163106370717287 2023-01-23 13:13:03.494340: step: 180/1541, loss: 0.02236449159681797 2023-01-23 13:13:04.377439: step: 182/1541, loss: 0.19021323323249817 2023-01-23 13:13:05.257693: step: 184/1541, loss: 0.04950343444943428 2023-01-23 13:13:06.108519: step: 186/1541, loss: 0.03623499721288681 2023-01-23 13:13:07.039147: step: 188/1541, loss: 0.13138410449028015 2023-01-23 13:13:07.968883: step: 190/1541, loss: 0.10206883400678635 2023-01-23 13:13:08.869965: step: 192/1541, loss: 0.011254720389842987 2023-01-23 13:13:09.740340: step: 194/1541, loss: 0.002491669962182641 2023-01-23 13:13:10.633354: step: 196/1541, loss: 0.007739105727523565 2023-01-23 13:13:11.526953: step: 198/1541, loss: 0.009647518396377563 2023-01-23 13:13:12.428172: step: 200/1541, loss: 0.0017581202555447817 2023-01-23 13:13:13.319398: step: 202/1541, loss: 0.012036487460136414 2023-01-23 13:13:14.223769: step: 204/1541, loss: 0.009727994911372662 2023-01-23 13:13:15.099980: step: 206/1541, loss: 0.002550698583945632 2023-01-23 13:13:16.020391: step: 208/1541, loss: 0.1199609637260437 2023-01-23 13:13:16.919233: step: 210/1541, loss: 0.03212680295109749 2023-01-23 13:13:17.767840: step: 212/1541, loss: 0.05277290567755699 2023-01-23 13:13:18.634510: step: 214/1541, loss: 0.12302184104919434 2023-01-23 13:13:19.505543: step: 216/1541, loss: 0.00790120754390955 2023-01-23 13:13:20.362125: step: 218/1541, loss: 0.03859895095229149 2023-01-23 13:13:21.266032: step: 220/1541, loss: 0.011373250745236874 2023-01-23 13:13:22.123986: step: 222/1541, loss: 0.08637259155511856 2023-01-23 13:13:23.019891: step: 224/1541, loss: 0.11179209500551224 2023-01-23 13:13:23.960924: step: 226/1541, loss: 0.13519422709941864 2023-01-23 13:13:24.821440: step: 228/1541, loss: 0.02513863332569599 2023-01-23 13:13:25.696033: step: 230/1541, loss: 0.00677085667848587 2023-01-23 13:13:26.549932: step: 232/1541, loss: 0.011958037503063679 2023-01-23 13:13:27.467110: step: 234/1541, loss: 0.0013630142202600837 2023-01-23 13:13:28.328401: step: 236/1541, loss: 0.0024995652493089437 2023-01-23 13:13:29.222651: step: 238/1541, loss: 0.0034968676045536995 2023-01-23 13:13:30.067891: step: 240/1541, loss: 0.030055580660700798 2023-01-23 13:13:30.922161: step: 242/1541, loss: 0.10143277049064636 2023-01-23 13:13:31.866573: step: 244/1541, loss: 0.0612819641828537 2023-01-23 13:13:32.734168: step: 246/1541, loss: 0.010765839368104935 2023-01-23 13:13:33.592763: step: 248/1541, loss: 0.017888225615024567 2023-01-23 13:13:34.502412: step: 250/1541, loss: 0.007672085892409086 2023-01-23 13:13:35.455537: step: 252/1541, loss: 0.0018291042651981115 2023-01-23 13:13:36.372715: step: 254/1541, loss: 0.1273723989725113 2023-01-23 13:13:37.285301: step: 256/1541, loss: 0.0703803077340126 2023-01-23 13:13:38.178104: step: 258/1541, loss: 0.06955435127019882 2023-01-23 13:13:39.071169: step: 260/1541, loss: 0.01364007405936718 2023-01-23 13:13:40.035058: step: 262/1541, loss: 0.12496859580278397 2023-01-23 13:13:40.885358: step: 264/1541, loss: 0.12101192027330399 2023-01-23 13:13:41.771208: step: 266/1541, loss: 0.03316393122076988 2023-01-23 13:13:42.704075: step: 268/1541, loss: 0.04979296773672104 2023-01-23 13:13:43.625343: step: 270/1541, loss: 0.1426677405834198 2023-01-23 13:13:44.525191: step: 272/1541, loss: 0.02098092809319496 2023-01-23 13:13:45.414068: step: 274/1541, loss: 0.3299717605113983 2023-01-23 13:13:46.351325: step: 276/1541, loss: 0.06523246318101883 2023-01-23 13:13:47.280175: step: 278/1541, loss: 0.006921971682459116 2023-01-23 13:13:48.166028: step: 280/1541, loss: 0.39168187975883484 2023-01-23 13:13:49.067371: step: 282/1541, loss: 0.0008211208041757345 2023-01-23 13:13:49.997469: step: 284/1541, loss: 0.033104728907346725 2023-01-23 13:13:50.892552: step: 286/1541, loss: 0.032079633325338364 2023-01-23 13:13:51.815806: step: 288/1541, loss: 0.6239430904388428 2023-01-23 13:13:52.683577: step: 290/1541, loss: 0.1182745099067688 2023-01-23 13:13:53.551859: step: 292/1541, loss: 0.022312458604574203 2023-01-23 13:13:54.470272: step: 294/1541, loss: 0.1283448189496994 2023-01-23 13:13:55.344807: step: 296/1541, loss: 0.008340666070580482 2023-01-23 13:13:56.189764: step: 298/1541, loss: 0.042744554579257965 2023-01-23 13:13:57.058294: step: 300/1541, loss: 0.1905391961336136 2023-01-23 13:13:57.948723: step: 302/1541, loss: 0.0239050704985857 2023-01-23 13:13:58.821444: step: 304/1541, loss: 0.07488715648651123 2023-01-23 13:13:59.774940: step: 306/1541, loss: 0.050656117498874664 2023-01-23 13:14:00.650239: step: 308/1541, loss: 0.020196668803691864 2023-01-23 13:14:01.500169: step: 310/1541, loss: 0.05619847774505615 2023-01-23 13:14:02.436884: step: 312/1541, loss: 0.020331794396042824 2023-01-23 13:14:03.334991: step: 314/1541, loss: 0.001892792759463191 2023-01-23 13:14:04.210993: step: 316/1541, loss: 0.0037725395523011684 2023-01-23 13:14:05.076258: step: 318/1541, loss: 0.050093986093997955 2023-01-23 13:14:05.941956: step: 320/1541, loss: 0.12933918833732605 2023-01-23 13:14:06.803081: step: 322/1541, loss: 0.08472442626953125 2023-01-23 13:14:07.702671: step: 324/1541, loss: 0.05550583451986313 2023-01-23 13:14:08.545625: step: 326/1541, loss: 0.042352039366960526 2023-01-23 13:14:09.428900: step: 328/1541, loss: 0.0005784376990050077 2023-01-23 13:14:10.327175: step: 330/1541, loss: 0.231533020734787 2023-01-23 13:14:11.246967: step: 332/1541, loss: 0.1273551881313324 2023-01-23 13:14:12.147417: step: 334/1541, loss: 0.12400264292955399 2023-01-23 13:14:13.059506: step: 336/1541, loss: 0.02385353483259678 2023-01-23 13:14:13.968374: step: 338/1541, loss: 0.05171962082386017 2023-01-23 13:14:14.911843: step: 340/1541, loss: 0.17510110139846802 2023-01-23 13:14:15.865921: step: 342/1541, loss: 0.12122102081775665 2023-01-23 13:14:16.772815: step: 344/1541, loss: 0.020850952714681625 2023-01-23 13:14:17.712528: step: 346/1541, loss: 0.38115033507347107 2023-01-23 13:14:18.655705: step: 348/1541, loss: 0.20131361484527588 2023-01-23 13:14:19.591978: step: 350/1541, loss: 0.04100744426250458 2023-01-23 13:14:20.470427: step: 352/1541, loss: 0.004574654623866081 2023-01-23 13:14:21.356313: step: 354/1541, loss: 0.001984362956136465 2023-01-23 13:14:22.272237: step: 356/1541, loss: 0.04342930391430855 2023-01-23 13:14:23.153744: step: 358/1541, loss: 0.1254783272743225 2023-01-23 13:14:24.041619: step: 360/1541, loss: 0.17016085982322693 2023-01-23 13:14:24.936858: step: 362/1541, loss: 0.07567794620990753 2023-01-23 13:14:25.821005: step: 364/1541, loss: 0.013461025431752205 2023-01-23 13:14:26.743838: step: 366/1541, loss: 0.32113364338874817 2023-01-23 13:14:27.653105: step: 368/1541, loss: 0.02278517186641693 2023-01-23 13:14:28.548242: step: 370/1541, loss: 0.005846342071890831 2023-01-23 13:14:29.498703: step: 372/1541, loss: 0.0026037925854325294 2023-01-23 13:14:30.372316: step: 374/1541, loss: 0.3145577907562256 2023-01-23 13:14:31.266937: step: 376/1541, loss: 0.03062773123383522 2023-01-23 13:14:32.277762: step: 378/1541, loss: 0.0045675658620893955 2023-01-23 13:14:33.170422: step: 380/1541, loss: 0.0007514451281167567 2023-01-23 13:14:34.030961: step: 382/1541, loss: 0.025017831474542618 2023-01-23 13:14:34.937165: step: 384/1541, loss: 0.7040007710456848 2023-01-23 13:14:35.845663: step: 386/1541, loss: 0.08765818923711777 2023-01-23 13:14:36.755647: step: 388/1541, loss: 0.08632253855466843 2023-01-23 13:14:37.638975: step: 390/1541, loss: 0.07163075357675552 2023-01-23 13:14:38.580441: step: 392/1541, loss: 0.006388225592672825 2023-01-23 13:14:39.548219: step: 394/1541, loss: 0.0509164035320282 2023-01-23 13:14:40.427066: step: 396/1541, loss: 0.2200157344341278 2023-01-23 13:14:41.286462: step: 398/1541, loss: 0.010247222147881985 2023-01-23 13:14:42.181059: step: 400/1541, loss: 0.015162996016442776 2023-01-23 13:14:43.065679: step: 402/1541, loss: 0.04015033319592476 2023-01-23 13:14:44.057796: step: 404/1541, loss: 0.18576335906982422 2023-01-23 13:14:44.951054: step: 406/1541, loss: 0.2597793936729431 2023-01-23 13:14:45.803196: step: 408/1541, loss: 0.013601619750261307 2023-01-23 13:14:46.680625: step: 410/1541, loss: 0.001924270298331976 2023-01-23 13:14:47.577294: step: 412/1541, loss: 0.1858125776052475 2023-01-23 13:14:48.501479: step: 414/1541, loss: 0.10947094857692719 2023-01-23 13:14:49.404904: step: 416/1541, loss: 0.9113889932632446 2023-01-23 13:14:50.294565: step: 418/1541, loss: 0.005804963409900665 2023-01-23 13:14:51.143981: step: 420/1541, loss: 0.5701504945755005 2023-01-23 13:14:52.047344: step: 422/1541, loss: 0.07563351094722748 2023-01-23 13:14:52.988907: step: 424/1541, loss: 0.025721915066242218 2023-01-23 13:14:53.888404: step: 426/1541, loss: 0.5155048966407776 2023-01-23 13:14:54.774401: step: 428/1541, loss: 0.11615440249443054 2023-01-23 13:14:55.673434: step: 430/1541, loss: 0.08859451860189438 2023-01-23 13:14:56.634789: step: 432/1541, loss: 0.07589344680309296 2023-01-23 13:14:57.511368: step: 434/1541, loss: 0.11613765358924866 2023-01-23 13:14:58.443042: step: 436/1541, loss: 0.05779512971639633 2023-01-23 13:14:59.360478: step: 438/1541, loss: 0.20021799206733704 2023-01-23 13:15:00.261922: step: 440/1541, loss: 0.04281631112098694 2023-01-23 13:15:01.179523: step: 442/1541, loss: 0.11660642921924591 2023-01-23 13:15:02.054043: step: 444/1541, loss: 0.052661679685115814 2023-01-23 13:15:02.958632: step: 446/1541, loss: 0.06275466084480286 2023-01-23 13:15:03.858420: step: 448/1541, loss: 0.1366511434316635 2023-01-23 13:15:04.773167: step: 450/1541, loss: 0.07968856394290924 2023-01-23 13:15:05.636709: step: 452/1541, loss: 0.25713998079299927 2023-01-23 13:15:06.499136: step: 454/1541, loss: 0.012210873886942863 2023-01-23 13:15:07.413454: step: 456/1541, loss: 0.06675802171230316 2023-01-23 13:15:08.293366: step: 458/1541, loss: 0.006813528016209602 2023-01-23 13:15:09.199846: step: 460/1541, loss: 0.019562041386961937 2023-01-23 13:15:10.119716: step: 462/1541, loss: 0.014044535346329212 2023-01-23 13:15:11.035862: step: 464/1541, loss: 0.3026881217956543 2023-01-23 13:15:11.943085: step: 466/1541, loss: 0.07811924070119858 2023-01-23 13:15:12.837673: step: 468/1541, loss: 0.22466407716274261 2023-01-23 13:15:13.757943: step: 470/1541, loss: 0.05211891606450081 2023-01-23 13:15:14.657972: step: 472/1541, loss: 0.0026304726488888264 2023-01-23 13:15:15.525417: step: 474/1541, loss: 0.19221119582653046 2023-01-23 13:15:16.390894: step: 476/1541, loss: 0.0062649063766002655 2023-01-23 13:15:17.303873: step: 478/1541, loss: 0.043697554618120193 2023-01-23 13:15:18.185733: step: 480/1541, loss: 0.06614244729280472 2023-01-23 13:15:19.126739: step: 482/1541, loss: 0.06756594032049179 2023-01-23 13:15:20.019032: step: 484/1541, loss: 0.9071804881095886 2023-01-23 13:15:20.958516: step: 486/1541, loss: 0.007474147714674473 2023-01-23 13:15:21.817587: step: 488/1541, loss: 0.03466964140534401 2023-01-23 13:15:22.726292: step: 490/1541, loss: 0.07002197206020355 2023-01-23 13:15:23.651995: step: 492/1541, loss: 0.29866158962249756 2023-01-23 13:15:24.533118: step: 494/1541, loss: 0.00564575707539916 2023-01-23 13:15:25.419777: step: 496/1541, loss: 0.023806370794773102 2023-01-23 13:15:26.366071: step: 498/1541, loss: 0.05549122020602226 2023-01-23 13:15:27.322026: step: 500/1541, loss: 0.07044035196304321 2023-01-23 13:15:28.221159: step: 502/1541, loss: 0.2643941342830658 2023-01-23 13:15:29.156148: step: 504/1541, loss: 0.1646927297115326 2023-01-23 13:15:30.028911: step: 506/1541, loss: 0.008433074690401554 2023-01-23 13:15:30.922187: step: 508/1541, loss: 0.005053425673395395 2023-01-23 13:15:31.853474: step: 510/1541, loss: 0.1693253368139267 2023-01-23 13:15:32.751435: step: 512/1541, loss: 0.004911074414849281 2023-01-23 13:15:33.623807: step: 514/1541, loss: 0.11432237178087234 2023-01-23 13:15:34.560669: step: 516/1541, loss: 0.14591601490974426 2023-01-23 13:15:35.461584: step: 518/1541, loss: 0.21067847311496735 2023-01-23 13:15:36.351464: step: 520/1541, loss: 0.01847182773053646 2023-01-23 13:15:37.236767: step: 522/1541, loss: 0.07076986134052277 2023-01-23 13:15:38.143307: step: 524/1541, loss: 0.07378102838993073 2023-01-23 13:15:39.041047: step: 526/1541, loss: 0.01887853629887104 2023-01-23 13:15:39.933897: step: 528/1541, loss: 0.18138080835342407 2023-01-23 13:15:40.852916: step: 530/1541, loss: 0.006518143229186535 2023-01-23 13:15:41.754220: step: 532/1541, loss: 0.029962578788399696 2023-01-23 13:15:42.663156: step: 534/1541, loss: 0.17386187613010406 2023-01-23 13:15:43.570462: step: 536/1541, loss: 0.10145531594753265 2023-01-23 13:15:44.504051: step: 538/1541, loss: 0.17677608132362366 2023-01-23 13:15:45.418068: step: 540/1541, loss: 0.0019416952272877097 2023-01-23 13:15:46.425467: step: 542/1541, loss: 0.03398469090461731 2023-01-23 13:15:47.322429: step: 544/1541, loss: 0.012979067862033844 2023-01-23 13:15:48.218292: step: 546/1541, loss: 0.04614932835102081 2023-01-23 13:15:49.096690: step: 548/1541, loss: 0.4004869759082794 2023-01-23 13:15:49.978479: step: 550/1541, loss: 0.5461052656173706 2023-01-23 13:15:50.910682: step: 552/1541, loss: 0.05885918065905571 2023-01-23 13:15:51.809171: step: 554/1541, loss: 0.06884634494781494 2023-01-23 13:15:52.691972: step: 556/1541, loss: 0.05129464343190193 2023-01-23 13:15:53.576473: step: 558/1541, loss: 0.02915334142744541 2023-01-23 13:15:54.464999: step: 560/1541, loss: 0.5236309766769409 2023-01-23 13:15:55.340980: step: 562/1541, loss: 0.3392239511013031 2023-01-23 13:15:56.235838: step: 564/1541, loss: 0.003989610821008682 2023-01-23 13:15:57.120190: step: 566/1541, loss: 0.009524248540401459 2023-01-23 13:15:58.024865: step: 568/1541, loss: 0.127654567360878 2023-01-23 13:15:58.965070: step: 570/1541, loss: 0.023816514760255814 2023-01-23 13:15:59.845503: step: 572/1541, loss: 1.010927438735962 2023-01-23 13:16:00.741780: step: 574/1541, loss: 0.08303532749414444 2023-01-23 13:16:01.635043: step: 576/1541, loss: 0.12676919996738434 2023-01-23 13:16:02.622639: step: 578/1541, loss: 0.12842561304569244 2023-01-23 13:16:03.500918: step: 580/1541, loss: 0.00042644940549507737 2023-01-23 13:16:04.418347: step: 582/1541, loss: 0.17245282232761383 2023-01-23 13:16:05.321047: step: 584/1541, loss: 0.21671265363693237 2023-01-23 13:16:06.205270: step: 586/1541, loss: 0.007820278406143188 2023-01-23 13:16:07.080108: step: 588/1541, loss: 0.021137235686182976 2023-01-23 13:16:07.985378: step: 590/1541, loss: 0.16265390813350677 2023-01-23 13:16:08.921884: step: 592/1541, loss: 0.1508551687002182 2023-01-23 13:16:09.834602: step: 594/1541, loss: 0.07327046245336533 2023-01-23 13:16:10.707253: step: 596/1541, loss: 0.016217971220612526 2023-01-23 13:16:11.641255: step: 598/1541, loss: 0.11566101014614105 2023-01-23 13:16:12.523305: step: 600/1541, loss: 0.021723918616771698 2023-01-23 13:16:13.428032: step: 602/1541, loss: 0.03463612124323845 2023-01-23 13:16:14.316213: step: 604/1541, loss: 0.4640634059906006 2023-01-23 13:16:15.224402: step: 606/1541, loss: 0.4834745228290558 2023-01-23 13:16:16.128267: step: 608/1541, loss: 0.07505625486373901 2023-01-23 13:16:17.012097: step: 610/1541, loss: 0.3176939785480499 2023-01-23 13:16:17.896371: step: 612/1541, loss: 0.04104199260473251 2023-01-23 13:16:18.815804: step: 614/1541, loss: 0.006834205240011215 2023-01-23 13:16:19.740002: step: 616/1541, loss: 0.11130981147289276 2023-01-23 13:16:20.657413: step: 618/1541, loss: 0.011764477007091045 2023-01-23 13:16:21.565451: step: 620/1541, loss: 0.01605243794620037 2023-01-23 13:16:22.497909: step: 622/1541, loss: 0.056519437581300735 2023-01-23 13:16:23.378955: step: 624/1541, loss: 0.017870251089334488 2023-01-23 13:16:24.265394: step: 626/1541, loss: 0.00963474903255701 2023-01-23 13:16:25.121962: step: 628/1541, loss: 0.8657944202423096 2023-01-23 13:16:25.999514: step: 630/1541, loss: 0.25111502408981323 2023-01-23 13:16:26.883456: step: 632/1541, loss: 0.11442916095256805 2023-01-23 13:16:27.873780: step: 634/1541, loss: 0.008623220026493073 2023-01-23 13:16:28.733794: step: 636/1541, loss: 0.013858847320079803 2023-01-23 13:16:29.620718: step: 638/1541, loss: 0.21002548933029175 2023-01-23 13:16:30.493663: step: 640/1541, loss: 0.08338141441345215 2023-01-23 13:16:31.430536: step: 642/1541, loss: 0.18559911847114563 2023-01-23 13:16:32.314617: step: 644/1541, loss: 0.010964857414364815 2023-01-23 13:16:33.197390: step: 646/1541, loss: 0.09149401634931564 2023-01-23 13:16:34.067986: step: 648/1541, loss: 0.6269261240959167 2023-01-23 13:16:34.981396: step: 650/1541, loss: 0.12280185520648956 2023-01-23 13:16:35.917368: step: 652/1541, loss: 0.0304156132042408 2023-01-23 13:16:36.818555: step: 654/1541, loss: 0.006771347019821405 2023-01-23 13:16:37.742025: step: 656/1541, loss: 0.01980961300432682 2023-01-23 13:16:38.649695: step: 658/1541, loss: 0.041579294949769974 2023-01-23 13:16:39.522363: step: 660/1541, loss: 0.027013450860977173 2023-01-23 13:16:40.407070: step: 662/1541, loss: 0.010231840424239635 2023-01-23 13:16:41.328905: step: 664/1541, loss: 0.020755421370267868 2023-01-23 13:16:42.268645: step: 666/1541, loss: 0.017519040033221245 2023-01-23 13:16:43.171902: step: 668/1541, loss: 0.05344212055206299 2023-01-23 13:16:44.051622: step: 670/1541, loss: 0.0044530238956213 2023-01-23 13:16:44.976294: step: 672/1541, loss: 0.04073942080140114 2023-01-23 13:16:45.871004: step: 674/1541, loss: 0.015461381524801254 2023-01-23 13:16:46.813489: step: 676/1541, loss: 0.17371031641960144 2023-01-23 13:16:47.690987: step: 678/1541, loss: 0.03134732320904732 2023-01-23 13:16:48.563321: step: 680/1541, loss: 0.045595381408929825 2023-01-23 13:16:49.448501: step: 682/1541, loss: 0.039614103734493256 2023-01-23 13:16:50.375846: step: 684/1541, loss: 0.07836824655532837 2023-01-23 13:16:51.278545: step: 686/1541, loss: 0.0013104017125442624 2023-01-23 13:16:52.164642: step: 688/1541, loss: 0.011345231905579567 2023-01-23 13:16:53.082143: step: 690/1541, loss: 0.17100460827350616 2023-01-23 13:16:53.986086: step: 692/1541, loss: 0.02066144533455372 2023-01-23 13:16:54.873675: step: 694/1541, loss: 0.29427778720855713 2023-01-23 13:16:55.826130: step: 696/1541, loss: 0.013348768465220928 2023-01-23 13:16:56.820595: step: 698/1541, loss: 0.07001550495624542 2023-01-23 13:16:57.733565: step: 700/1541, loss: 0.16545145213603973 2023-01-23 13:16:58.732308: step: 702/1541, loss: 0.0032134654466062784 2023-01-23 13:16:59.648206: step: 704/1541, loss: 0.05043766275048256 2023-01-23 13:17:00.518121: step: 706/1541, loss: 0.18018977344036102 2023-01-23 13:17:01.392504: step: 708/1541, loss: 0.00498165050521493 2023-01-23 13:17:02.273465: step: 710/1541, loss: 0.13564527034759521 2023-01-23 13:17:03.199467: step: 712/1541, loss: 0.18393981456756592 2023-01-23 13:17:04.057478: step: 714/1541, loss: 0.14923040568828583 2023-01-23 13:17:05.010566: step: 716/1541, loss: 0.005422748625278473 2023-01-23 13:17:05.879023: step: 718/1541, loss: 0.006824526470154524 2023-01-23 13:17:06.802126: step: 720/1541, loss: 0.03948064520955086 2023-01-23 13:17:07.697996: step: 722/1541, loss: 0.010665354318916798 2023-01-23 13:17:08.602805: step: 724/1541, loss: 0.02776268869638443 2023-01-23 13:17:09.572975: step: 726/1541, loss: 0.14106431603431702 2023-01-23 13:17:10.442897: step: 728/1541, loss: 0.06679695844650269 2023-01-23 13:17:11.294153: step: 730/1541, loss: 0.04989100247621536 2023-01-23 13:17:12.210767: step: 732/1541, loss: 0.03653576225042343 2023-01-23 13:17:13.129631: step: 734/1541, loss: 0.024934284389019012 2023-01-23 13:17:13.996478: step: 736/1541, loss: 0.1642642766237259 2023-01-23 13:17:14.876141: step: 738/1541, loss: 0.10278841108083725 2023-01-23 13:17:15.805726: step: 740/1541, loss: 0.0912117063999176 2023-01-23 13:17:16.673243: step: 742/1541, loss: 0.004065868444740772 2023-01-23 13:17:17.553449: step: 744/1541, loss: 0.12424866110086441 2023-01-23 13:17:18.466583: step: 746/1541, loss: 0.3832933008670807 2023-01-23 13:17:19.361969: step: 748/1541, loss: 0.03231837972998619 2023-01-23 13:17:20.244978: step: 750/1541, loss: 0.02422720566391945 2023-01-23 13:17:21.161321: step: 752/1541, loss: 0.0151997534558177 2023-01-23 13:17:22.080471: step: 754/1541, loss: 0.003213518299162388 2023-01-23 13:17:22.941524: step: 756/1541, loss: 0.5728037357330322 2023-01-23 13:17:23.811766: step: 758/1541, loss: 0.005970542784780264 2023-01-23 13:17:24.762570: step: 760/1541, loss: 0.07388678193092346 2023-01-23 13:17:25.716012: step: 762/1541, loss: 0.17014017701148987 2023-01-23 13:17:26.617890: step: 764/1541, loss: 0.10693033039569855 2023-01-23 13:17:27.518815: step: 766/1541, loss: 0.004766104277223349 2023-01-23 13:17:28.441629: step: 768/1541, loss: 0.061655834317207336 2023-01-23 13:17:29.341769: step: 770/1541, loss: 0.41923999786376953 2023-01-23 13:17:30.223571: step: 772/1541, loss: 0.02323102578520775 2023-01-23 13:17:31.093040: step: 774/1541, loss: 0.02031773142516613 2023-01-23 13:17:31.985566: step: 776/1541, loss: 0.30764269828796387 2023-01-23 13:17:32.864154: step: 778/1541, loss: 0.09607541561126709 2023-01-23 13:17:33.773680: step: 780/1541, loss: 0.03783353790640831 2023-01-23 13:17:34.691303: step: 782/1541, loss: 0.08871351927518845 2023-01-23 13:17:35.599482: step: 784/1541, loss: 0.031345970928668976 2023-01-23 13:17:36.472537: step: 786/1541, loss: 0.26151710748672485 2023-01-23 13:17:37.391824: step: 788/1541, loss: 0.024086199700832367 2023-01-23 13:17:38.273336: step: 790/1541, loss: 0.051537539809942245 2023-01-23 13:17:39.121844: step: 792/1541, loss: 0.04759884625673294 2023-01-23 13:17:40.049513: step: 794/1541, loss: 1.6499288082122803 2023-01-23 13:17:40.933657: step: 796/1541, loss: 0.014800909906625748 2023-01-23 13:17:41.950934: step: 798/1541, loss: 0.19215740263462067 2023-01-23 13:17:42.862757: step: 800/1541, loss: 0.00776332663372159 2023-01-23 13:17:43.757456: step: 802/1541, loss: 0.012565119192004204 2023-01-23 13:17:44.655695: step: 804/1541, loss: 0.15910914540290833 2023-01-23 13:17:45.526880: step: 806/1541, loss: 0.2668585777282715 2023-01-23 13:17:46.410211: step: 808/1541, loss: 0.002610441530123353 2023-01-23 13:17:47.302940: step: 810/1541, loss: 0.017783887684345245 2023-01-23 13:17:48.203979: step: 812/1541, loss: 0.006163714453577995 2023-01-23 13:17:49.071784: step: 814/1541, loss: 0.10477820038795471 2023-01-23 13:17:50.045885: step: 816/1541, loss: 0.015250623226165771 2023-01-23 13:17:50.960682: step: 818/1541, loss: 0.12925000488758087 2023-01-23 13:17:51.863414: step: 820/1541, loss: 0.16139492392539978 2023-01-23 13:17:52.763130: step: 822/1541, loss: 0.1841122955083847 2023-01-23 13:17:53.643916: step: 824/1541, loss: 0.3275120258331299 2023-01-23 13:17:54.530833: step: 826/1541, loss: 0.015256223268806934 2023-01-23 13:17:55.419077: step: 828/1541, loss: 0.19999036192893982 2023-01-23 13:17:56.373069: step: 830/1541, loss: 0.0652628019452095 2023-01-23 13:17:57.310715: step: 832/1541, loss: 0.21764153242111206 2023-01-23 13:17:58.210288: step: 834/1541, loss: 0.04825722798705101 2023-01-23 13:17:59.112887: step: 836/1541, loss: 0.222250834107399 2023-01-23 13:18:00.036997: step: 838/1541, loss: 0.004957143217325211 2023-01-23 13:18:00.925097: step: 840/1541, loss: 0.06553681194782257 2023-01-23 13:18:01.844862: step: 842/1541, loss: 0.21115034818649292 2023-01-23 13:18:02.737572: step: 844/1541, loss: 0.02847551368176937 2023-01-23 13:18:03.624097: step: 846/1541, loss: 0.1271306872367859 2023-01-23 13:18:04.627998: step: 848/1541, loss: 0.21822486817836761 2023-01-23 13:18:05.463515: step: 850/1541, loss: 0.36513400077819824 2023-01-23 13:18:06.338917: step: 852/1541, loss: 0.0028762209694832563 2023-01-23 13:18:07.226429: step: 854/1541, loss: 0.01898910105228424 2023-01-23 13:18:08.096997: step: 856/1541, loss: 0.14016176760196686 2023-01-23 13:18:09.001712: step: 858/1541, loss: 0.2879003882408142 2023-01-23 13:18:09.930108: step: 860/1541, loss: 0.009314239025115967 2023-01-23 13:18:10.828465: step: 862/1541, loss: 0.015695741400122643 2023-01-23 13:18:11.770722: step: 864/1541, loss: 0.02397802472114563 2023-01-23 13:18:12.783355: step: 866/1541, loss: 0.2996315062046051 2023-01-23 13:18:13.657616: step: 868/1541, loss: 0.004756936803460121 2023-01-23 13:18:14.546851: step: 870/1541, loss: 0.12349354475736618 2023-01-23 13:18:15.512460: step: 872/1541, loss: 0.04223296791315079 2023-01-23 13:18:16.380405: step: 874/1541, loss: 0.017367251217365265 2023-01-23 13:18:17.256649: step: 876/1541, loss: 0.03393436223268509 2023-01-23 13:18:18.164545: step: 878/1541, loss: 0.020823083817958832 2023-01-23 13:18:19.071712: step: 880/1541, loss: 0.04110778123140335 2023-01-23 13:18:19.924557: step: 882/1541, loss: 0.006354108452796936 2023-01-23 13:18:20.902711: step: 884/1541, loss: 0.09796854108572006 2023-01-23 13:18:21.786080: step: 886/1541, loss: 0.05853315070271492 2023-01-23 13:18:22.724379: step: 888/1541, loss: 0.31612303853034973 2023-01-23 13:18:23.594069: step: 890/1541, loss: 0.0684494823217392 2023-01-23 13:18:24.503443: step: 892/1541, loss: 1.1163345575332642 2023-01-23 13:18:25.388188: step: 894/1541, loss: 0.020117750391364098 2023-01-23 13:18:26.274273: step: 896/1541, loss: 0.13571199774742126 2023-01-23 13:18:27.185533: step: 898/1541, loss: 0.039978139102458954 2023-01-23 13:18:28.141889: step: 900/1541, loss: 0.005985496565699577 2023-01-23 13:18:29.047700: step: 902/1541, loss: 0.07939185202121735 2023-01-23 13:18:29.921360: step: 904/1541, loss: 0.02292976714670658 2023-01-23 13:18:30.839845: step: 906/1541, loss: 0.013072159141302109 2023-01-23 13:18:31.710786: step: 908/1541, loss: 0.23328369855880737 2023-01-23 13:18:32.593892: step: 910/1541, loss: 0.029828406870365143 2023-01-23 13:18:33.465597: step: 912/1541, loss: 0.24118244647979736 2023-01-23 13:18:34.352505: step: 914/1541, loss: 0.04303163290023804 2023-01-23 13:18:35.232146: step: 916/1541, loss: 0.0014218484284356236 2023-01-23 13:18:36.209165: step: 918/1541, loss: 0.05167902633547783 2023-01-23 13:18:37.056725: step: 920/1541, loss: 0.00439149746671319 2023-01-23 13:18:37.975512: step: 922/1541, loss: 0.028492126613855362 2023-01-23 13:18:38.899947: step: 924/1541, loss: 0.005213480442762375 2023-01-23 13:18:39.822850: step: 926/1541, loss: 0.10809888690710068 2023-01-23 13:18:40.746172: step: 928/1541, loss: 0.0008490345207974315 2023-01-23 13:18:41.667268: step: 930/1541, loss: 0.1478426605463028 2023-01-23 13:18:42.539533: step: 932/1541, loss: 0.4311121702194214 2023-01-23 13:18:43.480845: step: 934/1541, loss: 0.021310165524482727 2023-01-23 13:18:44.365586: step: 936/1541, loss: 0.024716423824429512 2023-01-23 13:18:45.230101: step: 938/1541, loss: 0.07156320661306381 2023-01-23 13:18:46.122186: step: 940/1541, loss: 0.07636986672878265 2023-01-23 13:18:47.180266: step: 942/1541, loss: 0.12498950213193893 2023-01-23 13:18:48.113358: step: 944/1541, loss: 0.0411287285387516 2023-01-23 13:18:49.013949: step: 946/1541, loss: 0.07960078865289688 2023-01-23 13:18:49.956420: step: 948/1541, loss: 0.0004094674368388951 2023-01-23 13:18:50.827673: step: 950/1541, loss: 0.06472969055175781 2023-01-23 13:18:51.738472: step: 952/1541, loss: 0.07051781564950943 2023-01-23 13:18:52.652512: step: 954/1541, loss: 0.0829668939113617 2023-01-23 13:18:53.531461: step: 956/1541, loss: 0.0002948674373328686 2023-01-23 13:18:54.411770: step: 958/1541, loss: 0.02331748604774475 2023-01-23 13:18:55.364748: step: 960/1541, loss: 0.12403231114149094 2023-01-23 13:18:56.275040: step: 962/1541, loss: 0.008729968219995499 2023-01-23 13:18:57.174506: step: 964/1541, loss: 0.025379912927746773 2023-01-23 13:18:58.081318: step: 966/1541, loss: 0.03443967178463936 2023-01-23 13:18:58.925127: step: 968/1541, loss: 0.24560889601707458 2023-01-23 13:18:59.828832: step: 970/1541, loss: 0.21735930442810059 2023-01-23 13:19:00.744827: step: 972/1541, loss: 0.08282952755689621 2023-01-23 13:19:01.619462: step: 974/1541, loss: 0.2422586977481842 2023-01-23 13:19:02.591791: step: 976/1541, loss: 0.04729422926902771 2023-01-23 13:19:03.479275: step: 978/1541, loss: 0.03738318011164665 2023-01-23 13:19:04.359248: step: 980/1541, loss: 0.07164213061332703 2023-01-23 13:19:05.321800: step: 982/1541, loss: 0.25642135739326477 2023-01-23 13:19:06.211949: step: 984/1541, loss: 0.05447705090045929 2023-01-23 13:19:07.147575: step: 986/1541, loss: 0.3018224835395813 2023-01-23 13:19:08.108201: step: 988/1541, loss: 0.07228723168373108 2023-01-23 13:19:09.026861: step: 990/1541, loss: 0.18089228868484497 2023-01-23 13:19:09.922022: step: 992/1541, loss: 0.025456929579377174 2023-01-23 13:19:10.861703: step: 994/1541, loss: 0.038795556873083115 2023-01-23 13:19:11.831316: step: 996/1541, loss: 0.13822104036808014 2023-01-23 13:19:12.747079: step: 998/1541, loss: 0.09221699833869934 2023-01-23 13:19:13.671815: step: 1000/1541, loss: 0.10931505262851715 2023-01-23 13:19:14.558911: step: 1002/1541, loss: 0.007945956662297249 2023-01-23 13:19:15.571958: step: 1004/1541, loss: 0.00958828441798687 2023-01-23 13:19:16.506481: step: 1006/1541, loss: 0.10626472532749176 2023-01-23 13:19:17.404566: step: 1008/1541, loss: 0.005616665352135897 2023-01-23 13:19:18.282817: step: 1010/1541, loss: 0.031200209632515907 2023-01-23 13:19:19.167013: step: 1012/1541, loss: 0.06856995075941086 2023-01-23 13:19:20.100051: step: 1014/1541, loss: 0.04461256042122841 2023-01-23 13:19:20.990937: step: 1016/1541, loss: 0.13765355944633484 2023-01-23 13:19:21.922323: step: 1018/1541, loss: 0.03339853510260582 2023-01-23 13:19:22.842547: step: 1020/1541, loss: 0.057664595544338226 2023-01-23 13:19:23.717835: step: 1022/1541, loss: 0.045342955738306046 2023-01-23 13:19:24.620190: step: 1024/1541, loss: 0.010907329618930817 2023-01-23 13:19:25.504854: step: 1026/1541, loss: 0.006321973167359829 2023-01-23 13:19:26.407418: step: 1028/1541, loss: 0.582557201385498 2023-01-23 13:19:27.286677: step: 1030/1541, loss: 0.23636996746063232 2023-01-23 13:19:28.151655: step: 1032/1541, loss: 0.19994854927062988 2023-01-23 13:19:29.036394: step: 1034/1541, loss: 0.010020453482866287 2023-01-23 13:19:29.961174: step: 1036/1541, loss: 0.0055518727749586105 2023-01-23 13:19:30.880396: step: 1038/1541, loss: 0.010954076424241066 2023-01-23 13:19:31.761094: step: 1040/1541, loss: 0.043013665825128555 2023-01-23 13:19:32.666580: step: 1042/1541, loss: 0.13304385542869568 2023-01-23 13:19:33.648287: step: 1044/1541, loss: 0.07439865171909332 2023-01-23 13:19:34.521126: step: 1046/1541, loss: 0.014444536529481411 2023-01-23 13:19:35.411509: step: 1048/1541, loss: 0.5707491636276245 2023-01-23 13:19:36.321755: step: 1050/1541, loss: 0.008325711823999882 2023-01-23 13:19:37.224866: step: 1052/1541, loss: 0.026415016502141953 2023-01-23 13:19:38.115329: step: 1054/1541, loss: 1.0989279747009277 2023-01-23 13:19:38.992933: step: 1056/1541, loss: 0.03171064704656601 2023-01-23 13:19:39.914943: step: 1058/1541, loss: 0.09401664137840271 2023-01-23 13:19:40.806677: step: 1060/1541, loss: 0.004856400191783905 2023-01-23 13:19:41.692268: step: 1062/1541, loss: 0.032295357435941696 2023-01-23 13:19:42.573956: step: 1064/1541, loss: 0.0008176096598617733 2023-01-23 13:19:43.436482: step: 1066/1541, loss: 0.005817277356982231 2023-01-23 13:19:44.346851: step: 1068/1541, loss: 0.49361613392829895 2023-01-23 13:19:45.303101: step: 1070/1541, loss: 0.012075724080204964 2023-01-23 13:19:46.242678: step: 1072/1541, loss: 0.09049274772405624 2023-01-23 13:19:47.174606: step: 1074/1541, loss: 0.04390905797481537 2023-01-23 13:19:48.042607: step: 1076/1541, loss: 0.004895934369415045 2023-01-23 13:19:48.951095: step: 1078/1541, loss: 0.010105658322572708 2023-01-23 13:19:49.802111: step: 1080/1541, loss: 0.3154439628124237 2023-01-23 13:19:50.708876: step: 1082/1541, loss: 0.04373719543218613 2023-01-23 13:19:51.692877: step: 1084/1541, loss: 0.04966335371136665 2023-01-23 13:19:52.572752: step: 1086/1541, loss: 0.001829140936024487 2023-01-23 13:19:53.528052: step: 1088/1541, loss: 0.000512718572281301 2023-01-23 13:19:54.426699: step: 1090/1541, loss: 0.004114057868719101 2023-01-23 13:19:55.496388: step: 1092/1541, loss: 0.05661249905824661 2023-01-23 13:19:56.406909: step: 1094/1541, loss: 0.022276639938354492 2023-01-23 13:19:57.316252: step: 1096/1541, loss: 0.16531483829021454 2023-01-23 13:19:58.205177: step: 1098/1541, loss: 0.05908655375242233 2023-01-23 13:19:59.223137: step: 1100/1541, loss: 0.26792988181114197 2023-01-23 13:20:00.129912: step: 1102/1541, loss: 0.018401462584733963 2023-01-23 13:20:01.032051: step: 1104/1541, loss: 0.7362544536590576 2023-01-23 13:20:01.947750: step: 1106/1541, loss: 0.021327266469597816 2023-01-23 13:20:02.840284: step: 1108/1541, loss: 0.08465167135000229 2023-01-23 13:20:03.787442: step: 1110/1541, loss: 0.02984076738357544 2023-01-23 13:20:04.769413: step: 1112/1541, loss: 0.33818331360816956 2023-01-23 13:20:05.689223: step: 1114/1541, loss: 0.01990583911538124 2023-01-23 13:20:06.638153: step: 1116/1541, loss: 0.04977618157863617 2023-01-23 13:20:07.532950: step: 1118/1541, loss: 0.23767158389091492 2023-01-23 13:20:08.487255: step: 1120/1541, loss: 0.028829891234636307 2023-01-23 13:20:09.346832: step: 1122/1541, loss: 0.01985594630241394 2023-01-23 13:20:10.328003: step: 1124/1541, loss: 0.24116972088813782 2023-01-23 13:20:11.181966: step: 1126/1541, loss: 0.012157508172094822 2023-01-23 13:20:12.094052: step: 1128/1541, loss: 0.00848105177283287 2023-01-23 13:20:13.018553: step: 1130/1541, loss: 0.0014149065827950835 2023-01-23 13:20:13.928635: step: 1132/1541, loss: 0.023748552426695824 2023-01-23 13:20:14.837059: step: 1134/1541, loss: 0.03315339982509613 2023-01-23 13:20:15.739249: step: 1136/1541, loss: 0.0015526444185525179 2023-01-23 13:20:16.651295: step: 1138/1541, loss: 0.004281903151422739 2023-01-23 13:20:17.557392: step: 1140/1541, loss: 0.1657373011112213 2023-01-23 13:20:18.449273: step: 1142/1541, loss: 0.023847421631217003 2023-01-23 13:20:19.320081: step: 1144/1541, loss: 0.03035685233771801 2023-01-23 13:20:20.248731: step: 1146/1541, loss: 0.15888634324073792 2023-01-23 13:20:21.188697: step: 1148/1541, loss: 0.2597794532775879 2023-01-23 13:20:22.064534: step: 1150/1541, loss: 0.08205755800008774 2023-01-23 13:20:22.956676: step: 1152/1541, loss: 0.06204358860850334 2023-01-23 13:20:23.855045: step: 1154/1541, loss: 0.010188490152359009 2023-01-23 13:20:24.768923: step: 1156/1541, loss: 0.013407478109002113 2023-01-23 13:20:25.705409: step: 1158/1541, loss: 0.11466729640960693 2023-01-23 13:20:26.581118: step: 1160/1541, loss: 0.1054057627916336 2023-01-23 13:20:27.500240: step: 1162/1541, loss: 0.6960899233818054 2023-01-23 13:20:28.426023: step: 1164/1541, loss: 0.0016743669984862208 2023-01-23 13:20:29.347895: step: 1166/1541, loss: 0.011875038966536522 2023-01-23 13:20:30.208652: step: 1168/1541, loss: 0.061981525272130966 2023-01-23 13:20:31.095443: step: 1170/1541, loss: 0.061308354139328 2023-01-23 13:20:32.026984: step: 1172/1541, loss: 0.0033978717401623726 2023-01-23 13:20:32.965918: step: 1174/1541, loss: 0.1562420278787613 2023-01-23 13:20:33.889715: step: 1176/1541, loss: 0.0016647777520120144 2023-01-23 13:20:34.778487: step: 1178/1541, loss: 0.12745755910873413 2023-01-23 13:20:35.628065: step: 1180/1541, loss: 0.0006721148965880275 2023-01-23 13:20:36.486926: step: 1182/1541, loss: 0.07786582410335541 2023-01-23 13:20:37.433471: step: 1184/1541, loss: 0.0146236727014184 2023-01-23 13:20:38.347890: step: 1186/1541, loss: 0.02755221724510193 2023-01-23 13:20:39.268392: step: 1188/1541, loss: 0.0006399246631190181 2023-01-23 13:20:40.177373: step: 1190/1541, loss: 0.01005490217357874 2023-01-23 13:20:41.050371: step: 1192/1541, loss: 0.008943283930420876 2023-01-23 13:20:41.951741: step: 1194/1541, loss: 0.1406763195991516 2023-01-23 13:20:42.934673: step: 1196/1541, loss: 0.07543447613716125 2023-01-23 13:20:43.819273: step: 1198/1541, loss: 0.16809053719043732 2023-01-23 13:20:44.674433: step: 1200/1541, loss: 0.018586400896310806 2023-01-23 13:20:45.547594: step: 1202/1541, loss: 0.40140336751937866 2023-01-23 13:20:46.452719: step: 1204/1541, loss: 0.002395350020378828 2023-01-23 13:20:47.354962: step: 1206/1541, loss: 0.09399373829364777 2023-01-23 13:20:48.335972: step: 1208/1541, loss: 0.0929272398352623 2023-01-23 13:20:49.211951: step: 1210/1541, loss: 0.0038665509782731533 2023-01-23 13:20:50.155296: step: 1212/1541, loss: 0.04323064908385277 2023-01-23 13:20:51.072499: step: 1214/1541, loss: 0.0010621018009260297 2023-01-23 13:20:51.966578: step: 1216/1541, loss: 0.38827353715896606 2023-01-23 13:20:52.900644: step: 1218/1541, loss: 0.006235811393707991 2023-01-23 13:20:53.827776: step: 1220/1541, loss: 0.002084065694361925 2023-01-23 13:20:54.758029: step: 1222/1541, loss: 0.01953435130417347 2023-01-23 13:20:55.660184: step: 1224/1541, loss: 0.18086229264736176 2023-01-23 13:20:56.565599: step: 1226/1541, loss: 0.14130212366580963 2023-01-23 13:20:57.435043: step: 1228/1541, loss: 0.036937177181243896 2023-01-23 13:20:58.331095: step: 1230/1541, loss: 0.26088961958885193 2023-01-23 13:20:59.222660: step: 1232/1541, loss: 0.12401027977466583 2023-01-23 13:21:00.161979: step: 1234/1541, loss: 0.055558621883392334 2023-01-23 13:21:01.040169: step: 1236/1541, loss: 0.008238913491368294 2023-01-23 13:21:01.947399: step: 1238/1541, loss: 0.1807030588388443 2023-01-23 13:21:02.788908: step: 1240/1541, loss: 0.011412535794079304 2023-01-23 13:21:03.686874: step: 1242/1541, loss: 0.5774466395378113 2023-01-23 13:21:04.572588: step: 1244/1541, loss: 0.05029130354523659 2023-01-23 13:21:05.482604: step: 1246/1541, loss: 0.048569776117801666 2023-01-23 13:21:06.429718: step: 1248/1541, loss: 0.02313295379281044 2023-01-23 13:21:07.316943: step: 1250/1541, loss: 0.17704930901527405 2023-01-23 13:21:08.271143: step: 1252/1541, loss: 0.1466778963804245 2023-01-23 13:21:09.211081: step: 1254/1541, loss: 0.02387051098048687 2023-01-23 13:21:10.079091: step: 1256/1541, loss: 0.06459858268499374 2023-01-23 13:21:10.973016: step: 1258/1541, loss: 0.034445032477378845 2023-01-23 13:21:11.819862: step: 1260/1541, loss: 0.05216214433312416 2023-01-23 13:21:12.686044: step: 1262/1541, loss: 0.021889928728342056 2023-01-23 13:21:13.561370: step: 1264/1541, loss: 0.06129954010248184 2023-01-23 13:21:14.419434: step: 1266/1541, loss: 0.04676266387104988 2023-01-23 13:21:15.298880: step: 1268/1541, loss: 0.17107245326042175 2023-01-23 13:21:16.166319: step: 1270/1541, loss: 0.0015520816668868065 2023-01-23 13:21:17.072552: step: 1272/1541, loss: 0.12875917553901672 2023-01-23 13:21:17.961112: step: 1274/1541, loss: 0.07854516804218292 2023-01-23 13:21:18.817914: step: 1276/1541, loss: 0.019801098853349686 2023-01-23 13:21:19.725450: step: 1278/1541, loss: 0.002113678492605686 2023-01-23 13:21:20.621904: step: 1280/1541, loss: 0.09423168003559113 2023-01-23 13:21:21.512172: step: 1282/1541, loss: 0.18681567907333374 2023-01-23 13:21:22.423898: step: 1284/1541, loss: 0.016060341149568558 2023-01-23 13:21:23.270194: step: 1286/1541, loss: 0.013240509666502476 2023-01-23 13:21:24.163299: step: 1288/1541, loss: 0.3832680881023407 2023-01-23 13:21:25.016417: step: 1290/1541, loss: 0.01880752108991146 2023-01-23 13:21:25.923772: step: 1292/1541, loss: 0.11155514419078827 2023-01-23 13:21:26.843587: step: 1294/1541, loss: 0.21056216955184937 2023-01-23 13:21:27.728074: step: 1296/1541, loss: 0.10484105348587036 2023-01-23 13:21:28.611977: step: 1298/1541, loss: 0.03157151862978935 2023-01-23 13:21:29.521749: step: 1300/1541, loss: 0.23259568214416504 2023-01-23 13:21:30.430685: step: 1302/1541, loss: 0.03388087823987007 2023-01-23 13:21:31.331091: step: 1304/1541, loss: 0.02194192260503769 2023-01-23 13:21:32.284683: step: 1306/1541, loss: 0.09114895761013031 2023-01-23 13:21:33.164922: step: 1308/1541, loss: 0.02407264895737171 2023-01-23 13:21:34.062209: step: 1310/1541, loss: 0.041382499039173126 2023-01-23 13:21:34.991609: step: 1312/1541, loss: 0.15028779208660126 2023-01-23 13:21:35.863503: step: 1314/1541, loss: 0.034768395125865936 2023-01-23 13:21:36.751060: step: 1316/1541, loss: 0.0084220627322793 2023-01-23 13:21:37.669751: step: 1318/1541, loss: 0.20736660063266754 2023-01-23 13:21:38.562127: step: 1320/1541, loss: 0.037168048322200775 2023-01-23 13:21:39.478532: step: 1322/1541, loss: 0.01974881999194622 2023-01-23 13:21:40.420065: step: 1324/1541, loss: 0.009769368916749954 2023-01-23 13:21:41.299411: step: 1326/1541, loss: 0.0019145067781209946 2023-01-23 13:21:42.277747: step: 1328/1541, loss: 0.09522038698196411 2023-01-23 13:21:43.152312: step: 1330/1541, loss: 0.03905206546187401 2023-01-23 13:21:44.087375: step: 1332/1541, loss: 0.010831634514033794 2023-01-23 13:21:44.949675: step: 1334/1541, loss: 0.027120525017380714 2023-01-23 13:21:45.827160: step: 1336/1541, loss: 0.0761134922504425 2023-01-23 13:21:46.725210: step: 1338/1541, loss: 0.38408493995666504 2023-01-23 13:21:47.602992: step: 1340/1541, loss: 0.005626339931041002 2023-01-23 13:21:48.471666: step: 1342/1541, loss: 0.012821693904697895 2023-01-23 13:21:49.440238: step: 1344/1541, loss: 0.22854489088058472 2023-01-23 13:21:50.363116: step: 1346/1541, loss: 0.06174284592270851 2023-01-23 13:21:51.294052: step: 1348/1541, loss: 0.060214996337890625 2023-01-23 13:21:52.163685: step: 1350/1541, loss: 0.0001507510314695537 2023-01-23 13:21:53.048293: step: 1352/1541, loss: 0.0016488103428855538 2023-01-23 13:21:53.961209: step: 1354/1541, loss: 0.0005999734858050942 2023-01-23 13:21:54.892142: step: 1356/1541, loss: 0.0214699599891901 2023-01-23 13:21:55.754188: step: 1358/1541, loss: 0.013449624180793762 2023-01-23 13:21:56.654962: step: 1360/1541, loss: 0.004104962572455406 2023-01-23 13:21:57.520426: step: 1362/1541, loss: 0.0009639157797209918 2023-01-23 13:21:58.379505: step: 1364/1541, loss: 0.002045039087533951 2023-01-23 13:21:59.249349: step: 1366/1541, loss: 0.3827258050441742 2023-01-23 13:22:00.202316: step: 1368/1541, loss: 0.0007724305614829063 2023-01-23 13:22:01.076511: step: 1370/1541, loss: 0.003328014397993684 2023-01-23 13:22:01.947932: step: 1372/1541, loss: 0.02028988115489483 2023-01-23 13:22:02.834180: step: 1374/1541, loss: 0.17653429508209229 2023-01-23 13:22:03.699619: step: 1376/1541, loss: 0.012322774156928062 2023-01-23 13:22:04.569512: step: 1378/1541, loss: 0.16971029341220856 2023-01-23 13:22:05.477172: step: 1380/1541, loss: 0.17367373406887054 2023-01-23 13:22:06.393194: step: 1382/1541, loss: 0.0016561605734750628 2023-01-23 13:22:07.275466: step: 1384/1541, loss: 0.0852598324418068 2023-01-23 13:22:08.137435: step: 1386/1541, loss: 0.1505809724330902 2023-01-23 13:22:09.019341: step: 1388/1541, loss: 0.0005968534969724715 2023-01-23 13:22:09.901820: step: 1390/1541, loss: 0.001871079090051353 2023-01-23 13:22:10.773533: step: 1392/1541, loss: 0.31047728657722473 2023-01-23 13:22:11.741824: step: 1394/1541, loss: 0.0031389796640723944 2023-01-23 13:22:12.632835: step: 1396/1541, loss: 0.052833665162324905 2023-01-23 13:22:13.506928: step: 1398/1541, loss: 0.23581717908382416 2023-01-23 13:22:14.455320: step: 1400/1541, loss: 0.056524042040109634 2023-01-23 13:22:15.405310: step: 1402/1541, loss: 0.007751258555799723 2023-01-23 13:22:16.347141: step: 1404/1541, loss: 0.05848980322480202 2023-01-23 13:22:17.248288: step: 1406/1541, loss: 0.007277533411979675 2023-01-23 13:22:18.138710: step: 1408/1541, loss: 0.005042483564466238 2023-01-23 13:22:19.058582: step: 1410/1541, loss: 0.08441472053527832 2023-01-23 13:22:19.925947: step: 1412/1541, loss: 0.0996207520365715 2023-01-23 13:22:20.974117: step: 1414/1541, loss: 0.12506645917892456 2023-01-23 13:22:21.904992: step: 1416/1541, loss: 0.20778505504131317 2023-01-23 13:22:22.826841: step: 1418/1541, loss: 0.3222147524356842 2023-01-23 13:22:23.706049: step: 1420/1541, loss: 0.015062001533806324 2023-01-23 13:22:24.609335: step: 1422/1541, loss: 0.049929264932870865 2023-01-23 13:22:25.500657: step: 1424/1541, loss: 0.030363745987415314 2023-01-23 13:22:26.358123: step: 1426/1541, loss: 0.13731661438941956 2023-01-23 13:22:27.254008: step: 1428/1541, loss: 0.027601972222328186 2023-01-23 13:22:28.127920: step: 1430/1541, loss: 0.07834615558385849 2023-01-23 13:22:29.036489: step: 1432/1541, loss: 0.016715314239263535 2023-01-23 13:22:29.954757: step: 1434/1541, loss: 0.018802113831043243 2023-01-23 13:22:30.861767: step: 1436/1541, loss: 0.029626306146383286 2023-01-23 13:22:31.755717: step: 1438/1541, loss: 0.5094988942146301 2023-01-23 13:22:32.697091: step: 1440/1541, loss: 0.17679978907108307 2023-01-23 13:22:33.614114: step: 1442/1541, loss: 0.002147621475160122 2023-01-23 13:22:34.490613: step: 1444/1541, loss: 0.04717045649886131 2023-01-23 13:22:35.384090: step: 1446/1541, loss: 0.0149467084556818 2023-01-23 13:22:36.246822: step: 1448/1541, loss: 0.2300068587064743 2023-01-23 13:22:37.150403: step: 1450/1541, loss: 0.026994962245225906 2023-01-23 13:22:38.129379: step: 1452/1541, loss: 0.002550954231992364 2023-01-23 13:22:39.038594: step: 1454/1541, loss: 2.156686544418335 2023-01-23 13:22:39.938941: step: 1456/1541, loss: 0.013657170347869396 2023-01-23 13:22:40.815214: step: 1458/1541, loss: 0.0714966431260109 2023-01-23 13:22:41.719949: step: 1460/1541, loss: 0.21889451146125793 2023-01-23 13:22:42.625287: step: 1462/1541, loss: 0.1373293399810791 2023-01-23 13:22:43.511259: step: 1464/1541, loss: 0.07066337764263153 2023-01-23 13:22:44.388434: step: 1466/1541, loss: 0.00041889649583026767 2023-01-23 13:22:45.272049: step: 1468/1541, loss: 0.006430384702980518 2023-01-23 13:22:46.173236: step: 1470/1541, loss: 0.062394410371780396 2023-01-23 13:22:47.052826: step: 1472/1541, loss: 0.058674171566963196 2023-01-23 13:22:47.934757: step: 1474/1541, loss: 0.03164822235703468 2023-01-23 13:22:48.886590: step: 1476/1541, loss: 0.29828232526779175 2023-01-23 13:22:49.776451: step: 1478/1541, loss: 0.35846397280693054 2023-01-23 13:22:50.657521: step: 1480/1541, loss: 0.006081937346607447 2023-01-23 13:22:51.518926: step: 1482/1541, loss: 0.006836132612079382 2023-01-23 13:22:52.399159: step: 1484/1541, loss: 0.0034070084802806377 2023-01-23 13:22:53.334514: step: 1486/1541, loss: 0.006768969353288412 2023-01-23 13:22:54.293206: step: 1488/1541, loss: 0.11707174777984619 2023-01-23 13:22:55.196124: step: 1490/1541, loss: 0.04867934435606003 2023-01-23 13:22:56.121660: step: 1492/1541, loss: 0.16720597445964813 2023-01-23 13:22:57.041857: step: 1494/1541, loss: 0.30625128746032715 2023-01-23 13:22:57.980232: step: 1496/1541, loss: 0.0003398554108571261 2023-01-23 13:22:58.912014: step: 1498/1541, loss: 0.08009295165538788 2023-01-23 13:22:59.827393: step: 1500/1541, loss: 0.018127981573343277 2023-01-23 13:23:00.681760: step: 1502/1541, loss: 0.0015371732879430056 2023-01-23 13:23:01.545648: step: 1504/1541, loss: 0.06418639421463013 2023-01-23 13:23:02.547769: step: 1506/1541, loss: 0.05156966671347618 2023-01-23 13:23:03.478075: step: 1508/1541, loss: 0.272576242685318 2023-01-23 13:23:04.349298: step: 1510/1541, loss: 0.10610067844390869 2023-01-23 13:23:05.262327: step: 1512/1541, loss: 0.0014433047035709023 2023-01-23 13:23:06.175356: step: 1514/1541, loss: 0.02928965911269188 2023-01-23 13:23:07.058595: step: 1516/1541, loss: 0.04237956553697586 2023-01-23 13:23:07.971465: step: 1518/1541, loss: 0.02686968445777893 2023-01-23 13:23:08.879121: step: 1520/1541, loss: 0.08735917508602142 2023-01-23 13:23:09.787274: step: 1522/1541, loss: 0.42076191306114197 2023-01-23 13:23:10.722359: step: 1524/1541, loss: 0.2880915701389313 2023-01-23 13:23:11.579762: step: 1526/1541, loss: 0.17704807221889496 2023-01-23 13:23:12.513557: step: 1528/1541, loss: 0.0006639892817474902 2023-01-23 13:23:13.389221: step: 1530/1541, loss: 0.0007453273283317685 2023-01-23 13:23:14.270754: step: 1532/1541, loss: 0.28301742672920227 2023-01-23 13:23:15.164940: step: 1534/1541, loss: 0.2138848602771759 2023-01-23 13:23:16.064518: step: 1536/1541, loss: 0.09730264544487 2023-01-23 13:23:16.961631: step: 1538/1541, loss: 0.031743619590997696 2023-01-23 13:23:17.880529: step: 1540/1541, loss: 0.06503773480653763 2023-01-23 13:23:18.861867: step: 1542/1541, loss: 0.07652998715639114 2023-01-23 13:23:19.788261: step: 1544/1541, loss: 0.1107519268989563 2023-01-23 13:23:20.702273: step: 1546/1541, loss: 0.035048313438892365 2023-01-23 13:23:21.583285: step: 1548/1541, loss: 0.6050601601600647 2023-01-23 13:23:22.512322: step: 1550/1541, loss: 0.09597384929656982 2023-01-23 13:23:23.384124: step: 1552/1541, loss: 0.1318424493074417 2023-01-23 13:23:24.299039: step: 1554/1541, loss: 0.013941655866801739 2023-01-23 13:23:25.139295: step: 1556/1541, loss: 0.03129700943827629 2023-01-23 13:23:25.987320: step: 1558/1541, loss: 0.04883868992328644 2023-01-23 13:23:26.859308: step: 1560/1541, loss: 0.07244002819061279 2023-01-23 13:23:27.753434: step: 1562/1541, loss: 0.07828481495380402 2023-01-23 13:23:28.667302: step: 1564/1541, loss: 0.026310957968235016 2023-01-23 13:23:29.605007: step: 1566/1541, loss: 0.13034817576408386 2023-01-23 13:23:30.498964: step: 1568/1541, loss: 0.196660116314888 2023-01-23 13:23:31.431384: step: 1570/1541, loss: 0.050312213599681854 2023-01-23 13:23:32.379531: step: 1572/1541, loss: 0.4088444113731384 2023-01-23 13:23:33.291690: step: 1574/1541, loss: 0.12109736353158951 2023-01-23 13:23:34.168834: step: 1576/1541, loss: 0.05840281397104263 2023-01-23 13:23:35.096908: step: 1578/1541, loss: 0.01227295957505703 2023-01-23 13:23:36.003660: step: 1580/1541, loss: 0.3431071639060974 2023-01-23 13:23:36.868679: step: 1582/1541, loss: 0.018882209435105324 2023-01-23 13:23:37.802442: step: 1584/1541, loss: 0.025753211230039597 2023-01-23 13:23:38.725401: step: 1586/1541, loss: 0.22824259102344513 2023-01-23 13:23:39.700048: step: 1588/1541, loss: 0.09018564969301224 2023-01-23 13:23:40.595084: step: 1590/1541, loss: 0.057785023003816605 2023-01-23 13:23:41.496481: step: 1592/1541, loss: 1.2666773796081543 2023-01-23 13:23:42.375913: step: 1594/1541, loss: 0.03368164971470833 2023-01-23 13:23:43.267638: step: 1596/1541, loss: 0.39614495635032654 2023-01-23 13:23:44.136571: step: 1598/1541, loss: 0.12547621130943298 2023-01-23 13:23:45.034464: step: 1600/1541, loss: 0.1747778207063675 2023-01-23 13:23:45.936430: step: 1602/1541, loss: 0.011222116649150848 2023-01-23 13:23:46.835618: step: 1604/1541, loss: 0.269514799118042 2023-01-23 13:23:47.689342: step: 1606/1541, loss: 0.005298818927258253 2023-01-23 13:23:48.564794: step: 1608/1541, loss: 0.01053615566343069 2023-01-23 13:23:49.431900: step: 1610/1541, loss: 0.03310626745223999 2023-01-23 13:23:50.413377: step: 1612/1541, loss: 0.029398230835795403 2023-01-23 13:23:51.335758: step: 1614/1541, loss: 0.03064156323671341 2023-01-23 13:23:52.227839: step: 1616/1541, loss: 0.027616068720817566 2023-01-23 13:23:53.099628: step: 1618/1541, loss: 0.039557795971632004 2023-01-23 13:23:53.992127: step: 1620/1541, loss: 0.12447790801525116 2023-01-23 13:23:54.960450: step: 1622/1541, loss: 0.0009458402637392282 2023-01-23 13:23:55.931124: step: 1624/1541, loss: 0.03126974031329155 2023-01-23 13:23:56.822466: step: 1626/1541, loss: 0.11707887798547745 2023-01-23 13:23:57.766172: step: 1628/1541, loss: 0.009751810692250729 2023-01-23 13:23:58.731673: step: 1630/1541, loss: 0.031382620334625244 2023-01-23 13:23:59.630831: step: 1632/1541, loss: 0.11442345380783081 2023-01-23 13:24:00.492750: step: 1634/1541, loss: 0.1979667693376541 2023-01-23 13:24:01.399748: step: 1636/1541, loss: 0.02496139518916607 2023-01-23 13:24:02.256789: step: 1638/1541, loss: 0.002578835468739271 2023-01-23 13:24:03.131867: step: 1640/1541, loss: 0.000655827228911221 2023-01-23 13:24:04.077710: step: 1642/1541, loss: 0.0007081179646775126 2023-01-23 13:24:04.970394: step: 1644/1541, loss: 0.00940034817904234 2023-01-23 13:24:05.831773: step: 1646/1541, loss: 1.055717945098877 2023-01-23 13:24:06.730409: step: 1648/1541, loss: 0.13421477377414703 2023-01-23 13:24:07.629015: step: 1650/1541, loss: 0.0002499611000530422 2023-01-23 13:24:08.548399: step: 1652/1541, loss: 0.10544562339782715 2023-01-23 13:24:09.458536: step: 1654/1541, loss: 0.0035549006424844265 2023-01-23 13:24:10.349609: step: 1656/1541, loss: 0.0017767488025128841 2023-01-23 13:24:11.219109: step: 1658/1541, loss: 0.021132808178663254 2023-01-23 13:24:12.087010: step: 1660/1541, loss: 0.08749856054782867 2023-01-23 13:24:13.025572: step: 1662/1541, loss: 0.1743752658367157 2023-01-23 13:24:13.951889: step: 1664/1541, loss: 0.002000597305595875 2023-01-23 13:24:14.848936: step: 1666/1541, loss: 1.1773741245269775 2023-01-23 13:24:15.731555: step: 1668/1541, loss: 0.006405658554285765 2023-01-23 13:24:16.713799: step: 1670/1541, loss: 0.18150776624679565 2023-01-23 13:24:17.626605: step: 1672/1541, loss: 0.003835328621789813 2023-01-23 13:24:18.508958: step: 1674/1541, loss: 0.037487294524908066 2023-01-23 13:24:19.454296: step: 1676/1541, loss: 0.04651595652103424 2023-01-23 13:24:20.318215: step: 1678/1541, loss: 0.011551778763532639 2023-01-23 13:24:21.224204: step: 1680/1541, loss: 0.06585836410522461 2023-01-23 13:24:22.072605: step: 1682/1541, loss: 0.2069855034351349 2023-01-23 13:24:22.950374: step: 1684/1541, loss: 0.24154549837112427 2023-01-23 13:24:23.841935: step: 1686/1541, loss: 0.06843435764312744 2023-01-23 13:24:24.759071: step: 1688/1541, loss: 0.11918547004461288 2023-01-23 13:24:25.638461: step: 1690/1541, loss: 0.06794002652168274 2023-01-23 13:24:26.513659: step: 1692/1541, loss: 0.024245351552963257 2023-01-23 13:24:27.373328: step: 1694/1541, loss: 0.041559863835573196 2023-01-23 13:24:28.321488: step: 1696/1541, loss: 0.01163859199732542 2023-01-23 13:24:29.179289: step: 1698/1541, loss: 0.4034099876880646 2023-01-23 13:24:30.071226: step: 1700/1541, loss: 0.00840966310352087 2023-01-23 13:24:30.992411: step: 1702/1541, loss: 0.2351929247379303 2023-01-23 13:24:32.006764: step: 1704/1541, loss: 0.06632445752620697 2023-01-23 13:24:32.920802: step: 1706/1541, loss: 0.3850589394569397 2023-01-23 13:24:33.879345: step: 1708/1541, loss: 0.12318325787782669 2023-01-23 13:24:34.780862: step: 1710/1541, loss: 0.018900850787758827 2023-01-23 13:24:35.655361: step: 1712/1541, loss: 1.061939001083374 2023-01-23 13:24:36.548641: step: 1714/1541, loss: 0.18305809795856476 2023-01-23 13:24:37.428884: step: 1716/1541, loss: 0.00702032120898366 2023-01-23 13:24:38.353877: step: 1718/1541, loss: 0.11066678911447525 2023-01-23 13:24:39.341166: step: 1720/1541, loss: 0.23873333632946014 2023-01-23 13:24:40.248566: step: 1722/1541, loss: 0.04485991969704628 2023-01-23 13:24:41.121081: step: 1724/1541, loss: 0.006006645038723946 2023-01-23 13:24:42.003132: step: 1726/1541, loss: 0.04659740999341011 2023-01-23 13:24:42.877492: step: 1728/1541, loss: 0.09652192890644073 2023-01-23 13:24:43.787214: step: 1730/1541, loss: 0.0031016406137496233 2023-01-23 13:24:44.702614: step: 1732/1541, loss: 0.09197567403316498 2023-01-23 13:24:45.619132: step: 1734/1541, loss: 0.030458584427833557 2023-01-23 13:24:46.487162: step: 1736/1541, loss: 0.6071921586990356 2023-01-23 13:24:47.381386: step: 1738/1541, loss: 0.05043993890285492 2023-01-23 13:24:48.249870: step: 1740/1541, loss: 0.033310167491436005 2023-01-23 13:24:49.158477: step: 1742/1541, loss: 0.017187828198075294 2023-01-23 13:24:50.089539: step: 1744/1541, loss: 0.018143214285373688 2023-01-23 13:24:50.972545: step: 1746/1541, loss: 0.1463497132062912 2023-01-23 13:24:51.826271: step: 1748/1541, loss: 0.0162052009254694 2023-01-23 13:24:52.701399: step: 1750/1541, loss: 0.00663332361727953 2023-01-23 13:24:53.625968: step: 1752/1541, loss: 0.022903023287653923 2023-01-23 13:24:54.572688: step: 1754/1541, loss: 0.05971274897456169 2023-01-23 13:24:55.556599: step: 1756/1541, loss: 0.02112996019423008 2023-01-23 13:24:56.457883: step: 1758/1541, loss: 0.05142147094011307 2023-01-23 13:24:57.381368: step: 1760/1541, loss: 0.019559312611818314 2023-01-23 13:24:58.306315: step: 1762/1541, loss: 0.014085864648222923 2023-01-23 13:24:59.236257: step: 1764/1541, loss: 0.03774266317486763 2023-01-23 13:25:00.169632: step: 1766/1541, loss: 0.0014095602091401815 2023-01-23 13:25:01.023689: step: 1768/1541, loss: 0.0046224710531532764 2023-01-23 13:25:01.948045: step: 1770/1541, loss: 0.05709097906947136 2023-01-23 13:25:02.845385: step: 1772/1541, loss: 0.14365237951278687 2023-01-23 13:25:03.759257: step: 1774/1541, loss: 0.007255307864397764 2023-01-23 13:25:04.615210: step: 1776/1541, loss: 0.4316622018814087 2023-01-23 13:25:05.546845: step: 1778/1541, loss: 0.002310684882104397 2023-01-23 13:25:06.436096: step: 1780/1541, loss: 0.05728807672858238 2023-01-23 13:25:07.455984: step: 1782/1541, loss: 0.0013380537275224924 2023-01-23 13:25:08.336401: step: 1784/1541, loss: 0.03123779222369194 2023-01-23 13:25:09.232693: step: 1786/1541, loss: 0.14883384108543396 2023-01-23 13:25:10.136262: step: 1788/1541, loss: 0.007014410570263863 2023-01-23 13:25:11.044958: step: 1790/1541, loss: 0.048258837312459946 2023-01-23 13:25:11.955395: step: 1792/1541, loss: 0.4595321714878082 2023-01-23 13:25:12.854042: step: 1794/1541, loss: 0.16304494440555573 2023-01-23 13:25:13.734642: step: 1796/1541, loss: 0.038212284445762634 2023-01-23 13:25:14.624900: step: 1798/1541, loss: 0.33114686608314514 2023-01-23 13:25:15.527709: step: 1800/1541, loss: 0.028797922655940056 2023-01-23 13:25:16.440160: step: 1802/1541, loss: 0.013587121851742268 2023-01-23 13:25:17.367467: step: 1804/1541, loss: 0.12941698729991913 2023-01-23 13:25:18.228165: step: 1806/1541, loss: 0.043225519359111786 2023-01-23 13:25:19.133629: step: 1808/1541, loss: 0.12364441156387329 2023-01-23 13:25:19.999635: step: 1810/1541, loss: 0.025515075773000717 2023-01-23 13:25:20.877365: step: 1812/1541, loss: 0.19788098335266113 2023-01-23 13:25:21.772854: step: 1814/1541, loss: 0.05030180513858795 2023-01-23 13:25:22.716369: step: 1816/1541, loss: 1.1554332971572876 2023-01-23 13:25:23.680948: step: 1818/1541, loss: 0.776170015335083 2023-01-23 13:25:24.565734: step: 1820/1541, loss: 0.07690556347370148 2023-01-23 13:25:25.494055: step: 1822/1541, loss: 0.026847660541534424 2023-01-23 13:25:26.377296: step: 1824/1541, loss: 0.0006014914251863956 2023-01-23 13:25:27.282278: step: 1826/1541, loss: 0.18264970183372498 2023-01-23 13:25:28.161291: step: 1828/1541, loss: 0.010939240455627441 2023-01-23 13:25:29.034672: step: 1830/1541, loss: 0.120020292699337 2023-01-23 13:25:29.923652: step: 1832/1541, loss: 0.024223491549491882 2023-01-23 13:25:30.803469: step: 1834/1541, loss: 0.28911077976226807 2023-01-23 13:25:31.727059: step: 1836/1541, loss: 0.009316226467490196 2023-01-23 13:25:32.639677: step: 1838/1541, loss: 0.5497550964355469 2023-01-23 13:25:33.543036: step: 1840/1541, loss: 0.11335516721010208 2023-01-23 13:25:34.435972: step: 1842/1541, loss: 0.09636427462100983 2023-01-23 13:25:35.343782: step: 1844/1541, loss: 0.003273461712524295 2023-01-23 13:25:36.200533: step: 1846/1541, loss: 0.13299480080604553 2023-01-23 13:25:37.109233: step: 1848/1541, loss: 0.25456586480140686 2023-01-23 13:25:38.004260: step: 1850/1541, loss: 0.03686009347438812 2023-01-23 13:25:38.865547: step: 1852/1541, loss: 0.007374418433755636 2023-01-23 13:25:39.705999: step: 1854/1541, loss: 0.04305959492921829 2023-01-23 13:25:40.582961: step: 1856/1541, loss: 0.0333915539085865 2023-01-23 13:25:41.481454: step: 1858/1541, loss: 0.006233341991901398 2023-01-23 13:25:42.380467: step: 1860/1541, loss: 0.0007086943369358778 2023-01-23 13:25:43.287649: step: 1862/1541, loss: 0.05746805667877197 2023-01-23 13:25:44.167347: step: 1864/1541, loss: 0.0015492524253204465 2023-01-23 13:25:45.046486: step: 1866/1541, loss: 0.151303231716156 2023-01-23 13:25:46.018498: step: 1868/1541, loss: 0.20371031761169434 2023-01-23 13:25:46.895810: step: 1870/1541, loss: 0.0010191034525632858 2023-01-23 13:25:47.772536: step: 1872/1541, loss: 0.24180778861045837 2023-01-23 13:25:48.644904: step: 1874/1541, loss: 0.0008536901441402733 2023-01-23 13:25:49.521331: step: 1876/1541, loss: 0.5580377578735352 2023-01-23 13:25:50.430898: step: 1878/1541, loss: 0.019763028249144554 2023-01-23 13:25:51.346086: step: 1880/1541, loss: 0.03529159352183342 2023-01-23 13:25:52.241227: step: 1882/1541, loss: 0.39040470123291016 2023-01-23 13:25:53.134877: step: 1884/1541, loss: 0.5459210872650146 2023-01-23 13:25:54.019042: step: 1886/1541, loss: 0.08868534862995148 2023-01-23 13:25:54.933049: step: 1888/1541, loss: 0.28224465250968933 2023-01-23 13:25:55.843019: step: 1890/1541, loss: 0.027448559179902077 2023-01-23 13:25:56.814706: step: 1892/1541, loss: 0.17192642390727997 2023-01-23 13:25:57.688807: step: 1894/1541, loss: 0.1079803854227066 2023-01-23 13:25:58.563477: step: 1896/1541, loss: 0.0038767801597714424 2023-01-23 13:25:59.434972: step: 1898/1541, loss: 0.05553476884961128 2023-01-23 13:26:00.334327: step: 1900/1541, loss: 0.014929691329598427 2023-01-23 13:26:01.222512: step: 1902/1541, loss: 0.0018101413734257221 2023-01-23 13:26:02.163178: step: 1904/1541, loss: 0.0016258335672318935 2023-01-23 13:26:03.034654: step: 1906/1541, loss: 0.16499347984790802 2023-01-23 13:26:03.913583: step: 1908/1541, loss: 0.006354386918246746 2023-01-23 13:26:04.754815: step: 1910/1541, loss: 0.01860087364912033 2023-01-23 13:26:05.640803: step: 1912/1541, loss: 0.005806013010442257 2023-01-23 13:26:06.496223: step: 1914/1541, loss: 0.011498061008751392 2023-01-23 13:26:07.407037: step: 1916/1541, loss: 0.29835373163223267 2023-01-23 13:26:08.289021: step: 1918/1541, loss: 0.006781705655157566 2023-01-23 13:26:09.146133: step: 1920/1541, loss: 0.13699986040592194 2023-01-23 13:26:10.214817: step: 1922/1541, loss: 0.0022376354318112135 2023-01-23 13:26:11.123029: step: 1924/1541, loss: 0.2220495045185089 2023-01-23 13:26:11.997638: step: 1926/1541, loss: 0.42349785566329956 2023-01-23 13:26:12.865981: step: 1928/1541, loss: 0.0012168074026703835 2023-01-23 13:26:13.776800: step: 1930/1541, loss: 0.00347571331076324 2023-01-23 13:26:14.719528: step: 1932/1541, loss: 0.04742145538330078 2023-01-23 13:26:15.588416: step: 1934/1541, loss: 0.03007023222744465 2023-01-23 13:26:16.497763: step: 1936/1541, loss: 0.008246514946222305 2023-01-23 13:26:17.359427: step: 1938/1541, loss: 0.01878410018980503 2023-01-23 13:26:18.253126: step: 1940/1541, loss: 0.008124271407723427 2023-01-23 13:26:19.161461: step: 1942/1541, loss: 0.01940893568098545 2023-01-23 13:26:20.041996: step: 1944/1541, loss: 0.03685691952705383 2023-01-23 13:26:21.020294: step: 1946/1541, loss: 0.002839495660737157 2023-01-23 13:26:21.924390: step: 1948/1541, loss: 0.0008630827069282532 2023-01-23 13:26:22.904877: step: 1950/1541, loss: 0.2205807864665985 2023-01-23 13:26:23.826746: step: 1952/1541, loss: 0.022276587784290314 2023-01-23 13:26:24.712742: step: 1954/1541, loss: 0.0005457100342027843 2023-01-23 13:26:25.613274: step: 1956/1541, loss: 0.03205416351556778 2023-01-23 13:26:26.479695: step: 1958/1541, loss: 0.02198505774140358 2023-01-23 13:26:27.365475: step: 1960/1541, loss: 0.0022154655307531357 2023-01-23 13:26:28.238331: step: 1962/1541, loss: 0.14355024695396423 2023-01-23 13:26:29.097241: step: 1964/1541, loss: 0.23064979910850525 2023-01-23 13:26:29.940760: step: 1966/1541, loss: 0.06151990219950676 2023-01-23 13:26:30.852879: step: 1968/1541, loss: 0.056682080030441284 2023-01-23 13:26:31.739979: step: 1970/1541, loss: 0.046583279967308044 2023-01-23 13:26:32.598773: step: 1972/1541, loss: 0.2992234230041504 2023-01-23 13:26:33.488796: step: 1974/1541, loss: 0.006284581031650305 2023-01-23 13:26:34.415710: step: 1976/1541, loss: 0.3493778705596924 2023-01-23 13:26:35.318144: step: 1978/1541, loss: 0.004898838698863983 2023-01-23 13:26:36.203407: step: 1980/1541, loss: 0.09626460075378418 2023-01-23 13:26:37.112451: step: 1982/1541, loss: 0.13198444247245789 2023-01-23 13:26:38.001741: step: 1984/1541, loss: 0.0806054174900055 2023-01-23 13:26:38.860529: step: 1986/1541, loss: 0.0022908318787813187 2023-01-23 13:26:39.806123: step: 1988/1541, loss: 0.01948918029665947 2023-01-23 13:26:40.682570: step: 1990/1541, loss: 0.23889119923114777 2023-01-23 13:26:41.566766: step: 1992/1541, loss: 0.003531662281602621 2023-01-23 13:26:42.471995: step: 1994/1541, loss: 0.0018165489891543984 2023-01-23 13:26:43.371187: step: 1996/1541, loss: 0.04493894428014755 2023-01-23 13:26:44.290190: step: 1998/1541, loss: 0.21846380829811096 2023-01-23 13:26:45.293553: step: 2000/1541, loss: 0.0032192510552704334 2023-01-23 13:26:46.276987: step: 2002/1541, loss: 0.03925346955657005 2023-01-23 13:26:47.190038: step: 2004/1541, loss: 0.1605617254972458 2023-01-23 13:26:48.092518: step: 2006/1541, loss: 0.10999435186386108 2023-01-23 13:26:48.976099: step: 2008/1541, loss: 0.06340859830379486 2023-01-23 13:26:49.866791: step: 2010/1541, loss: 0.09505012631416321 2023-01-23 13:26:50.739545: step: 2012/1541, loss: 0.012742365710437298 2023-01-23 13:26:51.594469: step: 2014/1541, loss: 0.47829610109329224 2023-01-23 13:26:52.478151: step: 2016/1541, loss: 0.040780771523714066 2023-01-23 13:26:53.376956: step: 2018/1541, loss: 0.6558758616447449 2023-01-23 13:26:54.269010: step: 2020/1541, loss: 0.012601961381733418 2023-01-23 13:26:55.180501: step: 2022/1541, loss: 0.07823481410741806 2023-01-23 13:26:56.095087: step: 2024/1541, loss: 0.2165614664554596 2023-01-23 13:26:57.008992: step: 2026/1541, loss: 0.4073321521282196 2023-01-23 13:26:57.927293: step: 2028/1541, loss: 0.018045589327812195 2023-01-23 13:26:58.840457: step: 2030/1541, loss: 0.019175522029399872 2023-01-23 13:26:59.755701: step: 2032/1541, loss: 0.06998036801815033 2023-01-23 13:27:00.655768: step: 2034/1541, loss: 0.470512717962265 2023-01-23 13:27:01.664210: step: 2036/1541, loss: 0.4633577764034271 2023-01-23 13:27:02.558361: step: 2038/1541, loss: 0.024141257628798485 2023-01-23 13:27:03.422750: step: 2040/1541, loss: 0.011088556610047817 2023-01-23 13:27:04.298079: step: 2042/1541, loss: 0.028711015358567238 2023-01-23 13:27:05.229514: step: 2044/1541, loss: 0.10333414375782013 2023-01-23 13:27:06.149041: step: 2046/1541, loss: 1.1348230838775635 2023-01-23 13:27:07.030217: step: 2048/1541, loss: 0.27009472250938416 2023-01-23 13:27:07.930317: step: 2050/1541, loss: 0.023456472903490067 2023-01-23 13:27:08.839906: step: 2052/1541, loss: 0.010312872007489204 2023-01-23 13:27:09.735425: step: 2054/1541, loss: 0.14601872861385345 2023-01-23 13:27:10.579304: step: 2056/1541, loss: 0.05023409426212311 2023-01-23 13:27:11.511763: step: 2058/1541, loss: 0.08132089674472809 2023-01-23 13:27:12.417468: step: 2060/1541, loss: 0.02818399854004383 2023-01-23 13:27:13.284344: step: 2062/1541, loss: 0.022280313074588776 2023-01-23 13:27:14.139101: step: 2064/1541, loss: 0.36210525035858154 2023-01-23 13:27:15.036699: step: 2066/1541, loss: 0.05196067690849304 2023-01-23 13:27:15.959953: step: 2068/1541, loss: 0.08693350106477737 2023-01-23 13:27:16.822517: step: 2070/1541, loss: 0.38910168409347534 2023-01-23 13:27:17.725162: step: 2072/1541, loss: 0.08862853050231934 2023-01-23 13:27:18.604251: step: 2074/1541, loss: 0.45206552743911743 2023-01-23 13:27:19.536405: step: 2076/1541, loss: 0.3045422434806824 2023-01-23 13:27:20.451163: step: 2078/1541, loss: 0.06570036709308624 2023-01-23 13:27:21.324888: step: 2080/1541, loss: 0.05532398074865341 2023-01-23 13:27:22.262813: step: 2082/1541, loss: 0.03425411880016327 2023-01-23 13:27:23.176673: step: 2084/1541, loss: 0.014657450839877129 2023-01-23 13:27:24.057814: step: 2086/1541, loss: 0.03236192837357521 2023-01-23 13:27:24.952919: step: 2088/1541, loss: 0.005954239517450333 2023-01-23 13:27:25.895436: step: 2090/1541, loss: 0.051141541451215744 2023-01-23 13:27:26.751016: step: 2092/1541, loss: 0.10366813093423843 2023-01-23 13:27:27.643562: step: 2094/1541, loss: 0.005312520544975996 2023-01-23 13:27:28.525864: step: 2096/1541, loss: 0.042210668325424194 2023-01-23 13:27:29.415486: step: 2098/1541, loss: 0.007967500947415829 2023-01-23 13:27:30.335544: step: 2100/1541, loss: 0.021221445873379707 2023-01-23 13:27:31.271740: step: 2102/1541, loss: 0.015817079693078995 2023-01-23 13:27:32.226498: step: 2104/1541, loss: 0.18760699033737183 2023-01-23 13:27:33.149025: step: 2106/1541, loss: 0.004664067178964615 2023-01-23 13:27:34.033917: step: 2108/1541, loss: 0.516364336013794 2023-01-23 13:27:34.907130: step: 2110/1541, loss: 0.21005834639072418 2023-01-23 13:27:35.793035: step: 2112/1541, loss: 0.006554937455803156 2023-01-23 13:27:36.683203: step: 2114/1541, loss: 0.03339575603604317 2023-01-23 13:27:37.589331: step: 2116/1541, loss: 0.3264215886592865 2023-01-23 13:27:38.527552: step: 2118/1541, loss: 0.0838550478219986 2023-01-23 13:27:39.421794: step: 2120/1541, loss: 0.3179932236671448 2023-01-23 13:27:40.339438: step: 2122/1541, loss: 0.4633437991142273 2023-01-23 13:27:41.268353: step: 2124/1541, loss: 0.00481631513684988 2023-01-23 13:27:42.166477: step: 2126/1541, loss: 0.006407633423805237 2023-01-23 13:27:43.103810: step: 2128/1541, loss: 0.07116030901670456 2023-01-23 13:27:44.020593: step: 2130/1541, loss: 0.16681306064128876 2023-01-23 13:27:44.936631: step: 2132/1541, loss: 0.016789143905043602 2023-01-23 13:27:45.854526: step: 2134/1541, loss: 0.06610868126153946 2023-01-23 13:27:46.721988: step: 2136/1541, loss: 0.05957185849547386 2023-01-23 13:27:47.618113: step: 2138/1541, loss: 0.002093194518238306 2023-01-23 13:27:48.497478: step: 2140/1541, loss: 0.019982947036623955 2023-01-23 13:27:49.382345: step: 2142/1541, loss: 0.003470461815595627 2023-01-23 13:27:50.262048: step: 2144/1541, loss: 0.001757838879711926 2023-01-23 13:27:51.165016: step: 2146/1541, loss: 0.006416101939976215 2023-01-23 13:27:52.013557: step: 2148/1541, loss: 0.328772634267807 2023-01-23 13:27:52.944350: step: 2150/1541, loss: 0.22291071712970734 2023-01-23 13:27:53.831108: step: 2152/1541, loss: 0.06180936098098755 2023-01-23 13:27:54.706991: step: 2154/1541, loss: 0.20031549036502838 2023-01-23 13:27:55.588055: step: 2156/1541, loss: 0.036294564604759216 2023-01-23 13:27:56.512581: step: 2158/1541, loss: 0.0012320717796683311 2023-01-23 13:27:57.395537: step: 2160/1541, loss: 0.013097759336233139 2023-01-23 13:27:58.277117: step: 2162/1541, loss: 0.0455588661134243 2023-01-23 13:27:59.115186: step: 2164/1541, loss: 0.3722847104072571 2023-01-23 13:27:59.959555: step: 2166/1541, loss: 0.1959984451532364 2023-01-23 13:28:00.781956: step: 2168/1541, loss: 0.1853993684053421 2023-01-23 13:28:01.665547: step: 2170/1541, loss: 0.0054342979565262794 2023-01-23 13:28:02.595582: step: 2172/1541, loss: 0.0028305065352469683 2023-01-23 13:28:03.448574: step: 2174/1541, loss: 0.1514509916305542 2023-01-23 13:28:04.314264: step: 2176/1541, loss: 0.09143413603305817 2023-01-23 13:28:05.247554: step: 2178/1541, loss: 0.004349726717919111 2023-01-23 13:28:06.129382: step: 2180/1541, loss: 0.014022013172507286 2023-01-23 13:28:07.001706: step: 2182/1541, loss: 0.05419051647186279 2023-01-23 13:28:07.871020: step: 2184/1541, loss: 0.002475922228768468 2023-01-23 13:28:08.795894: step: 2186/1541, loss: 0.07532675564289093 2023-01-23 13:28:09.718933: step: 2188/1541, loss: 0.002226933604106307 2023-01-23 13:28:10.620853: step: 2190/1541, loss: 0.2862243354320526 2023-01-23 13:28:11.483572: step: 2192/1541, loss: 0.5392299294471741 2023-01-23 13:28:12.405329: step: 2194/1541, loss: 0.22322161495685577 2023-01-23 13:28:13.323794: step: 2196/1541, loss: 0.02037070319056511 2023-01-23 13:28:14.211284: step: 2198/1541, loss: 0.006261641159653664 2023-01-23 13:28:15.116928: step: 2200/1541, loss: 0.3749578595161438 2023-01-23 13:28:15.945580: step: 2202/1541, loss: 0.5547122359275818 2023-01-23 13:28:16.824412: step: 2204/1541, loss: 0.0024629877880215645 2023-01-23 13:28:17.844663: step: 2206/1541, loss: 0.0005127150798216462 2023-01-23 13:28:18.726347: step: 2208/1541, loss: 0.19475732743740082 2023-01-23 13:28:19.605143: step: 2210/1541, loss: 0.1773599088191986 2023-01-23 13:28:20.523996: step: 2212/1541, loss: 0.30458420515060425 2023-01-23 13:28:21.404372: step: 2214/1541, loss: 0.20587673783302307 2023-01-23 13:28:22.289701: step: 2216/1541, loss: 0.01193096861243248 2023-01-23 13:28:23.260873: step: 2218/1541, loss: 0.01018901914358139 2023-01-23 13:28:24.133486: step: 2220/1541, loss: 0.5399552583694458 2023-01-23 13:28:25.030247: step: 2222/1541, loss: 0.10864483565092087 2023-01-23 13:28:25.899894: step: 2224/1541, loss: 0.0575905479490757 2023-01-23 13:28:26.812291: step: 2226/1541, loss: 0.0766119584441185 2023-01-23 13:28:27.677810: step: 2228/1541, loss: 0.23566968739032745 2023-01-23 13:28:28.575859: step: 2230/1541, loss: 0.12995363771915436 2023-01-23 13:28:29.496301: step: 2232/1541, loss: 0.02403651736676693 2023-01-23 13:28:30.402722: step: 2234/1541, loss: 0.09264495968818665 2023-01-23 13:28:31.301970: step: 2236/1541, loss: 0.06560436636209488 2023-01-23 13:28:32.194301: step: 2238/1541, loss: 0.25115767121315 2023-01-23 13:28:33.052905: step: 2240/1541, loss: 0.02882031351327896 2023-01-23 13:28:33.978201: step: 2242/1541, loss: 0.8131330013275146 2023-01-23 13:28:34.875224: step: 2244/1541, loss: 0.6434248089790344 2023-01-23 13:28:35.749321: step: 2246/1541, loss: 0.2237693965435028 2023-01-23 13:28:36.605147: step: 2248/1541, loss: 0.008929814212024212 2023-01-23 13:28:37.509462: step: 2250/1541, loss: 0.13047313690185547 2023-01-23 13:28:38.466366: step: 2252/1541, loss: 0.051993656903505325 2023-01-23 13:28:39.410384: step: 2254/1541, loss: 0.25397977232933044 2023-01-23 13:28:40.333219: step: 2256/1541, loss: 0.08363661170005798 2023-01-23 13:28:41.214533: step: 2258/1541, loss: 0.04451102390885353 2023-01-23 13:28:42.093395: step: 2260/1541, loss: 0.004281326197087765 2023-01-23 13:28:42.993254: step: 2262/1541, loss: 0.18104605376720428 2023-01-23 13:28:43.895854: step: 2264/1541, loss: 0.030841387808322906 2023-01-23 13:28:44.808786: step: 2266/1541, loss: 0.5882239937782288 2023-01-23 13:28:45.728959: step: 2268/1541, loss: 0.0011730717960745096 2023-01-23 13:28:46.668795: step: 2270/1541, loss: 0.006608176045119762 2023-01-23 13:28:47.538214: step: 2272/1541, loss: 0.01970859430730343 2023-01-23 13:28:48.426950: step: 2274/1541, loss: 0.007811482530087233 2023-01-23 13:28:49.339904: step: 2276/1541, loss: 0.0013697381364181638 2023-01-23 13:28:50.191514: step: 2278/1541, loss: 0.31655576825141907 2023-01-23 13:28:51.165977: step: 2280/1541, loss: 0.03242190182209015 2023-01-23 13:28:52.032280: step: 2282/1541, loss: 0.2080167531967163 2023-01-23 13:28:53.008983: step: 2284/1541, loss: 0.08916722238063812 2023-01-23 13:28:53.938823: step: 2286/1541, loss: 0.011364754289388657 2023-01-23 13:28:54.841197: step: 2288/1541, loss: 0.14338868856430054 2023-01-23 13:28:55.726825: step: 2290/1541, loss: 0.22651322185993195 2023-01-23 13:28:56.664978: step: 2292/1541, loss: 0.15899036824703217 2023-01-23 13:28:57.587692: step: 2294/1541, loss: 0.22446639835834503 2023-01-23 13:28:58.525604: step: 2296/1541, loss: 0.024213340133428574 2023-01-23 13:28:59.410466: step: 2298/1541, loss: 0.024199699983000755 2023-01-23 13:29:00.290415: step: 2300/1541, loss: 0.03642839193344116 2023-01-23 13:29:01.162509: step: 2302/1541, loss: 0.018485482782125473 2023-01-23 13:29:02.075542: step: 2304/1541, loss: 0.017439819872379303 2023-01-23 13:29:03.019611: step: 2306/1541, loss: 0.0015901490114629269 2023-01-23 13:29:03.910400: step: 2308/1541, loss: 0.0008788778795860708 2023-01-23 13:29:04.813458: step: 2310/1541, loss: 0.07079044729471207 2023-01-23 13:29:05.717375: step: 2312/1541, loss: 0.018854452297091484 2023-01-23 13:29:06.654290: step: 2314/1541, loss: 0.16202278435230255 2023-01-23 13:29:07.604029: step: 2316/1541, loss: 0.0014102348359301686 2023-01-23 13:29:08.536813: step: 2318/1541, loss: 0.9330284595489502 2023-01-23 13:29:09.532490: step: 2320/1541, loss: 0.01880178600549698 2023-01-23 13:29:10.437074: step: 2322/1541, loss: 0.14961379766464233 2023-01-23 13:29:11.348151: step: 2324/1541, loss: 0.20118212699890137 2023-01-23 13:29:12.246763: step: 2326/1541, loss: 0.1637607216835022 2023-01-23 13:29:13.130335: step: 2328/1541, loss: 0.01067230012267828 2023-01-23 13:29:14.035489: step: 2330/1541, loss: 0.15716543793678284 2023-01-23 13:29:14.903523: step: 2332/1541, loss: 0.010143930092453957 2023-01-23 13:29:15.790334: step: 2334/1541, loss: 0.008617300540208817 2023-01-23 13:29:16.719151: step: 2336/1541, loss: 0.009363390505313873 2023-01-23 13:29:17.625286: step: 2338/1541, loss: 0.014176563359797001 2023-01-23 13:29:18.515486: step: 2340/1541, loss: 0.1158350259065628 2023-01-23 13:29:19.384997: step: 2342/1541, loss: 0.026771295815706253 2023-01-23 13:29:20.265886: step: 2344/1541, loss: 0.013148806057870388 2023-01-23 13:29:21.170885: step: 2346/1541, loss: 0.002125679748132825 2023-01-23 13:29:22.050505: step: 2348/1541, loss: 1.5923709869384766 2023-01-23 13:29:22.960948: step: 2350/1541, loss: 0.018935780972242355 2023-01-23 13:29:23.826878: step: 2352/1541, loss: 0.0029710636008530855 2023-01-23 13:29:24.715833: step: 2354/1541, loss: 0.02230776846408844 2023-01-23 13:29:25.612924: step: 2356/1541, loss: 0.03315354511141777 2023-01-23 13:29:26.549093: step: 2358/1541, loss: 0.05520084500312805 2023-01-23 13:29:27.426104: step: 2360/1541, loss: 0.29736578464508057 2023-01-23 13:29:28.304769: step: 2362/1541, loss: 0.0020698353182524443 2023-01-23 13:29:29.201741: step: 2364/1541, loss: 0.03804399445652962 2023-01-23 13:29:30.142944: step: 2366/1541, loss: 0.022582504898309708 2023-01-23 13:29:30.990073: step: 2368/1541, loss: 0.3525781035423279 2023-01-23 13:29:31.899833: step: 2370/1541, loss: 0.01365760900080204 2023-01-23 13:29:32.828613: step: 2372/1541, loss: 0.011044373735785484 2023-01-23 13:29:33.813473: step: 2374/1541, loss: 0.005079344380646944 2023-01-23 13:29:34.685206: step: 2376/1541, loss: 0.12858277559280396 2023-01-23 13:29:35.546083: step: 2378/1541, loss: 0.34010276198387146 2023-01-23 13:29:36.425143: step: 2380/1541, loss: 0.14987888932228088 2023-01-23 13:29:37.382347: step: 2382/1541, loss: 0.26227959990501404 2023-01-23 13:29:38.336822: step: 2384/1541, loss: 0.09278755635023117 2023-01-23 13:29:39.210119: step: 2386/1541, loss: 0.1314513385295868 2023-01-23 13:29:40.126479: step: 2388/1541, loss: 0.001702077453956008 2023-01-23 13:29:41.032545: step: 2390/1541, loss: 0.08105625212192535 2023-01-23 13:29:41.908138: step: 2392/1541, loss: 0.003984313923865557 2023-01-23 13:29:42.814450: step: 2394/1541, loss: 0.00490307342261076 2023-01-23 13:29:43.738126: step: 2396/1541, loss: 0.18454191088676453 2023-01-23 13:29:44.606264: step: 2398/1541, loss: 0.3446226418018341 2023-01-23 13:29:45.490520: step: 2400/1541, loss: 0.0459461510181427 2023-01-23 13:29:46.439267: step: 2402/1541, loss: 0.09742790460586548 2023-01-23 13:29:47.364372: step: 2404/1541, loss: 0.1565696746110916 2023-01-23 13:29:48.280620: step: 2406/1541, loss: 0.01607365719974041 2023-01-23 13:29:49.178808: step: 2408/1541, loss: 0.03871019557118416 2023-01-23 13:29:50.044577: step: 2410/1541, loss: 0.20636184513568878 2023-01-23 13:29:50.968707: step: 2412/1541, loss: 0.13547202944755554 2023-01-23 13:29:51.831846: step: 2414/1541, loss: 0.008186300285160542 2023-01-23 13:29:52.727398: step: 2416/1541, loss: 0.10422055423259735 2023-01-23 13:29:53.637214: step: 2418/1541, loss: 0.03608193248510361 2023-01-23 13:29:54.510895: step: 2420/1541, loss: 0.04762798920273781 2023-01-23 13:29:55.398986: step: 2422/1541, loss: 0.08660177886486053 2023-01-23 13:29:56.281769: step: 2424/1541, loss: 0.35077574849128723 2023-01-23 13:29:57.204277: step: 2426/1541, loss: 0.002884188899770379 2023-01-23 13:29:58.084057: step: 2428/1541, loss: 0.06013417989015579 2023-01-23 13:29:59.091434: step: 2430/1541, loss: 0.37910720705986023 2023-01-23 13:29:59.997015: step: 2432/1541, loss: 0.02306322008371353 2023-01-23 13:30:00.918641: step: 2434/1541, loss: 0.004862906411290169 2023-01-23 13:30:01.862997: step: 2436/1541, loss: 0.04637288302183151 2023-01-23 13:30:02.776019: step: 2438/1541, loss: 0.1415688395500183 2023-01-23 13:30:03.628425: step: 2440/1541, loss: 0.038749855011701584 2023-01-23 13:30:04.537507: step: 2442/1541, loss: 0.0837727040052414 2023-01-23 13:30:05.430278: step: 2444/1541, loss: 0.7703099250793457 2023-01-23 13:30:06.346451: step: 2446/1541, loss: 0.021916337311267853 2023-01-23 13:30:07.199920: step: 2448/1541, loss: 0.027416883036494255 2023-01-23 13:30:08.092534: step: 2450/1541, loss: 0.24409180879592896 2023-01-23 13:30:09.016876: step: 2452/1541, loss: 0.16121794283390045 2023-01-23 13:30:09.962545: step: 2454/1541, loss: 0.028247810900211334 2023-01-23 13:30:10.837922: step: 2456/1541, loss: 0.02623821049928665 2023-01-23 13:30:11.734473: step: 2458/1541, loss: 0.062359798699617386 2023-01-23 13:30:12.612024: step: 2460/1541, loss: 0.0485445037484169 2023-01-23 13:30:13.497235: step: 2462/1541, loss: 0.0075970725156366825 2023-01-23 13:30:14.422956: step: 2464/1541, loss: 0.04659472405910492 2023-01-23 13:30:15.307382: step: 2466/1541, loss: 0.012199008837342262 2023-01-23 13:30:16.182477: step: 2468/1541, loss: 0.05270225182175636 2023-01-23 13:30:17.041009: step: 2470/1541, loss: 0.14117693901062012 2023-01-23 13:30:17.912814: step: 2472/1541, loss: 0.00026658683782443404 2023-01-23 13:30:18.771804: step: 2474/1541, loss: 0.04097289592027664 2023-01-23 13:30:19.713777: step: 2476/1541, loss: 0.03701062500476837 2023-01-23 13:30:20.626476: step: 2478/1541, loss: 0.24372509121894836 2023-01-23 13:30:21.526229: step: 2480/1541, loss: 0.034459132701158524 2023-01-23 13:30:22.375395: step: 2482/1541, loss: 0.08647972345352173 2023-01-23 13:30:23.268021: step: 2484/1541, loss: 0.08255451917648315 2023-01-23 13:30:24.201472: step: 2486/1541, loss: 0.04164716601371765 2023-01-23 13:30:25.123075: step: 2488/1541, loss: 0.4609640836715698 2023-01-23 13:30:26.023704: step: 2490/1541, loss: 0.01758461259305477 2023-01-23 13:30:27.024043: step: 2492/1541, loss: 0.06755983829498291 2023-01-23 13:30:27.931009: step: 2494/1541, loss: 0.026467978954315186 2023-01-23 13:30:28.916202: step: 2496/1541, loss: 0.0027891257777810097 2023-01-23 13:30:29.797425: step: 2498/1541, loss: 1.5713859796524048 2023-01-23 13:30:30.737157: step: 2500/1541, loss: 0.018182653933763504 2023-01-23 13:30:31.625010: step: 2502/1541, loss: 0.052410565316677094 2023-01-23 13:30:32.538795: step: 2504/1541, loss: 0.024285824969410896 2023-01-23 13:30:33.424925: step: 2506/1541, loss: 0.006999323610216379 2023-01-23 13:30:34.322071: step: 2508/1541, loss: 0.15669696033000946 2023-01-23 13:30:35.206015: step: 2510/1541, loss: 0.012733196839690208 2023-01-23 13:30:36.075794: step: 2512/1541, loss: 0.002012951299548149 2023-01-23 13:30:36.932412: step: 2514/1541, loss: 0.0006587515817955136 2023-01-23 13:30:37.824138: step: 2516/1541, loss: 0.061971958726644516 2023-01-23 13:30:38.824097: step: 2518/1541, loss: 0.14210230112075806 2023-01-23 13:30:39.690952: step: 2520/1541, loss: 0.08176276832818985 2023-01-23 13:30:40.576032: step: 2522/1541, loss: 0.06471174210309982 2023-01-23 13:30:41.542461: step: 2524/1541, loss: 0.026291552931070328 2023-01-23 13:30:42.414936: step: 2526/1541, loss: 0.029316371306777 2023-01-23 13:30:43.325005: step: 2528/1541, loss: 0.03934100270271301 2023-01-23 13:30:44.184489: step: 2530/1541, loss: 0.005047784186899662 2023-01-23 13:30:45.089819: step: 2532/1541, loss: 0.01697508431971073 2023-01-23 13:30:46.017433: step: 2534/1541, loss: 0.020926348865032196 2023-01-23 13:30:46.925734: step: 2536/1541, loss: 0.055763114243745804 2023-01-23 13:30:47.821909: step: 2538/1541, loss: 0.028686419129371643 2023-01-23 13:30:48.732971: step: 2540/1541, loss: 0.24609872698783875 2023-01-23 13:30:49.664032: step: 2542/1541, loss: 0.2017723023891449 2023-01-23 13:30:50.601899: step: 2544/1541, loss: 1.0795724391937256 2023-01-23 13:30:51.472890: step: 2546/1541, loss: 0.05231604352593422 2023-01-23 13:30:52.379390: step: 2548/1541, loss: 0.1025046780705452 2023-01-23 13:30:53.258358: step: 2550/1541, loss: 0.2697257399559021 2023-01-23 13:30:54.139326: step: 2552/1541, loss: 0.06311595439910889 2023-01-23 13:30:55.037476: step: 2554/1541, loss: 0.008339512161910534 2023-01-23 13:30:56.010593: step: 2556/1541, loss: 0.0069153765216469765 2023-01-23 13:30:56.975277: step: 2558/1541, loss: 0.06529707461595535 2023-01-23 13:30:57.879615: step: 2560/1541, loss: 0.007409083191305399 2023-01-23 13:30:58.797428: step: 2562/1541, loss: 0.09370359033346176 2023-01-23 13:30:59.690741: step: 2564/1541, loss: 0.07663578540086746 2023-01-23 13:31:00.595246: step: 2566/1541, loss: 0.16484233736991882 2023-01-23 13:31:01.506157: step: 2568/1541, loss: 0.12349309772253036 2023-01-23 13:31:02.388054: step: 2570/1541, loss: 0.1285080462694168 2023-01-23 13:31:03.320992: step: 2572/1541, loss: 0.06057856231927872 2023-01-23 13:31:04.199454: step: 2574/1541, loss: 0.09361010789871216 2023-01-23 13:31:05.074226: step: 2576/1541, loss: 0.6908661723136902 2023-01-23 13:31:05.981188: step: 2578/1541, loss: 0.1587202101945877 2023-01-23 13:31:06.865400: step: 2580/1541, loss: 0.0029562402050942183 2023-01-23 13:31:07.854640: step: 2582/1541, loss: 0.20506247878074646 2023-01-23 13:31:08.780476: step: 2584/1541, loss: 0.09014420211315155 2023-01-23 13:31:09.648078: step: 2586/1541, loss: 0.4451562166213989 2023-01-23 13:31:10.531133: step: 2588/1541, loss: 0.00438456516712904 2023-01-23 13:31:11.407239: step: 2590/1541, loss: 0.20990301668643951 2023-01-23 13:31:12.268183: step: 2592/1541, loss: 0.012914927676320076 2023-01-23 13:31:13.152962: step: 2594/1541, loss: 0.7987510561943054 2023-01-23 13:31:14.070285: step: 2596/1541, loss: 0.05413239449262619 2023-01-23 13:31:14.982542: step: 2598/1541, loss: 0.012664212845265865 2023-01-23 13:31:15.873759: step: 2600/1541, loss: 0.006121294107288122 2023-01-23 13:31:16.765936: step: 2602/1541, loss: 0.022012535482645035 2023-01-23 13:31:17.640383: step: 2604/1541, loss: 0.12063468992710114 2023-01-23 13:31:18.555892: step: 2606/1541, loss: 0.07775156199932098 2023-01-23 13:31:19.483510: step: 2608/1541, loss: 0.10077781230211258 2023-01-23 13:31:20.412166: step: 2610/1541, loss: 0.1820513904094696 2023-01-23 13:31:21.298955: step: 2612/1541, loss: 0.23130597174167633 2023-01-23 13:31:22.157756: step: 2614/1541, loss: 0.006213963031768799 2023-01-23 13:31:23.056731: step: 2616/1541, loss: 0.02160077728331089 2023-01-23 13:31:23.957119: step: 2618/1541, loss: 0.3125333786010742 2023-01-23 13:31:24.856501: step: 2620/1541, loss: 0.006583400536328554 2023-01-23 13:31:25.743156: step: 2622/1541, loss: 0.028764983639121056 2023-01-23 13:31:26.599227: step: 2624/1541, loss: 0.004559632856398821 2023-01-23 13:31:27.473778: step: 2626/1541, loss: 0.21129798889160156 2023-01-23 13:31:28.360145: step: 2628/1541, loss: 0.13572624325752258 2023-01-23 13:31:29.244391: step: 2630/1541, loss: 0.006091154646128416 2023-01-23 13:31:30.133462: step: 2632/1541, loss: 0.052562031894922256 2023-01-23 13:31:31.018230: step: 2634/1541, loss: 0.0215141698718071 2023-01-23 13:31:31.940245: step: 2636/1541, loss: 0.029826125130057335 2023-01-23 13:31:32.830933: step: 2638/1541, loss: 0.09404144436120987 2023-01-23 13:31:33.716967: step: 2640/1541, loss: 0.012569130398333073 2023-01-23 13:31:34.609098: step: 2642/1541, loss: 0.009001344442367554 2023-01-23 13:31:35.503370: step: 2644/1541, loss: 0.10965908318758011 2023-01-23 13:31:36.412654: step: 2646/1541, loss: 0.2572696805000305 2023-01-23 13:31:37.349637: step: 2648/1541, loss: 0.026955487206578255 2023-01-23 13:31:38.235659: step: 2650/1541, loss: 0.007214278448373079 2023-01-23 13:31:39.103788: step: 2652/1541, loss: 0.003010216634720564 2023-01-23 13:31:40.041204: step: 2654/1541, loss: 0.15884366631507874 2023-01-23 13:31:40.964111: step: 2656/1541, loss: 0.11686346679925919 2023-01-23 13:31:41.927711: step: 2658/1541, loss: 0.011275378987193108 2023-01-23 13:31:42.799426: step: 2660/1541, loss: 0.13453306257724762 2023-01-23 13:31:43.724503: step: 2662/1541, loss: 0.38437190651893616 2023-01-23 13:31:44.652265: step: 2664/1541, loss: 0.0615885891020298 2023-01-23 13:31:45.549070: step: 2666/1541, loss: 0.014072511345148087 2023-01-23 13:31:46.424986: step: 2668/1541, loss: 0.029921362176537514 2023-01-23 13:31:47.318095: step: 2670/1541, loss: 0.0004452097637113184 2023-01-23 13:31:48.209116: step: 2672/1541, loss: 1.6298145055770874 2023-01-23 13:31:49.087212: step: 2674/1541, loss: 0.05326288565993309 2023-01-23 13:31:49.986327: step: 2676/1541, loss: 0.0067852395586669445 2023-01-23 13:31:50.880272: step: 2678/1541, loss: 0.017349349334836006 2023-01-23 13:31:51.786337: step: 2680/1541, loss: 0.32455459237098694 2023-01-23 13:31:52.683027: step: 2682/1541, loss: 0.20659691095352173 2023-01-23 13:31:53.563494: step: 2684/1541, loss: 0.07372280210256577 2023-01-23 13:31:54.486324: step: 2686/1541, loss: 0.0015035587130114436 2023-01-23 13:31:55.401837: step: 2688/1541, loss: 0.009225023910403252 2023-01-23 13:31:56.288045: step: 2690/1541, loss: 0.03463327884674072 2023-01-23 13:31:57.146531: step: 2692/1541, loss: 0.20045599341392517 2023-01-23 13:31:58.081351: step: 2694/1541, loss: 0.0011512850178405643 2023-01-23 13:31:58.985145: step: 2696/1541, loss: 0.5912777781486511 2023-01-23 13:31:59.855391: step: 2698/1541, loss: 0.3326188325881958 2023-01-23 13:32:00.716112: step: 2700/1541, loss: 0.21241195499897003 2023-01-23 13:32:01.601044: step: 2702/1541, loss: 0.12219364941120148 2023-01-23 13:32:02.493792: step: 2704/1541, loss: 0.041299860924482346 2023-01-23 13:32:03.368399: step: 2706/1541, loss: 0.08093284070491791 2023-01-23 13:32:04.288702: step: 2708/1541, loss: 0.0752045214176178 2023-01-23 13:32:05.164425: step: 2710/1541, loss: 0.0016700936248525977 2023-01-23 13:32:06.063666: step: 2712/1541, loss: 0.7212167382240295 2023-01-23 13:32:06.934642: step: 2714/1541, loss: 0.23547421395778656 2023-01-23 13:32:07.798456: step: 2716/1541, loss: 2.7787325382232666 2023-01-23 13:32:08.732189: step: 2718/1541, loss: 0.11457962542772293 2023-01-23 13:32:09.656312: step: 2720/1541, loss: 0.054900530725717545 2023-01-23 13:32:10.580286: step: 2722/1541, loss: 0.015994912013411522 2023-01-23 13:32:11.453070: step: 2724/1541, loss: 0.013680891133844852 2023-01-23 13:32:12.331706: step: 2726/1541, loss: 0.14205704629421234 2023-01-23 13:32:13.239203: step: 2728/1541, loss: 0.0271901935338974 2023-01-23 13:32:14.156135: step: 2730/1541, loss: 0.009879360906779766 2023-01-23 13:32:15.081864: step: 2732/1541, loss: 0.0484209842979908 2023-01-23 13:32:16.000409: step: 2734/1541, loss: 0.0586107037961483 2023-01-23 13:32:16.866687: step: 2736/1541, loss: 0.13144128024578094 2023-01-23 13:32:17.746835: step: 2738/1541, loss: 0.041685882955789566 2023-01-23 13:32:18.687183: step: 2740/1541, loss: 0.08767169713973999 2023-01-23 13:32:19.683185: step: 2742/1541, loss: 0.015530141070485115 2023-01-23 13:32:20.600183: step: 2744/1541, loss: 0.10355158150196075 2023-01-23 13:32:21.526017: step: 2746/1541, loss: 0.6815831065177917 2023-01-23 13:32:22.414205: step: 2748/1541, loss: 0.0473937913775444 2023-01-23 13:32:23.308189: step: 2750/1541, loss: 0.053996749222278595 2023-01-23 13:32:24.173527: step: 2752/1541, loss: 0.002964879386126995 2023-01-23 13:32:25.053384: step: 2754/1541, loss: 0.01646331511437893 2023-01-23 13:32:25.936462: step: 2756/1541, loss: 0.08974044024944305 2023-01-23 13:32:26.799625: step: 2758/1541, loss: 0.10477717220783234 2023-01-23 13:32:27.702430: step: 2760/1541, loss: 0.024058926850557327 2023-01-23 13:32:28.750629: step: 2762/1541, loss: 0.0027702751103788614 2023-01-23 13:32:29.718395: step: 2764/1541, loss: 0.002738512586802244 2023-01-23 13:32:30.611070: step: 2766/1541, loss: 0.06973826140165329 2023-01-23 13:32:31.518868: step: 2768/1541, loss: 0.025923417881131172 2023-01-23 13:32:32.388168: step: 2770/1541, loss: 0.16340698301792145 2023-01-23 13:32:33.313519: step: 2772/1541, loss: 0.20840862393379211 2023-01-23 13:32:34.235174: step: 2774/1541, loss: 0.11355840414762497 2023-01-23 13:32:35.132389: step: 2776/1541, loss: 0.0016458991449326277 2023-01-23 13:32:36.117114: step: 2778/1541, loss: 0.011169382371008396 2023-01-23 13:32:37.028892: step: 2780/1541, loss: 0.009781121276319027 2023-01-23 13:32:37.961765: step: 2782/1541, loss: 0.0195864737033844 2023-01-23 13:32:38.871810: step: 2784/1541, loss: 0.06534158438444138 2023-01-23 13:32:39.788351: step: 2786/1541, loss: 0.017231324687600136 2023-01-23 13:32:40.697076: step: 2788/1541, loss: 0.019862569868564606 2023-01-23 13:32:41.556024: step: 2790/1541, loss: 0.07684632390737534 2023-01-23 13:32:42.416216: step: 2792/1541, loss: 0.008985933847725391 2023-01-23 13:32:43.302932: step: 2794/1541, loss: 0.08680988103151321 2023-01-23 13:32:44.207789: step: 2796/1541, loss: 0.016647852957248688 2023-01-23 13:32:45.090918: step: 2798/1541, loss: 0.014201411046087742 2023-01-23 13:32:46.000143: step: 2800/1541, loss: 0.09456305205821991 2023-01-23 13:32:46.907505: step: 2802/1541, loss: 0.10251103341579437 2023-01-23 13:32:47.791120: step: 2804/1541, loss: 0.22944965958595276 2023-01-23 13:32:48.658647: step: 2806/1541, loss: 0.0013401471078395844 2023-01-23 13:32:49.530665: step: 2808/1541, loss: 0.005424576811492443 2023-01-23 13:32:50.395552: step: 2810/1541, loss: 0.048588186502456665 2023-01-23 13:32:51.270437: step: 2812/1541, loss: 0.027401044964790344 2023-01-23 13:32:52.174889: step: 2814/1541, loss: 0.5878602862358093 2023-01-23 13:32:53.034753: step: 2816/1541, loss: 0.05740518495440483 2023-01-23 13:32:53.962874: step: 2818/1541, loss: 0.016529671847820282 2023-01-23 13:32:54.903611: step: 2820/1541, loss: 0.08568502962589264 2023-01-23 13:32:55.803539: step: 2822/1541, loss: 0.13637369871139526 2023-01-23 13:32:56.698488: step: 2824/1541, loss: 0.08135868608951569 2023-01-23 13:32:57.572480: step: 2826/1541, loss: 0.17537245154380798 2023-01-23 13:32:58.493867: step: 2828/1541, loss: 0.08420301973819733 2023-01-23 13:32:59.363948: step: 2830/1541, loss: 0.10303421318531036 2023-01-23 13:33:00.269438: step: 2832/1541, loss: 0.5915151834487915 2023-01-23 13:33:01.164186: step: 2834/1541, loss: 0.16769036650657654 2023-01-23 13:33:02.035691: step: 2836/1541, loss: 0.04996141418814659 2023-01-23 13:33:02.942034: step: 2838/1541, loss: 0.07590459287166595 2023-01-23 13:33:03.883547: step: 2840/1541, loss: 0.03396027162671089 2023-01-23 13:33:04.802568: step: 2842/1541, loss: 0.020311228930950165 2023-01-23 13:33:05.718900: step: 2844/1541, loss: 0.11251986771821976 2023-01-23 13:33:06.594631: step: 2846/1541, loss: 0.09865684062242508 2023-01-23 13:33:07.536875: step: 2848/1541, loss: 0.4663926362991333 2023-01-23 13:33:08.452283: step: 2850/1541, loss: 0.045131780207157135 2023-01-23 13:33:09.370838: step: 2852/1541, loss: 0.26532262563705444 2023-01-23 13:33:10.240593: step: 2854/1541, loss: 0.0717984214425087 2023-01-23 13:33:11.109501: step: 2856/1541, loss: 0.2484789490699768 2023-01-23 13:33:12.012263: step: 2858/1541, loss: 0.10255742818117142 2023-01-23 13:33:12.914504: step: 2860/1541, loss: 0.017559072002768517 2023-01-23 13:33:13.831878: step: 2862/1541, loss: 0.004875365179032087 2023-01-23 13:33:14.750737: step: 2864/1541, loss: 0.0679057389497757 2023-01-23 13:33:15.654652: step: 2866/1541, loss: 0.027875473722815514 2023-01-23 13:33:16.532845: step: 2868/1541, loss: 0.005152657628059387 2023-01-23 13:33:17.434321: step: 2870/1541, loss: 0.09064792841672897 2023-01-23 13:33:18.323005: step: 2872/1541, loss: 0.08080688118934631 2023-01-23 13:33:19.255179: step: 2874/1541, loss: 0.05381791293621063 2023-01-23 13:33:20.154541: step: 2876/1541, loss: 0.05854932218790054 2023-01-23 13:33:21.045290: step: 2878/1541, loss: 0.5516523122787476 2023-01-23 13:33:21.894862: step: 2880/1541, loss: 0.049057066440582275 2023-01-23 13:33:22.783490: step: 2882/1541, loss: 0.07966727763414383 2023-01-23 13:33:23.626368: step: 2884/1541, loss: 0.3295106887817383 2023-01-23 13:33:24.516488: step: 2886/1541, loss: 0.13865935802459717 2023-01-23 13:33:25.424888: step: 2888/1541, loss: 0.0337943397462368 2023-01-23 13:33:26.288085: step: 2890/1541, loss: 0.01599794067442417 2023-01-23 13:33:27.171595: step: 2892/1541, loss: 0.36735162138938904 2023-01-23 13:33:28.058476: step: 2894/1541, loss: 0.006095837336033583 2023-01-23 13:33:28.954514: step: 2896/1541, loss: 0.029876159504055977 2023-01-23 13:33:29.910040: step: 2898/1541, loss: 0.38226279616355896 2023-01-23 13:33:30.766388: step: 2900/1541, loss: 0.027924487367272377 2023-01-23 13:33:31.625331: step: 2902/1541, loss: 0.05163521692156792 2023-01-23 13:33:32.481419: step: 2904/1541, loss: 0.002108021406456828 2023-01-23 13:33:33.353439: step: 2906/1541, loss: 0.03478695824742317 2023-01-23 13:33:34.258721: step: 2908/1541, loss: 0.04839549958705902 2023-01-23 13:33:35.120804: step: 2910/1541, loss: 0.006668416317552328 2023-01-23 13:33:36.035288: step: 2912/1541, loss: 0.03675432130694389 2023-01-23 13:33:36.978115: step: 2914/1541, loss: 0.19004803895950317 2023-01-23 13:33:37.864463: step: 2916/1541, loss: 0.03793233260512352 2023-01-23 13:33:38.746508: step: 2918/1541, loss: 0.09670344740152359 2023-01-23 13:33:39.727114: step: 2920/1541, loss: 0.15410009026527405 2023-01-23 13:33:40.615787: step: 2922/1541, loss: 0.031333327293395996 2023-01-23 13:33:41.501332: step: 2924/1541, loss: 0.022624274715781212 2023-01-23 13:33:42.431207: step: 2926/1541, loss: 0.036314576864242554 2023-01-23 13:33:43.372949: step: 2928/1541, loss: 0.09669142216444016 2023-01-23 13:33:44.276074: step: 2930/1541, loss: 0.11317084729671478 2023-01-23 13:33:45.187212: step: 2932/1541, loss: 0.09302320331335068 2023-01-23 13:33:46.074850: step: 2934/1541, loss: 0.13382022082805634 2023-01-23 13:33:46.947709: step: 2936/1541, loss: 0.1630672961473465 2023-01-23 13:33:47.821042: step: 2938/1541, loss: 0.02745886892080307 2023-01-23 13:33:48.726951: step: 2940/1541, loss: 0.008984394371509552 2023-01-23 13:33:49.635589: step: 2942/1541, loss: 0.2938075065612793 2023-01-23 13:33:50.526879: step: 2944/1541, loss: 0.013839860446751118 2023-01-23 13:33:51.407959: step: 2946/1541, loss: 0.015170163474977016 2023-01-23 13:33:52.331945: step: 2948/1541, loss: 0.0016442665364593267 2023-01-23 13:33:53.320985: step: 2950/1541, loss: 0.10109329223632812 2023-01-23 13:33:54.207542: step: 2952/1541, loss: 0.07654835283756256 2023-01-23 13:33:55.054634: step: 2954/1541, loss: 0.005928581114858389 2023-01-23 13:33:55.954762: step: 2956/1541, loss: 0.0903153344988823 2023-01-23 13:33:56.851535: step: 2958/1541, loss: 0.006576709449291229 2023-01-23 13:33:57.725614: step: 2960/1541, loss: 0.10088109970092773 2023-01-23 13:33:58.596022: step: 2962/1541, loss: 0.006872275844216347 2023-01-23 13:33:59.509979: step: 2964/1541, loss: 0.10442568361759186 2023-01-23 13:34:00.418586: step: 2966/1541, loss: 0.10094975680112839 2023-01-23 13:34:01.333423: step: 2968/1541, loss: 0.11709624528884888 2023-01-23 13:34:02.283287: step: 2970/1541, loss: 0.027862144634127617 2023-01-23 13:34:03.213777: step: 2972/1541, loss: 0.44989538192749023 2023-01-23 13:34:04.118561: step: 2974/1541, loss: 0.049501098692417145 2023-01-23 13:34:05.052651: step: 2976/1541, loss: 0.04405096545815468 2023-01-23 13:34:05.953413: step: 2978/1541, loss: 0.03737078979611397 2023-01-23 13:34:06.850016: step: 2980/1541, loss: 0.24293383955955505 2023-01-23 13:34:07.752054: step: 2982/1541, loss: 0.1281648874282837 2023-01-23 13:34:08.617633: step: 2984/1541, loss: 0.05777006596326828 2023-01-23 13:34:09.511823: step: 2986/1541, loss: 0.04597015306353569 2023-01-23 13:34:10.385388: step: 2988/1541, loss: 0.034866075962781906 2023-01-23 13:34:11.265216: step: 2990/1541, loss: 0.011411454528570175 2023-01-23 13:34:12.135322: step: 2992/1541, loss: 0.039000872522592545 2023-01-23 13:34:13.037732: step: 2994/1541, loss: 0.009131587110459805 2023-01-23 13:34:13.954719: step: 2996/1541, loss: 0.009528676979243755 2023-01-23 13:34:14.861975: step: 2998/1541, loss: 0.004444832447916269 2023-01-23 13:34:15.889051: step: 3000/1541, loss: 0.293466717004776 2023-01-23 13:34:16.765853: step: 3002/1541, loss: 0.4620399475097656 2023-01-23 13:34:17.641921: step: 3004/1541, loss: 0.4622495472431183 2023-01-23 13:34:18.526984: step: 3006/1541, loss: 0.09683705866336823 2023-01-23 13:34:19.506049: step: 3008/1541, loss: 0.04502713307738304 2023-01-23 13:34:20.436916: step: 3010/1541, loss: 0.03938554972410202 2023-01-23 13:34:21.329353: step: 3012/1541, loss: 0.2784087657928467 2023-01-23 13:34:22.312967: step: 3014/1541, loss: 0.018860282376408577 2023-01-23 13:34:23.191144: step: 3016/1541, loss: 0.19744481146335602 2023-01-23 13:34:24.092581: step: 3018/1541, loss: 0.16288261115550995 2023-01-23 13:34:24.990502: step: 3020/1541, loss: 0.0009074907866306603 2023-01-23 13:34:25.880932: step: 3022/1541, loss: 0.0016650250181555748 2023-01-23 13:34:26.853476: step: 3024/1541, loss: 0.012695495039224625 2023-01-23 13:34:27.711347: step: 3026/1541, loss: 0.05516774207353592 2023-01-23 13:34:28.561811: step: 3028/1541, loss: 0.005496777594089508 2023-01-23 13:34:29.449917: step: 3030/1541, loss: 0.0340716652572155 2023-01-23 13:34:30.340475: step: 3032/1541, loss: 0.13451248407363892 2023-01-23 13:34:31.224757: step: 3034/1541, loss: 0.11812527477741241 2023-01-23 13:34:32.142817: step: 3036/1541, loss: 0.2195294201374054 2023-01-23 13:34:33.050855: step: 3038/1541, loss: 0.017060454934835434 2023-01-23 13:34:34.002691: step: 3040/1541, loss: 0.4700644314289093 2023-01-23 13:34:34.929531: step: 3042/1541, loss: 0.015028027817606926 2023-01-23 13:34:35.798577: step: 3044/1541, loss: 0.12852640450000763 2023-01-23 13:34:36.733599: step: 3046/1541, loss: 0.005955435335636139 2023-01-23 13:34:37.630861: step: 3048/1541, loss: 0.0026573871728032827 2023-01-23 13:34:38.467059: step: 3050/1541, loss: 0.09253375977277756 2023-01-23 13:34:39.348614: step: 3052/1541, loss: 0.24983330070972443 2023-01-23 13:34:40.219956: step: 3054/1541, loss: 0.22292283177375793 2023-01-23 13:34:41.131520: step: 3056/1541, loss: 0.1031327024102211 2023-01-23 13:34:42.006664: step: 3058/1541, loss: 0.039155010133981705 2023-01-23 13:34:42.879650: step: 3060/1541, loss: 0.7622398138046265 2023-01-23 13:34:43.755952: step: 3062/1541, loss: 0.09359554946422577 2023-01-23 13:34:44.662449: step: 3064/1541, loss: 0.3618670701980591 2023-01-23 13:34:45.597408: step: 3066/1541, loss: 0.004998777061700821 2023-01-23 13:34:46.490750: step: 3068/1541, loss: 0.07622431963682175 2023-01-23 13:34:47.446074: step: 3070/1541, loss: 0.0052468315698206425 2023-01-23 13:34:48.448925: step: 3072/1541, loss: 0.5301809906959534 2023-01-23 13:34:49.392350: step: 3074/1541, loss: 0.008651860989630222 2023-01-23 13:34:50.312792: step: 3076/1541, loss: 0.2728029787540436 2023-01-23 13:34:51.216805: step: 3078/1541, loss: 0.0837574377655983 2023-01-23 13:34:52.137250: step: 3080/1541, loss: 0.1798139065504074 2023-01-23 13:34:53.035394: step: 3082/1541, loss: 0.08002857118844986 2023-01-23 13:34:53.667298: step: 3084/1541, loss: 0.0912754088640213 ================================================== Loss: 0.118 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6640785176976081, 'r': 0.644001725302099, 'f1': 0.6538860493031938}, 'combined': 0.4315318511774227, 'epoch': 17} Test Chinese: {'event': {'p': 0.6093585699263933, 'r': 0.6944278010784901, 'f1': 0.6491178941472977}, 'argument': {'p': 0.7242476573073298, 'r': 0.6973259149438649, 'f1': 0.7105318643171055}, 'combined': 0.461218947490073, 'epoch': 17} Dev Korean: {'event': {'p': 0.5696078431372549, 'r': 0.7736351531291611, 'f1': 0.6561264822134386}, 'argument': {'p': 0.6588832184640967, 'r': 0.6489645463581856, 'f1': 0.6538862709784535}, 'combined': 0.429032098744756, 'epoch': 17} Test Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.7177867167777032, 'r': 0.712991376546066, 'f1': 0.7153810107239413}, 'combined': 0.4888215348031391, 'epoch': 17} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6642531440510157, 'r': 0.6507683057883259, 'f1': 0.657441585034921}, 'combined': 0.4550567460053622, 'epoch': 17} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.7176641171614032, 'r': 0.7019450826382988, 'f1': 0.7097175731158694}, 'combined': 0.4920444180093821, 'epoch': 17} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.63335837716003, 'r': 0.6781979082864038, 'f1': 0.6550116550116549}, 'combined': 0.4702647779570856, 'epoch': 17} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6135819107517221, 'r': 0.41692104192104196, 'f1': 0.4964861262571187}, 'combined': 0.28929112868525025, 'epoch': 17} Russian: {'event': {'p': 0.47058823529411764, 'r': 0.4444444444444444, 'f1': 0.45714285714285713}, 'argument': {'p': 0.5756560967767864, 'r': 0.5890434478646186, 'f1': 0.5822728335213472}, 'combined': 0.26618186675261585, 'epoch': 17} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6406723427552294, 'r': 0.6675328291750916, 'f1': 0.6538268319220668}, 'combined': 0.43149277068912223, 'epoch': 5} Test for Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.6917776725673452, 'r': 0.7202241343577782, 'f1': 0.7057143594990817}, 'combined': 0.460068056262633, 'epoch': 5} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6433357969943335, 'r': 0.7002681684097612, 'f1': 0.6705957883923985}, 'combined': 0.4929165624080878, 'epoch': 5} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 18 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 13:40:17.146115: step: 2/1541, loss: 0.03167138993740082 2023-01-23 13:40:18.035009: step: 4/1541, loss: 0.031252723187208176 2023-01-23 13:40:18.919218: step: 6/1541, loss: 0.12045005708932877 2023-01-23 13:40:19.852470: step: 8/1541, loss: 0.01120576448738575 2023-01-23 13:40:20.734212: step: 10/1541, loss: 0.1666063517332077 2023-01-23 13:40:21.599640: step: 12/1541, loss: 0.09283392876386642 2023-01-23 13:40:22.497426: step: 14/1541, loss: 0.0740724727511406 2023-01-23 13:40:23.404538: step: 16/1541, loss: 0.02452191896736622 2023-01-23 13:40:24.299359: step: 18/1541, loss: 0.028287608176469803 2023-01-23 13:40:25.231149: step: 20/1541, loss: 0.04039207473397255 2023-01-23 13:40:26.074495: step: 22/1541, loss: 0.05965889245271683 2023-01-23 13:40:26.935325: step: 24/1541, loss: 0.043523240834474564 2023-01-23 13:40:27.885576: step: 26/1541, loss: 0.044879138469696045 2023-01-23 13:40:28.747715: step: 28/1541, loss: 0.04016757011413574 2023-01-23 13:40:29.682131: step: 30/1541, loss: 0.04481187090277672 2023-01-23 13:40:30.546781: step: 32/1541, loss: 0.023505184799432755 2023-01-23 13:40:31.526055: step: 34/1541, loss: 0.033588383346796036 2023-01-23 13:40:32.407611: step: 36/1541, loss: 0.018087923526763916 2023-01-23 13:40:33.253023: step: 38/1541, loss: 0.14428389072418213 2023-01-23 13:40:34.220074: step: 40/1541, loss: 0.08640512079000473 2023-01-23 13:40:35.070509: step: 42/1541, loss: 0.05632287263870239 2023-01-23 13:40:35.986157: step: 44/1541, loss: 0.023415017873048782 2023-01-23 13:40:36.927347: step: 46/1541, loss: 0.013445996679365635 2023-01-23 13:40:37.840584: step: 48/1541, loss: 0.02030394971370697 2023-01-23 13:40:38.769918: step: 50/1541, loss: 0.020298687741160393 2023-01-23 13:40:39.671766: step: 52/1541, loss: 0.003354056039825082 2023-01-23 13:40:40.571120: step: 54/1541, loss: 0.3944990932941437 2023-01-23 13:40:41.512142: step: 56/1541, loss: 0.05184536054730415 2023-01-23 13:40:42.403374: step: 58/1541, loss: 0.11862887442111969 2023-01-23 13:40:43.268946: step: 60/1541, loss: 0.01179054006934166 2023-01-23 13:40:44.171817: step: 62/1541, loss: 0.1638486683368683 2023-01-23 13:40:45.003779: step: 64/1541, loss: 0.07698006182909012 2023-01-23 13:40:45.895783: step: 66/1541, loss: 0.05794934928417206 2023-01-23 13:40:46.850587: step: 68/1541, loss: 0.03863717243075371 2023-01-23 13:40:47.740844: step: 70/1541, loss: 0.0018667387776076794 2023-01-23 13:40:48.615620: step: 72/1541, loss: 0.002216184278950095 2023-01-23 13:40:49.490093: step: 74/1541, loss: 0.010352753102779388 2023-01-23 13:40:50.348420: step: 76/1541, loss: 0.0035813236609101295 2023-01-23 13:40:51.200560: step: 78/1541, loss: 0.09454351663589478 2023-01-23 13:40:52.091926: step: 80/1541, loss: 0.0078035336919128895 2023-01-23 13:40:53.013648: step: 82/1541, loss: 0.5205455422401428 2023-01-23 13:40:53.938458: step: 84/1541, loss: 0.07475526630878448 2023-01-23 13:40:54.778297: step: 86/1541, loss: 0.04646795988082886 2023-01-23 13:40:55.658156: step: 88/1541, loss: 0.005197844002395868 2023-01-23 13:40:56.601547: step: 90/1541, loss: 0.003941833507269621 2023-01-23 13:40:57.484292: step: 92/1541, loss: 0.053620677441358566 2023-01-23 13:40:58.415382: step: 94/1541, loss: 0.1888563483953476 2023-01-23 13:40:59.402137: step: 96/1541, loss: 0.5799818634986877 2023-01-23 13:41:00.304195: step: 98/1541, loss: 0.022854749113321304 2023-01-23 13:41:01.218277: step: 100/1541, loss: 0.0042548878118395805 2023-01-23 13:41:02.184265: step: 102/1541, loss: 0.006871955469250679 2023-01-23 13:41:03.088851: step: 104/1541, loss: 0.15678837895393372 2023-01-23 13:41:03.979746: step: 106/1541, loss: 0.008902897126972675 2023-01-23 13:41:04.891313: step: 108/1541, loss: 0.29337814450263977 2023-01-23 13:41:05.810839: step: 110/1541, loss: 0.04553539305925369 2023-01-23 13:41:06.710874: step: 112/1541, loss: 0.1444515436887741 2023-01-23 13:41:07.637696: step: 114/1541, loss: 0.008004043251276016 2023-01-23 13:41:08.487556: step: 116/1541, loss: 0.019381530582904816 2023-01-23 13:41:09.344006: step: 118/1541, loss: 0.001448211376555264 2023-01-23 13:41:10.271147: step: 120/1541, loss: 0.03206487372517586 2023-01-23 13:41:11.187080: step: 122/1541, loss: 0.2129950374364853 2023-01-23 13:41:12.115916: step: 124/1541, loss: 0.007131516933441162 2023-01-23 13:41:13.007773: step: 126/1541, loss: 0.03319280222058296 2023-01-23 13:41:13.956249: step: 128/1541, loss: 0.14381875097751617 2023-01-23 13:41:14.861595: step: 130/1541, loss: 1.0531423091888428 2023-01-23 13:41:15.783544: step: 132/1541, loss: 0.0027782912366092205 2023-01-23 13:41:16.690778: step: 134/1541, loss: 0.01713484711945057 2023-01-23 13:41:17.542398: step: 136/1541, loss: 0.0006825944292359054 2023-01-23 13:41:18.414721: step: 138/1541, loss: 0.0005364463431760669 2023-01-23 13:41:19.288846: step: 140/1541, loss: 0.003587862476706505 2023-01-23 13:41:20.145679: step: 142/1541, loss: 0.10191520303487778 2023-01-23 13:41:21.016260: step: 144/1541, loss: 0.06368841230869293 2023-01-23 13:41:21.960876: step: 146/1541, loss: 0.023890851065516472 2023-01-23 13:41:22.838893: step: 148/1541, loss: 0.026165343821048737 2023-01-23 13:41:23.760295: step: 150/1541, loss: 0.003694022772833705 2023-01-23 13:41:24.662605: step: 152/1541, loss: 0.018875334411859512 2023-01-23 13:41:25.565600: step: 154/1541, loss: 0.028772469609975815 2023-01-23 13:41:26.429134: step: 156/1541, loss: 0.004754864610731602 2023-01-23 13:41:27.300301: step: 158/1541, loss: 0.5555144548416138 2023-01-23 13:41:28.187089: step: 160/1541, loss: 0.008750852197408676 2023-01-23 13:41:29.084956: step: 162/1541, loss: 0.26795580983161926 2023-01-23 13:41:29.958790: step: 164/1541, loss: 0.06031786650419235 2023-01-23 13:41:30.839265: step: 166/1541, loss: 0.018374359235167503 2023-01-23 13:41:31.734750: step: 168/1541, loss: 0.5359371304512024 2023-01-23 13:41:32.653494: step: 170/1541, loss: 0.7701743841171265 2023-01-23 13:41:33.618837: step: 172/1541, loss: 0.0021193581633269787 2023-01-23 13:41:34.538800: step: 174/1541, loss: 0.2332596480846405 2023-01-23 13:41:35.539647: step: 176/1541, loss: 0.09242469072341919 2023-01-23 13:41:36.442110: step: 178/1541, loss: 0.04779333621263504 2023-01-23 13:41:37.345023: step: 180/1541, loss: 0.0010945461690425873 2023-01-23 13:41:38.269926: step: 182/1541, loss: 0.08468297868967056 2023-01-23 13:41:39.102035: step: 184/1541, loss: 0.0032733820844441652 2023-01-23 13:41:40.060180: step: 186/1541, loss: 0.004524590447545052 2023-01-23 13:41:40.924653: step: 188/1541, loss: 0.051940493285655975 2023-01-23 13:41:41.826271: step: 190/1541, loss: 0.10192398726940155 2023-01-23 13:41:42.670848: step: 192/1541, loss: 0.0003672754974104464 2023-01-23 13:41:43.603168: step: 194/1541, loss: 0.02001466229557991 2023-01-23 13:41:44.493678: step: 196/1541, loss: 0.59686279296875 2023-01-23 13:41:45.362906: step: 198/1541, loss: 0.018477197736501694 2023-01-23 13:41:46.271590: step: 200/1541, loss: 0.0198025181889534 2023-01-23 13:41:47.205038: step: 202/1541, loss: 0.11188554018735886 2023-01-23 13:41:48.084675: step: 204/1541, loss: 0.002441598568111658 2023-01-23 13:41:49.017244: step: 206/1541, loss: 0.016970716416835785 2023-01-23 13:41:49.908198: step: 208/1541, loss: 0.08372700214385986 2023-01-23 13:41:50.801764: step: 210/1541, loss: 0.04571813717484474 2023-01-23 13:41:51.690139: step: 212/1541, loss: 0.3031550645828247 2023-01-23 13:41:52.581211: step: 214/1541, loss: 0.3001607656478882 2023-01-23 13:41:53.499700: step: 216/1541, loss: 0.026237409561872482 2023-01-23 13:41:54.371348: step: 218/1541, loss: 0.014175524935126305 2023-01-23 13:41:55.224300: step: 220/1541, loss: 0.05373436212539673 2023-01-23 13:41:56.146738: step: 222/1541, loss: 0.10773556679487228 2023-01-23 13:41:57.018088: step: 224/1541, loss: 0.04030788689851761 2023-01-23 13:41:57.966090: step: 226/1541, loss: 0.005087403580546379 2023-01-23 13:41:58.850267: step: 228/1541, loss: 0.09760823100805283 2023-01-23 13:41:59.714226: step: 230/1541, loss: 0.015509521588683128 2023-01-23 13:42:00.581171: step: 232/1541, loss: 0.1374167650938034 2023-01-23 13:42:01.465430: step: 234/1541, loss: 0.14378394186496735 2023-01-23 13:42:02.347390: step: 236/1541, loss: 0.0016869991086423397 2023-01-23 13:42:03.206669: step: 238/1541, loss: 0.014420066960155964 2023-01-23 13:42:04.135026: step: 240/1541, loss: 0.16339173913002014 2023-01-23 13:42:04.992874: step: 242/1541, loss: 1.701258659362793 2023-01-23 13:42:05.889617: step: 244/1541, loss: 0.21987950801849365 2023-01-23 13:42:06.805061: step: 246/1541, loss: 0.003689684672281146 2023-01-23 13:42:07.722764: step: 248/1541, loss: 0.274677038192749 2023-01-23 13:42:08.591894: step: 250/1541, loss: 0.005432185251265764 2023-01-23 13:42:09.491932: step: 252/1541, loss: 0.22448237240314484 2023-01-23 13:42:10.455432: step: 254/1541, loss: 0.0038933814503252506 2023-01-23 13:42:11.360700: step: 256/1541, loss: 0.04561635106801987 2023-01-23 13:42:12.272348: step: 258/1541, loss: 0.007521952502429485 2023-01-23 13:42:13.128785: step: 260/1541, loss: 0.14437924325466156 2023-01-23 13:42:14.064975: step: 262/1541, loss: 0.026745125651359558 2023-01-23 13:42:14.938414: step: 264/1541, loss: 0.002379072131589055 2023-01-23 13:42:15.878115: step: 266/1541, loss: 0.044783469289541245 2023-01-23 13:42:16.795501: step: 268/1541, loss: 0.17113782465457916 2023-01-23 13:42:17.697221: step: 270/1541, loss: 0.11431426554918289 2023-01-23 13:42:18.621283: step: 272/1541, loss: 0.1175670400261879 2023-01-23 13:42:19.565507: step: 274/1541, loss: 0.10870248824357986 2023-01-23 13:42:20.421738: step: 276/1541, loss: 0.11576926708221436 2023-01-23 13:42:21.317247: step: 278/1541, loss: 0.022760484367609024 2023-01-23 13:42:22.243791: step: 280/1541, loss: 0.005376818589866161 2023-01-23 13:42:23.112189: step: 282/1541, loss: 0.07537765055894852 2023-01-23 13:42:24.052223: step: 284/1541, loss: 0.12068654596805573 2023-01-23 13:42:24.973643: step: 286/1541, loss: 0.18088464438915253 2023-01-23 13:42:25.850738: step: 288/1541, loss: 0.1388048380613327 2023-01-23 13:42:26.764847: step: 290/1541, loss: 0.2599007487297058 2023-01-23 13:42:27.691696: step: 292/1541, loss: 0.1623908281326294 2023-01-23 13:42:28.601555: step: 294/1541, loss: 0.006694047711789608 2023-01-23 13:42:29.500328: step: 296/1541, loss: 0.03885795548558235 2023-01-23 13:42:30.376690: step: 298/1541, loss: 0.002871888689696789 2023-01-23 13:42:31.295507: step: 300/1541, loss: 0.004705744795501232 2023-01-23 13:42:32.229858: step: 302/1541, loss: 2.2530882358551025 2023-01-23 13:42:33.088831: step: 304/1541, loss: 0.04520944133400917 2023-01-23 13:42:33.994227: step: 306/1541, loss: 0.08204144239425659 2023-01-23 13:42:34.873764: step: 308/1541, loss: 0.0013716209214180708 2023-01-23 13:42:35.768012: step: 310/1541, loss: 0.03886165842413902 2023-01-23 13:42:36.635311: step: 312/1541, loss: 0.015643319115042686 2023-01-23 13:42:37.556046: step: 314/1541, loss: 0.024586297571659088 2023-01-23 13:42:38.427992: step: 316/1541, loss: 0.9871943593025208 2023-01-23 13:42:39.337120: step: 318/1541, loss: 0.009712541475892067 2023-01-23 13:42:40.268702: step: 320/1541, loss: 0.2680545747280121 2023-01-23 13:42:41.153730: step: 322/1541, loss: 0.09075714647769928 2023-01-23 13:42:42.061564: step: 324/1541, loss: 0.01136994082480669 2023-01-23 13:42:42.957480: step: 326/1541, loss: 0.04841097071766853 2023-01-23 13:42:43.855852: step: 328/1541, loss: 0.16480037569999695 2023-01-23 13:42:44.724571: step: 330/1541, loss: 0.0027586144860833883 2023-01-23 13:42:45.650730: step: 332/1541, loss: 0.006775274872779846 2023-01-23 13:42:46.530692: step: 334/1541, loss: 0.0014781096251681447 2023-01-23 13:42:47.375322: step: 336/1541, loss: 0.008720762096345425 2023-01-23 13:42:48.326140: step: 338/1541, loss: 0.2529064416885376 2023-01-23 13:42:49.152748: step: 340/1541, loss: 0.03776928782463074 2023-01-23 13:42:50.055911: step: 342/1541, loss: 2.265246868133545 2023-01-23 13:42:50.964112: step: 344/1541, loss: 0.0034401349257677794 2023-01-23 13:42:51.835247: step: 346/1541, loss: 0.0011754065053537488 2023-01-23 13:42:52.726040: step: 348/1541, loss: 0.009324345737695694 2023-01-23 13:42:53.594930: step: 350/1541, loss: 0.022643079981207848 2023-01-23 13:42:54.496452: step: 352/1541, loss: 0.13527336716651917 2023-01-23 13:42:55.358418: step: 354/1541, loss: 0.0060224272310733795 2023-01-23 13:42:56.230473: step: 356/1541, loss: 0.00043817824916914105 2023-01-23 13:42:57.100046: step: 358/1541, loss: 0.0019699635449796915 2023-01-23 13:42:58.019776: step: 360/1541, loss: 0.30664780735969543 2023-01-23 13:42:58.907245: step: 362/1541, loss: 0.06712885946035385 2023-01-23 13:42:59.825131: step: 364/1541, loss: 0.2013663798570633 2023-01-23 13:43:00.725594: step: 366/1541, loss: 0.567334771156311 2023-01-23 13:43:01.630098: step: 368/1541, loss: 0.06595244258642197 2023-01-23 13:43:02.474880: step: 370/1541, loss: 0.00543657410889864 2023-01-23 13:43:03.397864: step: 372/1541, loss: 0.005211867392063141 2023-01-23 13:43:04.316321: step: 374/1541, loss: 0.0006777165690436959 2023-01-23 13:43:05.178136: step: 376/1541, loss: 0.17380115389823914 2023-01-23 13:43:06.045025: step: 378/1541, loss: 0.03280390799045563 2023-01-23 13:43:06.930040: step: 380/1541, loss: 0.100759357213974 2023-01-23 13:43:07.835409: step: 382/1541, loss: 0.02859988622367382 2023-01-23 13:43:08.758607: step: 384/1541, loss: 0.035405099391937256 2023-01-23 13:43:09.657277: step: 386/1541, loss: 0.01699351891875267 2023-01-23 13:43:10.567731: step: 388/1541, loss: 0.06883090734481812 2023-01-23 13:43:11.466024: step: 390/1541, loss: 0.021487301215529442 2023-01-23 13:43:12.420716: step: 392/1541, loss: 0.05897705256938934 2023-01-23 13:43:13.355612: step: 394/1541, loss: 0.05783717334270477 2023-01-23 13:43:14.282470: step: 396/1541, loss: 0.0006234690081328154 2023-01-23 13:43:15.162969: step: 398/1541, loss: 0.016179252415895462 2023-01-23 13:43:16.053701: step: 400/1541, loss: 0.010870292782783508 2023-01-23 13:43:16.930670: step: 402/1541, loss: 0.06589603424072266 2023-01-23 13:43:17.835100: step: 404/1541, loss: 0.2890029847621918 2023-01-23 13:43:18.741141: step: 406/1541, loss: 0.008208822458982468 2023-01-23 13:43:19.623273: step: 408/1541, loss: 0.008649587631225586 2023-01-23 13:43:20.530319: step: 410/1541, loss: 0.0343882292509079 2023-01-23 13:43:21.396600: step: 412/1541, loss: 0.002631555078551173 2023-01-23 13:43:22.255358: step: 414/1541, loss: 0.8101260662078857 2023-01-23 13:43:23.158100: step: 416/1541, loss: 0.020374111831188202 2023-01-23 13:43:24.049056: step: 418/1541, loss: 0.000506754731759429 2023-01-23 13:43:24.938438: step: 420/1541, loss: 0.011743126437067986 2023-01-23 13:43:25.899758: step: 422/1541, loss: 0.07007783651351929 2023-01-23 13:43:26.757497: step: 424/1541, loss: 0.026078952476382256 2023-01-23 13:43:27.637275: step: 426/1541, loss: 0.10966729372739792 2023-01-23 13:43:28.552019: step: 428/1541, loss: 0.1504870355129242 2023-01-23 13:43:29.459418: step: 430/1541, loss: 0.0914187803864479 2023-01-23 13:43:30.414192: step: 432/1541, loss: 0.0008953818469308317 2023-01-23 13:43:31.291655: step: 434/1541, loss: 0.568489670753479 2023-01-23 13:43:32.235623: step: 436/1541, loss: 0.031998224556446075 2023-01-23 13:43:33.073864: step: 438/1541, loss: 0.004127781838178635 2023-01-23 13:43:33.977230: step: 440/1541, loss: 0.09752589464187622 2023-01-23 13:43:34.953128: step: 442/1541, loss: 0.11604975163936615 2023-01-23 13:43:35.875125: step: 444/1541, loss: 0.06915535032749176 2023-01-23 13:43:36.764022: step: 446/1541, loss: 0.0027857900131493807 2023-01-23 13:43:37.662269: step: 448/1541, loss: 0.07944732904434204 2023-01-23 13:43:38.652684: step: 450/1541, loss: 0.24353644251823425 2023-01-23 13:43:39.517869: step: 452/1541, loss: 0.3217882812023163 2023-01-23 13:43:40.388377: step: 454/1541, loss: 0.07805657386779785 2023-01-23 13:43:41.248081: step: 456/1541, loss: 0.14454178512096405 2023-01-23 13:43:42.161750: step: 458/1541, loss: 0.03731463477015495 2023-01-23 13:43:43.150941: step: 460/1541, loss: 0.2169698029756546 2023-01-23 13:43:44.043626: step: 462/1541, loss: 0.03347097337245941 2023-01-23 13:43:44.989986: step: 464/1541, loss: 0.07161908596754074 2023-01-23 13:43:45.876363: step: 466/1541, loss: 0.08263411372900009 2023-01-23 13:43:46.818792: step: 468/1541, loss: 0.04685647785663605 2023-01-23 13:43:47.771664: step: 470/1541, loss: 0.022168587893247604 2023-01-23 13:43:48.668085: step: 472/1541, loss: 0.38291096687316895 2023-01-23 13:43:49.590867: step: 474/1541, loss: 0.1789609044790268 2023-01-23 13:43:50.539582: step: 476/1541, loss: 0.14439423382282257 2023-01-23 13:43:51.393222: step: 478/1541, loss: 0.0003703912952914834 2023-01-23 13:43:52.285707: step: 480/1541, loss: 0.113945372402668 2023-01-23 13:43:53.234051: step: 482/1541, loss: 0.0488165020942688 2023-01-23 13:43:54.195173: step: 484/1541, loss: 0.010311068035662174 2023-01-23 13:43:55.098880: step: 486/1541, loss: 0.0039005870930850506 2023-01-23 13:43:55.995869: step: 488/1541, loss: 0.10010526329278946 2023-01-23 13:43:56.900162: step: 490/1541, loss: 0.004624937195330858 2023-01-23 13:43:57.799883: step: 492/1541, loss: 0.2734541594982147 2023-01-23 13:43:58.709423: step: 494/1541, loss: 0.1974729299545288 2023-01-23 13:43:59.602938: step: 496/1541, loss: 0.0686512216925621 2023-01-23 13:44:00.469045: step: 498/1541, loss: 0.0705181285738945 2023-01-23 13:44:01.376611: step: 500/1541, loss: 0.027456944808363914 2023-01-23 13:44:02.280431: step: 502/1541, loss: 0.0050945160910487175 2023-01-23 13:44:03.194989: step: 504/1541, loss: 0.01920604705810547 2023-01-23 13:44:04.089688: step: 506/1541, loss: 0.1273924857378006 2023-01-23 13:44:04.975474: step: 508/1541, loss: 0.019213495776057243 2023-01-23 13:44:05.920600: step: 510/1541, loss: 0.018927646800875664 2023-01-23 13:44:06.830873: step: 512/1541, loss: 0.1009230688214302 2023-01-23 13:44:07.714789: step: 514/1541, loss: 0.2160305678844452 2023-01-23 13:44:08.617222: step: 516/1541, loss: 0.18574194610118866 2023-01-23 13:44:09.508507: step: 518/1541, loss: 0.673141598701477 2023-01-23 13:44:10.414428: step: 520/1541, loss: 0.00839014071971178 2023-01-23 13:44:11.296252: step: 522/1541, loss: 0.05798783898353577 2023-01-23 13:44:12.289459: step: 524/1541, loss: 0.03460098057985306 2023-01-23 13:44:13.298103: step: 526/1541, loss: 0.012909317389130592 2023-01-23 13:44:14.156871: step: 528/1541, loss: 0.12424344569444656 2023-01-23 13:44:15.070239: step: 530/1541, loss: 0.0030059844721108675 2023-01-23 13:44:15.980691: step: 532/1541, loss: 0.1292862892150879 2023-01-23 13:44:16.908670: step: 534/1541, loss: 2.324120044708252 2023-01-23 13:44:17.904619: step: 536/1541, loss: 0.2944205105304718 2023-01-23 13:44:18.837130: step: 538/1541, loss: 0.2628215551376343 2023-01-23 13:44:19.768744: step: 540/1541, loss: 0.0424942821264267 2023-01-23 13:44:20.725707: step: 542/1541, loss: 0.013327321037650108 2023-01-23 13:44:21.684123: step: 544/1541, loss: 0.004175846464931965 2023-01-23 13:44:22.536239: step: 546/1541, loss: 0.08292068541049957 2023-01-23 13:44:23.397560: step: 548/1541, loss: 0.024830324575304985 2023-01-23 13:44:24.343590: step: 550/1541, loss: 0.0036294220481067896 2023-01-23 13:44:25.319080: step: 552/1541, loss: 0.39744240045547485 2023-01-23 13:44:26.222252: step: 554/1541, loss: 0.02107456885278225 2023-01-23 13:44:27.114785: step: 556/1541, loss: 0.2632489800453186 2023-01-23 13:44:27.993917: step: 558/1541, loss: 0.048272982239723206 2023-01-23 13:44:28.834653: step: 560/1541, loss: 0.0016901196213439107 2023-01-23 13:44:29.733353: step: 562/1541, loss: 0.11940109729766846 2023-01-23 13:44:30.636276: step: 564/1541, loss: 0.18479768931865692 2023-01-23 13:44:31.548975: step: 566/1541, loss: 0.053942956030368805 2023-01-23 13:44:32.419032: step: 568/1541, loss: 0.017892582342028618 2023-01-23 13:44:33.335632: step: 570/1541, loss: 0.13849535584449768 2023-01-23 13:44:34.216048: step: 572/1541, loss: 0.17375420033931732 2023-01-23 13:44:35.214855: step: 574/1541, loss: 0.0014666621573269367 2023-01-23 13:44:36.087016: step: 576/1541, loss: 0.008639959618449211 2023-01-23 13:44:36.974096: step: 578/1541, loss: 0.1408296525478363 2023-01-23 13:44:37.878080: step: 580/1541, loss: 0.013327508233487606 2023-01-23 13:44:38.868060: step: 582/1541, loss: 0.31799206137657166 2023-01-23 13:44:39.868676: step: 584/1541, loss: 0.01174866035580635 2023-01-23 13:44:40.742174: step: 586/1541, loss: 0.004607317969202995 2023-01-23 13:44:41.655913: step: 588/1541, loss: 0.24435777962207794 2023-01-23 13:44:42.541265: step: 590/1541, loss: 0.031526222825050354 2023-01-23 13:44:43.462857: step: 592/1541, loss: 0.28896310925483704 2023-01-23 13:44:44.402541: step: 594/1541, loss: 0.8027299046516418 2023-01-23 13:44:45.282134: step: 596/1541, loss: 0.034358374774456024 2023-01-23 13:44:46.153385: step: 598/1541, loss: 0.16464565694332123 2023-01-23 13:44:47.053358: step: 600/1541, loss: 0.029434850439429283 2023-01-23 13:44:48.030452: step: 602/1541, loss: 0.012653099372982979 2023-01-23 13:44:48.901713: step: 604/1541, loss: 0.033301785588264465 2023-01-23 13:44:49.863172: step: 606/1541, loss: 0.17200222611427307 2023-01-23 13:44:50.795918: step: 608/1541, loss: 0.01475493423640728 2023-01-23 13:44:51.680798: step: 610/1541, loss: 0.046330276876688004 2023-01-23 13:44:52.549191: step: 612/1541, loss: 0.009881246834993362 2023-01-23 13:44:53.429243: step: 614/1541, loss: 0.02771923877298832 2023-01-23 13:44:54.297610: step: 616/1541, loss: 0.0022334707900881767 2023-01-23 13:44:55.185749: step: 618/1541, loss: 0.0020753908902406693 2023-01-23 13:44:56.096306: step: 620/1541, loss: 0.1325647383928299 2023-01-23 13:44:56.964894: step: 622/1541, loss: 0.18252329528331757 2023-01-23 13:44:57.889225: step: 624/1541, loss: 0.01451291237026453 2023-01-23 13:44:58.792578: step: 626/1541, loss: 0.027213532477617264 2023-01-23 13:44:59.721258: step: 628/1541, loss: 0.02338140644133091 2023-01-23 13:45:00.612578: step: 630/1541, loss: 0.15288323163986206 2023-01-23 13:45:01.521366: step: 632/1541, loss: 0.021981708705425262 2023-01-23 13:45:02.507126: step: 634/1541, loss: 0.0005555914249271154 2023-01-23 13:45:03.390326: step: 636/1541, loss: 0.021452059969305992 2023-01-23 13:45:04.283725: step: 638/1541, loss: 0.16993902623653412 2023-01-23 13:45:05.264474: step: 640/1541, loss: 0.016838552430272102 2023-01-23 13:45:06.160009: step: 642/1541, loss: 1.2142747640609741 2023-01-23 13:45:07.067137: step: 644/1541, loss: 0.008364157751202583 2023-01-23 13:45:07.997008: step: 646/1541, loss: 0.015141448006033897 2023-01-23 13:45:08.949980: step: 648/1541, loss: 0.22327204048633575 2023-01-23 13:45:09.832466: step: 650/1541, loss: 1.9571034908294678 2023-01-23 13:45:10.738103: step: 652/1541, loss: 0.06363751739263535 2023-01-23 13:45:11.652242: step: 654/1541, loss: 0.03902821987867355 2023-01-23 13:45:12.536825: step: 656/1541, loss: 0.11467930674552917 2023-01-23 13:45:13.436365: step: 658/1541, loss: 0.21343959867954254 2023-01-23 13:45:14.346933: step: 660/1541, loss: 0.052577778697013855 2023-01-23 13:45:15.226747: step: 662/1541, loss: 0.007944525219500065 2023-01-23 13:45:16.110171: step: 664/1541, loss: 0.08106973022222519 2023-01-23 13:45:17.017272: step: 666/1541, loss: 0.0017253892729058862 2023-01-23 13:45:17.892065: step: 668/1541, loss: 0.19235552847385406 2023-01-23 13:45:18.761534: step: 670/1541, loss: 0.16274598240852356 2023-01-23 13:45:19.777863: step: 672/1541, loss: 0.09374597668647766 2023-01-23 13:45:20.655622: step: 674/1541, loss: 0.1288571059703827 2023-01-23 13:45:21.504302: step: 676/1541, loss: 0.03597090393304825 2023-01-23 13:45:22.411562: step: 678/1541, loss: 0.6257969737052917 2023-01-23 13:45:23.315364: step: 680/1541, loss: 0.10398691892623901 2023-01-23 13:45:24.231280: step: 682/1541, loss: 0.0011440615635365248 2023-01-23 13:45:25.138715: step: 684/1541, loss: 0.8525472283363342 2023-01-23 13:45:26.076912: step: 686/1541, loss: 0.0008262282935902476 2023-01-23 13:45:26.978941: step: 688/1541, loss: 1.517127513885498 2023-01-23 13:45:27.853080: step: 690/1541, loss: 0.008462222293019295 2023-01-23 13:45:28.787419: step: 692/1541, loss: 0.011921565048396587 2023-01-23 13:45:29.669116: step: 694/1541, loss: 0.013857966288924217 2023-01-23 13:45:30.579550: step: 696/1541, loss: 0.026341207325458527 2023-01-23 13:45:31.501716: step: 698/1541, loss: 0.019284134730696678 2023-01-23 13:45:32.471373: step: 700/1541, loss: 0.09872955083847046 2023-01-23 13:45:33.384242: step: 702/1541, loss: 0.13716529309749603 2023-01-23 13:45:34.255840: step: 704/1541, loss: 0.09924763441085815 2023-01-23 13:45:35.197924: step: 706/1541, loss: 0.1943928301334381 2023-01-23 13:45:36.083084: step: 708/1541, loss: 0.003870465327054262 2023-01-23 13:45:37.011070: step: 710/1541, loss: 0.17800742387771606 2023-01-23 13:45:37.972817: step: 712/1541, loss: 0.08984506875276566 2023-01-23 13:45:38.833165: step: 714/1541, loss: 0.02046402357518673 2023-01-23 13:45:39.742084: step: 716/1541, loss: 0.09129280596971512 2023-01-23 13:45:40.631663: step: 718/1541, loss: 0.07907211035490036 2023-01-23 13:45:41.532029: step: 720/1541, loss: 0.4179629981517792 2023-01-23 13:45:42.512260: step: 722/1541, loss: 0.001663272618316114 2023-01-23 13:45:43.384753: step: 724/1541, loss: 0.0029955203644931316 2023-01-23 13:45:44.267818: step: 726/1541, loss: 0.038028962910175323 2023-01-23 13:45:45.157893: step: 728/1541, loss: 0.0037028719671070576 2023-01-23 13:45:46.091116: step: 730/1541, loss: 0.09447617828845978 2023-01-23 13:45:47.044923: step: 732/1541, loss: 0.053289994597435 2023-01-23 13:45:47.960235: step: 734/1541, loss: 0.016065146774053574 2023-01-23 13:45:48.836716: step: 736/1541, loss: 0.002781136427074671 2023-01-23 13:45:49.737086: step: 738/1541, loss: 0.007557546719908714 2023-01-23 13:45:50.626331: step: 740/1541, loss: 0.03698279708623886 2023-01-23 13:45:51.510912: step: 742/1541, loss: 0.023053249344229698 2023-01-23 13:45:52.384449: step: 744/1541, loss: 0.005392503459006548 2023-01-23 13:45:53.297479: step: 746/1541, loss: 0.005242730490863323 2023-01-23 13:45:54.161710: step: 748/1541, loss: 0.0029714268166571856 2023-01-23 13:45:55.045234: step: 750/1541, loss: 0.3128361403942108 2023-01-23 13:45:55.993599: step: 752/1541, loss: 0.028281861916184425 2023-01-23 13:45:56.876171: step: 754/1541, loss: 0.03545592352747917 2023-01-23 13:45:57.823474: step: 756/1541, loss: 0.42849671840667725 2023-01-23 13:45:58.702790: step: 758/1541, loss: 0.2264781892299652 2023-01-23 13:45:59.609867: step: 760/1541, loss: 0.3131261169910431 2023-01-23 13:46:00.514024: step: 762/1541, loss: 0.04086082801222801 2023-01-23 13:46:01.439605: step: 764/1541, loss: 0.08923818171024323 2023-01-23 13:46:02.438202: step: 766/1541, loss: 0.02882053516805172 2023-01-23 13:46:03.330359: step: 768/1541, loss: 0.007531320676207542 2023-01-23 13:46:04.181897: step: 770/1541, loss: 0.3712306022644043 2023-01-23 13:46:05.100619: step: 772/1541, loss: 0.008026636205613613 2023-01-23 13:46:06.022821: step: 774/1541, loss: 0.0013917306205257773 2023-01-23 13:46:06.913093: step: 776/1541, loss: 0.018582850694656372 2023-01-23 13:46:07.815158: step: 778/1541, loss: 0.057280123233795166 2023-01-23 13:46:08.712172: step: 780/1541, loss: 0.0035091228783130646 2023-01-23 13:46:09.630733: step: 782/1541, loss: 0.08744553476572037 2023-01-23 13:46:10.520947: step: 784/1541, loss: 0.07915737479925156 2023-01-23 13:46:11.426203: step: 786/1541, loss: 0.00118174497038126 2023-01-23 13:46:12.330412: step: 788/1541, loss: 0.07491274923086166 2023-01-23 13:46:13.272694: step: 790/1541, loss: 0.008905691094696522 2023-01-23 13:46:14.162466: step: 792/1541, loss: 0.2573586702346802 2023-01-23 13:46:15.044566: step: 794/1541, loss: 0.1399036943912506 2023-01-23 13:46:16.018244: step: 796/1541, loss: 0.07011990994215012 2023-01-23 13:46:16.905786: step: 798/1541, loss: 0.011174469254910946 2023-01-23 13:46:17.820821: step: 800/1541, loss: 0.012445351108908653 2023-01-23 13:46:18.710927: step: 802/1541, loss: 0.003445929381996393 2023-01-23 13:46:19.672281: step: 804/1541, loss: 0.008145988918840885 2023-01-23 13:46:20.572357: step: 806/1541, loss: 0.014569563791155815 2023-01-23 13:46:21.460978: step: 808/1541, loss: 0.0037413418758660555 2023-01-23 13:46:22.341584: step: 810/1541, loss: 0.16886714100837708 2023-01-23 13:46:23.215865: step: 812/1541, loss: 0.2648633122444153 2023-01-23 13:46:24.121373: step: 814/1541, loss: 0.032391101121902466 2023-01-23 13:46:25.021186: step: 816/1541, loss: 0.02357361651957035 2023-01-23 13:46:25.923328: step: 818/1541, loss: 0.26272884011268616 2023-01-23 13:46:26.772615: step: 820/1541, loss: 0.002569508273154497 2023-01-23 13:46:27.645456: step: 822/1541, loss: 0.020862611010670662 2023-01-23 13:46:28.542821: step: 824/1541, loss: 0.1618192344903946 2023-01-23 13:46:29.398088: step: 826/1541, loss: 0.14685404300689697 2023-01-23 13:46:30.300102: step: 828/1541, loss: 0.04107276722788811 2023-01-23 13:46:31.163229: step: 830/1541, loss: 0.010439794510602951 2023-01-23 13:46:32.062781: step: 832/1541, loss: 0.026993483304977417 2023-01-23 13:46:32.964372: step: 834/1541, loss: 0.09317512810230255 2023-01-23 13:46:33.894129: step: 836/1541, loss: 0.06580539792776108 2023-01-23 13:46:34.842522: step: 838/1541, loss: 0.03266920894384384 2023-01-23 13:46:35.742667: step: 840/1541, loss: 0.10705414414405823 2023-01-23 13:46:36.685345: step: 842/1541, loss: 0.04002569615840912 2023-01-23 13:46:37.611107: step: 844/1541, loss: 0.007076891139149666 2023-01-23 13:46:38.630129: step: 846/1541, loss: 0.025137802585959435 2023-01-23 13:46:39.528475: step: 848/1541, loss: 0.15784312784671783 2023-01-23 13:46:40.512184: step: 850/1541, loss: 0.16845333576202393 2023-01-23 13:46:41.479721: step: 852/1541, loss: 0.028844628483057022 2023-01-23 13:46:42.334558: step: 854/1541, loss: 0.009963423013687134 2023-01-23 13:46:43.284131: step: 856/1541, loss: 0.13480785489082336 2023-01-23 13:46:44.158629: step: 858/1541, loss: 0.017207790166139603 2023-01-23 13:46:45.036739: step: 860/1541, loss: 0.03288906812667847 2023-01-23 13:46:45.935108: step: 862/1541, loss: 0.04666819050908089 2023-01-23 13:46:46.825777: step: 864/1541, loss: 0.1478477269411087 2023-01-23 13:46:47.751754: step: 866/1541, loss: 0.1705843210220337 2023-01-23 13:46:48.660182: step: 868/1541, loss: 0.45422008633613586 2023-01-23 13:46:49.524048: step: 870/1541, loss: 0.001990019576624036 2023-01-23 13:46:50.436409: step: 872/1541, loss: 0.002790072001516819 2023-01-23 13:46:51.397385: step: 874/1541, loss: 0.023328658193349838 2023-01-23 13:46:52.320478: step: 876/1541, loss: 0.0014473379123955965 2023-01-23 13:46:53.229806: step: 878/1541, loss: 0.0007775822305120528 2023-01-23 13:46:54.105486: step: 880/1541, loss: 0.008096902631223202 2023-01-23 13:46:55.073797: step: 882/1541, loss: 0.010028445161879063 2023-01-23 13:46:56.004806: step: 884/1541, loss: 0.0007818815647624433 2023-01-23 13:46:56.891064: step: 886/1541, loss: 0.02430824562907219 2023-01-23 13:46:57.777979: step: 888/1541, loss: 0.025341195985674858 2023-01-23 13:46:58.662466: step: 890/1541, loss: 0.004494498949497938 2023-01-23 13:46:59.566924: step: 892/1541, loss: 0.12808315455913544 2023-01-23 13:47:00.428125: step: 894/1541, loss: 0.01392448041588068 2023-01-23 13:47:01.304514: step: 896/1541, loss: 0.24313390254974365 2023-01-23 13:47:02.234514: step: 898/1541, loss: 0.252944678068161 2023-01-23 13:47:03.169942: step: 900/1541, loss: 0.17030425369739532 2023-01-23 13:47:04.057391: step: 902/1541, loss: 0.18128125369548798 2023-01-23 13:47:04.952937: step: 904/1541, loss: 0.057188309729099274 2023-01-23 13:47:05.902911: step: 906/1541, loss: 0.08254411816596985 2023-01-23 13:47:06.819608: step: 908/1541, loss: 0.10944943130016327 2023-01-23 13:47:07.685256: step: 910/1541, loss: 0.10165455937385559 2023-01-23 13:47:08.636114: step: 912/1541, loss: 0.26056772470474243 2023-01-23 13:47:09.521210: step: 914/1541, loss: 0.006104598753154278 2023-01-23 13:47:10.407200: step: 916/1541, loss: 0.037180449813604355 2023-01-23 13:47:11.273141: step: 918/1541, loss: 0.047534290701150894 2023-01-23 13:47:12.190967: step: 920/1541, loss: 0.13486351072788239 2023-01-23 13:47:13.117481: step: 922/1541, loss: 0.011094995774328709 2023-01-23 13:47:14.020701: step: 924/1541, loss: 0.4018276631832123 2023-01-23 13:47:14.911396: step: 926/1541, loss: 0.10673202574253082 2023-01-23 13:47:15.810079: step: 928/1541, loss: 0.05573830381035805 2023-01-23 13:47:16.725749: step: 930/1541, loss: 0.0064310901798307896 2023-01-23 13:47:17.639452: step: 932/1541, loss: 0.13219881057739258 2023-01-23 13:47:18.502622: step: 934/1541, loss: 0.004171323962509632 2023-01-23 13:47:19.373648: step: 936/1541, loss: 0.039166610687971115 2023-01-23 13:47:20.244405: step: 938/1541, loss: 0.11278244853019714 2023-01-23 13:47:21.131772: step: 940/1541, loss: 0.0992962121963501 2023-01-23 13:47:22.000817: step: 942/1541, loss: 0.03980115056037903 2023-01-23 13:47:22.921637: step: 944/1541, loss: 0.0423603430390358 2023-01-23 13:47:23.807559: step: 946/1541, loss: 0.023649321869015694 2023-01-23 13:47:24.697314: step: 948/1541, loss: 0.021139057353138924 2023-01-23 13:47:25.599392: step: 950/1541, loss: 0.05650101229548454 2023-01-23 13:47:26.498623: step: 952/1541, loss: 0.005212497431784868 2023-01-23 13:47:27.379100: step: 954/1541, loss: 0.005589260719716549 2023-01-23 13:47:28.336809: step: 956/1541, loss: 0.03220261633396149 2023-01-23 13:47:29.243526: step: 958/1541, loss: 0.06321306526660919 2023-01-23 13:47:30.133802: step: 960/1541, loss: 0.027245771139860153 2023-01-23 13:47:31.001401: step: 962/1541, loss: 0.0010643531568348408 2023-01-23 13:47:31.903509: step: 964/1541, loss: 0.010270975530147552 2023-01-23 13:47:32.810487: step: 966/1541, loss: 0.024268368259072304 2023-01-23 13:47:33.743050: step: 968/1541, loss: 0.13279590010643005 2023-01-23 13:47:34.614212: step: 970/1541, loss: 0.00010963380918838084 2023-01-23 13:47:35.508245: step: 972/1541, loss: 0.05093638598918915 2023-01-23 13:47:36.398268: step: 974/1541, loss: 0.0013057133182883263 2023-01-23 13:47:37.276322: step: 976/1541, loss: 0.0004031231510452926 2023-01-23 13:47:38.157864: step: 978/1541, loss: 0.00446285167708993 2023-01-23 13:47:39.059450: step: 980/1541, loss: 0.12562665343284607 2023-01-23 13:47:39.945344: step: 982/1541, loss: 0.10122684389352798 2023-01-23 13:47:40.848406: step: 984/1541, loss: 0.04853068292140961 2023-01-23 13:47:41.764981: step: 986/1541, loss: 0.0007472630823031068 2023-01-23 13:47:42.715193: step: 988/1541, loss: 0.07025963068008423 2023-01-23 13:47:43.643384: step: 990/1541, loss: 0.052518609911203384 2023-01-23 13:47:44.574503: step: 992/1541, loss: 0.0008736887248232961 2023-01-23 13:47:45.518417: step: 994/1541, loss: 0.0015234671300277114 2023-01-23 13:47:46.413951: step: 996/1541, loss: 0.0002515031665097922 2023-01-23 13:47:47.288252: step: 998/1541, loss: 0.23159503936767578 2023-01-23 13:47:48.171347: step: 1000/1541, loss: 0.005750239361077547 2023-01-23 13:47:49.083640: step: 1002/1541, loss: 0.40262189507484436 2023-01-23 13:47:49.975619: step: 1004/1541, loss: 0.0052469889633357525 2023-01-23 13:47:50.858276: step: 1006/1541, loss: 0.04063579440116882 2023-01-23 13:47:51.756654: step: 1008/1541, loss: 0.07271448522806168 2023-01-23 13:47:52.683022: step: 1010/1541, loss: 0.028282606974244118 2023-01-23 13:47:53.587192: step: 1012/1541, loss: 0.07955127954483032 2023-01-23 13:47:54.534211: step: 1014/1541, loss: 0.10255607962608337 2023-01-23 13:47:55.419518: step: 1016/1541, loss: 0.020318830385804176 2023-01-23 13:47:56.326629: step: 1018/1541, loss: 0.664966881275177 2023-01-23 13:47:57.213832: step: 1020/1541, loss: 0.00044088016147725284 2023-01-23 13:47:58.135740: step: 1022/1541, loss: 0.019328972324728966 2023-01-23 13:47:59.030806: step: 1024/1541, loss: 0.03563138097524643 2023-01-23 13:47:59.930835: step: 1026/1541, loss: 0.038402438163757324 2023-01-23 13:48:00.870011: step: 1028/1541, loss: 0.16800907254219055 2023-01-23 13:48:01.794855: step: 1030/1541, loss: 0.018447663635015488 2023-01-23 13:48:02.673961: step: 1032/1541, loss: 0.0022497265599668026 2023-01-23 13:48:03.699331: step: 1034/1541, loss: 0.0034468811936676502 2023-01-23 13:48:04.540270: step: 1036/1541, loss: 0.005768901668488979 2023-01-23 13:48:05.408853: step: 1038/1541, loss: 0.42532968521118164 2023-01-23 13:48:06.306824: step: 1040/1541, loss: 0.04253058880567551 2023-01-23 13:48:07.215070: step: 1042/1541, loss: 0.0007632244378328323 2023-01-23 13:48:08.074106: step: 1044/1541, loss: 0.039759986102581024 2023-01-23 13:48:08.967311: step: 1046/1541, loss: 0.033132053911685944 2023-01-23 13:48:09.857620: step: 1048/1541, loss: 0.12386126816272736 2023-01-23 13:48:10.775542: step: 1050/1541, loss: 0.009634098038077354 2023-01-23 13:48:11.699193: step: 1052/1541, loss: 0.002778031397610903 2023-01-23 13:48:12.596380: step: 1054/1541, loss: 0.014519982971251011 2023-01-23 13:48:13.508812: step: 1056/1541, loss: 0.09391579777002335 2023-01-23 13:48:14.437354: step: 1058/1541, loss: 0.09221022576093674 2023-01-23 13:48:15.352458: step: 1060/1541, loss: 0.2006833255290985 2023-01-23 13:48:16.290927: step: 1062/1541, loss: 0.6078553199768066 2023-01-23 13:48:17.147781: step: 1064/1541, loss: 0.039390116930007935 2023-01-23 13:48:18.044387: step: 1066/1541, loss: 0.2793983221054077 2023-01-23 13:48:18.892076: step: 1068/1541, loss: 0.04274805262684822 2023-01-23 13:48:19.838725: step: 1070/1541, loss: 0.1270114630460739 2023-01-23 13:48:20.724013: step: 1072/1541, loss: 0.0649448111653328 2023-01-23 13:48:21.599015: step: 1074/1541, loss: 0.07075513899326324 2023-01-23 13:48:22.483652: step: 1076/1541, loss: 0.012090066447854042 2023-01-23 13:48:23.350437: step: 1078/1541, loss: 0.019925516098737717 2023-01-23 13:48:24.243666: step: 1080/1541, loss: 0.15134723484516144 2023-01-23 13:48:25.173092: step: 1082/1541, loss: 0.01567944511771202 2023-01-23 13:48:26.068583: step: 1084/1541, loss: 0.36088934540748596 2023-01-23 13:48:27.022866: step: 1086/1541, loss: 0.1171928197145462 2023-01-23 13:48:27.914139: step: 1088/1541, loss: 0.026195194572210312 2023-01-23 13:48:28.872487: step: 1090/1541, loss: 0.013366206549108028 2023-01-23 13:48:29.785391: step: 1092/1541, loss: 0.1072927936911583 2023-01-23 13:48:30.654708: step: 1094/1541, loss: 0.22109584510326385 2023-01-23 13:48:31.551757: step: 1096/1541, loss: 0.056227847933769226 2023-01-23 13:48:32.434793: step: 1098/1541, loss: 0.024327300488948822 2023-01-23 13:48:33.276917: step: 1100/1541, loss: 0.10382251441478729 2023-01-23 13:48:34.172063: step: 1102/1541, loss: 0.010615181177854538 2023-01-23 13:48:35.024811: step: 1104/1541, loss: 0.2787073254585266 2023-01-23 13:48:35.899063: step: 1106/1541, loss: 0.011991113424301147 2023-01-23 13:48:36.787574: step: 1108/1541, loss: 0.1709972620010376 2023-01-23 13:48:37.680645: step: 1110/1541, loss: 0.22118407487869263 2023-01-23 13:48:38.564090: step: 1112/1541, loss: 0.13070212304592133 2023-01-23 13:48:39.453176: step: 1114/1541, loss: 0.22195538878440857 2023-01-23 13:48:40.350047: step: 1116/1541, loss: 0.15350033342838287 2023-01-23 13:48:41.208140: step: 1118/1541, loss: 0.008107892237603664 2023-01-23 13:48:42.112965: step: 1120/1541, loss: 0.17988178133964539 2023-01-23 13:48:43.016174: step: 1122/1541, loss: 0.06766431778669357 2023-01-23 13:48:43.932101: step: 1124/1541, loss: 0.18557405471801758 2023-01-23 13:48:44.864548: step: 1126/1541, loss: 0.019521377980709076 2023-01-23 13:48:45.826836: step: 1128/1541, loss: 0.09287773817777634 2023-01-23 13:48:46.659158: step: 1130/1541, loss: 0.010120512917637825 2023-01-23 13:48:47.552317: step: 1132/1541, loss: 0.023588713258504868 2023-01-23 13:48:48.419809: step: 1134/1541, loss: 0.018276285380125046 2023-01-23 13:48:49.327134: step: 1136/1541, loss: 0.11128009855747223 2023-01-23 13:48:50.220205: step: 1138/1541, loss: 0.008398870937526226 2023-01-23 13:48:51.067511: step: 1140/1541, loss: 0.037013594061136246 2023-01-23 13:48:52.047933: step: 1142/1541, loss: 0.07029888778924942 2023-01-23 13:48:52.941766: step: 1144/1541, loss: 0.17643976211547852 2023-01-23 13:48:53.847919: step: 1146/1541, loss: 0.13405850529670715 2023-01-23 13:48:54.769615: step: 1148/1541, loss: 0.010578439570963383 2023-01-23 13:48:55.637344: step: 1150/1541, loss: 0.003054578322917223 2023-01-23 13:48:56.479397: step: 1152/1541, loss: 0.47567418217658997 2023-01-23 13:48:57.365988: step: 1154/1541, loss: 0.4469793438911438 2023-01-23 13:48:58.292360: step: 1156/1541, loss: 0.11082825809717178 2023-01-23 13:48:59.209441: step: 1158/1541, loss: 2.5195648670196533 2023-01-23 13:49:00.075394: step: 1160/1541, loss: 0.04688503220677376 2023-01-23 13:49:00.953856: step: 1162/1541, loss: 0.5031635761260986 2023-01-23 13:49:01.859685: step: 1164/1541, loss: 0.05983990430831909 2023-01-23 13:49:02.754020: step: 1166/1541, loss: 0.17317040264606476 2023-01-23 13:49:03.685001: step: 1168/1541, loss: 0.018056001514196396 2023-01-23 13:49:04.564919: step: 1170/1541, loss: 0.026988347992300987 2023-01-23 13:49:05.419262: step: 1172/1541, loss: 0.004257995169609785 2023-01-23 13:49:06.286391: step: 1174/1541, loss: 0.06969109177589417 2023-01-23 13:49:07.190042: step: 1176/1541, loss: 0.005515883676707745 2023-01-23 13:49:08.085088: step: 1178/1541, loss: 0.016989806666970253 2023-01-23 13:49:08.933766: step: 1180/1541, loss: 0.20929251611232758 2023-01-23 13:49:09.814320: step: 1182/1541, loss: 0.00043612130684778094 2023-01-23 13:49:10.706563: step: 1184/1541, loss: 0.0009416227112524211 2023-01-23 13:49:11.620000: step: 1186/1541, loss: 0.7729610204696655 2023-01-23 13:49:12.499368: step: 1188/1541, loss: 0.02622940018773079 2023-01-23 13:49:13.397025: step: 1190/1541, loss: 0.1257881373167038 2023-01-23 13:49:14.274109: step: 1192/1541, loss: 0.12846817076206207 2023-01-23 13:49:15.181836: step: 1194/1541, loss: 0.13827650249004364 2023-01-23 13:49:16.096666: step: 1196/1541, loss: 0.005593533162027597 2023-01-23 13:49:17.032892: step: 1198/1541, loss: 0.003273226087912917 2023-01-23 13:49:17.906490: step: 1200/1541, loss: 0.003070464124903083 2023-01-23 13:49:18.805456: step: 1202/1541, loss: 0.24412117898464203 2023-01-23 13:49:19.684352: step: 1204/1541, loss: 0.0009670940926298499 2023-01-23 13:49:20.553857: step: 1206/1541, loss: 0.0012680896325036883 2023-01-23 13:49:21.461049: step: 1208/1541, loss: 0.008086340501904488 2023-01-23 13:49:22.332660: step: 1210/1541, loss: 0.0006555754807777703 2023-01-23 13:49:23.201073: step: 1212/1541, loss: 0.037787895649671555 2023-01-23 13:49:24.137677: step: 1214/1541, loss: 0.3075808882713318 2023-01-23 13:49:25.063684: step: 1216/1541, loss: 0.09243491291999817 2023-01-23 13:49:25.946909: step: 1218/1541, loss: 1.1911284923553467 2023-01-23 13:49:26.829438: step: 1220/1541, loss: 0.0053827292285859585 2023-01-23 13:49:27.713707: step: 1222/1541, loss: 0.05537646636366844 2023-01-23 13:49:28.605849: step: 1224/1541, loss: 0.24232153594493866 2023-01-23 13:49:29.519445: step: 1226/1541, loss: 0.32074886560440063 2023-01-23 13:49:30.454987: step: 1228/1541, loss: 0.061414435505867004 2023-01-23 13:49:31.332454: step: 1230/1541, loss: 0.011428320780396461 2023-01-23 13:49:32.277678: step: 1232/1541, loss: 0.010613876394927502 2023-01-23 13:49:33.132627: step: 1234/1541, loss: 0.10948833078145981 2023-01-23 13:49:34.016853: step: 1236/1541, loss: 0.21405233442783356 2023-01-23 13:49:34.891570: step: 1238/1541, loss: 0.1311015784740448 2023-01-23 13:49:35.765531: step: 1240/1541, loss: 0.018756041303277016 2023-01-23 13:49:36.647188: step: 1242/1541, loss: 0.22843433916568756 2023-01-23 13:49:37.623857: step: 1244/1541, loss: 0.0020600638817995787 2023-01-23 13:49:38.515623: step: 1246/1541, loss: 0.11115092039108276 2023-01-23 13:49:39.393863: step: 1248/1541, loss: 0.04351053386926651 2023-01-23 13:49:40.265951: step: 1250/1541, loss: 0.23380932211875916 2023-01-23 13:49:41.170172: step: 1252/1541, loss: 0.008830292150378227 2023-01-23 13:49:42.013488: step: 1254/1541, loss: 0.05702408403158188 2023-01-23 13:49:42.916697: step: 1256/1541, loss: 0.3089410066604614 2023-01-23 13:49:43.864429: step: 1258/1541, loss: 0.3200015127658844 2023-01-23 13:49:44.801627: step: 1260/1541, loss: 0.02977409027516842 2023-01-23 13:49:45.752340: step: 1262/1541, loss: 0.032569680362939835 2023-01-23 13:49:46.631121: step: 1264/1541, loss: 0.2206566482782364 2023-01-23 13:49:47.525174: step: 1266/1541, loss: 0.04610970616340637 2023-01-23 13:49:48.457386: step: 1268/1541, loss: 0.10672781616449356 2023-01-23 13:49:49.319387: step: 1270/1541, loss: 0.09091871976852417 2023-01-23 13:49:50.281199: step: 1272/1541, loss: 0.0701715499162674 2023-01-23 13:49:51.209502: step: 1274/1541, loss: 0.0019056638702750206 2023-01-23 13:49:52.078166: step: 1276/1541, loss: 1.2223286628723145 2023-01-23 13:49:53.009102: step: 1278/1541, loss: 0.0015511823585256934 2023-01-23 13:49:53.887517: step: 1280/1541, loss: 0.018585771322250366 2023-01-23 13:49:54.793778: step: 1282/1541, loss: 1.1706571578979492 2023-01-23 13:49:55.759447: step: 1284/1541, loss: 0.002001524204388261 2023-01-23 13:49:56.681990: step: 1286/1541, loss: 0.007746764458715916 2023-01-23 13:49:57.567154: step: 1288/1541, loss: 0.04936820641160011 2023-01-23 13:49:58.458250: step: 1290/1541, loss: 0.38941988348960876 2023-01-23 13:49:59.351282: step: 1292/1541, loss: 0.00727837672457099 2023-01-23 13:50:00.237078: step: 1294/1541, loss: 0.010406305082142353 2023-01-23 13:50:01.091483: step: 1296/1541, loss: 0.11270023882389069 2023-01-23 13:50:02.069553: step: 1298/1541, loss: 0.09949423372745514 2023-01-23 13:50:02.960661: step: 1300/1541, loss: 0.24899286031723022 2023-01-23 13:50:03.827691: step: 1302/1541, loss: 0.015560094267129898 2023-01-23 13:50:04.736907: step: 1304/1541, loss: 0.204633891582489 2023-01-23 13:50:05.665408: step: 1306/1541, loss: 0.003927095327526331 2023-01-23 13:50:06.548879: step: 1308/1541, loss: 0.13840459287166595 2023-01-23 13:50:07.431090: step: 1310/1541, loss: 0.006067188456654549 2023-01-23 13:50:08.318525: step: 1312/1541, loss: 0.013284137472510338 2023-01-23 13:50:09.221788: step: 1314/1541, loss: 0.19141367077827454 2023-01-23 13:50:10.121710: step: 1316/1541, loss: 0.3110189139842987 2023-01-23 13:50:11.000819: step: 1318/1541, loss: 0.004802853334695101 2023-01-23 13:50:11.877251: step: 1320/1541, loss: 0.07784724235534668 2023-01-23 13:50:12.767249: step: 1322/1541, loss: 0.0005527895409613848 2023-01-23 13:50:13.732081: step: 1324/1541, loss: 0.007084481418132782 2023-01-23 13:50:14.585839: step: 1326/1541, loss: 0.05495910719037056 2023-01-23 13:50:15.441043: step: 1328/1541, loss: 0.14168724417686462 2023-01-23 13:50:16.340011: step: 1330/1541, loss: 0.15366096794605255 2023-01-23 13:50:17.237817: step: 1332/1541, loss: 0.2553420662879944 2023-01-23 13:50:18.106536: step: 1334/1541, loss: 0.09264848381280899 2023-01-23 13:50:18.991028: step: 1336/1541, loss: 0.14526504278182983 2023-01-23 13:50:19.872076: step: 1338/1541, loss: 0.01796007715165615 2023-01-23 13:50:20.792198: step: 1340/1541, loss: 0.01866740733385086 2023-01-23 13:50:21.721935: step: 1342/1541, loss: 0.002816471504047513 2023-01-23 13:50:22.620605: step: 1344/1541, loss: 0.006390634924173355 2023-01-23 13:50:23.520181: step: 1346/1541, loss: 0.2860090136528015 2023-01-23 13:50:24.431076: step: 1348/1541, loss: 0.02633778005838394 2023-01-23 13:50:25.321463: step: 1350/1541, loss: 0.09334146976470947 2023-01-23 13:50:26.222844: step: 1352/1541, loss: 0.08318118751049042 2023-01-23 13:50:27.102109: step: 1354/1541, loss: 0.23190169036388397 2023-01-23 13:50:27.990819: step: 1356/1541, loss: 0.07193995267152786 2023-01-23 13:50:28.912556: step: 1358/1541, loss: 0.026525627821683884 2023-01-23 13:50:29.775353: step: 1360/1541, loss: 0.003640363924205303 2023-01-23 13:50:30.667323: step: 1362/1541, loss: 0.14152435958385468 2023-01-23 13:50:31.537865: step: 1364/1541, loss: 0.04529128968715668 2023-01-23 13:50:32.423397: step: 1366/1541, loss: 0.021522156894207 2023-01-23 13:50:33.316159: step: 1368/1541, loss: 0.023346848785877228 2023-01-23 13:50:34.220469: step: 1370/1541, loss: 0.025278039276599884 2023-01-23 13:50:35.125706: step: 1372/1541, loss: 0.015458198264241219 2023-01-23 13:50:36.048484: step: 1374/1541, loss: 0.1086038202047348 2023-01-23 13:50:36.935667: step: 1376/1541, loss: 0.006501100491732359 2023-01-23 13:50:37.883309: step: 1378/1541, loss: 0.01141156442463398 2023-01-23 13:50:38.799831: step: 1380/1541, loss: 0.024936493486166 2023-01-23 13:50:39.712971: step: 1382/1541, loss: 0.2486206740140915 2023-01-23 13:50:40.647722: step: 1384/1541, loss: 0.012024548836052418 2023-01-23 13:50:41.544094: step: 1386/1541, loss: 0.07989849895238876 2023-01-23 13:50:42.457426: step: 1388/1541, loss: 0.01451414730399847 2023-01-23 13:50:43.364132: step: 1390/1541, loss: 0.14521315693855286 2023-01-23 13:50:44.264234: step: 1392/1541, loss: 0.005243238527327776 2023-01-23 13:50:45.130438: step: 1394/1541, loss: 0.020432623103260994 2023-01-23 13:50:46.030807: step: 1396/1541, loss: 0.0709814727306366 2023-01-23 13:50:46.951684: step: 1398/1541, loss: 0.17491808533668518 2023-01-23 13:50:47.823843: step: 1400/1541, loss: 0.012715958058834076 2023-01-23 13:50:48.710021: step: 1402/1541, loss: 0.1370498389005661 2023-01-23 13:50:49.589348: step: 1404/1541, loss: 0.0027279192581772804 2023-01-23 13:50:50.518041: step: 1406/1541, loss: 0.025017045438289642 2023-01-23 13:50:51.397443: step: 1408/1541, loss: 0.03679659962654114 2023-01-23 13:50:52.311642: step: 1410/1541, loss: 0.012565458193421364 2023-01-23 13:50:53.196836: step: 1412/1541, loss: 0.03192506358027458 2023-01-23 13:50:54.097456: step: 1414/1541, loss: 0.2729776203632355 2023-01-23 13:50:54.988895: step: 1416/1541, loss: 0.05564601346850395 2023-01-23 13:50:55.873392: step: 1418/1541, loss: 0.005805878899991512 2023-01-23 13:50:56.757236: step: 1420/1541, loss: 0.02395675703883171 2023-01-23 13:50:57.686429: step: 1422/1541, loss: 0.004977133125066757 2023-01-23 13:50:58.614148: step: 1424/1541, loss: 0.038205284625291824 2023-01-23 13:50:59.468971: step: 1426/1541, loss: 0.12126496434211731 2023-01-23 13:51:00.427320: step: 1428/1541, loss: 0.39341771602630615 2023-01-23 13:51:01.302368: step: 1430/1541, loss: 0.010460371151566505 2023-01-23 13:51:02.236358: step: 1432/1541, loss: 0.011211884208023548 2023-01-23 13:51:03.132084: step: 1434/1541, loss: 0.24398089945316315 2023-01-23 13:51:04.072591: step: 1436/1541, loss: 0.16010764241218567 2023-01-23 13:51:04.966135: step: 1438/1541, loss: 0.10788280516862869 2023-01-23 13:51:05.893984: step: 1440/1541, loss: 0.01180784311145544 2023-01-23 13:51:06.798705: step: 1442/1541, loss: 0.14753952622413635 2023-01-23 13:51:07.705714: step: 1444/1541, loss: 0.009109413251280785 2023-01-23 13:51:08.603473: step: 1446/1541, loss: 0.03481364995241165 2023-01-23 13:51:09.542353: step: 1448/1541, loss: 0.006654584780335426 2023-01-23 13:51:10.437705: step: 1450/1541, loss: 0.03603819012641907 2023-01-23 13:51:11.296319: step: 1452/1541, loss: 0.041050590574741364 2023-01-23 13:51:12.179222: step: 1454/1541, loss: 0.015538666397333145 2023-01-23 13:51:13.090663: step: 1456/1541, loss: 0.00858769565820694 2023-01-23 13:51:13.954396: step: 1458/1541, loss: 0.22349020838737488 2023-01-23 13:51:14.815046: step: 1460/1541, loss: 0.03902633115649223 2023-01-23 13:51:15.719023: step: 1462/1541, loss: 0.02714218758046627 2023-01-23 13:51:16.618403: step: 1464/1541, loss: 0.009999044239521027 2023-01-23 13:51:17.543876: step: 1466/1541, loss: 0.006435755640268326 2023-01-23 13:51:18.432553: step: 1468/1541, loss: 0.005882742814719677 2023-01-23 13:51:19.274599: step: 1470/1541, loss: 0.08972463756799698 2023-01-23 13:51:20.158398: step: 1472/1541, loss: 0.09670059382915497 2023-01-23 13:51:21.100234: step: 1474/1541, loss: 0.062369320541620255 2023-01-23 13:51:21.970034: step: 1476/1541, loss: 0.045896977186203 2023-01-23 13:51:22.873965: step: 1478/1541, loss: 0.10240461677312851 2023-01-23 13:51:23.745863: step: 1480/1541, loss: 0.02728465385735035 2023-01-23 13:51:24.675117: step: 1482/1541, loss: 0.0073363385163247585 2023-01-23 13:51:25.569744: step: 1484/1541, loss: 0.13295038044452667 2023-01-23 13:51:26.531994: step: 1486/1541, loss: 0.03270035237073898 2023-01-23 13:51:27.423238: step: 1488/1541, loss: 0.7548205256462097 2023-01-23 13:51:28.279925: step: 1490/1541, loss: 0.04322776570916176 2023-01-23 13:51:29.294511: step: 1492/1541, loss: 0.0043805805034935474 2023-01-23 13:51:30.192434: step: 1494/1541, loss: 0.06522413343191147 2023-01-23 13:51:31.087791: step: 1496/1541, loss: 0.6764643788337708 2023-01-23 13:51:32.039777: step: 1498/1541, loss: 0.0013626462314277887 2023-01-23 13:51:32.930609: step: 1500/1541, loss: 0.03119550831615925 2023-01-23 13:51:33.817024: step: 1502/1541, loss: 0.010735983029007912 2023-01-23 13:51:34.730866: step: 1504/1541, loss: 0.0038726942148059607 2023-01-23 13:51:35.591100: step: 1506/1541, loss: 0.006417033728212118 2023-01-23 13:51:36.477890: step: 1508/1541, loss: 0.41171732544898987 2023-01-23 13:51:37.387811: step: 1510/1541, loss: 0.001094859093427658 2023-01-23 13:51:38.292999: step: 1512/1541, loss: 0.0015105867059901357 2023-01-23 13:51:39.239685: step: 1514/1541, loss: 0.035040467977523804 2023-01-23 13:51:40.160979: step: 1516/1541, loss: 0.008911394514143467 2023-01-23 13:51:41.020372: step: 1518/1541, loss: 0.03451605141162872 2023-01-23 13:51:41.919617: step: 1520/1541, loss: 0.04516228660941124 2023-01-23 13:51:42.776963: step: 1522/1541, loss: 0.0007461255881935358 2023-01-23 13:51:43.662089: step: 1524/1541, loss: 0.07739850878715515 2023-01-23 13:51:44.569848: step: 1526/1541, loss: 0.2795867323875427 2023-01-23 13:51:45.457589: step: 1528/1541, loss: 0.2636241912841797 2023-01-23 13:51:46.312378: step: 1530/1541, loss: 0.006313674617558718 2023-01-23 13:51:47.253125: step: 1532/1541, loss: 0.15105970203876495 2023-01-23 13:51:48.146856: step: 1534/1541, loss: 0.011402379721403122 2023-01-23 13:51:49.052954: step: 1536/1541, loss: 0.2228054702281952 2023-01-23 13:51:49.977965: step: 1538/1541, loss: 0.0019098209450021386 2023-01-23 13:51:50.865944: step: 1540/1541, loss: 0.0495571494102478 2023-01-23 13:51:51.754039: step: 1542/1541, loss: 0.0008200726588256657 2023-01-23 13:51:52.606850: step: 1544/1541, loss: 0.0009869778295978904 2023-01-23 13:51:53.483023: step: 1546/1541, loss: 0.10580582171678543 2023-01-23 13:51:54.354483: step: 1548/1541, loss: 0.22964489459991455 2023-01-23 13:51:55.282698: step: 1550/1541, loss: 0.7519454956054688 2023-01-23 13:51:56.174794: step: 1552/1541, loss: 0.0023253890685737133 2023-01-23 13:51:57.072250: step: 1554/1541, loss: 0.009112834930419922 2023-01-23 13:51:57.978584: step: 1556/1541, loss: 0.07405531406402588 2023-01-23 13:51:58.870905: step: 1558/1541, loss: 0.13716979324817657 2023-01-23 13:51:59.771070: step: 1560/1541, loss: 0.009163793176412582 2023-01-23 13:52:00.656077: step: 1562/1541, loss: 0.0024770149029791355 2023-01-23 13:52:01.628037: step: 1564/1541, loss: 0.15983366966247559 2023-01-23 13:52:02.520395: step: 1566/1541, loss: 0.23730725049972534 2023-01-23 13:52:03.395064: step: 1568/1541, loss: 0.0018472010269761086 2023-01-23 13:52:04.260676: step: 1570/1541, loss: 0.008689751848578453 2023-01-23 13:52:05.134616: step: 1572/1541, loss: 0.04361684247851372 2023-01-23 13:52:06.098426: step: 1574/1541, loss: 0.006224913522601128 2023-01-23 13:52:07.045621: step: 1576/1541, loss: 0.013925373554229736 2023-01-23 13:52:07.971045: step: 1578/1541, loss: 0.004252703860402107 2023-01-23 13:52:08.893136: step: 1580/1541, loss: 0.01459326408803463 2023-01-23 13:52:09.796529: step: 1582/1541, loss: 0.005372129380702972 2023-01-23 13:52:10.652836: step: 1584/1541, loss: 0.07345418632030487 2023-01-23 13:52:11.554052: step: 1586/1541, loss: 0.016637027263641357 2023-01-23 13:52:12.467711: step: 1588/1541, loss: 0.118306003510952 2023-01-23 13:52:13.352547: step: 1590/1541, loss: 0.040749888867139816 2023-01-23 13:52:14.231518: step: 1592/1541, loss: 0.14959952235221863 2023-01-23 13:52:15.139850: step: 1594/1541, loss: 0.05661753565073013 2023-01-23 13:52:16.029029: step: 1596/1541, loss: 0.4293776750564575 2023-01-23 13:52:16.898408: step: 1598/1541, loss: 0.014124070294201374 2023-01-23 13:52:17.782661: step: 1600/1541, loss: 0.004589707124978304 2023-01-23 13:52:18.688271: step: 1602/1541, loss: 0.06719052791595459 2023-01-23 13:52:19.594262: step: 1604/1541, loss: 0.014271754771471024 2023-01-23 13:52:20.493459: step: 1606/1541, loss: 0.006322608329355717 2023-01-23 13:52:21.370753: step: 1608/1541, loss: 0.000860665226355195 2023-01-23 13:52:22.247527: step: 1610/1541, loss: 0.03286782652139664 2023-01-23 13:52:23.158616: step: 1612/1541, loss: 0.143119677901268 2023-01-23 13:52:24.070941: step: 1614/1541, loss: 0.14124922454357147 2023-01-23 13:52:24.953031: step: 1616/1541, loss: 0.03681224584579468 2023-01-23 13:52:25.852910: step: 1618/1541, loss: 0.01752808317542076 2023-01-23 13:52:26.757671: step: 1620/1541, loss: 0.19378888607025146 2023-01-23 13:52:27.694512: step: 1622/1541, loss: 0.07529474049806595 2023-01-23 13:52:28.549040: step: 1624/1541, loss: 0.14134323596954346 2023-01-23 13:52:29.498171: step: 1626/1541, loss: 0.8631657361984253 2023-01-23 13:52:30.372427: step: 1628/1541, loss: 0.0025916623417288065 2023-01-23 13:52:31.267995: step: 1630/1541, loss: 0.018251020461320877 2023-01-23 13:52:32.174313: step: 1632/1541, loss: 0.1666615754365921 2023-01-23 13:52:33.060204: step: 1634/1541, loss: 0.044888488948345184 2023-01-23 13:52:33.927862: step: 1636/1541, loss: 0.012707230634987354 2023-01-23 13:52:34.827778: step: 1638/1541, loss: 0.0005321461358107626 2023-01-23 13:52:35.712696: step: 1640/1541, loss: 0.11132626235485077 2023-01-23 13:52:36.621144: step: 1642/1541, loss: 0.012381982058286667 2023-01-23 13:52:37.540790: step: 1644/1541, loss: 0.036958396434783936 2023-01-23 13:52:38.502866: step: 1646/1541, loss: 0.1607763171195984 2023-01-23 13:52:39.435235: step: 1648/1541, loss: 0.021113606169819832 2023-01-23 13:52:40.322209: step: 1650/1541, loss: 0.002355261705815792 2023-01-23 13:52:41.260507: step: 1652/1541, loss: 0.0395466648042202 2023-01-23 13:52:42.132850: step: 1654/1541, loss: 0.15506359934806824 2023-01-23 13:52:43.072556: step: 1656/1541, loss: 0.01814577728509903 2023-01-23 13:52:43.954921: step: 1658/1541, loss: 0.3639950752258301 2023-01-23 13:52:44.832005: step: 1660/1541, loss: 0.012448267079889774 2023-01-23 13:52:45.700056: step: 1662/1541, loss: 0.17347007989883423 2023-01-23 13:52:46.592272: step: 1664/1541, loss: 2.3524506092071533 2023-01-23 13:52:47.490700: step: 1666/1541, loss: 0.0101325623691082 2023-01-23 13:52:48.360090: step: 1668/1541, loss: 0.06992980092763901 2023-01-23 13:52:49.255954: step: 1670/1541, loss: 0.07094164937734604 2023-01-23 13:52:50.231989: step: 1672/1541, loss: 0.005016685929149389 2023-01-23 13:52:51.154982: step: 1674/1541, loss: 0.37573912739753723 2023-01-23 13:52:52.176636: step: 1676/1541, loss: 0.04733767360448837 2023-01-23 13:52:53.067551: step: 1678/1541, loss: 0.024756550788879395 2023-01-23 13:52:53.982827: step: 1680/1541, loss: 0.007060361094772816 2023-01-23 13:52:54.862156: step: 1682/1541, loss: 0.1463789939880371 2023-01-23 13:52:55.783196: step: 1684/1541, loss: 0.2839793264865875 2023-01-23 13:52:56.662574: step: 1686/1541, loss: 0.011564607731997967 2023-01-23 13:52:57.571849: step: 1688/1541, loss: 0.029992489144206047 2023-01-23 13:52:58.481515: step: 1690/1541, loss: 0.21046507358551025 2023-01-23 13:52:59.351511: step: 1692/1541, loss: 0.02242862991988659 2023-01-23 13:53:00.256142: step: 1694/1541, loss: 0.09312772005796432 2023-01-23 13:53:01.182681: step: 1696/1541, loss: 0.02596467174589634 2023-01-23 13:53:02.193411: step: 1698/1541, loss: 0.07192492485046387 2023-01-23 13:53:03.076806: step: 1700/1541, loss: 0.38844799995422363 2023-01-23 13:53:03.943312: step: 1702/1541, loss: 0.0016214895294979215 2023-01-23 13:53:04.818219: step: 1704/1541, loss: 0.004088807385414839 2023-01-23 13:53:05.760977: step: 1706/1541, loss: 0.004351894836872816 2023-01-23 13:53:06.659630: step: 1708/1541, loss: 0.0018030868377536535 2023-01-23 13:53:07.599505: step: 1710/1541, loss: 0.06575045734643936 2023-01-23 13:53:08.461938: step: 1712/1541, loss: 0.00821829866617918 2023-01-23 13:53:09.312335: step: 1714/1541, loss: 0.2451414167881012 2023-01-23 13:53:10.228367: step: 1716/1541, loss: 0.03490203246474266 2023-01-23 13:53:11.122799: step: 1718/1541, loss: 0.008267232216894627 2023-01-23 13:53:12.039142: step: 1720/1541, loss: 0.05970313027501106 2023-01-23 13:53:12.922206: step: 1722/1541, loss: 0.22803570330142975 2023-01-23 13:53:13.824523: step: 1724/1541, loss: 0.11333595216274261 2023-01-23 13:53:14.730754: step: 1726/1541, loss: 0.11831510812044144 2023-01-23 13:53:15.627498: step: 1728/1541, loss: 0.09337607771158218 2023-01-23 13:53:16.524202: step: 1730/1541, loss: 0.2787192165851593 2023-01-23 13:53:17.420233: step: 1732/1541, loss: 0.01101865153759718 2023-01-23 13:53:18.317355: step: 1734/1541, loss: 0.09154574573040009 2023-01-23 13:53:19.245128: step: 1736/1541, loss: 0.02023325115442276 2023-01-23 13:53:20.155689: step: 1738/1541, loss: 0.006396290846168995 2023-01-23 13:53:20.994131: step: 1740/1541, loss: 0.16722151637077332 2023-01-23 13:53:21.908673: step: 1742/1541, loss: 0.018412059172987938 2023-01-23 13:53:22.786989: step: 1744/1541, loss: 0.09422669559717178 2023-01-23 13:53:23.663242: step: 1746/1541, loss: 0.009884566999971867 2023-01-23 13:53:24.610295: step: 1748/1541, loss: 0.05466711148619652 2023-01-23 13:53:25.548199: step: 1750/1541, loss: 0.012781446799635887 2023-01-23 13:53:26.482232: step: 1752/1541, loss: 0.008764313533902168 2023-01-23 13:53:27.464170: step: 1754/1541, loss: 0.00329984282143414 2023-01-23 13:53:28.349888: step: 1756/1541, loss: 0.015266355127096176 2023-01-23 13:53:29.242636: step: 1758/1541, loss: 0.002544356742873788 2023-01-23 13:53:30.166795: step: 1760/1541, loss: 0.28491640090942383 2023-01-23 13:53:31.076530: step: 1762/1541, loss: 0.09262404590845108 2023-01-23 13:53:31.954362: step: 1764/1541, loss: 0.030571959912776947 2023-01-23 13:53:32.816482: step: 1766/1541, loss: 0.08795850723981857 2023-01-23 13:53:33.696994: step: 1768/1541, loss: 0.009799148887395859 2023-01-23 13:53:34.621565: step: 1770/1541, loss: 0.3826794922351837 2023-01-23 13:53:35.514789: step: 1772/1541, loss: 0.24093814194202423 2023-01-23 13:53:36.445782: step: 1774/1541, loss: 0.01155153475701809 2023-01-23 13:53:37.345674: step: 1776/1541, loss: 0.17310234904289246 2023-01-23 13:53:38.200896: step: 1778/1541, loss: 0.2590431869029999 2023-01-23 13:53:39.108580: step: 1780/1541, loss: 0.011677958071231842 2023-01-23 13:53:40.084455: step: 1782/1541, loss: 0.12995614111423492 2023-01-23 13:53:40.985503: step: 1784/1541, loss: 0.021041177213191986 2023-01-23 13:53:41.891630: step: 1786/1541, loss: 0.3093246519565582 2023-01-23 13:53:42.761102: step: 1788/1541, loss: 0.04530685767531395 2023-01-23 13:53:43.686639: step: 1790/1541, loss: 0.17157362401485443 2023-01-23 13:53:44.574878: step: 1792/1541, loss: 0.09568136930465698 2023-01-23 13:53:45.482975: step: 1794/1541, loss: 0.024038655683398247 2023-01-23 13:53:46.373142: step: 1796/1541, loss: 0.00027941790176555514 2023-01-23 13:53:47.316918: step: 1798/1541, loss: 0.1429162472486496 2023-01-23 13:53:48.322928: step: 1800/1541, loss: 0.09139340370893478 2023-01-23 13:53:49.274585: step: 1802/1541, loss: 0.36251893639564514 2023-01-23 13:53:50.165092: step: 1804/1541, loss: 0.47716307640075684 2023-01-23 13:53:51.030928: step: 1806/1541, loss: 0.02496267855167389 2023-01-23 13:53:51.911669: step: 1808/1541, loss: 0.0019748525228351355 2023-01-23 13:53:52.800828: step: 1810/1541, loss: 0.15511469542980194 2023-01-23 13:53:53.760650: step: 1812/1541, loss: 0.015562442131340504 2023-01-23 13:53:54.643696: step: 1814/1541, loss: 0.08226703107357025 2023-01-23 13:53:55.561791: step: 1816/1541, loss: 0.026492085307836533 2023-01-23 13:53:56.499866: step: 1818/1541, loss: 0.04084540158510208 2023-01-23 13:53:57.373868: step: 1820/1541, loss: 0.27567195892333984 2023-01-23 13:53:58.282132: step: 1822/1541, loss: 0.2097344994544983 2023-01-23 13:53:59.159704: step: 1824/1541, loss: 0.07152089476585388 2023-01-23 13:54:00.078935: step: 1826/1541, loss: 0.21875186264514923 2023-01-23 13:54:00.990840: step: 1828/1541, loss: 0.02453640289604664 2023-01-23 13:54:01.945944: step: 1830/1541, loss: 0.20455797016620636 2023-01-23 13:54:02.846182: step: 1832/1541, loss: 0.18359822034835815 2023-01-23 13:54:03.725923: step: 1834/1541, loss: 0.2077162265777588 2023-01-23 13:54:04.638361: step: 1836/1541, loss: 0.1830747425556183 2023-01-23 13:54:05.576166: step: 1838/1541, loss: 0.006753591820597649 2023-01-23 13:54:06.452466: step: 1840/1541, loss: 0.0972810834646225 2023-01-23 13:54:07.372736: step: 1842/1541, loss: 0.030362393707036972 2023-01-23 13:54:08.284780: step: 1844/1541, loss: 0.01204657182097435 2023-01-23 13:54:09.156854: step: 1846/1541, loss: 0.8648999333381653 2023-01-23 13:54:10.064425: step: 1848/1541, loss: 0.034099072217941284 2023-01-23 13:54:10.936580: step: 1850/1541, loss: 0.2990572452545166 2023-01-23 13:54:11.830173: step: 1852/1541, loss: 0.18088418245315552 2023-01-23 13:54:12.814749: step: 1854/1541, loss: 0.1119888424873352 2023-01-23 13:54:13.705088: step: 1856/1541, loss: 0.003467787755653262 2023-01-23 13:54:14.596596: step: 1858/1541, loss: 0.0150911845266819 2023-01-23 13:54:15.480083: step: 1860/1541, loss: 0.04967474937438965 2023-01-23 13:54:16.377212: step: 1862/1541, loss: 0.03623102977871895 2023-01-23 13:54:17.299731: step: 1864/1541, loss: 0.005578772164881229 2023-01-23 13:54:18.195949: step: 1866/1541, loss: 0.5776165127754211 2023-01-23 13:54:19.141629: step: 1868/1541, loss: 0.06585343927145004 2023-01-23 13:54:19.977661: step: 1870/1541, loss: 0.013152139261364937 2023-01-23 13:54:20.877091: step: 1872/1541, loss: 0.008198819123208523 2023-01-23 13:54:21.774303: step: 1874/1541, loss: 0.05704259127378464 2023-01-23 13:54:22.744641: step: 1876/1541, loss: 0.013764875940978527 2023-01-23 13:54:23.653298: step: 1878/1541, loss: 0.2761753797531128 2023-01-23 13:54:24.526204: step: 1880/1541, loss: 0.01864388771355152 2023-01-23 13:54:25.437535: step: 1882/1541, loss: 0.32505470514297485 2023-01-23 13:54:26.363322: step: 1884/1541, loss: 0.004387436900287867 2023-01-23 13:54:27.259732: step: 1886/1541, loss: 0.04241263493895531 2023-01-23 13:54:28.139143: step: 1888/1541, loss: 0.019039258360862732 2023-01-23 13:54:29.081760: step: 1890/1541, loss: 0.10084550082683563 2023-01-23 13:54:29.991424: step: 1892/1541, loss: 0.025885164737701416 2023-01-23 13:54:30.950948: step: 1894/1541, loss: 0.009148719720542431 2023-01-23 13:54:31.938053: step: 1896/1541, loss: 0.08689752221107483 2023-01-23 13:54:32.859295: step: 1898/1541, loss: 0.012796293012797832 2023-01-23 13:54:33.767291: step: 1900/1541, loss: 1.3331230878829956 2023-01-23 13:54:34.664441: step: 1902/1541, loss: 0.2779817581176758 2023-01-23 13:54:35.623615: step: 1904/1541, loss: 0.02316620573401451 2023-01-23 13:54:36.570300: step: 1906/1541, loss: 0.040874630212783813 2023-01-23 13:54:37.535040: step: 1908/1541, loss: 0.06463762372732162 2023-01-23 13:54:38.479679: step: 1910/1541, loss: 0.002203871263191104 2023-01-23 13:54:39.433605: step: 1912/1541, loss: 0.014719548635184765 2023-01-23 13:54:40.286256: step: 1914/1541, loss: 0.027900634333491325 2023-01-23 13:54:41.175832: step: 1916/1541, loss: 0.013836732134222984 2023-01-23 13:54:42.087931: step: 1918/1541, loss: 0.013713355176150799 2023-01-23 13:54:42.989818: step: 1920/1541, loss: 0.022433282807469368 2023-01-23 13:54:43.902720: step: 1922/1541, loss: 0.020577777177095413 2023-01-23 13:54:44.800015: step: 1924/1541, loss: 0.1028762012720108 2023-01-23 13:54:45.676643: step: 1926/1541, loss: 0.1484283208847046 2023-01-23 13:54:46.566277: step: 1928/1541, loss: 0.6236924529075623 2023-01-23 13:54:47.534410: step: 1930/1541, loss: 0.16369818150997162 2023-01-23 13:54:48.443874: step: 1932/1541, loss: 0.01455769408494234 2023-01-23 13:54:49.303904: step: 1934/1541, loss: 0.025052640587091446 2023-01-23 13:54:50.193317: step: 1936/1541, loss: 0.010112019255757332 2023-01-23 13:54:51.074692: step: 1938/1541, loss: 0.002696627750992775 2023-01-23 13:54:51.966193: step: 1940/1541, loss: 0.1231241226196289 2023-01-23 13:54:52.937353: step: 1942/1541, loss: 0.015582002699375153 2023-01-23 13:54:53.849822: step: 1944/1541, loss: 0.56369549036026 2023-01-23 13:54:54.768626: step: 1946/1541, loss: 0.0030954882968217134 2023-01-23 13:54:55.651055: step: 1948/1541, loss: 0.03508135303854942 2023-01-23 13:54:56.571269: step: 1950/1541, loss: 0.11656718701124191 2023-01-23 13:54:57.436064: step: 1952/1541, loss: 0.01282709650695324 2023-01-23 13:54:58.318070: step: 1954/1541, loss: 0.019007708877325058 2023-01-23 13:54:59.200380: step: 1956/1541, loss: 0.0024731787852942944 2023-01-23 13:55:00.085741: step: 1958/1541, loss: 0.0013079793425276875 2023-01-23 13:55:00.980213: step: 1960/1541, loss: 0.011411354877054691 2023-01-23 13:55:01.934806: step: 1962/1541, loss: 0.22354643046855927 2023-01-23 13:55:02.846725: step: 1964/1541, loss: 0.003008573781698942 2023-01-23 13:55:03.720190: step: 1966/1541, loss: 0.03479743003845215 2023-01-23 13:55:04.598983: step: 1968/1541, loss: 0.02709280699491501 2023-01-23 13:55:05.483238: step: 1970/1541, loss: 0.12461584061384201 2023-01-23 13:55:06.358344: step: 1972/1541, loss: 0.013491091318428516 2023-01-23 13:55:07.273409: step: 1974/1541, loss: 0.0057317763566970825 2023-01-23 13:55:08.199590: step: 1976/1541, loss: 0.07886990159749985 2023-01-23 13:55:09.086102: step: 1978/1541, loss: 0.006288223899900913 2023-01-23 13:55:10.012842: step: 1980/1541, loss: 0.026631120592355728 2023-01-23 13:55:10.891936: step: 1982/1541, loss: 0.002956712618470192 2023-01-23 13:55:11.750537: step: 1984/1541, loss: 0.007648573722690344 2023-01-23 13:55:12.659037: step: 1986/1541, loss: 0.0623050257563591 2023-01-23 13:55:13.566966: step: 1988/1541, loss: 0.007195488549768925 2023-01-23 13:55:14.472854: step: 1990/1541, loss: 0.11015693843364716 2023-01-23 13:55:15.359287: step: 1992/1541, loss: 0.03449738025665283 2023-01-23 13:55:16.241233: step: 1994/1541, loss: 0.0069829546846449375 2023-01-23 13:55:17.158963: step: 1996/1541, loss: 0.004179663024842739 2023-01-23 13:55:18.022571: step: 1998/1541, loss: 0.002700645476579666 2023-01-23 13:55:18.893633: step: 2000/1541, loss: 0.001518316799774766 2023-01-23 13:55:19.789325: step: 2002/1541, loss: 0.005400010850280523 2023-01-23 13:55:20.670190: step: 2004/1541, loss: 0.03343428298830986 2023-01-23 13:55:21.599378: step: 2006/1541, loss: 0.06961869448423386 2023-01-23 13:55:22.451747: step: 2008/1541, loss: 0.0017720966134220362 2023-01-23 13:55:23.322657: step: 2010/1541, loss: 0.0005702256457880139 2023-01-23 13:55:24.195102: step: 2012/1541, loss: 0.336657851934433 2023-01-23 13:55:25.070755: step: 2014/1541, loss: 0.20219920575618744 2023-01-23 13:55:25.942078: step: 2016/1541, loss: 0.08610707521438599 2023-01-23 13:55:26.823504: step: 2018/1541, loss: 0.009196813218295574 2023-01-23 13:55:27.750774: step: 2020/1541, loss: 0.1503063291311264 2023-01-23 13:55:28.727354: step: 2022/1541, loss: 0.0020055377390235662 2023-01-23 13:55:29.594992: step: 2024/1541, loss: 0.039137836545705795 2023-01-23 13:55:30.494268: step: 2026/1541, loss: 0.031322989612817764 2023-01-23 13:55:31.347131: step: 2028/1541, loss: 0.3474265933036804 2023-01-23 13:55:32.262292: step: 2030/1541, loss: 0.025526151061058044 2023-01-23 13:55:33.127561: step: 2032/1541, loss: 2.074739456176758 2023-01-23 13:55:34.062664: step: 2034/1541, loss: 0.13673201203346252 2023-01-23 13:55:34.949454: step: 2036/1541, loss: 0.006376875564455986 2023-01-23 13:55:35.877961: step: 2038/1541, loss: 0.02279086969792843 2023-01-23 13:55:36.740115: step: 2040/1541, loss: 0.17999514937400818 2023-01-23 13:55:37.638010: step: 2042/1541, loss: 1.3480318784713745 2023-01-23 13:55:38.512279: step: 2044/1541, loss: 0.14240851998329163 2023-01-23 13:55:39.443536: step: 2046/1541, loss: 0.009068192914128304 2023-01-23 13:55:40.337607: step: 2048/1541, loss: 0.03526700660586357 2023-01-23 13:55:41.276765: step: 2050/1541, loss: 0.40890491008758545 2023-01-23 13:55:42.134331: step: 2052/1541, loss: 0.19514703750610352 2023-01-23 13:55:43.020754: step: 2054/1541, loss: 0.007035284768790007 2023-01-23 13:55:43.943214: step: 2056/1541, loss: 0.001953280996531248 2023-01-23 13:55:44.826914: step: 2058/1541, loss: 0.14065398275852203 2023-01-23 13:55:45.735900: step: 2060/1541, loss: 0.00584710668772459 2023-01-23 13:55:46.602175: step: 2062/1541, loss: 0.0081870062276721 2023-01-23 13:55:47.461903: step: 2064/1541, loss: 0.06359392404556274 2023-01-23 13:55:48.376368: step: 2066/1541, loss: 0.001110730110667646 2023-01-23 13:55:49.286671: step: 2068/1541, loss: 0.006989039946347475 2023-01-23 13:55:50.174475: step: 2070/1541, loss: 0.014545871876180172 2023-01-23 13:55:51.058670: step: 2072/1541, loss: 0.008343604393303394 2023-01-23 13:55:52.008472: step: 2074/1541, loss: 0.10065170377492905 2023-01-23 13:55:52.893403: step: 2076/1541, loss: 0.16537295281887054 2023-01-23 13:55:53.784969: step: 2078/1541, loss: 0.06329221278429031 2023-01-23 13:55:54.676557: step: 2080/1541, loss: 0.6596778631210327 2023-01-23 13:55:55.651271: step: 2082/1541, loss: 0.0030784904956817627 2023-01-23 13:55:56.561723: step: 2084/1541, loss: 0.20106634497642517 2023-01-23 13:55:57.521085: step: 2086/1541, loss: 0.17159314453601837 2023-01-23 13:55:58.446965: step: 2088/1541, loss: 0.01033039391040802 2023-01-23 13:55:59.358832: step: 2090/1541, loss: 0.006675757467746735 2023-01-23 13:56:00.283014: step: 2092/1541, loss: 0.005637442227452993 2023-01-23 13:56:01.134570: step: 2094/1541, loss: 0.0012457574484869838 2023-01-23 13:56:02.088446: step: 2096/1541, loss: 0.03142043203115463 2023-01-23 13:56:03.045282: step: 2098/1541, loss: 0.024269457906484604 2023-01-23 13:56:04.021180: step: 2100/1541, loss: 0.021330133080482483 2023-01-23 13:56:04.917078: step: 2102/1541, loss: 0.015400748699903488 2023-01-23 13:56:05.826206: step: 2104/1541, loss: 0.1668912172317505 2023-01-23 13:56:06.748495: step: 2106/1541, loss: 0.19031602144241333 2023-01-23 13:56:07.642435: step: 2108/1541, loss: 0.3811422884464264 2023-01-23 13:56:08.537264: step: 2110/1541, loss: 0.02062860131263733 2023-01-23 13:56:09.425500: step: 2112/1541, loss: 0.0022681495174765587 2023-01-23 13:56:10.303539: step: 2114/1541, loss: 0.05228957533836365 2023-01-23 13:56:11.192210: step: 2116/1541, loss: 0.08214633911848068 2023-01-23 13:56:12.086036: step: 2118/1541, loss: 0.03545140475034714 2023-01-23 13:56:13.006158: step: 2120/1541, loss: 0.008206518366932869 2023-01-23 13:56:13.912021: step: 2122/1541, loss: 3.468595027923584 2023-01-23 13:56:14.822253: step: 2124/1541, loss: 0.0009105342323891819 2023-01-23 13:56:15.805520: step: 2126/1541, loss: 0.07619493454694748 2023-01-23 13:56:16.715567: step: 2128/1541, loss: 0.0023267269134521484 2023-01-23 13:56:17.578358: step: 2130/1541, loss: 0.011058975011110306 2023-01-23 13:56:18.463980: step: 2132/1541, loss: 0.398556113243103 2023-01-23 13:56:19.329170: step: 2134/1541, loss: 0.16994842886924744 2023-01-23 13:56:20.243870: step: 2136/1541, loss: 0.006783036049455404 2023-01-23 13:56:21.135619: step: 2138/1541, loss: 0.0062148659490048885 2023-01-23 13:56:22.013300: step: 2140/1541, loss: 0.03828222304582596 2023-01-23 13:56:22.937383: step: 2142/1541, loss: 0.10597944259643555 2023-01-23 13:56:23.838116: step: 2144/1541, loss: 0.012161126360297203 2023-01-23 13:56:24.713553: step: 2146/1541, loss: 0.05056140944361687 2023-01-23 13:56:25.597914: step: 2148/1541, loss: 0.0026309657841920853 2023-01-23 13:56:26.548484: step: 2150/1541, loss: 0.02760053612291813 2023-01-23 13:56:27.448282: step: 2152/1541, loss: 0.04817895591259003 2023-01-23 13:56:28.387202: step: 2154/1541, loss: 0.012869033962488174 2023-01-23 13:56:29.250175: step: 2156/1541, loss: 0.0014453433686867356 2023-01-23 13:56:30.159393: step: 2158/1541, loss: 0.024307584390044212 2023-01-23 13:56:31.072578: step: 2160/1541, loss: 0.05839160829782486 2023-01-23 13:56:31.949863: step: 2162/1541, loss: 0.18094533681869507 2023-01-23 13:56:32.820422: step: 2164/1541, loss: 0.12920181453227997 2023-01-23 13:56:33.704190: step: 2166/1541, loss: 0.0032752826809883118 2023-01-23 13:56:34.653363: step: 2168/1541, loss: 0.187177374958992 2023-01-23 13:56:35.616513: step: 2170/1541, loss: 0.005983002483844757 2023-01-23 13:56:36.578158: step: 2172/1541, loss: 0.11446206271648407 2023-01-23 13:56:37.451693: step: 2174/1541, loss: 0.10498301684856415 2023-01-23 13:56:38.342718: step: 2176/1541, loss: 0.506338894367218 2023-01-23 13:56:39.216533: step: 2178/1541, loss: 0.08652496337890625 2023-01-23 13:56:40.117024: step: 2180/1541, loss: 0.0007062872173264623 2023-01-23 13:56:41.022455: step: 2182/1541, loss: 0.10516755282878876 2023-01-23 13:56:41.915661: step: 2184/1541, loss: 0.04144292324781418 2023-01-23 13:56:42.815223: step: 2186/1541, loss: 0.007301343604922295 2023-01-23 13:56:43.687063: step: 2188/1541, loss: 0.023075079545378685 2023-01-23 13:56:44.571674: step: 2190/1541, loss: 0.15595915913581848 2023-01-23 13:56:45.467453: step: 2192/1541, loss: 0.02791862189769745 2023-01-23 13:56:46.351250: step: 2194/1541, loss: 0.03007386438548565 2023-01-23 13:56:47.364857: step: 2196/1541, loss: 0.010151881724596024 2023-01-23 13:56:48.283020: step: 2198/1541, loss: 0.021889977157115936 2023-01-23 13:56:49.315254: step: 2200/1541, loss: 0.0038118786178529263 2023-01-23 13:56:50.216958: step: 2202/1541, loss: 0.018140757456421852 2023-01-23 13:56:51.146575: step: 2204/1541, loss: 0.26633211970329285 2023-01-23 13:56:52.086968: step: 2206/1541, loss: 0.015927573665976524 2023-01-23 13:56:52.984988: step: 2208/1541, loss: 0.00516885332763195 2023-01-23 13:56:53.860922: step: 2210/1541, loss: 0.09409988671541214 2023-01-23 13:56:54.717328: step: 2212/1541, loss: 0.026614628732204437 2023-01-23 13:56:55.586504: step: 2214/1541, loss: 0.0073180426843464375 2023-01-23 13:56:56.516372: step: 2216/1541, loss: 0.07018817961215973 2023-01-23 13:56:57.391073: step: 2218/1541, loss: 0.032195668667554855 2023-01-23 13:56:58.318950: step: 2220/1541, loss: 0.0006553775747306645 2023-01-23 13:56:59.204920: step: 2222/1541, loss: 0.02232353575527668 2023-01-23 13:57:00.061209: step: 2224/1541, loss: 0.062237393110990524 2023-01-23 13:57:00.933406: step: 2226/1541, loss: 0.0436524823307991 2023-01-23 13:57:01.801942: step: 2228/1541, loss: 0.18367838859558105 2023-01-23 13:57:02.684545: step: 2230/1541, loss: 0.006525114644318819 2023-01-23 13:57:03.592589: step: 2232/1541, loss: 0.21501630544662476 2023-01-23 13:57:04.499946: step: 2234/1541, loss: 0.03690531477332115 2023-01-23 13:57:05.375132: step: 2236/1541, loss: 0.014687259681522846 2023-01-23 13:57:06.256490: step: 2238/1541, loss: 0.43507149815559387 2023-01-23 13:57:07.123413: step: 2240/1541, loss: 0.004703406244516373 2023-01-23 13:57:08.024417: step: 2242/1541, loss: 0.017657900229096413 2023-01-23 13:57:08.911380: step: 2244/1541, loss: 0.0006348452297970653 2023-01-23 13:57:09.765092: step: 2246/1541, loss: 0.3867260813713074 2023-01-23 13:57:10.735047: step: 2248/1541, loss: 0.0014348243130370975 2023-01-23 13:57:11.606891: step: 2250/1541, loss: 0.4976275861263275 2023-01-23 13:57:12.485374: step: 2252/1541, loss: 0.07705119997262955 2023-01-23 13:57:13.382740: step: 2254/1541, loss: 0.23006273806095123 2023-01-23 13:57:14.246346: step: 2256/1541, loss: 0.10347851365804672 2023-01-23 13:57:15.159973: step: 2258/1541, loss: 0.18053385615348816 2023-01-23 13:57:16.037084: step: 2260/1541, loss: 0.008687441237270832 2023-01-23 13:57:16.910322: step: 2262/1541, loss: 0.07415398210287094 2023-01-23 13:57:17.793536: step: 2264/1541, loss: 0.03508385643362999 2023-01-23 13:57:18.760882: step: 2266/1541, loss: 0.004527932032942772 2023-01-23 13:57:19.647325: step: 2268/1541, loss: 0.2076130509376526 2023-01-23 13:57:20.599538: step: 2270/1541, loss: 0.31198686361312866 2023-01-23 13:57:21.513703: step: 2272/1541, loss: 0.07210588455200195 2023-01-23 13:57:22.380470: step: 2274/1541, loss: 0.2547752857208252 2023-01-23 13:57:23.303391: step: 2276/1541, loss: 0.004004706162959337 2023-01-23 13:57:24.176709: step: 2278/1541, loss: 0.3821481466293335 2023-01-23 13:57:25.074002: step: 2280/1541, loss: 0.04943105950951576 2023-01-23 13:57:25.955766: step: 2282/1541, loss: 0.17900946736335754 2023-01-23 13:57:26.869212: step: 2284/1541, loss: 0.07548375427722931 2023-01-23 13:57:27.738179: step: 2286/1541, loss: 0.09882526844739914 2023-01-23 13:57:28.650415: step: 2288/1541, loss: 0.010457594878971577 2023-01-23 13:57:29.555866: step: 2290/1541, loss: 0.0038914592005312443 2023-01-23 13:57:30.466907: step: 2292/1541, loss: 0.349715918302536 2023-01-23 13:57:31.322175: step: 2294/1541, loss: 0.19568964838981628 2023-01-23 13:57:32.222515: step: 2296/1541, loss: 0.04593972861766815 2023-01-23 13:57:33.166458: step: 2298/1541, loss: 0.12145113199949265 2023-01-23 13:57:34.024532: step: 2300/1541, loss: 0.008500450290739536 2023-01-23 13:57:34.926658: step: 2302/1541, loss: 0.010831251740455627 2023-01-23 13:57:35.804486: step: 2304/1541, loss: 0.016527460888028145 2023-01-23 13:57:36.710233: step: 2306/1541, loss: 0.026311641559004784 2023-01-23 13:57:37.648756: step: 2308/1541, loss: 0.13994662463665009 2023-01-23 13:57:38.514627: step: 2310/1541, loss: 0.13943853974342346 2023-01-23 13:57:39.372002: step: 2312/1541, loss: 0.005480692721903324 2023-01-23 13:57:40.253059: step: 2314/1541, loss: 0.07951635122299194 2023-01-23 13:57:41.141397: step: 2316/1541, loss: 0.021663611754775047 2023-01-23 13:57:42.028931: step: 2318/1541, loss: 0.1365155130624771 2023-01-23 13:57:43.004883: step: 2320/1541, loss: 0.03885456174612045 2023-01-23 13:57:43.891238: step: 2322/1541, loss: 0.03208979219198227 2023-01-23 13:57:44.761934: step: 2324/1541, loss: 0.36992546916007996 2023-01-23 13:57:45.649669: step: 2326/1541, loss: 0.07772085815668106 2023-01-23 13:57:46.622403: step: 2328/1541, loss: 0.006483321078121662 2023-01-23 13:57:47.506456: step: 2330/1541, loss: 0.017757145687937737 2023-01-23 13:57:48.407033: step: 2332/1541, loss: 0.15908324718475342 2023-01-23 13:57:49.265102: step: 2334/1541, loss: 0.019110340625047684 2023-01-23 13:57:50.178115: step: 2336/1541, loss: 0.15583834052085876 2023-01-23 13:57:51.108414: step: 2338/1541, loss: 0.09414426237344742 2023-01-23 13:57:52.029634: step: 2340/1541, loss: 0.05680352449417114 2023-01-23 13:57:52.899590: step: 2342/1541, loss: 0.008083969354629517 2023-01-23 13:57:53.783432: step: 2344/1541, loss: 0.01685251109302044 2023-01-23 13:57:54.656671: step: 2346/1541, loss: 0.038661882281303406 2023-01-23 13:57:55.531578: step: 2348/1541, loss: 0.04504942521452904 2023-01-23 13:57:56.431501: step: 2350/1541, loss: 0.001217059907503426 2023-01-23 13:57:57.319955: step: 2352/1541, loss: 0.2571791410446167 2023-01-23 13:57:58.197831: step: 2354/1541, loss: 0.10129697620868683 2023-01-23 13:57:59.115013: step: 2356/1541, loss: 0.04508807510137558 2023-01-23 13:58:00.092771: step: 2358/1541, loss: 0.19261321425437927 2023-01-23 13:58:00.960901: step: 2360/1541, loss: 0.3428558111190796 2023-01-23 13:58:01.884849: step: 2362/1541, loss: 0.012255275622010231 2023-01-23 13:58:02.764338: step: 2364/1541, loss: 0.05787574499845505 2023-01-23 13:58:03.688004: step: 2366/1541, loss: 0.022480929270386696 2023-01-23 13:58:04.577792: step: 2368/1541, loss: 0.018100161105394363 2023-01-23 13:58:05.487274: step: 2370/1541, loss: 0.4227488040924072 2023-01-23 13:58:06.382669: step: 2372/1541, loss: 0.3897826373577118 2023-01-23 13:58:07.287568: step: 2374/1541, loss: 0.1370008885860443 2023-01-23 13:58:08.206085: step: 2376/1541, loss: 0.00664899218827486 2023-01-23 13:58:09.123885: step: 2378/1541, loss: 0.06726379692554474 2023-01-23 13:58:10.058844: step: 2380/1541, loss: 0.025943573564291 2023-01-23 13:58:10.963128: step: 2382/1541, loss: 0.0010552983731031418 2023-01-23 13:58:11.865393: step: 2384/1541, loss: 0.037139907479286194 2023-01-23 13:58:12.731682: step: 2386/1541, loss: 0.01524459756910801 2023-01-23 13:58:13.632698: step: 2388/1541, loss: 0.07304703444242477 2023-01-23 13:58:14.511226: step: 2390/1541, loss: 0.014022784307599068 2023-01-23 13:58:15.464677: step: 2392/1541, loss: 0.11339455842971802 2023-01-23 13:58:16.336149: step: 2394/1541, loss: 0.0014050083700567484 2023-01-23 13:58:17.295094: step: 2396/1541, loss: 0.0013861428014934063 2023-01-23 13:58:18.204044: step: 2398/1541, loss: 0.04053761810064316 2023-01-23 13:58:19.232094: step: 2400/1541, loss: 0.01857452467083931 2023-01-23 13:58:20.085087: step: 2402/1541, loss: 0.00984022207558155 2023-01-23 13:58:21.053553: step: 2404/1541, loss: 0.0944405198097229 2023-01-23 13:58:21.981080: step: 2406/1541, loss: 0.05853435769677162 2023-01-23 13:58:22.907061: step: 2408/1541, loss: 0.144709050655365 2023-01-23 13:58:23.787812: step: 2410/1541, loss: 0.006619004067033529 2023-01-23 13:58:24.676316: step: 2412/1541, loss: 0.24655459821224213 2023-01-23 13:58:25.578539: step: 2414/1541, loss: 0.2696431577205658 2023-01-23 13:58:26.460795: step: 2416/1541, loss: 0.576810896396637 2023-01-23 13:58:27.352815: step: 2418/1541, loss: 0.2067824900150299 2023-01-23 13:58:28.245901: step: 2420/1541, loss: 0.050569090992212296 2023-01-23 13:58:29.125395: step: 2422/1541, loss: 0.032014258205890656 2023-01-23 13:58:30.012965: step: 2424/1541, loss: 0.18980905413627625 2023-01-23 13:58:30.862810: step: 2426/1541, loss: 0.23771348595619202 2023-01-23 13:58:31.747937: step: 2428/1541, loss: 0.23943696916103363 2023-01-23 13:58:32.654013: step: 2430/1541, loss: 0.06032700091600418 2023-01-23 13:58:33.557288: step: 2432/1541, loss: 0.029445216059684753 2023-01-23 13:58:34.462498: step: 2434/1541, loss: 0.21820563077926636 2023-01-23 13:58:35.366455: step: 2436/1541, loss: 0.057801082730293274 2023-01-23 13:58:36.281992: step: 2438/1541, loss: 0.33600854873657227 2023-01-23 13:58:37.223174: step: 2440/1541, loss: 0.15036419034004211 2023-01-23 13:58:38.090894: step: 2442/1541, loss: 0.032437536865472794 2023-01-23 13:58:39.009002: step: 2444/1541, loss: 0.03460666537284851 2023-01-23 13:58:39.942758: step: 2446/1541, loss: 1.5159764289855957 2023-01-23 13:58:40.832825: step: 2448/1541, loss: 0.32207876443862915 2023-01-23 13:58:41.756738: step: 2450/1541, loss: 0.3533972501754761 2023-01-23 13:58:42.691539: step: 2452/1541, loss: 0.8577343821525574 2023-01-23 13:58:43.657270: step: 2454/1541, loss: 0.5870413780212402 2023-01-23 13:58:44.531122: step: 2456/1541, loss: 0.02309577167034149 2023-01-23 13:58:45.504496: step: 2458/1541, loss: 2.6816420555114746 2023-01-23 13:58:46.403530: step: 2460/1541, loss: 0.006720582023262978 2023-01-23 13:58:47.354766: step: 2462/1541, loss: 0.2570784389972687 2023-01-23 13:58:48.257076: step: 2464/1541, loss: 0.019814809784293175 2023-01-23 13:58:49.156228: step: 2466/1541, loss: 0.007232311647385359 2023-01-23 13:58:50.072030: step: 2468/1541, loss: 0.07395780086517334 2023-01-23 13:58:50.958228: step: 2470/1541, loss: 0.01524664368480444 2023-01-23 13:58:51.896150: step: 2472/1541, loss: 1.0152549743652344 2023-01-23 13:58:52.818529: step: 2474/1541, loss: 0.00843226257711649 2023-01-23 13:58:53.767338: step: 2476/1541, loss: 0.025374844670295715 2023-01-23 13:58:54.635892: step: 2478/1541, loss: 0.027288058772683144 2023-01-23 13:58:55.534290: step: 2480/1541, loss: 0.03149669989943504 2023-01-23 13:58:56.509607: step: 2482/1541, loss: 0.044087205082178116 2023-01-23 13:58:57.416501: step: 2484/1541, loss: 0.10561323910951614 2023-01-23 13:58:58.289875: step: 2486/1541, loss: 0.0676058977842331 2023-01-23 13:58:59.162803: step: 2488/1541, loss: 0.3011837601661682 2023-01-23 13:59:00.076332: step: 2490/1541, loss: 0.17485462129116058 2023-01-23 13:59:01.078393: step: 2492/1541, loss: 0.11154323071241379 2023-01-23 13:59:01.969763: step: 2494/1541, loss: 0.09069971740245819 2023-01-23 13:59:02.920951: step: 2496/1541, loss: 0.06847142428159714 2023-01-23 13:59:03.840623: step: 2498/1541, loss: 0.10253242403268814 2023-01-23 13:59:04.730507: step: 2500/1541, loss: 0.14205734431743622 2023-01-23 13:59:05.614628: step: 2502/1541, loss: 0.0027855313383042812 2023-01-23 13:59:06.504968: step: 2504/1541, loss: 0.026526696979999542 2023-01-23 13:59:07.453883: step: 2506/1541, loss: 0.16776341199874878 2023-01-23 13:59:08.349377: step: 2508/1541, loss: 0.01869254931807518 2023-01-23 13:59:09.247595: step: 2510/1541, loss: 0.016837874427437782 2023-01-23 13:59:10.136124: step: 2512/1541, loss: 0.12111961096525192 2023-01-23 13:59:11.114758: step: 2514/1541, loss: 0.014047276228666306 2023-01-23 13:59:12.013778: step: 2516/1541, loss: 0.023225096985697746 2023-01-23 13:59:12.956289: step: 2518/1541, loss: 0.16789007186889648 2023-01-23 13:59:13.836475: step: 2520/1541, loss: 0.006475478410720825 2023-01-23 13:59:14.731678: step: 2522/1541, loss: 0.004374522250145674 2023-01-23 13:59:15.597705: step: 2524/1541, loss: 0.0775475725531578 2023-01-23 13:59:16.465879: step: 2526/1541, loss: 0.008296188898384571 2023-01-23 13:59:17.406595: step: 2528/1541, loss: 0.12656664848327637 2023-01-23 13:59:18.360402: step: 2530/1541, loss: 0.052543047815561295 2023-01-23 13:59:19.236504: step: 2532/1541, loss: 0.017087383195757866 2023-01-23 13:59:20.169631: step: 2534/1541, loss: 2.131448268890381 2023-01-23 13:59:21.073274: step: 2536/1541, loss: 0.003595472313463688 2023-01-23 13:59:22.005833: step: 2538/1541, loss: 0.10373825579881668 2023-01-23 13:59:22.892825: step: 2540/1541, loss: 0.09775185585021973 2023-01-23 13:59:23.785610: step: 2542/1541, loss: 0.003228941932320595 2023-01-23 13:59:24.674635: step: 2544/1541, loss: 0.03404537960886955 2023-01-23 13:59:25.627594: step: 2546/1541, loss: 0.05457933992147446 2023-01-23 13:59:26.548876: step: 2548/1541, loss: 0.15710140764713287 2023-01-23 13:59:27.412997: step: 2550/1541, loss: 0.015934329479932785 2023-01-23 13:59:28.400130: step: 2552/1541, loss: 0.036350276321172714 2023-01-23 13:59:29.323169: step: 2554/1541, loss: 0.12855565547943115 2023-01-23 13:59:30.176785: step: 2556/1541, loss: 0.16782830655574799 2023-01-23 13:59:31.094878: step: 2558/1541, loss: 0.0443844273686409 2023-01-23 13:59:31.970381: step: 2560/1541, loss: 0.23384232819080353 2023-01-23 13:59:32.833896: step: 2562/1541, loss: 0.05839791148900986 2023-01-23 13:59:33.715469: step: 2564/1541, loss: 0.012533423490822315 2023-01-23 13:59:34.650974: step: 2566/1541, loss: 0.02922842837870121 2023-01-23 13:59:35.552693: step: 2568/1541, loss: 0.019334828481078148 2023-01-23 13:59:36.439877: step: 2570/1541, loss: 0.025997117161750793 2023-01-23 13:59:37.441017: step: 2572/1541, loss: 0.32466965913772583 2023-01-23 13:59:38.376527: step: 2574/1541, loss: 0.00831502303481102 2023-01-23 13:59:39.255683: step: 2576/1541, loss: 0.8168749809265137 2023-01-23 13:59:40.199718: step: 2578/1541, loss: 0.014933456666767597 2023-01-23 13:59:41.145834: step: 2580/1541, loss: 0.0859113559126854 2023-01-23 13:59:42.071043: step: 2582/1541, loss: 0.003202788531780243 2023-01-23 13:59:42.944476: step: 2584/1541, loss: 0.022798221558332443 2023-01-23 13:59:43.918974: step: 2586/1541, loss: 0.011035117320716381 2023-01-23 13:59:44.827046: step: 2588/1541, loss: 0.0845019668340683 2023-01-23 13:59:45.765198: step: 2590/1541, loss: 0.0019145916448906064 2023-01-23 13:59:46.639128: step: 2592/1541, loss: 0.04778861254453659 2023-01-23 13:59:47.597513: step: 2594/1541, loss: 0.04051327332854271 2023-01-23 13:59:48.457726: step: 2596/1541, loss: 0.002758912742137909 2023-01-23 13:59:49.382793: step: 2598/1541, loss: 0.14651328325271606 2023-01-23 13:59:50.261325: step: 2600/1541, loss: 0.2135404646396637 2023-01-23 13:59:51.106721: step: 2602/1541, loss: 0.06766960769891739 2023-01-23 13:59:51.991900: step: 2604/1541, loss: 0.04334710165858269 2023-01-23 13:59:52.878679: step: 2606/1541, loss: 0.04546865075826645 2023-01-23 13:59:53.780021: step: 2608/1541, loss: 0.16750416159629822 2023-01-23 13:59:54.650242: step: 2610/1541, loss: 0.0038214013911783695 2023-01-23 13:59:55.560360: step: 2612/1541, loss: 0.5381208062171936 2023-01-23 13:59:56.448759: step: 2614/1541, loss: 0.10656782239675522 2023-01-23 13:59:57.353318: step: 2616/1541, loss: 0.19743509590625763 2023-01-23 13:59:58.287027: step: 2618/1541, loss: 0.08570172637701035 2023-01-23 13:59:59.161203: step: 2620/1541, loss: 0.09088955074548721 2023-01-23 14:00:00.088417: step: 2622/1541, loss: 0.08749288320541382 2023-01-23 14:00:00.981997: step: 2624/1541, loss: 0.039253078401088715 2023-01-23 14:00:01.906843: step: 2626/1541, loss: 0.05853981152176857 2023-01-23 14:00:02.826304: step: 2628/1541, loss: 0.3015047311782837 2023-01-23 14:00:03.681473: step: 2630/1541, loss: 0.49172964692115784 2023-01-23 14:00:04.563109: step: 2632/1541, loss: 0.02957630716264248 2023-01-23 14:00:05.471693: step: 2634/1541, loss: 0.013594934716820717 2023-01-23 14:00:06.411461: step: 2636/1541, loss: 0.03190716356039047 2023-01-23 14:00:07.288440: step: 2638/1541, loss: 0.6661400198936462 2023-01-23 14:00:08.200255: step: 2640/1541, loss: 0.30094948410987854 2023-01-23 14:00:09.157675: step: 2642/1541, loss: 0.22559307515621185 2023-01-23 14:00:10.042704: step: 2644/1541, loss: 0.025696318596601486 2023-01-23 14:00:10.921930: step: 2646/1541, loss: 0.0008726480300538242 2023-01-23 14:00:11.780829: step: 2648/1541, loss: 0.4374997019767761 2023-01-23 14:00:12.734830: step: 2650/1541, loss: 0.002645072527229786 2023-01-23 14:00:13.641786: step: 2652/1541, loss: 0.058077652007341385 2023-01-23 14:00:14.519532: step: 2654/1541, loss: 0.1564338058233261 2023-01-23 14:00:15.439227: step: 2656/1541, loss: 0.05022241920232773 2023-01-23 14:00:16.317486: step: 2658/1541, loss: 0.002128944732248783 2023-01-23 14:00:17.199505: step: 2660/1541, loss: 0.0017293249256908894 2023-01-23 14:00:18.058065: step: 2662/1541, loss: 0.16785147786140442 2023-01-23 14:00:18.972191: step: 2664/1541, loss: 0.028015993535518646 2023-01-23 14:00:19.845772: step: 2666/1541, loss: 0.006621657870709896 2023-01-23 14:00:20.708090: step: 2668/1541, loss: 0.13872455060482025 2023-01-23 14:00:21.609620: step: 2670/1541, loss: 0.32253021001815796 2023-01-23 14:00:22.496933: step: 2672/1541, loss: 0.012094688601791859 2023-01-23 14:00:23.407418: step: 2674/1541, loss: 0.03191390633583069 2023-01-23 14:00:24.294686: step: 2676/1541, loss: 0.09119012951850891 2023-01-23 14:00:25.195957: step: 2678/1541, loss: 0.006685855332762003 2023-01-23 14:00:26.157592: step: 2680/1541, loss: 0.02478199452161789 2023-01-23 14:00:27.043470: step: 2682/1541, loss: 0.1257781684398651 2023-01-23 14:00:27.932674: step: 2684/1541, loss: 0.07575640082359314 2023-01-23 14:00:28.847202: step: 2686/1541, loss: 0.10740549862384796 2023-01-23 14:00:29.784821: step: 2688/1541, loss: 0.019429486244916916 2023-01-23 14:00:30.672453: step: 2690/1541, loss: 0.03525572642683983 2023-01-23 14:00:31.587633: step: 2692/1541, loss: 0.3974149525165558 2023-01-23 14:00:32.458000: step: 2694/1541, loss: 0.15193021297454834 2023-01-23 14:00:33.369441: step: 2696/1541, loss: 0.002500186674296856 2023-01-23 14:00:34.246541: step: 2698/1541, loss: 0.061166293919086456 2023-01-23 14:00:35.148380: step: 2700/1541, loss: 0.027070583775639534 2023-01-23 14:00:36.018474: step: 2702/1541, loss: 0.03643856570124626 2023-01-23 14:00:36.947050: step: 2704/1541, loss: 0.0023344242945313454 2023-01-23 14:00:37.824412: step: 2706/1541, loss: 0.0019278411054983735 2023-01-23 14:00:38.768715: step: 2708/1541, loss: 0.10892560333013535 2023-01-23 14:00:39.680283: step: 2710/1541, loss: 0.004437249153852463 2023-01-23 14:00:40.574448: step: 2712/1541, loss: 0.1170048639178276 2023-01-23 14:00:41.478043: step: 2714/1541, loss: 0.004984814208000898 2023-01-23 14:00:42.407397: step: 2716/1541, loss: 0.18956059217453003 2023-01-23 14:00:43.319171: step: 2718/1541, loss: 0.04181739687919617 2023-01-23 14:00:44.198906: step: 2720/1541, loss: 0.5520344972610474 2023-01-23 14:00:45.085130: step: 2722/1541, loss: 0.12840694189071655 2023-01-23 14:00:46.028382: step: 2724/1541, loss: 0.24876157939434052 2023-01-23 14:00:46.931038: step: 2726/1541, loss: 0.34649962186813354 2023-01-23 14:00:47.898665: step: 2728/1541, loss: 0.02107164077460766 2023-01-23 14:00:48.805082: step: 2730/1541, loss: 0.0028469297103583813 2023-01-23 14:00:49.728825: step: 2732/1541, loss: 0.03409109637141228 2023-01-23 14:00:50.619863: step: 2734/1541, loss: 0.07239269465208054 2023-01-23 14:00:51.512336: step: 2736/1541, loss: 0.13379615545272827 2023-01-23 14:00:52.408718: step: 2738/1541, loss: 0.05826166272163391 2023-01-23 14:00:53.277974: step: 2740/1541, loss: 0.13424761593341827 2023-01-23 14:00:54.234266: step: 2742/1541, loss: 0.009221998043358326 2023-01-23 14:00:55.125040: step: 2744/1541, loss: 0.3343107998371124 2023-01-23 14:00:55.997216: step: 2746/1541, loss: 0.4674799144268036 2023-01-23 14:00:56.905596: step: 2748/1541, loss: 0.10518639534711838 2023-01-23 14:00:57.843077: step: 2750/1541, loss: 0.4366323947906494 2023-01-23 14:00:58.700951: step: 2752/1541, loss: 0.014184839092195034 2023-01-23 14:00:59.594423: step: 2754/1541, loss: 0.019127922132611275 2023-01-23 14:01:00.476788: step: 2756/1541, loss: 0.3483906686306 2023-01-23 14:01:01.363083: step: 2758/1541, loss: 0.11063899099826813 2023-01-23 14:01:02.307564: step: 2760/1541, loss: 0.035082507878541946 2023-01-23 14:01:03.183237: step: 2762/1541, loss: 0.00271345442160964 2023-01-23 14:01:04.073062: step: 2764/1541, loss: 0.026341650635004044 2023-01-23 14:01:04.981588: step: 2766/1541, loss: 0.20479030907154083 2023-01-23 14:01:05.866898: step: 2768/1541, loss: 0.007899967953562737 2023-01-23 14:01:06.738903: step: 2770/1541, loss: 0.05634470283985138 2023-01-23 14:01:07.623177: step: 2772/1541, loss: 0.002608424983918667 2023-01-23 14:01:08.527545: step: 2774/1541, loss: 0.07738536596298218 2023-01-23 14:01:09.440457: step: 2776/1541, loss: 0.0023462409153580666 2023-01-23 14:01:10.314290: step: 2778/1541, loss: 0.20571376383304596 2023-01-23 14:01:11.192302: step: 2780/1541, loss: 0.00936359353363514 2023-01-23 14:01:12.100618: step: 2782/1541, loss: 0.18924584984779358 2023-01-23 14:01:13.015459: step: 2784/1541, loss: 0.20943628251552582 2023-01-23 14:01:13.893197: step: 2786/1541, loss: 0.07807636260986328 2023-01-23 14:01:14.746057: step: 2788/1541, loss: 0.023705320432782173 2023-01-23 14:01:15.652407: step: 2790/1541, loss: 0.07036195695400238 2023-01-23 14:01:16.507345: step: 2792/1541, loss: 0.10155332833528519 2023-01-23 14:01:17.383262: step: 2794/1541, loss: 0.09230057895183563 2023-01-23 14:01:18.283281: step: 2796/1541, loss: 0.010979505255818367 2023-01-23 14:01:19.191246: step: 2798/1541, loss: 0.032298240810632706 2023-01-23 14:01:20.134457: step: 2800/1541, loss: 0.057735003530979156 2023-01-23 14:01:21.000065: step: 2802/1541, loss: 0.033841945230960846 2023-01-23 14:01:21.856911: step: 2804/1541, loss: 0.0031092248391360044 2023-01-23 14:01:22.709673: step: 2806/1541, loss: 0.42226162552833557 2023-01-23 14:01:23.579573: step: 2808/1541, loss: 0.015118773095309734 2023-01-23 14:01:24.484749: step: 2810/1541, loss: 0.17493538558483124 2023-01-23 14:01:25.456681: step: 2812/1541, loss: 0.36890196800231934 2023-01-23 14:01:26.333845: step: 2814/1541, loss: 0.009793408215045929 2023-01-23 14:01:27.245175: step: 2816/1541, loss: 0.035847511142492294 2023-01-23 14:01:28.133927: step: 2818/1541, loss: 0.006178900599479675 2023-01-23 14:01:29.034152: step: 2820/1541, loss: 0.21259643137454987 2023-01-23 14:01:29.922363: step: 2822/1541, loss: 0.08803702145814896 2023-01-23 14:01:30.795178: step: 2824/1541, loss: 0.01435153093189001 2023-01-23 14:01:31.690645: step: 2826/1541, loss: 0.006895377766340971 2023-01-23 14:01:32.617512: step: 2828/1541, loss: 0.08442172408103943 2023-01-23 14:01:33.462189: step: 2830/1541, loss: 0.04527837783098221 2023-01-23 14:01:34.337026: step: 2832/1541, loss: 0.021921508014202118 2023-01-23 14:01:35.223076: step: 2834/1541, loss: 0.008158550597727299 2023-01-23 14:01:36.099145: step: 2836/1541, loss: 0.0036682377103716135 2023-01-23 14:01:36.987927: step: 2838/1541, loss: 0.8272219896316528 2023-01-23 14:01:37.874067: step: 2840/1541, loss: 0.1176614910364151 2023-01-23 14:01:38.808626: step: 2842/1541, loss: 0.13157999515533447 2023-01-23 14:01:39.671113: step: 2844/1541, loss: 0.14286653697490692 2023-01-23 14:01:40.571252: step: 2846/1541, loss: 0.006748029496520758 2023-01-23 14:01:41.489367: step: 2848/1541, loss: 0.08779813349246979 2023-01-23 14:01:42.354790: step: 2850/1541, loss: 0.04271307960152626 2023-01-23 14:01:43.284783: step: 2852/1541, loss: 0.07598268985748291 2023-01-23 14:01:44.212740: step: 2854/1541, loss: 0.4899033308029175 2023-01-23 14:01:45.077106: step: 2856/1541, loss: 0.34017035365104675 2023-01-23 14:01:46.062694: step: 2858/1541, loss: 0.059400007128715515 2023-01-23 14:01:46.970794: step: 2860/1541, loss: 0.45013874769210815 2023-01-23 14:01:47.867819: step: 2862/1541, loss: 0.06934197247028351 2023-01-23 14:01:48.767947: step: 2864/1541, loss: 0.17113015055656433 2023-01-23 14:01:49.643439: step: 2866/1541, loss: 0.045760493725538254 2023-01-23 14:01:50.547148: step: 2868/1541, loss: 0.008146025240421295 2023-01-23 14:01:51.454235: step: 2870/1541, loss: 0.3431890308856964 2023-01-23 14:01:52.389722: step: 2872/1541, loss: 0.07817700505256653 2023-01-23 14:01:53.268737: step: 2874/1541, loss: 0.128688782453537 2023-01-23 14:01:54.153092: step: 2876/1541, loss: 0.017331331968307495 2023-01-23 14:01:55.104198: step: 2878/1541, loss: 0.1957763284444809 2023-01-23 14:01:56.097975: step: 2880/1541, loss: 0.7072577476501465 2023-01-23 14:01:56.979921: step: 2882/1541, loss: 0.25076648592948914 2023-01-23 14:01:57.883196: step: 2884/1541, loss: 0.1360708326101303 2023-01-23 14:01:58.753421: step: 2886/1541, loss: 0.11988551169633865 2023-01-23 14:01:59.662676: step: 2888/1541, loss: 0.03613334521651268 2023-01-23 14:02:00.508874: step: 2890/1541, loss: 0.08317132294178009 2023-01-23 14:02:01.397515: step: 2892/1541, loss: 0.0412982776761055 2023-01-23 14:02:02.366722: step: 2894/1541, loss: 0.03385389596223831 2023-01-23 14:02:03.219950: step: 2896/1541, loss: 0.015507034957408905 2023-01-23 14:02:04.082825: step: 2898/1541, loss: 0.328144907951355 2023-01-23 14:02:04.957075: step: 2900/1541, loss: 0.11678165197372437 2023-01-23 14:02:05.825259: step: 2902/1541, loss: 0.0031643807888031006 2023-01-23 14:02:06.827490: step: 2904/1541, loss: 0.06034282222390175 2023-01-23 14:02:07.862276: step: 2906/1541, loss: 0.04733438044786453 2023-01-23 14:02:08.742821: step: 2908/1541, loss: 0.02264234609901905 2023-01-23 14:02:09.645337: step: 2910/1541, loss: 0.02956981025636196 2023-01-23 14:02:10.550807: step: 2912/1541, loss: 0.28521254658699036 2023-01-23 14:02:11.464754: step: 2914/1541, loss: 0.25121456384658813 2023-01-23 14:02:12.356129: step: 2916/1541, loss: 0.037604425102472305 2023-01-23 14:02:13.233037: step: 2918/1541, loss: 0.1465356945991516 2023-01-23 14:02:14.124059: step: 2920/1541, loss: 0.027709024026989937 2023-01-23 14:02:15.022152: step: 2922/1541, loss: 0.007421792950481176 2023-01-23 14:02:15.893105: step: 2924/1541, loss: 0.09783942997455597 2023-01-23 14:02:16.787861: step: 2926/1541, loss: 0.02201654016971588 2023-01-23 14:02:17.661062: step: 2928/1541, loss: 0.11516444385051727 2023-01-23 14:02:18.591600: step: 2930/1541, loss: 0.03874868154525757 2023-01-23 14:02:19.472361: step: 2932/1541, loss: 0.031072422862052917 2023-01-23 14:02:20.353001: step: 2934/1541, loss: 0.03295714408159256 2023-01-23 14:02:21.242925: step: 2936/1541, loss: 0.32553160190582275 2023-01-23 14:02:22.087796: step: 2938/1541, loss: 0.049032922834157944 2023-01-23 14:02:22.926335: step: 2940/1541, loss: 0.5780029296875 2023-01-23 14:02:23.798387: step: 2942/1541, loss: 0.004303500521928072 2023-01-23 14:02:24.717760: step: 2944/1541, loss: 0.0007261644350364804 2023-01-23 14:02:25.596939: step: 2946/1541, loss: 0.02934255450963974 2023-01-23 14:02:26.499018: step: 2948/1541, loss: 0.13989554345607758 2023-01-23 14:02:27.358519: step: 2950/1541, loss: 0.0045053595677018166 2023-01-23 14:02:28.252370: step: 2952/1541, loss: 0.0419485904276371 2023-01-23 14:02:29.109561: step: 2954/1541, loss: 0.12840263545513153 2023-01-23 14:02:29.986597: step: 2956/1541, loss: 0.0067695374600589275 2023-01-23 14:02:30.861745: step: 2958/1541, loss: 0.11996018141508102 2023-01-23 14:02:31.759660: step: 2960/1541, loss: 0.48727962374687195 2023-01-23 14:02:32.695404: step: 2962/1541, loss: 0.007619503885507584 2023-01-23 14:02:33.601809: step: 2964/1541, loss: 0.019763970747590065 2023-01-23 14:02:34.537596: step: 2966/1541, loss: 0.08006273210048676 2023-01-23 14:02:35.506160: step: 2968/1541, loss: 1.0942578315734863 2023-01-23 14:02:36.384499: step: 2970/1541, loss: 0.029672956094145775 2023-01-23 14:02:37.286677: step: 2972/1541, loss: 0.07222762703895569 2023-01-23 14:02:38.184220: step: 2974/1541, loss: 0.03349202498793602 2023-01-23 14:02:39.063690: step: 2976/1541, loss: 0.0028372907545417547 2023-01-23 14:02:39.932554: step: 2978/1541, loss: 0.04581623151898384 2023-01-23 14:02:40.803181: step: 2980/1541, loss: 0.08591553568840027 2023-01-23 14:02:41.690982: step: 2982/1541, loss: 0.01898043230175972 2023-01-23 14:02:42.577573: step: 2984/1541, loss: 0.02600725367665291 2023-01-23 14:02:43.592703: step: 2986/1541, loss: 0.043363526463508606 2023-01-23 14:02:44.509174: step: 2988/1541, loss: 0.3743121922016144 2023-01-23 14:02:45.401994: step: 2990/1541, loss: 0.2953755259513855 2023-01-23 14:02:46.320996: step: 2992/1541, loss: 0.15417610108852386 2023-01-23 14:02:47.190328: step: 2994/1541, loss: 0.002496634842827916 2023-01-23 14:02:48.045003: step: 2996/1541, loss: 0.013469274155795574 2023-01-23 14:02:48.913925: step: 2998/1541, loss: 0.04787495732307434 2023-01-23 14:02:49.851243: step: 3000/1541, loss: 0.05466896668076515 2023-01-23 14:02:50.713462: step: 3002/1541, loss: 0.03956658020615578 2023-01-23 14:02:51.578543: step: 3004/1541, loss: 0.016286488622426987 2023-01-23 14:02:52.440097: step: 3006/1541, loss: 0.30905401706695557 2023-01-23 14:02:53.305737: step: 3008/1541, loss: 0.026316294446587563 2023-01-23 14:02:54.209333: step: 3010/1541, loss: 0.025143759325146675 2023-01-23 14:02:55.148696: step: 3012/1541, loss: 0.017562441527843475 2023-01-23 14:02:56.101715: step: 3014/1541, loss: 0.3871620297431946 2023-01-23 14:02:56.939859: step: 3016/1541, loss: 0.28705260157585144 2023-01-23 14:02:57.847544: step: 3018/1541, loss: 0.04820851609110832 2023-01-23 14:02:58.722725: step: 3020/1541, loss: 0.27417904138565063 2023-01-23 14:02:59.612857: step: 3022/1541, loss: 0.27538245916366577 2023-01-23 14:03:00.518251: step: 3024/1541, loss: 0.08157826960086823 2023-01-23 14:03:01.401390: step: 3026/1541, loss: 0.05498020350933075 2023-01-23 14:03:02.293658: step: 3028/1541, loss: 0.23264813423156738 2023-01-23 14:03:03.143408: step: 3030/1541, loss: 0.009351534768939018 2023-01-23 14:03:04.191542: step: 3032/1541, loss: 0.1164257600903511 2023-01-23 14:03:05.056984: step: 3034/1541, loss: 0.14672595262527466 2023-01-23 14:03:05.939088: step: 3036/1541, loss: 0.1089651957154274 2023-01-23 14:03:06.891900: step: 3038/1541, loss: 0.1311662197113037 2023-01-23 14:03:07.790056: step: 3040/1541, loss: 0.0517667718231678 2023-01-23 14:03:08.711469: step: 3042/1541, loss: 0.008225592784583569 2023-01-23 14:03:09.582428: step: 3044/1541, loss: 1.517844557762146 2023-01-23 14:03:10.451809: step: 3046/1541, loss: 0.023457001894712448 2023-01-23 14:03:11.330611: step: 3048/1541, loss: 0.01737617328763008 2023-01-23 14:03:12.207077: step: 3050/1541, loss: 0.10479500144720078 2023-01-23 14:03:13.086222: step: 3052/1541, loss: 0.18662340939044952 2023-01-23 14:03:13.950419: step: 3054/1541, loss: 0.4966622292995453 2023-01-23 14:03:14.835869: step: 3056/1541, loss: 0.03841211274266243 2023-01-23 14:03:15.703647: step: 3058/1541, loss: 0.013686875812709332 2023-01-23 14:03:16.582789: step: 3060/1541, loss: 0.1655149906873703 2023-01-23 14:03:17.475864: step: 3062/1541, loss: 0.09059970825910568 2023-01-23 14:03:18.348778: step: 3064/1541, loss: 0.03215779364109039 2023-01-23 14:03:19.248287: step: 3066/1541, loss: 0.2474099099636078 2023-01-23 14:03:20.150578: step: 3068/1541, loss: 0.019593380391597748 2023-01-23 14:03:20.999961: step: 3070/1541, loss: 0.015501338057219982 2023-01-23 14:03:21.912365: step: 3072/1541, loss: 0.07309693098068237 2023-01-23 14:03:22.813562: step: 3074/1541, loss: 0.020654303953051567 2023-01-23 14:03:23.680879: step: 3076/1541, loss: 0.007938170805573463 2023-01-23 14:03:24.560680: step: 3078/1541, loss: 0.028259046375751495 2023-01-23 14:03:25.530380: step: 3080/1541, loss: 0.020655667409300804 2023-01-23 14:03:26.411332: step: 3082/1541, loss: 0.0575631782412529 2023-01-23 14:03:27.030706: step: 3084/1541, loss: 0.01470599602907896 ================================================== Loss: 0.123 -------------------- Dev Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Dev Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6851670471807466, 'r': 0.6438504413206011, 'f1': 0.6638665172165784}, 'combined': 0.4378295268825317, 'epoch': 18} Test Korean: {'event': {'p': 0.6110581506196378, 'r': 0.7681246255242661, 'f1': 0.6806477302893549}, 'argument': {'p': 0.7273354643967593, 'r': 0.6979868403947849, 'f1': 0.712358995568536}, 'combined': 0.4848655334849286, 'epoch': 18} Dev Russian: {'event': {'p': 0.6563981042654028, 'r': 0.7376830892143809, 'f1': 0.6946708463949843}, 'argument': {'p': 0.6913057944643356, 'r': 0.6449758127570119, 'f1': 0.6673376543368226}, 'combined': 0.463580013169404, 'epoch': 18} Test Russian: {'event': {'p': 0.6514913657770801, 'r': 0.7459556620730977, 'f1': 0.6955307262569833}, 'argument': {'p': 0.726321562078031, 'r': 0.6878064810362723, 'f1': 0.7065395246861376}, 'combined': 0.4914199487342131, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6270833333333333, 'r': 0.33444444444444443, 'f1': 0.43623188405797103}, 'combined': 0.2610521510898094, 'epoch': 18} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5550541813172126, 'r': 0.5550541813172126, 'f1': 0.5550541813172126}, 'combined': 0.26959774521121754, 'epoch': 18} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test for Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 19 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 14:08:54.868336: step: 2/1541, loss: 0.006840829737484455 2023-01-23 14:08:55.777706: step: 4/1541, loss: 0.040367335081100464 2023-01-23 14:08:56.640259: step: 6/1541, loss: 0.0007323858444578946 2023-01-23 14:08:57.651207: step: 8/1541, loss: 0.007294128183275461 2023-01-23 14:08:58.513565: step: 10/1541, loss: 0.15124794840812683 2023-01-23 14:08:59.392010: step: 12/1541, loss: 0.041203852742910385 2023-01-23 14:09:00.256925: step: 14/1541, loss: 0.07364069670438766 2023-01-23 14:09:01.121554: step: 16/1541, loss: 0.0537496916949749 2023-01-23 14:09:01.992337: step: 18/1541, loss: 0.025874746963381767 2023-01-23 14:09:02.891291: step: 20/1541, loss: 0.036253541707992554 2023-01-23 14:09:03.773109: step: 22/1541, loss: 0.02623746544122696 2023-01-23 14:09:04.640697: step: 24/1541, loss: 0.001265228958800435 2023-01-23 14:09:05.578637: step: 26/1541, loss: 0.1356028914451599 2023-01-23 14:09:06.423180: step: 28/1541, loss: 0.1806621104478836 2023-01-23 14:09:07.379576: step: 30/1541, loss: 0.4861183166503906 2023-01-23 14:09:08.267688: step: 32/1541, loss: 0.0034757694229483604 2023-01-23 14:09:09.167268: step: 34/1541, loss: 0.0010571201564744115 2023-01-23 14:09:10.153023: step: 36/1541, loss: 0.024414237588644028 2023-01-23 14:09:11.067296: step: 38/1541, loss: 0.09139170497655869 2023-01-23 14:09:12.013122: step: 40/1541, loss: 0.08590928465127945 2023-01-23 14:09:12.910774: step: 42/1541, loss: 0.2077408730983734 2023-01-23 14:09:13.833248: step: 44/1541, loss: 0.08669878542423248 2023-01-23 14:09:14.708079: step: 46/1541, loss: 0.00810203980654478 2023-01-23 14:09:15.605683: step: 48/1541, loss: 0.037748921662569046 2023-01-23 14:09:16.547546: step: 50/1541, loss: 0.04086484760046005 2023-01-23 14:09:17.410171: step: 52/1541, loss: 0.10321012884378433 2023-01-23 14:09:18.290612: step: 54/1541, loss: 0.20111702382564545 2023-01-23 14:09:19.179172: step: 56/1541, loss: 0.012548195198178291 2023-01-23 14:09:20.097070: step: 58/1541, loss: 0.000904937565792352 2023-01-23 14:09:20.969188: step: 60/1541, loss: 0.0412137545645237 2023-01-23 14:09:21.911962: step: 62/1541, loss: 0.15586550533771515 2023-01-23 14:09:22.804978: step: 64/1541, loss: 0.0024217991158366203 2023-01-23 14:09:23.649394: step: 66/1541, loss: 0.2143872082233429 2023-01-23 14:09:24.601458: step: 68/1541, loss: 0.029979202896356583 2023-01-23 14:09:25.459252: step: 70/1541, loss: 0.05628613755106926 2023-01-23 14:09:26.431309: step: 72/1541, loss: 0.23381885886192322 2023-01-23 14:09:27.333009: step: 74/1541, loss: 0.011132079176604748 2023-01-23 14:09:28.237701: step: 76/1541, loss: 0.1456516534090042 2023-01-23 14:09:29.137002: step: 78/1541, loss: 0.017415720969438553 2023-01-23 14:09:30.066953: step: 80/1541, loss: 0.18799979984760284 2023-01-23 14:09:30.981329: step: 82/1541, loss: 0.07114081084728241 2023-01-23 14:09:31.924445: step: 84/1541, loss: 0.0048638684675097466 2023-01-23 14:09:32.900964: step: 86/1541, loss: 0.003417287254706025 2023-01-23 14:09:33.809704: step: 88/1541, loss: 0.0364559032022953 2023-01-23 14:09:34.706442: step: 90/1541, loss: 0.02522142231464386 2023-01-23 14:09:35.580647: step: 92/1541, loss: 0.027187205851078033 2023-01-23 14:09:36.497635: step: 94/1541, loss: 0.018151242285966873 2023-01-23 14:09:37.360345: step: 96/1541, loss: 0.0003890358784701675 2023-01-23 14:09:38.239679: step: 98/1541, loss: 0.016628213226795197 2023-01-23 14:09:39.140734: step: 100/1541, loss: 0.07832306623458862 2023-01-23 14:09:40.041083: step: 102/1541, loss: 0.084125816822052 2023-01-23 14:09:40.999471: step: 104/1541, loss: 0.060092583298683167 2023-01-23 14:09:41.868384: step: 106/1541, loss: 0.11145110428333282 2023-01-23 14:09:42.741810: step: 108/1541, loss: 0.29788264632225037 2023-01-23 14:09:43.612187: step: 110/1541, loss: 0.0029968356247991323 2023-01-23 14:09:44.522827: step: 112/1541, loss: 0.0002826134441420436 2023-01-23 14:09:45.427603: step: 114/1541, loss: 0.01887044869363308 2023-01-23 14:09:46.335826: step: 116/1541, loss: 0.0017345882952213287 2023-01-23 14:09:47.314816: step: 118/1541, loss: 0.0024150956887751818 2023-01-23 14:09:48.203068: step: 120/1541, loss: 0.11361192166805267 2023-01-23 14:09:49.073578: step: 122/1541, loss: 0.43202367424964905 2023-01-23 14:09:49.932214: step: 124/1541, loss: 0.13464301824569702 2023-01-23 14:09:50.857941: step: 126/1541, loss: 0.0021861859131604433 2023-01-23 14:09:51.768958: step: 128/1541, loss: 0.0013958357740193605 2023-01-23 14:09:52.644141: step: 130/1541, loss: 0.11090469360351562 2023-01-23 14:09:53.599803: step: 132/1541, loss: 0.09767448157072067 2023-01-23 14:09:54.494526: step: 134/1541, loss: 0.015955323353409767 2023-01-23 14:09:55.354084: step: 136/1541, loss: 0.7606996893882751 2023-01-23 14:09:56.213460: step: 138/1541, loss: 0.001636892557144165 2023-01-23 14:09:57.182731: step: 140/1541, loss: 0.0014043181436136365 2023-01-23 14:09:58.070945: step: 142/1541, loss: 0.004770076833665371 2023-01-23 14:09:58.933291: step: 144/1541, loss: 0.019355854019522667 2023-01-23 14:09:59.814802: step: 146/1541, loss: 0.13152213394641876 2023-01-23 14:10:00.674448: step: 148/1541, loss: 2.6560583114624023 2023-01-23 14:10:01.635627: step: 150/1541, loss: 0.06873039901256561 2023-01-23 14:10:02.527573: step: 152/1541, loss: 0.01099498849362135 2023-01-23 14:10:03.405002: step: 154/1541, loss: 0.13756924867630005 2023-01-23 14:10:04.312585: step: 156/1541, loss: 0.0008709438261575997 2023-01-23 14:10:05.188837: step: 158/1541, loss: 0.10453835874795914 2023-01-23 14:10:06.084286: step: 160/1541, loss: 0.001481763320043683 2023-01-23 14:10:06.957773: step: 162/1541, loss: 0.9090039134025574 2023-01-23 14:10:07.887217: step: 164/1541, loss: 0.09700421988964081 2023-01-23 14:10:08.754932: step: 166/1541, loss: 0.021261028945446014 2023-01-23 14:10:09.640235: step: 168/1541, loss: 0.005449088290333748 2023-01-23 14:10:10.497085: step: 170/1541, loss: 0.1699041724205017 2023-01-23 14:10:11.443372: step: 172/1541, loss: 0.07044477015733719 2023-01-23 14:10:12.342979: step: 174/1541, loss: 0.03822760283946991 2023-01-23 14:10:13.220516: step: 176/1541, loss: 0.05608749017119408 2023-01-23 14:10:14.127675: step: 178/1541, loss: 0.0009239721694029868 2023-01-23 14:10:15.016233: step: 180/1541, loss: 0.026134127750992775 2023-01-23 14:10:15.902039: step: 182/1541, loss: 0.04435925930738449 2023-01-23 14:10:16.788776: step: 184/1541, loss: 0.04098604619503021 2023-01-23 14:10:17.694172: step: 186/1541, loss: 0.01204367820173502 2023-01-23 14:10:18.607333: step: 188/1541, loss: 0.0026138366665691137 2023-01-23 14:10:19.495422: step: 190/1541, loss: 0.03476361557841301 2023-01-23 14:10:20.360923: step: 192/1541, loss: 0.004590250551700592 2023-01-23 14:10:21.211156: step: 194/1541, loss: 0.07475150376558304 2023-01-23 14:10:22.116153: step: 196/1541, loss: 0.021029949188232422 2023-01-23 14:10:23.028838: step: 198/1541, loss: 0.006518982816487551 2023-01-23 14:10:23.914103: step: 200/1541, loss: 0.07010011374950409 2023-01-23 14:10:24.861739: step: 202/1541, loss: 0.002858005929738283 2023-01-23 14:10:25.777753: step: 204/1541, loss: 0.11801641434431076 2023-01-23 14:10:26.699594: step: 206/1541, loss: 0.055373914539813995 2023-01-23 14:10:27.590233: step: 208/1541, loss: 0.008373265154659748 2023-01-23 14:10:28.526583: step: 210/1541, loss: 0.002181770745664835 2023-01-23 14:10:29.383466: step: 212/1541, loss: 0.007560813799500465 2023-01-23 14:10:30.260766: step: 214/1541, loss: 0.020653950050473213 2023-01-23 14:10:31.125015: step: 216/1541, loss: 0.01600203849375248 2023-01-23 14:10:32.024891: step: 218/1541, loss: 0.08382964879274368 2023-01-23 14:10:32.952003: step: 220/1541, loss: 0.07811086624860764 2023-01-23 14:10:33.844014: step: 222/1541, loss: 0.3343915343284607 2023-01-23 14:10:34.765719: step: 224/1541, loss: 0.07146095484495163 2023-01-23 14:10:35.680195: step: 226/1541, loss: 0.028385814279317856 2023-01-23 14:10:36.582584: step: 228/1541, loss: 0.058994486927986145 2023-01-23 14:10:37.558810: step: 230/1541, loss: 0.0037122301291674376 2023-01-23 14:10:38.436420: step: 232/1541, loss: 0.0034146185498684645 2023-01-23 14:10:39.313134: step: 234/1541, loss: 0.01489188615232706 2023-01-23 14:10:40.206011: step: 236/1541, loss: 0.2996954023838043 2023-01-23 14:10:41.109231: step: 238/1541, loss: 0.03876380994915962 2023-01-23 14:10:42.033984: step: 240/1541, loss: 0.14262782037258148 2023-01-23 14:10:42.898396: step: 242/1541, loss: 0.0907682552933693 2023-01-23 14:10:43.784020: step: 244/1541, loss: 0.04478205367922783 2023-01-23 14:10:44.696425: step: 246/1541, loss: 0.013370675034821033 2023-01-23 14:10:45.529042: step: 248/1541, loss: 0.017697138711810112 2023-01-23 14:10:46.425003: step: 250/1541, loss: 0.06689522415399551 2023-01-23 14:10:47.323753: step: 252/1541, loss: 0.008446060121059418 2023-01-23 14:10:48.285264: step: 254/1541, loss: 0.085657998919487 2023-01-23 14:10:49.193685: step: 256/1541, loss: 0.22088506817817688 2023-01-23 14:10:50.118692: step: 258/1541, loss: 0.07021859288215637 2023-01-23 14:10:50.993952: step: 260/1541, loss: 0.1996653825044632 2023-01-23 14:10:51.907143: step: 262/1541, loss: 0.0002306547248736024 2023-01-23 14:10:52.851733: step: 264/1541, loss: 0.009747855365276337 2023-01-23 14:10:53.786353: step: 266/1541, loss: 0.003512374823912978 2023-01-23 14:10:54.749086: step: 268/1541, loss: 0.20959508419036865 2023-01-23 14:10:55.683581: step: 270/1541, loss: 0.34060120582580566 2023-01-23 14:10:56.573680: step: 272/1541, loss: 0.0010767203057184815 2023-01-23 14:10:57.475543: step: 274/1541, loss: 1.1445280313491821 2023-01-23 14:10:58.387082: step: 276/1541, loss: 0.003024660050868988 2023-01-23 14:10:59.277375: step: 278/1541, loss: 0.2518146336078644 2023-01-23 14:11:00.156162: step: 280/1541, loss: 0.012587959878146648 2023-01-23 14:11:01.120887: step: 282/1541, loss: 0.041392892599105835 2023-01-23 14:11:02.006777: step: 284/1541, loss: 0.0385383740067482 2023-01-23 14:11:02.939681: step: 286/1541, loss: 0.014825062826275826 2023-01-23 14:11:03.806683: step: 288/1541, loss: 0.008073842152953148 2023-01-23 14:11:04.694059: step: 290/1541, loss: 0.006481650285422802 2023-01-23 14:11:05.585489: step: 292/1541, loss: 0.26784226298332214 2023-01-23 14:11:06.456160: step: 294/1541, loss: 0.004613989498466253 2023-01-23 14:11:07.372278: step: 296/1541, loss: 0.28859034180641174 2023-01-23 14:11:08.265427: step: 298/1541, loss: 0.012054420076310635 2023-01-23 14:11:09.154555: step: 300/1541, loss: 0.0012745774583891034 2023-01-23 14:11:10.051242: step: 302/1541, loss: 0.006059629842638969 2023-01-23 14:11:11.006626: step: 304/1541, loss: 0.08918789774179459 2023-01-23 14:11:11.875318: step: 306/1541, loss: 0.173869326710701 2023-01-23 14:11:12.784545: step: 308/1541, loss: 0.1763545423746109 2023-01-23 14:11:13.684342: step: 310/1541, loss: 0.08904216438531876 2023-01-23 14:11:14.684238: step: 312/1541, loss: 0.025651300325989723 2023-01-23 14:11:15.673439: step: 314/1541, loss: 0.11281001567840576 2023-01-23 14:11:16.639774: step: 316/1541, loss: 0.0006107885856181383 2023-01-23 14:11:17.550396: step: 318/1541, loss: 0.0012050573714077473 2023-01-23 14:11:18.450843: step: 320/1541, loss: 0.0028578119818121195 2023-01-23 14:11:19.319015: step: 322/1541, loss: 0.013976125977933407 2023-01-23 14:11:20.205010: step: 324/1541, loss: 0.004844714887440205 2023-01-23 14:11:21.086931: step: 326/1541, loss: 0.06740359216928482 2023-01-23 14:11:22.027188: step: 328/1541, loss: 0.000699551310390234 2023-01-23 14:11:22.899730: step: 330/1541, loss: 0.008002225309610367 2023-01-23 14:11:23.812046: step: 332/1541, loss: 0.02539239265024662 2023-01-23 14:11:24.667409: step: 334/1541, loss: 0.0730142593383789 2023-01-23 14:11:25.598798: step: 336/1541, loss: 0.07125365734100342 2023-01-23 14:11:26.473828: step: 338/1541, loss: 0.1724403351545334 2023-01-23 14:11:27.370273: step: 340/1541, loss: 0.009501478634774685 2023-01-23 14:11:28.233753: step: 342/1541, loss: 0.12148158252239227 2023-01-23 14:11:29.102832: step: 344/1541, loss: 0.13111518323421478 2023-01-23 14:11:29.991585: step: 346/1541, loss: 0.0017371916910633445 2023-01-23 14:11:30.882903: step: 348/1541, loss: 0.0014267127262428403 2023-01-23 14:11:31.792301: step: 350/1541, loss: 0.1824961155653 2023-01-23 14:11:32.714799: step: 352/1541, loss: 0.3287135064601898 2023-01-23 14:11:33.588810: step: 354/1541, loss: 1.1267180442810059 2023-01-23 14:11:34.465348: step: 356/1541, loss: 0.0018038711277768016 2023-01-23 14:11:35.347816: step: 358/1541, loss: 0.20279745757579803 2023-01-23 14:11:36.281721: step: 360/1541, loss: 0.008781959302723408 2023-01-23 14:11:37.200723: step: 362/1541, loss: 0.026398031041026115 2023-01-23 14:11:38.073595: step: 364/1541, loss: 0.08725106716156006 2023-01-23 14:11:38.978272: step: 366/1541, loss: 0.009967723861336708 2023-01-23 14:11:39.871070: step: 368/1541, loss: 0.03542924299836159 2023-01-23 14:11:40.757535: step: 370/1541, loss: 0.13195057213306427 2023-01-23 14:11:41.631719: step: 372/1541, loss: 0.01760764792561531 2023-01-23 14:11:42.548073: step: 374/1541, loss: 0.007269497029483318 2023-01-23 14:11:43.469289: step: 376/1541, loss: 1.0688844919204712 2023-01-23 14:11:44.340018: step: 378/1541, loss: 0.003632102394476533 2023-01-23 14:11:45.270344: step: 380/1541, loss: 0.013673050329089165 2023-01-23 14:11:46.176226: step: 382/1541, loss: 0.033165011554956436 2023-01-23 14:11:47.106390: step: 384/1541, loss: 0.04596991837024689 2023-01-23 14:11:48.044710: step: 386/1541, loss: 0.13115812838077545 2023-01-23 14:11:48.946050: step: 388/1541, loss: 0.2566492557525635 2023-01-23 14:11:49.836894: step: 390/1541, loss: 0.07478619366884232 2023-01-23 14:11:50.736031: step: 392/1541, loss: 0.015239792875945568 2023-01-23 14:11:51.645200: step: 394/1541, loss: 0.06627656519412994 2023-01-23 14:11:52.561420: step: 396/1541, loss: 0.0004029665142297745 2023-01-23 14:11:53.429597: step: 398/1541, loss: 0.001658988301642239 2023-01-23 14:11:54.322900: step: 400/1541, loss: 0.08125725388526917 2023-01-23 14:11:55.224792: step: 402/1541, loss: 0.07317018508911133 2023-01-23 14:11:56.190484: step: 404/1541, loss: 0.24633410573005676 2023-01-23 14:11:57.095756: step: 406/1541, loss: 0.011146720498800278 2023-01-23 14:11:58.025275: step: 408/1541, loss: 0.12428318709135056 2023-01-23 14:11:58.939378: step: 410/1541, loss: 0.02376268431544304 2023-01-23 14:11:59.825242: step: 412/1541, loss: 0.32940953969955444 2023-01-23 14:12:00.734977: step: 414/1541, loss: 0.03130790963768959 2023-01-23 14:12:01.624924: step: 416/1541, loss: 0.037390630692243576 2023-01-23 14:12:02.552116: step: 418/1541, loss: 0.0317574217915535 2023-01-23 14:12:03.436812: step: 420/1541, loss: 0.008186902850866318 2023-01-23 14:12:04.317548: step: 422/1541, loss: 0.03363414108753204 2023-01-23 14:12:05.222574: step: 424/1541, loss: 0.01960226707160473 2023-01-23 14:12:06.092755: step: 426/1541, loss: 0.5701298713684082 2023-01-23 14:12:06.966895: step: 428/1541, loss: 0.6641824245452881 2023-01-23 14:12:07.847890: step: 430/1541, loss: 0.11761171370744705 2023-01-23 14:12:08.752327: step: 432/1541, loss: 0.008732864633202553 2023-01-23 14:12:09.689841: step: 434/1541, loss: 0.008679170161485672 2023-01-23 14:12:10.584712: step: 436/1541, loss: 0.10414771735668182 2023-01-23 14:12:11.484304: step: 438/1541, loss: 0.0005080229020677507 2023-01-23 14:12:12.376620: step: 440/1541, loss: 0.07571485638618469 2023-01-23 14:12:13.260285: step: 442/1541, loss: 0.007136153057217598 2023-01-23 14:12:14.101046: step: 444/1541, loss: 0.0019597262144088745 2023-01-23 14:12:14.976422: step: 446/1541, loss: 0.16802288591861725 2023-01-23 14:12:15.898685: step: 448/1541, loss: 0.024029023945331573 2023-01-23 14:12:16.816280: step: 450/1541, loss: 0.015768343582749367 2023-01-23 14:12:17.762837: step: 452/1541, loss: 0.1496095210313797 2023-01-23 14:12:18.683517: step: 454/1541, loss: 0.01613960787653923 2023-01-23 14:12:19.592279: step: 456/1541, loss: 0.03820876404643059 2023-01-23 14:12:20.499119: step: 458/1541, loss: 0.02104012668132782 2023-01-23 14:12:21.505115: step: 460/1541, loss: 0.024899175390601158 2023-01-23 14:12:22.376087: step: 462/1541, loss: 0.12633958458900452 2023-01-23 14:12:23.300591: step: 464/1541, loss: 0.28701338171958923 2023-01-23 14:12:24.258852: step: 466/1541, loss: 0.16580958664417267 2023-01-23 14:12:25.153075: step: 468/1541, loss: 0.0346025750041008 2023-01-23 14:12:26.073501: step: 470/1541, loss: 0.027306580916047096 2023-01-23 14:12:27.053776: step: 472/1541, loss: 0.010097975842654705 2023-01-23 14:12:27.912873: step: 474/1541, loss: 0.0992480143904686 2023-01-23 14:12:28.805655: step: 476/1541, loss: 0.09442628920078278 2023-01-23 14:12:29.688242: step: 478/1541, loss: 0.04372538626194 2023-01-23 14:12:30.574240: step: 480/1541, loss: 0.16959640383720398 2023-01-23 14:12:31.465000: step: 482/1541, loss: 0.10041357576847076 2023-01-23 14:12:32.352023: step: 484/1541, loss: 0.037499185651540756 2023-01-23 14:12:33.278751: step: 486/1541, loss: 0.008355815894901752 2023-01-23 14:12:34.152207: step: 488/1541, loss: 0.013183551840484142 2023-01-23 14:12:35.052691: step: 490/1541, loss: 0.032721687108278275 2023-01-23 14:12:35.917644: step: 492/1541, loss: 0.012975478544831276 2023-01-23 14:12:36.868765: step: 494/1541, loss: 0.30350539088249207 2023-01-23 14:12:37.771711: step: 496/1541, loss: 0.04123743623495102 2023-01-23 14:12:38.691897: step: 498/1541, loss: 0.04925207793712616 2023-01-23 14:12:39.560760: step: 500/1541, loss: 0.1650976687669754 2023-01-23 14:12:40.476836: step: 502/1541, loss: 0.027280861511826515 2023-01-23 14:12:41.335838: step: 504/1541, loss: 0.04828439652919769 2023-01-23 14:12:42.232792: step: 506/1541, loss: 0.0072444467805325985 2023-01-23 14:12:43.141110: step: 508/1541, loss: 0.006554941181093454 2023-01-23 14:12:44.032826: step: 510/1541, loss: 0.03865527734160423 2023-01-23 14:12:44.951512: step: 512/1541, loss: 0.016096001490950584 2023-01-23 14:12:45.902571: step: 514/1541, loss: 0.009478316642343998 2023-01-23 14:12:46.838787: step: 516/1541, loss: 0.19917680323123932 2023-01-23 14:12:47.740875: step: 518/1541, loss: 0.008196617476642132 2023-01-23 14:12:48.633606: step: 520/1541, loss: 0.0009077187278307974 2023-01-23 14:12:49.555888: step: 522/1541, loss: 0.14131386578083038 2023-01-23 14:12:50.430720: step: 524/1541, loss: 0.011077404022216797 2023-01-23 14:12:51.322957: step: 526/1541, loss: 0.10222513228654861 2023-01-23 14:12:52.248033: step: 528/1541, loss: 0.032381851226091385 2023-01-23 14:12:53.149248: step: 530/1541, loss: 0.3683755397796631 2023-01-23 14:12:53.989212: step: 532/1541, loss: 0.0020749513059854507 2023-01-23 14:12:54.898521: step: 534/1541, loss: 0.056846555322408676 2023-01-23 14:12:55.835186: step: 536/1541, loss: 0.005631374195218086 2023-01-23 14:12:56.728868: step: 538/1541, loss: 0.2672826647758484 2023-01-23 14:12:57.602243: step: 540/1541, loss: 0.0023353705182671547 2023-01-23 14:12:58.537552: step: 542/1541, loss: 0.03684106096625328 2023-01-23 14:12:59.457803: step: 544/1541, loss: 0.06169937923550606 2023-01-23 14:13:00.413893: step: 546/1541, loss: 0.03804774209856987 2023-01-23 14:13:01.308588: step: 548/1541, loss: 0.02787112258374691 2023-01-23 14:13:02.262103: step: 550/1541, loss: 0.06573978811502457 2023-01-23 14:13:03.137618: step: 552/1541, loss: 0.0034516113810241222 2023-01-23 14:13:03.968031: step: 554/1541, loss: 0.006257480941712856 2023-01-23 14:13:04.838185: step: 556/1541, loss: 0.04241759330034256 2023-01-23 14:13:05.790295: step: 558/1541, loss: 0.0018500578589737415 2023-01-23 14:13:06.689069: step: 560/1541, loss: 0.06268831342458725 2023-01-23 14:13:07.580808: step: 562/1541, loss: 0.12899529933929443 2023-01-23 14:13:08.482393: step: 564/1541, loss: 0.122866690158844 2023-01-23 14:13:09.352501: step: 566/1541, loss: 0.0004092956951353699 2023-01-23 14:13:10.244776: step: 568/1541, loss: 0.47684839367866516 2023-01-23 14:13:11.144451: step: 570/1541, loss: 0.02297895960509777 2023-01-23 14:13:12.109334: step: 572/1541, loss: 0.07480262964963913 2023-01-23 14:13:12.987030: step: 574/1541, loss: 0.34592998027801514 2023-01-23 14:13:13.848620: step: 576/1541, loss: 0.00018885050667449832 2023-01-23 14:13:14.726593: step: 578/1541, loss: 0.008898770436644554 2023-01-23 14:13:15.585112: step: 580/1541, loss: 0.03114832192659378 2023-01-23 14:13:16.487427: step: 582/1541, loss: 0.5890554189682007 2023-01-23 14:13:17.393561: step: 584/1541, loss: 0.011941581033170223 2023-01-23 14:13:18.329514: step: 586/1541, loss: 0.0918821319937706 2023-01-23 14:13:19.365904: step: 588/1541, loss: 0.0009893012465909123 2023-01-23 14:13:20.261843: step: 590/1541, loss: 0.07285910099744797 2023-01-23 14:13:21.166230: step: 592/1541, loss: 0.03813472017645836 2023-01-23 14:13:22.066908: step: 594/1541, loss: 0.02108285389840603 2023-01-23 14:13:22.969753: step: 596/1541, loss: 0.01795687898993492 2023-01-23 14:13:23.902624: step: 598/1541, loss: 0.022302640601992607 2023-01-23 14:13:24.806229: step: 600/1541, loss: 0.47670838236808777 2023-01-23 14:13:25.736801: step: 602/1541, loss: 0.22584745287895203 2023-01-23 14:13:26.664417: step: 604/1541, loss: 0.06956525146961212 2023-01-23 14:13:27.647527: step: 606/1541, loss: 0.011914217844605446 2023-01-23 14:13:28.572094: step: 608/1541, loss: 0.012513760477304459 2023-01-23 14:13:29.502870: step: 610/1541, loss: 0.014949439093470573 2023-01-23 14:13:30.472302: step: 612/1541, loss: 0.00910663791000843 2023-01-23 14:13:31.378680: step: 614/1541, loss: 0.31080520153045654 2023-01-23 14:13:32.281830: step: 616/1541, loss: 0.00011768780677812174 2023-01-23 14:13:33.129421: step: 618/1541, loss: 0.0455767922103405 2023-01-23 14:13:33.991846: step: 620/1541, loss: 0.7190796732902527 2023-01-23 14:13:34.893348: step: 622/1541, loss: 0.10173318535089493 2023-01-23 14:13:35.791108: step: 624/1541, loss: 0.24984468519687653 2023-01-23 14:13:36.719117: step: 626/1541, loss: 0.08822129666805267 2023-01-23 14:13:37.608261: step: 628/1541, loss: 0.013134979642927647 2023-01-23 14:13:38.533848: step: 630/1541, loss: 0.000368805427569896 2023-01-23 14:13:39.456630: step: 632/1541, loss: 0.010551095008850098 2023-01-23 14:13:40.361042: step: 634/1541, loss: 0.14581620693206787 2023-01-23 14:13:41.248528: step: 636/1541, loss: 0.057936251163482666 2023-01-23 14:13:42.133340: step: 638/1541, loss: 0.025201721116900444 2023-01-23 14:13:43.040892: step: 640/1541, loss: 0.029535487294197083 2023-01-23 14:13:43.909938: step: 642/1541, loss: 0.00039017823291942477 2023-01-23 14:13:44.814449: step: 644/1541, loss: 0.09351978451013565 2023-01-23 14:13:45.712337: step: 646/1541, loss: 0.16154365241527557 2023-01-23 14:13:46.589489: step: 648/1541, loss: 0.007550468668341637 2023-01-23 14:13:47.458943: step: 650/1541, loss: 0.06222234666347504 2023-01-23 14:13:48.362153: step: 652/1541, loss: 0.030641604214906693 2023-01-23 14:13:49.236868: step: 654/1541, loss: 0.0030363101977854967 2023-01-23 14:13:50.162055: step: 656/1541, loss: 0.01224206667393446 2023-01-23 14:13:51.021936: step: 658/1541, loss: 0.04391255974769592 2023-01-23 14:13:51.963439: step: 660/1541, loss: 0.0028435152489691973 2023-01-23 14:13:52.921453: step: 662/1541, loss: 0.002190619008615613 2023-01-23 14:13:53.798232: step: 664/1541, loss: 0.04512029141187668 2023-01-23 14:13:54.721722: step: 666/1541, loss: 0.3578680753707886 2023-01-23 14:13:55.573226: step: 668/1541, loss: 0.008981378749012947 2023-01-23 14:13:56.589175: step: 670/1541, loss: 0.0012066637864336371 2023-01-23 14:13:57.539352: step: 672/1541, loss: 0.2969602048397064 2023-01-23 14:13:58.455844: step: 674/1541, loss: 0.13056990504264832 2023-01-23 14:13:59.328126: step: 676/1541, loss: 0.031487029045820236 2023-01-23 14:14:00.227192: step: 678/1541, loss: 0.17099781334400177 2023-01-23 14:14:01.122716: step: 680/1541, loss: 0.5483419895172119 2023-01-23 14:14:01.992284: step: 682/1541, loss: 0.024837331846356392 2023-01-23 14:14:02.910362: step: 684/1541, loss: 0.0788377970457077 2023-01-23 14:14:03.842508: step: 686/1541, loss: 0.29833438992500305 2023-01-23 14:14:04.799972: step: 688/1541, loss: 0.08901110291481018 2023-01-23 14:14:05.740452: step: 690/1541, loss: 0.0005893887137062848 2023-01-23 14:14:06.637758: step: 692/1541, loss: 0.05800052359700203 2023-01-23 14:14:07.542067: step: 694/1541, loss: 0.03751714900135994 2023-01-23 14:14:08.404856: step: 696/1541, loss: 0.0007441360503435135 2023-01-23 14:14:09.277823: step: 698/1541, loss: 0.0008670261595398188 2023-01-23 14:14:10.163825: step: 700/1541, loss: 0.0069167474284768105 2023-01-23 14:14:11.041845: step: 702/1541, loss: 0.0010748449712991714 2023-01-23 14:14:11.926904: step: 704/1541, loss: 0.03652012720704079 2023-01-23 14:14:12.829506: step: 706/1541, loss: 0.02139975316822529 2023-01-23 14:14:13.773429: step: 708/1541, loss: 0.013993465341627598 2023-01-23 14:14:14.643047: step: 710/1541, loss: 0.05937807261943817 2023-01-23 14:14:15.502870: step: 712/1541, loss: 0.093086376786232 2023-01-23 14:14:16.362263: step: 714/1541, loss: 0.006571030244231224 2023-01-23 14:14:17.249642: step: 716/1541, loss: 0.038606926798820496 2023-01-23 14:14:18.151989: step: 718/1541, loss: 0.06824749708175659 2023-01-23 14:14:19.071582: step: 720/1541, loss: 0.002338886260986328 2023-01-23 14:14:19.925713: step: 722/1541, loss: 0.0167354978621006 2023-01-23 14:14:20.803054: step: 724/1541, loss: 0.14220446348190308 2023-01-23 14:14:21.741492: step: 726/1541, loss: 0.7966555953025818 2023-01-23 14:14:22.625984: step: 728/1541, loss: 0.0011269768001511693 2023-01-23 14:14:23.480363: step: 730/1541, loss: 0.014856810681521893 2023-01-23 14:14:24.385956: step: 732/1541, loss: 0.19165684282779694 2023-01-23 14:14:25.320793: step: 734/1541, loss: 0.07529016584157944 2023-01-23 14:14:26.236845: step: 736/1541, loss: 0.03771274909377098 2023-01-23 14:14:27.110167: step: 738/1541, loss: 0.1598612368106842 2023-01-23 14:14:28.048140: step: 740/1541, loss: 0.0343526266515255 2023-01-23 14:14:28.983966: step: 742/1541, loss: 0.0026176758110523224 2023-01-23 14:14:29.863276: step: 744/1541, loss: 1.1910691261291504 2023-01-23 14:14:30.765124: step: 746/1541, loss: 0.02226872555911541 2023-01-23 14:14:31.731709: step: 748/1541, loss: 0.1392821967601776 2023-01-23 14:14:32.638538: step: 750/1541, loss: 0.010969199240207672 2023-01-23 14:14:33.605330: step: 752/1541, loss: 0.005618890281766653 2023-01-23 14:14:34.494283: step: 754/1541, loss: 0.018233774229884148 2023-01-23 14:14:35.420359: step: 756/1541, loss: 0.005814037751406431 2023-01-23 14:14:36.348968: step: 758/1541, loss: 0.3099656105041504 2023-01-23 14:14:37.249403: step: 760/1541, loss: 0.008256305009126663 2023-01-23 14:14:38.158203: step: 762/1541, loss: 0.002923267660662532 2023-01-23 14:14:39.070513: step: 764/1541, loss: 0.028958484530448914 2023-01-23 14:14:39.980397: step: 766/1541, loss: 0.011769245378673077 2023-01-23 14:14:40.874255: step: 768/1541, loss: 0.02862417697906494 2023-01-23 14:14:41.823065: step: 770/1541, loss: 0.023071860894560814 2023-01-23 14:14:42.666743: step: 772/1541, loss: 0.004003632813692093 2023-01-23 14:14:43.575597: step: 774/1541, loss: 0.20411229133605957 2023-01-23 14:14:44.508064: step: 776/1541, loss: 0.3748527467250824 2023-01-23 14:14:45.440950: step: 778/1541, loss: 0.002817860571667552 2023-01-23 14:14:46.311341: step: 780/1541, loss: 0.1408853381872177 2023-01-23 14:14:47.250310: step: 782/1541, loss: 0.43074873089790344 2023-01-23 14:14:48.147274: step: 784/1541, loss: 0.06411497294902802 2023-01-23 14:14:49.089432: step: 786/1541, loss: 0.09981412440538406 2023-01-23 14:14:49.984174: step: 788/1541, loss: 0.020840276032686234 2023-01-23 14:14:50.854016: step: 790/1541, loss: 0.011936821974813938 2023-01-23 14:14:51.744805: step: 792/1541, loss: 0.002594290766865015 2023-01-23 14:14:52.592698: step: 794/1541, loss: 0.18289028108119965 2023-01-23 14:14:53.493544: step: 796/1541, loss: 0.0008263294002972543 2023-01-23 14:14:54.371791: step: 798/1541, loss: 0.03704765811562538 2023-01-23 14:14:55.285326: step: 800/1541, loss: 0.10412431508302689 2023-01-23 14:14:56.138758: step: 802/1541, loss: 0.005175029393285513 2023-01-23 14:14:57.038394: step: 804/1541, loss: 0.05252726003527641 2023-01-23 14:14:57.890430: step: 806/1541, loss: 0.008414492011070251 2023-01-23 14:14:58.753512: step: 808/1541, loss: 0.026448287069797516 2023-01-23 14:14:59.637245: step: 810/1541, loss: 0.3097690939903259 2023-01-23 14:15:00.498752: step: 812/1541, loss: 0.036184508353471756 2023-01-23 14:15:01.393057: step: 814/1541, loss: 0.04901780188083649 2023-01-23 14:15:02.279939: step: 816/1541, loss: 0.09503621608018875 2023-01-23 14:15:03.193969: step: 818/1541, loss: 0.04631217196583748 2023-01-23 14:15:04.072264: step: 820/1541, loss: 0.0027177222073078156 2023-01-23 14:15:04.957392: step: 822/1541, loss: 0.004618196748197079 2023-01-23 14:15:05.844735: step: 824/1541, loss: 0.0018943548202514648 2023-01-23 14:15:06.733089: step: 826/1541, loss: 0.011140797287225723 2023-01-23 14:15:07.691387: step: 828/1541, loss: 0.2471076250076294 2023-01-23 14:15:08.585421: step: 830/1541, loss: 0.0064604757353663445 2023-01-23 14:15:09.493918: step: 832/1541, loss: 0.09542816132307053 2023-01-23 14:15:10.417925: step: 834/1541, loss: 0.04143818840384483 2023-01-23 14:15:11.300625: step: 836/1541, loss: 0.1598784327507019 2023-01-23 14:15:12.213975: step: 838/1541, loss: 0.004192814230918884 2023-01-23 14:15:13.128461: step: 840/1541, loss: 0.0211174376308918 2023-01-23 14:15:14.043479: step: 842/1541, loss: 0.1268792450428009 2023-01-23 14:15:14.936349: step: 844/1541, loss: 0.22190876305103302 2023-01-23 14:15:15.838400: step: 846/1541, loss: 0.15153703093528748 2023-01-23 14:15:16.717156: step: 848/1541, loss: 0.00506852101534605 2023-01-23 14:15:17.645529: step: 850/1541, loss: 0.030054187402129173 2023-01-23 14:15:18.489026: step: 852/1541, loss: 0.0432487390935421 2023-01-23 14:15:19.391735: step: 854/1541, loss: 0.10334019362926483 2023-01-23 14:15:20.283205: step: 856/1541, loss: 0.6138371825218201 2023-01-23 14:15:21.165086: step: 858/1541, loss: 0.3951861262321472 2023-01-23 14:15:22.069056: step: 860/1541, loss: 0.1819702833890915 2023-01-23 14:15:23.023913: step: 862/1541, loss: 0.0596892423927784 2023-01-23 14:15:23.972525: step: 864/1541, loss: 0.015148352831602097 2023-01-23 14:15:24.862835: step: 866/1541, loss: 0.412092924118042 2023-01-23 14:15:25.730862: step: 868/1541, loss: 0.004668254870921373 2023-01-23 14:15:26.655549: step: 870/1541, loss: 0.5519520044326782 2023-01-23 14:15:27.518710: step: 872/1541, loss: 0.0011286498047411442 2023-01-23 14:15:28.430077: step: 874/1541, loss: 0.053412891924381256 2023-01-23 14:15:29.341980: step: 876/1541, loss: 0.014312983490526676 2023-01-23 14:15:30.214313: step: 878/1541, loss: 0.0011039162054657936 2023-01-23 14:15:31.061846: step: 880/1541, loss: 0.9281341433525085 2023-01-23 14:15:31.973014: step: 882/1541, loss: 0.0012299004010856152 2023-01-23 14:15:32.853739: step: 884/1541, loss: 0.0012766222935169935 2023-01-23 14:15:33.723835: step: 886/1541, loss: 0.0594535768032074 2023-01-23 14:15:34.597080: step: 888/1541, loss: 0.23099109530448914 2023-01-23 14:15:35.443776: step: 890/1541, loss: 0.005557533353567123 2023-01-23 14:15:36.321475: step: 892/1541, loss: 0.0034068890381604433 2023-01-23 14:15:37.224580: step: 894/1541, loss: 0.4484725892543793 2023-01-23 14:15:38.070937: step: 896/1541, loss: 0.002637644996866584 2023-01-23 14:15:39.054927: step: 898/1541, loss: 0.0015914423856884241 2023-01-23 14:15:39.944514: step: 900/1541, loss: 0.08655256032943726 2023-01-23 14:15:40.833442: step: 902/1541, loss: 0.17126336693763733 2023-01-23 14:15:41.752113: step: 904/1541, loss: 0.1417098492383957 2023-01-23 14:15:42.614027: step: 906/1541, loss: 0.0029866467230021954 2023-01-23 14:15:43.521875: step: 908/1541, loss: 0.0761360451579094 2023-01-23 14:15:44.417875: step: 910/1541, loss: 0.07055066525936127 2023-01-23 14:15:45.286444: step: 912/1541, loss: 0.01742524281144142 2023-01-23 14:15:46.213625: step: 914/1541, loss: 0.001969712320715189 2023-01-23 14:15:47.114060: step: 916/1541, loss: 0.009747430682182312 2023-01-23 14:15:47.999789: step: 918/1541, loss: 0.09956435114145279 2023-01-23 14:15:48.926447: step: 920/1541, loss: 0.0030845818109810352 2023-01-23 14:15:49.847762: step: 922/1541, loss: 0.0012776937801390886 2023-01-23 14:15:50.725526: step: 924/1541, loss: 0.6568020582199097 2023-01-23 14:15:51.618301: step: 926/1541, loss: 0.0001750742958392948 2023-01-23 14:15:52.473161: step: 928/1541, loss: 0.012170827947556973 2023-01-23 14:15:53.410029: step: 930/1541, loss: 0.3755401074886322 2023-01-23 14:15:54.300766: step: 932/1541, loss: 0.8990718126296997 2023-01-23 14:15:55.190586: step: 934/1541, loss: 0.010977869853377342 2023-01-23 14:15:56.073965: step: 936/1541, loss: 0.06007228419184685 2023-01-23 14:15:56.965248: step: 938/1541, loss: 0.0008927925955504179 2023-01-23 14:15:57.870939: step: 940/1541, loss: 0.005959540605545044 2023-01-23 14:15:58.791671: step: 942/1541, loss: 0.04605838283896446 2023-01-23 14:15:59.734067: step: 944/1541, loss: 0.011185686103999615 2023-01-23 14:16:00.614656: step: 946/1541, loss: 0.002033415250480175 2023-01-23 14:16:01.499255: step: 948/1541, loss: 0.00907213892787695 2023-01-23 14:16:02.430889: step: 950/1541, loss: 0.046052493155002594 2023-01-23 14:16:03.288220: step: 952/1541, loss: 0.0023050070740282536 2023-01-23 14:16:04.202814: step: 954/1541, loss: 0.11603005975484848 2023-01-23 14:16:05.083539: step: 956/1541, loss: 0.01559829618781805 2023-01-23 14:16:05.943176: step: 958/1541, loss: 0.025165637955069542 2023-01-23 14:16:06.865834: step: 960/1541, loss: 0.034558914601802826 2023-01-23 14:16:07.769077: step: 962/1541, loss: 0.10270064324140549 2023-01-23 14:16:08.639966: step: 964/1541, loss: 0.014140799641609192 2023-01-23 14:16:09.510275: step: 966/1541, loss: 0.005471274722367525 2023-01-23 14:16:10.417151: step: 968/1541, loss: 0.09880885481834412 2023-01-23 14:16:11.291761: step: 970/1541, loss: 0.0074617015197873116 2023-01-23 14:16:12.201759: step: 972/1541, loss: 0.00331102148629725 2023-01-23 14:16:13.160069: step: 974/1541, loss: 0.26330646872520447 2023-01-23 14:16:14.039205: step: 976/1541, loss: 0.21903172135353088 2023-01-23 14:16:14.948765: step: 978/1541, loss: 0.0016670726472511888 2023-01-23 14:16:15.842495: step: 980/1541, loss: 0.007165216375142336 2023-01-23 14:16:16.738366: step: 982/1541, loss: 0.05484318733215332 2023-01-23 14:16:17.652874: step: 984/1541, loss: 0.0005745058879256248 2023-01-23 14:16:18.566767: step: 986/1541, loss: 0.08813843131065369 2023-01-23 14:16:19.440786: step: 988/1541, loss: 0.02114121802151203 2023-01-23 14:16:20.346429: step: 990/1541, loss: 0.008241694420576096 2023-01-23 14:16:21.221188: step: 992/1541, loss: 0.002513365587219596 2023-01-23 14:16:22.086389: step: 994/1541, loss: 0.0002571187214925885 2023-01-23 14:16:22.949635: step: 996/1541, loss: 0.0021500405855476856 2023-01-23 14:16:23.852194: step: 998/1541, loss: 0.12086987495422363 2023-01-23 14:16:24.754090: step: 1000/1541, loss: 0.12153440713882446 2023-01-23 14:16:25.641412: step: 1002/1541, loss: 0.27587977051734924 2023-01-23 14:16:26.552112: step: 1004/1541, loss: 0.09438435733318329 2023-01-23 14:16:27.442523: step: 1006/1541, loss: 0.003222208935767412 2023-01-23 14:16:28.303400: step: 1008/1541, loss: 0.006741050165146589 2023-01-23 14:16:29.223457: step: 1010/1541, loss: 0.04406218230724335 2023-01-23 14:16:30.253620: step: 1012/1541, loss: 0.05954353138804436 2023-01-23 14:16:31.133309: step: 1014/1541, loss: 0.19332067668437958 2023-01-23 14:16:32.043578: step: 1016/1541, loss: 0.5969523191452026 2023-01-23 14:16:32.988466: step: 1018/1541, loss: 0.2929092347621918 2023-01-23 14:16:33.877427: step: 1020/1541, loss: 0.008196281269192696 2023-01-23 14:16:34.759004: step: 1022/1541, loss: 0.020132852718234062 2023-01-23 14:16:35.657625: step: 1024/1541, loss: 0.004532395862042904 2023-01-23 14:16:36.539550: step: 1026/1541, loss: 0.3733829855918884 2023-01-23 14:16:37.433279: step: 1028/1541, loss: 0.015984471887350082 2023-01-23 14:16:38.306098: step: 1030/1541, loss: 0.04584013670682907 2023-01-23 14:16:39.213236: step: 1032/1541, loss: 0.0011983473086729646 2023-01-23 14:16:40.117238: step: 1034/1541, loss: 0.0005839393706992269 2023-01-23 14:16:41.008106: step: 1036/1541, loss: 0.006730641704052687 2023-01-23 14:16:41.852180: step: 1038/1541, loss: 0.01157945767045021 2023-01-23 14:16:42.726212: step: 1040/1541, loss: 0.03333570808172226 2023-01-23 14:16:43.602096: step: 1042/1541, loss: 0.0015294228214770555 2023-01-23 14:16:44.477803: step: 1044/1541, loss: 0.09025227278470993 2023-01-23 14:16:45.350779: step: 1046/1541, loss: 0.40132173895835876 2023-01-23 14:16:46.214329: step: 1048/1541, loss: 0.0008501226548105478 2023-01-23 14:16:47.114526: step: 1050/1541, loss: 0.6621274948120117 2023-01-23 14:16:48.053699: step: 1052/1541, loss: 0.07232652604579926 2023-01-23 14:16:49.039955: step: 1054/1541, loss: 0.029096899554133415 2023-01-23 14:16:49.914568: step: 1056/1541, loss: 0.007122037000954151 2023-01-23 14:16:50.815089: step: 1058/1541, loss: 0.45476675033569336 2023-01-23 14:16:51.713413: step: 1060/1541, loss: 0.012388303875923157 2023-01-23 14:16:52.658281: step: 1062/1541, loss: 0.38092881441116333 2023-01-23 14:16:53.494588: step: 1064/1541, loss: 0.012647122144699097 2023-01-23 14:16:54.371543: step: 1066/1541, loss: 0.019905511289834976 2023-01-23 14:16:55.262065: step: 1068/1541, loss: 0.10931843519210815 2023-01-23 14:16:56.183948: step: 1070/1541, loss: 0.6108524203300476 2023-01-23 14:16:57.084367: step: 1072/1541, loss: 0.014981322921812534 2023-01-23 14:16:57.966221: step: 1074/1541, loss: 0.019014911726117134 2023-01-23 14:16:58.832167: step: 1076/1541, loss: 0.29001638293266296 2023-01-23 14:16:59.718148: step: 1078/1541, loss: 0.755748987197876 2023-01-23 14:17:00.628847: step: 1080/1541, loss: 0.47417473793029785 2023-01-23 14:17:01.528677: step: 1082/1541, loss: 0.01807493530213833 2023-01-23 14:17:02.470719: step: 1084/1541, loss: 0.003946379292756319 2023-01-23 14:17:03.409708: step: 1086/1541, loss: 0.04692811146378517 2023-01-23 14:17:04.283196: step: 1088/1541, loss: 0.11078958958387375 2023-01-23 14:17:05.157323: step: 1090/1541, loss: 0.09366226941347122 2023-01-23 14:17:06.074732: step: 1092/1541, loss: 0.014413011260330677 2023-01-23 14:17:06.997688: step: 1094/1541, loss: 0.012067708186805248 2023-01-23 14:17:07.869360: step: 1096/1541, loss: 0.009462353773415089 2023-01-23 14:17:08.767913: step: 1098/1541, loss: 0.007778630591928959 2023-01-23 14:17:09.711948: step: 1100/1541, loss: 0.004586765076965094 2023-01-23 14:17:10.581637: step: 1102/1541, loss: 0.025105604901909828 2023-01-23 14:17:11.515103: step: 1104/1541, loss: 0.020970184355974197 2023-01-23 14:17:12.443186: step: 1106/1541, loss: 0.0555136576294899 2023-01-23 14:17:13.426459: step: 1108/1541, loss: 0.026193637400865555 2023-01-23 14:17:14.317128: step: 1110/1541, loss: 0.040528345853090286 2023-01-23 14:17:15.299196: step: 1112/1541, loss: 0.25318247079849243 2023-01-23 14:17:16.216173: step: 1114/1541, loss: 0.30124080181121826 2023-01-23 14:17:17.171831: step: 1116/1541, loss: 0.0005477939266711473 2023-01-23 14:17:18.079501: step: 1118/1541, loss: 0.012025855481624603 2023-01-23 14:17:18.964259: step: 1120/1541, loss: 0.0018586762016639113 2023-01-23 14:17:19.868450: step: 1122/1541, loss: 0.022025680169463158 2023-01-23 14:17:20.766625: step: 1124/1541, loss: 0.4061425030231476 2023-01-23 14:17:21.624898: step: 1126/1541, loss: 0.0015691595617681742 2023-01-23 14:17:22.510748: step: 1128/1541, loss: 0.3085462152957916 2023-01-23 14:17:23.432414: step: 1130/1541, loss: 0.04491104185581207 2023-01-23 14:17:24.346159: step: 1132/1541, loss: 0.02158481627702713 2023-01-23 14:17:25.235821: step: 1134/1541, loss: 0.2882329821586609 2023-01-23 14:17:26.134879: step: 1136/1541, loss: 0.06825193762779236 2023-01-23 14:17:27.063257: step: 1138/1541, loss: 0.12878385186195374 2023-01-23 14:17:27.978807: step: 1140/1541, loss: 0.0256502702832222 2023-01-23 14:17:28.846271: step: 1142/1541, loss: 0.016340650618076324 2023-01-23 14:17:29.736040: step: 1144/1541, loss: 0.0511750690639019 2023-01-23 14:17:30.663669: step: 1146/1541, loss: 0.28021368384361267 2023-01-23 14:17:31.547923: step: 1148/1541, loss: 0.02794484794139862 2023-01-23 14:17:32.422176: step: 1150/1541, loss: 0.12221322953701019 2023-01-23 14:17:33.294818: step: 1152/1541, loss: 0.013928892090916634 2023-01-23 14:17:34.190917: step: 1154/1541, loss: 0.10704264044761658 2023-01-23 14:17:35.107400: step: 1156/1541, loss: 0.012279298156499863 2023-01-23 14:17:35.982018: step: 1158/1541, loss: 0.07325281947851181 2023-01-23 14:17:36.884248: step: 1160/1541, loss: 0.03357774764299393 2023-01-23 14:17:37.728486: step: 1162/1541, loss: 0.1049022525548935 2023-01-23 14:17:38.588801: step: 1164/1541, loss: 0.021797873079776764 2023-01-23 14:17:39.462784: step: 1166/1541, loss: 0.0017447606660425663 2023-01-23 14:17:40.362646: step: 1168/1541, loss: 0.01759824901819229 2023-01-23 14:17:41.213417: step: 1170/1541, loss: 0.01275869831442833 2023-01-23 14:17:42.102279: step: 1172/1541, loss: 0.008875508792698383 2023-01-23 14:17:43.025049: step: 1174/1541, loss: 0.0006361842970363796 2023-01-23 14:17:43.913629: step: 1176/1541, loss: 0.14336851239204407 2023-01-23 14:17:44.828571: step: 1178/1541, loss: 0.025298409163951874 2023-01-23 14:17:45.690932: step: 1180/1541, loss: 0.0005302042118273675 2023-01-23 14:17:46.569501: step: 1182/1541, loss: 0.19197918474674225 2023-01-23 14:17:47.473156: step: 1184/1541, loss: 1.1253166198730469 2023-01-23 14:17:48.519153: step: 1186/1541, loss: 0.07363351434469223 2023-01-23 14:17:49.443035: step: 1188/1541, loss: 0.010372474789619446 2023-01-23 14:17:50.402665: step: 1190/1541, loss: 0.048280019313097 2023-01-23 14:17:51.308843: step: 1192/1541, loss: 0.02126612514257431 2023-01-23 14:17:52.187134: step: 1194/1541, loss: 0.02620311640202999 2023-01-23 14:17:53.078714: step: 1196/1541, loss: 0.008868546225130558 2023-01-23 14:17:53.992733: step: 1198/1541, loss: 0.04846220090985298 2023-01-23 14:17:54.904132: step: 1200/1541, loss: 0.08779220283031464 2023-01-23 14:17:55.798023: step: 1202/1541, loss: 0.13704514503479004 2023-01-23 14:17:56.669636: step: 1204/1541, loss: 0.031326331198215485 2023-01-23 14:17:57.547425: step: 1206/1541, loss: 0.00742282485589385 2023-01-23 14:17:58.409204: step: 1208/1541, loss: 0.5685746669769287 2023-01-23 14:17:59.310947: step: 1210/1541, loss: 0.0298236683011055 2023-01-23 14:18:00.221274: step: 1212/1541, loss: 0.0029425169341266155 2023-01-23 14:18:01.174700: step: 1214/1541, loss: 0.25519880652427673 2023-01-23 14:18:02.040780: step: 1216/1541, loss: 0.006268471945077181 2023-01-23 14:18:02.926380: step: 1218/1541, loss: 0.01279367133975029 2023-01-23 14:18:03.821774: step: 1220/1541, loss: 0.03443770110607147 2023-01-23 14:18:04.661524: step: 1222/1541, loss: 0.05745341256260872 2023-01-23 14:18:05.543318: step: 1224/1541, loss: 0.09681784361600876 2023-01-23 14:18:06.529012: step: 1226/1541, loss: 0.0026562195271253586 2023-01-23 14:18:07.420280: step: 1228/1541, loss: 0.022131336852908134 2023-01-23 14:18:08.327388: step: 1230/1541, loss: 0.21555033326148987 2023-01-23 14:18:09.250594: step: 1232/1541, loss: 0.0021634139120578766 2023-01-23 14:18:10.211551: step: 1234/1541, loss: 0.0077743190340697765 2023-01-23 14:18:11.092866: step: 1236/1541, loss: 0.04302595183253288 2023-01-23 14:18:12.009567: step: 1238/1541, loss: 0.008520055562257767 2023-01-23 14:18:12.918791: step: 1240/1541, loss: 0.009090008214116096 2023-01-23 14:18:13.868014: step: 1242/1541, loss: 0.028370987623929977 2023-01-23 14:18:14.784180: step: 1244/1541, loss: 0.06391964852809906 2023-01-23 14:18:15.672253: step: 1246/1541, loss: 0.0010837053414434195 2023-01-23 14:18:16.619523: step: 1248/1541, loss: 0.19441252946853638 2023-01-23 14:18:17.512696: step: 1250/1541, loss: 0.0153032336384058 2023-01-23 14:18:18.418185: step: 1252/1541, loss: 0.11953029036521912 2023-01-23 14:18:19.392615: step: 1254/1541, loss: 0.08663810789585114 2023-01-23 14:18:20.334374: step: 1256/1541, loss: 0.011132107116281986 2023-01-23 14:18:21.278240: step: 1258/1541, loss: 0.050053082406520844 2023-01-23 14:18:22.184744: step: 1260/1541, loss: 0.013173922896385193 2023-01-23 14:18:23.074362: step: 1262/1541, loss: 0.046237580478191376 2023-01-23 14:18:23.997027: step: 1264/1541, loss: 0.3192403018474579 2023-01-23 14:18:24.894458: step: 1266/1541, loss: 0.006275571417063475 2023-01-23 14:18:25.793532: step: 1268/1541, loss: 0.0015077944844961166 2023-01-23 14:18:26.708243: step: 1270/1541, loss: 0.118768110871315 2023-01-23 14:18:27.595034: step: 1272/1541, loss: 0.0002735867165029049 2023-01-23 14:18:28.497252: step: 1274/1541, loss: 0.0007673403597436845 2023-01-23 14:18:29.379509: step: 1276/1541, loss: 0.6181181073188782 2023-01-23 14:18:30.334383: step: 1278/1541, loss: 0.006849802099168301 2023-01-23 14:18:31.198574: step: 1280/1541, loss: 0.0021971925161778927 2023-01-23 14:18:32.123806: step: 1282/1541, loss: 0.0009230539435520768 2023-01-23 14:18:32.981423: step: 1284/1541, loss: 0.0018663655500859022 2023-01-23 14:18:33.866710: step: 1286/1541, loss: 0.048192914575338364 2023-01-23 14:18:34.795243: step: 1288/1541, loss: 0.11436200141906738 2023-01-23 14:18:35.686275: step: 1290/1541, loss: 0.025837907567620277 2023-01-23 14:18:36.568462: step: 1292/1541, loss: 0.24345709383487701 2023-01-23 14:18:37.451506: step: 1294/1541, loss: 0.11424266546964645 2023-01-23 14:18:38.318863: step: 1296/1541, loss: 0.0517127625644207 2023-01-23 14:18:39.206617: step: 1298/1541, loss: 0.0025140016805380583 2023-01-23 14:18:40.094867: step: 1300/1541, loss: 0.014665881171822548 2023-01-23 14:18:40.995740: step: 1302/1541, loss: 0.0023478653747588396 2023-01-23 14:18:41.960414: step: 1304/1541, loss: 0.41721588373184204 2023-01-23 14:18:42.936488: step: 1306/1541, loss: 0.2068191021680832 2023-01-23 14:18:43.836213: step: 1308/1541, loss: 0.004653909243643284 2023-01-23 14:18:44.732145: step: 1310/1541, loss: 0.11516449600458145 2023-01-23 14:18:45.691425: step: 1312/1541, loss: 0.01921744830906391 2023-01-23 14:18:46.599518: step: 1314/1541, loss: 0.19301101565361023 2023-01-23 14:18:47.542097: step: 1316/1541, loss: 0.0011604160536080599 2023-01-23 14:18:48.448271: step: 1318/1541, loss: 0.19503153860569 2023-01-23 14:18:49.318439: step: 1320/1541, loss: 0.062321197241544724 2023-01-23 14:18:50.223424: step: 1322/1541, loss: 0.008074413053691387 2023-01-23 14:18:51.100093: step: 1324/1541, loss: 0.35321784019470215 2023-01-23 14:18:52.008618: step: 1326/1541, loss: 0.3743601441383362 2023-01-23 14:18:52.872318: step: 1328/1541, loss: 0.00561209861189127 2023-01-23 14:18:53.780489: step: 1330/1541, loss: 2.323291778564453 2023-01-23 14:18:54.659184: step: 1332/1541, loss: 0.06193472072482109 2023-01-23 14:18:55.605295: step: 1334/1541, loss: 0.0017516661901026964 2023-01-23 14:18:56.476186: step: 1336/1541, loss: 0.07573892921209335 2023-01-23 14:18:57.399154: step: 1338/1541, loss: 0.026930684223771095 2023-01-23 14:18:58.361715: step: 1340/1541, loss: 0.03746692091226578 2023-01-23 14:18:59.284670: step: 1342/1541, loss: 0.18913942575454712 2023-01-23 14:19:00.194404: step: 1344/1541, loss: 0.42162081599235535 2023-01-23 14:19:01.082801: step: 1346/1541, loss: 0.005729974713176489 2023-01-23 14:19:01.960747: step: 1348/1541, loss: 0.0061131133697927 2023-01-23 14:19:02.821229: step: 1350/1541, loss: 0.008420757949352264 2023-01-23 14:19:03.705571: step: 1352/1541, loss: 0.1494671106338501 2023-01-23 14:19:04.551882: step: 1354/1541, loss: 0.0005491299089044333 2023-01-23 14:19:05.448611: step: 1356/1541, loss: 0.1985543668270111 2023-01-23 14:19:06.352319: step: 1358/1541, loss: 0.17447824776172638 2023-01-23 14:19:07.272605: step: 1360/1541, loss: 0.04574582725763321 2023-01-23 14:19:08.158958: step: 1362/1541, loss: 0.069280706346035 2023-01-23 14:19:09.193871: step: 1364/1541, loss: 0.019869161769747734 2023-01-23 14:19:10.048437: step: 1366/1541, loss: 0.03605639189481735 2023-01-23 14:19:10.967360: step: 1368/1541, loss: 0.0632053017616272 2023-01-23 14:19:11.886014: step: 1370/1541, loss: 0.07227849215269089 2023-01-23 14:19:12.787162: step: 1372/1541, loss: 0.14355240762233734 2023-01-23 14:19:13.691465: step: 1374/1541, loss: 0.03880409151315689 2023-01-23 14:19:14.627837: step: 1376/1541, loss: 0.06469662487506866 2023-01-23 14:19:15.515192: step: 1378/1541, loss: 0.01919056475162506 2023-01-23 14:19:16.416472: step: 1380/1541, loss: 0.023267369717359543 2023-01-23 14:19:17.319087: step: 1382/1541, loss: 0.5467102527618408 2023-01-23 14:19:18.218404: step: 1384/1541, loss: 0.15280835330486298 2023-01-23 14:19:19.103824: step: 1386/1541, loss: 0.0004634230281226337 2023-01-23 14:19:20.134118: step: 1388/1541, loss: 0.050664085894823074 2023-01-23 14:19:21.051245: step: 1390/1541, loss: 0.008000245317816734 2023-01-23 14:19:21.959006: step: 1392/1541, loss: 0.04804781451821327 2023-01-23 14:19:22.858859: step: 1394/1541, loss: 0.0051312679424881935 2023-01-23 14:19:23.755620: step: 1396/1541, loss: 0.033181969076395035 2023-01-23 14:19:24.635354: step: 1398/1541, loss: 0.031178269535303116 2023-01-23 14:19:25.535405: step: 1400/1541, loss: 0.020535703748464584 2023-01-23 14:19:26.421260: step: 1402/1541, loss: 0.018195828422904015 2023-01-23 14:19:27.388055: step: 1404/1541, loss: 0.08141764998435974 2023-01-23 14:19:28.269906: step: 1406/1541, loss: 0.5863285660743713 2023-01-23 14:19:29.163753: step: 1408/1541, loss: 0.4569101929664612 2023-01-23 14:19:30.091021: step: 1410/1541, loss: 0.01312613021582365 2023-01-23 14:19:30.946720: step: 1412/1541, loss: 0.44198131561279297 2023-01-23 14:19:31.893541: step: 1414/1541, loss: 0.23133417963981628 2023-01-23 14:19:32.775188: step: 1416/1541, loss: 0.0046235378831624985 2023-01-23 14:19:33.636867: step: 1418/1541, loss: 0.8822972774505615 2023-01-23 14:19:34.519823: step: 1420/1541, loss: 0.04205870255827904 2023-01-23 14:19:35.434662: step: 1422/1541, loss: 0.09406188130378723 2023-01-23 14:19:36.371287: step: 1424/1541, loss: 0.15424980223178864 2023-01-23 14:19:37.295307: step: 1426/1541, loss: 0.012523717246949673 2023-01-23 14:19:38.223928: step: 1428/1541, loss: 0.04638564586639404 2023-01-23 14:19:39.152130: step: 1430/1541, loss: 0.011355206370353699 2023-01-23 14:19:40.042345: step: 1432/1541, loss: 0.07410623133182526 2023-01-23 14:19:40.927383: step: 1434/1541, loss: 0.3762640058994293 2023-01-23 14:19:41.825121: step: 1436/1541, loss: 0.02678326703608036 2023-01-23 14:19:42.679990: step: 1438/1541, loss: 0.03622463345527649 2023-01-23 14:19:43.685584: step: 1440/1541, loss: 0.024898406118154526 2023-01-23 14:19:44.591287: step: 1442/1541, loss: 0.14828723669052124 2023-01-23 14:19:45.479952: step: 1444/1541, loss: 0.13687759637832642 2023-01-23 14:19:46.425764: step: 1446/1541, loss: 0.06130693107843399 2023-01-23 14:19:47.293950: step: 1448/1541, loss: 0.5129718780517578 2023-01-23 14:19:48.146716: step: 1450/1541, loss: 0.01715805009007454 2023-01-23 14:19:49.007109: step: 1452/1541, loss: 0.00184864632319659 2023-01-23 14:19:49.902293: step: 1454/1541, loss: 0.013861988671123981 2023-01-23 14:19:50.813293: step: 1456/1541, loss: 0.10524948686361313 2023-01-23 14:19:51.743226: step: 1458/1541, loss: 0.018043577671051025 2023-01-23 14:19:52.629471: step: 1460/1541, loss: 0.021152324974536896 2023-01-23 14:19:53.497826: step: 1462/1541, loss: 3.6124584674835205 2023-01-23 14:19:54.417310: step: 1464/1541, loss: 0.00419680867344141 2023-01-23 14:19:55.300094: step: 1466/1541, loss: 0.016683004796504974 2023-01-23 14:19:56.177250: step: 1468/1541, loss: 0.019590768963098526 2023-01-23 14:19:57.053565: step: 1470/1541, loss: 0.05386573448777199 2023-01-23 14:19:57.932498: step: 1472/1541, loss: 0.0021799891255795956 2023-01-23 14:19:58.787765: step: 1474/1541, loss: 0.005056781228631735 2023-01-23 14:19:59.691716: step: 1476/1541, loss: 0.04191350191831589 2023-01-23 14:20:00.628249: step: 1478/1541, loss: 0.1503850221633911 2023-01-23 14:20:01.568489: step: 1480/1541, loss: 0.03593091294169426 2023-01-23 14:20:02.510859: step: 1482/1541, loss: 0.11220979690551758 2023-01-23 14:20:03.431333: step: 1484/1541, loss: 0.26549944281578064 2023-01-23 14:20:04.317329: step: 1486/1541, loss: 0.048993419855833054 2023-01-23 14:20:05.221901: step: 1488/1541, loss: 0.14736397564411163 2023-01-23 14:20:06.090298: step: 1490/1541, loss: 0.15894567966461182 2023-01-23 14:20:07.012259: step: 1492/1541, loss: 0.23514892160892487 2023-01-23 14:20:07.900414: step: 1494/1541, loss: 0.3453029990196228 2023-01-23 14:20:08.802347: step: 1496/1541, loss: 0.05632419139146805 2023-01-23 14:20:09.751950: step: 1498/1541, loss: 0.2033897042274475 2023-01-23 14:20:10.741650: step: 1500/1541, loss: 0.005227889399975538 2023-01-23 14:20:11.679766: step: 1502/1541, loss: 0.008176030591130257 2023-01-23 14:20:12.605063: step: 1504/1541, loss: 0.023297572508454323 2023-01-23 14:20:13.508795: step: 1506/1541, loss: 0.015869271010160446 2023-01-23 14:20:14.418966: step: 1508/1541, loss: 0.0036181355826556683 2023-01-23 14:20:15.263533: step: 1510/1541, loss: 0.0021397722885012627 2023-01-23 14:20:16.204659: step: 1512/1541, loss: 0.006539541762322187 2023-01-23 14:20:17.092407: step: 1514/1541, loss: 0.42993587255477905 2023-01-23 14:20:17.951090: step: 1516/1541, loss: 0.0030749947763979435 2023-01-23 14:20:18.838579: step: 1518/1541, loss: 0.005233317613601685 2023-01-23 14:20:19.748372: step: 1520/1541, loss: 0.031825218349695206 2023-01-23 14:20:20.674495: step: 1522/1541, loss: 0.02511584758758545 2023-01-23 14:20:21.629362: step: 1524/1541, loss: 0.17047947645187378 2023-01-23 14:20:22.479838: step: 1526/1541, loss: 0.05319500342011452 2023-01-23 14:20:23.375387: step: 1528/1541, loss: 0.06736414134502411 2023-01-23 14:20:24.248851: step: 1530/1541, loss: 0.262554794549942 2023-01-23 14:20:25.119114: step: 1532/1541, loss: 0.0004109106957912445 2023-01-23 14:20:26.011382: step: 1534/1541, loss: 0.15902474522590637 2023-01-23 14:20:26.883457: step: 1536/1541, loss: 0.05377090722322464 2023-01-23 14:20:27.775092: step: 1538/1541, loss: 0.0009363923454657197 2023-01-23 14:20:28.675131: step: 1540/1541, loss: 0.09149353951215744 2023-01-23 14:20:29.621015: step: 1542/1541, loss: 0.013585257343947887 2023-01-23 14:20:30.566679: step: 1544/1541, loss: 0.06987312436103821 2023-01-23 14:20:31.479091: step: 1546/1541, loss: 0.18782861530780792 2023-01-23 14:20:32.383026: step: 1548/1541, loss: 0.11639590561389923 2023-01-23 14:20:33.325557: step: 1550/1541, loss: 0.11225780099630356 2023-01-23 14:20:34.192202: step: 1552/1541, loss: 0.12156464904546738 2023-01-23 14:20:35.136777: step: 1554/1541, loss: 0.10932756215333939 2023-01-23 14:20:36.069806: step: 1556/1541, loss: 0.27841663360595703 2023-01-23 14:20:36.933532: step: 1558/1541, loss: 0.6522138118743896 2023-01-23 14:20:37.890375: step: 1560/1541, loss: 0.2484341263771057 2023-01-23 14:20:38.875509: step: 1562/1541, loss: 0.2974283993244171 2023-01-23 14:20:39.776912: step: 1564/1541, loss: 0.10718736052513123 2023-01-23 14:20:40.673737: step: 1566/1541, loss: 0.004387273453176022 2023-01-23 14:20:41.559048: step: 1568/1541, loss: 0.003968379460275173 2023-01-23 14:20:42.460123: step: 1570/1541, loss: 0.19797031581401825 2023-01-23 14:20:43.384463: step: 1572/1541, loss: 0.010012602433562279 2023-01-23 14:20:44.334074: step: 1574/1541, loss: 0.020494185388088226 2023-01-23 14:20:45.249849: step: 1576/1541, loss: 0.09967431426048279 2023-01-23 14:20:46.114147: step: 1578/1541, loss: 0.07918280363082886 2023-01-23 14:20:47.041679: step: 1580/1541, loss: 0.1637406200170517 2023-01-23 14:20:47.940041: step: 1582/1541, loss: 0.12682849168777466 2023-01-23 14:20:48.870072: step: 1584/1541, loss: 0.004135312046855688 2023-01-23 14:20:49.809565: step: 1586/1541, loss: 0.04086877778172493 2023-01-23 14:20:50.796250: step: 1588/1541, loss: 0.08705979585647583 2023-01-23 14:20:51.700531: step: 1590/1541, loss: 0.03160206228494644 2023-01-23 14:20:52.601979: step: 1592/1541, loss: 0.06880120933055878 2023-01-23 14:20:53.468027: step: 1594/1541, loss: 0.48566052317619324 2023-01-23 14:20:54.307493: step: 1596/1541, loss: 0.09923281520605087 2023-01-23 14:20:55.204488: step: 1598/1541, loss: 0.059996508061885834 2023-01-23 14:20:56.081972: step: 1600/1541, loss: 0.1651666760444641 2023-01-23 14:20:57.017928: step: 1602/1541, loss: 0.0361478291451931 2023-01-23 14:20:57.945177: step: 1604/1541, loss: 0.0467313788831234 2023-01-23 14:20:58.824753: step: 1606/1541, loss: 0.15816786885261536 2023-01-23 14:20:59.692149: step: 1608/1541, loss: 0.015032934956252575 2023-01-23 14:21:00.580380: step: 1610/1541, loss: 0.8432632684707642 2023-01-23 14:21:01.497834: step: 1612/1541, loss: 0.040712982416152954 2023-01-23 14:21:02.373430: step: 1614/1541, loss: 0.05704854801297188 2023-01-23 14:21:03.280927: step: 1616/1541, loss: 0.008303694427013397 2023-01-23 14:21:04.218323: step: 1618/1541, loss: 0.13192008435726166 2023-01-23 14:21:05.072808: step: 1620/1541, loss: 0.007658757735043764 2023-01-23 14:21:06.046600: step: 1622/1541, loss: 0.3308708667755127 2023-01-23 14:21:06.939960: step: 1624/1541, loss: 0.019981976598501205 2023-01-23 14:21:07.859012: step: 1626/1541, loss: 0.0457623153924942 2023-01-23 14:21:08.770506: step: 1628/1541, loss: 0.11207490414381027 2023-01-23 14:21:09.649951: step: 1630/1541, loss: 0.14602549374103546 2023-01-23 14:21:10.524279: step: 1632/1541, loss: 0.3355695903301239 2023-01-23 14:21:11.418655: step: 1634/1541, loss: 0.1263478547334671 2023-01-23 14:21:12.286761: step: 1636/1541, loss: 0.013094332069158554 2023-01-23 14:21:13.149112: step: 1638/1541, loss: 0.06978926062583923 2023-01-23 14:21:14.047781: step: 1640/1541, loss: 0.001178173115476966 2023-01-23 14:21:14.926276: step: 1642/1541, loss: 0.07391324639320374 2023-01-23 14:21:15.790498: step: 1644/1541, loss: 0.021577954292297363 2023-01-23 14:21:16.676533: step: 1646/1541, loss: 0.07909654080867767 2023-01-23 14:21:17.563842: step: 1648/1541, loss: 0.021889863535761833 2023-01-23 14:21:18.456291: step: 1650/1541, loss: 0.015647299587726593 2023-01-23 14:21:19.391760: step: 1652/1541, loss: 0.009239456616342068 2023-01-23 14:21:20.274121: step: 1654/1541, loss: 0.14688000082969666 2023-01-23 14:21:21.147377: step: 1656/1541, loss: 0.0013084462843835354 2023-01-23 14:21:22.132440: step: 1658/1541, loss: 0.24056535959243774 2023-01-23 14:21:23.023808: step: 1660/1541, loss: 0.020661696791648865 2023-01-23 14:21:23.923015: step: 1662/1541, loss: 0.0018102673348039389 2023-01-23 14:21:24.840050: step: 1664/1541, loss: 0.05021614953875542 2023-01-23 14:21:25.725358: step: 1666/1541, loss: 0.43399927020072937 2023-01-23 14:21:26.700214: step: 1668/1541, loss: 0.004733896814286709 2023-01-23 14:21:27.611583: step: 1670/1541, loss: 0.13317465782165527 2023-01-23 14:21:28.527685: step: 1672/1541, loss: 0.003070527222007513 2023-01-23 14:21:29.450610: step: 1674/1541, loss: 0.21389912068843842 2023-01-23 14:21:30.330078: step: 1676/1541, loss: 0.01063494011759758 2023-01-23 14:21:31.238560: step: 1678/1541, loss: 0.0008101218263618648 2023-01-23 14:21:32.203558: step: 1680/1541, loss: 0.03985840827226639 2023-01-23 14:21:33.091382: step: 1682/1541, loss: 0.13706058263778687 2023-01-23 14:21:34.079280: step: 1684/1541, loss: 0.006808031816035509 2023-01-23 14:21:34.955254: step: 1686/1541, loss: 0.0175581406801939 2023-01-23 14:21:35.854718: step: 1688/1541, loss: 0.1701197475194931 2023-01-23 14:21:36.735947: step: 1690/1541, loss: 0.2308453917503357 2023-01-23 14:21:37.622146: step: 1692/1541, loss: 0.0027110595256090164 2023-01-23 14:21:38.555959: step: 1694/1541, loss: 0.05079560726881027 2023-01-23 14:21:39.422708: step: 1696/1541, loss: 0.001307911123149097 2023-01-23 14:21:40.341155: step: 1698/1541, loss: 0.08937328308820724 2023-01-23 14:21:41.262355: step: 1700/1541, loss: 0.15605071187019348 2023-01-23 14:21:42.152601: step: 1702/1541, loss: 0.08085521310567856 2023-01-23 14:21:43.029737: step: 1704/1541, loss: 0.005810217000544071 2023-01-23 14:21:43.900046: step: 1706/1541, loss: 0.0546431839466095 2023-01-23 14:21:44.840463: step: 1708/1541, loss: 0.0030337651260197163 2023-01-23 14:21:45.724936: step: 1710/1541, loss: 0.00123029004316777 2023-01-23 14:21:46.660167: step: 1712/1541, loss: 0.00428441958501935 2023-01-23 14:21:47.560880: step: 1714/1541, loss: 0.06120012328028679 2023-01-23 14:21:48.469024: step: 1716/1541, loss: 0.007646612823009491 2023-01-23 14:21:49.366520: step: 1718/1541, loss: 0.03661838918924332 2023-01-23 14:21:50.229053: step: 1720/1541, loss: 0.004995575174689293 2023-01-23 14:21:51.174184: step: 1722/1541, loss: 0.011784611269831657 2023-01-23 14:21:52.070174: step: 1724/1541, loss: 0.265352338552475 2023-01-23 14:21:52.963833: step: 1726/1541, loss: 0.171234130859375 2023-01-23 14:21:53.811811: step: 1728/1541, loss: 0.02400541491806507 2023-01-23 14:21:54.717836: step: 1730/1541, loss: 0.005000611767172813 2023-01-23 14:21:55.572454: step: 1732/1541, loss: 0.07392661273479462 2023-01-23 14:21:56.426300: step: 1734/1541, loss: 0.0295913927257061 2023-01-23 14:21:57.359476: step: 1736/1541, loss: 0.000568366376683116 2023-01-23 14:21:58.273976: step: 1738/1541, loss: 0.001271244022063911 2023-01-23 14:21:59.153865: step: 1740/1541, loss: 0.021678496152162552 2023-01-23 14:21:59.997939: step: 1742/1541, loss: 0.0019344217143952847 2023-01-23 14:22:00.887154: step: 1744/1541, loss: 0.028435368090867996 2023-01-23 14:22:01.774149: step: 1746/1541, loss: 1.4033628702163696 2023-01-23 14:22:02.643446: step: 1748/1541, loss: 0.02575252763926983 2023-01-23 14:22:03.529345: step: 1750/1541, loss: 0.22965718805789948 2023-01-23 14:22:04.444792: step: 1752/1541, loss: 0.01490727812051773 2023-01-23 14:22:05.373893: step: 1754/1541, loss: 0.0022478573955595493 2023-01-23 14:22:06.258639: step: 1756/1541, loss: 0.005985175259411335 2023-01-23 14:22:07.199532: step: 1758/1541, loss: 0.0019811864476650953 2023-01-23 14:22:08.063890: step: 1760/1541, loss: 0.034791722893714905 2023-01-23 14:22:08.936870: step: 1762/1541, loss: 0.004928275011479855 2023-01-23 14:22:09.827757: step: 1764/1541, loss: 0.034033577889204025 2023-01-23 14:22:10.719228: step: 1766/1541, loss: 0.10037658363580704 2023-01-23 14:22:11.642837: step: 1768/1541, loss: 0.11752455681562424 2023-01-23 14:22:12.532917: step: 1770/1541, loss: 0.06264100968837738 2023-01-23 14:22:13.459152: step: 1772/1541, loss: 0.031957197934389114 2023-01-23 14:22:14.380380: step: 1774/1541, loss: 0.07580579817295074 2023-01-23 14:22:15.261520: step: 1776/1541, loss: 0.0063719903118908405 2023-01-23 14:22:16.179725: step: 1778/1541, loss: 0.08074221760034561 2023-01-23 14:22:17.096428: step: 1780/1541, loss: 0.04435533285140991 2023-01-23 14:22:18.014324: step: 1782/1541, loss: 0.0010778391733765602 2023-01-23 14:22:18.850402: step: 1784/1541, loss: 0.06281670182943344 2023-01-23 14:22:19.714499: step: 1786/1541, loss: 0.02874073013663292 2023-01-23 14:22:20.570343: step: 1788/1541, loss: 0.02559143677353859 2023-01-23 14:22:21.470925: step: 1790/1541, loss: 0.2351444661617279 2023-01-23 14:22:22.352348: step: 1792/1541, loss: 0.017387352883815765 2023-01-23 14:22:23.256592: step: 1794/1541, loss: 0.0030650999397039413 2023-01-23 14:22:24.154174: step: 1796/1541, loss: 0.27888965606689453 2023-01-23 14:22:25.073904: step: 1798/1541, loss: 0.02373666875064373 2023-01-23 14:22:25.991609: step: 1800/1541, loss: 0.26354220509529114 2023-01-23 14:22:26.948514: step: 1802/1541, loss: 0.24893833696842194 2023-01-23 14:22:27.846399: step: 1804/1541, loss: 0.03017820045351982 2023-01-23 14:22:28.747455: step: 1806/1541, loss: 0.004221450071781874 2023-01-23 14:22:29.632291: step: 1808/1541, loss: 0.0006364500150084496 2023-01-23 14:22:30.504388: step: 1810/1541, loss: 0.021938815712928772 2023-01-23 14:22:31.399570: step: 1812/1541, loss: 0.3425673246383667 2023-01-23 14:22:32.306797: step: 1814/1541, loss: 1.7284492254257202 2023-01-23 14:22:33.218131: step: 1816/1541, loss: 0.007449397351592779 2023-01-23 14:22:34.151155: step: 1818/1541, loss: 0.17995955049991608 2023-01-23 14:22:35.034807: step: 1820/1541, loss: 0.009912114590406418 2023-01-23 14:22:35.903327: step: 1822/1541, loss: 0.00369925401173532 2023-01-23 14:22:36.803188: step: 1824/1541, loss: 0.3765684962272644 2023-01-23 14:22:37.692050: step: 1826/1541, loss: 0.13370825350284576 2023-01-23 14:22:38.600270: step: 1828/1541, loss: 0.08376535028219223 2023-01-23 14:22:39.488486: step: 1830/1541, loss: 0.0014592106454074383 2023-01-23 14:22:40.401686: step: 1832/1541, loss: 0.27273091673851013 2023-01-23 14:22:41.358063: step: 1834/1541, loss: 0.013365240767598152 2023-01-23 14:22:42.258320: step: 1836/1541, loss: 1.881858229637146 2023-01-23 14:22:43.142053: step: 1838/1541, loss: 0.07208792865276337 2023-01-23 14:22:44.039955: step: 1840/1541, loss: 0.007995463907718658 2023-01-23 14:22:44.929494: step: 1842/1541, loss: 0.22695784270763397 2023-01-23 14:22:45.821193: step: 1844/1541, loss: 0.001652261009439826 2023-01-23 14:22:46.712401: step: 1846/1541, loss: 0.105140820145607 2023-01-23 14:22:47.691608: step: 1848/1541, loss: 0.005915215238928795 2023-01-23 14:22:48.621960: step: 1850/1541, loss: 0.0971696749329567 2023-01-23 14:22:49.531032: step: 1852/1541, loss: 0.06990358978509903 2023-01-23 14:22:50.413994: step: 1854/1541, loss: 0.00042506950558163226 2023-01-23 14:22:51.311766: step: 1856/1541, loss: 0.0014169936766847968 2023-01-23 14:22:52.209038: step: 1858/1541, loss: 0.02060423232614994 2023-01-23 14:22:53.072112: step: 1860/1541, loss: 0.006403124891221523 2023-01-23 14:22:54.069021: step: 1862/1541, loss: 0.0640178993344307 2023-01-23 14:22:54.992734: step: 1864/1541, loss: 0.30558398365974426 2023-01-23 14:22:55.889555: step: 1866/1541, loss: 0.02932848408818245 2023-01-23 14:22:56.755551: step: 1868/1541, loss: 0.17072248458862305 2023-01-23 14:22:57.677121: step: 1870/1541, loss: 0.004840912763029337 2023-01-23 14:22:58.614553: step: 1872/1541, loss: 0.004175083711743355 2023-01-23 14:22:59.513739: step: 1874/1541, loss: 0.010952619835734367 2023-01-23 14:23:00.435871: step: 1876/1541, loss: 0.042593102902173996 2023-01-23 14:23:01.294343: step: 1878/1541, loss: 0.4800741970539093 2023-01-23 14:23:02.298565: step: 1880/1541, loss: 0.001135572325438261 2023-01-23 14:23:03.150424: step: 1882/1541, loss: 0.47496145963668823 2023-01-23 14:23:04.063902: step: 1884/1541, loss: 0.0025380714796483517 2023-01-23 14:23:04.937425: step: 1886/1541, loss: 0.20941095054149628 2023-01-23 14:23:05.798547: step: 1888/1541, loss: 0.00261903484351933 2023-01-23 14:23:06.668774: step: 1890/1541, loss: 0.14917884767055511 2023-01-23 14:23:07.545894: step: 1892/1541, loss: 0.18805061280727386 2023-01-23 14:23:08.458802: step: 1894/1541, loss: 0.12631995975971222 2023-01-23 14:23:09.345112: step: 1896/1541, loss: 0.011319255456328392 2023-01-23 14:23:10.256747: step: 1898/1541, loss: 0.020291034132242203 2023-01-23 14:23:11.157466: step: 1900/1541, loss: 0.002878590952605009 2023-01-23 14:23:12.052301: step: 1902/1541, loss: 0.027341129258275032 2023-01-23 14:23:12.988663: step: 1904/1541, loss: 0.057356953620910645 2023-01-23 14:23:13.927608: step: 1906/1541, loss: 0.011452188715338707 2023-01-23 14:23:14.809722: step: 1908/1541, loss: 0.08510810881853104 2023-01-23 14:23:15.726861: step: 1910/1541, loss: 0.1083402931690216 2023-01-23 14:23:16.600025: step: 1912/1541, loss: 0.03120722435414791 2023-01-23 14:23:17.510168: step: 1914/1541, loss: 0.050172336399555206 2023-01-23 14:23:18.382720: step: 1916/1541, loss: 0.012515212409198284 2023-01-23 14:23:19.263130: step: 1918/1541, loss: 0.017262397333979607 2023-01-23 14:23:20.134542: step: 1920/1541, loss: 0.16212379932403564 2023-01-23 14:23:21.007958: step: 1922/1541, loss: 0.004750330001115799 2023-01-23 14:23:21.887166: step: 1924/1541, loss: 0.0036884532310068607 2023-01-23 14:23:22.762376: step: 1926/1541, loss: 0.004247688222676516 2023-01-23 14:23:23.658391: step: 1928/1541, loss: 0.05049682408571243 2023-01-23 14:23:24.537292: step: 1930/1541, loss: 0.04153824597597122 2023-01-23 14:23:25.396927: step: 1932/1541, loss: 0.008065460249781609 2023-01-23 14:23:26.279044: step: 1934/1541, loss: 0.030924435704946518 2023-01-23 14:23:27.122010: step: 1936/1541, loss: 0.08672984689474106 2023-01-23 14:23:27.992332: step: 1938/1541, loss: 0.0016581236850470304 2023-01-23 14:23:28.914392: step: 1940/1541, loss: 0.04879210144281387 2023-01-23 14:23:29.801424: step: 1942/1541, loss: 0.006098780781030655 2023-01-23 14:23:30.697699: step: 1944/1541, loss: 0.03284892439842224 2023-01-23 14:23:31.590165: step: 1946/1541, loss: 0.030782828107476234 2023-01-23 14:23:32.481264: step: 1948/1541, loss: 0.0010794226545840502 2023-01-23 14:23:33.385924: step: 1950/1541, loss: 0.07849927991628647 2023-01-23 14:23:34.266523: step: 1952/1541, loss: 0.0342511311173439 2023-01-23 14:23:35.208219: step: 1954/1541, loss: 0.12594348192214966 2023-01-23 14:23:36.077973: step: 1956/1541, loss: 0.1889326125383377 2023-01-23 14:23:36.953517: step: 1958/1541, loss: 0.18172189593315125 2023-01-23 14:23:37.838908: step: 1960/1541, loss: 0.36002227663993835 2023-01-23 14:23:38.706476: step: 1962/1541, loss: 0.03152956813573837 2023-01-23 14:23:39.604276: step: 1964/1541, loss: 0.2008645087480545 2023-01-23 14:23:40.459432: step: 1966/1541, loss: 0.0019415982533246279 2023-01-23 14:23:41.367364: step: 1968/1541, loss: 0.3844973146915436 2023-01-23 14:23:42.276225: step: 1970/1541, loss: 0.10205645859241486 2023-01-23 14:23:43.162218: step: 1972/1541, loss: 0.0006301969988271594 2023-01-23 14:23:44.039506: step: 1974/1541, loss: 0.04298808053135872 2023-01-23 14:23:44.931546: step: 1976/1541, loss: 0.0355856753885746 2023-01-23 14:23:45.818952: step: 1978/1541, loss: 0.45164555311203003 2023-01-23 14:23:46.701543: step: 1980/1541, loss: 0.0024523467291146517 2023-01-23 14:23:47.590601: step: 1982/1541, loss: 0.37700286507606506 2023-01-23 14:23:48.482348: step: 1984/1541, loss: 0.3347828686237335 2023-01-23 14:23:49.397309: step: 1986/1541, loss: 0.034524012356996536 2023-01-23 14:23:50.298343: step: 1988/1541, loss: 0.0015190071426331997 2023-01-23 14:23:51.174369: step: 1990/1541, loss: 0.22841642796993256 2023-01-23 14:23:52.069273: step: 1992/1541, loss: 0.19236116111278534 2023-01-23 14:23:53.069436: step: 1994/1541, loss: 0.09129824489355087 2023-01-23 14:23:54.010621: step: 1996/1541, loss: 0.008415047079324722 2023-01-23 14:23:54.929709: step: 1998/1541, loss: 0.01048329845070839 2023-01-23 14:23:55.818412: step: 2000/1541, loss: 0.0004069555434398353 2023-01-23 14:23:56.723122: step: 2002/1541, loss: 0.07250712066888809 2023-01-23 14:23:57.642086: step: 2004/1541, loss: 0.06921879947185516 2023-01-23 14:23:58.552167: step: 2006/1541, loss: 0.10023558139801025 2023-01-23 14:23:59.423015: step: 2008/1541, loss: 0.005193215794861317 2023-01-23 14:24:00.365763: step: 2010/1541, loss: 0.01647339016199112 2023-01-23 14:24:01.281730: step: 2012/1541, loss: 0.0015253014862537384 2023-01-23 14:24:02.220741: step: 2014/1541, loss: 0.06894304603338242 2023-01-23 14:24:03.111236: step: 2016/1541, loss: 0.08231233805418015 2023-01-23 14:24:04.067209: step: 2018/1541, loss: 0.0498834028840065 2023-01-23 14:24:05.120122: step: 2020/1541, loss: 0.09286296367645264 2023-01-23 14:24:06.101569: step: 2022/1541, loss: 0.0030468180775642395 2023-01-23 14:24:07.036092: step: 2024/1541, loss: 0.005813459865748882 2023-01-23 14:24:07.896037: step: 2026/1541, loss: 0.052386969327926636 2023-01-23 14:24:08.816417: step: 2028/1541, loss: 0.018562326207756996 2023-01-23 14:24:09.707664: step: 2030/1541, loss: 0.6433858275413513 2023-01-23 14:24:10.603954: step: 2032/1541, loss: 0.007996853440999985 2023-01-23 14:24:11.517683: step: 2034/1541, loss: 0.07579246163368225 2023-01-23 14:24:12.438032: step: 2036/1541, loss: 0.28362366557121277 2023-01-23 14:24:13.321152: step: 2038/1541, loss: 0.011307073757052422 2023-01-23 14:24:14.244230: step: 2040/1541, loss: 0.004399225115776062 2023-01-23 14:24:15.177952: step: 2042/1541, loss: 0.0013234452344477177 2023-01-23 14:24:16.111696: step: 2044/1541, loss: 0.02784702368080616 2023-01-23 14:24:17.007183: step: 2046/1541, loss: 0.011847375892102718 2023-01-23 14:24:17.952547: step: 2048/1541, loss: 0.05455063283443451 2023-01-23 14:24:18.865074: step: 2050/1541, loss: 0.005330177955329418 2023-01-23 14:24:19.759539: step: 2052/1541, loss: 0.05483957380056381 2023-01-23 14:24:20.644222: step: 2054/1541, loss: 0.5957295298576355 2023-01-23 14:24:21.551354: step: 2056/1541, loss: 0.18058118224143982 2023-01-23 14:24:22.451994: step: 2058/1541, loss: 0.008982889354228973 2023-01-23 14:24:23.341543: step: 2060/1541, loss: 0.0006638760678470135 2023-01-23 14:24:24.198594: step: 2062/1541, loss: 0.0017741926712915301 2023-01-23 14:24:25.153684: step: 2064/1541, loss: 0.05737312510609627 2023-01-23 14:24:26.101859: step: 2066/1541, loss: 0.19648191332817078 2023-01-23 14:24:27.009695: step: 2068/1541, loss: 0.1974198967218399 2023-01-23 14:24:27.901803: step: 2070/1541, loss: 0.0029299871530383825 2023-01-23 14:24:28.902976: step: 2072/1541, loss: 0.037837665528059006 2023-01-23 14:24:29.816902: step: 2074/1541, loss: 0.0046365694142878056 2023-01-23 14:24:30.719361: step: 2076/1541, loss: 0.0016544453101232648 2023-01-23 14:24:31.579126: step: 2078/1541, loss: 0.013285079039633274 2023-01-23 14:24:32.475428: step: 2080/1541, loss: 0.04823166877031326 2023-01-23 14:24:33.363300: step: 2082/1541, loss: 0.48173046112060547 2023-01-23 14:24:34.256608: step: 2084/1541, loss: 0.0028084469959139824 2023-01-23 14:24:35.167436: step: 2086/1541, loss: 0.0017582932487130165 2023-01-23 14:24:36.077483: step: 2088/1541, loss: 0.1557607203722 2023-01-23 14:24:36.937790: step: 2090/1541, loss: 0.0029760152101516724 2023-01-23 14:24:37.800565: step: 2092/1541, loss: 0.2504844069480896 2023-01-23 14:24:38.696906: step: 2094/1541, loss: 0.18229828774929047 2023-01-23 14:24:39.572592: step: 2096/1541, loss: 0.14767718315124512 2023-01-23 14:24:40.444549: step: 2098/1541, loss: 0.018741199746727943 2023-01-23 14:24:41.332331: step: 2100/1541, loss: 0.06747641414403915 2023-01-23 14:24:42.335099: step: 2102/1541, loss: 0.151476189494133 2023-01-23 14:24:43.202314: step: 2104/1541, loss: 0.03296038880944252 2023-01-23 14:24:44.097079: step: 2106/1541, loss: 0.02179262414574623 2023-01-23 14:24:45.007468: step: 2108/1541, loss: 0.00401750672608614 2023-01-23 14:24:45.873822: step: 2110/1541, loss: 0.1145562082529068 2023-01-23 14:24:46.782077: step: 2112/1541, loss: 0.34701600670814514 2023-01-23 14:24:47.664287: step: 2114/1541, loss: 0.04381339251995087 2023-01-23 14:24:48.561388: step: 2116/1541, loss: 0.1380159556865692 2023-01-23 14:24:49.453518: step: 2118/1541, loss: 0.021390525624155998 2023-01-23 14:24:50.348420: step: 2120/1541, loss: 0.06675220280885696 2023-01-23 14:24:51.257923: step: 2122/1541, loss: 0.027514539659023285 2023-01-23 14:24:52.172921: step: 2124/1541, loss: 0.03010205738246441 2023-01-23 14:24:53.064888: step: 2126/1541, loss: 0.031142784282565117 2023-01-23 14:24:54.007114: step: 2128/1541, loss: 0.0006297917570918798 2023-01-23 14:24:54.888450: step: 2130/1541, loss: 0.1562158167362213 2023-01-23 14:24:55.768204: step: 2132/1541, loss: 0.013581840321421623 2023-01-23 14:24:56.741830: step: 2134/1541, loss: 0.16527341306209564 2023-01-23 14:24:57.631963: step: 2136/1541, loss: 0.0009121389593929052 2023-01-23 14:24:58.512796: step: 2138/1541, loss: 0.016370993107557297 2023-01-23 14:24:59.392040: step: 2140/1541, loss: 0.008286595344543457 2023-01-23 14:25:00.268742: step: 2142/1541, loss: 0.17231200635433197 2023-01-23 14:25:01.245309: step: 2144/1541, loss: 0.04879073426127434 2023-01-23 14:25:02.199356: step: 2146/1541, loss: 0.11134877800941467 2023-01-23 14:25:03.128660: step: 2148/1541, loss: 0.0009587676031515002 2023-01-23 14:25:03.991521: step: 2150/1541, loss: 0.007030772510915995 2023-01-23 14:25:04.934723: step: 2152/1541, loss: 0.003647079225629568 2023-01-23 14:25:05.813070: step: 2154/1541, loss: 0.07776331156492233 2023-01-23 14:25:06.750365: step: 2156/1541, loss: 0.16099956631660461 2023-01-23 14:25:07.634778: step: 2158/1541, loss: 0.023251159116625786 2023-01-23 14:25:08.535659: step: 2160/1541, loss: 0.06473288685083389 2023-01-23 14:25:09.399756: step: 2162/1541, loss: 0.013699292205274105 2023-01-23 14:25:10.306333: step: 2164/1541, loss: 0.032607149332761765 2023-01-23 14:25:11.201008: step: 2166/1541, loss: 0.004091009497642517 2023-01-23 14:25:12.056381: step: 2168/1541, loss: 0.0016689753392711282 2023-01-23 14:25:12.947265: step: 2170/1541, loss: 0.076145701110363 2023-01-23 14:25:13.899824: step: 2172/1541, loss: 0.04140029475092888 2023-01-23 14:25:14.788881: step: 2174/1541, loss: 0.0029800795018672943 2023-01-23 14:25:15.739657: step: 2176/1541, loss: 0.04538391903042793 2023-01-23 14:25:16.639884: step: 2178/1541, loss: 0.02027961239218712 2023-01-23 14:25:17.498934: step: 2180/1541, loss: 0.00080846156924963 2023-01-23 14:25:18.409004: step: 2182/1541, loss: 0.4916696548461914 2023-01-23 14:25:19.305206: step: 2184/1541, loss: 0.02053872123360634 2023-01-23 14:25:20.228136: step: 2186/1541, loss: 0.00855330377817154 2023-01-23 14:25:21.189672: step: 2188/1541, loss: 0.000713894609361887 2023-01-23 14:25:22.062910: step: 2190/1541, loss: 0.026697320863604546 2023-01-23 14:25:22.982708: step: 2192/1541, loss: 0.13953101634979248 2023-01-23 14:25:23.897949: step: 2194/1541, loss: 0.08838559687137604 2023-01-23 14:25:24.836180: step: 2196/1541, loss: 0.07540837675333023 2023-01-23 14:25:25.758173: step: 2198/1541, loss: 0.026495609432458878 2023-01-23 14:25:26.751456: step: 2200/1541, loss: 0.0076150791719555855 2023-01-23 14:25:27.673665: step: 2202/1541, loss: 0.16246160864830017 2023-01-23 14:25:28.686124: step: 2204/1541, loss: 0.05375196784734726 2023-01-23 14:25:29.627113: step: 2206/1541, loss: 0.02665003202855587 2023-01-23 14:25:30.526295: step: 2208/1541, loss: 0.0011306499363854527 2023-01-23 14:25:31.391599: step: 2210/1541, loss: 0.04233214259147644 2023-01-23 14:25:32.285655: step: 2212/1541, loss: 0.19236494600772858 2023-01-23 14:25:33.197923: step: 2214/1541, loss: 0.35484910011291504 2023-01-23 14:25:34.136724: step: 2216/1541, loss: 0.006451777648180723 2023-01-23 14:25:35.053598: step: 2218/1541, loss: 0.2563555836677551 2023-01-23 14:25:35.979632: step: 2220/1541, loss: 0.07187531143426895 2023-01-23 14:25:36.877670: step: 2222/1541, loss: 0.019874002784490585 2023-01-23 14:25:37.782684: step: 2224/1541, loss: 0.06617342680692673 2023-01-23 14:25:38.672164: step: 2226/1541, loss: 0.05384184420108795 2023-01-23 14:25:39.559472: step: 2228/1541, loss: 0.14059334993362427 2023-01-23 14:25:40.437124: step: 2230/1541, loss: 0.017174633219838142 2023-01-23 14:25:41.411721: step: 2232/1541, loss: 0.02596936747431755 2023-01-23 14:25:42.317701: step: 2234/1541, loss: 0.027286900207400322 2023-01-23 14:25:43.227234: step: 2236/1541, loss: 0.5826295018196106 2023-01-23 14:25:44.126989: step: 2238/1541, loss: 0.024142343550920486 2023-01-23 14:25:45.076065: step: 2240/1541, loss: 0.007413071114569902 2023-01-23 14:25:45.936657: step: 2242/1541, loss: 0.08758633583784103 2023-01-23 14:25:46.822532: step: 2244/1541, loss: 0.004866606555879116 2023-01-23 14:25:47.677546: step: 2246/1541, loss: 0.23651696741580963 2023-01-23 14:25:48.594949: step: 2248/1541, loss: 0.36757102608680725 2023-01-23 14:25:49.562510: step: 2250/1541, loss: 0.2739916741847992 2023-01-23 14:25:50.520089: step: 2252/1541, loss: 0.05545209348201752 2023-01-23 14:25:51.426275: step: 2254/1541, loss: 0.40400585532188416 2023-01-23 14:25:52.333753: step: 2256/1541, loss: 0.04348975419998169 2023-01-23 14:25:53.230353: step: 2258/1541, loss: 0.0018401766428723931 2023-01-23 14:25:54.136967: step: 2260/1541, loss: 0.0009497231803834438 2023-01-23 14:25:54.998157: step: 2262/1541, loss: 0.19416841864585876 2023-01-23 14:25:55.907476: step: 2264/1541, loss: 0.10482808947563171 2023-01-23 14:25:56.812854: step: 2266/1541, loss: 0.5251117944717407 2023-01-23 14:25:57.696094: step: 2268/1541, loss: 0.2504787743091583 2023-01-23 14:25:58.618917: step: 2270/1541, loss: 0.006204793229699135 2023-01-23 14:25:59.506774: step: 2272/1541, loss: 0.04343700036406517 2023-01-23 14:26:00.377826: step: 2274/1541, loss: 0.12626396119594574 2023-01-23 14:26:01.255612: step: 2276/1541, loss: 0.22570393979549408 2023-01-23 14:26:02.176611: step: 2278/1541, loss: 0.003418696578592062 2023-01-23 14:26:03.040288: step: 2280/1541, loss: 0.0030911213252693415 2023-01-23 14:26:03.968619: step: 2282/1541, loss: 0.03322022035717964 2023-01-23 14:26:04.902303: step: 2284/1541, loss: 0.025804422795772552 2023-01-23 14:26:05.791835: step: 2286/1541, loss: 0.30016863346099854 2023-01-23 14:26:06.643689: step: 2288/1541, loss: 0.2180774211883545 2023-01-23 14:26:07.535671: step: 2290/1541, loss: 0.21169671416282654 2023-01-23 14:26:08.404903: step: 2292/1541, loss: 0.24650631844997406 2023-01-23 14:26:09.316180: step: 2294/1541, loss: 0.0027993349358439445 2023-01-23 14:26:10.248301: step: 2296/1541, loss: 0.01790800504386425 2023-01-23 14:26:11.126272: step: 2298/1541, loss: 0.08899195492267609 2023-01-23 14:26:12.019877: step: 2300/1541, loss: 0.032773979008197784 2023-01-23 14:26:13.006915: step: 2302/1541, loss: 0.13624675571918488 2023-01-23 14:26:13.984712: step: 2304/1541, loss: 0.12974895536899567 2023-01-23 14:26:14.887826: step: 2306/1541, loss: 0.004826443735510111 2023-01-23 14:26:15.742841: step: 2308/1541, loss: 0.08538442850112915 2023-01-23 14:26:16.632878: step: 2310/1541, loss: 0.10760234296321869 2023-01-23 14:26:17.494930: step: 2312/1541, loss: 0.06462180614471436 2023-01-23 14:26:18.400807: step: 2314/1541, loss: 0.22678811848163605 2023-01-23 14:26:19.316304: step: 2316/1541, loss: 0.016485445201396942 2023-01-23 14:26:20.263955: step: 2318/1541, loss: 0.02409505844116211 2023-01-23 14:26:21.204820: step: 2320/1541, loss: 0.1593443602323532 2023-01-23 14:26:22.112876: step: 2322/1541, loss: 0.014127500355243683 2023-01-23 14:26:23.120334: step: 2324/1541, loss: 0.20403112471103668 2023-01-23 14:26:23.975913: step: 2326/1541, loss: 0.0006603110814467072 2023-01-23 14:26:24.840065: step: 2328/1541, loss: 0.14496129751205444 2023-01-23 14:26:25.754886: step: 2330/1541, loss: 0.019757570698857307 2023-01-23 14:26:26.657271: step: 2332/1541, loss: 0.007907679304480553 2023-01-23 14:26:27.524923: step: 2334/1541, loss: 0.005479884799569845 2023-01-23 14:26:28.448888: step: 2336/1541, loss: 0.006333393044769764 2023-01-23 14:26:29.365158: step: 2338/1541, loss: 0.04361199587583542 2023-01-23 14:26:30.292845: step: 2340/1541, loss: 0.33440735936164856 2023-01-23 14:26:31.206721: step: 2342/1541, loss: 0.20094460248947144 2023-01-23 14:26:32.141842: step: 2344/1541, loss: 0.0061956183053553104 2023-01-23 14:26:33.022642: step: 2346/1541, loss: 0.005690207239240408 2023-01-23 14:26:33.877426: step: 2348/1541, loss: 0.0009964120108634233 2023-01-23 14:26:34.830935: step: 2350/1541, loss: 0.17210344970226288 2023-01-23 14:26:35.714298: step: 2352/1541, loss: 0.15461453795433044 2023-01-23 14:26:36.532378: step: 2354/1541, loss: 0.11435107886791229 2023-01-23 14:26:37.473814: step: 2356/1541, loss: 0.024227751418948174 2023-01-23 14:26:38.388217: step: 2358/1541, loss: 0.5286649465560913 2023-01-23 14:26:39.268413: step: 2360/1541, loss: 0.011114479973912239 2023-01-23 14:26:40.157806: step: 2362/1541, loss: 0.15474963188171387 2023-01-23 14:26:41.123929: step: 2364/1541, loss: 0.039317063987255096 2023-01-23 14:26:41.982586: step: 2366/1541, loss: 0.16874714195728302 2023-01-23 14:26:42.885261: step: 2368/1541, loss: 0.21558374166488647 2023-01-23 14:26:43.768768: step: 2370/1541, loss: 0.12846511602401733 2023-01-23 14:26:44.674374: step: 2372/1541, loss: 0.05142996087670326 2023-01-23 14:26:45.565376: step: 2374/1541, loss: 0.06191970407962799 2023-01-23 14:26:46.472316: step: 2376/1541, loss: 0.1344054788351059 2023-01-23 14:26:47.340712: step: 2378/1541, loss: 0.061152663081884384 2023-01-23 14:26:48.265787: step: 2380/1541, loss: 0.18507960438728333 2023-01-23 14:26:49.146472: step: 2382/1541, loss: 0.003916791174560785 2023-01-23 14:26:50.089112: step: 2384/1541, loss: 0.011482681147754192 2023-01-23 14:26:50.982633: step: 2386/1541, loss: 0.03068653680384159 2023-01-23 14:26:51.878519: step: 2388/1541, loss: 0.1706390678882599 2023-01-23 14:26:52.753875: step: 2390/1541, loss: 0.04254418611526489 2023-01-23 14:26:53.640444: step: 2392/1541, loss: 0.009851023554801941 2023-01-23 14:26:54.533920: step: 2394/1541, loss: 0.23689714074134827 2023-01-23 14:26:55.407683: step: 2396/1541, loss: 0.2690816819667816 2023-01-23 14:26:56.286615: step: 2398/1541, loss: 0.07530439645051956 2023-01-23 14:26:57.218994: step: 2400/1541, loss: 0.03829614073038101 2023-01-23 14:26:58.106302: step: 2402/1541, loss: 0.003568917280063033 2023-01-23 14:26:58.993266: step: 2404/1541, loss: 0.15296104550361633 2023-01-23 14:26:59.870923: step: 2406/1541, loss: 0.006985758896917105 2023-01-23 14:27:00.720091: step: 2408/1541, loss: 0.06010570377111435 2023-01-23 14:27:01.664503: step: 2410/1541, loss: 0.0216588843613863 2023-01-23 14:27:02.636040: step: 2412/1541, loss: 0.018310030922293663 2023-01-23 14:27:03.535393: step: 2414/1541, loss: 0.00612199492752552 2023-01-23 14:27:04.443775: step: 2416/1541, loss: 0.014289729297161102 2023-01-23 14:27:05.388753: step: 2418/1541, loss: 0.0423847958445549 2023-01-23 14:27:06.275675: step: 2420/1541, loss: 0.030621379613876343 2023-01-23 14:27:07.155965: step: 2422/1541, loss: 0.03901814669370651 2023-01-23 14:27:08.069783: step: 2424/1541, loss: 0.024059193208813667 2023-01-23 14:27:08.966929: step: 2426/1541, loss: 0.0012690664734691381 2023-01-23 14:27:09.848691: step: 2428/1541, loss: 0.1854569911956787 2023-01-23 14:27:10.759041: step: 2430/1541, loss: 0.2411404550075531 2023-01-23 14:27:11.658427: step: 2432/1541, loss: 0.2099153697490692 2023-01-23 14:27:12.556964: step: 2434/1541, loss: 0.10523752868175507 2023-01-23 14:27:13.441907: step: 2436/1541, loss: 0.0037271801847964525 2023-01-23 14:27:14.350950: step: 2438/1541, loss: 0.016863441094756126 2023-01-23 14:27:15.258076: step: 2440/1541, loss: 0.04452583193778992 2023-01-23 14:27:16.174100: step: 2442/1541, loss: 0.0011635980335995555 2023-01-23 14:27:17.054017: step: 2444/1541, loss: 0.0011430577142164111 2023-01-23 14:27:17.963446: step: 2446/1541, loss: 0.0013419223250821233 2023-01-23 14:27:18.843173: step: 2448/1541, loss: 0.003868487197905779 2023-01-23 14:27:19.791067: step: 2450/1541, loss: 0.010133619420230389 2023-01-23 14:27:20.728352: step: 2452/1541, loss: 0.04143180325627327 2023-01-23 14:27:21.573464: step: 2454/1541, loss: 0.0015462625306099653 2023-01-23 14:27:22.481743: step: 2456/1541, loss: 0.04059559479355812 2023-01-23 14:27:23.363695: step: 2458/1541, loss: 0.005251492839306593 2023-01-23 14:27:24.308882: step: 2460/1541, loss: 0.019418012350797653 2023-01-23 14:27:25.179327: step: 2462/1541, loss: 0.0037555734161287546 2023-01-23 14:27:26.097747: step: 2464/1541, loss: 0.013033850118517876 2023-01-23 14:27:26.964123: step: 2466/1541, loss: 0.019293228164315224 2023-01-23 14:27:27.882343: step: 2468/1541, loss: 0.10722889006137848 2023-01-23 14:27:28.809898: step: 2470/1541, loss: 0.1441357135772705 2023-01-23 14:27:29.727541: step: 2472/1541, loss: 0.47365516424179077 2023-01-23 14:27:30.625355: step: 2474/1541, loss: 0.2103521078824997 2023-01-23 14:27:31.503587: step: 2476/1541, loss: 0.35322609543800354 2023-01-23 14:27:32.393005: step: 2478/1541, loss: 0.026699723675847054 2023-01-23 14:27:33.307702: step: 2480/1541, loss: 0.00987829640507698 2023-01-23 14:27:34.260188: step: 2482/1541, loss: 0.08267510682344437 2023-01-23 14:27:35.157349: step: 2484/1541, loss: 0.38600972294807434 2023-01-23 14:27:36.005739: step: 2486/1541, loss: 0.20708727836608887 2023-01-23 14:27:36.895682: step: 2488/1541, loss: 0.2093614786863327 2023-01-23 14:27:37.801649: step: 2490/1541, loss: 0.038035862147808075 2023-01-23 14:27:38.735240: step: 2492/1541, loss: 0.01117095910012722 2023-01-23 14:27:39.688235: step: 2494/1541, loss: 0.2812758982181549 2023-01-23 14:27:40.567765: step: 2496/1541, loss: 0.04896145686507225 2023-01-23 14:27:41.454546: step: 2498/1541, loss: 0.0713900476694107 2023-01-23 14:27:42.355215: step: 2500/1541, loss: 0.059223420917987823 2023-01-23 14:27:43.241659: step: 2502/1541, loss: 0.03047577664256096 2023-01-23 14:27:44.137724: step: 2504/1541, loss: 0.21720248460769653 2023-01-23 14:27:45.053189: step: 2506/1541, loss: 0.003913445398211479 2023-01-23 14:27:45.940394: step: 2508/1541, loss: 0.056447017937898636 2023-01-23 14:27:46.855624: step: 2510/1541, loss: 0.047653503715991974 2023-01-23 14:27:47.776560: step: 2512/1541, loss: 0.07218678295612335 2023-01-23 14:27:48.700359: step: 2514/1541, loss: 0.12231504917144775 2023-01-23 14:27:49.566918: step: 2516/1541, loss: 0.016482003033161163 2023-01-23 14:27:50.463021: step: 2518/1541, loss: 0.2279033064842224 2023-01-23 14:27:51.380131: step: 2520/1541, loss: 0.03234478831291199 2023-01-23 14:27:52.289487: step: 2522/1541, loss: 0.08561884611845016 2023-01-23 14:27:53.144745: step: 2524/1541, loss: 0.04833587631583214 2023-01-23 14:27:54.013923: step: 2526/1541, loss: 0.06633913516998291 2023-01-23 14:27:55.049841: step: 2528/1541, loss: 0.11208629608154297 2023-01-23 14:27:55.981783: step: 2530/1541, loss: 0.5714380145072937 2023-01-23 14:27:56.929972: step: 2532/1541, loss: 0.07463069260120392 2023-01-23 14:27:57.814674: step: 2534/1541, loss: 0.03334716334939003 2023-01-23 14:27:58.721586: step: 2536/1541, loss: 0.008055535145103931 2023-01-23 14:27:59.672602: step: 2538/1541, loss: 0.011491788551211357 2023-01-23 14:28:00.552450: step: 2540/1541, loss: 0.25761082768440247 2023-01-23 14:28:01.457508: step: 2542/1541, loss: 0.0010511045111343265 2023-01-23 14:28:02.342574: step: 2544/1541, loss: 0.1547536551952362 2023-01-23 14:28:03.239239: step: 2546/1541, loss: 0.012439063750207424 2023-01-23 14:28:04.116063: step: 2548/1541, loss: 0.004686018452048302 2023-01-23 14:28:05.014047: step: 2550/1541, loss: 0.013490953482687473 2023-01-23 14:28:05.948592: step: 2552/1541, loss: 0.003603191114962101 2023-01-23 14:28:06.875587: step: 2554/1541, loss: 0.0014011645689606667 2023-01-23 14:28:07.789383: step: 2556/1541, loss: 0.06256692856550217 2023-01-23 14:28:08.653459: step: 2558/1541, loss: 0.05553685128688812 2023-01-23 14:28:09.551116: step: 2560/1541, loss: 0.0037135721649974585 2023-01-23 14:28:10.437554: step: 2562/1541, loss: 0.01098275650292635 2023-01-23 14:28:11.405015: step: 2564/1541, loss: 0.02063746005296707 2023-01-23 14:28:12.310426: step: 2566/1541, loss: 0.09874505549669266 2023-01-23 14:28:13.307642: step: 2568/1541, loss: 0.010097799822688103 2023-01-23 14:28:14.193415: step: 2570/1541, loss: 0.045692674815654755 2023-01-23 14:28:15.068509: step: 2572/1541, loss: 1.2244850397109985 2023-01-23 14:28:15.981853: step: 2574/1541, loss: 0.0035671675577759743 2023-01-23 14:28:16.877664: step: 2576/1541, loss: 0.04721824824810028 2023-01-23 14:28:17.780119: step: 2578/1541, loss: 0.004189366940408945 2023-01-23 14:28:18.678363: step: 2580/1541, loss: 0.0006754483329132199 2023-01-23 14:28:19.606251: step: 2582/1541, loss: 0.033689603209495544 2023-01-23 14:28:20.475427: step: 2584/1541, loss: 0.0932808518409729 2023-01-23 14:28:21.432732: step: 2586/1541, loss: 0.8206276893615723 2023-01-23 14:28:22.311063: step: 2588/1541, loss: 0.02585495635867119 2023-01-23 14:28:23.208863: step: 2590/1541, loss: 0.3441183567047119 2023-01-23 14:28:24.132032: step: 2592/1541, loss: 0.01252303458750248 2023-01-23 14:28:25.056017: step: 2594/1541, loss: 0.01756765879690647 2023-01-23 14:28:25.924186: step: 2596/1541, loss: 0.01280541718006134 2023-01-23 14:28:26.836915: step: 2598/1541, loss: 0.04468044638633728 2023-01-23 14:28:27.708116: step: 2600/1541, loss: 0.010016268119215965 2023-01-23 14:28:28.606546: step: 2602/1541, loss: 0.016390666365623474 2023-01-23 14:28:29.492563: step: 2604/1541, loss: 0.014788866974413395 2023-01-23 14:28:30.388589: step: 2606/1541, loss: 0.008631512522697449 2023-01-23 14:28:31.266522: step: 2608/1541, loss: 0.7697489857673645 2023-01-23 14:28:32.265648: step: 2610/1541, loss: 0.3309965133666992 2023-01-23 14:28:33.182693: step: 2612/1541, loss: 0.0006240541115403175 2023-01-23 14:28:34.115619: step: 2614/1541, loss: 0.005507146939635277 2023-01-23 14:28:34.971398: step: 2616/1541, loss: 0.2246003895998001 2023-01-23 14:28:35.821526: step: 2618/1541, loss: 0.23631525039672852 2023-01-23 14:28:36.702177: step: 2620/1541, loss: 0.001829443033784628 2023-01-23 14:28:37.644234: step: 2622/1541, loss: 0.18834146857261658 2023-01-23 14:28:38.503562: step: 2624/1541, loss: 1.1706699132919312 2023-01-23 14:28:39.471277: step: 2626/1541, loss: 0.01801474206149578 2023-01-23 14:28:40.333563: step: 2628/1541, loss: 0.051325723528862 2023-01-23 14:28:41.231879: step: 2630/1541, loss: 0.04677816107869148 2023-01-23 14:28:42.111173: step: 2632/1541, loss: 0.08682719618082047 2023-01-23 14:28:42.991810: step: 2634/1541, loss: 0.07283589988946915 2023-01-23 14:28:43.877969: step: 2636/1541, loss: 0.01824846677482128 2023-01-23 14:28:44.765096: step: 2638/1541, loss: 0.029195647686719894 2023-01-23 14:28:45.707914: step: 2640/1541, loss: 0.025583311915397644 2023-01-23 14:28:46.585613: step: 2642/1541, loss: 0.0914483368396759 2023-01-23 14:28:47.458647: step: 2644/1541, loss: 0.05193319171667099 2023-01-23 14:28:48.364380: step: 2646/1541, loss: 0.07868237048387527 2023-01-23 14:28:49.344316: step: 2648/1541, loss: 0.004133968148380518 2023-01-23 14:28:50.248733: step: 2650/1541, loss: 0.033427245914936066 2023-01-23 14:28:51.154416: step: 2652/1541, loss: 0.27482736110687256 2023-01-23 14:28:52.015725: step: 2654/1541, loss: 0.07715576887130737 2023-01-23 14:28:52.905222: step: 2656/1541, loss: 0.024569695815443993 2023-01-23 14:28:53.814042: step: 2658/1541, loss: 0.030857359990477562 2023-01-23 14:28:54.750333: step: 2660/1541, loss: 0.25756821036338806 2023-01-23 14:28:55.644548: step: 2662/1541, loss: 0.008891415782272816 2023-01-23 14:28:56.508192: step: 2664/1541, loss: 0.0018101013265550137 2023-01-23 14:28:57.375175: step: 2666/1541, loss: 0.04450196772813797 2023-01-23 14:28:58.279152: step: 2668/1541, loss: 0.010209430009126663 2023-01-23 14:28:59.185875: step: 2670/1541, loss: 0.24316364526748657 2023-01-23 14:29:00.098120: step: 2672/1541, loss: 0.003549120854586363 2023-01-23 14:29:00.967387: step: 2674/1541, loss: 0.004555210471153259 2023-01-23 14:29:01.806614: step: 2676/1541, loss: 0.0038586468435823917 2023-01-23 14:29:02.704300: step: 2678/1541, loss: 0.0069725955836474895 2023-01-23 14:29:03.588264: step: 2680/1541, loss: 0.08052515238523483 2023-01-23 14:29:04.462124: step: 2682/1541, loss: 0.01548875030130148 2023-01-23 14:29:05.329768: step: 2684/1541, loss: 0.0011200245935469866 2023-01-23 14:29:06.236936: step: 2686/1541, loss: 0.0360744334757328 2023-01-23 14:29:07.149629: step: 2688/1541, loss: 0.013138530775904655 2023-01-23 14:29:08.031635: step: 2690/1541, loss: 0.2007901817560196 2023-01-23 14:29:08.908284: step: 2692/1541, loss: 0.26202353835105896 2023-01-23 14:29:09.806647: step: 2694/1541, loss: 0.04259442538022995 2023-01-23 14:29:10.705666: step: 2696/1541, loss: 0.08833473175764084 2023-01-23 14:29:11.734488: step: 2698/1541, loss: 0.027121325954794884 2023-01-23 14:29:12.624452: step: 2700/1541, loss: 0.018245544284582138 2023-01-23 14:29:13.580588: step: 2702/1541, loss: 0.002135822083801031 2023-01-23 14:29:14.501590: step: 2704/1541, loss: 0.0012322685215622187 2023-01-23 14:29:15.439413: step: 2706/1541, loss: 0.010182485915720463 2023-01-23 14:29:16.323202: step: 2708/1541, loss: 0.03988854959607124 2023-01-23 14:29:17.220747: step: 2710/1541, loss: 0.2834201753139496 2023-01-23 14:29:18.105019: step: 2712/1541, loss: 0.012031206861138344 2023-01-23 14:29:19.066838: step: 2714/1541, loss: 0.33422428369522095 2023-01-23 14:29:19.984803: step: 2716/1541, loss: 0.12182258069515228 2023-01-23 14:29:20.891373: step: 2718/1541, loss: 0.013082146644592285 2023-01-23 14:29:21.798408: step: 2720/1541, loss: 0.12312042713165283 2023-01-23 14:29:22.654010: step: 2722/1541, loss: 0.7067546248435974 2023-01-23 14:29:23.557715: step: 2724/1541, loss: 0.010100528597831726 2023-01-23 14:29:24.435321: step: 2726/1541, loss: 0.0005000962992198765 2023-01-23 14:29:25.354561: step: 2728/1541, loss: 0.05902973562479019 2023-01-23 14:29:26.205716: step: 2730/1541, loss: 0.10024446994066238 2023-01-23 14:29:27.075753: step: 2732/1541, loss: 0.016114339232444763 2023-01-23 14:29:27.957182: step: 2734/1541, loss: 0.22178427875041962 2023-01-23 14:29:28.862970: step: 2736/1541, loss: 0.045574720948934555 2023-01-23 14:29:29.750089: step: 2738/1541, loss: 0.5213814377784729 2023-01-23 14:29:30.682994: step: 2740/1541, loss: 0.022070633247494698 2023-01-23 14:29:31.596424: step: 2742/1541, loss: 0.16334392130374908 2023-01-23 14:29:32.511497: step: 2744/1541, loss: 0.012554271146655083 2023-01-23 14:29:33.505724: step: 2746/1541, loss: 0.834284245967865 2023-01-23 14:29:34.391776: step: 2748/1541, loss: 0.011901230551302433 2023-01-23 14:29:35.353297: step: 2750/1541, loss: 0.4214922785758972 2023-01-23 14:29:36.263908: step: 2752/1541, loss: 0.0065605822019279 2023-01-23 14:29:37.187972: step: 2754/1541, loss: 0.03360406309366226 2023-01-23 14:29:38.150767: step: 2756/1541, loss: 5.311716079711914 2023-01-23 14:29:39.037947: step: 2758/1541, loss: 0.39773452281951904 2023-01-23 14:29:39.910352: step: 2760/1541, loss: 0.2531640827655792 2023-01-23 14:29:40.801636: step: 2762/1541, loss: 0.08995946496725082 2023-01-23 14:29:41.698606: step: 2764/1541, loss: 1.5997446775436401 2023-01-23 14:29:42.651834: step: 2766/1541, loss: 1.0144213438034058 2023-01-23 14:29:43.536857: step: 2768/1541, loss: 0.18368275463581085 2023-01-23 14:29:44.412292: step: 2770/1541, loss: 0.1155925989151001 2023-01-23 14:29:45.296843: step: 2772/1541, loss: 0.07058102637529373 2023-01-23 14:29:46.187878: step: 2774/1541, loss: 0.18659967184066772 2023-01-23 14:29:47.067461: step: 2776/1541, loss: 0.08250416815280914 2023-01-23 14:29:47.917629: step: 2778/1541, loss: 0.0017787594115361571 2023-01-23 14:29:48.781698: step: 2780/1541, loss: 0.015516603365540504 2023-01-23 14:29:49.647328: step: 2782/1541, loss: 0.0191606767475605 2023-01-23 14:29:50.539347: step: 2784/1541, loss: 0.13567693531513214 2023-01-23 14:29:51.452180: step: 2786/1541, loss: 0.06157941371202469 2023-01-23 14:29:52.383222: step: 2788/1541, loss: 0.09652146697044373 2023-01-23 14:29:53.282405: step: 2790/1541, loss: 0.018391044810414314 2023-01-23 14:29:54.161698: step: 2792/1541, loss: 0.013808486051857471 2023-01-23 14:29:55.073726: step: 2794/1541, loss: 0.2613140046596527 2023-01-23 14:29:56.017713: step: 2796/1541, loss: 0.008716804906725883 2023-01-23 14:29:56.933169: step: 2798/1541, loss: 0.03046264499425888 2023-01-23 14:29:57.810773: step: 2800/1541, loss: 0.004875644575804472 2023-01-23 14:29:58.766501: step: 2802/1541, loss: 0.003964089788496494 2023-01-23 14:29:59.643902: step: 2804/1541, loss: 0.005385813303291798 2023-01-23 14:30:00.504882: step: 2806/1541, loss: 0.044206537306308746 2023-01-23 14:30:01.475889: step: 2808/1541, loss: 0.08496911823749542 2023-01-23 14:30:02.370181: step: 2810/1541, loss: 0.012450351379811764 2023-01-23 14:30:03.311054: step: 2812/1541, loss: 0.008835436776280403 2023-01-23 14:30:04.189439: step: 2814/1541, loss: 0.02839016355574131 2023-01-23 14:30:05.093866: step: 2816/1541, loss: 0.014513684436678886 2023-01-23 14:30:06.049389: step: 2818/1541, loss: 0.04490786790847778 2023-01-23 14:30:07.002131: step: 2820/1541, loss: 0.06322741508483887 2023-01-23 14:30:07.879267: step: 2822/1541, loss: 0.06871025264263153 2023-01-23 14:30:08.778104: step: 2824/1541, loss: 0.011370684020221233 2023-01-23 14:30:09.718374: step: 2826/1541, loss: 0.5058742761611938 2023-01-23 14:30:10.617374: step: 2828/1541, loss: 0.08631178736686707 2023-01-23 14:30:11.489174: step: 2830/1541, loss: 0.06778618693351746 2023-01-23 14:30:12.382109: step: 2832/1541, loss: 0.042834654450416565 2023-01-23 14:30:13.260768: step: 2834/1541, loss: 0.20563003420829773 2023-01-23 14:30:14.241886: step: 2836/1541, loss: 0.011565578170120716 2023-01-23 14:30:15.190432: step: 2838/1541, loss: 0.03206267207860947 2023-01-23 14:30:16.105838: step: 2840/1541, loss: 0.0066482494585216045 2023-01-23 14:30:17.020284: step: 2842/1541, loss: 0.11609742790460587 2023-01-23 14:30:17.887843: step: 2844/1541, loss: 0.0031744714360684156 2023-01-23 14:30:18.740044: step: 2846/1541, loss: 0.015843041241168976 2023-01-23 14:30:19.677201: step: 2848/1541, loss: 0.11807087063789368 2023-01-23 14:30:20.658917: step: 2850/1541, loss: 0.0013620903482660651 2023-01-23 14:30:21.607360: step: 2852/1541, loss: 0.17668084800243378 2023-01-23 14:30:22.505394: step: 2854/1541, loss: 0.019825929775834084 2023-01-23 14:30:23.433398: step: 2856/1541, loss: 0.0014934251084923744 2023-01-23 14:30:24.366822: step: 2858/1541, loss: 0.009645847603678703 2023-01-23 14:30:25.283678: step: 2860/1541, loss: 0.10558478534221649 2023-01-23 14:30:26.176536: step: 2862/1541, loss: 0.018995454534888268 2023-01-23 14:30:27.063197: step: 2864/1541, loss: 0.000721273070666939 2023-01-23 14:30:27.950243: step: 2866/1541, loss: 0.5706853270530701 2023-01-23 14:30:28.842460: step: 2868/1541, loss: 0.17587949335575104 2023-01-23 14:30:29.712826: step: 2870/1541, loss: 0.8705295920372009 2023-01-23 14:30:30.603639: step: 2872/1541, loss: 0.03664668649435043 2023-01-23 14:30:31.544697: step: 2874/1541, loss: 0.052941348403692245 2023-01-23 14:30:32.462425: step: 2876/1541, loss: 0.006598265841603279 2023-01-23 14:30:33.383789: step: 2878/1541, loss: 0.003609397215768695 2023-01-23 14:30:34.279455: step: 2880/1541, loss: 0.08698729425668716 2023-01-23 14:30:35.209812: step: 2882/1541, loss: 0.004563175607472658 2023-01-23 14:30:36.119872: step: 2884/1541, loss: 0.38168081641197205 2023-01-23 14:30:37.015296: step: 2886/1541, loss: 0.014069771394133568 2023-01-23 14:30:37.885956: step: 2888/1541, loss: 0.2587631046772003 2023-01-23 14:30:38.773024: step: 2890/1541, loss: 0.0052452078089118 2023-01-23 14:30:39.640099: step: 2892/1541, loss: 0.0035643004812300205 2023-01-23 14:30:40.543765: step: 2894/1541, loss: 0.014875243417918682 2023-01-23 14:30:41.421098: step: 2896/1541, loss: 0.0011070435866713524 2023-01-23 14:30:42.328650: step: 2898/1541, loss: 0.009611331857740879 2023-01-23 14:30:43.192797: step: 2900/1541, loss: 0.04989007115364075 2023-01-23 14:30:44.149062: step: 2902/1541, loss: 0.007669121492654085 2023-01-23 14:30:45.119755: step: 2904/1541, loss: 0.21924744546413422 2023-01-23 14:30:45.999948: step: 2906/1541, loss: 0.1303994208574295 2023-01-23 14:30:46.918325: step: 2908/1541, loss: 0.011416460387408733 2023-01-23 14:30:47.810205: step: 2910/1541, loss: 0.016921324655413628 2023-01-23 14:30:48.732122: step: 2912/1541, loss: 0.20170457661151886 2023-01-23 14:30:49.629056: step: 2914/1541, loss: 0.5750266313552856 2023-01-23 14:30:50.520918: step: 2916/1541, loss: 0.013705998659133911 2023-01-23 14:30:51.438206: step: 2918/1541, loss: 0.07571543008089066 2023-01-23 14:30:52.319003: step: 2920/1541, loss: 0.0006592290010303259 2023-01-23 14:30:53.263058: step: 2922/1541, loss: 0.004415647126734257 2023-01-23 14:30:54.196260: step: 2924/1541, loss: 0.1932874172925949 2023-01-23 14:30:55.068296: step: 2926/1541, loss: 0.016490178182721138 2023-01-23 14:30:55.932417: step: 2928/1541, loss: 0.18817168474197388 2023-01-23 14:30:56.838252: step: 2930/1541, loss: 0.022803857922554016 2023-01-23 14:30:57.736584: step: 2932/1541, loss: 0.08972574770450592 2023-01-23 14:30:58.677745: step: 2934/1541, loss: 0.009444830939173698 2023-01-23 14:30:59.553889: step: 2936/1541, loss: 0.0006721225217916071 2023-01-23 14:31:00.473555: step: 2938/1541, loss: 0.032145630568265915 2023-01-23 14:31:01.358212: step: 2940/1541, loss: 0.09077713638544083 2023-01-23 14:31:02.241731: step: 2942/1541, loss: 1.3943688869476318 2023-01-23 14:31:03.195815: step: 2944/1541, loss: 0.06141170486807823 2023-01-23 14:31:04.085068: step: 2946/1541, loss: 0.045472994446754456 2023-01-23 14:31:04.969969: step: 2948/1541, loss: 0.021035414189100266 2023-01-23 14:31:05.876362: step: 2950/1541, loss: 0.09226444363594055 2023-01-23 14:31:06.756878: step: 2952/1541, loss: 0.21112164855003357 2023-01-23 14:31:07.666417: step: 2954/1541, loss: 0.03519939258694649 2023-01-23 14:31:08.537088: step: 2956/1541, loss: 0.474243700504303 2023-01-23 14:31:09.484551: step: 2958/1541, loss: 0.02955322153866291 2023-01-23 14:31:10.348338: step: 2960/1541, loss: 0.2846032381057739 2023-01-23 14:31:11.219036: step: 2962/1541, loss: 0.08499595522880554 2023-01-23 14:31:12.132300: step: 2964/1541, loss: 0.16872607171535492 2023-01-23 14:31:13.111543: step: 2966/1541, loss: 0.10781361162662506 2023-01-23 14:31:13.989699: step: 2968/1541, loss: 0.016050346195697784 2023-01-23 14:31:14.869344: step: 2970/1541, loss: 0.0005743026849813759 2023-01-23 14:31:15.752849: step: 2972/1541, loss: 0.046217188239097595 2023-01-23 14:31:16.690037: step: 2974/1541, loss: 0.026974445208907127 2023-01-23 14:31:17.562739: step: 2976/1541, loss: 0.0785592794418335 2023-01-23 14:31:18.425171: step: 2978/1541, loss: 0.3279232084751129 2023-01-23 14:31:19.354436: step: 2980/1541, loss: 0.286877304315567 2023-01-23 14:31:20.299227: step: 2982/1541, loss: 0.14631952345371246 2023-01-23 14:31:21.193796: step: 2984/1541, loss: 0.7866782546043396 2023-01-23 14:31:22.074811: step: 2986/1541, loss: 0.0913606658577919 2023-01-23 14:31:22.964805: step: 2988/1541, loss: 0.1412551999092102 2023-01-23 14:31:23.858911: step: 2990/1541, loss: 0.23728816211223602 2023-01-23 14:31:24.729821: step: 2992/1541, loss: 0.09618819504976273 2023-01-23 14:31:25.641306: step: 2994/1541, loss: 0.03948437049984932 2023-01-23 14:31:26.529192: step: 2996/1541, loss: 0.14371787011623383 2023-01-23 14:31:27.404603: step: 2998/1541, loss: 0.13709816336631775 2023-01-23 14:31:28.321633: step: 3000/1541, loss: 0.025225181132555008 2023-01-23 14:31:29.268473: step: 3002/1541, loss: 0.1412876695394516 2023-01-23 14:31:30.161038: step: 3004/1541, loss: 0.0067994738928973675 2023-01-23 14:31:31.085009: step: 3006/1541, loss: 0.052835654467344284 2023-01-23 14:31:31.999311: step: 3008/1541, loss: 0.02275489643216133 2023-01-23 14:31:32.940774: step: 3010/1541, loss: 0.2135123461484909 2023-01-23 14:31:33.893994: step: 3012/1541, loss: 0.12354106456041336 2023-01-23 14:31:34.800721: step: 3014/1541, loss: 0.08005888015031815 2023-01-23 14:31:35.704895: step: 3016/1541, loss: 0.2664591372013092 2023-01-23 14:31:36.598637: step: 3018/1541, loss: 0.003738206345587969 2023-01-23 14:31:37.479678: step: 3020/1541, loss: 0.02468537725508213 2023-01-23 14:31:38.338474: step: 3022/1541, loss: 0.008080204017460346 2023-01-23 14:31:39.242329: step: 3024/1541, loss: 0.028955776244401932 2023-01-23 14:31:40.139287: step: 3026/1541, loss: 0.05162983760237694 2023-01-23 14:31:41.033714: step: 3028/1541, loss: 0.06330092996358871 2023-01-23 14:31:41.899158: step: 3030/1541, loss: 0.6477805972099304 2023-01-23 14:31:42.781187: step: 3032/1541, loss: 0.015025155618786812 2023-01-23 14:31:43.678669: step: 3034/1541, loss: 0.0903439000248909 2023-01-23 14:31:44.557892: step: 3036/1541, loss: 0.16031333804130554 2023-01-23 14:31:45.454610: step: 3038/1541, loss: 0.026234878227114677 2023-01-23 14:31:46.359608: step: 3040/1541, loss: 0.16532553732395172 2023-01-23 14:31:47.281981: step: 3042/1541, loss: 0.5808386206626892 2023-01-23 14:31:48.233170: step: 3044/1541, loss: 0.008564391173422337 2023-01-23 14:31:49.151374: step: 3046/1541, loss: 0.0063416920602321625 2023-01-23 14:31:50.052572: step: 3048/1541, loss: 0.13151395320892334 2023-01-23 14:31:50.911796: step: 3050/1541, loss: 0.008889219723641872 2023-01-23 14:31:51.839861: step: 3052/1541, loss: 0.08771167695522308 2023-01-23 14:31:52.700989: step: 3054/1541, loss: 0.03766860067844391 2023-01-23 14:31:53.611311: step: 3056/1541, loss: 0.19015103578567505 2023-01-23 14:31:54.481281: step: 3058/1541, loss: 0.5971313714981079 2023-01-23 14:31:55.354353: step: 3060/1541, loss: 0.2778802514076233 2023-01-23 14:31:56.233882: step: 3062/1541, loss: 0.0478278249502182 2023-01-23 14:31:57.148817: step: 3064/1541, loss: 0.002529993187636137 2023-01-23 14:31:58.065529: step: 3066/1541, loss: 0.5200748443603516 2023-01-23 14:31:58.985099: step: 3068/1541, loss: 0.5668439269065857 2023-01-23 14:31:59.849208: step: 3070/1541, loss: 0.13641373813152313 2023-01-23 14:32:00.774666: step: 3072/1541, loss: 0.047345373779535294 2023-01-23 14:32:01.664130: step: 3074/1541, loss: 0.0014701337786391377 2023-01-23 14:32:02.557237: step: 3076/1541, loss: 0.06915567815303802 2023-01-23 14:32:03.440061: step: 3078/1541, loss: 0.0024954553227871656 2023-01-23 14:32:04.349907: step: 3080/1541, loss: 0.06250207126140594 2023-01-23 14:32:05.260084: step: 3082/1541, loss: 0.003682513255625963 2023-01-23 14:32:05.855985: step: 3084/1541, loss: 0.008871801197528839 ================================================== Loss: 0.114 -------------------- Dev Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6763328078722167, 'r': 0.6590044051860497, 'f1': 0.6675561729293212}, 'combined': 0.44223494579448736, 'epoch': 19} Test Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7190813155360521, 'r': 0.6983939798014024, 'f1': 0.708586687062498}, 'combined': 0.4623374351317896, 'epoch': 19} Dev Korean: {'event': {'p': 0.5735294117647058, 'r': 0.7789613848202397, 'f1': 0.6606437041219649}, 'argument': {'p': 0.6713720494406288, 'r': 0.6636662009535053, 'f1': 0.6674968860960941}, 'combined': 0.440977615320401, 'epoch': 19} Test Korean: {'event': {'p': 0.6124880838894184, 'r': 0.7699221090473337, 'f1': 0.6822405096894081}, 'argument': {'p': 0.7136783443203177, 'r': 0.7134281064716781, 'f1': 0.7135532034568562}, 'combined': 0.4868149012169155, 'epoch': 19} Dev Russian: {'event': {'p': 0.6575829383886256, 'r': 0.7390146471371505, 'f1': 0.695924764890282}, 'argument': {'p': 0.6711868599987194, 'r': 0.6661289333973726, 'f1': 0.6686483317838983}, 'combined': 0.46532893309098866, 'epoch': 19} Test Russian: {'event': {'p': 0.652014652014652, 'r': 0.7465548232474536, 'f1': 0.6960893854748604}, 'argument': {'p': 0.7115327423082188, 'r': 0.7015723054991662, 'f1': 0.7065174202496574}, 'combined': 0.4917992768888677, 'epoch': 19} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6325115562403697, 'r': 0.660498793242156, 'f1': 0.6462022825659188}, 'combined': 0.47498629316811125, 'epoch': 19} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5301587301587302, 'r': 0.34425891568748707, 'f1': 0.4174478190226221}, 'combined': 0.2432373118714491, 'epoch': 19} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5613247843719343, 'r': 0.6396491728889484, 'f1': 0.5979329224831473}, 'combined': 0.3075083601341901, 'epoch': 19} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test for Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 20 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 14:37:29.419024: step: 2/1541, loss: 0.26052722334861755 2023-01-23 14:37:30.259519: step: 4/1541, loss: 0.004312042146921158 2023-01-23 14:37:31.099378: step: 6/1541, loss: 0.0015609543770551682 2023-01-23 14:37:31.981242: step: 8/1541, loss: 0.14582599699497223 2023-01-23 14:37:32.850618: step: 10/1541, loss: 0.01595141366124153 2023-01-23 14:37:33.732632: step: 12/1541, loss: 0.015729419887065887 2023-01-23 14:37:34.600219: step: 14/1541, loss: 0.003864104161038995 2023-01-23 14:37:35.465827: step: 16/1541, loss: 0.002820666180923581 2023-01-23 14:37:36.346305: step: 18/1541, loss: 0.0659559965133667 2023-01-23 14:37:37.232033: step: 20/1541, loss: 0.025487393140792847 2023-01-23 14:37:38.182732: step: 22/1541, loss: 0.00852445513010025 2023-01-23 14:37:39.084730: step: 24/1541, loss: 0.0017599965212866664 2023-01-23 14:37:39.969121: step: 26/1541, loss: 0.31973299384117126 2023-01-23 14:37:40.865467: step: 28/1541, loss: 0.10300710052251816 2023-01-23 14:37:41.750642: step: 30/1541, loss: 0.0028266950976103544 2023-01-23 14:37:42.659683: step: 32/1541, loss: 0.020595602691173553 2023-01-23 14:37:43.527545: step: 34/1541, loss: 0.10379064083099365 2023-01-23 14:37:44.424635: step: 36/1541, loss: 0.006883212830871344 2023-01-23 14:37:45.343152: step: 38/1541, loss: 0.036623042076826096 2023-01-23 14:37:46.247043: step: 40/1541, loss: 0.48429784178733826 2023-01-23 14:37:47.163463: step: 42/1541, loss: 0.004307319410145283 2023-01-23 14:37:48.018808: step: 44/1541, loss: 0.2955109179019928 2023-01-23 14:37:48.898788: step: 46/1541, loss: 0.23374223709106445 2023-01-23 14:37:49.786106: step: 48/1541, loss: 0.000624489679466933 2023-01-23 14:37:50.676349: step: 50/1541, loss: 0.0023635185789316893 2023-01-23 14:37:51.556269: step: 52/1541, loss: 0.0013833703706040978 2023-01-23 14:37:52.506988: step: 54/1541, loss: 0.04134543240070343 2023-01-23 14:37:53.360186: step: 56/1541, loss: 0.008015324361622334 2023-01-23 14:37:54.272677: step: 58/1541, loss: 0.00026156852254644036 2023-01-23 14:37:55.199752: step: 60/1541, loss: 0.038705918937921524 2023-01-23 14:37:56.109505: step: 62/1541, loss: 0.3498249650001526 2023-01-23 14:37:56.989675: step: 64/1541, loss: 0.11038661003112793 2023-01-23 14:37:57.928619: step: 66/1541, loss: 0.0007975719636306167 2023-01-23 14:37:58.810759: step: 68/1541, loss: 0.003938476089388132 2023-01-23 14:37:59.678058: step: 70/1541, loss: 0.07285416126251221 2023-01-23 14:38:00.565257: step: 72/1541, loss: 0.035287175327539444 2023-01-23 14:38:01.522999: step: 74/1541, loss: 0.21757209300994873 2023-01-23 14:38:02.420529: step: 76/1541, loss: 0.009751600213348866 2023-01-23 14:38:03.255349: step: 78/1541, loss: 0.022619711235165596 2023-01-23 14:38:04.134609: step: 80/1541, loss: 0.14801527559757233 2023-01-23 14:38:05.080468: step: 82/1541, loss: 0.02652822621166706 2023-01-23 14:38:05.954169: step: 84/1541, loss: 0.09003207832574844 2023-01-23 14:38:06.832238: step: 86/1541, loss: 0.039553746581077576 2023-01-23 14:38:07.744558: step: 88/1541, loss: 0.009206793271005154 2023-01-23 14:38:08.588393: step: 90/1541, loss: 0.006028847303241491 2023-01-23 14:38:09.500076: step: 92/1541, loss: 0.0680864155292511 2023-01-23 14:38:10.460189: step: 94/1541, loss: 0.47576019167900085 2023-01-23 14:38:11.304544: step: 96/1541, loss: 0.03071100078523159 2023-01-23 14:38:12.193022: step: 98/1541, loss: 0.008475064300000668 2023-01-23 14:38:13.119945: step: 100/1541, loss: 0.005568194203078747 2023-01-23 14:38:13.994276: step: 102/1541, loss: 0.2740955054759979 2023-01-23 14:38:14.873249: step: 104/1541, loss: 0.005871633533388376 2023-01-23 14:38:15.786803: step: 106/1541, loss: 0.010593019425868988 2023-01-23 14:38:16.752225: step: 108/1541, loss: 0.20327164232730865 2023-01-23 14:38:17.691741: step: 110/1541, loss: 0.009132735431194305 2023-01-23 14:38:18.574492: step: 112/1541, loss: 0.006372724659740925 2023-01-23 14:38:19.489852: step: 114/1541, loss: 0.00048291479470208287 2023-01-23 14:38:20.375289: step: 116/1541, loss: 0.0011896587675437331 2023-01-23 14:38:21.274174: step: 118/1541, loss: 0.0797819122672081 2023-01-23 14:38:22.146438: step: 120/1541, loss: 0.2088932991027832 2023-01-23 14:38:23.013312: step: 122/1541, loss: 0.023416750133037567 2023-01-23 14:38:23.889540: step: 124/1541, loss: 0.0030216691084206104 2023-01-23 14:38:24.795760: step: 126/1541, loss: 0.0042836591601371765 2023-01-23 14:38:25.711213: step: 128/1541, loss: 0.008496868424117565 2023-01-23 14:38:26.599370: step: 130/1541, loss: 0.0033712584991008043 2023-01-23 14:38:27.486563: step: 132/1541, loss: 0.042376868426799774 2023-01-23 14:38:28.400867: step: 134/1541, loss: 0.009241335093975067 2023-01-23 14:38:29.261520: step: 136/1541, loss: 0.028480999171733856 2023-01-23 14:38:30.159759: step: 138/1541, loss: 0.018057549372315407 2023-01-23 14:38:31.117233: step: 140/1541, loss: 0.00026021714438684285 2023-01-23 14:38:32.013583: step: 142/1541, loss: 0.009054329246282578 2023-01-23 14:38:32.857962: step: 144/1541, loss: 0.0020306576043367386 2023-01-23 14:38:33.799762: step: 146/1541, loss: 0.0009305702988058329 2023-01-23 14:38:34.753220: step: 148/1541, loss: 0.005408919416368008 2023-01-23 14:38:35.616371: step: 150/1541, loss: 0.00379978958517313 2023-01-23 14:38:36.516302: step: 152/1541, loss: 0.49679034948349 2023-01-23 14:38:37.412877: step: 154/1541, loss: 0.013510606251657009 2023-01-23 14:38:38.293031: step: 156/1541, loss: 0.00018540775636211038 2023-01-23 14:38:39.195763: step: 158/1541, loss: 0.0015003925655037165 2023-01-23 14:38:40.135625: step: 160/1541, loss: 0.03214363753795624 2023-01-23 14:38:40.999945: step: 162/1541, loss: 0.1131444051861763 2023-01-23 14:38:41.875412: step: 164/1541, loss: 0.00263426941819489 2023-01-23 14:38:42.791854: step: 166/1541, loss: 0.04809746518731117 2023-01-23 14:38:43.634636: step: 168/1541, loss: 0.004574735648930073 2023-01-23 14:38:44.566008: step: 170/1541, loss: 0.5662730932235718 2023-01-23 14:38:45.453149: step: 172/1541, loss: 0.16383609175682068 2023-01-23 14:38:46.337046: step: 174/1541, loss: 0.1255757212638855 2023-01-23 14:38:47.207904: step: 176/1541, loss: 0.14430014789104462 2023-01-23 14:38:48.144033: step: 178/1541, loss: 0.020524801686406136 2023-01-23 14:38:49.041109: step: 180/1541, loss: 0.0033451307099312544 2023-01-23 14:38:49.942394: step: 182/1541, loss: 0.0595240518450737 2023-01-23 14:38:50.810119: step: 184/1541, loss: 0.0006686819833703339 2023-01-23 14:38:51.718127: step: 186/1541, loss: 0.009310015477240086 2023-01-23 14:38:52.604299: step: 188/1541, loss: 0.010271593928337097 2023-01-23 14:38:53.489608: step: 190/1541, loss: 0.001166630070656538 2023-01-23 14:38:54.352035: step: 192/1541, loss: 0.014414891600608826 2023-01-23 14:38:55.291676: step: 194/1541, loss: 0.0053884428925812244 2023-01-23 14:38:56.195765: step: 196/1541, loss: 0.07122215628623962 2023-01-23 14:38:57.109079: step: 198/1541, loss: 0.011580136604607105 2023-01-23 14:38:58.031538: step: 200/1541, loss: 0.030603060498833656 2023-01-23 14:38:58.959867: step: 202/1541, loss: 0.010448930785059929 2023-01-23 14:38:59.873268: step: 204/1541, loss: 0.025475339964032173 2023-01-23 14:39:00.701564: step: 206/1541, loss: 0.0003231983573641628 2023-01-23 14:39:01.623242: step: 208/1541, loss: 0.32892316579818726 2023-01-23 14:39:02.487673: step: 210/1541, loss: 0.06108943372964859 2023-01-23 14:39:03.496999: step: 212/1541, loss: 0.10235926508903503 2023-01-23 14:39:04.377836: step: 214/1541, loss: 0.0004136250354349613 2023-01-23 14:39:05.229923: step: 216/1541, loss: 0.0012120557948946953 2023-01-23 14:39:06.137656: step: 218/1541, loss: 0.040278926491737366 2023-01-23 14:39:06.997046: step: 220/1541, loss: 0.0002847366558853537 2023-01-23 14:39:07.924438: step: 222/1541, loss: 1.2553809938253835e-05 2023-01-23 14:39:08.814683: step: 224/1541, loss: 0.03431353345513344 2023-01-23 14:39:09.692339: step: 226/1541, loss: 0.0025831556413322687 2023-01-23 14:39:10.590725: step: 228/1541, loss: 0.015747735276818275 2023-01-23 14:39:11.494510: step: 230/1541, loss: 0.004222837742418051 2023-01-23 14:39:12.361550: step: 232/1541, loss: 0.09693960845470428 2023-01-23 14:39:13.247254: step: 234/1541, loss: 0.0017551094060763717 2023-01-23 14:39:14.138153: step: 236/1541, loss: 0.00034788408083841205 2023-01-23 14:39:15.108396: step: 238/1541, loss: 0.039569057524204254 2023-01-23 14:39:16.033225: step: 240/1541, loss: 0.009251503273844719 2023-01-23 14:39:16.984131: step: 242/1541, loss: 0.0070761083625257015 2023-01-23 14:39:17.867082: step: 244/1541, loss: 0.017478616908192635 2023-01-23 14:39:18.770032: step: 246/1541, loss: 0.06553170084953308 2023-01-23 14:39:19.701811: step: 248/1541, loss: 0.0016680896515026689 2023-01-23 14:39:20.625340: step: 250/1541, loss: 0.040865156799554825 2023-01-23 14:39:21.485626: step: 252/1541, loss: 0.03857989236712456 2023-01-23 14:39:22.366270: step: 254/1541, loss: 0.0008529964252375066 2023-01-23 14:39:23.259131: step: 256/1541, loss: 0.1529330015182495 2023-01-23 14:39:24.119075: step: 258/1541, loss: 0.0017774453153833747 2023-01-23 14:39:25.010667: step: 260/1541, loss: 0.6326757073402405 2023-01-23 14:39:25.918907: step: 262/1541, loss: 0.008424441330134869 2023-01-23 14:39:26.818093: step: 264/1541, loss: 0.20613417029380798 2023-01-23 14:39:27.694236: step: 266/1541, loss: 0.07042527943849564 2023-01-23 14:39:28.586763: step: 268/1541, loss: 0.0024017998948693275 2023-01-23 14:39:29.483589: step: 270/1541, loss: 0.322501540184021 2023-01-23 14:39:30.396205: step: 272/1541, loss: 0.08441367745399475 2023-01-23 14:39:31.307756: step: 274/1541, loss: 0.02335173636674881 2023-01-23 14:39:32.250342: step: 276/1541, loss: 0.754289448261261 2023-01-23 14:39:33.174672: step: 278/1541, loss: 0.4302430748939514 2023-01-23 14:39:34.107079: step: 280/1541, loss: 0.19214372336864471 2023-01-23 14:39:35.064233: step: 282/1541, loss: 0.01110201608389616 2023-01-23 14:39:35.999446: step: 284/1541, loss: 0.06337303668260574 2023-01-23 14:39:36.899790: step: 286/1541, loss: 0.09749133139848709 2023-01-23 14:39:37.807101: step: 288/1541, loss: 0.01395460870116949 2023-01-23 14:39:38.704731: step: 290/1541, loss: 0.015542733483016491 2023-01-23 14:39:39.615179: step: 292/1541, loss: 0.0001975251652766019 2023-01-23 14:39:40.510838: step: 294/1541, loss: 0.0003491009119898081 2023-01-23 14:39:41.404942: step: 296/1541, loss: 0.01627402938902378 2023-01-23 14:39:42.367024: step: 298/1541, loss: 0.07418464869260788 2023-01-23 14:39:43.246600: step: 300/1541, loss: 0.02146293595433235 2023-01-23 14:39:44.182810: step: 302/1541, loss: 0.062167756259441376 2023-01-23 14:39:45.172353: step: 304/1541, loss: 0.28929126262664795 2023-01-23 14:39:46.091674: step: 306/1541, loss: 0.032972753047943115 2023-01-23 14:39:47.031441: step: 308/1541, loss: 0.0016205398133024573 2023-01-23 14:39:47.935991: step: 310/1541, loss: 0.04693244397640228 2023-01-23 14:39:48.831707: step: 312/1541, loss: 0.07058319449424744 2023-01-23 14:39:49.731215: step: 314/1541, loss: 0.12654374539852142 2023-01-23 14:39:50.620734: step: 316/1541, loss: 0.06844421476125717 2023-01-23 14:39:51.560040: step: 318/1541, loss: 0.005533336661756039 2023-01-23 14:39:52.450661: step: 320/1541, loss: 0.027521232143044472 2023-01-23 14:39:53.402140: step: 322/1541, loss: 0.05381683260202408 2023-01-23 14:39:54.284519: step: 324/1541, loss: 0.0029895121697336435 2023-01-23 14:39:55.214370: step: 326/1541, loss: 0.04870901629328728 2023-01-23 14:39:56.087922: step: 328/1541, loss: 0.0021567342337220907 2023-01-23 14:39:57.017338: step: 330/1541, loss: 0.016023389995098114 2023-01-23 14:39:57.961183: step: 332/1541, loss: 0.004106288310140371 2023-01-23 14:39:58.867810: step: 334/1541, loss: 0.03359534591436386 2023-01-23 14:39:59.764876: step: 336/1541, loss: 0.03889274224638939 2023-01-23 14:40:00.636584: step: 338/1541, loss: 0.04329176992177963 2023-01-23 14:40:01.524707: step: 340/1541, loss: 0.1300399899482727 2023-01-23 14:40:02.417625: step: 342/1541, loss: 0.005736941937357187 2023-01-23 14:40:03.276912: step: 344/1541, loss: 0.10867950320243835 2023-01-23 14:40:04.159241: step: 346/1541, loss: 0.03322408348321915 2023-01-23 14:40:05.052087: step: 348/1541, loss: 0.0049471426755189896 2023-01-23 14:40:05.971841: step: 350/1541, loss: 0.0060414522886276245 2023-01-23 14:40:06.874543: step: 352/1541, loss: 0.00203197100199759 2023-01-23 14:40:07.733495: step: 354/1541, loss: 0.0004254558007232845 2023-01-23 14:40:08.635471: step: 356/1541, loss: 0.09120732545852661 2023-01-23 14:40:09.542019: step: 358/1541, loss: 0.02184911258518696 2023-01-23 14:40:10.438313: step: 360/1541, loss: 0.004264919087290764 2023-01-23 14:40:11.352594: step: 362/1541, loss: 0.005809272639453411 2023-01-23 14:40:12.244845: step: 364/1541, loss: 0.0018554575508460402 2023-01-23 14:40:13.176620: step: 366/1541, loss: 0.27582916617393494 2023-01-23 14:40:14.048160: step: 368/1541, loss: 0.24800436198711395 2023-01-23 14:40:14.950000: step: 370/1541, loss: 0.18700900673866272 2023-01-23 14:40:15.814627: step: 372/1541, loss: 0.0023896158672869205 2023-01-23 14:40:16.673453: step: 374/1541, loss: 0.004736054223030806 2023-01-23 14:40:17.598974: step: 376/1541, loss: 0.05942230299115181 2023-01-23 14:40:18.544827: step: 378/1541, loss: 0.00039380573434755206 2023-01-23 14:40:19.431031: step: 380/1541, loss: 0.015873733907938004 2023-01-23 14:40:20.315498: step: 382/1541, loss: 0.015387365594506264 2023-01-23 14:40:21.209632: step: 384/1541, loss: 0.010561851784586906 2023-01-23 14:40:22.180022: step: 386/1541, loss: 0.015932684764266014 2023-01-23 14:40:23.059100: step: 388/1541, loss: 0.720106303691864 2023-01-23 14:40:23.909300: step: 390/1541, loss: 0.361054927110672 2023-01-23 14:40:24.789283: step: 392/1541, loss: 0.10421533137559891 2023-01-23 14:40:25.691035: step: 394/1541, loss: 0.00985540822148323 2023-01-23 14:40:26.551265: step: 396/1541, loss: 0.004404325969517231 2023-01-23 14:40:27.428452: step: 398/1541, loss: 0.019587237387895584 2023-01-23 14:40:28.339684: step: 400/1541, loss: 0.01981387659907341 2023-01-23 14:40:29.226926: step: 402/1541, loss: 0.09601565450429916 2023-01-23 14:40:30.088568: step: 404/1541, loss: 0.0006019726861268282 2023-01-23 14:40:31.008153: step: 406/1541, loss: 0.007770684082061052 2023-01-23 14:40:31.955223: step: 408/1541, loss: 0.000903372943866998 2023-01-23 14:40:32.855380: step: 410/1541, loss: 0.5975468754768372 2023-01-23 14:40:33.754022: step: 412/1541, loss: 0.10682407766580582 2023-01-23 14:40:34.611265: step: 414/1541, loss: 0.004407711327075958 2023-01-23 14:40:35.508765: step: 416/1541, loss: 0.05183235928416252 2023-01-23 14:40:36.436659: step: 418/1541, loss: 0.00041283838800154626 2023-01-23 14:40:37.329281: step: 420/1541, loss: 0.0705690085887909 2023-01-23 14:40:38.275438: step: 422/1541, loss: 0.0273742713034153 2023-01-23 14:40:39.234340: step: 424/1541, loss: 0.04121452569961548 2023-01-23 14:40:40.148501: step: 426/1541, loss: 0.009284518659114838 2023-01-23 14:40:41.034305: step: 428/1541, loss: 0.0035093831829726696 2023-01-23 14:40:41.933581: step: 430/1541, loss: 2.3808133602142334 2023-01-23 14:40:42.838203: step: 432/1541, loss: 0.031052179634571075 2023-01-23 14:40:43.720387: step: 434/1541, loss: 0.013485316187143326 2023-01-23 14:40:44.592783: step: 436/1541, loss: 0.001959496643394232 2023-01-23 14:40:45.471207: step: 438/1541, loss: 0.16427195072174072 2023-01-23 14:40:46.444650: step: 440/1541, loss: 0.0048873545601964 2023-01-23 14:40:47.336739: step: 442/1541, loss: 0.008348612114787102 2023-01-23 14:40:48.209618: step: 444/1541, loss: 0.03985283523797989 2023-01-23 14:40:49.266952: step: 446/1541, loss: 0.028586911037564278 2023-01-23 14:40:50.164338: step: 448/1541, loss: 0.10266415774822235 2023-01-23 14:40:51.195808: step: 450/1541, loss: 0.05594159662723541 2023-01-23 14:40:52.142510: step: 452/1541, loss: 0.0021166533697396517 2023-01-23 14:40:53.034067: step: 454/1541, loss: 0.009615117684006691 2023-01-23 14:40:53.868810: step: 456/1541, loss: 0.050415173172950745 2023-01-23 14:40:54.735199: step: 458/1541, loss: 0.11320128291845322 2023-01-23 14:40:55.581935: step: 460/1541, loss: 0.3798447251319885 2023-01-23 14:40:56.486228: step: 462/1541, loss: 0.004134126473218203 2023-01-23 14:40:57.410040: step: 464/1541, loss: 0.0792321115732193 2023-01-23 14:40:58.316306: step: 466/1541, loss: 0.26899364590644836 2023-01-23 14:40:59.226788: step: 468/1541, loss: 0.01264805905520916 2023-01-23 14:41:00.121471: step: 470/1541, loss: 0.002997811883687973 2023-01-23 14:41:00.964540: step: 472/1541, loss: 0.39732661843299866 2023-01-23 14:41:01.878304: step: 474/1541, loss: 0.04536206275224686 2023-01-23 14:41:02.772889: step: 476/1541, loss: 0.2131555825471878 2023-01-23 14:41:03.642074: step: 478/1541, loss: 0.0037987991236150265 2023-01-23 14:41:04.510143: step: 480/1541, loss: 0.07945414632558823 2023-01-23 14:41:05.362513: step: 482/1541, loss: 0.037281811237335205 2023-01-23 14:41:06.283075: step: 484/1541, loss: 0.009196650236845016 2023-01-23 14:41:07.200131: step: 486/1541, loss: 0.0204470232129097 2023-01-23 14:41:08.110681: step: 488/1541, loss: 0.05006467550992966 2023-01-23 14:41:08.984782: step: 490/1541, loss: 0.3176155686378479 2023-01-23 14:41:09.879291: step: 492/1541, loss: 0.025693560019135475 2023-01-23 14:41:10.805816: step: 494/1541, loss: 0.22499582171440125 2023-01-23 14:41:11.698415: step: 496/1541, loss: 0.02489759773015976 2023-01-23 14:41:12.622892: step: 498/1541, loss: 0.14071565866470337 2023-01-23 14:41:13.489880: step: 500/1541, loss: 0.12517638504505157 2023-01-23 14:41:14.384793: step: 502/1541, loss: 0.14398787915706635 2023-01-23 14:41:15.241502: step: 504/1541, loss: 0.013900361023843288 2023-01-23 14:41:16.105970: step: 506/1541, loss: 0.10433325171470642 2023-01-23 14:41:17.029578: step: 508/1541, loss: 0.05142971873283386 2023-01-23 14:41:17.936782: step: 510/1541, loss: 0.0056724874302744865 2023-01-23 14:41:18.890644: step: 512/1541, loss: 0.12578023970127106 2023-01-23 14:41:19.848082: step: 514/1541, loss: 0.5861780643463135 2023-01-23 14:41:20.776992: step: 516/1541, loss: 0.38147827982902527 2023-01-23 14:41:21.708982: step: 518/1541, loss: 0.03007921203970909 2023-01-23 14:41:22.619451: step: 520/1541, loss: 0.015804652124643326 2023-01-23 14:41:23.499361: step: 522/1541, loss: 0.0036809309385716915 2023-01-23 14:41:24.372890: step: 524/1541, loss: 0.04410224407911301 2023-01-23 14:41:25.260471: step: 526/1541, loss: 0.004692526534199715 2023-01-23 14:41:26.193924: step: 528/1541, loss: 0.16155801713466644 2023-01-23 14:41:27.067186: step: 530/1541, loss: 0.06669943779706955 2023-01-23 14:41:27.940775: step: 532/1541, loss: 0.005444600246846676 2023-01-23 14:41:28.827389: step: 534/1541, loss: 0.007384049240499735 2023-01-23 14:41:29.696750: step: 536/1541, loss: 0.01934458315372467 2023-01-23 14:41:30.585615: step: 538/1541, loss: 0.2897636592388153 2023-01-23 14:41:31.486749: step: 540/1541, loss: 0.03951757028698921 2023-01-23 14:41:32.423196: step: 542/1541, loss: 0.04051179438829422 2023-01-23 14:41:33.327202: step: 544/1541, loss: 0.14553627371788025 2023-01-23 14:41:34.181937: step: 546/1541, loss: 0.23466765880584717 2023-01-23 14:41:35.112972: step: 548/1541, loss: 0.005334379151463509 2023-01-23 14:41:35.992531: step: 550/1541, loss: 0.07879555970430374 2023-01-23 14:41:36.893632: step: 552/1541, loss: 0.20764796435832977 2023-01-23 14:41:37.760582: step: 554/1541, loss: 0.0010080926585942507 2023-01-23 14:41:38.653400: step: 556/1541, loss: 0.06138166785240173 2023-01-23 14:41:39.503432: step: 558/1541, loss: 0.03010406345129013 2023-01-23 14:41:40.361365: step: 560/1541, loss: 0.05997081100940704 2023-01-23 14:41:41.299157: step: 562/1541, loss: 0.006623000372201204 2023-01-23 14:41:42.157976: step: 564/1541, loss: 0.23504993319511414 2023-01-23 14:41:43.056207: step: 566/1541, loss: 0.4965655207633972 2023-01-23 14:41:43.936707: step: 568/1541, loss: 0.09458818286657333 2023-01-23 14:41:44.813505: step: 570/1541, loss: 0.15926529467105865 2023-01-23 14:41:45.715661: step: 572/1541, loss: 0.018894100561738014 2023-01-23 14:41:46.598322: step: 574/1541, loss: 0.14407385885715485 2023-01-23 14:41:47.516053: step: 576/1541, loss: 0.06822721660137177 2023-01-23 14:41:48.384845: step: 578/1541, loss: 0.004849599674344063 2023-01-23 14:41:49.253414: step: 580/1541, loss: 0.3277113437652588 2023-01-23 14:41:50.109463: step: 582/1541, loss: 0.2789633870124817 2023-01-23 14:41:51.054735: step: 584/1541, loss: 0.021918268874287605 2023-01-23 14:41:52.035403: step: 586/1541, loss: 0.0009404286392964423 2023-01-23 14:41:52.941037: step: 588/1541, loss: 0.004680080804973841 2023-01-23 14:41:53.809058: step: 590/1541, loss: 0.003088371828198433 2023-01-23 14:41:54.710007: step: 592/1541, loss: 0.10124772042036057 2023-01-23 14:41:55.621508: step: 594/1541, loss: 0.002064306987449527 2023-01-23 14:41:56.522482: step: 596/1541, loss: 0.007187586277723312 2023-01-23 14:41:57.440798: step: 598/1541, loss: 0.05923362076282501 2023-01-23 14:41:58.357899: step: 600/1541, loss: 0.02539757639169693 2023-01-23 14:41:59.296419: step: 602/1541, loss: 0.02842211350798607 2023-01-23 14:42:00.219193: step: 604/1541, loss: 0.05625755339860916 2023-01-23 14:42:01.163338: step: 606/1541, loss: 0.04957851395010948 2023-01-23 14:42:02.026823: step: 608/1541, loss: 0.047343406826257706 2023-01-23 14:42:03.042525: step: 610/1541, loss: 0.0046416157856583595 2023-01-23 14:42:03.982681: step: 612/1541, loss: 0.0006079301820136607 2023-01-23 14:42:04.935268: step: 614/1541, loss: 0.06006580963730812 2023-01-23 14:42:05.857292: step: 616/1541, loss: 0.039360351860523224 2023-01-23 14:42:06.751982: step: 618/1541, loss: 0.06942006200551987 2023-01-23 14:42:07.664018: step: 620/1541, loss: 0.00030586926732212305 2023-01-23 14:42:08.592078: step: 622/1541, loss: 0.03439493849873543 2023-01-23 14:42:09.523539: step: 624/1541, loss: 0.09903620928525925 2023-01-23 14:42:10.413820: step: 626/1541, loss: 0.00849018432199955 2023-01-23 14:42:11.301696: step: 628/1541, loss: 0.01037557516247034 2023-01-23 14:42:12.179057: step: 630/1541, loss: 0.0007100197253748775 2023-01-23 14:42:13.111811: step: 632/1541, loss: 0.028199009597301483 2023-01-23 14:42:14.039990: step: 634/1541, loss: 0.0592581070959568 2023-01-23 14:42:14.903938: step: 636/1541, loss: 0.05261462926864624 2023-01-23 14:42:15.832215: step: 638/1541, loss: 0.07660619914531708 2023-01-23 14:42:16.730354: step: 640/1541, loss: 0.0010648472234606743 2023-01-23 14:42:17.602078: step: 642/1541, loss: 0.0004494223976507783 2023-01-23 14:42:18.501745: step: 644/1541, loss: 0.0877455547451973 2023-01-23 14:42:19.399680: step: 646/1541, loss: 0.15858852863311768 2023-01-23 14:42:20.294719: step: 648/1541, loss: 0.9587832093238831 2023-01-23 14:42:21.167531: step: 650/1541, loss: 0.04466380178928375 2023-01-23 14:42:22.034878: step: 652/1541, loss: 0.08955695480108261 2023-01-23 14:42:22.990250: step: 654/1541, loss: 0.0028186405543237925 2023-01-23 14:42:23.852074: step: 656/1541, loss: 0.42639899253845215 2023-01-23 14:42:24.747306: step: 658/1541, loss: 0.29322442412376404 2023-01-23 14:42:25.669671: step: 660/1541, loss: 0.029442209750413895 2023-01-23 14:42:26.583502: step: 662/1541, loss: 0.06645262986421585 2023-01-23 14:42:27.454581: step: 664/1541, loss: 0.14380541443824768 2023-01-23 14:42:28.458108: step: 666/1541, loss: 0.019669324159622192 2023-01-23 14:42:29.379939: step: 668/1541, loss: 0.05971710383892059 2023-01-23 14:42:30.314838: step: 670/1541, loss: 0.007125318981707096 2023-01-23 14:42:31.181488: step: 672/1541, loss: 0.019385986030101776 2023-01-23 14:42:32.124252: step: 674/1541, loss: 0.008215161971747875 2023-01-23 14:42:32.996114: step: 676/1541, loss: 0.02354647032916546 2023-01-23 14:42:33.871349: step: 678/1541, loss: 0.33324742317199707 2023-01-23 14:42:34.833690: step: 680/1541, loss: 0.024108441546559334 2023-01-23 14:42:35.715329: step: 682/1541, loss: 0.009441704489290714 2023-01-23 14:42:36.609633: step: 684/1541, loss: 0.06070411950349808 2023-01-23 14:42:37.563304: step: 686/1541, loss: 0.23408706486225128 2023-01-23 14:42:38.435384: step: 688/1541, loss: 0.02616354636847973 2023-01-23 14:42:39.326235: step: 690/1541, loss: 0.05710308626294136 2023-01-23 14:42:40.220683: step: 692/1541, loss: 0.1312725841999054 2023-01-23 14:42:41.113276: step: 694/1541, loss: 0.5262748003005981 2023-01-23 14:42:42.014650: step: 696/1541, loss: 0.0011424930999055505 2023-01-23 14:42:42.880920: step: 698/1541, loss: 0.73890620470047 2023-01-23 14:42:43.766892: step: 700/1541, loss: 0.01071170438081026 2023-01-23 14:42:44.705068: step: 702/1541, loss: 0.00632027443498373 2023-01-23 14:42:45.599788: step: 704/1541, loss: 0.014456070959568024 2023-01-23 14:42:46.498760: step: 706/1541, loss: 0.017933711409568787 2023-01-23 14:42:47.434161: step: 708/1541, loss: 0.06508679687976837 2023-01-23 14:42:48.293640: step: 710/1541, loss: 0.06700970977544785 2023-01-23 14:42:49.181898: step: 712/1541, loss: 0.1277466118335724 2023-01-23 14:42:50.100437: step: 714/1541, loss: 1.3822364807128906 2023-01-23 14:42:50.946233: step: 716/1541, loss: 5.6123037211364135e-05 2023-01-23 14:42:51.840767: step: 718/1541, loss: 0.15546806156635284 2023-01-23 14:42:52.736758: step: 720/1541, loss: 0.019494887441396713 2023-01-23 14:42:53.689427: step: 722/1541, loss: 0.036943867802619934 2023-01-23 14:42:54.594702: step: 724/1541, loss: 0.03168697655200958 2023-01-23 14:42:55.536081: step: 726/1541, loss: 0.001250762608833611 2023-01-23 14:42:56.402023: step: 728/1541, loss: 0.0026153544895350933 2023-01-23 14:42:57.355864: step: 730/1541, loss: 0.007198788225650787 2023-01-23 14:42:58.266516: step: 732/1541, loss: 0.06686458736658096 2023-01-23 14:42:59.164773: step: 734/1541, loss: 0.009212433360517025 2023-01-23 14:43:00.035484: step: 736/1541, loss: 1.4676861763000488 2023-01-23 14:43:00.962892: step: 738/1541, loss: 0.0011023529805243015 2023-01-23 14:43:01.944389: step: 740/1541, loss: 0.27553918957710266 2023-01-23 14:43:02.827749: step: 742/1541, loss: 0.03542141243815422 2023-01-23 14:43:03.728532: step: 744/1541, loss: 0.490664005279541 2023-01-23 14:43:04.647623: step: 746/1541, loss: 0.05641352757811546 2023-01-23 14:43:05.551001: step: 748/1541, loss: 0.06679896265268326 2023-01-23 14:43:06.506440: step: 750/1541, loss: 0.016487719491124153 2023-01-23 14:43:07.377983: step: 752/1541, loss: 0.03214520215988159 2023-01-23 14:43:08.363980: step: 754/1541, loss: 0.029999814927577972 2023-01-23 14:43:09.237931: step: 756/1541, loss: 0.0033253722358494997 2023-01-23 14:43:10.114033: step: 758/1541, loss: 0.016895189881324768 2023-01-23 14:43:11.020667: step: 760/1541, loss: 0.006820939481258392 2023-01-23 14:43:11.944560: step: 762/1541, loss: 0.2609625458717346 2023-01-23 14:43:12.799781: step: 764/1541, loss: 0.0006816461682319641 2023-01-23 14:43:13.708430: step: 766/1541, loss: 0.3168528079986572 2023-01-23 14:43:14.687734: step: 768/1541, loss: 0.0023108262103050947 2023-01-23 14:43:15.554008: step: 770/1541, loss: 0.007917290553450584 2023-01-23 14:43:16.449315: step: 772/1541, loss: 0.0022426254581660032 2023-01-23 14:43:17.356904: step: 774/1541, loss: 0.016165675595402718 2023-01-23 14:43:18.260016: step: 776/1541, loss: 0.640424370765686 2023-01-23 14:43:19.144054: step: 778/1541, loss: 0.870622456073761 2023-01-23 14:43:20.110687: step: 780/1541, loss: 0.02107613906264305 2023-01-23 14:43:21.028570: step: 782/1541, loss: 0.004342002794146538 2023-01-23 14:43:21.904318: step: 784/1541, loss: 0.06987813115119934 2023-01-23 14:43:22.801125: step: 786/1541, loss: 0.0027466057799756527 2023-01-23 14:43:23.699690: step: 788/1541, loss: 0.002508896868675947 2023-01-23 14:43:24.580808: step: 790/1541, loss: 0.007863522507250309 2023-01-23 14:43:25.535483: step: 792/1541, loss: 0.003575985785573721 2023-01-23 14:43:26.424698: step: 794/1541, loss: 0.04993351548910141 2023-01-23 14:43:27.311238: step: 796/1541, loss: 0.13635052740573883 2023-01-23 14:43:28.214144: step: 798/1541, loss: 0.006371407303959131 2023-01-23 14:43:29.128150: step: 800/1541, loss: 0.013367211446166039 2023-01-23 14:43:30.051545: step: 802/1541, loss: 0.007547438610345125 2023-01-23 14:43:30.933907: step: 804/1541, loss: 0.011371343396604061 2023-01-23 14:43:31.889860: step: 806/1541, loss: 0.4447466731071472 2023-01-23 14:43:32.819442: step: 808/1541, loss: 0.0006271371385082603 2023-01-23 14:43:33.693259: step: 810/1541, loss: 0.03236799314618111 2023-01-23 14:43:34.556815: step: 812/1541, loss: 0.092569500207901 2023-01-23 14:43:35.428983: step: 814/1541, loss: 0.0003967431257478893 2023-01-23 14:43:36.278520: step: 816/1541, loss: 0.0002118261472787708 2023-01-23 14:43:37.161189: step: 818/1541, loss: 0.0004009085241705179 2023-01-23 14:43:38.056502: step: 820/1541, loss: 0.05508503317832947 2023-01-23 14:43:38.913166: step: 822/1541, loss: 0.0016504665836691856 2023-01-23 14:43:39.818622: step: 824/1541, loss: 0.060562506318092346 2023-01-23 14:43:40.666277: step: 826/1541, loss: 0.015387450344860554 2023-01-23 14:43:41.533955: step: 828/1541, loss: 0.052662722766399384 2023-01-23 14:43:42.456942: step: 830/1541, loss: 0.03240709751844406 2023-01-23 14:43:43.363210: step: 832/1541, loss: 0.0008119994308799505 2023-01-23 14:43:44.247095: step: 834/1541, loss: 0.040407873690128326 2023-01-23 14:43:45.182094: step: 836/1541, loss: 0.2825890779495239 2023-01-23 14:43:46.073802: step: 838/1541, loss: 0.01649344526231289 2023-01-23 14:43:46.977089: step: 840/1541, loss: 0.00024185799702536315 2023-01-23 14:43:47.850825: step: 842/1541, loss: 0.0005923142307437956 2023-01-23 14:43:48.814557: step: 844/1541, loss: 0.006587035022675991 2023-01-23 14:43:49.670959: step: 846/1541, loss: 0.0007753377431072295 2023-01-23 14:43:50.593171: step: 848/1541, loss: 0.09425446391105652 2023-01-23 14:43:51.616655: step: 850/1541, loss: 0.021543893963098526 2023-01-23 14:43:52.513628: step: 852/1541, loss: 0.0017273438861593604 2023-01-23 14:43:53.424775: step: 854/1541, loss: 0.08099022507667542 2023-01-23 14:43:54.316392: step: 856/1541, loss: 0.003372320905327797 2023-01-23 14:43:55.211958: step: 858/1541, loss: 0.18681523203849792 2023-01-23 14:43:56.150928: step: 860/1541, loss: 0.004759871866554022 2023-01-23 14:43:57.049347: step: 862/1541, loss: 0.2195926457643509 2023-01-23 14:43:57.935601: step: 864/1541, loss: 0.0015716676134616137 2023-01-23 14:43:58.848949: step: 866/1541, loss: 0.022521916776895523 2023-01-23 14:43:59.735723: step: 868/1541, loss: 0.006670196540653706 2023-01-23 14:44:00.608771: step: 870/1541, loss: 0.11968283355236053 2023-01-23 14:44:01.484542: step: 872/1541, loss: 0.01712563820183277 2023-01-23 14:44:02.391398: step: 874/1541, loss: 0.3646920919418335 2023-01-23 14:44:03.321527: step: 876/1541, loss: 0.20924197137355804 2023-01-23 14:44:04.174762: step: 878/1541, loss: 0.022025685757398605 2023-01-23 14:44:05.133435: step: 880/1541, loss: 0.6899424195289612 2023-01-23 14:44:06.015773: step: 882/1541, loss: 0.04794035106897354 2023-01-23 14:44:06.891550: step: 884/1541, loss: 0.017417510971426964 2023-01-23 14:44:07.798406: step: 886/1541, loss: 0.027620550245046616 2023-01-23 14:44:08.691096: step: 888/1541, loss: 0.000285780435660854 2023-01-23 14:44:09.585980: step: 890/1541, loss: 0.10738246887922287 2023-01-23 14:44:10.503144: step: 892/1541, loss: 0.0027960622683167458 2023-01-23 14:44:11.380609: step: 894/1541, loss: 0.6740610003471375 2023-01-23 14:44:12.323369: step: 896/1541, loss: 0.0009986620862036943 2023-01-23 14:44:13.216139: step: 898/1541, loss: 0.0028189995791763067 2023-01-23 14:44:14.121130: step: 900/1541, loss: 0.1893172264099121 2023-01-23 14:44:14.985748: step: 902/1541, loss: 0.11919453740119934 2023-01-23 14:44:15.847904: step: 904/1541, loss: 0.0007235939847305417 2023-01-23 14:44:16.723882: step: 906/1541, loss: 0.1563260853290558 2023-01-23 14:44:17.687475: step: 908/1541, loss: 0.13353680074214935 2023-01-23 14:44:18.564388: step: 910/1541, loss: 0.0956960916519165 2023-01-23 14:44:19.462674: step: 912/1541, loss: 0.16833360493183136 2023-01-23 14:44:20.383996: step: 914/1541, loss: 0.002708792220801115 2023-01-23 14:44:21.285854: step: 916/1541, loss: 0.004066662397235632 2023-01-23 14:44:22.169341: step: 918/1541, loss: 0.06260163336992264 2023-01-23 14:44:23.040909: step: 920/1541, loss: 0.009323719888925552 2023-01-23 14:44:23.904914: step: 922/1541, loss: 0.20126613974571228 2023-01-23 14:44:24.839012: step: 924/1541, loss: 0.03919665515422821 2023-01-23 14:44:25.695670: step: 926/1541, loss: 0.35468631982803345 2023-01-23 14:44:26.572892: step: 928/1541, loss: 0.006728358566761017 2023-01-23 14:44:27.513057: step: 930/1541, loss: 0.06905370950698853 2023-01-23 14:44:28.451977: step: 932/1541, loss: 0.014615154825150967 2023-01-23 14:44:29.401147: step: 934/1541, loss: 0.0012997366720810533 2023-01-23 14:44:30.300362: step: 936/1541, loss: 0.0007988314027898014 2023-01-23 14:44:31.174926: step: 938/1541, loss: 0.3519473969936371 2023-01-23 14:44:32.065608: step: 940/1541, loss: 0.003542701480910182 2023-01-23 14:44:32.933838: step: 942/1541, loss: 0.07243490219116211 2023-01-23 14:44:33.850058: step: 944/1541, loss: 0.0061518470756709576 2023-01-23 14:44:34.723237: step: 946/1541, loss: 0.018462156876921654 2023-01-23 14:44:35.614947: step: 948/1541, loss: 0.005034018307924271 2023-01-23 14:44:36.547904: step: 950/1541, loss: 0.11938820034265518 2023-01-23 14:44:37.445120: step: 952/1541, loss: 0.001634112442843616 2023-01-23 14:44:38.448676: step: 954/1541, loss: 0.01712021976709366 2023-01-23 14:44:39.326146: step: 956/1541, loss: 0.07056924700737 2023-01-23 14:44:40.234102: step: 958/1541, loss: 0.021891415119171143 2023-01-23 14:44:41.184675: step: 960/1541, loss: 0.3988152742385864 2023-01-23 14:44:42.107056: step: 962/1541, loss: 0.03608271852135658 2023-01-23 14:44:43.023632: step: 964/1541, loss: 0.0033263652585446835 2023-01-23 14:44:43.933398: step: 966/1541, loss: 0.06177534535527229 2023-01-23 14:44:44.874715: step: 968/1541, loss: 0.010552933439612389 2023-01-23 14:44:45.787519: step: 970/1541, loss: 0.0015408325707539916 2023-01-23 14:44:46.659739: step: 972/1541, loss: 0.14387954771518707 2023-01-23 14:44:47.589730: step: 974/1541, loss: 0.0697750449180603 2023-01-23 14:44:48.530548: step: 976/1541, loss: 0.12978903949260712 2023-01-23 14:44:49.458911: step: 978/1541, loss: 0.010772795416414738 2023-01-23 14:44:50.379092: step: 980/1541, loss: 0.006253973580896854 2023-01-23 14:44:51.244730: step: 982/1541, loss: 0.0027984464541077614 2023-01-23 14:44:52.160076: step: 984/1541, loss: 0.0350869856774807 2023-01-23 14:44:53.060889: step: 986/1541, loss: 0.06528948992490768 2023-01-23 14:44:53.958040: step: 988/1541, loss: 0.0019711146596819162 2023-01-23 14:44:54.846930: step: 990/1541, loss: 0.0030765472911298275 2023-01-23 14:44:55.744369: step: 992/1541, loss: 0.22315239906311035 2023-01-23 14:44:56.643594: step: 994/1541, loss: 0.002833254635334015 2023-01-23 14:44:57.539332: step: 996/1541, loss: 0.18494470417499542 2023-01-23 14:44:58.482951: step: 998/1541, loss: 0.007933597080409527 2023-01-23 14:44:59.433036: step: 1000/1541, loss: 0.05812647566199303 2023-01-23 14:45:00.311481: step: 1002/1541, loss: 0.1801704615354538 2023-01-23 14:45:01.197036: step: 1004/1541, loss: 0.016861261799931526 2023-01-23 14:45:02.123555: step: 1006/1541, loss: 0.6448569893836975 2023-01-23 14:45:03.008767: step: 1008/1541, loss: 0.06784265488386154 2023-01-23 14:45:03.958153: step: 1010/1541, loss: 0.3081810772418976 2023-01-23 14:45:04.822774: step: 1012/1541, loss: 0.0032435795292258263 2023-01-23 14:45:05.676257: step: 1014/1541, loss: 0.011882035993039608 2023-01-23 14:45:06.586907: step: 1016/1541, loss: 0.0005583363818004727 2023-01-23 14:45:07.496200: step: 1018/1541, loss: 0.0015417687827721238 2023-01-23 14:45:08.389305: step: 1020/1541, loss: 0.06631942838430405 2023-01-23 14:45:09.288175: step: 1022/1541, loss: 0.056812088936567307 2023-01-23 14:45:10.192438: step: 1024/1541, loss: 0.09721461683511734 2023-01-23 14:45:11.128030: step: 1026/1541, loss: 0.25383710861206055 2023-01-23 14:45:12.067181: step: 1028/1541, loss: 0.4103975296020508 2023-01-23 14:45:13.029849: step: 1030/1541, loss: 0.1361197531223297 2023-01-23 14:45:13.953288: step: 1032/1541, loss: 0.378515362739563 2023-01-23 14:45:14.826473: step: 1034/1541, loss: 0.7550113797187805 2023-01-23 14:45:15.707640: step: 1036/1541, loss: 0.00100473512429744 2023-01-23 14:45:16.627318: step: 1038/1541, loss: 0.10017115622758865 2023-01-23 14:45:17.544306: step: 1040/1541, loss: 0.2531064450740814 2023-01-23 14:45:18.428396: step: 1042/1541, loss: 0.014724460430443287 2023-01-23 14:45:19.314050: step: 1044/1541, loss: 0.13002067804336548 2023-01-23 14:45:20.217546: step: 1046/1541, loss: 0.042928729206323624 2023-01-23 14:45:21.106541: step: 1048/1541, loss: 0.01904451660811901 2023-01-23 14:45:21.997230: step: 1050/1541, loss: 0.013640697114169598 2023-01-23 14:45:22.880110: step: 1052/1541, loss: 0.00769293587654829 2023-01-23 14:45:23.796114: step: 1054/1541, loss: 0.01853465847671032 2023-01-23 14:45:24.693121: step: 1056/1541, loss: 0.1091749519109726 2023-01-23 14:45:25.569343: step: 1058/1541, loss: 0.013643360696732998 2023-01-23 14:45:26.446367: step: 1060/1541, loss: 0.11145784705877304 2023-01-23 14:45:27.322702: step: 1062/1541, loss: 0.37451857328414917 2023-01-23 14:45:28.274687: step: 1064/1541, loss: 0.001077757216989994 2023-01-23 14:45:29.163415: step: 1066/1541, loss: 0.006971501745283604 2023-01-23 14:45:30.029512: step: 1068/1541, loss: 0.0005260758334770799 2023-01-23 14:45:30.952680: step: 1070/1541, loss: 0.007715362124145031 2023-01-23 14:45:31.891312: step: 1072/1541, loss: 0.11302333325147629 2023-01-23 14:45:32.856100: step: 1074/1541, loss: 0.11845434457063675 2023-01-23 14:45:33.734370: step: 1076/1541, loss: 0.23462693393230438 2023-01-23 14:45:34.638482: step: 1078/1541, loss: 0.000884953944478184 2023-01-23 14:45:35.563635: step: 1080/1541, loss: 0.036105573177337646 2023-01-23 14:45:36.515424: step: 1082/1541, loss: 0.0041875168681144714 2023-01-23 14:45:37.410070: step: 1084/1541, loss: 0.008354298770427704 2023-01-23 14:45:38.308668: step: 1086/1541, loss: 0.009887631051242352 2023-01-23 14:45:39.187393: step: 1088/1541, loss: 0.021708695217967033 2023-01-23 14:45:40.090106: step: 1090/1541, loss: 0.00019776762928813696 2023-01-23 14:45:41.018197: step: 1092/1541, loss: 0.010575405322015285 2023-01-23 14:45:41.906114: step: 1094/1541, loss: 0.001133573125116527 2023-01-23 14:45:42.833660: step: 1096/1541, loss: 0.1384424865245819 2023-01-23 14:45:43.731924: step: 1098/1541, loss: 0.007971018552780151 2023-01-23 14:45:44.658671: step: 1100/1541, loss: 0.0002379644865868613 2023-01-23 14:45:45.574717: step: 1102/1541, loss: 1.8613733053207397 2023-01-23 14:45:46.463068: step: 1104/1541, loss: 0.005740402732044458 2023-01-23 14:45:47.451869: step: 1106/1541, loss: 0.000997438095510006 2023-01-23 14:45:48.326987: step: 1108/1541, loss: 0.05495848506689072 2023-01-23 14:45:49.211849: step: 1110/1541, loss: 0.1066826730966568 2023-01-23 14:45:50.135758: step: 1112/1541, loss: 0.0010062790242955089 2023-01-23 14:45:51.004235: step: 1114/1541, loss: 0.0012903637252748013 2023-01-23 14:45:51.904968: step: 1116/1541, loss: 0.22878439724445343 2023-01-23 14:45:52.822150: step: 1118/1541, loss: 0.06889858841896057 2023-01-23 14:45:53.779085: step: 1120/1541, loss: 0.00046263146214187145 2023-01-23 14:45:54.709530: step: 1122/1541, loss: 0.11123943328857422 2023-01-23 14:45:55.607309: step: 1124/1541, loss: 0.09865579009056091 2023-01-23 14:45:56.548895: step: 1126/1541, loss: 0.5275824069976807 2023-01-23 14:45:57.433969: step: 1128/1541, loss: 0.27018266916275024 2023-01-23 14:45:58.321936: step: 1130/1541, loss: 0.02761758118867874 2023-01-23 14:45:59.262721: step: 1132/1541, loss: 0.02760159969329834 2023-01-23 14:46:00.182325: step: 1134/1541, loss: 0.001579919713549316 2023-01-23 14:46:01.073724: step: 1136/1541, loss: 0.00981958582997322 2023-01-23 14:46:02.031353: step: 1138/1541, loss: 0.1164517030119896 2023-01-23 14:46:02.912466: step: 1140/1541, loss: 0.004976809956133366 2023-01-23 14:46:03.846943: step: 1142/1541, loss: 0.027907636016607285 2023-01-23 14:46:04.730985: step: 1144/1541, loss: 0.0341205894947052 2023-01-23 14:46:05.684540: step: 1146/1541, loss: 0.0010095163015648723 2023-01-23 14:46:06.585507: step: 1148/1541, loss: 0.0398908294737339 2023-01-23 14:46:07.446094: step: 1150/1541, loss: 2.3787877559661865 2023-01-23 14:46:08.335131: step: 1152/1541, loss: 0.0020601539872586727 2023-01-23 14:46:09.295702: step: 1154/1541, loss: 0.035838931798934937 2023-01-23 14:46:10.241576: step: 1156/1541, loss: 0.0002067982713924721 2023-01-23 14:46:11.133643: step: 1158/1541, loss: 0.04135212302207947 2023-01-23 14:46:12.006897: step: 1160/1541, loss: 0.04619007185101509 2023-01-23 14:46:12.911768: step: 1162/1541, loss: 0.0007084602839313447 2023-01-23 14:46:13.797956: step: 1164/1541, loss: 0.07134788483381271 2023-01-23 14:46:14.690372: step: 1166/1541, loss: 0.3509811758995056 2023-01-23 14:46:15.646797: step: 1168/1541, loss: 0.006608231924474239 2023-01-23 14:46:16.516820: step: 1170/1541, loss: 0.04234747216105461 2023-01-23 14:46:17.393756: step: 1172/1541, loss: 0.014836229383945465 2023-01-23 14:46:18.261367: step: 1174/1541, loss: 0.18452802300453186 2023-01-23 14:46:19.163977: step: 1176/1541, loss: 0.06082315370440483 2023-01-23 14:46:20.103353: step: 1178/1541, loss: 0.003860452212393284 2023-01-23 14:46:20.977876: step: 1180/1541, loss: 0.0003005138714797795 2023-01-23 14:46:21.867035: step: 1182/1541, loss: 0.27837565541267395 2023-01-23 14:46:22.817711: step: 1184/1541, loss: 0.003438810119405389 2023-01-23 14:46:23.699069: step: 1186/1541, loss: 0.2344909906387329 2023-01-23 14:46:24.625878: step: 1188/1541, loss: 0.33760589361190796 2023-01-23 14:46:25.534805: step: 1190/1541, loss: 0.017557945102453232 2023-01-23 14:46:26.445240: step: 1192/1541, loss: 0.43004167079925537 2023-01-23 14:46:27.351694: step: 1194/1541, loss: 0.19170795381069183 2023-01-23 14:46:28.287713: step: 1196/1541, loss: 0.4369644224643707 2023-01-23 14:46:29.187920: step: 1198/1541, loss: 0.29583749175071716 2023-01-23 14:46:30.080960: step: 1200/1541, loss: 0.0026181875728070736 2023-01-23 14:46:30.968424: step: 1202/1541, loss: 0.0021711315494030714 2023-01-23 14:46:31.875559: step: 1204/1541, loss: 0.01309957355260849 2023-01-23 14:46:32.781070: step: 1206/1541, loss: 0.05990570783615112 2023-01-23 14:46:33.724022: step: 1208/1541, loss: 0.09827933460474014 2023-01-23 14:46:34.615061: step: 1210/1541, loss: 0.5247399210929871 2023-01-23 14:46:35.542839: step: 1212/1541, loss: 0.02455936372280121 2023-01-23 14:46:36.464028: step: 1214/1541, loss: 0.1377803087234497 2023-01-23 14:46:37.352022: step: 1216/1541, loss: 0.1392533928155899 2023-01-23 14:46:38.260243: step: 1218/1541, loss: 0.02635376900434494 2023-01-23 14:46:39.146097: step: 1220/1541, loss: 0.004166792146861553 2023-01-23 14:46:40.065045: step: 1222/1541, loss: 0.26795899868011475 2023-01-23 14:46:40.957516: step: 1224/1541, loss: 0.026727069169282913 2023-01-23 14:46:41.847318: step: 1226/1541, loss: 0.09503374993801117 2023-01-23 14:46:42.848234: step: 1228/1541, loss: 0.1098235622048378 2023-01-23 14:46:43.749650: step: 1230/1541, loss: 0.001223036553710699 2023-01-23 14:46:44.658673: step: 1232/1541, loss: 0.009864171035587788 2023-01-23 14:46:45.580921: step: 1234/1541, loss: 0.26308757066726685 2023-01-23 14:46:46.551992: step: 1236/1541, loss: 0.16836585104465485 2023-01-23 14:46:47.474496: step: 1238/1541, loss: 0.004320581443607807 2023-01-23 14:46:48.375447: step: 1240/1541, loss: 0.03916039690375328 2023-01-23 14:46:49.257170: step: 1242/1541, loss: 0.06983067095279694 2023-01-23 14:46:50.108200: step: 1244/1541, loss: 0.0009660836076363921 2023-01-23 14:46:51.005754: step: 1246/1541, loss: 0.0011893545743077993 2023-01-23 14:46:51.890343: step: 1248/1541, loss: 0.07290101051330566 2023-01-23 14:46:52.832143: step: 1250/1541, loss: 0.00411789957433939 2023-01-23 14:46:53.698436: step: 1252/1541, loss: 0.04125020280480385 2023-01-23 14:46:54.584059: step: 1254/1541, loss: 0.10714036226272583 2023-01-23 14:46:55.464300: step: 1256/1541, loss: 0.06908870488405228 2023-01-23 14:46:56.331830: step: 1258/1541, loss: 0.08166728168725967 2023-01-23 14:46:57.277714: step: 1260/1541, loss: 0.014350762590765953 2023-01-23 14:46:58.174379: step: 1262/1541, loss: 0.3372860550880432 2023-01-23 14:46:59.067701: step: 1264/1541, loss: 0.000706991646438837 2023-01-23 14:46:59.915324: step: 1266/1541, loss: 0.20477057993412018 2023-01-23 14:47:00.841847: step: 1268/1541, loss: 0.008610039949417114 2023-01-23 14:47:01.758280: step: 1270/1541, loss: 0.002034042729064822 2023-01-23 14:47:02.674885: step: 1272/1541, loss: 0.02523896098136902 2023-01-23 14:47:03.545879: step: 1274/1541, loss: 0.24021616578102112 2023-01-23 14:47:04.422011: step: 1276/1541, loss: 0.15273793041706085 2023-01-23 14:47:05.320101: step: 1278/1541, loss: 0.09683007746934891 2023-01-23 14:47:06.198774: step: 1280/1541, loss: 0.012290974147617817 2023-01-23 14:47:07.070725: step: 1282/1541, loss: 0.004062030464410782 2023-01-23 14:47:07.998878: step: 1284/1541, loss: 0.007765873335301876 2023-01-23 14:47:08.948278: step: 1286/1541, loss: 0.15516836941242218 2023-01-23 14:47:09.833753: step: 1288/1541, loss: 0.04479031264781952 2023-01-23 14:47:10.789968: step: 1290/1541, loss: 0.011888336390256882 2023-01-23 14:47:11.685058: step: 1292/1541, loss: 0.05089859664440155 2023-01-23 14:47:12.563125: step: 1294/1541, loss: 0.04382079094648361 2023-01-23 14:47:13.439431: step: 1296/1541, loss: 0.003704225178807974 2023-01-23 14:47:14.328924: step: 1298/1541, loss: 0.03437979146838188 2023-01-23 14:47:15.236598: step: 1300/1541, loss: 0.04564036428928375 2023-01-23 14:47:16.146360: step: 1302/1541, loss: 0.20789045095443726 2023-01-23 14:47:17.078722: step: 1304/1541, loss: 0.000624755397439003 2023-01-23 14:47:18.065553: step: 1306/1541, loss: 0.003347643418237567 2023-01-23 14:47:18.960069: step: 1308/1541, loss: 0.03604603186249733 2023-01-23 14:47:19.868987: step: 1310/1541, loss: 0.012337636202573776 2023-01-23 14:47:20.784545: step: 1312/1541, loss: 0.1705573946237564 2023-01-23 14:47:21.676508: step: 1314/1541, loss: 0.22204193472862244 2023-01-23 14:47:22.573400: step: 1316/1541, loss: 0.03567938506603241 2023-01-23 14:47:23.425330: step: 1318/1541, loss: 0.016535550355911255 2023-01-23 14:47:24.328775: step: 1320/1541, loss: 0.0036906329914927483 2023-01-23 14:47:25.173925: step: 1322/1541, loss: 0.018793856725096703 2023-01-23 14:47:26.054057: step: 1324/1541, loss: 0.1644763946533203 2023-01-23 14:47:26.940974: step: 1326/1541, loss: 0.0204005166888237 2023-01-23 14:47:27.807927: step: 1328/1541, loss: 0.01974393241107464 2023-01-23 14:47:28.682169: step: 1330/1541, loss: 0.0006325810099951923 2023-01-23 14:47:29.567568: step: 1332/1541, loss: 0.0021455292589962482 2023-01-23 14:47:30.455275: step: 1334/1541, loss: 0.003250367706641555 2023-01-23 14:47:31.367690: step: 1336/1541, loss: 0.00030772859463468194 2023-01-23 14:47:32.286647: step: 1338/1541, loss: 0.0010754127288237214 2023-01-23 14:47:33.150149: step: 1340/1541, loss: 0.07071628421545029 2023-01-23 14:47:34.055461: step: 1342/1541, loss: 0.0030960673466324806 2023-01-23 14:47:34.932688: step: 1344/1541, loss: 0.37172457575798035 2023-01-23 14:47:35.772106: step: 1346/1541, loss: 0.0004312206001486629 2023-01-23 14:47:36.688586: step: 1348/1541, loss: 0.13775300979614258 2023-01-23 14:47:37.606592: step: 1350/1541, loss: 0.01873999834060669 2023-01-23 14:47:38.487379: step: 1352/1541, loss: 0.21708905696868896 2023-01-23 14:47:39.370642: step: 1354/1541, loss: 0.19189612567424774 2023-01-23 14:47:40.273771: step: 1356/1541, loss: 0.1937635838985443 2023-01-23 14:47:41.195226: step: 1358/1541, loss: 0.05968308821320534 2023-01-23 14:47:42.114665: step: 1360/1541, loss: 1.9645276552182622e-05 2023-01-23 14:47:43.002607: step: 1362/1541, loss: 0.0015897693810984492 2023-01-23 14:47:43.891302: step: 1364/1541, loss: 0.3937607705593109 2023-01-23 14:47:44.793897: step: 1366/1541, loss: 0.2823716104030609 2023-01-23 14:47:45.691915: step: 1368/1541, loss: 0.003421990666538477 2023-01-23 14:47:46.586246: step: 1370/1541, loss: 0.01940634660422802 2023-01-23 14:47:47.490442: step: 1372/1541, loss: 0.003082838375121355 2023-01-23 14:47:48.367345: step: 1374/1541, loss: 0.09530489891767502 2023-01-23 14:47:49.229422: step: 1376/1541, loss: 0.002472632797434926 2023-01-23 14:47:50.104126: step: 1378/1541, loss: 0.028276236727833748 2023-01-23 14:47:51.002741: step: 1380/1541, loss: 0.009498176164925098 2023-01-23 14:47:51.945926: step: 1382/1541, loss: 0.002203900832682848 2023-01-23 14:47:52.841352: step: 1384/1541, loss: 0.012106296606361866 2023-01-23 14:47:53.750559: step: 1386/1541, loss: 0.019786641001701355 2023-01-23 14:47:54.606085: step: 1388/1541, loss: 0.07924973964691162 2023-01-23 14:47:55.493066: step: 1390/1541, loss: 0.08808793872594833 2023-01-23 14:47:56.398163: step: 1392/1541, loss: 0.022862032055854797 2023-01-23 14:47:57.296587: step: 1394/1541, loss: 0.3864498734474182 2023-01-23 14:47:58.222626: step: 1396/1541, loss: 0.08356665819883347 2023-01-23 14:47:59.128352: step: 1398/1541, loss: 0.003879899624735117 2023-01-23 14:48:00.020952: step: 1400/1541, loss: 0.015206445939838886 2023-01-23 14:48:00.944859: step: 1402/1541, loss: 0.00028469867538660765 2023-01-23 14:48:01.808349: step: 1404/1541, loss: 0.03616965562105179 2023-01-23 14:48:02.753141: step: 1406/1541, loss: 0.1893312633037567 2023-01-23 14:48:03.625152: step: 1408/1541, loss: 0.04376664012670517 2023-01-23 14:48:04.484094: step: 1410/1541, loss: 0.0027151890099048615 2023-01-23 14:48:05.362416: step: 1412/1541, loss: 0.0016551068983972073 2023-01-23 14:48:06.273736: step: 1414/1541, loss: 0.010086707770824432 2023-01-23 14:48:07.134718: step: 1416/1541, loss: 0.00893520936369896 2023-01-23 14:48:08.056168: step: 1418/1541, loss: 0.03690318763256073 2023-01-23 14:48:08.924997: step: 1420/1541, loss: 0.10676746815443039 2023-01-23 14:48:09.815152: step: 1422/1541, loss: 0.0012683288659900427 2023-01-23 14:48:10.769560: step: 1424/1541, loss: 0.004302429035305977 2023-01-23 14:48:11.654702: step: 1426/1541, loss: 0.005492726340889931 2023-01-23 14:48:12.530404: step: 1428/1541, loss: 0.02781672589480877 2023-01-23 14:48:13.508751: step: 1430/1541, loss: 0.0030110194347798824 2023-01-23 14:48:14.393453: step: 1432/1541, loss: 0.22784334421157837 2023-01-23 14:48:15.327424: step: 1434/1541, loss: 0.08427228033542633 2023-01-23 14:48:16.287852: step: 1436/1541, loss: 0.0711357444524765 2023-01-23 14:48:17.163043: step: 1438/1541, loss: 0.0025600106455385685 2023-01-23 14:48:18.088084: step: 1440/1541, loss: 0.0033854239154607058 2023-01-23 14:48:19.001987: step: 1442/1541, loss: 0.013592502102255821 2023-01-23 14:48:19.846122: step: 1444/1541, loss: 0.005832447670400143 2023-01-23 14:48:20.753237: step: 1446/1541, loss: 0.06222078576683998 2023-01-23 14:48:21.652038: step: 1448/1541, loss: 0.026987671852111816 2023-01-23 14:48:22.557314: step: 1450/1541, loss: 0.18588627874851227 2023-01-23 14:48:23.446553: step: 1452/1541, loss: 0.012807859107851982 2023-01-23 14:48:24.320161: step: 1454/1541, loss: 0.021585354581475258 2023-01-23 14:48:25.224222: step: 1456/1541, loss: 0.01896965131163597 2023-01-23 14:48:26.091061: step: 1458/1541, loss: 0.13768397271633148 2023-01-23 14:48:27.077604: step: 1460/1541, loss: 0.016559850424528122 2023-01-23 14:48:27.961381: step: 1462/1541, loss: 0.00639474019408226 2023-01-23 14:48:28.860026: step: 1464/1541, loss: 0.06771984696388245 2023-01-23 14:48:29.790948: step: 1466/1541, loss: 0.06905066967010498 2023-01-23 14:48:30.655121: step: 1468/1541, loss: 0.001040456467308104 2023-01-23 14:48:31.559282: step: 1470/1541, loss: 0.29023805260658264 2023-01-23 14:48:32.485033: step: 1472/1541, loss: 0.01301993615925312 2023-01-23 14:48:33.354353: step: 1474/1541, loss: 0.007571237627416849 2023-01-23 14:48:34.294784: step: 1476/1541, loss: 0.0005513157229870558 2023-01-23 14:48:35.224938: step: 1478/1541, loss: 0.9927527904510498 2023-01-23 14:48:36.080975: step: 1480/1541, loss: 0.12939222157001495 2023-01-23 14:48:36.993273: step: 1482/1541, loss: 0.05203109607100487 2023-01-23 14:48:37.942356: step: 1484/1541, loss: 0.020565371960401535 2023-01-23 14:48:38.787171: step: 1486/1541, loss: 0.39489346742630005 2023-01-23 14:48:39.697043: step: 1488/1541, loss: 0.030936339870095253 2023-01-23 14:48:40.681927: step: 1490/1541, loss: 0.001779000274837017 2023-01-23 14:48:41.658626: step: 1492/1541, loss: 0.16357234120368958 2023-01-23 14:48:42.533790: step: 1494/1541, loss: 0.03610648587346077 2023-01-23 14:48:43.418852: step: 1496/1541, loss: 0.009465430863201618 2023-01-23 14:48:44.307570: step: 1498/1541, loss: 0.0028360760770738125 2023-01-23 14:48:45.178350: step: 1500/1541, loss: 0.001883591990917921 2023-01-23 14:48:46.054714: step: 1502/1541, loss: 0.0006717046489939094 2023-01-23 14:48:46.936554: step: 1504/1541, loss: 0.0216484647244215 2023-01-23 14:48:47.841552: step: 1506/1541, loss: 0.0098259337246418 2023-01-23 14:48:48.764220: step: 1508/1541, loss: 0.17250744998455048 2023-01-23 14:48:49.643698: step: 1510/1541, loss: 0.0010147595312446356 2023-01-23 14:48:50.549650: step: 1512/1541, loss: 0.01374758593738079 2023-01-23 14:48:51.407684: step: 1514/1541, loss: 0.17558704316616058 2023-01-23 14:48:52.319330: step: 1516/1541, loss: 0.09056895971298218 2023-01-23 14:48:53.173509: step: 1518/1541, loss: 0.0329769067466259 2023-01-23 14:48:54.043349: step: 1520/1541, loss: 0.0015457860426977277 2023-01-23 14:48:54.960910: step: 1522/1541, loss: 0.030606979504227638 2023-01-23 14:48:55.899071: step: 1524/1541, loss: 0.002331498544663191 2023-01-23 14:48:56.745576: step: 1526/1541, loss: 0.3173818588256836 2023-01-23 14:48:57.668811: step: 1528/1541, loss: 0.0003281858516857028 2023-01-23 14:48:58.543574: step: 1530/1541, loss: 0.026849962770938873 2023-01-23 14:48:59.422185: step: 1532/1541, loss: 0.000329506496200338 2023-01-23 14:49:00.421164: step: 1534/1541, loss: 0.05903571471571922 2023-01-23 14:49:01.308034: step: 1536/1541, loss: 0.06206291913986206 2023-01-23 14:49:02.276005: step: 1538/1541, loss: 0.223111093044281 2023-01-23 14:49:03.156110: step: 1540/1541, loss: 0.14841191470623016 2023-01-23 14:49:04.100447: step: 1542/1541, loss: 0.5744365453720093 2023-01-23 14:49:05.005567: step: 1544/1541, loss: 0.03288063406944275 2023-01-23 14:49:05.911509: step: 1546/1541, loss: 0.21803143620491028 2023-01-23 14:49:06.812281: step: 1548/1541, loss: 0.03835736960172653 2023-01-23 14:49:07.787970: step: 1550/1541, loss: 0.04239998757839203 2023-01-23 14:49:08.643108: step: 1552/1541, loss: 0.0010736520634964108 2023-01-23 14:49:09.491967: step: 1554/1541, loss: 0.0003221218939870596 2023-01-23 14:49:10.345397: step: 1556/1541, loss: 0.3626253008842468 2023-01-23 14:49:11.210743: step: 1558/1541, loss: 0.031931839883327484 2023-01-23 14:49:12.133192: step: 1560/1541, loss: 0.012636199593544006 2023-01-23 14:49:13.042039: step: 1562/1541, loss: 0.0016646247822791338 2023-01-23 14:49:13.884136: step: 1564/1541, loss: 0.002214929321780801 2023-01-23 14:49:14.802071: step: 1566/1541, loss: 0.5511120557785034 2023-01-23 14:49:15.672495: step: 1568/1541, loss: 0.005362670868635178 2023-01-23 14:49:16.565140: step: 1570/1541, loss: 0.1905435025691986 2023-01-23 14:49:17.472110: step: 1572/1541, loss: 0.010941790416836739 2023-01-23 14:49:18.351248: step: 1574/1541, loss: 0.05524061620235443 2023-01-23 14:49:19.243623: step: 1576/1541, loss: 0.041050538420677185 2023-01-23 14:49:20.095009: step: 1578/1541, loss: 0.21629726886749268 2023-01-23 14:49:20.993964: step: 1580/1541, loss: 0.031291164457798004 2023-01-23 14:49:22.023762: step: 1582/1541, loss: 0.026461536064743996 2023-01-23 14:49:22.886533: step: 1584/1541, loss: 0.204750657081604 2023-01-23 14:49:23.840834: step: 1586/1541, loss: 0.07807651907205582 2023-01-23 14:49:24.709322: step: 1588/1541, loss: 0.0647551491856575 2023-01-23 14:49:25.568011: step: 1590/1541, loss: 0.09116615355014801 2023-01-23 14:49:26.485734: step: 1592/1541, loss: 0.022336499765515327 2023-01-23 14:49:27.363689: step: 1594/1541, loss: 0.03469115123152733 2023-01-23 14:49:28.229560: step: 1596/1541, loss: 0.0007201306871138513 2023-01-23 14:49:29.137671: step: 1598/1541, loss: 0.055135007947683334 2023-01-23 14:49:30.026944: step: 1600/1541, loss: 0.0692383423447609 2023-01-23 14:49:30.921787: step: 1602/1541, loss: 0.0625973641872406 2023-01-23 14:49:31.811242: step: 1604/1541, loss: 0.024959977716207504 2023-01-23 14:49:32.682180: step: 1606/1541, loss: 0.032890062779188156 2023-01-23 14:49:33.605742: step: 1608/1541, loss: 0.09884555637836456 2023-01-23 14:49:34.483658: step: 1610/1541, loss: 0.0969148650765419 2023-01-23 14:49:35.381314: step: 1612/1541, loss: 0.007946847938001156 2023-01-23 14:49:36.260839: step: 1614/1541, loss: 0.05177059769630432 2023-01-23 14:49:37.148791: step: 1616/1541, loss: 0.015207316726446152 2023-01-23 14:49:38.049256: step: 1618/1541, loss: 0.020005857571959496 2023-01-23 14:49:38.952347: step: 1620/1541, loss: 0.0035584960132837296 2023-01-23 14:49:39.863250: step: 1622/1541, loss: 0.08678564429283142 2023-01-23 14:49:40.729699: step: 1624/1541, loss: 0.09319931268692017 2023-01-23 14:49:41.627177: step: 1626/1541, loss: 0.03334442898631096 2023-01-23 14:49:42.481480: step: 1628/1541, loss: 0.010166899301111698 2023-01-23 14:49:43.338957: step: 1630/1541, loss: 0.004012858495116234 2023-01-23 14:49:44.288970: step: 1632/1541, loss: 0.015488368459045887 2023-01-23 14:49:45.186700: step: 1634/1541, loss: 0.07597093284130096 2023-01-23 14:49:46.036155: step: 1636/1541, loss: 0.0038149021565914154 2023-01-23 14:49:46.968774: step: 1638/1541, loss: 0.004706730600446463 2023-01-23 14:49:47.836255: step: 1640/1541, loss: 0.1797333061695099 2023-01-23 14:49:48.725195: step: 1642/1541, loss: 0.13150319457054138 2023-01-23 14:49:49.649985: step: 1644/1541, loss: 0.011532824486494064 2023-01-23 14:49:50.493332: step: 1646/1541, loss: 0.18402837216854095 2023-01-23 14:49:51.403114: step: 1648/1541, loss: 0.11886189877986908 2023-01-23 14:49:52.332595: step: 1650/1541, loss: 0.04146342724561691 2023-01-23 14:49:53.201706: step: 1652/1541, loss: 0.030516214668750763 2023-01-23 14:49:54.058253: step: 1654/1541, loss: 0.03845468908548355 2023-01-23 14:49:54.948450: step: 1656/1541, loss: 0.012098458595573902 2023-01-23 14:49:55.847101: step: 1658/1541, loss: 0.02152753435075283 2023-01-23 14:49:56.740203: step: 1660/1541, loss: 0.0031456563156098127 2023-01-23 14:49:57.610824: step: 1662/1541, loss: 0.003078354289755225 2023-01-23 14:49:58.537409: step: 1664/1541, loss: 0.035552263259887695 2023-01-23 14:49:59.432100: step: 1666/1541, loss: 0.2106686383485794 2023-01-23 14:50:00.388290: step: 1668/1541, loss: 0.13953004777431488 2023-01-23 14:50:01.308651: step: 1670/1541, loss: 0.005451608449220657 2023-01-23 14:50:02.224453: step: 1672/1541, loss: 0.20665082335472107 2023-01-23 14:50:03.103571: step: 1674/1541, loss: 0.06356720626354218 2023-01-23 14:50:03.996697: step: 1676/1541, loss: 0.0012475289404392242 2023-01-23 14:50:04.840034: step: 1678/1541, loss: 0.03711393103003502 2023-01-23 14:50:05.769180: step: 1680/1541, loss: 0.0023246025666594505 2023-01-23 14:50:06.678070: step: 1682/1541, loss: 0.03665519878268242 2023-01-23 14:50:07.581216: step: 1684/1541, loss: 0.039394766092300415 2023-01-23 14:50:08.511188: step: 1686/1541, loss: 0.4646840989589691 2023-01-23 14:50:09.423359: step: 1688/1541, loss: 0.005358805414289236 2023-01-23 14:50:10.332018: step: 1690/1541, loss: 0.06557773053646088 2023-01-23 14:50:11.248111: step: 1692/1541, loss: 0.02603251300752163 2023-01-23 14:50:12.160711: step: 1694/1541, loss: 0.7480476498603821 2023-01-23 14:50:13.086069: step: 1696/1541, loss: 0.031140215694904327 2023-01-23 14:50:13.937900: step: 1698/1541, loss: 0.07305939495563507 2023-01-23 14:50:14.822663: step: 1700/1541, loss: 0.03802083805203438 2023-01-23 14:50:15.725257: step: 1702/1541, loss: 0.05397149920463562 2023-01-23 14:50:16.664873: step: 1704/1541, loss: 0.17505884170532227 2023-01-23 14:50:17.564971: step: 1706/1541, loss: 0.005988899152725935 2023-01-23 14:50:18.440411: step: 1708/1541, loss: 0.17796939611434937 2023-01-23 14:50:19.325610: step: 1710/1541, loss: 0.003950885497033596 2023-01-23 14:50:20.252502: step: 1712/1541, loss: 0.16054315865039825 2023-01-23 14:50:21.135614: step: 1714/1541, loss: 0.011279609054327011 2023-01-23 14:50:22.041774: step: 1716/1541, loss: 0.0308078546077013 2023-01-23 14:50:22.923786: step: 1718/1541, loss: 0.15496380627155304 2023-01-23 14:50:23.816617: step: 1720/1541, loss: 0.14621502161026 2023-01-23 14:50:24.681105: step: 1722/1541, loss: 0.0051935454830527306 2023-01-23 14:50:25.579600: step: 1724/1541, loss: 0.0024432556238025427 2023-01-23 14:50:26.486631: step: 1726/1541, loss: 0.024657726287841797 2023-01-23 14:50:27.376921: step: 1728/1541, loss: 0.24136073887348175 2023-01-23 14:50:28.282436: step: 1730/1541, loss: 0.08895483613014221 2023-01-23 14:50:29.262433: step: 1732/1541, loss: 0.1203744038939476 2023-01-23 14:50:30.154920: step: 1734/1541, loss: 0.42851290106773376 2023-01-23 14:50:31.034360: step: 1736/1541, loss: 0.020415905863046646 2023-01-23 14:50:31.947937: step: 1738/1541, loss: 0.036647286266088486 2023-01-23 14:50:32.820374: step: 1740/1541, loss: 0.23894836008548737 2023-01-23 14:50:33.788625: step: 1742/1541, loss: 0.03854914382100105 2023-01-23 14:50:34.646658: step: 1744/1541, loss: 0.29589006304740906 2023-01-23 14:50:35.551443: step: 1746/1541, loss: 0.020108507946133614 2023-01-23 14:50:36.437449: step: 1748/1541, loss: 0.041325632482767105 2023-01-23 14:50:37.363610: step: 1750/1541, loss: 0.027386285364627838 2023-01-23 14:50:38.238860: step: 1752/1541, loss: 0.015076879411935806 2023-01-23 14:50:39.131936: step: 1754/1541, loss: 0.041810210794210434 2023-01-23 14:50:40.127250: step: 1756/1541, loss: 0.017985740676522255 2023-01-23 14:50:41.029840: step: 1758/1541, loss: 0.11727441847324371 2023-01-23 14:50:41.882568: step: 1760/1541, loss: 0.17510098218917847 2023-01-23 14:50:42.767913: step: 1762/1541, loss: 0.006104953121393919 2023-01-23 14:50:43.675292: step: 1764/1541, loss: 1.625666856765747 2023-01-23 14:50:44.563522: step: 1766/1541, loss: 0.009841574355959892 2023-01-23 14:50:45.446658: step: 1768/1541, loss: 0.0022718501277267933 2023-01-23 14:50:46.321783: step: 1770/1541, loss: 0.0007262626313604414 2023-01-23 14:50:47.182863: step: 1772/1541, loss: 0.0011360669741407037 2023-01-23 14:50:48.149097: step: 1774/1541, loss: 0.05727340281009674 2023-01-23 14:50:49.058991: step: 1776/1541, loss: 0.1293431967496872 2023-01-23 14:50:49.926512: step: 1778/1541, loss: 0.0030825382564216852 2023-01-23 14:50:50.823108: step: 1780/1541, loss: 0.01763301156461239 2023-01-23 14:50:51.725120: step: 1782/1541, loss: 0.6603602170944214 2023-01-23 14:50:52.598394: step: 1784/1541, loss: 0.023428376764059067 2023-01-23 14:50:53.461647: step: 1786/1541, loss: 0.0011869892477989197 2023-01-23 14:50:54.361902: step: 1788/1541, loss: 0.09617190808057785 2023-01-23 14:50:55.266861: step: 1790/1541, loss: 0.031136061996221542 2023-01-23 14:50:56.205548: step: 1792/1541, loss: 0.23787139356136322 2023-01-23 14:50:57.071608: step: 1794/1541, loss: 0.05026145651936531 2023-01-23 14:50:57.958484: step: 1796/1541, loss: 0.1882222592830658 2023-01-23 14:50:58.821864: step: 1798/1541, loss: 0.019409729167819023 2023-01-23 14:50:59.745185: step: 1800/1541, loss: 0.004949421621859074 2023-01-23 14:51:00.684248: step: 1802/1541, loss: 0.010287461802363396 2023-01-23 14:51:01.614012: step: 1804/1541, loss: 0.06094623729586601 2023-01-23 14:51:02.527429: step: 1806/1541, loss: 0.1848420649766922 2023-01-23 14:51:03.427131: step: 1808/1541, loss: 0.21814289689064026 2023-01-23 14:51:04.274567: step: 1810/1541, loss: 0.009520888328552246 2023-01-23 14:51:05.161531: step: 1812/1541, loss: 0.003254788462072611 2023-01-23 14:51:06.101973: step: 1814/1541, loss: 0.006226176396012306 2023-01-23 14:51:06.997329: step: 1816/1541, loss: 0.003113329876214266 2023-01-23 14:51:07.958810: step: 1818/1541, loss: 0.13420097529888153 2023-01-23 14:51:08.853390: step: 1820/1541, loss: 0.019286787137389183 2023-01-23 14:51:09.734583: step: 1822/1541, loss: 0.0047376216389238834 2023-01-23 14:51:10.621512: step: 1824/1541, loss: 0.001473941607400775 2023-01-23 14:51:11.519057: step: 1826/1541, loss: 0.13076886534690857 2023-01-23 14:51:12.384249: step: 1828/1541, loss: 0.0087595134973526 2023-01-23 14:51:13.259311: step: 1830/1541, loss: 0.2211627960205078 2023-01-23 14:51:14.201192: step: 1832/1541, loss: 0.15801189839839935 2023-01-23 14:51:15.109041: step: 1834/1541, loss: 0.00022924784570932388 2023-01-23 14:51:15.980548: step: 1836/1541, loss: 0.0056908465921878815 2023-01-23 14:51:16.886311: step: 1838/1541, loss: 0.00037111196434125304 2023-01-23 14:51:17.836938: step: 1840/1541, loss: 0.003626273013651371 2023-01-23 14:51:18.729740: step: 1842/1541, loss: 0.013278440572321415 2023-01-23 14:51:19.670353: step: 1844/1541, loss: 0.17061996459960938 2023-01-23 14:51:20.626607: step: 1846/1541, loss: 0.753840446472168 2023-01-23 14:51:21.546531: step: 1848/1541, loss: 1.2376736402511597 2023-01-23 14:51:22.434896: step: 1850/1541, loss: 0.24063389003276825 2023-01-23 14:51:23.324397: step: 1852/1541, loss: 0.0225683581084013 2023-01-23 14:51:24.187710: step: 1854/1541, loss: 0.19850791990756989 2023-01-23 14:51:25.079847: step: 1856/1541, loss: 0.019589034840464592 2023-01-23 14:51:25.947572: step: 1858/1541, loss: 0.022924920544028282 2023-01-23 14:51:26.928360: step: 1860/1541, loss: 0.13727909326553345 2023-01-23 14:51:27.788429: step: 1862/1541, loss: 0.0009311704197898507 2023-01-23 14:51:28.739434: step: 1864/1541, loss: 0.0028301056008785963 2023-01-23 14:51:29.635938: step: 1866/1541, loss: 0.0012343702837824821 2023-01-23 14:51:30.510119: step: 1868/1541, loss: 0.012259801849722862 2023-01-23 14:51:31.387572: step: 1870/1541, loss: 0.04377282038331032 2023-01-23 14:51:32.294571: step: 1872/1541, loss: 0.031167276203632355 2023-01-23 14:51:33.205098: step: 1874/1541, loss: 0.015696126967668533 2023-01-23 14:51:34.084557: step: 1876/1541, loss: 0.10121112316846848 2023-01-23 14:51:34.945877: step: 1878/1541, loss: 0.18100860714912415 2023-01-23 14:51:35.849686: step: 1880/1541, loss: 0.01610606163740158 2023-01-23 14:51:36.740310: step: 1882/1541, loss: 0.002457632217556238 2023-01-23 14:51:37.618565: step: 1884/1541, loss: 0.0053979139775037766 2023-01-23 14:51:38.508971: step: 1886/1541, loss: 0.047490037977695465 2023-01-23 14:51:39.445104: step: 1888/1541, loss: 0.8405548930168152 2023-01-23 14:51:40.319133: step: 1890/1541, loss: 0.02941068261861801 2023-01-23 14:51:41.192761: step: 1892/1541, loss: 0.13808999955654144 2023-01-23 14:51:42.064772: step: 1894/1541, loss: 0.19576206803321838 2023-01-23 14:51:42.961984: step: 1896/1541, loss: 0.003325879108160734 2023-01-23 14:51:43.878196: step: 1898/1541, loss: 0.0023479238152503967 2023-01-23 14:51:44.760415: step: 1900/1541, loss: 0.11652793735265732 2023-01-23 14:51:45.690051: step: 1902/1541, loss: 0.7801007032394409 2023-01-23 14:51:46.580609: step: 1904/1541, loss: 0.22370277345180511 2023-01-23 14:51:47.475325: step: 1906/1541, loss: 0.15920549631118774 2023-01-23 14:51:48.372834: step: 1908/1541, loss: 0.006289755925536156 2023-01-23 14:51:49.265519: step: 1910/1541, loss: 0.0010824750643223524 2023-01-23 14:51:50.141391: step: 1912/1541, loss: 0.021892236545681953 2023-01-23 14:51:51.013888: step: 1914/1541, loss: 0.5372433066368103 2023-01-23 14:51:51.896509: step: 1916/1541, loss: 0.01541062444448471 2023-01-23 14:51:52.767742: step: 1918/1541, loss: 0.2022789865732193 2023-01-23 14:51:53.671644: step: 1920/1541, loss: 0.1393156349658966 2023-01-23 14:51:54.565142: step: 1922/1541, loss: 0.021942690014839172 2023-01-23 14:51:55.431956: step: 1924/1541, loss: 0.0035410448908805847 2023-01-23 14:51:56.340169: step: 1926/1541, loss: 0.17590144276618958 2023-01-23 14:51:57.264229: step: 1928/1541, loss: 0.005578326061367989 2023-01-23 14:51:58.122337: step: 1930/1541, loss: 0.086313895881176 2023-01-23 14:51:59.023625: step: 1932/1541, loss: 0.0011053020134568214 2023-01-23 14:51:59.905525: step: 1934/1541, loss: 0.0013264317531138659 2023-01-23 14:52:00.839132: step: 1936/1541, loss: 0.07779142260551453 2023-01-23 14:52:01.734172: step: 1938/1541, loss: 0.005406767129898071 2023-01-23 14:52:02.717621: step: 1940/1541, loss: 0.21903489530086517 2023-01-23 14:52:03.612654: step: 1942/1541, loss: 0.20666518807411194 2023-01-23 14:52:04.520285: step: 1944/1541, loss: 0.014932413585484028 2023-01-23 14:52:05.422904: step: 1946/1541, loss: 0.23877562582492828 2023-01-23 14:52:06.319658: step: 1948/1541, loss: 0.00827255379408598 2023-01-23 14:52:07.229395: step: 1950/1541, loss: 0.17674539983272552 2023-01-23 14:52:08.122740: step: 1952/1541, loss: 0.001312063541263342 2023-01-23 14:52:09.033308: step: 1954/1541, loss: 0.0023723822087049484 2023-01-23 14:52:09.940656: step: 1956/1541, loss: 0.003089007455855608 2023-01-23 14:52:10.897702: step: 1958/1541, loss: 0.02933012880384922 2023-01-23 14:52:11.793814: step: 1960/1541, loss: 0.04812924563884735 2023-01-23 14:52:12.710085: step: 1962/1541, loss: 0.03258667141199112 2023-01-23 14:52:13.587812: step: 1964/1541, loss: 0.0008046693401411176 2023-01-23 14:52:14.488698: step: 1966/1541, loss: 3.2499574444955215e-05 2023-01-23 14:52:15.367475: step: 1968/1541, loss: 0.1515752375125885 2023-01-23 14:52:16.273513: step: 1970/1541, loss: 0.06521917134523392 2023-01-23 14:52:17.153330: step: 1972/1541, loss: 0.0018795653013512492 2023-01-23 14:52:18.090456: step: 1974/1541, loss: 0.04549780488014221 2023-01-23 14:52:19.021039: step: 1976/1541, loss: 0.14046607911586761 2023-01-23 14:52:19.876040: step: 1978/1541, loss: 0.0032362283673137426 2023-01-23 14:52:20.772150: step: 1980/1541, loss: 0.17602813243865967 2023-01-23 14:52:21.665656: step: 1982/1541, loss: 0.558457612991333 2023-01-23 14:52:22.559189: step: 1984/1541, loss: 0.030884860083460808 2023-01-23 14:52:23.448051: step: 1986/1541, loss: 0.4311995506286621 2023-01-23 14:52:24.340434: step: 1988/1541, loss: 0.02027042955160141 2023-01-23 14:52:25.314241: step: 1990/1541, loss: 0.0008912637713365257 2023-01-23 14:52:26.201816: step: 1992/1541, loss: 0.02631687931716442 2023-01-23 14:52:27.108287: step: 1994/1541, loss: 0.03105524741113186 2023-01-23 14:52:27.974149: step: 1996/1541, loss: 0.444122850894928 2023-01-23 14:52:28.839369: step: 1998/1541, loss: 0.025688691064715385 2023-01-23 14:52:29.780214: step: 2000/1541, loss: 0.05086357519030571 2023-01-23 14:52:30.671122: step: 2002/1541, loss: 0.08661425858736038 2023-01-23 14:52:31.625170: step: 2004/1541, loss: 0.16971120238304138 2023-01-23 14:52:32.547931: step: 2006/1541, loss: 0.028608551248908043 2023-01-23 14:52:33.430054: step: 2008/1541, loss: 0.1618943214416504 2023-01-23 14:52:34.313283: step: 2010/1541, loss: 0.005279265809804201 2023-01-23 14:52:35.200741: step: 2012/1541, loss: 0.5664606690406799 2023-01-23 14:52:36.068233: step: 2014/1541, loss: 0.0063206651248037815 2023-01-23 14:52:36.967913: step: 2016/1541, loss: 0.0008830892620608211 2023-01-23 14:52:37.877350: step: 2018/1541, loss: 0.0011301652994006872 2023-01-23 14:52:38.814821: step: 2020/1541, loss: 0.0763089582324028 2023-01-23 14:52:39.708551: step: 2022/1541, loss: 0.003241674043238163 2023-01-23 14:52:40.623004: step: 2024/1541, loss: 0.02390648052096367 2023-01-23 14:52:41.496922: step: 2026/1541, loss: 0.01971549354493618 2023-01-23 14:52:42.388205: step: 2028/1541, loss: 0.05436837300658226 2023-01-23 14:52:43.300031: step: 2030/1541, loss: 0.34655138850212097 2023-01-23 14:52:44.157327: step: 2032/1541, loss: 0.0014600632712244987 2023-01-23 14:52:45.140313: step: 2034/1541, loss: 0.025856943801045418 2023-01-23 14:52:45.999579: step: 2036/1541, loss: 0.24305571615695953 2023-01-23 14:52:46.897106: step: 2038/1541, loss: 0.039396896958351135 2023-01-23 14:52:47.844076: step: 2040/1541, loss: 0.07908719033002853 2023-01-23 14:52:48.725402: step: 2042/1541, loss: 0.007671327795833349 2023-01-23 14:52:49.651106: step: 2044/1541, loss: 0.010425431653857231 2023-01-23 14:52:50.530155: step: 2046/1541, loss: 0.0016798351425677538 2023-01-23 14:52:51.452641: step: 2048/1541, loss: 0.011970201507210732 2023-01-23 14:52:52.356100: step: 2050/1541, loss: 0.003923983778804541 2023-01-23 14:52:53.279089: step: 2052/1541, loss: 0.2474774420261383 2023-01-23 14:52:54.169973: step: 2054/1541, loss: 0.05085920915007591 2023-01-23 14:52:55.060038: step: 2056/1541, loss: 0.0020805345848202705 2023-01-23 14:52:55.949973: step: 2058/1541, loss: 0.1608337163925171 2023-01-23 14:52:56.840205: step: 2060/1541, loss: 0.20505480468273163 2023-01-23 14:52:57.688938: step: 2062/1541, loss: 0.023182587698101997 2023-01-23 14:52:58.534847: step: 2064/1541, loss: 0.053640175610780716 2023-01-23 14:52:59.426590: step: 2066/1541, loss: 0.1287018060684204 2023-01-23 14:53:00.312634: step: 2068/1541, loss: 1.036729097366333 2023-01-23 14:53:01.294161: step: 2070/1541, loss: 0.05699240416288376 2023-01-23 14:53:02.259733: step: 2072/1541, loss: 0.005698154680430889 2023-01-23 14:53:03.137948: step: 2074/1541, loss: 0.03561811149120331 2023-01-23 14:53:04.032641: step: 2076/1541, loss: 0.01775103621184826 2023-01-23 14:53:04.935891: step: 2078/1541, loss: 0.09427357465028763 2023-01-23 14:53:05.849003: step: 2080/1541, loss: 0.02265825681388378 2023-01-23 14:53:06.818365: step: 2082/1541, loss: 0.2246987670660019 2023-01-23 14:53:07.712384: step: 2084/1541, loss: 0.12058053910732269 2023-01-23 14:53:08.615087: step: 2086/1541, loss: 0.02958082966506481 2023-01-23 14:53:09.494735: step: 2088/1541, loss: 0.5627107620239258 2023-01-23 14:53:10.534793: step: 2090/1541, loss: 0.04336955398321152 2023-01-23 14:53:11.419723: step: 2092/1541, loss: 0.0012846788158640265 2023-01-23 14:53:12.287327: step: 2094/1541, loss: 0.012439149431884289 2023-01-23 14:53:13.175382: step: 2096/1541, loss: 0.01697550155222416 2023-01-23 14:53:14.002108: step: 2098/1541, loss: 0.048243552446365356 2023-01-23 14:53:14.855291: step: 2100/1541, loss: 0.006056813057512045 2023-01-23 14:53:15.743351: step: 2102/1541, loss: 0.0223875530064106 2023-01-23 14:53:16.697352: step: 2104/1541, loss: 0.0018621780909597874 2023-01-23 14:53:17.611027: step: 2106/1541, loss: 0.016319552436470985 2023-01-23 14:53:18.477335: step: 2108/1541, loss: 0.0004905280657112598 2023-01-23 14:53:19.406061: step: 2110/1541, loss: 0.053111929446458817 2023-01-23 14:53:20.268433: step: 2112/1541, loss: 0.20743097364902496 2023-01-23 14:53:21.184530: step: 2114/1541, loss: 0.0061095827259123325 2023-01-23 14:53:22.085683: step: 2116/1541, loss: 0.3446820378303528 2023-01-23 14:53:22.961726: step: 2118/1541, loss: 0.2454647719860077 2023-01-23 14:53:23.899170: step: 2120/1541, loss: 0.051596082746982574 2023-01-23 14:53:24.724035: step: 2122/1541, loss: 0.0258918609470129 2023-01-23 14:53:25.595672: step: 2124/1541, loss: 0.5607160925865173 2023-01-23 14:53:26.461856: step: 2126/1541, loss: 0.055626146495342255 2023-01-23 14:53:27.390942: step: 2128/1541, loss: 0.007548685185611248 2023-01-23 14:53:28.274548: step: 2130/1541, loss: 0.0038399333134293556 2023-01-23 14:53:29.140817: step: 2132/1541, loss: 0.0022137530613690615 2023-01-23 14:53:30.073664: step: 2134/1541, loss: 0.11033478379249573 2023-01-23 14:53:30.954901: step: 2136/1541, loss: 0.016594547778367996 2023-01-23 14:53:31.896360: step: 2138/1541, loss: 0.06211506575345993 2023-01-23 14:53:32.820824: step: 2140/1541, loss: 0.007192818447947502 2023-01-23 14:53:33.780987: step: 2142/1541, loss: 0.013999562710523605 2023-01-23 14:53:34.680885: step: 2144/1541, loss: 0.057759203016757965 2023-01-23 14:53:35.587328: step: 2146/1541, loss: 0.16721783578395844 2023-01-23 14:53:36.464337: step: 2148/1541, loss: 0.09702923148870468 2023-01-23 14:53:37.342166: step: 2150/1541, loss: 0.015077224932610989 2023-01-23 14:53:38.268321: step: 2152/1541, loss: 0.05161929503083229 2023-01-23 14:53:39.175665: step: 2154/1541, loss: 0.0008450804743915796 2023-01-23 14:53:40.070866: step: 2156/1541, loss: 0.12457086145877838 2023-01-23 14:53:40.962124: step: 2158/1541, loss: 0.07479066401720047 2023-01-23 14:53:41.821114: step: 2160/1541, loss: 0.003082514740526676 2023-01-23 14:53:42.741661: step: 2162/1541, loss: 0.13016065955162048 2023-01-23 14:53:43.650628: step: 2164/1541, loss: 0.027801763266324997 2023-01-23 14:53:44.558609: step: 2166/1541, loss: 0.012355118989944458 2023-01-23 14:53:45.452402: step: 2168/1541, loss: 0.024368399754166603 2023-01-23 14:53:46.357684: step: 2170/1541, loss: 0.0032258369028568268 2023-01-23 14:53:47.257435: step: 2172/1541, loss: 0.024102341383695602 2023-01-23 14:53:48.266875: step: 2174/1541, loss: 0.11017893254756927 2023-01-23 14:53:49.169108: step: 2176/1541, loss: 0.026448696851730347 2023-01-23 14:53:50.089266: step: 2178/1541, loss: 0.12930218875408173 2023-01-23 14:53:50.937788: step: 2180/1541, loss: 0.0007543197134509683 2023-01-23 14:53:51.837831: step: 2182/1541, loss: 0.004440654069185257 2023-01-23 14:53:52.761660: step: 2184/1541, loss: 0.0031375382095575333 2023-01-23 14:53:53.650900: step: 2186/1541, loss: 0.0015739959198981524 2023-01-23 14:53:54.554765: step: 2188/1541, loss: 0.013768995180726051 2023-01-23 14:53:55.502802: step: 2190/1541, loss: 0.15648646652698517 2023-01-23 14:53:56.392671: step: 2192/1541, loss: 0.0021518359426409006 2023-01-23 14:53:57.256003: step: 2194/1541, loss: 0.011366806924343109 2023-01-23 14:53:58.107083: step: 2196/1541, loss: 0.12104998528957367 2023-01-23 14:53:59.066500: step: 2198/1541, loss: 0.03889764845371246 2023-01-23 14:53:59.947776: step: 2200/1541, loss: 0.23901808261871338 2023-01-23 14:54:00.860344: step: 2202/1541, loss: 0.012092727236449718 2023-01-23 14:54:01.729951: step: 2204/1541, loss: 0.04107450321316719 2023-01-23 14:54:02.671549: step: 2206/1541, loss: 0.009192622266709805 2023-01-23 14:54:03.547109: step: 2208/1541, loss: 0.023393267765641212 2023-01-23 14:54:04.407664: step: 2210/1541, loss: 0.04814985394477844 2023-01-23 14:54:05.319868: step: 2212/1541, loss: 0.2115429937839508 2023-01-23 14:54:06.218245: step: 2214/1541, loss: 0.0859154611825943 2023-01-23 14:54:07.129472: step: 2216/1541, loss: 0.005989289376884699 2023-01-23 14:54:08.053106: step: 2218/1541, loss: 0.017773227766156197 2023-01-23 14:54:08.957301: step: 2220/1541, loss: 0.14124761521816254 2023-01-23 14:54:09.914846: step: 2222/1541, loss: 0.03549240157008171 2023-01-23 14:54:10.809925: step: 2224/1541, loss: 0.001962220761924982 2023-01-23 14:54:11.715148: step: 2226/1541, loss: 0.003306201659142971 2023-01-23 14:54:12.617124: step: 2228/1541, loss: 0.09566272795200348 2023-01-23 14:54:13.487074: step: 2230/1541, loss: 0.38692474365234375 2023-01-23 14:54:14.411852: step: 2232/1541, loss: 0.49822521209716797 2023-01-23 14:54:15.356736: step: 2234/1541, loss: 0.09285132586956024 2023-01-23 14:54:16.253508: step: 2236/1541, loss: 0.12877774238586426 2023-01-23 14:54:17.194002: step: 2238/1541, loss: 0.10646367818117142 2023-01-23 14:54:18.077336: step: 2240/1541, loss: 0.06066196411848068 2023-01-23 14:54:18.975226: step: 2242/1541, loss: 0.08007532358169556 2023-01-23 14:54:19.850559: step: 2244/1541, loss: 0.0015845928573980927 2023-01-23 14:54:20.755505: step: 2246/1541, loss: 0.010475115850567818 2023-01-23 14:54:21.632949: step: 2248/1541, loss: 0.00973234511911869 2023-01-23 14:54:22.493048: step: 2250/1541, loss: 0.036462292075157166 2023-01-23 14:54:23.367901: step: 2252/1541, loss: 0.02376302145421505 2023-01-23 14:54:24.252148: step: 2254/1541, loss: 0.009071102365851402 2023-01-23 14:54:25.156171: step: 2256/1541, loss: 0.06627020239830017 2023-01-23 14:54:26.024731: step: 2258/1541, loss: 0.10506220906972885 2023-01-23 14:54:26.901273: step: 2260/1541, loss: 0.0007936840411275625 2023-01-23 14:54:27.791595: step: 2262/1541, loss: 0.11215343326330185 2023-01-23 14:54:28.666252: step: 2264/1541, loss: 0.022900428622961044 2023-01-23 14:54:29.568931: step: 2266/1541, loss: 0.004238556604832411 2023-01-23 14:54:30.467790: step: 2268/1541, loss: 0.010124627500772476 2023-01-23 14:54:31.381952: step: 2270/1541, loss: 0.10391164571046829 2023-01-23 14:54:32.258604: step: 2272/1541, loss: 0.009583141654729843 2023-01-23 14:54:33.163677: step: 2274/1541, loss: 0.011128293350338936 2023-01-23 14:54:34.049310: step: 2276/1541, loss: 0.004471362102776766 2023-01-23 14:54:34.921645: step: 2278/1541, loss: 0.0023573271464556456 2023-01-23 14:54:35.805354: step: 2280/1541, loss: 0.13002674281597137 2023-01-23 14:54:36.728781: step: 2282/1541, loss: 0.047316282987594604 2023-01-23 14:54:37.635393: step: 2284/1541, loss: 0.00024473629309795797 2023-01-23 14:54:38.496960: step: 2286/1541, loss: 0.0384853333234787 2023-01-23 14:54:39.430967: step: 2288/1541, loss: 0.16402389109134674 2023-01-23 14:54:40.304271: step: 2290/1541, loss: 0.13627204298973083 2023-01-23 14:54:41.170784: step: 2292/1541, loss: 0.030709344893693924 2023-01-23 14:54:42.097981: step: 2294/1541, loss: 0.08357752859592438 2023-01-23 14:54:42.969375: step: 2296/1541, loss: 0.0278677549213171 2023-01-23 14:54:43.853657: step: 2298/1541, loss: 0.0014603582676500082 2023-01-23 14:54:44.743679: step: 2300/1541, loss: 0.6800603270530701 2023-01-23 14:54:45.623304: step: 2302/1541, loss: 0.8535870313644409 2023-01-23 14:54:46.532385: step: 2304/1541, loss: 0.012221544049680233 2023-01-23 14:54:47.422103: step: 2306/1541, loss: 0.000963623751886189 2023-01-23 14:54:48.340266: step: 2308/1541, loss: 0.03480781987309456 2023-01-23 14:54:49.220245: step: 2310/1541, loss: 0.051503751426935196 2023-01-23 14:54:50.094615: step: 2312/1541, loss: 0.0067104557529091835 2023-01-23 14:54:51.009995: step: 2314/1541, loss: 0.19389386475086212 2023-01-23 14:54:51.951094: step: 2316/1541, loss: 0.0006941025494597852 2023-01-23 14:54:52.825093: step: 2318/1541, loss: 0.009320779703557491 2023-01-23 14:54:53.715923: step: 2320/1541, loss: 0.001661832444369793 2023-01-23 14:54:54.598148: step: 2322/1541, loss: 0.02785910852253437 2023-01-23 14:54:55.496513: step: 2324/1541, loss: 0.17270806431770325 2023-01-23 14:54:56.413215: step: 2326/1541, loss: 0.00423615425825119 2023-01-23 14:54:57.269016: step: 2328/1541, loss: 0.00267908931709826 2023-01-23 14:54:58.175760: step: 2330/1541, loss: 0.03799819573760033 2023-01-23 14:54:59.203153: step: 2332/1541, loss: 0.08495134115219116 2023-01-23 14:55:00.105777: step: 2334/1541, loss: 0.019726833328604698 2023-01-23 14:55:00.988049: step: 2336/1541, loss: 0.052025772631168365 2023-01-23 14:55:01.923758: step: 2338/1541, loss: 0.09040945768356323 2023-01-23 14:55:02.815919: step: 2340/1541, loss: 0.5516042709350586 2023-01-23 14:55:03.718170: step: 2342/1541, loss: 0.0005399503861553967 2023-01-23 14:55:04.632708: step: 2344/1541, loss: 0.00025722431018948555 2023-01-23 14:55:05.586881: step: 2346/1541, loss: 0.6730816960334778 2023-01-23 14:55:06.455474: step: 2348/1541, loss: 0.015165118500590324 2023-01-23 14:55:07.336126: step: 2350/1541, loss: 0.024869944900274277 2023-01-23 14:55:08.220398: step: 2352/1541, loss: 0.1713968962430954 2023-01-23 14:55:09.141115: step: 2354/1541, loss: 0.024188915267586708 2023-01-23 14:55:10.056078: step: 2356/1541, loss: 0.11223762482404709 2023-01-23 14:55:10.965959: step: 2358/1541, loss: 0.004123964812606573 2023-01-23 14:55:11.917047: step: 2360/1541, loss: 0.056243907660245895 2023-01-23 14:55:12.860746: step: 2362/1541, loss: 0.000575296813622117 2023-01-23 14:55:13.777195: step: 2364/1541, loss: 0.001832654234021902 2023-01-23 14:55:14.667280: step: 2366/1541, loss: 0.122042715549469 2023-01-23 14:55:15.569735: step: 2368/1541, loss: 0.2577767074108124 2023-01-23 14:55:16.428349: step: 2370/1541, loss: 0.01380847580730915 2023-01-23 14:55:17.304482: step: 2372/1541, loss: 0.003456735284999013 2023-01-23 14:55:18.178724: step: 2374/1541, loss: 0.0050200060941278934 2023-01-23 14:55:19.062310: step: 2376/1541, loss: 0.2935665249824524 2023-01-23 14:55:19.958917: step: 2378/1541, loss: 0.020840173587203026 2023-01-23 14:55:20.882436: step: 2380/1541, loss: 0.052427154034376144 2023-01-23 14:55:21.880257: step: 2382/1541, loss: 0.004018721170723438 2023-01-23 14:55:22.794334: step: 2384/1541, loss: 0.04193698987364769 2023-01-23 14:55:23.684827: step: 2386/1541, loss: 0.12631438672542572 2023-01-23 14:55:24.590567: step: 2388/1541, loss: 0.04917308688163757 2023-01-23 14:55:25.494466: step: 2390/1541, loss: 0.0016786332707852125 2023-01-23 14:55:26.356056: step: 2392/1541, loss: 0.1676911562681198 2023-01-23 14:55:27.255144: step: 2394/1541, loss: 0.039089519530534744 2023-01-23 14:55:28.107693: step: 2396/1541, loss: 0.02241251803934574 2023-01-23 14:55:28.985954: step: 2398/1541, loss: 0.13293948769569397 2023-01-23 14:55:29.898048: step: 2400/1541, loss: 0.00515945628285408 2023-01-23 14:55:30.829845: step: 2402/1541, loss: 0.11685876548290253 2023-01-23 14:55:31.693081: step: 2404/1541, loss: 0.27180296182632446 2023-01-23 14:55:32.637066: step: 2406/1541, loss: 0.005103992763906717 2023-01-23 14:55:33.519385: step: 2408/1541, loss: 0.001759159262292087 2023-01-23 14:55:34.409407: step: 2410/1541, loss: 0.0033320062793791294 2023-01-23 14:55:35.273157: step: 2412/1541, loss: 0.0038172518834471703 2023-01-23 14:55:36.196101: step: 2414/1541, loss: 0.04959040507674217 2023-01-23 14:55:37.079040: step: 2416/1541, loss: 0.0012163566425442696 2023-01-23 14:55:38.020668: step: 2418/1541, loss: 0.478801965713501 2023-01-23 14:55:38.974557: step: 2420/1541, loss: 0.03254525735974312 2023-01-23 14:55:39.882817: step: 2422/1541, loss: 0.10314453393220901 2023-01-23 14:55:40.805625: step: 2424/1541, loss: 0.010297751054167747 2023-01-23 14:55:41.699428: step: 2426/1541, loss: 0.006720947101712227 2023-01-23 14:55:42.570005: step: 2428/1541, loss: 0.0026776655577123165 2023-01-23 14:55:43.478833: step: 2430/1541, loss: 0.05674521625041962 2023-01-23 14:55:44.391116: step: 2432/1541, loss: 0.11773022264242172 2023-01-23 14:55:45.292387: step: 2434/1541, loss: 0.0529618076980114 2023-01-23 14:55:46.207786: step: 2436/1541, loss: 0.18057745695114136 2023-01-23 14:55:47.090178: step: 2438/1541, loss: 0.001577341347001493 2023-01-23 14:55:47.989840: step: 2440/1541, loss: 0.0065648555755615234 2023-01-23 14:55:48.886485: step: 2442/1541, loss: 0.20970375835895538 2023-01-23 14:55:49.745769: step: 2444/1541, loss: 0.10439158976078033 2023-01-23 14:55:50.663005: step: 2446/1541, loss: 0.24425913393497467 2023-01-23 14:55:51.566704: step: 2448/1541, loss: 0.4787808358669281 2023-01-23 14:55:52.481339: step: 2450/1541, loss: 0.09783443063497543 2023-01-23 14:55:53.367815: step: 2452/1541, loss: 0.001619736198335886 2023-01-23 14:55:54.302036: step: 2454/1541, loss: 0.01845349743962288 2023-01-23 14:55:55.244806: step: 2456/1541, loss: 0.08009319007396698 2023-01-23 14:55:56.121790: step: 2458/1541, loss: 0.057950571179389954 2023-01-23 14:55:57.092318: step: 2460/1541, loss: 0.01551899965852499 2023-01-23 14:55:58.035652: step: 2462/1541, loss: 0.055243831127882004 2023-01-23 14:55:58.958191: step: 2464/1541, loss: 0.021249718964099884 2023-01-23 14:55:59.947639: step: 2466/1541, loss: 0.3058217167854309 2023-01-23 14:56:00.812274: step: 2468/1541, loss: 0.0007437046151608229 2023-01-23 14:56:01.664280: step: 2470/1541, loss: 0.019062835723161697 2023-01-23 14:56:02.567557: step: 2472/1541, loss: 0.035469863563776016 2023-01-23 14:56:03.427838: step: 2474/1541, loss: 0.007076196372509003 2023-01-23 14:56:04.309079: step: 2476/1541, loss: 0.23312513530254364 2023-01-23 14:56:05.168915: step: 2478/1541, loss: 0.007329699117690325 2023-01-23 14:56:06.031996: step: 2480/1541, loss: 0.4277762472629547 2023-01-23 14:56:06.913922: step: 2482/1541, loss: 0.17041075229644775 2023-01-23 14:56:07.833403: step: 2484/1541, loss: 0.02063620090484619 2023-01-23 14:56:08.751205: step: 2486/1541, loss: 0.004550366662442684 2023-01-23 14:56:09.706697: step: 2488/1541, loss: 0.28292569518089294 2023-01-23 14:56:10.597044: step: 2490/1541, loss: 0.558793842792511 2023-01-23 14:56:11.513496: step: 2492/1541, loss: 0.22964882850646973 2023-01-23 14:56:12.387262: step: 2494/1541, loss: 0.051559872925281525 2023-01-23 14:56:13.302260: step: 2496/1541, loss: 0.010805750265717506 2023-01-23 14:56:14.185280: step: 2498/1541, loss: 0.0011680389288812876 2023-01-23 14:56:15.086685: step: 2500/1541, loss: 0.08965923637151718 2023-01-23 14:56:15.978501: step: 2502/1541, loss: 0.055631645023822784 2023-01-23 14:56:16.854620: step: 2504/1541, loss: 0.006857207976281643 2023-01-23 14:56:17.757420: step: 2506/1541, loss: 0.054466161876916885 2023-01-23 14:56:18.714615: step: 2508/1541, loss: 0.0281537938863039 2023-01-23 14:56:19.621780: step: 2510/1541, loss: 0.022615671157836914 2023-01-23 14:56:20.516801: step: 2512/1541, loss: 0.07317179441452026 2023-01-23 14:56:21.386952: step: 2514/1541, loss: 0.02325291372835636 2023-01-23 14:56:22.279068: step: 2516/1541, loss: 0.03608798608183861 2023-01-23 14:56:23.165136: step: 2518/1541, loss: 0.0244988314807415 2023-01-23 14:56:24.035782: step: 2520/1541, loss: 0.19086667895317078 2023-01-23 14:56:25.025417: step: 2522/1541, loss: 0.01666838675737381 2023-01-23 14:56:25.907061: step: 2524/1541, loss: 0.0412336103618145 2023-01-23 14:56:26.778187: step: 2526/1541, loss: 0.1462620347738266 2023-01-23 14:56:27.652399: step: 2528/1541, loss: 0.14384415745735168 2023-01-23 14:56:28.551218: step: 2530/1541, loss: 0.0034388259518891573 2023-01-23 14:56:29.423084: step: 2532/1541, loss: 0.16273005306720734 2023-01-23 14:56:30.370752: step: 2534/1541, loss: 0.005176844075322151 2023-01-23 14:56:31.312957: step: 2536/1541, loss: 0.15209831297397614 2023-01-23 14:56:32.209288: step: 2538/1541, loss: 0.007649002131074667 2023-01-23 14:56:33.133543: step: 2540/1541, loss: 0.012275660410523415 2023-01-23 14:56:34.050534: step: 2542/1541, loss: 0.008367364294826984 2023-01-23 14:56:34.940960: step: 2544/1541, loss: 0.033841680735349655 2023-01-23 14:56:35.850012: step: 2546/1541, loss: 1.435711145401001 2023-01-23 14:56:36.756483: step: 2548/1541, loss: 0.00866493210196495 2023-01-23 14:56:37.679253: step: 2550/1541, loss: 0.06087028235197067 2023-01-23 14:56:38.612251: step: 2552/1541, loss: 0.061327721923589706 2023-01-23 14:56:39.536425: step: 2554/1541, loss: 1.0702471733093262 2023-01-23 14:56:40.401198: step: 2556/1541, loss: 0.4292415678501129 2023-01-23 14:56:41.299319: step: 2558/1541, loss: 0.02316407300531864 2023-01-23 14:56:42.212660: step: 2560/1541, loss: 0.0013592599425464869 2023-01-23 14:56:43.110764: step: 2562/1541, loss: 0.05363268032670021 2023-01-23 14:56:44.017862: step: 2564/1541, loss: 0.0020845462568104267 2023-01-23 14:56:44.951691: step: 2566/1541, loss: 0.04289061203598976 2023-01-23 14:56:45.833742: step: 2568/1541, loss: 0.0002901387051679194 2023-01-23 14:56:46.729540: step: 2570/1541, loss: 0.043465495109558105 2023-01-23 14:56:47.678301: step: 2572/1541, loss: 0.010922739282250404 2023-01-23 14:56:48.562114: step: 2574/1541, loss: 0.003984601236879826 2023-01-23 14:56:49.498671: step: 2576/1541, loss: 0.16459687054157257 2023-01-23 14:56:50.490532: step: 2578/1541, loss: 0.032232243567705154 2023-01-23 14:56:51.390373: step: 2580/1541, loss: 0.3162035048007965 2023-01-23 14:56:52.424856: step: 2582/1541, loss: 0.030334463343024254 2023-01-23 14:56:53.358786: step: 2584/1541, loss: 0.003713259007781744 2023-01-23 14:56:54.223138: step: 2586/1541, loss: 0.004180682823061943 2023-01-23 14:56:55.122994: step: 2588/1541, loss: 0.20217987895011902 2023-01-23 14:56:56.072906: step: 2590/1541, loss: 0.07513339072465897 2023-01-23 14:56:56.956274: step: 2592/1541, loss: 0.1449218988418579 2023-01-23 14:56:57.834773: step: 2594/1541, loss: 0.001541769364848733 2023-01-23 14:56:58.756781: step: 2596/1541, loss: 0.1350420117378235 2023-01-23 14:56:59.683415: step: 2598/1541, loss: 0.006314007565379143 2023-01-23 14:57:00.583170: step: 2600/1541, loss: 0.007859422825276852 2023-01-23 14:57:01.477271: step: 2602/1541, loss: 0.31214872002601624 2023-01-23 14:57:02.360345: step: 2604/1541, loss: 0.0022235591895878315 2023-01-23 14:57:03.255473: step: 2606/1541, loss: 0.2604001760482788 2023-01-23 14:57:04.137561: step: 2608/1541, loss: 0.18911929428577423 2023-01-23 14:57:05.044189: step: 2610/1541, loss: 0.2841091752052307 2023-01-23 14:57:05.989819: step: 2612/1541, loss: 0.004438563250005245 2023-01-23 14:57:06.915228: step: 2614/1541, loss: 0.04076956585049629 2023-01-23 14:57:07.864338: step: 2616/1541, loss: 0.17951543629169464 2023-01-23 14:57:08.760576: step: 2618/1541, loss: 0.2109803706407547 2023-01-23 14:57:09.623023: step: 2620/1541, loss: 0.0008038735250011086 2023-01-23 14:57:10.495712: step: 2622/1541, loss: 0.11101865023374557 2023-01-23 14:57:11.413937: step: 2624/1541, loss: 0.6668934226036072 2023-01-23 14:57:12.325676: step: 2626/1541, loss: 0.2289951592683792 2023-01-23 14:57:13.246194: step: 2628/1541, loss: 0.20744375884532928 2023-01-23 14:57:14.155335: step: 2630/1541, loss: 0.006959069054573774 2023-01-23 14:57:15.117323: step: 2632/1541, loss: 0.0011942886048927903 2023-01-23 14:57:16.012205: step: 2634/1541, loss: 0.046151116490364075 2023-01-23 14:57:16.928745: step: 2636/1541, loss: 0.05802495777606964 2023-01-23 14:57:17.859868: step: 2638/1541, loss: 0.2258944809436798 2023-01-23 14:57:18.849346: step: 2640/1541, loss: 0.028839102014899254 2023-01-23 14:57:19.794619: step: 2642/1541, loss: 0.01840377040207386 2023-01-23 14:57:20.665869: step: 2644/1541, loss: 0.003907926380634308 2023-01-23 14:57:21.546740: step: 2646/1541, loss: 0.01915445365011692 2023-01-23 14:57:22.454483: step: 2648/1541, loss: 0.11763038486242294 2023-01-23 14:57:23.395608: step: 2650/1541, loss: 1.302730679512024 2023-01-23 14:57:24.261151: step: 2652/1541, loss: 0.00968475453555584 2023-01-23 14:57:25.171534: step: 2654/1541, loss: 0.08698169887065887 2023-01-23 14:57:26.060849: step: 2656/1541, loss: 0.0050104716792702675 2023-01-23 14:57:26.964027: step: 2658/1541, loss: 0.004334392957389355 2023-01-23 14:57:27.845909: step: 2660/1541, loss: 0.04586714133620262 2023-01-23 14:57:28.761716: step: 2662/1541, loss: 0.1464719921350479 2023-01-23 14:57:29.691431: step: 2664/1541, loss: 0.3722721338272095 2023-01-23 14:57:30.578131: step: 2666/1541, loss: 0.5294820666313171 2023-01-23 14:57:31.479944: step: 2668/1541, loss: 0.02993914857506752 2023-01-23 14:57:32.385950: step: 2670/1541, loss: 0.007089097052812576 2023-01-23 14:57:33.251388: step: 2672/1541, loss: 0.015108779072761536 2023-01-23 14:57:34.136305: step: 2674/1541, loss: 0.6905155181884766 2023-01-23 14:57:35.011003: step: 2676/1541, loss: 0.08026348799467087 2023-01-23 14:57:35.970603: step: 2678/1541, loss: 0.03894783556461334 2023-01-23 14:57:36.854132: step: 2680/1541, loss: 0.015852589160203934 2023-01-23 14:57:37.733536: step: 2682/1541, loss: 0.059181734919548035 2023-01-23 14:57:38.642250: step: 2684/1541, loss: 0.0013583372347056866 2023-01-23 14:57:39.506264: step: 2686/1541, loss: 0.002511550672352314 2023-01-23 14:57:40.391921: step: 2688/1541, loss: 0.009925602003932 2023-01-23 14:57:41.308636: step: 2690/1541, loss: 0.0356825515627861 2023-01-23 14:57:42.208888: step: 2692/1541, loss: 0.14882367849349976 2023-01-23 14:57:43.084447: step: 2694/1541, loss: 0.09184005856513977 2023-01-23 14:57:43.953725: step: 2696/1541, loss: 0.07068406045436859 2023-01-23 14:57:44.821592: step: 2698/1541, loss: 0.09044274687767029 2023-01-23 14:57:45.703320: step: 2700/1541, loss: 0.08126715570688248 2023-01-23 14:57:46.612404: step: 2702/1541, loss: 0.03812748193740845 2023-01-23 14:57:47.536248: step: 2704/1541, loss: 0.2268114984035492 2023-01-23 14:57:48.409035: step: 2706/1541, loss: 0.004034135490655899 2023-01-23 14:57:49.278237: step: 2708/1541, loss: 0.024817420169711113 2023-01-23 14:57:50.156509: step: 2710/1541, loss: 0.01659642904996872 2023-01-23 14:57:51.042091: step: 2712/1541, loss: 0.006873937789350748 2023-01-23 14:57:51.894063: step: 2714/1541, loss: 0.0034057197626680136 2023-01-23 14:57:52.817789: step: 2716/1541, loss: 0.22146940231323242 2023-01-23 14:57:53.697873: step: 2718/1541, loss: 0.0032457446213811636 2023-01-23 14:57:54.577840: step: 2720/1541, loss: 0.13668662309646606 2023-01-23 14:57:55.464848: step: 2722/1541, loss: 0.2734266221523285 2023-01-23 14:57:56.364917: step: 2724/1541, loss: 0.04560589790344238 2023-01-23 14:57:57.258372: step: 2726/1541, loss: 0.11498913168907166 2023-01-23 14:57:58.136270: step: 2728/1541, loss: 0.0033835116773843765 2023-01-23 14:57:59.019919: step: 2730/1541, loss: 0.4698968827724457 2023-01-23 14:57:59.935096: step: 2732/1541, loss: 0.05300412327051163 2023-01-23 14:58:00.810611: step: 2734/1541, loss: 0.028443919494748116 2023-01-23 14:58:01.672308: step: 2736/1541, loss: 0.001014145091176033 2023-01-23 14:58:02.661356: step: 2738/1541, loss: 0.012464524246752262 2023-01-23 14:58:03.541135: step: 2740/1541, loss: 0.01244304794818163 2023-01-23 14:58:04.431049: step: 2742/1541, loss: 0.06007201224565506 2023-01-23 14:58:05.329118: step: 2744/1541, loss: 0.05115649104118347 2023-01-23 14:58:06.207264: step: 2746/1541, loss: 0.19390752911567688 2023-01-23 14:58:07.145598: step: 2748/1541, loss: 0.03627650439739227 2023-01-23 14:58:08.142454: step: 2750/1541, loss: 0.02751932665705681 2023-01-23 14:58:09.034730: step: 2752/1541, loss: 0.00040315184742212296 2023-01-23 14:58:09.951773: step: 2754/1541, loss: 0.026617903262376785 2023-01-23 14:58:10.806444: step: 2756/1541, loss: 0.07451590895652771 2023-01-23 14:58:11.721891: step: 2758/1541, loss: 0.04644213989377022 2023-01-23 14:58:12.664023: step: 2760/1541, loss: 0.0015220555942505598 2023-01-23 14:58:13.552306: step: 2762/1541, loss: 0.0024781152606010437 2023-01-23 14:58:14.479007: step: 2764/1541, loss: 0.22066015005111694 2023-01-23 14:58:15.412732: step: 2766/1541, loss: 0.00927374605089426 2023-01-23 14:58:16.400551: step: 2768/1541, loss: 0.0027995051350444555 2023-01-23 14:58:17.283783: step: 2770/1541, loss: 0.03125801309943199 2023-01-23 14:58:18.219205: step: 2772/1541, loss: 0.005924984347075224 2023-01-23 14:58:19.134409: step: 2774/1541, loss: 0.03499224781990051 2023-01-23 14:58:20.012475: step: 2776/1541, loss: 0.03211138769984245 2023-01-23 14:58:20.885465: step: 2778/1541, loss: 0.22010311484336853 2023-01-23 14:58:21.742086: step: 2780/1541, loss: 0.3961426615715027 2023-01-23 14:58:22.676971: step: 2782/1541, loss: 0.021369488909840584 2023-01-23 14:58:23.532188: step: 2784/1541, loss: 0.006529415957629681 2023-01-23 14:58:24.407525: step: 2786/1541, loss: 0.00191067426931113 2023-01-23 14:58:25.301280: step: 2788/1541, loss: 0.0016596189234405756 2023-01-23 14:58:26.226989: step: 2790/1541, loss: 0.06708627194166183 2023-01-23 14:58:27.181452: step: 2792/1541, loss: 0.011597879230976105 2023-01-23 14:58:28.099470: step: 2794/1541, loss: 0.04272552579641342 2023-01-23 14:58:28.973366: step: 2796/1541, loss: 0.001127208350226283 2023-01-23 14:58:29.881682: step: 2798/1541, loss: 0.2495562583208084 2023-01-23 14:58:30.798732: step: 2800/1541, loss: 0.15533024072647095 2023-01-23 14:58:31.678533: step: 2802/1541, loss: 0.13218161463737488 2023-01-23 14:58:32.627759: step: 2804/1541, loss: 0.03431013599038124 2023-01-23 14:58:33.544922: step: 2806/1541, loss: 0.048792317509651184 2023-01-23 14:58:34.520205: step: 2808/1541, loss: 0.08137202262878418 2023-01-23 14:58:35.475015: step: 2810/1541, loss: 0.2343018651008606 2023-01-23 14:58:36.381484: step: 2812/1541, loss: 0.08139768242835999 2023-01-23 14:58:37.265635: step: 2814/1541, loss: 0.21005693078041077 2023-01-23 14:58:38.213532: step: 2816/1541, loss: 0.0010198475793004036 2023-01-23 14:58:39.108918: step: 2818/1541, loss: 0.009023095481097698 2023-01-23 14:58:39.995866: step: 2820/1541, loss: 0.004118498880416155 2023-01-23 14:58:40.913512: step: 2822/1541, loss: 0.006299855653196573 2023-01-23 14:58:41.823567: step: 2824/1541, loss: 0.04477488994598389 2023-01-23 14:58:42.724005: step: 2826/1541, loss: 0.0032553263008594513 2023-01-23 14:58:43.646310: step: 2828/1541, loss: 0.01728285476565361 2023-01-23 14:58:44.541075: step: 2830/1541, loss: 0.008146206848323345 2023-01-23 14:58:45.466345: step: 2832/1541, loss: 0.05759018287062645 2023-01-23 14:58:46.341934: step: 2834/1541, loss: 0.17585398256778717 2023-01-23 14:58:47.271637: step: 2836/1541, loss: 0.003632984356954694 2023-01-23 14:58:48.173684: step: 2838/1541, loss: 0.0034562654327601194 2023-01-23 14:58:49.045198: step: 2840/1541, loss: 0.6528621315956116 2023-01-23 14:58:49.915025: step: 2842/1541, loss: 0.006273627746850252 2023-01-23 14:58:50.815828: step: 2844/1541, loss: 0.007152046076953411 2023-01-23 14:58:51.700435: step: 2846/1541, loss: 0.014339257963001728 2023-01-23 14:58:52.666554: step: 2848/1541, loss: 0.08680178225040436 2023-01-23 14:58:53.541058: step: 2850/1541, loss: 0.08933660387992859 2023-01-23 14:58:54.437423: step: 2852/1541, loss: 0.12609384953975677 2023-01-23 14:58:55.404421: step: 2854/1541, loss: 0.0011365095851942897 2023-01-23 14:58:56.284022: step: 2856/1541, loss: 0.11240562051534653 2023-01-23 14:58:57.188480: step: 2858/1541, loss: 0.02239895425736904 2023-01-23 14:58:58.110665: step: 2860/1541, loss: 0.05867915228009224 2023-01-23 14:58:59.056767: step: 2862/1541, loss: 0.09278003871440887 2023-01-23 14:58:59.947151: step: 2864/1541, loss: 0.03421441838145256 2023-01-23 14:59:00.814683: step: 2866/1541, loss: 0.10491016507148743 2023-01-23 14:59:01.742648: step: 2868/1541, loss: 0.003859458025544882 2023-01-23 14:59:02.639084: step: 2870/1541, loss: 0.0823768898844719 2023-01-23 14:59:03.504413: step: 2872/1541, loss: 0.11990659683942795 2023-01-23 14:59:04.378979: step: 2874/1541, loss: 0.08376460522413254 2023-01-23 14:59:05.265677: step: 2876/1541, loss: 0.16256484389305115 2023-01-23 14:59:06.142488: step: 2878/1541, loss: 0.0014078961685299873 2023-01-23 14:59:07.065121: step: 2880/1541, loss: 0.20899489521980286 2023-01-23 14:59:07.918477: step: 2882/1541, loss: 0.04068193584680557 2023-01-23 14:59:08.880880: step: 2884/1541, loss: 0.002592571545392275 2023-01-23 14:59:09.790187: step: 2886/1541, loss: 0.3239312469959259 2023-01-23 14:59:10.724177: step: 2888/1541, loss: 0.013596813194453716 2023-01-23 14:59:11.635234: step: 2890/1541, loss: 0.010120989754796028 2023-01-23 14:59:12.539407: step: 2892/1541, loss: 0.014453781768679619 2023-01-23 14:59:13.398839: step: 2894/1541, loss: 0.0012029503704980016 2023-01-23 14:59:14.386520: step: 2896/1541, loss: 0.10325146466493607 2023-01-23 14:59:15.271484: step: 2898/1541, loss: 0.013717979192733765 2023-01-23 14:59:16.199306: step: 2900/1541, loss: 0.01632828079164028 2023-01-23 14:59:17.143962: step: 2902/1541, loss: 0.10927046835422516 2023-01-23 14:59:18.179417: step: 2904/1541, loss: 0.1920645833015442 2023-01-23 14:59:19.058797: step: 2906/1541, loss: 0.030231919139623642 2023-01-23 14:59:19.934338: step: 2908/1541, loss: 0.015515351668000221 2023-01-23 14:59:20.849979: step: 2910/1541, loss: 0.0478607639670372 2023-01-23 14:59:21.720696: step: 2912/1541, loss: 0.0027171592228114605 2023-01-23 14:59:22.589558: step: 2914/1541, loss: 0.058274898678064346 2023-01-23 14:59:23.491963: step: 2916/1541, loss: 0.1055503562092781 2023-01-23 14:59:24.367085: step: 2918/1541, loss: 0.17438410222530365 2023-01-23 14:59:25.236292: step: 2920/1541, loss: 0.0383063405752182 2023-01-23 14:59:26.118360: step: 2922/1541, loss: 0.022352144122123718 2023-01-23 14:59:26.987684: step: 2924/1541, loss: 0.004075824283063412 2023-01-23 14:59:27.839230: step: 2926/1541, loss: 0.061456240713596344 2023-01-23 14:59:28.726217: step: 2928/1541, loss: 0.15338493883609772 2023-01-23 14:59:29.659054: step: 2930/1541, loss: 0.34687575697898865 2023-01-23 14:59:30.542626: step: 2932/1541, loss: 0.08714588731527328 2023-01-23 14:59:31.407812: step: 2934/1541, loss: 0.024798836559057236 2023-01-23 14:59:32.286805: step: 2936/1541, loss: 0.002601565094664693 2023-01-23 14:59:33.160581: step: 2938/1541, loss: 0.0007752372184768319 2023-01-23 14:59:34.114304: step: 2940/1541, loss: 0.12251527607440948 2023-01-23 14:59:34.998399: step: 2942/1541, loss: 0.014808639883995056 2023-01-23 14:59:35.919065: step: 2944/1541, loss: 0.5421198606491089 2023-01-23 14:59:36.828474: step: 2946/1541, loss: 0.07623634487390518 2023-01-23 14:59:37.774712: step: 2948/1541, loss: 0.13078950345516205 2023-01-23 14:59:38.675071: step: 2950/1541, loss: 0.040155861526727676 2023-01-23 14:59:39.592170: step: 2952/1541, loss: 0.07593698799610138 2023-01-23 14:59:40.528346: step: 2954/1541, loss: 0.003941756673157215 2023-01-23 14:59:41.455011: step: 2956/1541, loss: 0.04240450635552406 2023-01-23 14:59:42.314326: step: 2958/1541, loss: 0.020329032093286514 2023-01-23 14:59:43.217080: step: 2960/1541, loss: 0.003924550488591194 2023-01-23 14:59:44.082311: step: 2962/1541, loss: 0.020334361121058464 2023-01-23 14:59:44.950261: step: 2964/1541, loss: 0.07669264823198318 2023-01-23 14:59:45.868252: step: 2966/1541, loss: 0.0016050880076363683 2023-01-23 14:59:46.702531: step: 2968/1541, loss: 0.015250638127326965 2023-01-23 14:59:47.583047: step: 2970/1541, loss: 0.04493734613060951 2023-01-23 14:59:48.444247: step: 2972/1541, loss: 0.0012614306760951877 2023-01-23 14:59:49.330721: step: 2974/1541, loss: 1.0621106624603271 2023-01-23 14:59:50.200699: step: 2976/1541, loss: 0.07131733000278473 2023-01-23 14:59:51.182759: step: 2978/1541, loss: 0.004439829383045435 2023-01-23 14:59:52.076005: step: 2980/1541, loss: 0.015477892011404037 2023-01-23 14:59:52.959257: step: 2982/1541, loss: 0.03791234642267227 2023-01-23 14:59:53.869924: step: 2984/1541, loss: 0.0008310953271575272 2023-01-23 14:59:54.778058: step: 2986/1541, loss: 0.006783580407500267 2023-01-23 14:59:55.681329: step: 2988/1541, loss: 0.02166316844522953 2023-01-23 14:59:56.560379: step: 2990/1541, loss: 0.010515798814594746 2023-01-23 14:59:57.431711: step: 2992/1541, loss: 0.01114215049892664 2023-01-23 14:59:58.336464: step: 2994/1541, loss: 0.41438132524490356 2023-01-23 14:59:59.195591: step: 2996/1541, loss: 0.3439897298812866 2023-01-23 15:00:00.135613: step: 2998/1541, loss: 0.10471116751432419 2023-01-23 15:00:01.082201: step: 3000/1541, loss: 0.010718286037445068 2023-01-23 15:00:01.984781: step: 3002/1541, loss: 0.08742348849773407 2023-01-23 15:00:02.876545: step: 3004/1541, loss: 0.009718884713947773 2023-01-23 15:00:03.768249: step: 3006/1541, loss: 0.5421129465103149 2023-01-23 15:00:04.666336: step: 3008/1541, loss: 0.10312153398990631 2023-01-23 15:00:05.621556: step: 3010/1541, loss: 0.0018611929845064878 2023-01-23 15:00:06.539761: step: 3012/1541, loss: 0.14704322814941406 2023-01-23 15:00:07.503321: step: 3014/1541, loss: 0.00523466756567359 2023-01-23 15:00:08.433348: step: 3016/1541, loss: 0.02458316460251808 2023-01-23 15:00:09.301579: step: 3018/1541, loss: 2.2427000999450684 2023-01-23 15:00:10.251830: step: 3020/1541, loss: 0.11684629321098328 2023-01-23 15:00:11.118483: step: 3022/1541, loss: 0.26754140853881836 2023-01-23 15:00:12.030505: step: 3024/1541, loss: 0.07577371597290039 2023-01-23 15:00:12.946172: step: 3026/1541, loss: 0.11517138034105301 2023-01-23 15:00:13.834782: step: 3028/1541, loss: 0.028702549636363983 2023-01-23 15:00:14.752193: step: 3030/1541, loss: 0.009930908679962158 2023-01-23 15:00:15.615448: step: 3032/1541, loss: 0.05687723308801651 2023-01-23 15:00:16.479882: step: 3034/1541, loss: 0.04268195480108261 2023-01-23 15:00:17.399987: step: 3036/1541, loss: 0.07040783017873764 2023-01-23 15:00:18.302897: step: 3038/1541, loss: 0.06595136970281601 2023-01-23 15:00:19.249253: step: 3040/1541, loss: 0.0005041819531470537 2023-01-23 15:00:20.130013: step: 3042/1541, loss: 0.0745711475610733 2023-01-23 15:00:21.028303: step: 3044/1541, loss: 0.014048121869564056 2023-01-23 15:00:21.946533: step: 3046/1541, loss: 0.0019113331800326705 2023-01-23 15:00:22.823916: step: 3048/1541, loss: 0.8437151312828064 2023-01-23 15:00:23.728166: step: 3050/1541, loss: 0.010072539560496807 2023-01-23 15:00:24.636940: step: 3052/1541, loss: 0.005364871118217707 2023-01-23 15:00:25.584315: step: 3054/1541, loss: 0.005777916871011257 2023-01-23 15:00:26.477262: step: 3056/1541, loss: 0.012439732439815998 2023-01-23 15:00:27.361220: step: 3058/1541, loss: 0.15375670790672302 2023-01-23 15:00:28.253356: step: 3060/1541, loss: 4.200720650260337e-05 2023-01-23 15:00:29.124585: step: 3062/1541, loss: 0.00578643660992384 2023-01-23 15:00:30.040832: step: 3064/1541, loss: 0.051215898245573044 2023-01-23 15:00:30.924523: step: 3066/1541, loss: 0.506492555141449 2023-01-23 15:00:31.791945: step: 3068/1541, loss: 0.009031414054334164 2023-01-23 15:00:32.654220: step: 3070/1541, loss: 0.03053055703639984 2023-01-23 15:00:33.546875: step: 3072/1541, loss: 0.019809141755104065 2023-01-23 15:00:34.441373: step: 3074/1541, loss: 0.0027076182886958122 2023-01-23 15:00:35.346672: step: 3076/1541, loss: 0.4735943675041199 2023-01-23 15:00:36.254057: step: 3078/1541, loss: 0.05603440850973129 2023-01-23 15:00:37.084061: step: 3080/1541, loss: 0.0005822200328111649 2023-01-23 15:00:37.961632: step: 3082/1541, loss: 0.00027684977976605296 2023-01-23 15:00:38.733586: step: 3084/1541, loss: 0.024751314893364906 ================================================== Loss: 0.098 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6770057855108976, 'r': 0.6476391519023692, 'f1': 0.6619969478273624}, 'combined': 0.43771838489844495, 'epoch': 20} Test Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.7142728955736691, 'r': 0.7007420429070641, 'f1': 0.7074427759396437}, 'combined': 0.4600061959484325, 'epoch': 20} Dev Korean: {'event': {'p': 0.5735294117647058, 'r': 0.7789613848202397, 'f1': 0.6606437041219649}, 'argument': {'p': 0.6711870580943137, 'r': 0.6601208915450885, 'f1': 0.6656079824397532}, 'combined': 0.4397297230121463, 'epoch': 20} Test Korean: {'event': {'p': 0.6101048617731173, 'r': 0.7669263031755542, 'f1': 0.6795858773559862}, 'argument': {'p': 0.7133488674413241, 'r': 0.7173550386870458, 'f1': 0.7153463441439948}, 'combined': 0.4861392728984939, 'epoch': 20} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6753575348562465, 'r': 0.6641946003958127, 'f1': 0.6697295553991112}, 'combined': 0.4644018108284746, 'epoch': 20} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.7129042236218599, 'r': 0.7052330877866175, 'f1': 0.7090479079513485}, 'combined': 0.4915801417696221, 'epoch': 20} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6663223140495866, 'r': 0.6486323411102171, 'f1': 0.6573583367305339}, 'combined': 0.4719495750885884, 'epoch': 20} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5610974274767379, 'r': 0.4226448155019584, 'f1': 0.4821281599059377}, 'combined': 0.28092506955149127, 'epoch': 20} Russian: {'event': {'p': 0.47058823529411764, 'r': 0.4444444444444444, 'f1': 0.45714285714285713}, 'argument': {'p': 0.5890434478646186, 'r': 0.5890434478646186, 'f1': 0.5890434478646186}, 'combined': 0.26927700473811134, 'epoch': 20} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test for Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 21 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 15:06:02.003303: step: 2/1541, loss: 0.0011990267084911466 2023-01-23 15:06:02.902138: step: 4/1541, loss: 0.00042786149424500763 2023-01-23 15:06:03.795226: step: 6/1541, loss: 0.10121322423219681 2023-01-23 15:06:04.697327: step: 8/1541, loss: 0.0012468999484553933 2023-01-23 15:06:05.605596: step: 10/1541, loss: 0.020397068932652473 2023-01-23 15:06:06.517387: step: 12/1541, loss: 0.0006474516121670604 2023-01-23 15:06:07.445181: step: 14/1541, loss: 0.101148322224617 2023-01-23 15:06:08.393722: step: 16/1541, loss: 0.0017270916141569614 2023-01-23 15:06:09.295696: step: 18/1541, loss: 0.01045075710862875 2023-01-23 15:06:10.184345: step: 20/1541, loss: 0.027819814160466194 2023-01-23 15:06:11.058924: step: 22/1541, loss: 0.0002745594538282603 2023-01-23 15:06:11.936252: step: 24/1541, loss: 0.004399345722049475 2023-01-23 15:06:12.877432: step: 26/1541, loss: 0.08101390302181244 2023-01-23 15:06:13.750959: step: 28/1541, loss: 0.1762433797121048 2023-01-23 15:06:14.658928: step: 30/1541, loss: 9.79471078608185e-05 2023-01-23 15:06:15.543179: step: 32/1541, loss: 8.51136283017695e-05 2023-01-23 15:06:16.382316: step: 34/1541, loss: 0.0009479433065280318 2023-01-23 15:06:17.351256: step: 36/1541, loss: 0.09414960443973541 2023-01-23 15:06:18.328893: step: 38/1541, loss: 0.28245803713798523 2023-01-23 15:06:19.250908: step: 40/1541, loss: 0.022897064685821533 2023-01-23 15:06:20.092615: step: 42/1541, loss: 0.0022249657195061445 2023-01-23 15:06:20.974801: step: 44/1541, loss: 7.0547430368606e-05 2023-01-23 15:06:21.915046: step: 46/1541, loss: 0.000556585262529552 2023-01-23 15:06:22.819651: step: 48/1541, loss: 0.037905409932136536 2023-01-23 15:06:23.717308: step: 50/1541, loss: 0.38887324929237366 2023-01-23 15:06:24.614764: step: 52/1541, loss: 0.002162596210837364 2023-01-23 15:06:25.546468: step: 54/1541, loss: 0.003554933937266469 2023-01-23 15:06:26.442811: step: 56/1541, loss: 0.11137191206216812 2023-01-23 15:06:27.414701: step: 58/1541, loss: 0.2669520974159241 2023-01-23 15:06:28.304901: step: 60/1541, loss: 0.058349236845970154 2023-01-23 15:06:29.195496: step: 62/1541, loss: 1.1632671356201172 2023-01-23 15:06:30.068027: step: 64/1541, loss: 0.0009888026397675276 2023-01-23 15:06:31.013117: step: 66/1541, loss: 0.0004744832986034453 2023-01-23 15:06:31.941944: step: 68/1541, loss: 0.06755787134170532 2023-01-23 15:06:32.882207: step: 70/1541, loss: 0.009365138597786427 2023-01-23 15:06:33.794666: step: 72/1541, loss: 0.351177841424942 2023-01-23 15:06:34.667668: step: 74/1541, loss: 0.0022500469349324703 2023-01-23 15:06:35.569656: step: 76/1541, loss: 0.00027731855516321957 2023-01-23 15:06:36.471690: step: 78/1541, loss: 0.0013600732199847698 2023-01-23 15:06:37.350807: step: 80/1541, loss: 9.97503666440025e-05 2023-01-23 15:06:38.227179: step: 82/1541, loss: 0.014345496892929077 2023-01-23 15:06:39.163448: step: 84/1541, loss: 0.00016363004397135228 2023-01-23 15:06:40.069439: step: 86/1541, loss: 0.015778005123138428 2023-01-23 15:06:41.039843: step: 88/1541, loss: 0.0017396665643900633 2023-01-23 15:06:41.941671: step: 90/1541, loss: 0.000345643114997074 2023-01-23 15:06:42.796508: step: 92/1541, loss: 0.0065606762655079365 2023-01-23 15:06:43.650149: step: 94/1541, loss: 0.34636712074279785 2023-01-23 15:06:44.571823: step: 96/1541, loss: 0.019146045669913292 2023-01-23 15:06:45.455953: step: 98/1541, loss: 1.2496562004089355 2023-01-23 15:06:46.379443: step: 100/1541, loss: 0.24774792790412903 2023-01-23 15:06:47.332394: step: 102/1541, loss: 0.014011047780513763 2023-01-23 15:06:48.272866: step: 104/1541, loss: 0.03177105635404587 2023-01-23 15:06:49.127929: step: 106/1541, loss: 0.19699063897132874 2023-01-23 15:06:50.086892: step: 108/1541, loss: 0.0024057046975940466 2023-01-23 15:06:50.924962: step: 110/1541, loss: 0.0012599639594554901 2023-01-23 15:06:51.837163: step: 112/1541, loss: 0.015374542213976383 2023-01-23 15:06:52.757700: step: 114/1541, loss: 0.07005660235881805 2023-01-23 15:06:53.621624: step: 116/1541, loss: 0.18576498329639435 2023-01-23 15:06:54.555135: step: 118/1541, loss: 0.38924640417099 2023-01-23 15:06:55.463351: step: 120/1541, loss: 0.2114572525024414 2023-01-23 15:06:56.373904: step: 122/1541, loss: 0.008413828909397125 2023-01-23 15:06:57.305146: step: 124/1541, loss: 0.11817087233066559 2023-01-23 15:06:58.205123: step: 126/1541, loss: 0.06619452685117722 2023-01-23 15:06:59.081487: step: 128/1541, loss: 0.04748934507369995 2023-01-23 15:07:00.004182: step: 130/1541, loss: 0.001916081178933382 2023-01-23 15:07:00.911487: step: 132/1541, loss: 0.00820173416286707 2023-01-23 15:07:01.818788: step: 134/1541, loss: 0.2690299451351166 2023-01-23 15:07:02.704116: step: 136/1541, loss: 0.0026788266841322184 2023-01-23 15:07:03.605569: step: 138/1541, loss: 0.011912638321518898 2023-01-23 15:07:04.507957: step: 140/1541, loss: 0.001692978898063302 2023-01-23 15:07:05.391597: step: 142/1541, loss: 0.028390316292643547 2023-01-23 15:07:06.269676: step: 144/1541, loss: 0.01949390396475792 2023-01-23 15:07:07.192527: step: 146/1541, loss: 0.43842461705207825 2023-01-23 15:07:08.078687: step: 148/1541, loss: 0.03052900731563568 2023-01-23 15:07:08.979989: step: 150/1541, loss: 0.002883505541831255 2023-01-23 15:07:09.876451: step: 152/1541, loss: 0.0012066828785464168 2023-01-23 15:07:10.825895: step: 154/1541, loss: 0.0010128291323781013 2023-01-23 15:07:11.748909: step: 156/1541, loss: 0.004937862046062946 2023-01-23 15:07:12.691948: step: 158/1541, loss: 0.004134683404117823 2023-01-23 15:07:13.592761: step: 160/1541, loss: 0.06627027690410614 2023-01-23 15:07:14.475283: step: 162/1541, loss: 0.003791474737226963 2023-01-23 15:07:15.357777: step: 164/1541, loss: 0.0033631448168307543 2023-01-23 15:07:16.295712: step: 166/1541, loss: 0.0862814411520958 2023-01-23 15:07:17.186167: step: 168/1541, loss: 0.0023829375859349966 2023-01-23 15:07:18.104112: step: 170/1541, loss: 0.011874639429152012 2023-01-23 15:07:18.996821: step: 172/1541, loss: 0.0487586110830307 2023-01-23 15:07:19.857882: step: 174/1541, loss: 0.08548683673143387 2023-01-23 15:07:20.715660: step: 176/1541, loss: 0.003095985157415271 2023-01-23 15:07:21.636158: step: 178/1541, loss: 0.0006548169767484069 2023-01-23 15:07:22.530029: step: 180/1541, loss: 0.000535095576196909 2023-01-23 15:07:23.391644: step: 182/1541, loss: 0.009127434343099594 2023-01-23 15:07:24.253438: step: 184/1541, loss: 0.7450195550918579 2023-01-23 15:07:25.139763: step: 186/1541, loss: 0.000905234890524298 2023-01-23 15:07:26.109918: step: 188/1541, loss: 0.010732199065387249 2023-01-23 15:07:27.077365: step: 190/1541, loss: 0.00048294427688233554 2023-01-23 15:07:27.950700: step: 192/1541, loss: 0.012497760355472565 2023-01-23 15:07:28.838597: step: 194/1541, loss: 0.0012936362763866782 2023-01-23 15:07:29.690770: step: 196/1541, loss: 0.0005978674162179232 2023-01-23 15:07:30.567241: step: 198/1541, loss: 0.00229706522077322 2023-01-23 15:07:31.467856: step: 200/1541, loss: 0.0006457269773818552 2023-01-23 15:07:32.376568: step: 202/1541, loss: 1.2857543230056763 2023-01-23 15:07:33.276667: step: 204/1541, loss: 0.0004462992656044662 2023-01-23 15:07:34.180394: step: 206/1541, loss: 0.11760401725769043 2023-01-23 15:07:35.079018: step: 208/1541, loss: 0.22261394560337067 2023-01-23 15:07:35.952345: step: 210/1541, loss: 0.010487932711839676 2023-01-23 15:07:36.918116: step: 212/1541, loss: 0.22909249365329742 2023-01-23 15:07:37.794445: step: 214/1541, loss: 0.16093489527702332 2023-01-23 15:07:38.687316: step: 216/1541, loss: 0.01873292773962021 2023-01-23 15:07:39.542903: step: 218/1541, loss: 0.005477495491504669 2023-01-23 15:07:40.523470: step: 220/1541, loss: 0.018609071150422096 2023-01-23 15:07:41.530408: step: 222/1541, loss: 0.0014006440760567784 2023-01-23 15:07:42.410095: step: 224/1541, loss: 0.030690571293234825 2023-01-23 15:07:43.313131: step: 226/1541, loss: 0.05086562782526016 2023-01-23 15:07:44.194857: step: 228/1541, loss: 0.0002931020862888545 2023-01-23 15:07:45.060749: step: 230/1541, loss: 0.16040226817131042 2023-01-23 15:07:45.966582: step: 232/1541, loss: 0.022589175030589104 2023-01-23 15:07:46.806587: step: 234/1541, loss: 0.005852223839610815 2023-01-23 15:07:47.687009: step: 236/1541, loss: 0.0010832735570147634 2023-01-23 15:07:48.586038: step: 238/1541, loss: 0.13038556277751923 2023-01-23 15:07:49.462878: step: 240/1541, loss: 0.6126219630241394 2023-01-23 15:07:50.326540: step: 242/1541, loss: 0.4800771474838257 2023-01-23 15:07:51.194584: step: 244/1541, loss: 0.0010986713459715247 2023-01-23 15:07:52.080925: step: 246/1541, loss: 0.037882834672927856 2023-01-23 15:07:52.961401: step: 248/1541, loss: 0.09362725913524628 2023-01-23 15:07:53.820438: step: 250/1541, loss: 0.0077874306589365005 2023-01-23 15:07:54.697278: step: 252/1541, loss: 0.00765667948871851 2023-01-23 15:07:55.575908: step: 254/1541, loss: 0.01470893993973732 2023-01-23 15:07:56.453008: step: 256/1541, loss: 0.001712208497337997 2023-01-23 15:07:57.333952: step: 258/1541, loss: 0.0038835725281387568 2023-01-23 15:07:58.244429: step: 260/1541, loss: 0.0033585275523364544 2023-01-23 15:07:59.174982: step: 262/1541, loss: 0.07666425406932831 2023-01-23 15:08:00.028412: step: 264/1541, loss: 0.22433213889598846 2023-01-23 15:08:00.898913: step: 266/1541, loss: 0.02042575180530548 2023-01-23 15:08:01.798809: step: 268/1541, loss: 0.016608189791440964 2023-01-23 15:08:02.656012: step: 270/1541, loss: 0.00144242774695158 2023-01-23 15:08:03.513668: step: 272/1541, loss: 0.0034311015624552965 2023-01-23 15:08:04.464725: step: 274/1541, loss: 0.0037544085644185543 2023-01-23 15:08:05.376366: step: 276/1541, loss: 1.4993818998336792 2023-01-23 15:08:06.254351: step: 278/1541, loss: 0.00465178769081831 2023-01-23 15:08:07.142448: step: 280/1541, loss: 0.0027188207022845745 2023-01-23 15:08:08.035116: step: 282/1541, loss: 0.012935079634189606 2023-01-23 15:08:08.930841: step: 284/1541, loss: 0.059758834540843964 2023-01-23 15:08:09.821236: step: 286/1541, loss: 0.03388135880231857 2023-01-23 15:08:10.780532: step: 288/1541, loss: 0.022333772853016853 2023-01-23 15:08:11.679967: step: 290/1541, loss: 0.00042055590893141925 2023-01-23 15:08:12.646830: step: 292/1541, loss: 0.08371367305517197 2023-01-23 15:08:13.556265: step: 294/1541, loss: 0.0029541547410190105 2023-01-23 15:08:14.459511: step: 296/1541, loss: 0.03520787134766579 2023-01-23 15:08:15.365517: step: 298/1541, loss: 0.11368819326162338 2023-01-23 15:08:16.254939: step: 300/1541, loss: 0.11594811081886292 2023-01-23 15:08:17.105148: step: 302/1541, loss: 0.1410096287727356 2023-01-23 15:08:17.997156: step: 304/1541, loss: 0.03859193995594978 2023-01-23 15:08:18.846153: step: 306/1541, loss: 0.4490327537059784 2023-01-23 15:08:19.768383: step: 308/1541, loss: 0.0018833179492503405 2023-01-23 15:08:20.668616: step: 310/1541, loss: 0.22009903192520142 2023-01-23 15:08:21.577594: step: 312/1541, loss: 0.014851275831460953 2023-01-23 15:08:22.515478: step: 314/1541, loss: 0.005942720919847488 2023-01-23 15:08:23.442432: step: 316/1541, loss: 0.05596879497170448 2023-01-23 15:08:24.397132: step: 318/1541, loss: 0.04911487549543381 2023-01-23 15:08:25.302669: step: 320/1541, loss: 0.004023282323032618 2023-01-23 15:08:26.201339: step: 322/1541, loss: 0.02931644394993782 2023-01-23 15:08:27.085920: step: 324/1541, loss: 0.0031165797263383865 2023-01-23 15:08:28.076702: step: 326/1541, loss: 0.0002236727304989472 2023-01-23 15:08:28.961046: step: 328/1541, loss: 0.11152026802301407 2023-01-23 15:08:29.835826: step: 330/1541, loss: 0.003432866418734193 2023-01-23 15:08:30.768798: step: 332/1541, loss: 0.2989652752876282 2023-01-23 15:08:31.679907: step: 334/1541, loss: 0.001529094995930791 2023-01-23 15:08:32.601159: step: 336/1541, loss: 0.005635582841932774 2023-01-23 15:08:33.521681: step: 338/1541, loss: 0.1349440962076187 2023-01-23 15:08:34.410168: step: 340/1541, loss: 0.07450185716152191 2023-01-23 15:08:35.293106: step: 342/1541, loss: 0.06885401159524918 2023-01-23 15:08:36.221516: step: 344/1541, loss: 0.2559170126914978 2023-01-23 15:08:37.125949: step: 346/1541, loss: 0.0061507392674684525 2023-01-23 15:08:37.997927: step: 348/1541, loss: 0.010455180890858173 2023-01-23 15:08:38.947041: step: 350/1541, loss: 0.006346748676151037 2023-01-23 15:08:39.829111: step: 352/1541, loss: 0.04940086975693703 2023-01-23 15:08:40.778748: step: 354/1541, loss: 0.008076767437160015 2023-01-23 15:08:41.649087: step: 356/1541, loss: 0.06210654601454735 2023-01-23 15:08:42.545008: step: 358/1541, loss: 0.0778910368680954 2023-01-23 15:08:43.441880: step: 360/1541, loss: 0.029408659785985947 2023-01-23 15:08:44.408355: step: 362/1541, loss: 0.0013437350280582905 2023-01-23 15:08:45.379690: step: 364/1541, loss: 0.09967788308858871 2023-01-23 15:08:46.269438: step: 366/1541, loss: 0.039800748229026794 2023-01-23 15:08:47.231244: step: 368/1541, loss: 0.013152480125427246 2023-01-23 15:08:48.111725: step: 370/1541, loss: 0.02926313690841198 2023-01-23 15:08:49.054827: step: 372/1541, loss: 0.01813555136322975 2023-01-23 15:08:49.935429: step: 374/1541, loss: 0.0006680453661829233 2023-01-23 15:08:50.825586: step: 376/1541, loss: 0.2679150104522705 2023-01-23 15:08:51.715345: step: 378/1541, loss: 0.0019676759839057922 2023-01-23 15:08:52.595450: step: 380/1541, loss: 0.14282360672950745 2023-01-23 15:08:53.452938: step: 382/1541, loss: 0.0016095072496682405 2023-01-23 15:08:54.339844: step: 384/1541, loss: 0.009469607844948769 2023-01-23 15:08:55.251783: step: 386/1541, loss: 0.6458147764205933 2023-01-23 15:08:56.157072: step: 388/1541, loss: 0.086551234126091 2023-01-23 15:08:57.013336: step: 390/1541, loss: 0.00052668183343485 2023-01-23 15:08:57.861853: step: 392/1541, loss: 0.08805432170629501 2023-01-23 15:08:58.761739: step: 394/1541, loss: 0.0047036330215632915 2023-01-23 15:08:59.629993: step: 396/1541, loss: 0.07499434053897858 2023-01-23 15:09:00.575772: step: 398/1541, loss: 0.16647474467754364 2023-01-23 15:09:01.464274: step: 400/1541, loss: 0.1452430784702301 2023-01-23 15:09:02.418950: step: 402/1541, loss: 0.002591511933133006 2023-01-23 15:09:03.296944: step: 404/1541, loss: 0.02591870166361332 2023-01-23 15:09:04.250852: step: 406/1541, loss: 0.8551899194717407 2023-01-23 15:09:05.137389: step: 408/1541, loss: 0.08695283532142639 2023-01-23 15:09:06.045583: step: 410/1541, loss: 0.002065641339868307 2023-01-23 15:09:06.947470: step: 412/1541, loss: 0.0070839812979102135 2023-01-23 15:09:07.852107: step: 414/1541, loss: 1.2415812015533447 2023-01-23 15:09:08.713121: step: 416/1541, loss: 0.08537326008081436 2023-01-23 15:09:09.628077: step: 418/1541, loss: 0.1566191017627716 2023-01-23 15:09:10.532945: step: 420/1541, loss: 0.23963065445423126 2023-01-23 15:09:11.522741: step: 422/1541, loss: 0.0017137707909569144 2023-01-23 15:09:12.423082: step: 424/1541, loss: 0.2730739116668701 2023-01-23 15:09:13.328041: step: 426/1541, loss: 0.035109832882881165 2023-01-23 15:09:14.192594: step: 428/1541, loss: 0.02556079439818859 2023-01-23 15:09:15.066031: step: 430/1541, loss: 0.07086921483278275 2023-01-23 15:09:15.961633: step: 432/1541, loss: 2.3140852451324463 2023-01-23 15:09:16.944575: step: 434/1541, loss: 0.8511379361152649 2023-01-23 15:09:17.825749: step: 436/1541, loss: 0.19943830370903015 2023-01-23 15:09:18.778096: step: 438/1541, loss: 0.013092043809592724 2023-01-23 15:09:19.657113: step: 440/1541, loss: 0.06279159337282181 2023-01-23 15:09:20.547276: step: 442/1541, loss: 0.013380142860114574 2023-01-23 15:09:21.430031: step: 444/1541, loss: 0.058334846049547195 2023-01-23 15:09:22.369754: step: 446/1541, loss: 0.11164559423923492 2023-01-23 15:09:23.255399: step: 448/1541, loss: 0.0022205396089702845 2023-01-23 15:09:24.190668: step: 450/1541, loss: 0.06131052225828171 2023-01-23 15:09:25.084993: step: 452/1541, loss: 0.08720431476831436 2023-01-23 15:09:26.000853: step: 454/1541, loss: 0.25271618366241455 2023-01-23 15:09:26.917744: step: 456/1541, loss: 0.13380488753318787 2023-01-23 15:09:27.792258: step: 458/1541, loss: 0.022008661180734634 2023-01-23 15:09:28.631174: step: 460/1541, loss: 0.00918544176965952 2023-01-23 15:09:29.495678: step: 462/1541, loss: 0.001129167270846665 2023-01-23 15:09:30.377001: step: 464/1541, loss: 0.01765168458223343 2023-01-23 15:09:31.296232: step: 466/1541, loss: 0.008215092122554779 2023-01-23 15:09:32.240461: step: 468/1541, loss: 0.00020727200899273157 2023-01-23 15:09:33.163035: step: 470/1541, loss: 0.06599970161914825 2023-01-23 15:09:34.097428: step: 472/1541, loss: 0.1851719617843628 2023-01-23 15:09:34.987531: step: 474/1541, loss: 0.1434284895658493 2023-01-23 15:09:35.855593: step: 476/1541, loss: 0.24368968605995178 2023-01-23 15:09:36.762026: step: 478/1541, loss: 0.16280612349510193 2023-01-23 15:09:37.669279: step: 480/1541, loss: 0.0032717955764383078 2023-01-23 15:09:38.605121: step: 482/1541, loss: 0.058278586715459824 2023-01-23 15:09:39.477764: step: 484/1541, loss: 0.0334722064435482 2023-01-23 15:09:40.362793: step: 486/1541, loss: 0.02207985147833824 2023-01-23 15:09:41.355890: step: 488/1541, loss: 0.025939008221030235 2023-01-23 15:09:42.233047: step: 490/1541, loss: 0.000266008049948141 2023-01-23 15:09:43.088611: step: 492/1541, loss: 0.43003377318382263 2023-01-23 15:09:43.946395: step: 494/1541, loss: 0.005835094954818487 2023-01-23 15:09:44.797850: step: 496/1541, loss: 0.1752420961856842 2023-01-23 15:09:45.700076: step: 498/1541, loss: 0.003541880752891302 2023-01-23 15:09:46.588632: step: 500/1541, loss: 0.12562163174152374 2023-01-23 15:09:47.453438: step: 502/1541, loss: 0.3411174714565277 2023-01-23 15:09:48.318670: step: 504/1541, loss: 0.01764141023159027 2023-01-23 15:09:49.175003: step: 506/1541, loss: 0.11115523427724838 2023-01-23 15:09:50.075202: step: 508/1541, loss: 0.00041867265827022493 2023-01-23 15:09:51.012896: step: 510/1541, loss: 1.2702126502990723 2023-01-23 15:09:51.891548: step: 512/1541, loss: 0.29114463925361633 2023-01-23 15:09:52.819279: step: 514/1541, loss: 0.0008013901533558965 2023-01-23 15:09:53.713983: step: 516/1541, loss: 0.0014652865938842297 2023-01-23 15:09:54.650009: step: 518/1541, loss: 0.009880991652607918 2023-01-23 15:09:55.539042: step: 520/1541, loss: 0.0006847115000709891 2023-01-23 15:09:56.496642: step: 522/1541, loss: 0.334651380777359 2023-01-23 15:09:57.358006: step: 524/1541, loss: 0.007911491207778454 2023-01-23 15:09:58.235870: step: 526/1541, loss: 0.14467595517635345 2023-01-23 15:09:59.148093: step: 528/1541, loss: 0.0017179420683532953 2023-01-23 15:10:00.073819: step: 530/1541, loss: 0.003528907895088196 2023-01-23 15:10:00.991588: step: 532/1541, loss: 0.02517286315560341 2023-01-23 15:10:01.892390: step: 534/1541, loss: 0.0034531597048044205 2023-01-23 15:10:02.769103: step: 536/1541, loss: 0.09892293810844421 2023-01-23 15:10:03.665906: step: 538/1541, loss: 0.043109774589538574 2023-01-23 15:10:04.562271: step: 540/1541, loss: 0.10375165194272995 2023-01-23 15:10:05.428846: step: 542/1541, loss: 0.0087313586845994 2023-01-23 15:10:06.314368: step: 544/1541, loss: 0.014228345826268196 2023-01-23 15:10:07.193235: step: 546/1541, loss: 0.11929180473089218 2023-01-23 15:10:08.043255: step: 548/1541, loss: 0.00384926563128829 2023-01-23 15:10:08.953962: step: 550/1541, loss: 0.15262866020202637 2023-01-23 15:10:09.853610: step: 552/1541, loss: 0.19057750701904297 2023-01-23 15:10:10.721973: step: 554/1541, loss: 0.011409149505198002 2023-01-23 15:10:11.598175: step: 556/1541, loss: 0.009791663847863674 2023-01-23 15:10:12.457922: step: 558/1541, loss: 0.06998512893915176 2023-01-23 15:10:13.335636: step: 560/1541, loss: 0.0013103687670081854 2023-01-23 15:10:14.223351: step: 562/1541, loss: 0.24730443954467773 2023-01-23 15:10:15.101175: step: 564/1541, loss: 0.08451177924871445 2023-01-23 15:10:15.970479: step: 566/1541, loss: 0.08708097785711288 2023-01-23 15:10:16.828424: step: 568/1541, loss: 0.021290691569447517 2023-01-23 15:10:17.761789: step: 570/1541, loss: 0.06651494652032852 2023-01-23 15:10:18.638182: step: 572/1541, loss: 0.0010168298613280058 2023-01-23 15:10:19.574102: step: 574/1541, loss: 0.020128611475229263 2023-01-23 15:10:20.422117: step: 576/1541, loss: 0.006813125219196081 2023-01-23 15:10:21.299015: step: 578/1541, loss: 0.01990417204797268 2023-01-23 15:10:22.173376: step: 580/1541, loss: 0.02213798090815544 2023-01-23 15:10:23.105591: step: 582/1541, loss: 0.006785633973777294 2023-01-23 15:10:24.106911: step: 584/1541, loss: 0.016671739518642426 2023-01-23 15:10:25.000428: step: 586/1541, loss: 0.0005561142461374402 2023-01-23 15:10:25.906258: step: 588/1541, loss: 0.00911678560078144 2023-01-23 15:10:26.832513: step: 590/1541, loss: 0.4846797585487366 2023-01-23 15:10:27.710479: step: 592/1541, loss: 0.12272603064775467 2023-01-23 15:10:28.589999: step: 594/1541, loss: 0.028578560799360275 2023-01-23 15:10:29.475152: step: 596/1541, loss: 0.0008745131781324744 2023-01-23 15:10:30.359200: step: 598/1541, loss: 0.01231854036450386 2023-01-23 15:10:31.236606: step: 600/1541, loss: 0.01258174329996109 2023-01-23 15:10:32.178296: step: 602/1541, loss: 0.02927273139357567 2023-01-23 15:10:33.063326: step: 604/1541, loss: 0.00267299497500062 2023-01-23 15:10:33.974270: step: 606/1541, loss: 0.0029813675209879875 2023-01-23 15:10:34.858717: step: 608/1541, loss: 0.003048415994271636 2023-01-23 15:10:35.764655: step: 610/1541, loss: 0.34143295884132385 2023-01-23 15:10:36.656113: step: 612/1541, loss: 0.03686757758259773 2023-01-23 15:10:37.623413: step: 614/1541, loss: 0.0012274947948753834 2023-01-23 15:10:38.603702: step: 616/1541, loss: 0.5073526501655579 2023-01-23 15:10:39.510726: step: 618/1541, loss: 0.7799814939498901 2023-01-23 15:10:40.427190: step: 620/1541, loss: 0.5611529350280762 2023-01-23 15:10:41.356352: step: 622/1541, loss: 0.25081852078437805 2023-01-23 15:10:42.212527: step: 624/1541, loss: 0.05309118703007698 2023-01-23 15:10:43.222661: step: 626/1541, loss: 0.14094354212284088 2023-01-23 15:10:44.122275: step: 628/1541, loss: 0.0072607663460075855 2023-01-23 15:10:45.020515: step: 630/1541, loss: 0.12200459092855453 2023-01-23 15:10:45.918783: step: 632/1541, loss: 0.005097947083413601 2023-01-23 15:10:46.761534: step: 634/1541, loss: 0.00045744929229840636 2023-01-23 15:10:47.653648: step: 636/1541, loss: 0.003435052465647459 2023-01-23 15:10:48.525886: step: 638/1541, loss: 0.01088061835616827 2023-01-23 15:10:49.405354: step: 640/1541, loss: 0.059293232858181 2023-01-23 15:10:50.293050: step: 642/1541, loss: 0.23120327293872833 2023-01-23 15:10:51.179607: step: 644/1541, loss: 0.09803294390439987 2023-01-23 15:10:52.070587: step: 646/1541, loss: 0.009853905998170376 2023-01-23 15:10:52.962966: step: 648/1541, loss: 0.1092943400144577 2023-01-23 15:10:53.846848: step: 650/1541, loss: 0.004936679266393185 2023-01-23 15:10:54.742178: step: 652/1541, loss: 0.03929657116532326 2023-01-23 15:10:55.625085: step: 654/1541, loss: 0.00039303075755015016 2023-01-23 15:10:56.536176: step: 656/1541, loss: 0.0020686050411313772 2023-01-23 15:10:57.416946: step: 658/1541, loss: 0.23698310554027557 2023-01-23 15:10:58.367664: step: 660/1541, loss: 0.0006147706881165504 2023-01-23 15:10:59.259426: step: 662/1541, loss: 0.071842722594738 2023-01-23 15:11:00.142979: step: 664/1541, loss: 0.00846558716148138 2023-01-23 15:11:01.099643: step: 666/1541, loss: 0.19330541789531708 2023-01-23 15:11:01.986961: step: 668/1541, loss: 0.032198768109083176 2023-01-23 15:11:02.888300: step: 670/1541, loss: 0.005645616911351681 2023-01-23 15:11:03.781353: step: 672/1541, loss: 0.0063369120471179485 2023-01-23 15:11:04.663842: step: 674/1541, loss: 0.09135620296001434 2023-01-23 15:11:05.542987: step: 676/1541, loss: 0.20653779804706573 2023-01-23 15:11:06.419867: step: 678/1541, loss: 0.04744414985179901 2023-01-23 15:11:07.444992: step: 680/1541, loss: 0.14563047885894775 2023-01-23 15:11:08.389416: step: 682/1541, loss: 0.006643394008278847 2023-01-23 15:11:09.246107: step: 684/1541, loss: 0.11652883887290955 2023-01-23 15:11:10.139389: step: 686/1541, loss: 0.4743061065673828 2023-01-23 15:11:10.994815: step: 688/1541, loss: 0.01862938143312931 2023-01-23 15:11:11.899868: step: 690/1541, loss: 0.011187397874891758 2023-01-23 15:11:12.773212: step: 692/1541, loss: 0.05044905096292496 2023-01-23 15:11:13.637006: step: 694/1541, loss: 0.007167048752307892 2023-01-23 15:11:14.520757: step: 696/1541, loss: 0.0341748483479023 2023-01-23 15:11:15.408692: step: 698/1541, loss: 0.03196345642209053 2023-01-23 15:11:16.292522: step: 700/1541, loss: 0.052721861749887466 2023-01-23 15:11:17.220539: step: 702/1541, loss: 0.01627536304295063 2023-01-23 15:11:18.108369: step: 704/1541, loss: 0.4756333827972412 2023-01-23 15:11:18.964726: step: 706/1541, loss: 0.011550710536539555 2023-01-23 15:11:19.927897: step: 708/1541, loss: 0.2749364972114563 2023-01-23 15:11:20.839800: step: 710/1541, loss: 0.007908741012215614 2023-01-23 15:11:21.734460: step: 712/1541, loss: 0.0036056337412446737 2023-01-23 15:11:22.607290: step: 714/1541, loss: 0.021530764177441597 2023-01-23 15:11:23.493171: step: 716/1541, loss: 0.00679780263453722 2023-01-23 15:11:24.403315: step: 718/1541, loss: 0.05093572661280632 2023-01-23 15:11:25.288347: step: 720/1541, loss: 0.17320534586906433 2023-01-23 15:11:26.153337: step: 722/1541, loss: 0.035376325249671936 2023-01-23 15:11:27.054412: step: 724/1541, loss: 0.0020370222628116608 2023-01-23 15:11:27.920873: step: 726/1541, loss: 0.6665530204772949 2023-01-23 15:11:28.815043: step: 728/1541, loss: 0.04379317909479141 2023-01-23 15:11:29.688209: step: 730/1541, loss: 0.09181468188762665 2023-01-23 15:11:30.569888: step: 732/1541, loss: 0.0008120510610751808 2023-01-23 15:11:31.444777: step: 734/1541, loss: 0.0007646644371561706 2023-01-23 15:11:32.353546: step: 736/1541, loss: 0.0008823717944324017 2023-01-23 15:11:33.232139: step: 738/1541, loss: 0.005692752543836832 2023-01-23 15:11:34.117286: step: 740/1541, loss: 0.0024052467197179794 2023-01-23 15:11:34.990203: step: 742/1541, loss: 0.05424629524350166 2023-01-23 15:11:35.905610: step: 744/1541, loss: 0.15923964977264404 2023-01-23 15:11:36.825048: step: 746/1541, loss: 0.006516771856695414 2023-01-23 15:11:37.755773: step: 748/1541, loss: 0.07249999791383743 2023-01-23 15:11:38.637446: step: 750/1541, loss: 0.008968139998614788 2023-01-23 15:11:39.543144: step: 752/1541, loss: 0.006572265177965164 2023-01-23 15:11:40.438361: step: 754/1541, loss: 0.0469832643866539 2023-01-23 15:11:41.312197: step: 756/1541, loss: 0.013785737566649914 2023-01-23 15:11:42.250721: step: 758/1541, loss: 0.004211990162730217 2023-01-23 15:11:43.147382: step: 760/1541, loss: 0.136879563331604 2023-01-23 15:11:44.018396: step: 762/1541, loss: 0.7615888118743896 2023-01-23 15:11:44.900935: step: 764/1541, loss: 0.01793927140533924 2023-01-23 15:11:45.794091: step: 766/1541, loss: 0.028659069910645485 2023-01-23 15:11:46.689294: step: 768/1541, loss: 0.36278268694877625 2023-01-23 15:11:47.619754: step: 770/1541, loss: 0.0004338957369327545 2023-01-23 15:11:48.503349: step: 772/1541, loss: 0.0004598631931003183 2023-01-23 15:11:49.390030: step: 774/1541, loss: 0.017684785649180412 2023-01-23 15:11:50.250454: step: 776/1541, loss: 0.03806672245264053 2023-01-23 15:11:51.146843: step: 778/1541, loss: 0.0037458259612321854 2023-01-23 15:11:52.094816: step: 780/1541, loss: 0.0013262873981148005 2023-01-23 15:11:52.968998: step: 782/1541, loss: 0.00852022785693407 2023-01-23 15:11:53.858327: step: 784/1541, loss: 0.1434069722890854 2023-01-23 15:11:54.750133: step: 786/1541, loss: 0.7961737513542175 2023-01-23 15:11:55.641505: step: 788/1541, loss: 0.005589101929217577 2023-01-23 15:11:56.542104: step: 790/1541, loss: 0.00014698573795612901 2023-01-23 15:11:57.461823: step: 792/1541, loss: 0.3865891098976135 2023-01-23 15:11:58.365004: step: 794/1541, loss: 0.0021199919283390045 2023-01-23 15:11:59.213526: step: 796/1541, loss: 0.0945088192820549 2023-01-23 15:12:00.128127: step: 798/1541, loss: 0.0005023150588385761 2023-01-23 15:12:01.026628: step: 800/1541, loss: 0.02551806904375553 2023-01-23 15:12:01.970597: step: 802/1541, loss: 0.004954714328050613 2023-01-23 15:12:02.862603: step: 804/1541, loss: 0.0009557405719533563 2023-01-23 15:12:03.748892: step: 806/1541, loss: 0.6632092595100403 2023-01-23 15:12:04.666929: step: 808/1541, loss: 0.16704492270946503 2023-01-23 15:12:05.577746: step: 810/1541, loss: 0.0006224154494702816 2023-01-23 15:12:06.496251: step: 812/1541, loss: 0.01800508052110672 2023-01-23 15:12:07.352704: step: 814/1541, loss: 0.00828884169459343 2023-01-23 15:12:08.239084: step: 816/1541, loss: 0.1724957674741745 2023-01-23 15:12:09.124273: step: 818/1541, loss: 0.002932213945314288 2023-01-23 15:12:09.976048: step: 820/1541, loss: 0.0005084481672383845 2023-01-23 15:12:10.886857: step: 822/1541, loss: 0.01970680244266987 2023-01-23 15:12:11.777534: step: 824/1541, loss: 0.048486363142728806 2023-01-23 15:12:12.674322: step: 826/1541, loss: 0.001031744759529829 2023-01-23 15:12:13.627495: step: 828/1541, loss: 0.7278354167938232 2023-01-23 15:12:14.527308: step: 830/1541, loss: 0.00048705688095651567 2023-01-23 15:12:15.476051: step: 832/1541, loss: 0.11179748922586441 2023-01-23 15:12:16.419045: step: 834/1541, loss: 0.0030904209706932306 2023-01-23 15:12:17.337320: step: 836/1541, loss: 0.030235450714826584 2023-01-23 15:12:18.279067: step: 838/1541, loss: 0.0035904659889638424 2023-01-23 15:12:19.132826: step: 840/1541, loss: 0.19818569719791412 2023-01-23 15:12:20.055027: step: 842/1541, loss: 0.022418811917304993 2023-01-23 15:12:20.936653: step: 844/1541, loss: 0.0015750350430607796 2023-01-23 15:12:21.854975: step: 846/1541, loss: 0.040528617799282074 2023-01-23 15:12:22.789600: step: 848/1541, loss: 0.049976132810115814 2023-01-23 15:12:23.707269: step: 850/1541, loss: 0.006958536338061094 2023-01-23 15:12:24.600595: step: 852/1541, loss: 0.36065399646759033 2023-01-23 15:12:25.558876: step: 854/1541, loss: 0.006173859816044569 2023-01-23 15:12:26.452966: step: 856/1541, loss: 0.002554834121838212 2023-01-23 15:12:27.369547: step: 858/1541, loss: 0.0018540986347943544 2023-01-23 15:12:28.255108: step: 860/1541, loss: 0.03354732692241669 2023-01-23 15:12:29.132610: step: 862/1541, loss: 0.0008816237677820027 2023-01-23 15:12:30.053592: step: 864/1541, loss: 0.006127183325588703 2023-01-23 15:12:31.005562: step: 866/1541, loss: 0.0004731749650090933 2023-01-23 15:12:31.951839: step: 868/1541, loss: 0.12190232425928116 2023-01-23 15:12:32.844025: step: 870/1541, loss: 0.00021449604537338018 2023-01-23 15:12:33.729053: step: 872/1541, loss: 0.2909848392009735 2023-01-23 15:12:34.626397: step: 874/1541, loss: 0.016850460320711136 2023-01-23 15:12:35.504782: step: 876/1541, loss: 0.002896638819947839 2023-01-23 15:12:36.382521: step: 878/1541, loss: 0.07419521361589432 2023-01-23 15:12:37.318341: step: 880/1541, loss: 0.3753017485141754 2023-01-23 15:12:38.221945: step: 882/1541, loss: 0.20035724341869354 2023-01-23 15:12:39.097378: step: 884/1541, loss: 0.3851487338542938 2023-01-23 15:12:40.079239: step: 886/1541, loss: 0.03397964686155319 2023-01-23 15:12:40.967639: step: 888/1541, loss: 0.045213717967271805 2023-01-23 15:12:41.909227: step: 890/1541, loss: 0.16431419551372528 2023-01-23 15:12:42.782416: step: 892/1541, loss: 0.21741773188114166 2023-01-23 15:12:43.704433: step: 894/1541, loss: 0.06805141270160675 2023-01-23 15:12:44.632343: step: 896/1541, loss: 0.15845464169979095 2023-01-23 15:12:45.538197: step: 898/1541, loss: 0.08607660979032516 2023-01-23 15:12:46.467946: step: 900/1541, loss: 0.09666237235069275 2023-01-23 15:12:47.368333: step: 902/1541, loss: 0.07957132160663605 2023-01-23 15:12:48.282172: step: 904/1541, loss: 0.055178672075271606 2023-01-23 15:12:49.210419: step: 906/1541, loss: 0.367095947265625 2023-01-23 15:12:50.100468: step: 908/1541, loss: 0.009334810078144073 2023-01-23 15:12:50.978839: step: 910/1541, loss: 0.13025790452957153 2023-01-23 15:12:51.890989: step: 912/1541, loss: 0.06220327317714691 2023-01-23 15:12:52.810021: step: 914/1541, loss: 0.03990568220615387 2023-01-23 15:12:53.684492: step: 916/1541, loss: 0.004698065109550953 2023-01-23 15:12:54.593717: step: 918/1541, loss: 0.038825251162052155 2023-01-23 15:12:55.508823: step: 920/1541, loss: 0.08148296922445297 2023-01-23 15:12:56.495335: step: 922/1541, loss: 0.1334799975156784 2023-01-23 15:12:57.412376: step: 924/1541, loss: 0.058076728135347366 2023-01-23 15:12:58.303746: step: 926/1541, loss: 0.07231292873620987 2023-01-23 15:12:59.181814: step: 928/1541, loss: 0.00945441983640194 2023-01-23 15:13:00.045485: step: 930/1541, loss: 0.0012794127687811852 2023-01-23 15:13:01.050383: step: 932/1541, loss: 0.11363072693347931 2023-01-23 15:13:01.982580: step: 934/1541, loss: 0.05804514139890671 2023-01-23 15:13:03.025835: step: 936/1541, loss: 0.03773319721221924 2023-01-23 15:13:03.926732: step: 938/1541, loss: 0.006986567284911871 2023-01-23 15:13:04.836197: step: 940/1541, loss: 0.006458488758653402 2023-01-23 15:13:05.746654: step: 942/1541, loss: 0.04714707285165787 2023-01-23 15:13:06.648874: step: 944/1541, loss: 0.01141305360943079 2023-01-23 15:13:07.539515: step: 946/1541, loss: 0.18752799928188324 2023-01-23 15:13:08.445733: step: 948/1541, loss: 0.8249396085739136 2023-01-23 15:13:09.457348: step: 950/1541, loss: 0.0061235493049025536 2023-01-23 15:13:10.345256: step: 952/1541, loss: 0.06473521143198013 2023-01-23 15:13:11.234891: step: 954/1541, loss: 0.3077547252178192 2023-01-23 15:13:12.139344: step: 956/1541, loss: 0.0016122767701745033 2023-01-23 15:13:13.074660: step: 958/1541, loss: 0.006062768399715424 2023-01-23 15:13:13.972284: step: 960/1541, loss: 0.001293402281589806 2023-01-23 15:13:14.839371: step: 962/1541, loss: 0.03963524475693703 2023-01-23 15:13:15.715848: step: 964/1541, loss: 0.002190153580158949 2023-01-23 15:13:16.643408: step: 966/1541, loss: 0.17470063269138336 2023-01-23 15:13:17.565634: step: 968/1541, loss: 0.053550608456134796 2023-01-23 15:13:18.455489: step: 970/1541, loss: 0.004209451377391815 2023-01-23 15:13:19.379158: step: 972/1541, loss: 0.04152669757604599 2023-01-23 15:13:20.259178: step: 974/1541, loss: 0.04229845479130745 2023-01-23 15:13:21.135520: step: 976/1541, loss: 0.006339768413454294 2023-01-23 15:13:22.039206: step: 978/1541, loss: 0.17788086831569672 2023-01-23 15:13:22.933518: step: 980/1541, loss: 0.005732354708015919 2023-01-23 15:13:23.829921: step: 982/1541, loss: 0.06824179738759995 2023-01-23 15:13:24.720636: step: 984/1541, loss: 0.09893042594194412 2023-01-23 15:13:25.668743: step: 986/1541, loss: 0.18602705001831055 2023-01-23 15:13:26.549378: step: 988/1541, loss: 0.034448109567165375 2023-01-23 15:13:27.419681: step: 990/1541, loss: 0.013261038810014725 2023-01-23 15:13:28.279185: step: 992/1541, loss: 0.11529054492712021 2023-01-23 15:13:29.185014: step: 994/1541, loss: 0.011201856657862663 2023-01-23 15:13:30.094207: step: 996/1541, loss: 0.0024191972333937883 2023-01-23 15:13:31.077443: step: 998/1541, loss: 0.04927126318216324 2023-01-23 15:13:31.961543: step: 1000/1541, loss: 0.004130347166210413 2023-01-23 15:13:32.881276: step: 1002/1541, loss: 0.0005036441143602133 2023-01-23 15:13:33.778216: step: 1004/1541, loss: 0.05474429577589035 2023-01-23 15:13:34.704610: step: 1006/1541, loss: 0.0004568178264889866 2023-01-23 15:13:35.640886: step: 1008/1541, loss: 0.018091119825839996 2023-01-23 15:13:36.563199: step: 1010/1541, loss: 0.001723811263218522 2023-01-23 15:13:37.437932: step: 1012/1541, loss: 0.08836065232753754 2023-01-23 15:13:38.414882: step: 1014/1541, loss: 0.021579543128609657 2023-01-23 15:13:39.318435: step: 1016/1541, loss: 0.007415829226374626 2023-01-23 15:13:40.187290: step: 1018/1541, loss: 0.004586322233080864 2023-01-23 15:13:41.104965: step: 1020/1541, loss: 0.011461085639894009 2023-01-23 15:13:42.045850: step: 1022/1541, loss: 0.037865862250328064 2023-01-23 15:13:42.895321: step: 1024/1541, loss: 0.010629304684698582 2023-01-23 15:13:43.767257: step: 1026/1541, loss: 0.02617485448718071 2023-01-23 15:13:44.761504: step: 1028/1541, loss: 0.2644825875759125 2023-01-23 15:13:45.683802: step: 1030/1541, loss: 0.06517747789621353 2023-01-23 15:13:46.600756: step: 1032/1541, loss: 0.4197404384613037 2023-01-23 15:13:47.486930: step: 1034/1541, loss: 0.002455874579027295 2023-01-23 15:13:48.352325: step: 1036/1541, loss: 0.02340877801179886 2023-01-23 15:13:49.230554: step: 1038/1541, loss: 0.002559083281084895 2023-01-23 15:13:50.160931: step: 1040/1541, loss: 0.46096742153167725 2023-01-23 15:13:51.054863: step: 1042/1541, loss: 0.0029665569309145212 2023-01-23 15:13:52.021288: step: 1044/1541, loss: 0.0010759884025901556 2023-01-23 15:13:52.929650: step: 1046/1541, loss: 0.11954247206449509 2023-01-23 15:13:53.807783: step: 1048/1541, loss: 0.008997864089906216 2023-01-23 15:13:54.679596: step: 1050/1541, loss: 0.14155207574367523 2023-01-23 15:13:55.549715: step: 1052/1541, loss: 0.003978329245001078 2023-01-23 15:13:56.433093: step: 1054/1541, loss: 0.27148157358169556 2023-01-23 15:13:57.283665: step: 1056/1541, loss: 0.1290607750415802 2023-01-23 15:13:58.192770: step: 1058/1541, loss: 0.001053734333254397 2023-01-23 15:13:59.090722: step: 1060/1541, loss: 0.0018936804262921214 2023-01-23 15:13:59.972769: step: 1062/1541, loss: 0.013831071555614471 2023-01-23 15:14:00.891154: step: 1064/1541, loss: 0.005690793972462416 2023-01-23 15:14:01.777190: step: 1066/1541, loss: 0.0016892293933779001 2023-01-23 15:14:02.688228: step: 1068/1541, loss: 0.026644155383110046 2023-01-23 15:14:03.615811: step: 1070/1541, loss: 0.0002343134838156402 2023-01-23 15:14:04.486938: step: 1072/1541, loss: 0.022785745561122894 2023-01-23 15:14:05.386038: step: 1074/1541, loss: 0.003825685940682888 2023-01-23 15:14:06.281688: step: 1076/1541, loss: 0.0007312914240173995 2023-01-23 15:14:07.210197: step: 1078/1541, loss: 0.00019796544802375138 2023-01-23 15:14:08.069073: step: 1080/1541, loss: 0.0022476434241980314 2023-01-23 15:14:08.928160: step: 1082/1541, loss: 0.3701659142971039 2023-01-23 15:14:09.825825: step: 1084/1541, loss: 0.06709215044975281 2023-01-23 15:14:10.742690: step: 1086/1541, loss: 0.05249050259590149 2023-01-23 15:14:11.650365: step: 1088/1541, loss: 0.00018842265126295388 2023-01-23 15:14:12.565249: step: 1090/1541, loss: 0.0329621285200119 2023-01-23 15:14:13.439617: step: 1092/1541, loss: 0.03438678756356239 2023-01-23 15:14:14.346203: step: 1094/1541, loss: 0.21241942048072815 2023-01-23 15:14:15.245107: step: 1096/1541, loss: 0.01257451344281435 2023-01-23 15:14:16.123001: step: 1098/1541, loss: 0.007906843908131123 2023-01-23 15:14:16.995939: step: 1100/1541, loss: 0.20366176962852478 2023-01-23 15:14:17.883057: step: 1102/1541, loss: 3.014428511960432e-05 2023-01-23 15:14:18.799258: step: 1104/1541, loss: 0.0020553546492010355 2023-01-23 15:14:19.652737: step: 1106/1541, loss: 0.12636172771453857 2023-01-23 15:14:20.545407: step: 1108/1541, loss: 0.0009845596505329013 2023-01-23 15:14:21.550990: step: 1110/1541, loss: 0.005401795729994774 2023-01-23 15:14:22.480949: step: 1112/1541, loss: 0.02454698458313942 2023-01-23 15:14:23.372637: step: 1114/1541, loss: 0.002930421382188797 2023-01-23 15:14:24.280347: step: 1116/1541, loss: 0.0021065310575067997 2023-01-23 15:14:25.194797: step: 1118/1541, loss: 0.0008856424246914685 2023-01-23 15:14:26.075213: step: 1120/1541, loss: 0.020970884710550308 2023-01-23 15:14:27.001638: step: 1122/1541, loss: 0.13574384152889252 2023-01-23 15:14:27.893561: step: 1124/1541, loss: 0.0004834343562833965 2023-01-23 15:14:28.780464: step: 1126/1541, loss: 0.19879594445228577 2023-01-23 15:14:29.700602: step: 1128/1541, loss: 0.03490487486124039 2023-01-23 15:14:30.608990: step: 1130/1541, loss: 0.011254236102104187 2023-01-23 15:14:31.545712: step: 1132/1541, loss: 0.003869804320856929 2023-01-23 15:14:32.497967: step: 1134/1541, loss: 0.003763670101761818 2023-01-23 15:14:33.367736: step: 1136/1541, loss: 0.031214552000164986 2023-01-23 15:14:34.328663: step: 1138/1541, loss: 0.027081243693828583 2023-01-23 15:14:35.193316: step: 1140/1541, loss: 0.015276312828063965 2023-01-23 15:14:36.093512: step: 1142/1541, loss: 0.36545971035957336 2023-01-23 15:14:36.982956: step: 1144/1541, loss: 0.2020483762025833 2023-01-23 15:14:37.881287: step: 1146/1541, loss: 0.004815470427274704 2023-01-23 15:14:38.812468: step: 1148/1541, loss: 0.15310734510421753 2023-01-23 15:14:39.703383: step: 1150/1541, loss: 0.2745389938354492 2023-01-23 15:14:40.618835: step: 1152/1541, loss: 0.0040659066289663315 2023-01-23 15:14:41.510739: step: 1154/1541, loss: 0.16702570021152496 2023-01-23 15:14:42.374709: step: 1156/1541, loss: 0.051245007663965225 2023-01-23 15:14:43.282065: step: 1158/1541, loss: 0.00043921524775214493 2023-01-23 15:14:44.171796: step: 1160/1541, loss: 0.4621126651763916 2023-01-23 15:14:45.059616: step: 1162/1541, loss: 0.15258333086967468 2023-01-23 15:14:45.951493: step: 1164/1541, loss: 0.002993278205394745 2023-01-23 15:14:46.856512: step: 1166/1541, loss: 0.2999827563762665 2023-01-23 15:14:47.735568: step: 1168/1541, loss: 0.0026355928275734186 2023-01-23 15:14:48.584318: step: 1170/1541, loss: 0.10091957449913025 2023-01-23 15:14:49.442497: step: 1172/1541, loss: 0.02459440566599369 2023-01-23 15:14:50.352022: step: 1174/1541, loss: 0.4654621183872223 2023-01-23 15:14:51.270005: step: 1176/1541, loss: 0.0455225333571434 2023-01-23 15:14:52.193113: step: 1178/1541, loss: 0.00514050479978323 2023-01-23 15:14:53.105769: step: 1180/1541, loss: 0.02521338313817978 2023-01-23 15:14:53.986499: step: 1182/1541, loss: 0.004071450792253017 2023-01-23 15:14:54.866036: step: 1184/1541, loss: 0.17990131676197052 2023-01-23 15:14:55.714901: step: 1186/1541, loss: 0.07849128544330597 2023-01-23 15:14:56.597730: step: 1188/1541, loss: 0.03737016022205353 2023-01-23 15:14:57.493256: step: 1190/1541, loss: 0.021802499890327454 2023-01-23 15:14:58.352265: step: 1192/1541, loss: 0.0014994085067883134 2023-01-23 15:14:59.255967: step: 1194/1541, loss: 0.0010849015088751912 2023-01-23 15:15:00.150272: step: 1196/1541, loss: 0.08476640284061432 2023-01-23 15:15:01.069296: step: 1198/1541, loss: 0.0022732270881533623 2023-01-23 15:15:01.954270: step: 1200/1541, loss: 0.006943284533917904 2023-01-23 15:15:02.821435: step: 1202/1541, loss: 0.0031015595886856318 2023-01-23 15:15:03.781155: step: 1204/1541, loss: 0.024222904816269875 2023-01-23 15:15:04.784760: step: 1206/1541, loss: 0.0004883997607976198 2023-01-23 15:15:05.688919: step: 1208/1541, loss: 0.06426744163036346 2023-01-23 15:15:06.610930: step: 1210/1541, loss: 0.0036224727518856525 2023-01-23 15:15:07.509306: step: 1212/1541, loss: 0.014675646089017391 2023-01-23 15:15:08.401709: step: 1214/1541, loss: 0.07298373430967331 2023-01-23 15:15:09.302415: step: 1216/1541, loss: 0.346790611743927 2023-01-23 15:15:10.218749: step: 1218/1541, loss: 0.09287425130605698 2023-01-23 15:15:11.140642: step: 1220/1541, loss: 0.0057611106894910336 2023-01-23 15:15:12.029616: step: 1222/1541, loss: 0.12420736253261566 2023-01-23 15:15:12.953961: step: 1224/1541, loss: 0.09776749461889267 2023-01-23 15:15:13.811315: step: 1226/1541, loss: 0.05108369514346123 2023-01-23 15:15:14.707038: step: 1228/1541, loss: 0.001525317900814116 2023-01-23 15:15:15.574323: step: 1230/1541, loss: 0.12686464190483093 2023-01-23 15:15:16.476357: step: 1232/1541, loss: 0.006447373423725367 2023-01-23 15:15:17.378047: step: 1234/1541, loss: 0.04725327342748642 2023-01-23 15:15:18.234136: step: 1236/1541, loss: 0.18180598318576813 2023-01-23 15:15:19.105309: step: 1238/1541, loss: 0.14592698216438293 2023-01-23 15:15:19.969898: step: 1240/1541, loss: 0.004245012067258358 2023-01-23 15:15:20.868423: step: 1242/1541, loss: 0.0034071416594088078 2023-01-23 15:15:21.757078: step: 1244/1541, loss: 0.005342969670891762 2023-01-23 15:15:22.629927: step: 1246/1541, loss: 0.0011501257540658116 2023-01-23 15:15:23.501957: step: 1248/1541, loss: 0.009021570906043053 2023-01-23 15:15:24.436160: step: 1250/1541, loss: 0.007614975795149803 2023-01-23 15:15:25.386026: step: 1252/1541, loss: 0.047562435269355774 2023-01-23 15:15:26.234149: step: 1254/1541, loss: 0.003006662940606475 2023-01-23 15:15:27.119886: step: 1256/1541, loss: 0.0012835052330046892 2023-01-23 15:15:28.058174: step: 1258/1541, loss: 0.11040247231721878 2023-01-23 15:15:29.059889: step: 1260/1541, loss: 0.062456969171762466 2023-01-23 15:15:29.960491: step: 1262/1541, loss: 0.0004485771059989929 2023-01-23 15:15:30.876583: step: 1264/1541, loss: 0.040270548313856125 2023-01-23 15:15:31.749211: step: 1266/1541, loss: 0.0013326044427230954 2023-01-23 15:15:32.655334: step: 1268/1541, loss: 0.023489980027079582 2023-01-23 15:15:33.552160: step: 1270/1541, loss: 0.008640054613351822 2023-01-23 15:15:34.429443: step: 1272/1541, loss: 0.0038697109557688236 2023-01-23 15:15:35.305586: step: 1274/1541, loss: 0.2634158134460449 2023-01-23 15:15:36.191274: step: 1276/1541, loss: 0.053068701177835464 2023-01-23 15:15:37.076176: step: 1278/1541, loss: 0.008110001683235168 2023-01-23 15:15:37.947589: step: 1280/1541, loss: 0.0006329318857751787 2023-01-23 15:15:38.820705: step: 1282/1541, loss: 0.0011924139689654112 2023-01-23 15:15:39.766705: step: 1284/1541, loss: 0.011879774741828442 2023-01-23 15:15:40.791335: step: 1286/1541, loss: 0.0945020467042923 2023-01-23 15:15:41.693872: step: 1288/1541, loss: 0.04447997361421585 2023-01-23 15:15:42.617510: step: 1290/1541, loss: 0.043340519070625305 2023-01-23 15:15:43.471145: step: 1292/1541, loss: 0.0006771035841666162 2023-01-23 15:15:44.353895: step: 1294/1541, loss: 0.002023170469328761 2023-01-23 15:15:45.224793: step: 1296/1541, loss: 0.028815407305955887 2023-01-23 15:15:46.095795: step: 1298/1541, loss: 0.41662126779556274 2023-01-23 15:15:46.982066: step: 1300/1541, loss: 0.01028523314744234 2023-01-23 15:15:47.902432: step: 1302/1541, loss: 0.21960556507110596 2023-01-23 15:15:48.811215: step: 1304/1541, loss: 0.008497782982885838 2023-01-23 15:15:49.718968: step: 1306/1541, loss: 0.0033747167326509953 2023-01-23 15:15:50.615788: step: 1308/1541, loss: 0.0015756331849843264 2023-01-23 15:15:51.507761: step: 1310/1541, loss: 0.015234149992465973 2023-01-23 15:15:52.448271: step: 1312/1541, loss: 0.00919760949909687 2023-01-23 15:15:53.345696: step: 1314/1541, loss: 0.003247878747060895 2023-01-23 15:15:54.239782: step: 1316/1541, loss: 0.04345054551959038 2023-01-23 15:15:55.150688: step: 1318/1541, loss: 0.0001337724388577044 2023-01-23 15:15:56.137059: step: 1320/1541, loss: 0.0015354955103248358 2023-01-23 15:15:57.071744: step: 1322/1541, loss: 0.09495804458856583 2023-01-23 15:15:57.926407: step: 1324/1541, loss: 0.0013843155466020107 2023-01-23 15:15:58.778734: step: 1326/1541, loss: 0.005961928516626358 2023-01-23 15:15:59.685112: step: 1328/1541, loss: 0.09309607744216919 2023-01-23 15:16:00.648217: step: 1330/1541, loss: 0.030371874570846558 2023-01-23 15:16:01.531865: step: 1332/1541, loss: 0.052769873291254044 2023-01-23 15:16:02.444034: step: 1334/1541, loss: 0.01182189304381609 2023-01-23 15:16:03.318882: step: 1336/1541, loss: 0.06804271042346954 2023-01-23 15:16:04.263014: step: 1338/1541, loss: 0.013219471089541912 2023-01-23 15:16:05.160802: step: 1340/1541, loss: 0.0035306138452142477 2023-01-23 15:16:06.000251: step: 1342/1541, loss: 0.5064759254455566 2023-01-23 15:16:06.886936: step: 1344/1541, loss: 0.027722958475351334 2023-01-23 15:16:07.782271: step: 1346/1541, loss: 0.543830394744873 2023-01-23 15:16:08.626054: step: 1348/1541, loss: 0.0063573941588401794 2023-01-23 15:16:09.517229: step: 1350/1541, loss: 0.007198314182460308 2023-01-23 15:16:10.413073: step: 1352/1541, loss: 0.02126127853989601 2023-01-23 15:16:11.366041: step: 1354/1541, loss: 0.018083037808537483 2023-01-23 15:16:12.252862: step: 1356/1541, loss: 0.26074373722076416 2023-01-23 15:16:13.165040: step: 1358/1541, loss: 0.001721300184726715 2023-01-23 15:16:14.052211: step: 1360/1541, loss: 0.0375041589140892 2023-01-23 15:16:14.937642: step: 1362/1541, loss: 0.0972839742898941 2023-01-23 15:16:15.802973: step: 1364/1541, loss: 0.04058467224240303 2023-01-23 15:16:16.689071: step: 1366/1541, loss: 0.017501266673207283 2023-01-23 15:16:17.646510: step: 1368/1541, loss: 0.019956961274147034 2023-01-23 15:16:18.555835: step: 1370/1541, loss: 0.008837244473397732 2023-01-23 15:16:19.460804: step: 1372/1541, loss: 0.07503902912139893 2023-01-23 15:16:20.349485: step: 1374/1541, loss: 0.3382948637008667 2023-01-23 15:16:21.299138: step: 1376/1541, loss: 0.00410831393674016 2023-01-23 15:16:22.204964: step: 1378/1541, loss: 0.10941239446401596 2023-01-23 15:16:23.140482: step: 1380/1541, loss: 0.07130137830972672 2023-01-23 15:16:24.089715: step: 1382/1541, loss: 0.0006942585459910333 2023-01-23 15:16:25.010830: step: 1384/1541, loss: 0.04654993861913681 2023-01-23 15:16:25.885308: step: 1386/1541, loss: 0.32930096983909607 2023-01-23 15:16:26.770298: step: 1388/1541, loss: 0.0021714912727475166 2023-01-23 15:16:27.654461: step: 1390/1541, loss: 0.003079739399254322 2023-01-23 15:16:28.512794: step: 1392/1541, loss: 0.06231606751680374 2023-01-23 15:16:29.429075: step: 1394/1541, loss: 0.008521363139152527 2023-01-23 15:16:30.333587: step: 1396/1541, loss: 0.003051474690437317 2023-01-23 15:16:31.293632: step: 1398/1541, loss: 0.030479593202471733 2023-01-23 15:16:32.182949: step: 1400/1541, loss: 0.3428973853588104 2023-01-23 15:16:33.102651: step: 1402/1541, loss: 0.13224661350250244 2023-01-23 15:16:34.013321: step: 1404/1541, loss: 0.0012552514672279358 2023-01-23 15:16:34.887767: step: 1406/1541, loss: 0.005449857097119093 2023-01-23 15:16:35.782158: step: 1408/1541, loss: 0.09611161798238754 2023-01-23 15:16:36.732442: step: 1410/1541, loss: 0.558841347694397 2023-01-23 15:16:37.679960: step: 1412/1541, loss: 0.35008829832077026 2023-01-23 15:16:38.559656: step: 1414/1541, loss: 0.0021580569446086884 2023-01-23 15:16:39.469686: step: 1416/1541, loss: 0.5101344585418701 2023-01-23 15:16:40.370697: step: 1418/1541, loss: 0.005148275755345821 2023-01-23 15:16:41.263206: step: 1420/1541, loss: 0.010682584717869759 2023-01-23 15:16:42.167583: step: 1422/1541, loss: 0.07202150672674179 2023-01-23 15:16:43.069541: step: 1424/1541, loss: 0.03902548551559448 2023-01-23 15:16:44.055847: step: 1426/1541, loss: 0.02807902731001377 2023-01-23 15:16:44.969087: step: 1428/1541, loss: 0.04946314916014671 2023-01-23 15:16:45.864901: step: 1430/1541, loss: 0.18126414716243744 2023-01-23 15:16:46.775309: step: 1432/1541, loss: 0.04948492720723152 2023-01-23 15:16:47.681222: step: 1434/1541, loss: 0.0033095097169280052 2023-01-23 15:16:48.602519: step: 1436/1541, loss: 0.04539315029978752 2023-01-23 15:16:49.507933: step: 1438/1541, loss: 0.2734442353248596 2023-01-23 15:16:50.420915: step: 1440/1541, loss: 0.006963464431464672 2023-01-23 15:16:51.333634: step: 1442/1541, loss: 0.1503136157989502 2023-01-23 15:16:52.242554: step: 1444/1541, loss: 0.10123288631439209 2023-01-23 15:16:53.135753: step: 1446/1541, loss: 0.01808716170489788 2023-01-23 15:16:54.085009: step: 1448/1541, loss: 0.1055794209241867 2023-01-23 15:16:54.974560: step: 1450/1541, loss: 0.2942376434803009 2023-01-23 15:16:55.888997: step: 1452/1541, loss: 0.014517522417008877 2023-01-23 15:16:56.787174: step: 1454/1541, loss: 0.8961776494979858 2023-01-23 15:16:57.703260: step: 1456/1541, loss: 0.005586724728345871 2023-01-23 15:16:58.594913: step: 1458/1541, loss: 0.11051840335130692 2023-01-23 15:16:59.560938: step: 1460/1541, loss: 0.04868344962596893 2023-01-23 15:17:00.474096: step: 1462/1541, loss: 0.03097658045589924 2023-01-23 15:17:01.340455: step: 1464/1541, loss: 0.005772842559963465 2023-01-23 15:17:02.243075: step: 1466/1541, loss: 0.014812060631811619 2023-01-23 15:17:03.149840: step: 1468/1541, loss: 0.012854678556323051 2023-01-23 15:17:04.022498: step: 1470/1541, loss: 0.002586216200143099 2023-01-23 15:17:04.922910: step: 1472/1541, loss: 0.062277551740407944 2023-01-23 15:17:05.800803: step: 1474/1541, loss: 0.05787656456232071 2023-01-23 15:17:06.682024: step: 1476/1541, loss: 0.21871134638786316 2023-01-23 15:17:07.562270: step: 1478/1541, loss: 0.06319666653871536 2023-01-23 15:17:08.422352: step: 1480/1541, loss: 0.05799601227045059 2023-01-23 15:17:09.294603: step: 1482/1541, loss: 0.02414773590862751 2023-01-23 15:17:10.171717: step: 1484/1541, loss: 0.0885370746254921 2023-01-23 15:17:11.110028: step: 1486/1541, loss: 0.04717829450964928 2023-01-23 15:17:11.996870: step: 1488/1541, loss: 0.1773417443037033 2023-01-23 15:17:12.877552: step: 1490/1541, loss: 0.004544224124401808 2023-01-23 15:17:13.815579: step: 1492/1541, loss: 0.02072511613368988 2023-01-23 15:17:14.694981: step: 1494/1541, loss: 0.18269948661327362 2023-01-23 15:17:15.594234: step: 1496/1541, loss: 0.016899075359106064 2023-01-23 15:17:16.494517: step: 1498/1541, loss: 0.002585198963060975 2023-01-23 15:17:17.379269: step: 1500/1541, loss: 0.08144240081310272 2023-01-23 15:17:18.252348: step: 1502/1541, loss: 0.01270102895796299 2023-01-23 15:17:19.175247: step: 1504/1541, loss: 0.023288430646061897 2023-01-23 15:17:20.091777: step: 1506/1541, loss: 0.22261600196361542 2023-01-23 15:17:20.981966: step: 1508/1541, loss: 0.0423971526324749 2023-01-23 15:17:21.999750: step: 1510/1541, loss: 0.03193766996264458 2023-01-23 15:17:22.910665: step: 1512/1541, loss: 0.013859597966074944 2023-01-23 15:17:23.789457: step: 1514/1541, loss: 0.00470448425039649 2023-01-23 15:17:24.656700: step: 1516/1541, loss: 0.005569943226873875 2023-01-23 15:17:25.520875: step: 1518/1541, loss: 0.0697251707315445 2023-01-23 15:17:26.443618: step: 1520/1541, loss: 0.002756180241703987 2023-01-23 15:17:27.381245: step: 1522/1541, loss: 0.03171033039689064 2023-01-23 15:17:28.277963: step: 1524/1541, loss: 0.09064942598342896 2023-01-23 15:17:29.164559: step: 1526/1541, loss: 0.0016910468693822622 2023-01-23 15:17:30.080236: step: 1528/1541, loss: 0.2669345736503601 2023-01-23 15:17:31.040568: step: 1530/1541, loss: 0.002441414864733815 2023-01-23 15:17:31.947382: step: 1532/1541, loss: 0.13525567948818207 2023-01-23 15:17:32.803361: step: 1534/1541, loss: 0.03789200633764267 2023-01-23 15:17:33.658184: step: 1536/1541, loss: 0.054198119789361954 2023-01-23 15:17:34.540483: step: 1538/1541, loss: 0.0015420139534398913 2023-01-23 15:17:35.456525: step: 1540/1541, loss: 0.013633822090923786 2023-01-23 15:17:36.413110: step: 1542/1541, loss: 0.0002005107671720907 2023-01-23 15:17:37.351239: step: 1544/1541, loss: 0.08228320628404617 2023-01-23 15:17:38.272305: step: 1546/1541, loss: 0.010401624254882336 2023-01-23 15:17:39.252711: step: 1548/1541, loss: 0.004251768812537193 2023-01-23 15:17:40.122629: step: 1550/1541, loss: 0.0026043641846626997 2023-01-23 15:17:41.006096: step: 1552/1541, loss: 0.2376638650894165 2023-01-23 15:17:41.905766: step: 1554/1541, loss: 0.0017496873624622822 2023-01-23 15:17:42.822026: step: 1556/1541, loss: 0.003544664941728115 2023-01-23 15:17:43.783208: step: 1558/1541, loss: 0.07293512672185898 2023-01-23 15:17:44.701471: step: 1560/1541, loss: 0.003290485357865691 2023-01-23 15:17:45.554314: step: 1562/1541, loss: 0.0009190199780277908 2023-01-23 15:17:46.429404: step: 1564/1541, loss: 0.0026954684872180223 2023-01-23 15:17:47.296498: step: 1566/1541, loss: 0.007475287653505802 2023-01-23 15:17:48.191597: step: 1568/1541, loss: 0.07455769926309586 2023-01-23 15:17:49.120110: step: 1570/1541, loss: 0.02096746116876602 2023-01-23 15:17:50.027462: step: 1572/1541, loss: 0.0010224085999652743 2023-01-23 15:17:50.914167: step: 1574/1541, loss: 0.09731557965278625 2023-01-23 15:17:51.783680: step: 1576/1541, loss: 0.0028727268800139427 2023-01-23 15:17:52.668076: step: 1578/1541, loss: 0.0011512459022924304 2023-01-23 15:17:53.663702: step: 1580/1541, loss: 0.33770105242729187 2023-01-23 15:17:54.556406: step: 1582/1541, loss: 0.10390827804803848 2023-01-23 15:17:55.448338: step: 1584/1541, loss: 0.0017250027740374207 2023-01-23 15:17:56.354651: step: 1586/1541, loss: 1.3700894117355347 2023-01-23 15:17:57.239326: step: 1588/1541, loss: 0.0029245782643556595 2023-01-23 15:17:58.113068: step: 1590/1541, loss: 0.027143225073814392 2023-01-23 15:17:59.032566: step: 1592/1541, loss: 0.3420751094818115 2023-01-23 15:17:59.940081: step: 1594/1541, loss: 0.010402848944067955 2023-01-23 15:18:00.850428: step: 1596/1541, loss: 0.07339980453252792 2023-01-23 15:18:01.715175: step: 1598/1541, loss: 0.004180591087788343 2023-01-23 15:18:02.664885: step: 1600/1541, loss: 0.033336665481328964 2023-01-23 15:18:03.588572: step: 1602/1541, loss: 0.10363778471946716 2023-01-23 15:18:04.448526: step: 1604/1541, loss: 0.005122465081512928 2023-01-23 15:18:05.359898: step: 1606/1541, loss: 0.015862055122852325 2023-01-23 15:18:06.286344: step: 1608/1541, loss: 0.01133865024894476 2023-01-23 15:18:07.151253: step: 1610/1541, loss: 0.015655048191547394 2023-01-23 15:18:08.036518: step: 1612/1541, loss: 0.16898344457149506 2023-01-23 15:18:08.993694: step: 1614/1541, loss: 0.011811611242592335 2023-01-23 15:18:09.878852: step: 1616/1541, loss: 0.0006953885313123465 2023-01-23 15:18:10.760354: step: 1618/1541, loss: 0.04821557179093361 2023-01-23 15:18:11.717200: step: 1620/1541, loss: 0.15936937928199768 2023-01-23 15:18:12.627254: step: 1622/1541, loss: 1.1979619264602661 2023-01-23 15:18:13.541130: step: 1624/1541, loss: 0.11580169200897217 2023-01-23 15:18:14.395576: step: 1626/1541, loss: 0.10117900371551514 2023-01-23 15:18:15.308254: step: 1628/1541, loss: 0.0020021735690534115 2023-01-23 15:18:16.175840: step: 1630/1541, loss: 0.003883913392201066 2023-01-23 15:18:17.057621: step: 1632/1541, loss: 0.04181507229804993 2023-01-23 15:18:17.997352: step: 1634/1541, loss: 0.02412114478647709 2023-01-23 15:18:18.884508: step: 1636/1541, loss: 0.025868365541100502 2023-01-23 15:18:19.748166: step: 1638/1541, loss: 0.007423856295645237 2023-01-23 15:18:20.613940: step: 1640/1541, loss: 0.008197139017283916 2023-01-23 15:18:21.481934: step: 1642/1541, loss: 0.045957934111356735 2023-01-23 15:18:22.331114: step: 1644/1541, loss: 0.003633457701653242 2023-01-23 15:18:23.293532: step: 1646/1541, loss: 0.0019786879420280457 2023-01-23 15:18:24.177483: step: 1648/1541, loss: 0.12166152894496918 2023-01-23 15:18:25.076431: step: 1650/1541, loss: 0.0349968746304512 2023-01-23 15:18:25.979548: step: 1652/1541, loss: 0.09287065267562866 2023-01-23 15:18:26.933208: step: 1654/1541, loss: 6.247135024750605e-05 2023-01-23 15:18:27.882262: step: 1656/1541, loss: 0.032939888536930084 2023-01-23 15:18:28.821970: step: 1658/1541, loss: 0.0027330077718943357 2023-01-23 15:18:29.754307: step: 1660/1541, loss: 0.005173880606889725 2023-01-23 15:18:30.623061: step: 1662/1541, loss: 0.014717599377036095 2023-01-23 15:18:31.557600: step: 1664/1541, loss: 0.0035902829840779305 2023-01-23 15:18:32.512471: step: 1666/1541, loss: 0.0006537449080497026 2023-01-23 15:18:33.409464: step: 1668/1541, loss: 0.04882137477397919 2023-01-23 15:18:34.354926: step: 1670/1541, loss: 0.025553470477461815 2023-01-23 15:18:35.261422: step: 1672/1541, loss: 0.04039148613810539 2023-01-23 15:18:36.157920: step: 1674/1541, loss: 0.06479017436504364 2023-01-23 15:18:37.126866: step: 1676/1541, loss: 0.18140797317028046 2023-01-23 15:18:37.978924: step: 1678/1541, loss: 0.0003986737283412367 2023-01-23 15:18:38.844503: step: 1680/1541, loss: 0.007388546597212553 2023-01-23 15:18:39.753252: step: 1682/1541, loss: 0.0017836957704275846 2023-01-23 15:18:40.728109: step: 1684/1541, loss: 0.0045522237196564674 2023-01-23 15:18:41.608341: step: 1686/1541, loss: 0.03664025291800499 2023-01-23 15:18:42.516189: step: 1688/1541, loss: 0.33881163597106934 2023-01-23 15:18:43.414030: step: 1690/1541, loss: 0.0010638851672410965 2023-01-23 15:18:44.288155: step: 1692/1541, loss: 0.11964231729507446 2023-01-23 15:18:45.194647: step: 1694/1541, loss: 0.2969772219657898 2023-01-23 15:18:46.076453: step: 1696/1541, loss: 0.0008975397213362157 2023-01-23 15:18:47.016484: step: 1698/1541, loss: 0.02936473861336708 2023-01-23 15:18:47.969275: step: 1700/1541, loss: 0.02913621813058853 2023-01-23 15:18:48.854707: step: 1702/1541, loss: 0.010574094019830227 2023-01-23 15:18:49.762226: step: 1704/1541, loss: 0.0010565613629296422 2023-01-23 15:18:50.666600: step: 1706/1541, loss: 0.03269660472869873 2023-01-23 15:18:51.586550: step: 1708/1541, loss: 0.25591588020324707 2023-01-23 15:18:52.552458: step: 1710/1541, loss: 0.0855715200304985 2023-01-23 15:18:53.456783: step: 1712/1541, loss: 0.020310256630182266 2023-01-23 15:18:54.381209: step: 1714/1541, loss: 0.0018200858030468225 2023-01-23 15:18:55.250729: step: 1716/1541, loss: 0.02020716667175293 2023-01-23 15:18:56.165787: step: 1718/1541, loss: 0.03466347977519035 2023-01-23 15:18:57.046404: step: 1720/1541, loss: 0.0008226932841353118 2023-01-23 15:18:57.958288: step: 1722/1541, loss: 0.11090100556612015 2023-01-23 15:18:58.953985: step: 1724/1541, loss: 0.006893196143209934 2023-01-23 15:18:59.818054: step: 1726/1541, loss: 0.18583829700946808 2023-01-23 15:19:00.759625: step: 1728/1541, loss: 0.028592582792043686 2023-01-23 15:19:01.698275: step: 1730/1541, loss: 0.020487740635871887 2023-01-23 15:19:02.598438: step: 1732/1541, loss: 0.001386292977258563 2023-01-23 15:19:03.506930: step: 1734/1541, loss: 0.12889181077480316 2023-01-23 15:19:04.423424: step: 1736/1541, loss: 0.0593712255358696 2023-01-23 15:19:05.342415: step: 1738/1541, loss: 0.2552907466888428 2023-01-23 15:19:06.218753: step: 1740/1541, loss: 0.01280569564551115 2023-01-23 15:19:07.062701: step: 1742/1541, loss: 0.0222945399582386 2023-01-23 15:19:07.966742: step: 1744/1541, loss: 0.017787175253033638 2023-01-23 15:19:08.860779: step: 1746/1541, loss: 0.25149574875831604 2023-01-23 15:19:09.745169: step: 1748/1541, loss: 0.4179149866104126 2023-01-23 15:19:10.621356: step: 1750/1541, loss: 0.008903717622160912 2023-01-23 15:19:11.504463: step: 1752/1541, loss: 0.026700135320425034 2023-01-23 15:19:12.382201: step: 1754/1541, loss: 0.026067083701491356 2023-01-23 15:19:13.297161: step: 1756/1541, loss: 0.03024778887629509 2023-01-23 15:19:14.180703: step: 1758/1541, loss: 0.011980884708464146 2023-01-23 15:19:15.091538: step: 1760/1541, loss: 0.19115190207958221 2023-01-23 15:19:15.976626: step: 1762/1541, loss: 0.1299746334552765 2023-01-23 15:19:16.849493: step: 1764/1541, loss: 0.0457489974796772 2023-01-23 15:19:17.717347: step: 1766/1541, loss: 0.03669755160808563 2023-01-23 15:19:18.578613: step: 1768/1541, loss: 0.30351245403289795 2023-01-23 15:19:19.463809: step: 1770/1541, loss: 0.020677141845226288 2023-01-23 15:19:20.388972: step: 1772/1541, loss: 0.0021349990274757147 2023-01-23 15:19:21.299287: step: 1774/1541, loss: 0.007026687264442444 2023-01-23 15:19:22.195140: step: 1776/1541, loss: 0.0028510456904768944 2023-01-23 15:19:23.033489: step: 1778/1541, loss: 0.03130119666457176 2023-01-23 15:19:23.937531: step: 1780/1541, loss: 0.028286725282669067 2023-01-23 15:19:24.793381: step: 1782/1541, loss: 0.0040702917613089085 2023-01-23 15:19:25.679243: step: 1784/1541, loss: 0.09389177709817886 2023-01-23 15:19:26.550798: step: 1786/1541, loss: 0.048515718430280685 2023-01-23 15:19:27.423535: step: 1788/1541, loss: 0.09666289389133453 2023-01-23 15:19:28.398165: step: 1790/1541, loss: 0.05157201364636421 2023-01-23 15:19:29.295801: step: 1792/1541, loss: 0.0619756244122982 2023-01-23 15:19:30.238624: step: 1794/1541, loss: 0.03216689079999924 2023-01-23 15:19:31.116637: step: 1796/1541, loss: 0.02796081081032753 2023-01-23 15:19:32.028323: step: 1798/1541, loss: 0.2403523474931717 2023-01-23 15:19:32.901391: step: 1800/1541, loss: 0.02477388083934784 2023-01-23 15:19:33.823415: step: 1802/1541, loss: 0.2069721668958664 2023-01-23 15:19:34.764681: step: 1804/1541, loss: 0.0006511249230243266 2023-01-23 15:19:35.651250: step: 1806/1541, loss: 0.022170552983880043 2023-01-23 15:19:36.524506: step: 1808/1541, loss: 0.019119057804346085 2023-01-23 15:19:37.445833: step: 1810/1541, loss: 0.05119628831744194 2023-01-23 15:19:38.357256: step: 1812/1541, loss: 0.025252869352698326 2023-01-23 15:19:39.269619: step: 1814/1541, loss: 0.021868133917450905 2023-01-23 15:19:40.146081: step: 1816/1541, loss: 0.08577164262533188 2023-01-23 15:19:41.042514: step: 1818/1541, loss: 0.015293735079467297 2023-01-23 15:19:41.951400: step: 1820/1541, loss: 0.008543867617845535 2023-01-23 15:19:42.822639: step: 1822/1541, loss: 0.00881225150078535 2023-01-23 15:19:43.736868: step: 1824/1541, loss: 0.0016458010068163276 2023-01-23 15:19:44.591496: step: 1826/1541, loss: 0.001657308079302311 2023-01-23 15:19:45.482335: step: 1828/1541, loss: 0.4611895680427551 2023-01-23 15:19:46.358828: step: 1830/1541, loss: 0.003906983882188797 2023-01-23 15:19:47.267869: step: 1832/1541, loss: 0.22055703401565552 2023-01-23 15:19:48.149216: step: 1834/1541, loss: 0.1713338941335678 2023-01-23 15:19:49.046377: step: 1836/1541, loss: 0.423229843378067 2023-01-23 15:19:49.931606: step: 1838/1541, loss: 0.13449321687221527 2023-01-23 15:19:50.843476: step: 1840/1541, loss: 0.003569052554666996 2023-01-23 15:19:51.752075: step: 1842/1541, loss: 0.022797297686338425 2023-01-23 15:19:52.686722: step: 1844/1541, loss: 0.003166893497109413 2023-01-23 15:19:53.593363: step: 1846/1541, loss: 0.0740242674946785 2023-01-23 15:19:54.511187: step: 1848/1541, loss: 0.008247988298535347 2023-01-23 15:19:55.394103: step: 1850/1541, loss: 0.01981170102953911 2023-01-23 15:19:56.312290: step: 1852/1541, loss: 0.005130631383508444 2023-01-23 15:19:57.228938: step: 1854/1541, loss: 0.1296698898077011 2023-01-23 15:19:58.161048: step: 1856/1541, loss: 0.13559319078922272 2023-01-23 15:19:59.077781: step: 1858/1541, loss: 0.0339214988052845 2023-01-23 15:20:00.030784: step: 1860/1541, loss: 0.012048546224832535 2023-01-23 15:20:00.911661: step: 1862/1541, loss: 0.026477210223674774 2023-01-23 15:20:01.922268: step: 1864/1541, loss: 0.3774886131286621 2023-01-23 15:20:02.843847: step: 1866/1541, loss: 0.35376644134521484 2023-01-23 15:20:03.792308: step: 1868/1541, loss: 0.0037015273701399565 2023-01-23 15:20:04.734981: step: 1870/1541, loss: 0.29668858647346497 2023-01-23 15:20:05.690148: step: 1872/1541, loss: 0.06656112521886826 2023-01-23 15:20:06.598963: step: 1874/1541, loss: 0.028384465724229813 2023-01-23 15:20:07.460638: step: 1876/1541, loss: 0.1861036717891693 2023-01-23 15:20:08.328442: step: 1878/1541, loss: 0.059217195957899094 2023-01-23 15:20:09.240919: step: 1880/1541, loss: 0.007161898072808981 2023-01-23 15:20:10.148799: step: 1882/1541, loss: 0.14646998047828674 2023-01-23 15:20:11.027112: step: 1884/1541, loss: 0.0017763092182576656 2023-01-23 15:20:11.940633: step: 1886/1541, loss: 0.017186030745506287 2023-01-23 15:20:12.782286: step: 1888/1541, loss: 0.0022483039647340775 2023-01-23 15:20:13.662966: step: 1890/1541, loss: 0.16695339977741241 2023-01-23 15:20:14.560183: step: 1892/1541, loss: 0.008894933387637138 2023-01-23 15:20:15.489534: step: 1894/1541, loss: 0.029338520020246506 2023-01-23 15:20:16.450577: step: 1896/1541, loss: 0.13410165905952454 2023-01-23 15:20:17.420420: step: 1898/1541, loss: 0.6670808792114258 2023-01-23 15:20:18.313015: step: 1900/1541, loss: 0.13588222861289978 2023-01-23 15:20:19.265116: step: 1902/1541, loss: 0.09316907078027725 2023-01-23 15:20:20.120693: step: 1904/1541, loss: 0.0005928139435127378 2023-01-23 15:20:21.020614: step: 1906/1541, loss: 0.008606509305536747 2023-01-23 15:20:21.908947: step: 1908/1541, loss: 0.2987154722213745 2023-01-23 15:20:22.782877: step: 1910/1541, loss: 0.20446211099624634 2023-01-23 15:20:23.651485: step: 1912/1541, loss: 0.005343284923583269 2023-01-23 15:20:24.533668: step: 1914/1541, loss: 0.04745471477508545 2023-01-23 15:20:25.433188: step: 1916/1541, loss: 0.07342158257961273 2023-01-23 15:20:26.309469: step: 1918/1541, loss: 0.04254574701189995 2023-01-23 15:20:27.199705: step: 1920/1541, loss: 0.49699512124061584 2023-01-23 15:20:28.078873: step: 1922/1541, loss: 0.2828829288482666 2023-01-23 15:20:28.988875: step: 1924/1541, loss: 0.2001311033964157 2023-01-23 15:20:29.877275: step: 1926/1541, loss: 0.014913424849510193 2023-01-23 15:20:30.773504: step: 1928/1541, loss: 0.22563467919826508 2023-01-23 15:20:31.685608: step: 1930/1541, loss: 0.21205240488052368 2023-01-23 15:20:32.649343: step: 1932/1541, loss: 0.0021767958533018827 2023-01-23 15:20:33.540557: step: 1934/1541, loss: 0.10833802074193954 2023-01-23 15:20:34.432344: step: 1936/1541, loss: 0.0586632564663887 2023-01-23 15:20:35.291384: step: 1938/1541, loss: 0.8321871161460876 2023-01-23 15:20:36.131154: step: 1940/1541, loss: 0.039653532207012177 2023-01-23 15:20:37.017421: step: 1942/1541, loss: 0.0016454998403787613 2023-01-23 15:20:37.936377: step: 1944/1541, loss: 0.2410949021577835 2023-01-23 15:20:38.882328: step: 1946/1541, loss: 0.2404835969209671 2023-01-23 15:20:39.797596: step: 1948/1541, loss: 0.0022616859059780836 2023-01-23 15:20:40.688392: step: 1950/1541, loss: 0.020379096269607544 2023-01-23 15:20:41.578769: step: 1952/1541, loss: 0.03170931711792946 2023-01-23 15:20:42.464840: step: 1954/1541, loss: 0.015154870226979256 2023-01-23 15:20:43.399826: step: 1956/1541, loss: 0.006656037177890539 2023-01-23 15:20:44.276201: step: 1958/1541, loss: 0.000999741954728961 2023-01-23 15:20:45.184888: step: 1960/1541, loss: 0.0028926143422722816 2023-01-23 15:20:46.052249: step: 1962/1541, loss: 0.052166737616062164 2023-01-23 15:20:47.089399: step: 1964/1541, loss: 0.07750189304351807 2023-01-23 15:20:47.956084: step: 1966/1541, loss: 0.02131754532456398 2023-01-23 15:20:48.839257: step: 1968/1541, loss: 0.06840918958187103 2023-01-23 15:20:49.767151: step: 1970/1541, loss: 0.031291648745536804 2023-01-23 15:20:50.646330: step: 1972/1541, loss: 0.02654258906841278 2023-01-23 15:20:51.617244: step: 1974/1541, loss: 0.0359310507774353 2023-01-23 15:20:52.507627: step: 1976/1541, loss: 0.0004842418129555881 2023-01-23 15:20:53.398788: step: 1978/1541, loss: 0.3622554540634155 2023-01-23 15:20:54.309783: step: 1980/1541, loss: 0.0026052435860037804 2023-01-23 15:20:55.241902: step: 1982/1541, loss: 0.15246890485286713 2023-01-23 15:20:56.194183: step: 1984/1541, loss: 1.2989144325256348 2023-01-23 15:20:57.077107: step: 1986/1541, loss: 0.05364475026726723 2023-01-23 15:20:58.016735: step: 1988/1541, loss: 0.18727995455265045 2023-01-23 15:20:58.953955: step: 1990/1541, loss: 0.0033180091995745897 2023-01-23 15:20:59.838157: step: 1992/1541, loss: 0.33509501814842224 2023-01-23 15:21:00.721149: step: 1994/1541, loss: 0.14001871645450592 2023-01-23 15:21:01.591267: step: 1996/1541, loss: 0.0012099652085453272 2023-01-23 15:21:02.533635: step: 1998/1541, loss: 0.0041788555681705475 2023-01-23 15:21:03.421852: step: 2000/1541, loss: 0.4531542658805847 2023-01-23 15:21:04.316571: step: 2002/1541, loss: 0.07928811758756638 2023-01-23 15:21:05.217958: step: 2004/1541, loss: 0.002152430359274149 2023-01-23 15:21:06.082785: step: 2006/1541, loss: 0.12271343171596527 2023-01-23 15:21:07.019864: step: 2008/1541, loss: 0.056901950389146805 2023-01-23 15:21:07.923517: step: 2010/1541, loss: 0.04797397926449776 2023-01-23 15:21:08.796385: step: 2012/1541, loss: 0.004996178671717644 2023-01-23 15:21:09.684112: step: 2014/1541, loss: 0.6914061307907104 2023-01-23 15:21:10.584481: step: 2016/1541, loss: 0.0040788003243505955 2023-01-23 15:21:11.456473: step: 2018/1541, loss: 0.04248609393835068 2023-01-23 15:21:12.330389: step: 2020/1541, loss: 0.0024378763046115637 2023-01-23 15:21:13.270983: step: 2022/1541, loss: 0.00031603043316863477 2023-01-23 15:21:14.176068: step: 2024/1541, loss: 0.03311077877879143 2023-01-23 15:21:15.119152: step: 2026/1541, loss: 0.08107215911149979 2023-01-23 15:21:16.021327: step: 2028/1541, loss: 0.055431488901376724 2023-01-23 15:21:16.901062: step: 2030/1541, loss: 0.007339492440223694 2023-01-23 15:21:17.795648: step: 2032/1541, loss: 0.005908743012696505 2023-01-23 15:21:18.677235: step: 2034/1541, loss: 0.00018801163241732866 2023-01-23 15:21:19.547662: step: 2036/1541, loss: 0.29943203926086426 2023-01-23 15:21:20.410749: step: 2038/1541, loss: 0.002953950548544526 2023-01-23 15:21:21.311400: step: 2040/1541, loss: 0.009513258934020996 2023-01-23 15:21:22.210935: step: 2042/1541, loss: 0.0064611840061843395 2023-01-23 15:21:23.065696: step: 2044/1541, loss: 0.0005646850913763046 2023-01-23 15:21:24.017353: step: 2046/1541, loss: 0.06112711504101753 2023-01-23 15:21:24.912371: step: 2048/1541, loss: 0.00026790655101649463 2023-01-23 15:21:25.813836: step: 2050/1541, loss: 1.5431454181671143 2023-01-23 15:21:26.708982: step: 2052/1541, loss: 0.004005273804068565 2023-01-23 15:21:27.655695: step: 2054/1541, loss: 0.058477502316236496 2023-01-23 15:21:28.543889: step: 2056/1541, loss: 0.014210712164640427 2023-01-23 15:21:29.469987: step: 2058/1541, loss: 0.003967012744396925 2023-01-23 15:21:30.365244: step: 2060/1541, loss: 0.14790689945220947 2023-01-23 15:21:31.313660: step: 2062/1541, loss: 0.11911389976739883 2023-01-23 15:21:32.219084: step: 2064/1541, loss: 0.0012160366168245673 2023-01-23 15:21:33.125198: step: 2066/1541, loss: 0.025418151170015335 2023-01-23 15:21:34.061688: step: 2068/1541, loss: 0.005518690217286348 2023-01-23 15:21:34.919763: step: 2070/1541, loss: 0.017735201865434647 2023-01-23 15:21:35.812822: step: 2072/1541, loss: 0.8022645711898804 2023-01-23 15:21:36.688487: step: 2074/1541, loss: 0.006643675267696381 2023-01-23 15:21:37.617535: step: 2076/1541, loss: 0.004883273504674435 2023-01-23 15:21:38.516237: step: 2078/1541, loss: 0.0254207830876112 2023-01-23 15:21:39.435551: step: 2080/1541, loss: 0.271453320980072 2023-01-23 15:21:40.314052: step: 2082/1541, loss: 0.008028722368180752 2023-01-23 15:21:41.256996: step: 2084/1541, loss: 0.015798788517713547 2023-01-23 15:21:42.172030: step: 2086/1541, loss: 0.010062409564852715 2023-01-23 15:21:43.110116: step: 2088/1541, loss: 0.009644618257880211 2023-01-23 15:21:44.028202: step: 2090/1541, loss: 0.4551592767238617 2023-01-23 15:21:44.922232: step: 2092/1541, loss: 0.7221885919570923 2023-01-23 15:21:45.842066: step: 2094/1541, loss: 0.0008126910543069243 2023-01-23 15:21:46.778406: step: 2096/1541, loss: 0.3605172634124756 2023-01-23 15:21:47.667778: step: 2098/1541, loss: 0.008548267185688019 2023-01-23 15:21:48.542929: step: 2100/1541, loss: 0.06770153343677521 2023-01-23 15:21:49.446557: step: 2102/1541, loss: 0.002212283667176962 2023-01-23 15:21:50.348967: step: 2104/1541, loss: 0.057885780930519104 2023-01-23 15:21:51.203134: step: 2106/1541, loss: 0.0683230310678482 2023-01-23 15:21:52.081551: step: 2108/1541, loss: 0.002260197652503848 2023-01-23 15:21:52.978616: step: 2110/1541, loss: 0.010327968746423721 2023-01-23 15:21:53.868265: step: 2112/1541, loss: 0.03759879246354103 2023-01-23 15:21:54.754069: step: 2114/1541, loss: 0.06167321279644966 2023-01-23 15:21:55.630994: step: 2116/1541, loss: 0.08687776327133179 2023-01-23 15:21:56.519992: step: 2118/1541, loss: 0.050001438707113266 2023-01-23 15:21:57.402229: step: 2120/1541, loss: 0.05893183499574661 2023-01-23 15:21:58.371061: step: 2122/1541, loss: 3.33078932762146 2023-01-23 15:21:59.253996: step: 2124/1541, loss: 0.045920420438051224 2023-01-23 15:22:00.111700: step: 2126/1541, loss: 0.07169154286384583 2023-01-23 15:22:01.038073: step: 2128/1541, loss: 0.011134427040815353 2023-01-23 15:22:01.939949: step: 2130/1541, loss: 0.0027574326377362013 2023-01-23 15:22:02.830324: step: 2132/1541, loss: 0.22544264793395996 2023-01-23 15:22:03.703556: step: 2134/1541, loss: 0.27838262915611267 2023-01-23 15:22:04.644861: step: 2136/1541, loss: 0.021989597007632256 2023-01-23 15:22:05.570872: step: 2138/1541, loss: 0.011690480634570122 2023-01-23 15:22:06.464637: step: 2140/1541, loss: 0.06382562965154648 2023-01-23 15:22:07.361182: step: 2142/1541, loss: 0.01582486368715763 2023-01-23 15:22:08.204311: step: 2144/1541, loss: 0.1765502244234085 2023-01-23 15:22:09.079163: step: 2146/1541, loss: 0.004277585539966822 2023-01-23 15:22:10.000174: step: 2148/1541, loss: 0.0023037928622215986 2023-01-23 15:22:10.901393: step: 2150/1541, loss: 0.00736613804474473 2023-01-23 15:22:11.834677: step: 2152/1541, loss: 0.015743842348456383 2023-01-23 15:22:12.708967: step: 2154/1541, loss: 0.036471981555223465 2023-01-23 15:22:13.687587: step: 2156/1541, loss: 0.13067632913589478 2023-01-23 15:22:14.581497: step: 2158/1541, loss: 0.00941124651581049 2023-01-23 15:22:15.521950: step: 2160/1541, loss: 0.19583779573440552 2023-01-23 15:22:16.381857: step: 2162/1541, loss: 0.007886994630098343 2023-01-23 15:22:17.315761: step: 2164/1541, loss: 0.00042790581937879324 2023-01-23 15:22:18.208408: step: 2166/1541, loss: 0.0013935503084212542 2023-01-23 15:22:19.136169: step: 2168/1541, loss: 0.001143052359111607 2023-01-23 15:22:20.015797: step: 2170/1541, loss: 0.045825615525245667 2023-01-23 15:22:20.883337: step: 2172/1541, loss: 0.00463692843914032 2023-01-23 15:22:21.751194: step: 2174/1541, loss: 0.18379254639148712 2023-01-23 15:22:22.697710: step: 2176/1541, loss: 1.517043948173523 2023-01-23 15:22:23.556188: step: 2178/1541, loss: 0.0011790200369432569 2023-01-23 15:22:24.473944: step: 2180/1541, loss: 0.13321292400360107 2023-01-23 15:22:25.375104: step: 2182/1541, loss: 0.08377004414796829 2023-01-23 15:22:26.261190: step: 2184/1541, loss: 0.03783464804291725 2023-01-23 15:22:27.116271: step: 2186/1541, loss: 0.000870368501637131 2023-01-23 15:22:28.027180: step: 2188/1541, loss: 0.0064483690075576305 2023-01-23 15:22:28.866535: step: 2190/1541, loss: 0.002111647976562381 2023-01-23 15:22:29.754544: step: 2192/1541, loss: 0.15321384370326996 2023-01-23 15:22:30.636270: step: 2194/1541, loss: 0.01952890306711197 2023-01-23 15:22:31.546514: step: 2196/1541, loss: 0.0003008541534654796 2023-01-23 15:22:32.472958: step: 2198/1541, loss: 0.22718796133995056 2023-01-23 15:22:33.393051: step: 2200/1541, loss: 0.21115636825561523 2023-01-23 15:22:34.274874: step: 2202/1541, loss: 0.002199005801230669 2023-01-23 15:22:35.156167: step: 2204/1541, loss: 0.046873822808265686 2023-01-23 15:22:36.092267: step: 2206/1541, loss: 0.006776036694645882 2023-01-23 15:22:37.019148: step: 2208/1541, loss: 0.13349586725234985 2023-01-23 15:22:37.923734: step: 2210/1541, loss: 0.12023966014385223 2023-01-23 15:22:38.803632: step: 2212/1541, loss: 0.2003541737794876 2023-01-23 15:22:39.672878: step: 2214/1541, loss: 0.16768798232078552 2023-01-23 15:22:40.541442: step: 2216/1541, loss: 0.12747029960155487 2023-01-23 15:22:41.457780: step: 2218/1541, loss: 0.010272916406393051 2023-01-23 15:22:42.382426: step: 2220/1541, loss: 0.006616753060370684 2023-01-23 15:22:43.297642: step: 2222/1541, loss: 0.22473502159118652 2023-01-23 15:22:44.200181: step: 2224/1541, loss: 0.003933085594326258 2023-01-23 15:22:45.105086: step: 2226/1541, loss: 0.006404202897101641 2023-01-23 15:22:46.011555: step: 2228/1541, loss: 0.23619474470615387 2023-01-23 15:22:46.974480: step: 2230/1541, loss: 0.1510353684425354 2023-01-23 15:22:47.842087: step: 2232/1541, loss: 0.030662912875413895 2023-01-23 15:22:48.749424: step: 2234/1541, loss: 0.0004958269419148564 2023-01-23 15:22:49.639361: step: 2236/1541, loss: 0.08315496891736984 2023-01-23 15:22:50.571247: step: 2238/1541, loss: 0.04747775197029114 2023-01-23 15:22:51.469051: step: 2240/1541, loss: 0.2493763267993927 2023-01-23 15:22:52.354543: step: 2242/1541, loss: 0.21709810197353363 2023-01-23 15:22:53.235308: step: 2244/1541, loss: 0.10946822911500931 2023-01-23 15:22:54.132014: step: 2246/1541, loss: 0.031451787799596786 2023-01-23 15:22:55.049783: step: 2248/1541, loss: 0.03264261782169342 2023-01-23 15:22:55.916151: step: 2250/1541, loss: 0.03001597337424755 2023-01-23 15:22:56.828254: step: 2252/1541, loss: 0.020151378586888313 2023-01-23 15:22:57.769330: step: 2254/1541, loss: 0.008918014355003834 2023-01-23 15:22:58.639230: step: 2256/1541, loss: 0.017217570915818214 2023-01-23 15:22:59.556630: step: 2258/1541, loss: 0.03388475254178047 2023-01-23 15:23:00.449169: step: 2260/1541, loss: 0.0006481375312432647 2023-01-23 15:23:01.357863: step: 2262/1541, loss: 0.5303584933280945 2023-01-23 15:23:02.288252: step: 2264/1541, loss: 0.020497499033808708 2023-01-23 15:23:03.177623: step: 2266/1541, loss: 0.38570088148117065 2023-01-23 15:23:04.081514: step: 2268/1541, loss: 0.237858384847641 2023-01-23 15:23:04.980032: step: 2270/1541, loss: 0.018038755282759666 2023-01-23 15:23:05.893433: step: 2272/1541, loss: 0.03014305792748928 2023-01-23 15:23:06.803890: step: 2274/1541, loss: 0.0034149931743741035 2023-01-23 15:23:07.704724: step: 2276/1541, loss: 0.08355073630809784 2023-01-23 15:23:08.626553: step: 2278/1541, loss: 0.14835011959075928 2023-01-23 15:23:09.522495: step: 2280/1541, loss: 0.08551445603370667 2023-01-23 15:23:10.439300: step: 2282/1541, loss: 0.23335212469100952 2023-01-23 15:23:11.339437: step: 2284/1541, loss: 0.03947236388921738 2023-01-23 15:23:12.236810: step: 2286/1541, loss: 0.0014325401280075312 2023-01-23 15:23:13.117267: step: 2288/1541, loss: 0.016073906794190407 2023-01-23 15:23:14.004786: step: 2290/1541, loss: 0.16215908527374268 2023-01-23 15:23:14.887626: step: 2292/1541, loss: 0.016963575035333633 2023-01-23 15:23:15.788996: step: 2294/1541, loss: 0.0009478272404521704 2023-01-23 15:23:16.693431: step: 2296/1541, loss: 0.07690204679965973 2023-01-23 15:23:17.603722: step: 2298/1541, loss: 0.11592008918523788 2023-01-23 15:23:18.510742: step: 2300/1541, loss: 0.12267972528934479 2023-01-23 15:23:19.400808: step: 2302/1541, loss: 0.0007167253061197698 2023-01-23 15:23:20.293307: step: 2304/1541, loss: 0.0014659473672509193 2023-01-23 15:23:21.196150: step: 2306/1541, loss: 0.015538109466433525 2023-01-23 15:23:22.059440: step: 2308/1541, loss: 0.31893661618232727 2023-01-23 15:23:22.988213: step: 2310/1541, loss: 0.005592022556811571 2023-01-23 15:23:23.872372: step: 2312/1541, loss: 0.17792078852653503 2023-01-23 15:23:24.790650: step: 2314/1541, loss: 0.022867921739816666 2023-01-23 15:23:25.719127: step: 2316/1541, loss: 0.021246491000056267 2023-01-23 15:23:26.653746: step: 2318/1541, loss: 0.039510514587163925 2023-01-23 15:23:27.595722: step: 2320/1541, loss: 0.22583408653736115 2023-01-23 15:23:28.477489: step: 2322/1541, loss: 0.01586693897843361 2023-01-23 15:23:29.395035: step: 2324/1541, loss: 0.000545744551345706 2023-01-23 15:23:30.255105: step: 2326/1541, loss: 0.0836431160569191 2023-01-23 15:23:31.141912: step: 2328/1541, loss: 0.09184379130601883 2023-01-23 15:23:32.066359: step: 2330/1541, loss: 0.09537263214588165 2023-01-23 15:23:33.014251: step: 2332/1541, loss: 0.009895832277834415 2023-01-23 15:23:34.010854: step: 2334/1541, loss: 0.00020822872465942055 2023-01-23 15:23:34.952326: step: 2336/1541, loss: 0.1868523359298706 2023-01-23 15:23:35.857072: step: 2338/1541, loss: 0.017835035920143127 2023-01-23 15:23:36.738646: step: 2340/1541, loss: 0.10845067352056503 2023-01-23 15:23:37.600411: step: 2342/1541, loss: 0.1363944560289383 2023-01-23 15:23:38.531518: step: 2344/1541, loss: 0.11453855782747269 2023-01-23 15:23:39.427320: step: 2346/1541, loss: 0.020115524530410767 2023-01-23 15:23:40.283528: step: 2348/1541, loss: 0.6612737774848938 2023-01-23 15:23:41.197990: step: 2350/1541, loss: 0.0037795724347233772 2023-01-23 15:23:42.210647: step: 2352/1541, loss: 0.00785169005393982 2023-01-23 15:23:43.109967: step: 2354/1541, loss: 0.014781405217945576 2023-01-23 15:23:44.024898: step: 2356/1541, loss: 0.0013555855257436633 2023-01-23 15:23:44.921331: step: 2358/1541, loss: 0.011708544567227364 2023-01-23 15:23:45.799853: step: 2360/1541, loss: 0.3977926969528198 2023-01-23 15:23:46.693438: step: 2362/1541, loss: 0.0003834724775515497 2023-01-23 15:23:47.569137: step: 2364/1541, loss: 0.13735412061214447 2023-01-23 15:23:48.438096: step: 2366/1541, loss: 0.0011339627671986818 2023-01-23 15:23:49.326932: step: 2368/1541, loss: 0.8087268471717834 2023-01-23 15:23:50.206345: step: 2370/1541, loss: 0.041209228336811066 2023-01-23 15:23:51.100873: step: 2372/1541, loss: 0.9670811295509338 2023-01-23 15:23:51.966565: step: 2374/1541, loss: 0.02269836515188217 2023-01-23 15:23:52.866211: step: 2376/1541, loss: 0.03883799910545349 2023-01-23 15:23:53.766139: step: 2378/1541, loss: 0.09403014183044434 2023-01-23 15:23:54.623826: step: 2380/1541, loss: 0.01749303750693798 2023-01-23 15:23:55.522409: step: 2382/1541, loss: 0.26465466618537903 2023-01-23 15:23:56.436456: step: 2384/1541, loss: 0.015357226133346558 2023-01-23 15:23:57.300159: step: 2386/1541, loss: 0.010571014136075974 2023-01-23 15:23:58.216303: step: 2388/1541, loss: 0.027390185743570328 2023-01-23 15:23:59.070750: step: 2390/1541, loss: 0.4298117756843567 2023-01-23 15:23:59.951976: step: 2392/1541, loss: 0.05468321964144707 2023-01-23 15:24:00.869376: step: 2394/1541, loss: 0.007048923056572676 2023-01-23 15:24:01.875725: step: 2396/1541, loss: 0.0022356868721544743 2023-01-23 15:24:02.786849: step: 2398/1541, loss: 0.015414711087942123 2023-01-23 15:24:03.688744: step: 2400/1541, loss: 0.19303931295871735 2023-01-23 15:24:04.583104: step: 2402/1541, loss: 0.01072979811578989 2023-01-23 15:24:05.467250: step: 2404/1541, loss: 0.00955916102975607 2023-01-23 15:24:06.333379: step: 2406/1541, loss: 0.03162779286503792 2023-01-23 15:24:07.228608: step: 2408/1541, loss: 0.0024843995925039053 2023-01-23 15:24:08.140030: step: 2410/1541, loss: 0.075116366147995 2023-01-23 15:24:09.034797: step: 2412/1541, loss: 0.040192991495132446 2023-01-23 15:24:09.958701: step: 2414/1541, loss: 0.04171750694513321 2023-01-23 15:24:10.868093: step: 2416/1541, loss: 0.013553980737924576 2023-01-23 15:24:11.813092: step: 2418/1541, loss: 0.0017926004948094487 2023-01-23 15:24:12.690533: step: 2420/1541, loss: 0.03420265018939972 2023-01-23 15:24:13.590246: step: 2422/1541, loss: 0.18927431106567383 2023-01-23 15:24:14.473500: step: 2424/1541, loss: 0.010672496631741524 2023-01-23 15:24:15.388394: step: 2426/1541, loss: 0.16081266105175018 2023-01-23 15:24:16.279452: step: 2428/1541, loss: 0.15853069722652435 2023-01-23 15:24:17.182818: step: 2430/1541, loss: 0.005131101235747337 2023-01-23 15:24:18.167327: step: 2432/1541, loss: 0.0005198430153541267 2023-01-23 15:24:19.076988: step: 2434/1541, loss: 0.1352534294128418 2023-01-23 15:24:19.965733: step: 2436/1541, loss: 0.006714282091706991 2023-01-23 15:24:20.849463: step: 2438/1541, loss: 0.007858352735638618 2023-01-23 15:24:21.716714: step: 2440/1541, loss: 0.0057489811442792416 2023-01-23 15:24:22.665400: step: 2442/1541, loss: 0.10167823731899261 2023-01-23 15:24:23.534605: step: 2444/1541, loss: 0.016209233552217484 2023-01-23 15:24:24.466654: step: 2446/1541, loss: 0.0004320454318076372 2023-01-23 15:24:25.361806: step: 2448/1541, loss: 0.12833847105503082 2023-01-23 15:24:26.265829: step: 2450/1541, loss: 0.07096296548843384 2023-01-23 15:24:27.181903: step: 2452/1541, loss: 0.003171631833538413 2023-01-23 15:24:28.070984: step: 2454/1541, loss: 0.0010841174516826868 2023-01-23 15:24:28.931159: step: 2456/1541, loss: 0.01919418014585972 2023-01-23 15:24:29.879995: step: 2458/1541, loss: 8.706876542419195e-05 2023-01-23 15:24:30.799613: step: 2460/1541, loss: 0.0013866252265870571 2023-01-23 15:24:31.707593: step: 2462/1541, loss: 0.30012276768684387 2023-01-23 15:24:32.672701: step: 2464/1541, loss: 0.13197638094425201 2023-01-23 15:24:33.564895: step: 2466/1541, loss: 0.0011913608759641647 2023-01-23 15:24:34.430676: step: 2468/1541, loss: 0.0016401614993810654 2023-01-23 15:24:35.356922: step: 2470/1541, loss: 0.02070523425936699 2023-01-23 15:24:36.319016: step: 2472/1541, loss: 0.0019512390717864037 2023-01-23 15:24:37.202253: step: 2474/1541, loss: 0.06778975576162338 2023-01-23 15:24:38.119599: step: 2476/1541, loss: 0.07454857975244522 2023-01-23 15:24:38.973710: step: 2478/1541, loss: 0.002272567246109247 2023-01-23 15:24:39.895949: step: 2480/1541, loss: 0.40647926926612854 2023-01-23 15:24:40.784562: step: 2482/1541, loss: 1.4535688161849976 2023-01-23 15:24:41.714360: step: 2484/1541, loss: 0.005390055011957884 2023-01-23 15:24:42.570371: step: 2486/1541, loss: 0.020153842866420746 2023-01-23 15:24:43.442099: step: 2488/1541, loss: 0.005713427439332008 2023-01-23 15:24:44.294697: step: 2490/1541, loss: 0.0009971188846975565 2023-01-23 15:24:45.181494: step: 2492/1541, loss: 0.0059486273676157 2023-01-23 15:24:46.208052: step: 2494/1541, loss: 0.0132808368653059 2023-01-23 15:24:47.116060: step: 2496/1541, loss: 0.220497265458107 2023-01-23 15:24:48.042960: step: 2498/1541, loss: 0.005019302945584059 2023-01-23 15:24:48.939154: step: 2500/1541, loss: 0.002809225581586361 2023-01-23 15:24:49.828916: step: 2502/1541, loss: 0.03573128953576088 2023-01-23 15:24:50.728351: step: 2504/1541, loss: 1.2061879634857178 2023-01-23 15:24:51.645940: step: 2506/1541, loss: 0.5435317754745483 2023-01-23 15:24:52.513267: step: 2508/1541, loss: 0.04087339714169502 2023-01-23 15:24:53.415503: step: 2510/1541, loss: 0.43061384558677673 2023-01-23 15:24:54.326990: step: 2512/1541, loss: 0.14209777116775513 2023-01-23 15:24:55.230467: step: 2514/1541, loss: 0.00047194870421662927 2023-01-23 15:24:56.157781: step: 2516/1541, loss: 0.06460634618997574 2023-01-23 15:24:57.073663: step: 2518/1541, loss: 0.02157628908753395 2023-01-23 15:24:57.981151: step: 2520/1541, loss: 0.43793877959251404 2023-01-23 15:24:58.874303: step: 2522/1541, loss: 0.06502510607242584 2023-01-23 15:24:59.763319: step: 2524/1541, loss: 0.07345429062843323 2023-01-23 15:25:00.663203: step: 2526/1541, loss: 0.051301151514053345 2023-01-23 15:25:01.610945: step: 2528/1541, loss: 0.005762577056884766 2023-01-23 15:25:02.492754: step: 2530/1541, loss: 0.0160240326076746 2023-01-23 15:25:03.394822: step: 2532/1541, loss: 0.045780256390571594 2023-01-23 15:25:04.267881: step: 2534/1541, loss: 0.03319982439279556 2023-01-23 15:25:05.186146: step: 2536/1541, loss: 0.03727710247039795 2023-01-23 15:25:06.073568: step: 2538/1541, loss: 0.013118362985551357 2023-01-23 15:25:06.965981: step: 2540/1541, loss: 0.0013530057622119784 2023-01-23 15:25:07.851064: step: 2542/1541, loss: 0.012254494242370129 2023-01-23 15:25:08.793981: step: 2544/1541, loss: 0.0019087530672550201 2023-01-23 15:25:09.688396: step: 2546/1541, loss: 0.7867255806922913 2023-01-23 15:25:10.638347: step: 2548/1541, loss: 0.012149590067565441 2023-01-23 15:25:11.538432: step: 2550/1541, loss: 0.62055504322052 2023-01-23 15:25:12.417797: step: 2552/1541, loss: 0.0013842330081388354 2023-01-23 15:25:13.346377: step: 2554/1541, loss: 0.016211913898587227 2023-01-23 15:25:14.263762: step: 2556/1541, loss: 0.0212777741253376 2023-01-23 15:25:15.154669: step: 2558/1541, loss: 0.08357928693294525 2023-01-23 15:25:16.035378: step: 2560/1541, loss: 0.027876345440745354 2023-01-23 15:25:16.960261: step: 2562/1541, loss: 0.04327898845076561 2023-01-23 15:25:17.847115: step: 2564/1541, loss: 0.3254438638687134 2023-01-23 15:25:18.756461: step: 2566/1541, loss: 0.07280358672142029 2023-01-23 15:25:19.673367: step: 2568/1541, loss: 0.05045310780405998 2023-01-23 15:25:20.657764: step: 2570/1541, loss: 0.3151857852935791 2023-01-23 15:25:21.560204: step: 2572/1541, loss: 0.08649353682994843 2023-01-23 15:25:22.481339: step: 2574/1541, loss: 0.006899126339703798 2023-01-23 15:25:23.371539: step: 2576/1541, loss: 0.3310322165489197 2023-01-23 15:25:24.322478: step: 2578/1541, loss: 0.027405092492699623 2023-01-23 15:25:25.199459: step: 2580/1541, loss: 0.07212966680526733 2023-01-23 15:25:26.077874: step: 2582/1541, loss: 0.1414777785539627 2023-01-23 15:25:26.987356: step: 2584/1541, loss: 0.004151119384914637 2023-01-23 15:25:27.891250: step: 2586/1541, loss: 0.03668823093175888 2023-01-23 15:25:28.779995: step: 2588/1541, loss: 0.0027011740021407604 2023-01-23 15:25:29.691743: step: 2590/1541, loss: 0.0013442256022244692 2023-01-23 15:25:30.555566: step: 2592/1541, loss: 0.16956879198551178 2023-01-23 15:25:31.552180: step: 2594/1541, loss: 0.0634746327996254 2023-01-23 15:25:32.491032: step: 2596/1541, loss: 0.01490460243076086 2023-01-23 15:25:33.377499: step: 2598/1541, loss: 0.009182044304907322 2023-01-23 15:25:34.248430: step: 2600/1541, loss: 0.004242438357323408 2023-01-23 15:25:35.149995: step: 2602/1541, loss: 0.029058074578642845 2023-01-23 15:25:36.156000: step: 2604/1541, loss: 0.13829784095287323 2023-01-23 15:25:37.037123: step: 2606/1541, loss: 0.01629411056637764 2023-01-23 15:25:37.925001: step: 2608/1541, loss: 0.003266312414780259 2023-01-23 15:25:38.839294: step: 2610/1541, loss: 0.006248933728784323 2023-01-23 15:25:39.746492: step: 2612/1541, loss: 0.14577354490756989 2023-01-23 15:25:40.599037: step: 2614/1541, loss: 0.01215719897300005 2023-01-23 15:25:41.556954: step: 2616/1541, loss: 0.004482412710785866 2023-01-23 15:25:42.463628: step: 2618/1541, loss: 0.00440121628344059 2023-01-23 15:25:43.407615: step: 2620/1541, loss: 0.14426641166210175 2023-01-23 15:25:44.365669: step: 2622/1541, loss: 0.07067541778087616 2023-01-23 15:25:45.276014: step: 2624/1541, loss: 0.21056213974952698 2023-01-23 15:25:46.167272: step: 2626/1541, loss: 0.7439041137695312 2023-01-23 15:25:47.037168: step: 2628/1541, loss: 0.18887268006801605 2023-01-23 15:25:47.950714: step: 2630/1541, loss: 0.01231754943728447 2023-01-23 15:25:48.828124: step: 2632/1541, loss: 0.19755226373672485 2023-01-23 15:25:49.687541: step: 2634/1541, loss: 0.011802766472101212 2023-01-23 15:25:50.613103: step: 2636/1541, loss: 0.14620999991893768 2023-01-23 15:25:51.469332: step: 2638/1541, loss: 0.04249509423971176 2023-01-23 15:25:52.430295: step: 2640/1541, loss: 0.02736775390803814 2023-01-23 15:25:53.347100: step: 2642/1541, loss: 0.0007895914604887366 2023-01-23 15:25:54.257115: step: 2644/1541, loss: 0.0003307706501800567 2023-01-23 15:25:55.083344: step: 2646/1541, loss: 0.00043517659651115537 2023-01-23 15:25:56.019369: step: 2648/1541, loss: 0.41844189167022705 2023-01-23 15:25:56.942260: step: 2650/1541, loss: 0.02175428345799446 2023-01-23 15:25:57.872791: step: 2652/1541, loss: 0.08839143067598343 2023-01-23 15:25:58.812554: step: 2654/1541, loss: 0.03241240233182907 2023-01-23 15:25:59.738762: step: 2656/1541, loss: 0.13186387717723846 2023-01-23 15:26:00.631299: step: 2658/1541, loss: 0.00815095566213131 2023-01-23 15:26:01.568391: step: 2660/1541, loss: 0.009360947646200657 2023-01-23 15:26:02.460295: step: 2662/1541, loss: 0.1861601620912552 2023-01-23 15:26:03.343969: step: 2664/1541, loss: 0.001178278704173863 2023-01-23 15:26:04.300943: step: 2666/1541, loss: 0.3416818380355835 2023-01-23 15:26:05.242741: step: 2668/1541, loss: 1.2578691244125366 2023-01-23 15:26:06.194824: step: 2670/1541, loss: 0.00032473981264047325 2023-01-23 15:26:07.096070: step: 2672/1541, loss: 0.001142075750976801 2023-01-23 15:26:08.090047: step: 2674/1541, loss: 0.06685430556535721 2023-01-23 15:26:08.944030: step: 2676/1541, loss: 0.283008873462677 2023-01-23 15:26:09.840053: step: 2678/1541, loss: 0.00735543854534626 2023-01-23 15:26:10.716294: step: 2680/1541, loss: 0.0010389222297817469 2023-01-23 15:26:11.587359: step: 2682/1541, loss: 0.004128730855882168 2023-01-23 15:26:12.544798: step: 2684/1541, loss: 0.21235381066799164 2023-01-23 15:26:13.457141: step: 2686/1541, loss: 0.48771312832832336 2023-01-23 15:26:14.316959: step: 2688/1541, loss: 0.0004933750024065375 2023-01-23 15:26:15.229299: step: 2690/1541, loss: 0.03576635196805 2023-01-23 15:26:16.111982: step: 2692/1541, loss: 0.0012869004858657718 2023-01-23 15:26:17.058846: step: 2694/1541, loss: 0.002555224345996976 2023-01-23 15:26:17.997028: step: 2696/1541, loss: 0.007059463299810886 2023-01-23 15:26:18.900859: step: 2698/1541, loss: 0.015531687065958977 2023-01-23 15:26:19.815356: step: 2700/1541, loss: 0.07806376367807388 2023-01-23 15:26:20.706284: step: 2702/1541, loss: 0.0003063712501898408 2023-01-23 15:26:21.635191: step: 2704/1541, loss: 0.4149368703365326 2023-01-23 15:26:22.668668: step: 2706/1541, loss: 0.007845965214073658 2023-01-23 15:26:23.547541: step: 2708/1541, loss: 0.15103933215141296 2023-01-23 15:26:24.470021: step: 2710/1541, loss: 0.04260963201522827 2023-01-23 15:26:25.333496: step: 2712/1541, loss: 0.25343450903892517 2023-01-23 15:26:26.261187: step: 2714/1541, loss: 0.008592176251113415 2023-01-23 15:26:27.084192: step: 2716/1541, loss: 0.04774594306945801 2023-01-23 15:26:27.967257: step: 2718/1541, loss: 0.03924812003970146 2023-01-23 15:26:28.844336: step: 2720/1541, loss: 0.07206819951534271 2023-01-23 15:26:29.733643: step: 2722/1541, loss: 0.14980055391788483 2023-01-23 15:26:30.603918: step: 2724/1541, loss: 0.18750745058059692 2023-01-23 15:26:31.545453: step: 2726/1541, loss: 0.12860386073589325 2023-01-23 15:26:32.451808: step: 2728/1541, loss: 0.008205858990550041 2023-01-23 15:26:33.297922: step: 2730/1541, loss: 0.26682668924331665 2023-01-23 15:26:34.167821: step: 2732/1541, loss: 0.07457947731018066 2023-01-23 15:26:35.081525: step: 2734/1541, loss: 0.003708715783432126 2023-01-23 15:26:36.007343: step: 2736/1541, loss: 0.00014739128528162837 2023-01-23 15:26:36.904899: step: 2738/1541, loss: 0.0009244008688256145 2023-01-23 15:26:37.878552: step: 2740/1541, loss: 0.04466333985328674 2023-01-23 15:26:38.785688: step: 2742/1541, loss: 0.0037793670780956745 2023-01-23 15:26:39.694734: step: 2744/1541, loss: 0.0018525926861912012 2023-01-23 15:26:40.613227: step: 2746/1541, loss: 0.4085570275783539 2023-01-23 15:26:41.514637: step: 2748/1541, loss: 0.12527599930763245 2023-01-23 15:26:42.456863: step: 2750/1541, loss: 0.2545296251773834 2023-01-23 15:26:43.353419: step: 2752/1541, loss: 0.20750792324543 2023-01-23 15:26:44.273864: step: 2754/1541, loss: 0.018345871940255165 2023-01-23 15:26:45.131514: step: 2756/1541, loss: 0.0003356239467393607 2023-01-23 15:26:46.040189: step: 2758/1541, loss: 0.09145672619342804 2023-01-23 15:26:46.950266: step: 2760/1541, loss: 0.003223462263122201 2023-01-23 15:26:47.844894: step: 2762/1541, loss: 0.025407323613762856 2023-01-23 15:26:48.711316: step: 2764/1541, loss: 0.001855459064245224 2023-01-23 15:26:49.625204: step: 2766/1541, loss: 0.011671331711113453 2023-01-23 15:26:50.509324: step: 2768/1541, loss: 0.019149616360664368 2023-01-23 15:26:51.393013: step: 2770/1541, loss: 0.0025214492343366146 2023-01-23 15:26:52.252929: step: 2772/1541, loss: 0.00839108694344759 2023-01-23 15:26:53.116480: step: 2774/1541, loss: 0.0020710215903818607 2023-01-23 15:26:54.009415: step: 2776/1541, loss: 0.12199332565069199 2023-01-23 15:26:54.945686: step: 2778/1541, loss: 0.001219910103827715 2023-01-23 15:26:55.869313: step: 2780/1541, loss: 0.00030940899159759283 2023-01-23 15:26:56.754898: step: 2782/1541, loss: 0.002410732675343752 2023-01-23 15:26:57.624771: step: 2784/1541, loss: 0.14903700351715088 2023-01-23 15:26:58.647889: step: 2786/1541, loss: 0.020889531821012497 2023-01-23 15:26:59.516517: step: 2788/1541, loss: 0.0028546538669615984 2023-01-23 15:27:00.436957: step: 2790/1541, loss: 0.03352862969040871 2023-01-23 15:27:01.369942: step: 2792/1541, loss: 0.0013212694320827723 2023-01-23 15:27:02.263724: step: 2794/1541, loss: 0.6285085082054138 2023-01-23 15:27:03.142796: step: 2796/1541, loss: 0.0061096930876374245 2023-01-23 15:27:04.055149: step: 2798/1541, loss: 0.011660989373922348 2023-01-23 15:27:04.999257: step: 2800/1541, loss: 0.02001539058983326 2023-01-23 15:27:05.907881: step: 2802/1541, loss: 0.11921022087335587 2023-01-23 15:27:06.880832: step: 2804/1541, loss: 0.0002862458350136876 2023-01-23 15:27:07.776578: step: 2806/1541, loss: 0.0017413079040125012 2023-01-23 15:27:08.681745: step: 2808/1541, loss: 0.21187160909175873 2023-01-23 15:27:09.609937: step: 2810/1541, loss: 0.032577283680438995 2023-01-23 15:27:10.499455: step: 2812/1541, loss: 0.10089825093746185 2023-01-23 15:27:11.366756: step: 2814/1541, loss: 0.018680697306990623 2023-01-23 15:27:12.236312: step: 2816/1541, loss: 0.048784539103507996 2023-01-23 15:27:13.098295: step: 2818/1541, loss: 0.1789606809616089 2023-01-23 15:27:13.957687: step: 2820/1541, loss: 0.07020949572324753 2023-01-23 15:27:14.840740: step: 2822/1541, loss: 0.0733446553349495 2023-01-23 15:27:15.745810: step: 2824/1541, loss: 0.065386563539505 2023-01-23 15:27:16.666746: step: 2826/1541, loss: 0.0011089409235864878 2023-01-23 15:27:17.610666: step: 2828/1541, loss: 0.12346965819597244 2023-01-23 15:27:18.512492: step: 2830/1541, loss: 0.06428179889917374 2023-01-23 15:27:19.363845: step: 2832/1541, loss: 0.004572676029056311 2023-01-23 15:27:20.270568: step: 2834/1541, loss: 0.0024542133323848248 2023-01-23 15:27:21.186178: step: 2836/1541, loss: 0.05545065924525261 2023-01-23 15:27:22.036852: step: 2838/1541, loss: 0.05773000419139862 2023-01-23 15:27:22.929711: step: 2840/1541, loss: 0.0024101424496620893 2023-01-23 15:27:23.822729: step: 2842/1541, loss: 0.0037659320514649153 2023-01-23 15:27:24.719252: step: 2844/1541, loss: 0.010381197556853294 2023-01-23 15:27:25.607957: step: 2846/1541, loss: 0.12505434453487396 2023-01-23 15:27:26.489955: step: 2848/1541, loss: 0.7336106300354004 2023-01-23 15:27:27.373203: step: 2850/1541, loss: 0.12916381657123566 2023-01-23 15:27:28.263544: step: 2852/1541, loss: 0.025768857449293137 2023-01-23 15:27:29.129974: step: 2854/1541, loss: 0.0005957502289675176 2023-01-23 15:27:30.037607: step: 2856/1541, loss: 0.0036425848957151175 2023-01-23 15:27:30.934590: step: 2858/1541, loss: 0.000664092309307307 2023-01-23 15:27:31.814321: step: 2860/1541, loss: 0.3422721028327942 2023-01-23 15:27:32.704987: step: 2862/1541, loss: 0.029865026473999023 2023-01-23 15:27:33.587934: step: 2864/1541, loss: 0.0006985223153606057 2023-01-23 15:27:34.485785: step: 2866/1541, loss: 0.007395718712359667 2023-01-23 15:27:35.356106: step: 2868/1541, loss: 0.05155131593346596 2023-01-23 15:27:36.227945: step: 2870/1541, loss: 0.08320734649896622 2023-01-23 15:27:37.106738: step: 2872/1541, loss: 0.008551623672246933 2023-01-23 15:27:37.993941: step: 2874/1541, loss: 0.009809212759137154 2023-01-23 15:27:38.900104: step: 2876/1541, loss: 0.0016105415998026729 2023-01-23 15:27:39.764196: step: 2878/1541, loss: 0.001276234514079988 2023-01-23 15:27:40.672768: step: 2880/1541, loss: 0.0019083363004028797 2023-01-23 15:27:41.546009: step: 2882/1541, loss: 0.05237584188580513 2023-01-23 15:27:42.419210: step: 2884/1541, loss: 0.0008902307017706335 2023-01-23 15:27:43.336494: step: 2886/1541, loss: 0.0014933700440451503 2023-01-23 15:27:44.182381: step: 2888/1541, loss: 0.006447800435125828 2023-01-23 15:27:45.131708: step: 2890/1541, loss: 0.011194933205842972 2023-01-23 15:27:46.092214: step: 2892/1541, loss: 0.00501829432323575 2023-01-23 15:27:47.033126: step: 2894/1541, loss: 0.15743111073970795 2023-01-23 15:27:47.927204: step: 2896/1541, loss: 0.052337564527988434 2023-01-23 15:27:48.806870: step: 2898/1541, loss: 1.6158922910690308 2023-01-23 15:27:49.743107: step: 2900/1541, loss: 0.040413644164800644 2023-01-23 15:27:50.671636: step: 2902/1541, loss: 0.005547805689275265 2023-01-23 15:27:51.650080: step: 2904/1541, loss: 0.13507413864135742 2023-01-23 15:27:52.541521: step: 2906/1541, loss: 0.18896879255771637 2023-01-23 15:27:53.440717: step: 2908/1541, loss: 0.0671725794672966 2023-01-23 15:27:54.397141: step: 2910/1541, loss: 0.04303329437971115 2023-01-23 15:27:55.275492: step: 2912/1541, loss: 0.11050011962652206 2023-01-23 15:27:56.145535: step: 2914/1541, loss: 0.16369658708572388 2023-01-23 15:27:57.047254: step: 2916/1541, loss: 0.005556304007768631 2023-01-23 15:27:57.973945: step: 2918/1541, loss: 0.043931420892477036 2023-01-23 15:27:58.928240: step: 2920/1541, loss: 0.05585720017552376 2023-01-23 15:27:59.893111: step: 2922/1541, loss: 0.003310319036245346 2023-01-23 15:28:00.799953: step: 2924/1541, loss: 0.0965816080570221 2023-01-23 15:28:01.665829: step: 2926/1541, loss: 0.004326257389038801 2023-01-23 15:28:02.569268: step: 2928/1541, loss: 0.007153264246881008 2023-01-23 15:28:03.454213: step: 2930/1541, loss: 0.0008862888789735734 2023-01-23 15:28:04.394593: step: 2932/1541, loss: 0.16293297708034515 2023-01-23 15:28:05.319103: step: 2934/1541, loss: 0.12938712537288666 2023-01-23 15:28:06.222222: step: 2936/1541, loss: 0.0003451881348155439 2023-01-23 15:28:07.155023: step: 2938/1541, loss: 0.003819960169494152 2023-01-23 15:28:08.037335: step: 2940/1541, loss: 0.054857902228832245 2023-01-23 15:28:08.929164: step: 2942/1541, loss: 0.010668602772057056 2023-01-23 15:28:09.825471: step: 2944/1541, loss: 0.6542710661888123 2023-01-23 15:28:10.767923: step: 2946/1541, loss: 0.03176563233137131 2023-01-23 15:28:11.679539: step: 2948/1541, loss: 0.1092563346028328 2023-01-23 15:28:12.589529: step: 2950/1541, loss: 0.0005075616063550115 2023-01-23 15:28:13.481821: step: 2952/1541, loss: 0.09430818259716034 2023-01-23 15:28:14.364707: step: 2954/1541, loss: 0.08080179989337921 2023-01-23 15:28:15.265641: step: 2956/1541, loss: 0.08223321288824081 2023-01-23 15:28:16.107912: step: 2958/1541, loss: 0.2186983972787857 2023-01-23 15:28:16.978326: step: 2960/1541, loss: 0.19804608821868896 2023-01-23 15:28:17.849050: step: 2962/1541, loss: 0.010693262331187725 2023-01-23 15:28:18.739972: step: 2964/1541, loss: 0.00030809949385002255 2023-01-23 15:28:19.611705: step: 2966/1541, loss: 0.0048450869508087635 2023-01-23 15:28:20.506725: step: 2968/1541, loss: 0.04056474193930626 2023-01-23 15:28:21.420390: step: 2970/1541, loss: 0.07856415212154388 2023-01-23 15:28:22.371882: step: 2972/1541, loss: 0.00035423357621766627 2023-01-23 15:28:23.250170: step: 2974/1541, loss: 0.004157893359661102 2023-01-23 15:28:24.170426: step: 2976/1541, loss: 0.05197976529598236 2023-01-23 15:28:25.071500: step: 2978/1541, loss: 0.021844157949090004 2023-01-23 15:28:25.940120: step: 2980/1541, loss: 0.3382064700126648 2023-01-23 15:28:26.828722: step: 2982/1541, loss: 0.002911218674853444 2023-01-23 15:28:27.688125: step: 2984/1541, loss: 0.04916376248002052 2023-01-23 15:28:28.572592: step: 2986/1541, loss: 0.0025577584747225046 2023-01-23 15:28:29.445429: step: 2988/1541, loss: 0.0019480668706819415 2023-01-23 15:28:30.356513: step: 2990/1541, loss: 0.007992720231413841 2023-01-23 15:28:31.258313: step: 2992/1541, loss: 0.032107047736644745 2023-01-23 15:28:32.179425: step: 2994/1541, loss: 0.05603983253240585 2023-01-23 15:28:33.103659: step: 2996/1541, loss: 0.018528707325458527 2023-01-23 15:28:34.011248: step: 2998/1541, loss: 0.0011148120975121856 2023-01-23 15:28:34.947072: step: 3000/1541, loss: 0.000669722503516823 2023-01-23 15:28:35.882464: step: 3002/1541, loss: 0.0025833144318312407 2023-01-23 15:28:36.762559: step: 3004/1541, loss: 0.7947881817817688 2023-01-23 15:28:37.717479: step: 3006/1541, loss: 0.09542691707611084 2023-01-23 15:28:38.614954: step: 3008/1541, loss: 0.0005119431880302727 2023-01-23 15:28:39.482669: step: 3010/1541, loss: 0.016482912003993988 2023-01-23 15:28:40.388383: step: 3012/1541, loss: 0.9279177188873291 2023-01-23 15:28:41.272071: step: 3014/1541, loss: 0.000449774059234187 2023-01-23 15:28:42.181328: step: 3016/1541, loss: 0.0057455855421721935 2023-01-23 15:28:43.033377: step: 3018/1541, loss: 0.019732918590307236 2023-01-23 15:28:44.018942: step: 3020/1541, loss: 0.13665175437927246 2023-01-23 15:28:45.008148: step: 3022/1541, loss: 0.014201851561665535 2023-01-23 15:28:45.955421: step: 3024/1541, loss: 0.0020875988993793726 2023-01-23 15:28:46.808610: step: 3026/1541, loss: 0.29906225204467773 2023-01-23 15:28:47.724990: step: 3028/1541, loss: 0.051334381103515625 2023-01-23 15:28:48.618144: step: 3030/1541, loss: 0.9166500568389893 2023-01-23 15:28:49.515439: step: 3032/1541, loss: 0.002875375095754862 2023-01-23 15:28:50.383134: step: 3034/1541, loss: 0.012993745505809784 2023-01-23 15:28:51.303804: step: 3036/1541, loss: 0.0446820892393589 2023-01-23 15:28:52.181582: step: 3038/1541, loss: 0.3341894745826721 2023-01-23 15:28:53.050845: step: 3040/1541, loss: 0.02052297443151474 2023-01-23 15:28:53.949337: step: 3042/1541, loss: 0.025311468169093132 2023-01-23 15:28:54.869954: step: 3044/1541, loss: 0.04226965457201004 2023-01-23 15:28:55.775781: step: 3046/1541, loss: 0.0013464929070323706 2023-01-23 15:28:56.666580: step: 3048/1541, loss: 0.1609574854373932 2023-01-23 15:28:57.599863: step: 3050/1541, loss: 0.10482891649007797 2023-01-23 15:28:58.501062: step: 3052/1541, loss: 0.00822414644062519 2023-01-23 15:28:59.457231: step: 3054/1541, loss: 0.23214012384414673 2023-01-23 15:29:00.315342: step: 3056/1541, loss: 0.01773558370769024 2023-01-23 15:29:01.182880: step: 3058/1541, loss: 0.05142778158187866 2023-01-23 15:29:02.163196: step: 3060/1541, loss: 0.13710960745811462 2023-01-23 15:29:03.103339: step: 3062/1541, loss: 0.18446998298168182 2023-01-23 15:29:03.997030: step: 3064/1541, loss: 0.2624516487121582 2023-01-23 15:29:04.890116: step: 3066/1541, loss: 0.19345898926258087 2023-01-23 15:29:05.748233: step: 3068/1541, loss: 0.05963024124503136 2023-01-23 15:29:06.659350: step: 3070/1541, loss: 0.0026979779358953238 2023-01-23 15:29:07.581829: step: 3072/1541, loss: 0.021091513335704803 2023-01-23 15:29:08.497784: step: 3074/1541, loss: 0.03336656838655472 2023-01-23 15:29:09.378433: step: 3076/1541, loss: 0.009232934564352036 2023-01-23 15:29:10.264965: step: 3078/1541, loss: 0.5008599162101746 2023-01-23 15:29:11.182145: step: 3080/1541, loss: 0.021145716309547424 2023-01-23 15:29:12.052903: step: 3082/1541, loss: 0.0725954920053482 2023-01-23 15:29:12.676776: step: 3084/1541, loss: 0.002820401219651103 ================================================== Loss: 0.100 -------------------- Dev Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6645425841164653, 'r': 0.642373979681238, 'f1': 0.6532702641492037}, 'combined': 0.4352392565931093, 'epoch': 21} Test Chinese: {'event': {'p': 0.613564668769716, 'r': 0.6992210904733374, 'f1': 0.6535984318118174}, 'argument': {'p': 0.7128404530056945, 'r': 0.6982767662605947, 'f1': 0.7054834560708773}, 'combined': 0.4611028805571066, 'epoch': 21} Dev Korean: {'event': {'p': 0.5676470588235294, 'r': 0.7709720372836218, 'f1': 0.6538678712591757}, 'argument': {'p': 0.6612324683819081, 'r': 0.6474666811722345, 'f1': 0.6542771758236653}, 'combined': 0.4278108241692854, 'epoch': 21} Test Korean: {'event': {'p': 0.6120114394661582, 'r': 0.7693229478729778, 'f1': 0.6817095832227236}, 'argument': {'p': 0.7112166621064147, 'r': 0.7122141609312483, 'f1': 0.7117150620097829}, 'combined': 0.4851829782960239, 'epoch': 21} Dev Russian: {'event': {'p': 0.6587677725118484, 'r': 0.7403462050599201, 'f1': 0.69717868338558}, 'argument': {'p': 0.66278432199299, 'r': 0.6498178867997487, 'f1': 0.6562370604535456}, 'combined': 0.45751448979582615, 'epoch': 21} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.7139491554959331, 'r': 0.701900462387025, 'f1': 0.7078735425485658}, 'combined': 0.49195233906727137, 'epoch': 21} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6268618387262455, 'r': 0.6545990882274068, 'f1': 0.6404302767939131}, 'combined': 0.45979609615973244, 'epoch': 21} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5768849206349207, 'r': 0.41955266955266957, 'f1': 0.4857978279030911}, 'combined': 0.2830633012978641, 'epoch': 21} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.561523351063581, 'r': 0.587640716229329, 'f1': 0.5742852454059351}, 'combined': 0.29534669763733806, 'epoch': 21} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test for Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 22 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 15:34:35.775940: step: 2/1541, loss: 0.015300535596907139 2023-01-23 15:34:36.716324: step: 4/1541, loss: 0.0009604439837858081 2023-01-23 15:34:37.571084: step: 6/1541, loss: 0.0003946206416003406 2023-01-23 15:34:38.489323: step: 8/1541, loss: 0.08638301491737366 2023-01-23 15:34:39.415228: step: 10/1541, loss: 0.09177932143211365 2023-01-23 15:34:40.353014: step: 12/1541, loss: 0.0009216945036314428 2023-01-23 15:34:41.318162: step: 14/1541, loss: 0.0014537021052092314 2023-01-23 15:34:42.222592: step: 16/1541, loss: 0.001902950112707913 2023-01-23 15:34:43.124988: step: 18/1541, loss: 0.008026043884456158 2023-01-23 15:34:44.072637: step: 20/1541, loss: 0.0012971041724085808 2023-01-23 15:34:44.948819: step: 22/1541, loss: 0.027433600276708603 2023-01-23 15:34:45.854675: step: 24/1541, loss: 0.1441715508699417 2023-01-23 15:34:46.743385: step: 26/1541, loss: 0.03160282224416733 2023-01-23 15:34:47.614265: step: 28/1541, loss: 0.004861755296587944 2023-01-23 15:34:48.493053: step: 30/1541, loss: 0.003702320624142885 2023-01-23 15:34:49.404049: step: 32/1541, loss: 0.3546944558620453 2023-01-23 15:34:50.324472: step: 34/1541, loss: 0.015912523493170738 2023-01-23 15:34:51.225497: step: 36/1541, loss: 0.29383355379104614 2023-01-23 15:34:52.057644: step: 38/1541, loss: 0.0020901074167340994 2023-01-23 15:34:52.929899: step: 40/1541, loss: 0.06274284422397614 2023-01-23 15:34:53.808571: step: 42/1541, loss: 0.0006496210116893053 2023-01-23 15:34:54.764447: step: 44/1541, loss: 0.001091110403649509 2023-01-23 15:34:55.607991: step: 46/1541, loss: 0.006856425665318966 2023-01-23 15:34:56.524948: step: 48/1541, loss: 0.00660090334713459 2023-01-23 15:34:57.433197: step: 50/1541, loss: 0.014995574951171875 2023-01-23 15:34:58.347754: step: 52/1541, loss: 0.00011032667680410668 2023-01-23 15:34:59.262565: step: 54/1541, loss: 0.006654310971498489 2023-01-23 15:35:00.232262: step: 56/1541, loss: 0.00288491602987051 2023-01-23 15:35:01.194169: step: 58/1541, loss: 0.002103119157254696 2023-01-23 15:35:02.079331: step: 60/1541, loss: 0.0026293033733963966 2023-01-23 15:35:02.987457: step: 62/1541, loss: 0.22387656569480896 2023-01-23 15:35:03.936122: step: 64/1541, loss: 0.03735353797674179 2023-01-23 15:35:04.803684: step: 66/1541, loss: 0.07710273563861847 2023-01-23 15:35:05.660688: step: 68/1541, loss: 0.1259918510913849 2023-01-23 15:35:06.600518: step: 70/1541, loss: 0.09359043091535568 2023-01-23 15:35:07.495217: step: 72/1541, loss: 0.7150434851646423 2023-01-23 15:35:08.356928: step: 74/1541, loss: 0.4649346172809601 2023-01-23 15:35:09.214854: step: 76/1541, loss: 0.0004917516489513218 2023-01-23 15:35:10.083816: step: 78/1541, loss: 0.00041897455230355263 2023-01-23 15:35:10.913344: step: 80/1541, loss: 0.0027991044335067272 2023-01-23 15:35:11.798571: step: 82/1541, loss: 0.28700703382492065 2023-01-23 15:35:12.671118: step: 84/1541, loss: 0.00032462819945067167 2023-01-23 15:35:13.654107: step: 86/1541, loss: 0.0018756892532110214 2023-01-23 15:35:14.537869: step: 88/1541, loss: 0.002916578436270356 2023-01-23 15:35:15.442483: step: 90/1541, loss: 0.000734793720766902 2023-01-23 15:35:16.343509: step: 92/1541, loss: 0.29499009251594543 2023-01-23 15:35:17.204658: step: 94/1541, loss: 0.018072960898280144 2023-01-23 15:35:18.071944: step: 96/1541, loss: 0.0065579721704125404 2023-01-23 15:35:18.925972: step: 98/1541, loss: 0.01030509453266859 2023-01-23 15:35:19.885524: step: 100/1541, loss: 0.2294408082962036 2023-01-23 15:35:20.801277: step: 102/1541, loss: 0.17676253616809845 2023-01-23 15:35:21.678421: step: 104/1541, loss: 0.5972090363502502 2023-01-23 15:35:22.546579: step: 106/1541, loss: 0.023868359625339508 2023-01-23 15:35:23.448187: step: 108/1541, loss: 0.03499625250697136 2023-01-23 15:35:24.318531: step: 110/1541, loss: 0.00320203579030931 2023-01-23 15:35:25.236899: step: 112/1541, loss: 0.0026574369985610247 2023-01-23 15:35:26.104621: step: 114/1541, loss: 0.2108362913131714 2023-01-23 15:35:26.990898: step: 116/1541, loss: 0.0014547037426382303 2023-01-23 15:35:27.921378: step: 118/1541, loss: 0.007253206800669432 2023-01-23 15:35:28.804285: step: 120/1541, loss: 0.0024364222772419453 2023-01-23 15:35:29.700976: step: 122/1541, loss: 0.005575246177613735 2023-01-23 15:35:30.629042: step: 124/1541, loss: 0.006179546471685171 2023-01-23 15:35:31.486082: step: 126/1541, loss: 0.013731323182582855 2023-01-23 15:35:32.370650: step: 128/1541, loss: 0.01772303134202957 2023-01-23 15:35:33.286058: step: 130/1541, loss: 0.02897394634783268 2023-01-23 15:35:34.183962: step: 132/1541, loss: 0.16702938079833984 2023-01-23 15:35:35.089866: step: 134/1541, loss: 0.022688983008265495 2023-01-23 15:35:36.030550: step: 136/1541, loss: 0.15834760665893555 2023-01-23 15:35:36.903673: step: 138/1541, loss: 0.06855620443820953 2023-01-23 15:35:37.814755: step: 140/1541, loss: 0.3156340718269348 2023-01-23 15:35:38.718654: step: 142/1541, loss: 0.008613613434135914 2023-01-23 15:35:39.614724: step: 144/1541, loss: 0.0036567645147442818 2023-01-23 15:35:40.539909: step: 146/1541, loss: 0.06442706286907196 2023-01-23 15:35:41.402758: step: 148/1541, loss: 0.019273003563284874 2023-01-23 15:35:42.266165: step: 150/1541, loss: 0.03589167073369026 2023-01-23 15:35:43.170928: step: 152/1541, loss: 0.0019041880732402205 2023-01-23 15:35:44.086825: step: 154/1541, loss: 0.0328756608068943 2023-01-23 15:35:44.947851: step: 156/1541, loss: 0.0064547923393547535 2023-01-23 15:35:45.922964: step: 158/1541, loss: 0.005412701517343521 2023-01-23 15:35:46.845650: step: 160/1541, loss: 0.002017623046413064 2023-01-23 15:35:47.757308: step: 162/1541, loss: 4.729811189463362e-05 2023-01-23 15:35:48.628166: step: 164/1541, loss: 0.13135385513305664 2023-01-23 15:35:49.513229: step: 166/1541, loss: 0.0024517809506505728 2023-01-23 15:35:50.399827: step: 168/1541, loss: 0.39913463592529297 2023-01-23 15:35:51.327457: step: 170/1541, loss: 0.05639205500483513 2023-01-23 15:35:52.242225: step: 172/1541, loss: 0.08132131397724152 2023-01-23 15:35:53.142053: step: 174/1541, loss: 0.000821072026155889 2023-01-23 15:35:54.094017: step: 176/1541, loss: 0.0005924159195274115 2023-01-23 15:35:55.082245: step: 178/1541, loss: 0.0024205727968364954 2023-01-23 15:35:55.976651: step: 180/1541, loss: 0.29141396284103394 2023-01-23 15:35:56.897208: step: 182/1541, loss: 0.0016962123336270452 2023-01-23 15:35:57.813995: step: 184/1541, loss: 0.02458515763282776 2023-01-23 15:35:58.727075: step: 186/1541, loss: 0.002769836224615574 2023-01-23 15:35:59.587507: step: 188/1541, loss: 0.004701548721641302 2023-01-23 15:36:00.469413: step: 190/1541, loss: 0.044569239020347595 2023-01-23 15:36:01.353813: step: 192/1541, loss: 0.02269962802529335 2023-01-23 15:36:02.270456: step: 194/1541, loss: 0.2181287705898285 2023-01-23 15:36:03.149785: step: 196/1541, loss: 0.0012312056496739388 2023-01-23 15:36:04.058318: step: 198/1541, loss: 0.0017316709272563457 2023-01-23 15:36:04.941547: step: 200/1541, loss: 0.0038112637121230364 2023-01-23 15:36:05.797273: step: 202/1541, loss: 0.3229347765445709 2023-01-23 15:36:06.688613: step: 204/1541, loss: 0.00012015781248919666 2023-01-23 15:36:07.605872: step: 206/1541, loss: 0.01273195818066597 2023-01-23 15:36:08.481945: step: 208/1541, loss: 0.0014551173662766814 2023-01-23 15:36:09.378502: step: 210/1541, loss: 0.04745034500956535 2023-01-23 15:36:10.241842: step: 212/1541, loss: 0.000553737161681056 2023-01-23 15:36:11.103241: step: 214/1541, loss: 0.10840224474668503 2023-01-23 15:36:12.002891: step: 216/1541, loss: 0.0023887776769697666 2023-01-23 15:36:12.934443: step: 218/1541, loss: 0.30124589800834656 2023-01-23 15:36:13.815865: step: 220/1541, loss: 0.01412928756326437 2023-01-23 15:36:14.768915: step: 222/1541, loss: 0.11638987064361572 2023-01-23 15:36:15.687740: step: 224/1541, loss: 0.0008816985064186156 2023-01-23 15:36:16.559985: step: 226/1541, loss: 0.0009489146759733558 2023-01-23 15:36:17.472602: step: 228/1541, loss: 0.023444024845957756 2023-01-23 15:36:18.370122: step: 230/1541, loss: 0.0023937358055263758 2023-01-23 15:36:19.296549: step: 232/1541, loss: 0.058658841997385025 2023-01-23 15:36:20.198316: step: 234/1541, loss: 0.10727740079164505 2023-01-23 15:36:21.108535: step: 236/1541, loss: 0.00048638356383889914 2023-01-23 15:36:21.994609: step: 238/1541, loss: 0.06810197234153748 2023-01-23 15:36:22.913524: step: 240/1541, loss: 0.04190440475940704 2023-01-23 15:36:23.873237: step: 242/1541, loss: 0.003476416226476431 2023-01-23 15:36:24.753418: step: 244/1541, loss: 0.06429721415042877 2023-01-23 15:36:25.632483: step: 246/1541, loss: 0.0012289953883737326 2023-01-23 15:36:26.559644: step: 248/1541, loss: 0.0010648120660334826 2023-01-23 15:36:27.486846: step: 250/1541, loss: 0.026459990069270134 2023-01-23 15:36:28.420531: step: 252/1541, loss: 0.21741525828838348 2023-01-23 15:36:29.409903: step: 254/1541, loss: 0.03941883519291878 2023-01-23 15:36:30.331924: step: 256/1541, loss: 0.004206167533993721 2023-01-23 15:36:31.262971: step: 258/1541, loss: 0.0005567509797401726 2023-01-23 15:36:32.215224: step: 260/1541, loss: 0.0006085358327254653 2023-01-23 15:36:33.108894: step: 262/1541, loss: 0.001228209468536079 2023-01-23 15:36:34.068279: step: 264/1541, loss: 0.191736102104187 2023-01-23 15:36:34.948143: step: 266/1541, loss: 0.03954432159662247 2023-01-23 15:36:35.887084: step: 268/1541, loss: 0.25561389327049255 2023-01-23 15:36:36.864210: step: 270/1541, loss: 0.05227431654930115 2023-01-23 15:36:37.785550: step: 272/1541, loss: 0.019957607612013817 2023-01-23 15:36:38.651464: step: 274/1541, loss: 0.03724155202507973 2023-01-23 15:36:39.556392: step: 276/1541, loss: 0.0217991191893816 2023-01-23 15:36:40.452155: step: 278/1541, loss: 0.23389650881290436 2023-01-23 15:36:41.332611: step: 280/1541, loss: 0.05996153503656387 2023-01-23 15:36:42.230377: step: 282/1541, loss: 0.014947508461773396 2023-01-23 15:36:43.115553: step: 284/1541, loss: 0.007740527391433716 2023-01-23 15:36:44.007772: step: 286/1541, loss: 0.004363568499684334 2023-01-23 15:36:44.880303: step: 288/1541, loss: 0.128758043050766 2023-01-23 15:36:45.783306: step: 290/1541, loss: 0.017903288826346397 2023-01-23 15:36:46.653666: step: 292/1541, loss: 0.0016686782473698258 2023-01-23 15:36:47.504337: step: 294/1541, loss: 0.0863364040851593 2023-01-23 15:36:48.374419: step: 296/1541, loss: 0.004714139737188816 2023-01-23 15:36:49.247616: step: 298/1541, loss: 1.042851209640503 2023-01-23 15:36:50.106082: step: 300/1541, loss: 0.009755814447999 2023-01-23 15:36:51.023828: step: 302/1541, loss: 0.01159083191305399 2023-01-23 15:36:51.914651: step: 304/1541, loss: 0.001099899411201477 2023-01-23 15:36:52.880815: step: 306/1541, loss: 0.0002985215687658638 2023-01-23 15:36:53.800189: step: 308/1541, loss: 0.11500431597232819 2023-01-23 15:36:54.676737: step: 310/1541, loss: 0.03151591494679451 2023-01-23 15:36:55.621158: step: 312/1541, loss: 0.43321260809898376 2023-01-23 15:36:56.526393: step: 314/1541, loss: 0.3872299790382385 2023-01-23 15:36:57.398093: step: 316/1541, loss: 0.029962453991174698 2023-01-23 15:36:58.272520: step: 318/1541, loss: 0.04967600479722023 2023-01-23 15:36:59.166684: step: 320/1541, loss: 0.0034818295389413834 2023-01-23 15:37:00.054169: step: 322/1541, loss: 0.05976514145731926 2023-01-23 15:37:00.911397: step: 324/1541, loss: 0.00199629133567214 2023-01-23 15:37:01.802908: step: 326/1541, loss: 0.03362613916397095 2023-01-23 15:37:02.686464: step: 328/1541, loss: 0.00551846157759428 2023-01-23 15:37:03.595718: step: 330/1541, loss: 0.0930972471833229 2023-01-23 15:37:04.456170: step: 332/1541, loss: 0.03440835699439049 2023-01-23 15:37:05.290803: step: 334/1541, loss: 0.015407832339406013 2023-01-23 15:37:06.181485: step: 336/1541, loss: 0.03659483417868614 2023-01-23 15:37:07.077749: step: 338/1541, loss: 0.14284776151180267 2023-01-23 15:37:08.034809: step: 340/1541, loss: 0.0006135471630841494 2023-01-23 15:37:08.936344: step: 342/1541, loss: 0.1253562718629837 2023-01-23 15:37:09.840588: step: 344/1541, loss: 0.282261461019516 2023-01-23 15:37:10.711656: step: 346/1541, loss: 0.213809996843338 2023-01-23 15:37:11.608754: step: 348/1541, loss: 0.03533785790205002 2023-01-23 15:37:12.527191: step: 350/1541, loss: 0.29285553097724915 2023-01-23 15:37:13.447840: step: 352/1541, loss: 0.19293247163295746 2023-01-23 15:37:14.368829: step: 354/1541, loss: 0.03414647653698921 2023-01-23 15:37:15.245095: step: 356/1541, loss: 0.00743416091427207 2023-01-23 15:37:16.132531: step: 358/1541, loss: 0.1745859980583191 2023-01-23 15:37:16.978850: step: 360/1541, loss: 0.0065483227372169495 2023-01-23 15:37:17.851476: step: 362/1541, loss: 0.015754926949739456 2023-01-23 15:37:18.740435: step: 364/1541, loss: 0.01102779246866703 2023-01-23 15:37:19.651464: step: 366/1541, loss: 0.0043164463713765144 2023-01-23 15:37:20.567860: step: 368/1541, loss: 0.0006786591839045286 2023-01-23 15:37:21.529157: step: 370/1541, loss: 0.031630951911211014 2023-01-23 15:37:22.441703: step: 372/1541, loss: 0.4750368297100067 2023-01-23 15:37:23.350422: step: 374/1541, loss: 0.31053411960601807 2023-01-23 15:37:24.215962: step: 376/1541, loss: 0.04442669451236725 2023-01-23 15:37:25.184272: step: 378/1541, loss: 0.00016040577611420304 2023-01-23 15:37:26.060281: step: 380/1541, loss: 0.0021390020847320557 2023-01-23 15:37:26.926943: step: 382/1541, loss: 0.0010751275112852454 2023-01-23 15:37:27.807959: step: 384/1541, loss: 0.19960282742977142 2023-01-23 15:37:28.684950: step: 386/1541, loss: 0.0026964826975017786 2023-01-23 15:37:29.583147: step: 388/1541, loss: 0.08832678198814392 2023-01-23 15:37:30.521417: step: 390/1541, loss: 0.019079986959695816 2023-01-23 15:37:31.407811: step: 392/1541, loss: 0.00010221985576208681 2023-01-23 15:37:32.330071: step: 394/1541, loss: 0.001564400619827211 2023-01-23 15:37:33.248940: step: 396/1541, loss: 0.0708572044968605 2023-01-23 15:37:34.177510: step: 398/1541, loss: 0.5058127641677856 2023-01-23 15:37:35.063788: step: 400/1541, loss: 0.003247069427743554 2023-01-23 15:37:35.960613: step: 402/1541, loss: 0.00034830914228223264 2023-01-23 15:37:36.945809: step: 404/1541, loss: 0.1197233572602272 2023-01-23 15:37:37.827006: step: 406/1541, loss: 0.050315503031015396 2023-01-23 15:37:38.699980: step: 408/1541, loss: 0.12999434769153595 2023-01-23 15:37:39.589435: step: 410/1541, loss: 0.0065672206692397594 2023-01-23 15:37:40.482079: step: 412/1541, loss: 0.08729466795921326 2023-01-23 15:37:41.358111: step: 414/1541, loss: 0.0030925797764211893 2023-01-23 15:37:42.199585: step: 416/1541, loss: 0.020884349942207336 2023-01-23 15:37:43.113453: step: 418/1541, loss: 0.020807670429348946 2023-01-23 15:37:44.031055: step: 420/1541, loss: 0.02914314717054367 2023-01-23 15:37:44.864565: step: 422/1541, loss: 0.001307455007918179 2023-01-23 15:37:45.749708: step: 424/1541, loss: 0.008324463851749897 2023-01-23 15:37:46.700433: step: 426/1541, loss: 0.049138009548187256 2023-01-23 15:37:47.637039: step: 428/1541, loss: 0.17713670432567596 2023-01-23 15:37:48.546359: step: 430/1541, loss: 0.24047259986400604 2023-01-23 15:37:49.417829: step: 432/1541, loss: 0.02287423238158226 2023-01-23 15:37:50.398513: step: 434/1541, loss: 0.0187204759567976 2023-01-23 15:37:51.359119: step: 436/1541, loss: 0.12271987646818161 2023-01-23 15:37:52.267132: step: 438/1541, loss: 0.006008432246744633 2023-01-23 15:37:53.205641: step: 440/1541, loss: 0.0027649831026792526 2023-01-23 15:37:54.087631: step: 442/1541, loss: 0.051355548202991486 2023-01-23 15:37:55.002437: step: 444/1541, loss: 0.009704772382974625 2023-01-23 15:37:55.926076: step: 446/1541, loss: 0.11767573654651642 2023-01-23 15:37:56.815244: step: 448/1541, loss: 0.05960521101951599 2023-01-23 15:37:57.681092: step: 450/1541, loss: 0.002038380363956094 2023-01-23 15:37:58.540243: step: 452/1541, loss: 0.0006008856580592692 2023-01-23 15:37:59.429018: step: 454/1541, loss: 0.14026321470737457 2023-01-23 15:38:00.349949: step: 456/1541, loss: 0.0015203722286969423 2023-01-23 15:38:01.209757: step: 458/1541, loss: 0.0008600373985245824 2023-01-23 15:38:02.127011: step: 460/1541, loss: 0.029940227046608925 2023-01-23 15:38:03.019564: step: 462/1541, loss: 0.17791393399238586 2023-01-23 15:38:03.911721: step: 464/1541, loss: 0.0027783901896327734 2023-01-23 15:38:04.773642: step: 466/1541, loss: 0.008297491818666458 2023-01-23 15:38:05.682251: step: 468/1541, loss: 0.09321466088294983 2023-01-23 15:38:06.583113: step: 470/1541, loss: 0.1360674500465393 2023-01-23 15:38:07.494537: step: 472/1541, loss: 0.008519736118614674 2023-01-23 15:38:08.366910: step: 474/1541, loss: 0.01285947859287262 2023-01-23 15:38:09.284538: step: 476/1541, loss: 0.3135996460914612 2023-01-23 15:38:10.176075: step: 478/1541, loss: 0.008074606768786907 2023-01-23 15:38:11.032418: step: 480/1541, loss: 0.0019602158572524786 2023-01-23 15:38:11.978797: step: 482/1541, loss: 0.14233283698558807 2023-01-23 15:38:12.869476: step: 484/1541, loss: 0.0025019925087690353 2023-01-23 15:38:13.784054: step: 486/1541, loss: 0.0015280673978850245 2023-01-23 15:38:14.671667: step: 488/1541, loss: 0.17353186011314392 2023-01-23 15:38:15.576687: step: 490/1541, loss: 0.2981252670288086 2023-01-23 15:38:16.516920: step: 492/1541, loss: 0.20669203996658325 2023-01-23 15:38:17.408467: step: 494/1541, loss: 0.015485269017517567 2023-01-23 15:38:18.275783: step: 496/1541, loss: 0.007479244843125343 2023-01-23 15:38:19.173582: step: 498/1541, loss: 0.10894138365983963 2023-01-23 15:38:20.041033: step: 500/1541, loss: 0.2598535418510437 2023-01-23 15:38:20.986056: step: 502/1541, loss: 0.0006970842368900776 2023-01-23 15:38:21.941452: step: 504/1541, loss: 0.24933229386806488 2023-01-23 15:38:22.842616: step: 506/1541, loss: 0.016430390998721123 2023-01-23 15:38:23.754981: step: 508/1541, loss: 0.2611241042613983 2023-01-23 15:38:24.654250: step: 510/1541, loss: 0.07532832026481628 2023-01-23 15:38:25.607030: step: 512/1541, loss: 0.000801795395091176 2023-01-23 15:38:26.562428: step: 514/1541, loss: 0.041702933609485626 2023-01-23 15:38:27.434527: step: 516/1541, loss: 0.008285962045192719 2023-01-23 15:38:28.368381: step: 518/1541, loss: 0.01445845142006874 2023-01-23 15:38:29.244563: step: 520/1541, loss: 0.0027610580436885357 2023-01-23 15:38:30.117217: step: 522/1541, loss: 0.37559372186660767 2023-01-23 15:38:30.988152: step: 524/1541, loss: 0.007646478712558746 2023-01-23 15:38:31.925567: step: 526/1541, loss: 1.1814759969711304 2023-01-23 15:38:32.814884: step: 528/1541, loss: 0.05922906473278999 2023-01-23 15:38:33.686095: step: 530/1541, loss: 0.004586657043546438 2023-01-23 15:38:34.583625: step: 532/1541, loss: 0.0018809036118909717 2023-01-23 15:38:35.512075: step: 534/1541, loss: 0.04785178229212761 2023-01-23 15:38:36.472946: step: 536/1541, loss: 0.031201716512441635 2023-01-23 15:38:37.361133: step: 538/1541, loss: 0.42399299144744873 2023-01-23 15:38:38.285015: step: 540/1541, loss: 0.00217597559094429 2023-01-23 15:38:39.179105: step: 542/1541, loss: 0.005059610586613417 2023-01-23 15:38:40.105428: step: 544/1541, loss: 0.014681674540042877 2023-01-23 15:38:41.034334: step: 546/1541, loss: 0.8763584494590759 2023-01-23 15:38:41.879649: step: 548/1541, loss: 0.0046599991619586945 2023-01-23 15:38:42.801924: step: 550/1541, loss: 0.0004478461924009025 2023-01-23 15:38:43.670882: step: 552/1541, loss: 0.028458505868911743 2023-01-23 15:38:44.591388: step: 554/1541, loss: 0.0007944087265059352 2023-01-23 15:38:45.462981: step: 556/1541, loss: 0.0031167420092970133 2023-01-23 15:38:46.349494: step: 558/1541, loss: 0.00138552812859416 2023-01-23 15:38:47.243617: step: 560/1541, loss: 0.012805943377315998 2023-01-23 15:38:48.143650: step: 562/1541, loss: 0.009813006967306137 2023-01-23 15:38:49.088305: step: 564/1541, loss: 0.06679932773113251 2023-01-23 15:38:49.984280: step: 566/1541, loss: 0.004328382667154074 2023-01-23 15:38:50.904697: step: 568/1541, loss: 0.2773871421813965 2023-01-23 15:38:51.758328: step: 570/1541, loss: 0.01303078979253769 2023-01-23 15:38:52.627764: step: 572/1541, loss: 0.0030185533687472343 2023-01-23 15:38:53.513246: step: 574/1541, loss: 0.06792037189006805 2023-01-23 15:38:54.410354: step: 576/1541, loss: 0.042586132884025574 2023-01-23 15:38:55.308692: step: 578/1541, loss: 0.002836148487403989 2023-01-23 15:38:56.234596: step: 580/1541, loss: 0.008842730894684792 2023-01-23 15:38:57.123094: step: 582/1541, loss: 0.014649279415607452 2023-01-23 15:38:57.974287: step: 584/1541, loss: 0.010906330309808254 2023-01-23 15:38:58.927231: step: 586/1541, loss: 0.37722331285476685 2023-01-23 15:38:59.807914: step: 588/1541, loss: 0.16243840754032135 2023-01-23 15:39:00.712668: step: 590/1541, loss: 0.053036514669656754 2023-01-23 15:39:01.602201: step: 592/1541, loss: 0.00036084518069401383 2023-01-23 15:39:02.477573: step: 594/1541, loss: 0.0010376323480159044 2023-01-23 15:39:03.342853: step: 596/1541, loss: 0.07315288484096527 2023-01-23 15:39:04.235667: step: 598/1541, loss: 0.00028122199000790715 2023-01-23 15:39:05.123073: step: 600/1541, loss: 0.010524768382310867 2023-01-23 15:39:06.006763: step: 602/1541, loss: 0.05887160450220108 2023-01-23 15:39:06.893148: step: 604/1541, loss: 0.0007391300750896335 2023-01-23 15:39:07.794588: step: 606/1541, loss: 0.00024437811225652695 2023-01-23 15:39:08.708412: step: 608/1541, loss: 0.0001156656289822422 2023-01-23 15:39:09.579862: step: 610/1541, loss: 0.2004428654909134 2023-01-23 15:39:10.499321: step: 612/1541, loss: 0.5825507640838623 2023-01-23 15:39:11.362964: step: 614/1541, loss: 0.0020843837410211563 2023-01-23 15:39:12.305566: step: 616/1541, loss: 0.003124615177512169 2023-01-23 15:39:13.199381: step: 618/1541, loss: 0.00013217962987255305 2023-01-23 15:39:14.068226: step: 620/1541, loss: 0.005544099025428295 2023-01-23 15:39:14.951532: step: 622/1541, loss: 0.0011304893996566534 2023-01-23 15:39:15.877351: step: 624/1541, loss: 0.058335673063993454 2023-01-23 15:39:16.770879: step: 626/1541, loss: 0.00012619001790881157 2023-01-23 15:39:17.631142: step: 628/1541, loss: 0.011537127196788788 2023-01-23 15:39:18.563994: step: 630/1541, loss: 0.045970045030117035 2023-01-23 15:39:19.496269: step: 632/1541, loss: 0.018906662240624428 2023-01-23 15:39:20.374138: step: 634/1541, loss: 0.03648194670677185 2023-01-23 15:39:21.239136: step: 636/1541, loss: 0.09471528977155685 2023-01-23 15:39:22.110488: step: 638/1541, loss: 0.0007792391115799546 2023-01-23 15:39:23.008059: step: 640/1541, loss: 0.0014559414703398943 2023-01-23 15:39:23.883151: step: 642/1541, loss: 0.0002692646230570972 2023-01-23 15:39:24.796113: step: 644/1541, loss: 0.04888572543859482 2023-01-23 15:39:25.651373: step: 646/1541, loss: 0.00290305120870471 2023-01-23 15:39:26.554159: step: 648/1541, loss: 0.043528590351343155 2023-01-23 15:39:27.475014: step: 650/1541, loss: 0.2186892181634903 2023-01-23 15:39:28.336421: step: 652/1541, loss: 0.030563194304704666 2023-01-23 15:39:29.240235: step: 654/1541, loss: 0.002308283466845751 2023-01-23 15:39:30.145443: step: 656/1541, loss: 0.03999001532793045 2023-01-23 15:39:31.041703: step: 658/1541, loss: 0.02308187447488308 2023-01-23 15:39:31.945602: step: 660/1541, loss: 0.0014621830778196454 2023-01-23 15:39:32.866696: step: 662/1541, loss: 0.4719213843345642 2023-01-23 15:39:33.704797: step: 664/1541, loss: 0.0033679166808724403 2023-01-23 15:39:34.595853: step: 666/1541, loss: 0.45804980397224426 2023-01-23 15:39:35.468482: step: 668/1541, loss: 0.0005306192906573415 2023-01-23 15:39:36.374943: step: 670/1541, loss: 0.023736434057354927 2023-01-23 15:39:37.298019: step: 672/1541, loss: 0.40653663873672485 2023-01-23 15:39:38.187499: step: 674/1541, loss: 0.00093102763639763 2023-01-23 15:39:39.051666: step: 676/1541, loss: 0.02846868522465229 2023-01-23 15:39:39.909360: step: 678/1541, loss: 0.0011485766153782606 2023-01-23 15:39:40.807609: step: 680/1541, loss: 0.002947620116174221 2023-01-23 15:39:41.730396: step: 682/1541, loss: 1.2332795449765399e-05 2023-01-23 15:39:42.635852: step: 684/1541, loss: 0.25846654176712036 2023-01-23 15:39:43.509702: step: 686/1541, loss: 4.967094355379231e-06 2023-01-23 15:39:44.477258: step: 688/1541, loss: 0.17620541155338287 2023-01-23 15:39:45.323441: step: 690/1541, loss: 0.024345451965928078 2023-01-23 15:39:46.235895: step: 692/1541, loss: 0.005897266324609518 2023-01-23 15:39:47.133159: step: 694/1541, loss: 0.13675557076931 2023-01-23 15:39:48.015024: step: 696/1541, loss: 0.023471476510167122 2023-01-23 15:39:48.901235: step: 698/1541, loss: 0.008565915748476982 2023-01-23 15:39:49.805167: step: 700/1541, loss: 0.17310082912445068 2023-01-23 15:39:50.659068: step: 702/1541, loss: 0.18551166355609894 2023-01-23 15:39:51.496236: step: 704/1541, loss: 0.0038463263772428036 2023-01-23 15:39:52.400485: step: 706/1541, loss: 0.010870221070945263 2023-01-23 15:39:53.295930: step: 708/1541, loss: 0.28174877166748047 2023-01-23 15:39:54.175138: step: 710/1541, loss: 0.0005104694282636046 2023-01-23 15:39:55.124655: step: 712/1541, loss: 0.03872711583971977 2023-01-23 15:39:56.036748: step: 714/1541, loss: 0.0033215153962373734 2023-01-23 15:39:57.070788: step: 716/1541, loss: 0.04073537886142731 2023-01-23 15:39:57.993190: step: 718/1541, loss: 0.0056901052594184875 2023-01-23 15:39:58.961067: step: 720/1541, loss: 0.2570381760597229 2023-01-23 15:39:59.866445: step: 722/1541, loss: 0.0008389179711230099 2023-01-23 15:40:00.817073: step: 724/1541, loss: 0.0006822932045906782 2023-01-23 15:40:01.785857: step: 726/1541, loss: 0.01818051189184189 2023-01-23 15:40:02.659644: step: 728/1541, loss: 0.008624536916613579 2023-01-23 15:40:03.608104: step: 730/1541, loss: 0.0006191303837113082 2023-01-23 15:40:04.532885: step: 732/1541, loss: 0.0007167455623857677 2023-01-23 15:40:05.450592: step: 734/1541, loss: 0.048682570457458496 2023-01-23 15:40:06.359410: step: 736/1541, loss: 0.0033508960623294115 2023-01-23 15:40:07.214338: step: 738/1541, loss: 0.022304905578494072 2023-01-23 15:40:08.064151: step: 740/1541, loss: 0.0029726880602538586 2023-01-23 15:40:08.939206: step: 742/1541, loss: 0.014490692876279354 2023-01-23 15:40:09.908869: step: 744/1541, loss: 0.03243699297308922 2023-01-23 15:40:10.745968: step: 746/1541, loss: 0.004063234198838472 2023-01-23 15:40:11.595484: step: 748/1541, loss: 0.014608154073357582 2023-01-23 15:40:12.470873: step: 750/1541, loss: 0.0054727597162127495 2023-01-23 15:40:13.425954: step: 752/1541, loss: 0.009741971269249916 2023-01-23 15:40:14.360792: step: 754/1541, loss: 0.0034082229249179363 2023-01-23 15:40:15.267408: step: 756/1541, loss: 0.011708010919392109 2023-01-23 15:40:16.151754: step: 758/1541, loss: 0.0010793599067255855 2023-01-23 15:40:17.062614: step: 760/1541, loss: 0.12677320837974548 2023-01-23 15:40:17.978655: step: 762/1541, loss: 0.00036195496795699 2023-01-23 15:40:18.878666: step: 764/1541, loss: 0.029469557106494904 2023-01-23 15:40:19.798781: step: 766/1541, loss: 0.010607527568936348 2023-01-23 15:40:20.700075: step: 768/1541, loss: 1.632637858390808 2023-01-23 15:40:21.639313: step: 770/1541, loss: 0.014567370526492596 2023-01-23 15:40:22.549069: step: 772/1541, loss: 0.0035550587344914675 2023-01-23 15:40:23.397162: step: 774/1541, loss: 0.0006886055343784392 2023-01-23 15:40:24.225620: step: 776/1541, loss: 0.032011110335588455 2023-01-23 15:40:25.133076: step: 778/1541, loss: 0.00026556203374639153 2023-01-23 15:40:26.014904: step: 780/1541, loss: 0.011703409254550934 2023-01-23 15:40:26.877779: step: 782/1541, loss: 0.15616461634635925 2023-01-23 15:40:27.749529: step: 784/1541, loss: 0.0014281400945037603 2023-01-23 15:40:28.647994: step: 786/1541, loss: 0.006469801068305969 2023-01-23 15:40:29.508143: step: 788/1541, loss: 0.0010002419585362077 2023-01-23 15:40:30.428735: step: 790/1541, loss: 0.006132756359875202 2023-01-23 15:40:31.307026: step: 792/1541, loss: 0.14561694860458374 2023-01-23 15:40:32.242574: step: 794/1541, loss: 0.004244827199727297 2023-01-23 15:40:33.138755: step: 796/1541, loss: 0.07673521339893341 2023-01-23 15:40:34.111052: step: 798/1541, loss: 0.11596889048814774 2023-01-23 15:40:34.979362: step: 800/1541, loss: 0.10547830909490585 2023-01-23 15:40:35.873514: step: 802/1541, loss: 0.034651562571525574 2023-01-23 15:40:36.785886: step: 804/1541, loss: 0.013778937049210072 2023-01-23 15:40:37.662746: step: 806/1541, loss: 0.06925033777952194 2023-01-23 15:40:38.511315: step: 808/1541, loss: 0.013045444153249264 2023-01-23 15:40:39.399011: step: 810/1541, loss: 0.20509186387062073 2023-01-23 15:40:40.330343: step: 812/1541, loss: 0.0007145864656195045 2023-01-23 15:40:41.208829: step: 814/1541, loss: 0.1863243281841278 2023-01-23 15:40:42.088350: step: 816/1541, loss: 0.002480454510077834 2023-01-23 15:40:43.070153: step: 818/1541, loss: 0.003853512229397893 2023-01-23 15:40:43.964754: step: 820/1541, loss: 0.07585537433624268 2023-01-23 15:40:44.845436: step: 822/1541, loss: 0.0005690618418157101 2023-01-23 15:40:45.730970: step: 824/1541, loss: 0.0003900880692526698 2023-01-23 15:40:46.615414: step: 826/1541, loss: 0.022185271605849266 2023-01-23 15:40:47.494640: step: 828/1541, loss: 0.0008659042650833726 2023-01-23 15:40:48.361295: step: 830/1541, loss: 0.022710300981998444 2023-01-23 15:40:49.247082: step: 832/1541, loss: 0.00017182520241476595 2023-01-23 15:40:50.201757: step: 834/1541, loss: 0.012320134788751602 2023-01-23 15:40:51.170383: step: 836/1541, loss: 0.05564544349908829 2023-01-23 15:40:52.074243: step: 838/1541, loss: 0.07350829988718033 2023-01-23 15:40:52.980807: step: 840/1541, loss: 0.0017718634335324168 2023-01-23 15:40:53.873761: step: 842/1541, loss: 0.03489896282553673 2023-01-23 15:40:54.718244: step: 844/1541, loss: 0.03435390070080757 2023-01-23 15:40:55.597565: step: 846/1541, loss: 0.004443713463842869 2023-01-23 15:40:56.477545: step: 848/1541, loss: 0.0017433167668059468 2023-01-23 15:40:57.402585: step: 850/1541, loss: 0.1982438713312149 2023-01-23 15:40:58.302493: step: 852/1541, loss: 0.009995426051318645 2023-01-23 15:40:59.169356: step: 854/1541, loss: 0.002769275801256299 2023-01-23 15:41:00.142277: step: 856/1541, loss: 0.006186977960169315 2023-01-23 15:41:01.010877: step: 858/1541, loss: 0.09353501349687576 2023-01-23 15:41:01.940707: step: 860/1541, loss: 0.0031439047306776047 2023-01-23 15:41:02.830262: step: 862/1541, loss: 0.00033378289663232863 2023-01-23 15:41:03.754783: step: 864/1541, loss: 0.0070167966187000275 2023-01-23 15:41:04.635108: step: 866/1541, loss: 0.03256335109472275 2023-01-23 15:41:05.602083: step: 868/1541, loss: 0.018688375130295753 2023-01-23 15:41:06.559247: step: 870/1541, loss: 0.011714642867445946 2023-01-23 15:41:07.504482: step: 872/1541, loss: 0.020810825750231743 2023-01-23 15:41:08.431944: step: 874/1541, loss: 0.14365288615226746 2023-01-23 15:41:09.331277: step: 876/1541, loss: 0.03204651176929474 2023-01-23 15:41:10.182644: step: 878/1541, loss: 0.0061209676787257195 2023-01-23 15:41:11.061975: step: 880/1541, loss: 0.012591050937771797 2023-01-23 15:41:11.999953: step: 882/1541, loss: 0.0011477854568511248 2023-01-23 15:41:12.883156: step: 884/1541, loss: 0.005968485958874226 2023-01-23 15:41:13.806523: step: 886/1541, loss: 0.011698342859745026 2023-01-23 15:41:14.689246: step: 888/1541, loss: 0.0031070844270288944 2023-01-23 15:41:15.575020: step: 890/1541, loss: 0.000524295202922076 2023-01-23 15:41:16.501240: step: 892/1541, loss: 0.06808734685182571 2023-01-23 15:41:17.403680: step: 894/1541, loss: 0.07892286777496338 2023-01-23 15:41:18.289591: step: 896/1541, loss: 0.07753516733646393 2023-01-23 15:41:19.186031: step: 898/1541, loss: 0.0016995834885165095 2023-01-23 15:41:20.041218: step: 900/1541, loss: 0.0010134648764505982 2023-01-23 15:41:20.908670: step: 902/1541, loss: 0.08761832118034363 2023-01-23 15:41:21.809787: step: 904/1541, loss: 0.14158950746059418 2023-01-23 15:41:22.680313: step: 906/1541, loss: 0.04327325522899628 2023-01-23 15:41:23.563974: step: 908/1541, loss: 0.011992108076810837 2023-01-23 15:41:24.452691: step: 910/1541, loss: 0.20360490679740906 2023-01-23 15:41:25.363677: step: 912/1541, loss: 0.6714680790901184 2023-01-23 15:41:26.260134: step: 914/1541, loss: 0.021195854991674423 2023-01-23 15:41:27.247555: step: 916/1541, loss: 0.001342867617495358 2023-01-23 15:41:28.223511: step: 918/1541, loss: 0.011455954052507877 2023-01-23 15:41:29.124358: step: 920/1541, loss: 0.00204728776589036 2023-01-23 15:41:30.026073: step: 922/1541, loss: 0.03559050336480141 2023-01-23 15:41:30.997729: step: 924/1541, loss: 0.28871238231658936 2023-01-23 15:41:31.892417: step: 926/1541, loss: 0.6333028674125671 2023-01-23 15:41:32.767780: step: 928/1541, loss: 0.011299646459519863 2023-01-23 15:41:33.714721: step: 930/1541, loss: 0.13407567143440247 2023-01-23 15:41:34.624852: step: 932/1541, loss: 0.007696622051298618 2023-01-23 15:41:35.506411: step: 934/1541, loss: 0.06582362204790115 2023-01-23 15:41:36.418064: step: 936/1541, loss: 0.0007290565990842879 2023-01-23 15:41:37.309466: step: 938/1541, loss: 0.017623400315642357 2023-01-23 15:41:38.313747: step: 940/1541, loss: 0.0025342130102217197 2023-01-23 15:41:39.202439: step: 942/1541, loss: 0.15654441714286804 2023-01-23 15:41:40.050846: step: 944/1541, loss: 0.1487923115491867 2023-01-23 15:41:40.975094: step: 946/1541, loss: 0.008865037001669407 2023-01-23 15:41:41.863738: step: 948/1541, loss: 0.0009073861874639988 2023-01-23 15:41:42.780599: step: 950/1541, loss: 0.1913743019104004 2023-01-23 15:41:43.656691: step: 952/1541, loss: 0.04269979149103165 2023-01-23 15:41:44.550171: step: 954/1541, loss: 0.001469526789151132 2023-01-23 15:41:45.464177: step: 956/1541, loss: 0.025078000500798225 2023-01-23 15:41:46.314994: step: 958/1541, loss: 5.1359497774683405e-06 2023-01-23 15:41:47.241113: step: 960/1541, loss: 0.0043472847901284695 2023-01-23 15:41:48.144220: step: 962/1541, loss: 0.0006723974365741014 2023-01-23 15:41:49.043126: step: 964/1541, loss: 0.010250811465084553 2023-01-23 15:41:49.943071: step: 966/1541, loss: 0.021913252770900726 2023-01-23 15:41:50.815036: step: 968/1541, loss: 0.001745810266584158 2023-01-23 15:41:51.675866: step: 970/1541, loss: 0.001128740725107491 2023-01-23 15:41:52.554855: step: 972/1541, loss: 0.010188835673034191 2023-01-23 15:41:53.452114: step: 974/1541, loss: 0.08904166519641876 2023-01-23 15:41:54.312698: step: 976/1541, loss: 0.05393954738974571 2023-01-23 15:41:55.189791: step: 978/1541, loss: 0.12410317361354828 2023-01-23 15:41:56.083146: step: 980/1541, loss: 0.14983542263507843 2023-01-23 15:41:56.971488: step: 982/1541, loss: 0.005967467091977596 2023-01-23 15:41:57.833995: step: 984/1541, loss: 0.11480892449617386 2023-01-23 15:41:58.821720: step: 986/1541, loss: 0.024714354425668716 2023-01-23 15:41:59.775553: step: 988/1541, loss: 0.0019865655340254307 2023-01-23 15:42:00.649818: step: 990/1541, loss: 0.0065589346922934055 2023-01-23 15:42:01.623759: step: 992/1541, loss: 0.11104439944028854 2023-01-23 15:42:02.525848: step: 994/1541, loss: 0.005557198077440262 2023-01-23 15:42:03.433356: step: 996/1541, loss: 0.006789627950638533 2023-01-23 15:42:04.311737: step: 998/1541, loss: 1.3260926008224487 2023-01-23 15:42:05.147737: step: 1000/1541, loss: 0.0014075112994760275 2023-01-23 15:42:06.043266: step: 1002/1541, loss: 0.004261194262653589 2023-01-23 15:42:06.964354: step: 1004/1541, loss: 0.0029531701002269983 2023-01-23 15:42:07.854653: step: 1006/1541, loss: 0.0010732148075476289 2023-01-23 15:42:08.732632: step: 1008/1541, loss: 0.0009253007592633367 2023-01-23 15:42:09.641048: step: 1010/1541, loss: 0.0015492499805986881 2023-01-23 15:42:10.514286: step: 1012/1541, loss: 0.0018464014865458012 2023-01-23 15:42:11.405568: step: 1014/1541, loss: 0.0001455526944482699 2023-01-23 15:42:12.282878: step: 1016/1541, loss: 0.7294583916664124 2023-01-23 15:42:13.243633: step: 1018/1541, loss: 0.05678289011120796 2023-01-23 15:42:14.141456: step: 1020/1541, loss: 0.05125743895769119 2023-01-23 15:42:15.003647: step: 1022/1541, loss: 0.00034973208676092327 2023-01-23 15:42:15.899891: step: 1024/1541, loss: 0.27753737568855286 2023-01-23 15:42:16.757109: step: 1026/1541, loss: 0.012417728081345558 2023-01-23 15:42:17.602814: step: 1028/1541, loss: 0.012648324482142925 2023-01-23 15:42:18.590291: step: 1030/1541, loss: 0.04266682639718056 2023-01-23 15:42:19.530929: step: 1032/1541, loss: 0.06935379654169083 2023-01-23 15:42:20.403428: step: 1034/1541, loss: 0.3335736095905304 2023-01-23 15:42:21.314565: step: 1036/1541, loss: 0.10914867371320724 2023-01-23 15:42:22.219868: step: 1038/1541, loss: 0.06410577148199081 2023-01-23 15:42:23.111485: step: 1040/1541, loss: 0.03534824028611183 2023-01-23 15:42:24.010275: step: 1042/1541, loss: 0.0005572002846747637 2023-01-23 15:42:24.937529: step: 1044/1541, loss: 0.010052659548819065 2023-01-23 15:42:25.828790: step: 1046/1541, loss: 0.030569512397050858 2023-01-23 15:42:26.733164: step: 1048/1541, loss: 0.0009772609919309616 2023-01-23 15:42:27.648275: step: 1050/1541, loss: 0.2900805175304413 2023-01-23 15:42:28.596098: step: 1052/1541, loss: 0.03491467982530594 2023-01-23 15:42:29.440053: step: 1054/1541, loss: 0.000553562247660011 2023-01-23 15:42:30.341291: step: 1056/1541, loss: 0.011722191236913204 2023-01-23 15:42:31.219252: step: 1058/1541, loss: 0.042985040694475174 2023-01-23 15:42:32.077676: step: 1060/1541, loss: 0.08531733602285385 2023-01-23 15:42:32.997801: step: 1062/1541, loss: 0.03476881608366966 2023-01-23 15:42:33.907208: step: 1064/1541, loss: 0.019594676792621613 2023-01-23 15:42:34.801350: step: 1066/1541, loss: 0.001819062395952642 2023-01-23 15:42:35.733064: step: 1068/1541, loss: 0.004835884086787701 2023-01-23 15:42:36.636541: step: 1070/1541, loss: 0.0027542628813534975 2023-01-23 15:42:37.512720: step: 1072/1541, loss: 0.008036807179450989 2023-01-23 15:42:38.460175: step: 1074/1541, loss: 0.012490917928516865 2023-01-23 15:42:39.331563: step: 1076/1541, loss: 0.10628519207239151 2023-01-23 15:42:40.256260: step: 1078/1541, loss: 0.04595714807510376 2023-01-23 15:42:41.133449: step: 1080/1541, loss: 0.024160154163837433 2023-01-23 15:42:42.037221: step: 1082/1541, loss: 0.09731938689947128 2023-01-23 15:42:42.916534: step: 1084/1541, loss: 0.6336036920547485 2023-01-23 15:42:43.829350: step: 1086/1541, loss: 0.04657336324453354 2023-01-23 15:42:44.723277: step: 1088/1541, loss: 0.2248685210943222 2023-01-23 15:42:45.585992: step: 1090/1541, loss: 0.2619301974773407 2023-01-23 15:42:46.438970: step: 1092/1541, loss: 0.0023082897532731295 2023-01-23 15:42:47.314427: step: 1094/1541, loss: 0.08242626488208771 2023-01-23 15:42:48.256346: step: 1096/1541, loss: 0.04266773536801338 2023-01-23 15:42:49.146539: step: 1098/1541, loss: 0.004849027842283249 2023-01-23 15:42:50.051665: step: 1100/1541, loss: 0.03525618463754654 2023-01-23 15:42:50.938186: step: 1102/1541, loss: 0.32304057478904724 2023-01-23 15:42:51.889698: step: 1104/1541, loss: 0.012746218591928482 2023-01-23 15:42:52.759484: step: 1106/1541, loss: 0.005297140218317509 2023-01-23 15:42:53.635965: step: 1108/1541, loss: 0.04271404817700386 2023-01-23 15:42:54.569786: step: 1110/1541, loss: 0.003966981545090675 2023-01-23 15:42:55.477394: step: 1112/1541, loss: 0.12695349752902985 2023-01-23 15:42:56.350399: step: 1114/1541, loss: 0.0005987243494018912 2023-01-23 15:42:57.263811: step: 1116/1541, loss: 0.20151546597480774 2023-01-23 15:42:58.155135: step: 1118/1541, loss: 0.002294867765158415 2023-01-23 15:42:59.019955: step: 1120/1541, loss: 0.013938067480921745 2023-01-23 15:42:59.933888: step: 1122/1541, loss: 4.748248100280762 2023-01-23 15:43:00.821300: step: 1124/1541, loss: 0.014946863986551762 2023-01-23 15:43:01.710334: step: 1126/1541, loss: 0.00039955993997864425 2023-01-23 15:43:02.667876: step: 1128/1541, loss: 0.001592300832271576 2023-01-23 15:43:03.566711: step: 1130/1541, loss: 0.12520258128643036 2023-01-23 15:43:04.484197: step: 1132/1541, loss: 0.009552741423249245 2023-01-23 15:43:05.396958: step: 1134/1541, loss: 0.2901588976383209 2023-01-23 15:43:06.334299: step: 1136/1541, loss: 0.01833314262330532 2023-01-23 15:43:07.219412: step: 1138/1541, loss: 0.001879486138932407 2023-01-23 15:43:08.133831: step: 1140/1541, loss: 0.007106376811861992 2023-01-23 15:43:09.000799: step: 1142/1541, loss: 0.08368659019470215 2023-01-23 15:43:09.933994: step: 1144/1541, loss: 0.14465002715587616 2023-01-23 15:43:10.848134: step: 1146/1541, loss: 0.013735882937908173 2023-01-23 15:43:11.730518: step: 1148/1541, loss: 0.01199259888380766 2023-01-23 15:43:12.562528: step: 1150/1541, loss: 0.00127523357514292 2023-01-23 15:43:13.478492: step: 1152/1541, loss: 0.009334937669336796 2023-01-23 15:43:14.387479: step: 1154/1541, loss: 0.046957217156887054 2023-01-23 15:43:15.283738: step: 1156/1541, loss: 0.0005348394042812288 2023-01-23 15:43:16.122261: step: 1158/1541, loss: 0.0017818229971453547 2023-01-23 15:43:17.066981: step: 1160/1541, loss: 0.05093904957175255 2023-01-23 15:43:17.970980: step: 1162/1541, loss: 0.0288010872900486 2023-01-23 15:43:18.869926: step: 1164/1541, loss: 0.0002610159863252193 2023-01-23 15:43:19.805440: step: 1166/1541, loss: 0.3951321542263031 2023-01-23 15:43:20.696756: step: 1168/1541, loss: 0.02313767373561859 2023-01-23 15:43:21.639027: step: 1170/1541, loss: 0.04208754003047943 2023-01-23 15:43:22.522041: step: 1172/1541, loss: 0.02002621814608574 2023-01-23 15:43:23.459180: step: 1174/1541, loss: 0.022451069205999374 2023-01-23 15:43:24.325354: step: 1176/1541, loss: 0.6716209053993225 2023-01-23 15:43:25.328517: step: 1178/1541, loss: 0.04077887162566185 2023-01-23 15:43:26.263037: step: 1180/1541, loss: 0.010634497739374638 2023-01-23 15:43:27.182398: step: 1182/1541, loss: 0.10040684789419174 2023-01-23 15:43:28.101021: step: 1184/1541, loss: 0.000639482750557363 2023-01-23 15:43:29.022614: step: 1186/1541, loss: 0.003979003056883812 2023-01-23 15:43:29.940330: step: 1188/1541, loss: 0.0001165675203083083 2023-01-23 15:43:30.838781: step: 1190/1541, loss: 0.26081258058547974 2023-01-23 15:43:31.726705: step: 1192/1541, loss: 0.023323625326156616 2023-01-23 15:43:32.639145: step: 1194/1541, loss: 0.00036454980727285147 2023-01-23 15:43:33.566733: step: 1196/1541, loss: 0.04654178023338318 2023-01-23 15:43:34.560021: step: 1198/1541, loss: 0.5924275517463684 2023-01-23 15:43:35.527866: step: 1200/1541, loss: 0.573039710521698 2023-01-23 15:43:36.455205: step: 1202/1541, loss: 0.1163698360323906 2023-01-23 15:43:37.342926: step: 1204/1541, loss: 0.20532101392745972 2023-01-23 15:43:38.239503: step: 1206/1541, loss: 0.009480875916779041 2023-01-23 15:43:39.171990: step: 1208/1541, loss: 0.23919571936130524 2023-01-23 15:43:40.094470: step: 1210/1541, loss: 0.002264974405989051 2023-01-23 15:43:40.934776: step: 1212/1541, loss: 1.5223115682601929 2023-01-23 15:43:41.832062: step: 1214/1541, loss: 0.10088317841291428 2023-01-23 15:43:42.804906: step: 1216/1541, loss: 0.07547038048505783 2023-01-23 15:43:43.684806: step: 1218/1541, loss: 0.05449599400162697 2023-01-23 15:43:44.589098: step: 1220/1541, loss: 0.030623823404312134 2023-01-23 15:43:45.582947: step: 1222/1541, loss: 0.11535158008337021 2023-01-23 15:43:46.467186: step: 1224/1541, loss: 0.0698317214846611 2023-01-23 15:43:47.405473: step: 1226/1541, loss: 0.00925469771027565 2023-01-23 15:43:48.297697: step: 1228/1541, loss: 0.009663763456046581 2023-01-23 15:43:49.156600: step: 1230/1541, loss: 0.005604348611086607 2023-01-23 15:43:50.066369: step: 1232/1541, loss: 0.09312209486961365 2023-01-23 15:43:51.070425: step: 1234/1541, loss: 0.003054675180464983 2023-01-23 15:43:52.004123: step: 1236/1541, loss: 0.00142189534381032 2023-01-23 15:43:52.849751: step: 1238/1541, loss: 0.024205263704061508 2023-01-23 15:43:53.817759: step: 1240/1541, loss: 0.008758820593357086 2023-01-23 15:43:54.756638: step: 1242/1541, loss: 0.009321802295744419 2023-01-23 15:43:55.642687: step: 1244/1541, loss: 0.01390277873724699 2023-01-23 15:43:56.496722: step: 1246/1541, loss: 0.0015428795013576746 2023-01-23 15:43:57.388571: step: 1248/1541, loss: 0.0036787025164812803 2023-01-23 15:43:58.292168: step: 1250/1541, loss: 0.13147184252738953 2023-01-23 15:43:59.165047: step: 1252/1541, loss: 0.0007544473046436906 2023-01-23 15:44:00.056385: step: 1254/1541, loss: 0.01711825467646122 2023-01-23 15:44:00.945218: step: 1256/1541, loss: 0.0463712215423584 2023-01-23 15:44:01.892224: step: 1258/1541, loss: 1.212417721748352 2023-01-23 15:44:02.797243: step: 1260/1541, loss: 0.02734544314444065 2023-01-23 15:44:03.655924: step: 1262/1541, loss: 0.07186029106378555 2023-01-23 15:44:04.613513: step: 1264/1541, loss: 0.11037348955869675 2023-01-23 15:44:05.525834: step: 1266/1541, loss: 0.14628444612026215 2023-01-23 15:44:06.400341: step: 1268/1541, loss: 0.0011224697809666395 2023-01-23 15:44:07.440678: step: 1270/1541, loss: 0.40560609102249146 2023-01-23 15:44:08.317508: step: 1272/1541, loss: 0.018556520342826843 2023-01-23 15:44:09.175022: step: 1274/1541, loss: 0.07795526832342148 2023-01-23 15:44:10.049141: step: 1276/1541, loss: 0.024522986263036728 2023-01-23 15:44:11.009898: step: 1278/1541, loss: 0.019635111093521118 2023-01-23 15:44:11.855740: step: 1280/1541, loss: 0.0007175641949288547 2023-01-23 15:44:12.719526: step: 1282/1541, loss: 0.0010277096880599856 2023-01-23 15:44:13.601379: step: 1284/1541, loss: 0.0004723462334368378 2023-01-23 15:44:14.471995: step: 1286/1541, loss: 0.17639702558517456 2023-01-23 15:44:15.323213: step: 1288/1541, loss: 0.021346010267734528 2023-01-23 15:44:16.191706: step: 1290/1541, loss: 0.03413134068250656 2023-01-23 15:44:17.120111: step: 1292/1541, loss: 0.13022251427173615 2023-01-23 15:44:18.006758: step: 1294/1541, loss: 0.04326188936829567 2023-01-23 15:44:18.871528: step: 1296/1541, loss: 0.05930148437619209 2023-01-23 15:44:19.742303: step: 1298/1541, loss: 0.048003021627664566 2023-01-23 15:44:20.638320: step: 1300/1541, loss: 0.18151172995567322 2023-01-23 15:44:21.537802: step: 1302/1541, loss: 0.00019435407011769712 2023-01-23 15:44:22.462803: step: 1304/1541, loss: 0.05741608515381813 2023-01-23 15:44:23.396641: step: 1306/1541, loss: 0.008112023584544659 2023-01-23 15:44:24.304316: step: 1308/1541, loss: 0.2584068477153778 2023-01-23 15:44:25.178654: step: 1310/1541, loss: 0.022168898954987526 2023-01-23 15:44:26.084592: step: 1312/1541, loss: 0.011885245330631733 2023-01-23 15:44:26.971605: step: 1314/1541, loss: 0.004924667067825794 2023-01-23 15:44:27.859089: step: 1316/1541, loss: 0.024349885061383247 2023-01-23 15:44:28.714999: step: 1318/1541, loss: 0.0014414319302886724 2023-01-23 15:44:29.618865: step: 1320/1541, loss: 0.20224624872207642 2023-01-23 15:44:30.564456: step: 1322/1541, loss: 0.08491571992635727 2023-01-23 15:44:31.457513: step: 1324/1541, loss: 0.3809888958930969 2023-01-23 15:44:32.389778: step: 1326/1541, loss: 0.12877382338047028 2023-01-23 15:44:33.267015: step: 1328/1541, loss: 0.0022381218150258064 2023-01-23 15:44:34.178089: step: 1330/1541, loss: 0.01735960878431797 2023-01-23 15:44:35.076924: step: 1332/1541, loss: 0.0013892919523641467 2023-01-23 15:44:35.962777: step: 1334/1541, loss: 0.0020972508937120438 2023-01-23 15:44:36.862270: step: 1336/1541, loss: 0.01004454493522644 2023-01-23 15:44:37.839989: step: 1338/1541, loss: 0.05627838149666786 2023-01-23 15:44:38.723740: step: 1340/1541, loss: 0.0001788049121387303 2023-01-23 15:44:39.619463: step: 1342/1541, loss: 0.11290786415338516 2023-01-23 15:44:40.540455: step: 1344/1541, loss: 0.036377914249897 2023-01-23 15:44:41.412551: step: 1346/1541, loss: 0.017985431477427483 2023-01-23 15:44:42.347779: step: 1348/1541, loss: 2.1118903160095215 2023-01-23 15:44:43.249863: step: 1350/1541, loss: 0.024062704294919968 2023-01-23 15:44:44.154778: step: 1352/1541, loss: 0.11295374482870102 2023-01-23 15:44:45.029163: step: 1354/1541, loss: 0.03428300842642784 2023-01-23 15:44:45.894938: step: 1356/1541, loss: 0.0012737475335597992 2023-01-23 15:44:46.756821: step: 1358/1541, loss: 0.005716381594538689 2023-01-23 15:44:47.663781: step: 1360/1541, loss: 0.7228692770004272 2023-01-23 15:44:48.575416: step: 1362/1541, loss: 0.3517131507396698 2023-01-23 15:44:49.475050: step: 1364/1541, loss: 0.001072004553861916 2023-01-23 15:44:50.382524: step: 1366/1541, loss: 0.011716279201209545 2023-01-23 15:44:51.257518: step: 1368/1541, loss: 0.00027890419005416334 2023-01-23 15:44:52.146089: step: 1370/1541, loss: 0.08456285297870636 2023-01-23 15:44:53.044887: step: 1372/1541, loss: 0.04696643725037575 2023-01-23 15:44:53.983974: step: 1374/1541, loss: 0.18665826320648193 2023-01-23 15:44:54.884622: step: 1376/1541, loss: 0.0022681120317429304 2023-01-23 15:44:55.785303: step: 1378/1541, loss: 0.2724111080169678 2023-01-23 15:44:56.653426: step: 1380/1541, loss: 0.005365685559809208 2023-01-23 15:44:57.581105: step: 1382/1541, loss: 0.42937079071998596 2023-01-23 15:44:58.469044: step: 1384/1541, loss: 0.09384635835886002 2023-01-23 15:44:59.340479: step: 1386/1541, loss: 0.28370627760887146 2023-01-23 15:45:00.225432: step: 1388/1541, loss: 0.3321060836315155 2023-01-23 15:45:01.144901: step: 1390/1541, loss: 0.005171791184693575 2023-01-23 15:45:02.096813: step: 1392/1541, loss: 0.01822707988321781 2023-01-23 15:45:02.971347: step: 1394/1541, loss: 0.05161063000559807 2023-01-23 15:45:03.883551: step: 1396/1541, loss: 0.03968049958348274 2023-01-23 15:45:04.802079: step: 1398/1541, loss: 0.0010348018258810043 2023-01-23 15:45:05.718971: step: 1400/1541, loss: 0.17247222363948822 2023-01-23 15:45:06.619289: step: 1402/1541, loss: 0.07896189391613007 2023-01-23 15:45:07.538613: step: 1404/1541, loss: 0.08531475812196732 2023-01-23 15:45:08.419475: step: 1406/1541, loss: 0.017925018444657326 2023-01-23 15:45:09.279719: step: 1408/1541, loss: 0.18828308582305908 2023-01-23 15:45:10.187201: step: 1410/1541, loss: 0.009193815290927887 2023-01-23 15:45:11.059678: step: 1412/1541, loss: 0.1453256905078888 2023-01-23 15:45:11.958572: step: 1414/1541, loss: 0.03308044373989105 2023-01-23 15:45:12.872372: step: 1416/1541, loss: 0.10790631175041199 2023-01-23 15:45:13.776537: step: 1418/1541, loss: 0.25433290004730225 2023-01-23 15:45:14.675343: step: 1420/1541, loss: 0.03814521059393883 2023-01-23 15:45:15.600846: step: 1422/1541, loss: 0.004881179425865412 2023-01-23 15:45:16.480067: step: 1424/1541, loss: 0.004926749970763922 2023-01-23 15:45:17.340279: step: 1426/1541, loss: 0.3430553078651428 2023-01-23 15:45:18.270639: step: 1428/1541, loss: 0.06126900017261505 2023-01-23 15:45:19.195842: step: 1430/1541, loss: 0.005308804102241993 2023-01-23 15:45:20.078024: step: 1432/1541, loss: 0.0022736755199730396 2023-01-23 15:45:20.995810: step: 1434/1541, loss: 0.05016796290874481 2023-01-23 15:45:21.859752: step: 1436/1541, loss: 0.28356173634529114 2023-01-23 15:45:22.778451: step: 1438/1541, loss: 0.07843242585659027 2023-01-23 15:45:23.660387: step: 1440/1541, loss: 0.010670153424143791 2023-01-23 15:45:24.541741: step: 1442/1541, loss: 0.005908952094614506 2023-01-23 15:45:25.414781: step: 1444/1541, loss: 0.017758894711732864 2023-01-23 15:45:26.298788: step: 1446/1541, loss: 0.2843528687953949 2023-01-23 15:45:27.185492: step: 1448/1541, loss: 0.0057846796698868275 2023-01-23 15:45:28.123687: step: 1450/1541, loss: 0.042541272938251495 2023-01-23 15:45:28.990385: step: 1452/1541, loss: 0.0011340915225446224 2023-01-23 15:45:29.925062: step: 1454/1541, loss: 0.24375133216381073 2023-01-23 15:45:30.823325: step: 1456/1541, loss: 0.03885091468691826 2023-01-23 15:45:31.714890: step: 1458/1541, loss: 0.03235882148146629 2023-01-23 15:45:32.681312: step: 1460/1541, loss: 0.005920750554651022 2023-01-23 15:45:33.527823: step: 1462/1541, loss: 0.06469563394784927 2023-01-23 15:45:34.475056: step: 1464/1541, loss: 0.0012130026007071137 2023-01-23 15:45:35.370473: step: 1466/1541, loss: 0.0035063307732343674 2023-01-23 15:45:36.269090: step: 1468/1541, loss: 0.042980119585990906 2023-01-23 15:45:37.184707: step: 1470/1541, loss: 0.44148385524749756 2023-01-23 15:45:38.092007: step: 1472/1541, loss: 0.004443781916052103 2023-01-23 15:45:38.966780: step: 1474/1541, loss: 0.0023690643720328808 2023-01-23 15:45:39.843177: step: 1476/1541, loss: 0.007921770215034485 2023-01-23 15:45:40.756281: step: 1478/1541, loss: 0.7040773034095764 2023-01-23 15:45:41.673577: step: 1480/1541, loss: 0.00592572009190917 2023-01-23 15:45:42.573610: step: 1482/1541, loss: 0.004800573922693729 2023-01-23 15:45:43.563004: step: 1484/1541, loss: 0.008383544161915779 2023-01-23 15:45:44.502749: step: 1486/1541, loss: 0.004854394122958183 2023-01-23 15:45:45.347825: step: 1488/1541, loss: 0.0013216156512498856 2023-01-23 15:45:46.239838: step: 1490/1541, loss: 0.025606608018279076 2023-01-23 15:45:47.214414: step: 1492/1541, loss: 0.01747247762978077 2023-01-23 15:45:48.116119: step: 1494/1541, loss: 0.006658685393631458 2023-01-23 15:45:48.999013: step: 1496/1541, loss: 0.0006264441180974245 2023-01-23 15:45:49.876156: step: 1498/1541, loss: 0.10829613357782364 2023-01-23 15:45:50.760795: step: 1500/1541, loss: 0.000686293700709939 2023-01-23 15:45:51.658458: step: 1502/1541, loss: 0.003875536611303687 2023-01-23 15:45:52.583154: step: 1504/1541, loss: 0.0007056877948343754 2023-01-23 15:45:53.457010: step: 1506/1541, loss: 0.04535554349422455 2023-01-23 15:45:54.354574: step: 1508/1541, loss: 1.0037012100219727 2023-01-23 15:45:55.221294: step: 1510/1541, loss: 0.009788114577531815 2023-01-23 15:45:56.103469: step: 1512/1541, loss: 0.13714949786663055 2023-01-23 15:45:56.998626: step: 1514/1541, loss: 0.15658608078956604 2023-01-23 15:45:57.863199: step: 1516/1541, loss: 0.2441788911819458 2023-01-23 15:45:58.747304: step: 1518/1541, loss: 0.791948676109314 2023-01-23 15:45:59.619717: step: 1520/1541, loss: 0.01175811979919672 2023-01-23 15:46:00.471523: step: 1522/1541, loss: 0.10170269757509232 2023-01-23 15:46:01.371305: step: 1524/1541, loss: 0.010137522593140602 2023-01-23 15:46:02.273858: step: 1526/1541, loss: 0.021408196538686752 2023-01-23 15:46:03.149875: step: 1528/1541, loss: 0.0401889905333519 2023-01-23 15:46:04.071003: step: 1530/1541, loss: 0.2546449899673462 2023-01-23 15:46:04.946676: step: 1532/1541, loss: 0.0027453801594674587 2023-01-23 15:46:05.820945: step: 1534/1541, loss: 0.012400769628584385 2023-01-23 15:46:06.755028: step: 1536/1541, loss: 0.05311654135584831 2023-01-23 15:46:07.667847: step: 1538/1541, loss: 0.03871896117925644 2023-01-23 15:46:08.579691: step: 1540/1541, loss: 0.44135645031929016 2023-01-23 15:46:09.463323: step: 1542/1541, loss: 0.008729072287678719 2023-01-23 15:46:10.524976: step: 1544/1541, loss: 0.14069052040576935 2023-01-23 15:46:11.491528: step: 1546/1541, loss: 0.01115637831389904 2023-01-23 15:46:12.378522: step: 1548/1541, loss: 0.01505152229219675 2023-01-23 15:46:13.271574: step: 1550/1541, loss: 0.1649997979402542 2023-01-23 15:46:14.236562: step: 1552/1541, loss: 0.01213043462485075 2023-01-23 15:46:15.119678: step: 1554/1541, loss: 0.02222074568271637 2023-01-23 15:46:16.009756: step: 1556/1541, loss: 0.04864944890141487 2023-01-23 15:46:16.955859: step: 1558/1541, loss: 0.003682895563542843 2023-01-23 15:46:17.948976: step: 1560/1541, loss: 0.011601632460951805 2023-01-23 15:46:18.840074: step: 1562/1541, loss: 0.0006016689585521817 2023-01-23 15:46:19.727496: step: 1564/1541, loss: 0.010930400341749191 2023-01-23 15:46:20.665624: step: 1566/1541, loss: 0.17589907348155975 2023-01-23 15:46:21.505990: step: 1568/1541, loss: 0.012859064154326916 2023-01-23 15:46:22.416908: step: 1570/1541, loss: 0.0453762523829937 2023-01-23 15:46:23.296364: step: 1572/1541, loss: 0.04004824161529541 2023-01-23 15:46:24.194409: step: 1574/1541, loss: 0.00870155356824398 2023-01-23 15:46:25.066641: step: 1576/1541, loss: 0.006552770268172026 2023-01-23 15:46:25.954126: step: 1578/1541, loss: 0.02963622286915779 2023-01-23 15:46:26.817345: step: 1580/1541, loss: 0.0031920569017529488 2023-01-23 15:46:27.709861: step: 1582/1541, loss: 0.052309174090623856 2023-01-23 15:46:28.622225: step: 1584/1541, loss: 0.2714037299156189 2023-01-23 15:46:29.521931: step: 1586/1541, loss: 0.22193512320518494 2023-01-23 15:46:30.417908: step: 1588/1541, loss: 0.01955438405275345 2023-01-23 15:46:31.307777: step: 1590/1541, loss: 0.050087474286556244 2023-01-23 15:46:32.241094: step: 1592/1541, loss: 0.04318393021821976 2023-01-23 15:46:33.188260: step: 1594/1541, loss: 0.1585947871208191 2023-01-23 15:46:34.076764: step: 1596/1541, loss: 0.0009089356753975153 2023-01-23 15:46:34.949738: step: 1598/1541, loss: 0.017672231420874596 2023-01-23 15:46:35.836471: step: 1600/1541, loss: 0.5448703765869141 2023-01-23 15:46:36.772456: step: 1602/1541, loss: 0.03248561918735504 2023-01-23 15:46:37.854194: step: 1604/1541, loss: 0.0931839570403099 2023-01-23 15:46:38.781405: step: 1606/1541, loss: 0.15290988981723785 2023-01-23 15:46:39.704702: step: 1608/1541, loss: 0.052810732275247574 2023-01-23 15:46:40.599353: step: 1610/1541, loss: 0.037673573940992355 2023-01-23 15:46:41.496429: step: 1612/1541, loss: 0.003290504915639758 2023-01-23 15:46:42.407385: step: 1614/1541, loss: 0.035235095769166946 2023-01-23 15:46:43.289124: step: 1616/1541, loss: 0.023969214409589767 2023-01-23 15:46:44.245861: step: 1618/1541, loss: 0.002162995981052518 2023-01-23 15:46:45.131876: step: 1620/1541, loss: 0.16526836156845093 2023-01-23 15:46:46.013309: step: 1622/1541, loss: 0.07677397131919861 2023-01-23 15:46:46.944138: step: 1624/1541, loss: 0.10992484539747238 2023-01-23 15:46:47.830533: step: 1626/1541, loss: 2.3893322944641113 2023-01-23 15:46:48.777511: step: 1628/1541, loss: 0.00862884521484375 2023-01-23 15:46:49.667776: step: 1630/1541, loss: 0.07324153929948807 2023-01-23 15:46:50.578312: step: 1632/1541, loss: 0.0014608395285904408 2023-01-23 15:46:51.439980: step: 1634/1541, loss: 0.08577339351177216 2023-01-23 15:46:52.344961: step: 1636/1541, loss: 0.028262939304113388 2023-01-23 15:46:53.209290: step: 1638/1541, loss: 0.002544563729315996 2023-01-23 15:46:54.122080: step: 1640/1541, loss: 0.0010449890978634357 2023-01-23 15:46:55.078037: step: 1642/1541, loss: 0.0021263521630316973 2023-01-23 15:46:55.944205: step: 1644/1541, loss: 0.0007611542823724449 2023-01-23 15:46:56.830938: step: 1646/1541, loss: 0.16799332201480865 2023-01-23 15:46:57.720619: step: 1648/1541, loss: 0.2420354038476944 2023-01-23 15:46:58.664878: step: 1650/1541, loss: 0.003110390156507492 2023-01-23 15:46:59.596523: step: 1652/1541, loss: 0.0019374516559764743 2023-01-23 15:47:00.503482: step: 1654/1541, loss: 0.2511504292488098 2023-01-23 15:47:01.370649: step: 1656/1541, loss: 0.004062313586473465 2023-01-23 15:47:02.258208: step: 1658/1541, loss: 0.021723566576838493 2023-01-23 15:47:03.143626: step: 1660/1541, loss: 0.30983108282089233 2023-01-23 15:47:04.164409: step: 1662/1541, loss: 0.00034050867543555796 2023-01-23 15:47:05.023046: step: 1664/1541, loss: 0.01946175843477249 2023-01-23 15:47:05.911936: step: 1666/1541, loss: 0.020308680832386017 2023-01-23 15:47:06.811661: step: 1668/1541, loss: 0.009676498360931873 2023-01-23 15:47:07.714447: step: 1670/1541, loss: 0.011988803744316101 2023-01-23 15:47:08.694884: step: 1672/1541, loss: 0.0008833120227791369 2023-01-23 15:47:09.589885: step: 1674/1541, loss: 0.0004228639299981296 2023-01-23 15:47:10.478760: step: 1676/1541, loss: 0.02929738350212574 2023-01-23 15:47:11.423574: step: 1678/1541, loss: 0.000466616649646312 2023-01-23 15:47:12.290250: step: 1680/1541, loss: 0.051257550716400146 2023-01-23 15:47:13.184834: step: 1682/1541, loss: 0.5887345671653748 2023-01-23 15:47:14.069810: step: 1684/1541, loss: 0.00737261027097702 2023-01-23 15:47:14.940572: step: 1686/1541, loss: 0.025635678321123123 2023-01-23 15:47:15.844469: step: 1688/1541, loss: 0.017164401710033417 2023-01-23 15:47:16.746432: step: 1690/1541, loss: 0.10350305587053299 2023-01-23 15:47:17.622776: step: 1692/1541, loss: 0.13970424234867096 2023-01-23 15:47:18.528244: step: 1694/1541, loss: 0.048461925238370895 2023-01-23 15:47:19.399865: step: 1696/1541, loss: 0.0013761069858446717 2023-01-23 15:47:20.281054: step: 1698/1541, loss: 0.017733527347445488 2023-01-23 15:47:21.183091: step: 1700/1541, loss: 0.006936816964298487 2023-01-23 15:47:22.055634: step: 1702/1541, loss: 0.24576488137245178 2023-01-23 15:47:22.922753: step: 1704/1541, loss: 0.012957115657627583 2023-01-23 15:47:23.789029: step: 1706/1541, loss: 0.06478933244943619 2023-01-23 15:47:24.677847: step: 1708/1541, loss: 0.30475637316703796 2023-01-23 15:47:25.539436: step: 1710/1541, loss: 0.00021285090770106763 2023-01-23 15:47:26.448911: step: 1712/1541, loss: 0.22074876725673676 2023-01-23 15:47:27.305559: step: 1714/1541, loss: 0.03000878170132637 2023-01-23 15:47:28.209096: step: 1716/1541, loss: 0.09467899799346924 2023-01-23 15:47:29.099159: step: 1718/1541, loss: 0.00017588249465916306 2023-01-23 15:47:29.963327: step: 1720/1541, loss: 0.0018827083986252546 2023-01-23 15:47:30.856374: step: 1722/1541, loss: 0.0010640786495059729 2023-01-23 15:47:31.757927: step: 1724/1541, loss: 0.0760561004281044 2023-01-23 15:47:32.677853: step: 1726/1541, loss: 0.005170438904315233 2023-01-23 15:47:33.606597: step: 1728/1541, loss: 0.22635453939437866 2023-01-23 15:47:34.497022: step: 1730/1541, loss: 0.0020000089425593615 2023-01-23 15:47:35.403391: step: 1732/1541, loss: 0.21575087308883667 2023-01-23 15:47:36.326455: step: 1734/1541, loss: 0.04907453432679176 2023-01-23 15:47:37.241023: step: 1736/1541, loss: 0.010568779893219471 2023-01-23 15:47:38.160440: step: 1738/1541, loss: 0.042898423969745636 2023-01-23 15:47:39.073395: step: 1740/1541, loss: 0.16275207698345184 2023-01-23 15:47:39.957394: step: 1742/1541, loss: 0.0028734079096466303 2023-01-23 15:47:41.010989: step: 1744/1541, loss: 0.002356881508603692 2023-01-23 15:47:41.885816: step: 1746/1541, loss: 0.030808957293629646 2023-01-23 15:47:42.758743: step: 1748/1541, loss: 0.1980571746826172 2023-01-23 15:47:43.649875: step: 1750/1541, loss: 0.04157045856118202 2023-01-23 15:47:44.539655: step: 1752/1541, loss: 0.01013915240764618 2023-01-23 15:47:45.370646: step: 1754/1541, loss: 0.01557622104883194 2023-01-23 15:47:46.240553: step: 1756/1541, loss: 0.0012496976414695382 2023-01-23 15:47:47.046671: step: 1758/1541, loss: 0.022503623738884926 2023-01-23 15:47:47.973753: step: 1760/1541, loss: 0.11847854405641556 2023-01-23 15:47:48.910759: step: 1762/1541, loss: 0.000994896050542593 2023-01-23 15:47:49.809326: step: 1764/1541, loss: 0.7209500074386597 2023-01-23 15:47:50.691239: step: 1766/1541, loss: 0.22645428776741028 2023-01-23 15:47:51.681096: step: 1768/1541, loss: 0.0021192452404648066 2023-01-23 15:47:52.554091: step: 1770/1541, loss: 0.0002601894666440785 2023-01-23 15:47:53.437766: step: 1772/1541, loss: 0.01065836288034916 2023-01-23 15:47:54.396598: step: 1774/1541, loss: 0.1903243064880371 2023-01-23 15:47:55.286065: step: 1776/1541, loss: 0.4460567235946655 2023-01-23 15:47:56.259737: step: 1778/1541, loss: 0.12118519097566605 2023-01-23 15:47:57.213829: step: 1780/1541, loss: 0.0016672280617058277 2023-01-23 15:47:58.094891: step: 1782/1541, loss: 0.08748739957809448 2023-01-23 15:47:59.003908: step: 1784/1541, loss: 0.0009838297264650464 2023-01-23 15:47:59.890170: step: 1786/1541, loss: 0.0043372102081775665 2023-01-23 15:48:00.807212: step: 1788/1541, loss: 0.5342979431152344 2023-01-23 15:48:01.753870: step: 1790/1541, loss: 0.10459505021572113 2023-01-23 15:48:02.615598: step: 1792/1541, loss: 0.02804679051041603 2023-01-23 15:48:03.520966: step: 1794/1541, loss: 0.011242983862757683 2023-01-23 15:48:04.391802: step: 1796/1541, loss: 0.0025874467100948095 2023-01-23 15:48:05.268366: step: 1798/1541, loss: 0.0007818539161235094 2023-01-23 15:48:06.125953: step: 1800/1541, loss: 0.003626100718975067 2023-01-23 15:48:07.034390: step: 1802/1541, loss: 0.0016117526683956385 2023-01-23 15:48:07.931822: step: 1804/1541, loss: 0.010923887602984905 2023-01-23 15:48:08.859836: step: 1806/1541, loss: 0.001015960588119924 2023-01-23 15:48:09.745834: step: 1808/1541, loss: 0.3052751421928406 2023-01-23 15:48:10.634748: step: 1810/1541, loss: 0.024860583245754242 2023-01-23 15:48:11.488351: step: 1812/1541, loss: 0.012618236243724823 2023-01-23 15:48:12.445144: step: 1814/1541, loss: 0.00036938287666998804 2023-01-23 15:48:13.307405: step: 1816/1541, loss: 0.001711311750113964 2023-01-23 15:48:14.207058: step: 1818/1541, loss: 0.04745325818657875 2023-01-23 15:48:15.062164: step: 1820/1541, loss: 0.004108648281544447 2023-01-23 15:48:15.951810: step: 1822/1541, loss: 0.00840333104133606 2023-01-23 15:48:16.886204: step: 1824/1541, loss: 0.0833311602473259 2023-01-23 15:48:17.791170: step: 1826/1541, loss: 0.027046170085668564 2023-01-23 15:48:18.661340: step: 1828/1541, loss: 0.0034607385750859976 2023-01-23 15:48:19.572288: step: 1830/1541, loss: 0.11957483738660812 2023-01-23 15:48:20.471349: step: 1832/1541, loss: 0.7103984355926514 2023-01-23 15:48:21.360216: step: 1834/1541, loss: 0.021941913291811943 2023-01-23 15:48:22.285192: step: 1836/1541, loss: 0.0007900748169049621 2023-01-23 15:48:23.176960: step: 1838/1541, loss: 0.2969145178794861 2023-01-23 15:48:24.067302: step: 1840/1541, loss: 0.0019991204608231783 2023-01-23 15:48:24.956491: step: 1842/1541, loss: 0.01707502454519272 2023-01-23 15:48:25.850032: step: 1844/1541, loss: 0.039462488144636154 2023-01-23 15:48:26.763918: step: 1846/1541, loss: 0.002301248488947749 2023-01-23 15:48:27.674911: step: 1848/1541, loss: 0.0444544292986393 2023-01-23 15:48:28.634240: step: 1850/1541, loss: 0.0910274088382721 2023-01-23 15:48:29.539805: step: 1852/1541, loss: 0.00010541669325903058 2023-01-23 15:48:30.435651: step: 1854/1541, loss: 0.018439266830682755 2023-01-23 15:48:31.327468: step: 1856/1541, loss: 0.1696324497461319 2023-01-23 15:48:32.243268: step: 1858/1541, loss: 0.0045213112607598305 2023-01-23 15:48:33.146988: step: 1860/1541, loss: 0.6085743308067322 2023-01-23 15:48:34.044747: step: 1862/1541, loss: 0.004467570688575506 2023-01-23 15:48:34.933646: step: 1864/1541, loss: 0.008604845963418484 2023-01-23 15:48:35.854802: step: 1866/1541, loss: 0.019412437453866005 2023-01-23 15:48:36.748818: step: 1868/1541, loss: 0.04335068166255951 2023-01-23 15:48:37.628723: step: 1870/1541, loss: 0.24478647112846375 2023-01-23 15:48:38.529564: step: 1872/1541, loss: 0.06348875910043716 2023-01-23 15:48:39.531473: step: 1874/1541, loss: 0.11072982847690582 2023-01-23 15:48:40.402661: step: 1876/1541, loss: 0.22060231864452362 2023-01-23 15:48:41.267795: step: 1878/1541, loss: 0.0003698430664371699 2023-01-23 15:48:42.214460: step: 1880/1541, loss: 0.003199396887794137 2023-01-23 15:48:43.094188: step: 1882/1541, loss: 0.05507997050881386 2023-01-23 15:48:43.950875: step: 1884/1541, loss: 0.001258688629604876 2023-01-23 15:48:44.856262: step: 1886/1541, loss: 0.035251107066869736 2023-01-23 15:48:45.744244: step: 1888/1541, loss: 0.042863231152296066 2023-01-23 15:48:46.630210: step: 1890/1541, loss: 0.27698394656181335 2023-01-23 15:48:47.533827: step: 1892/1541, loss: 0.00041921419324353337 2023-01-23 15:48:48.431132: step: 1894/1541, loss: 0.022428268566727638 2023-01-23 15:48:49.303503: step: 1896/1541, loss: 0.0334848053753376 2023-01-23 15:48:50.193626: step: 1898/1541, loss: 0.02416989393532276 2023-01-23 15:48:51.080739: step: 1900/1541, loss: 0.033510684967041016 2023-01-23 15:48:51.969510: step: 1902/1541, loss: 0.0008570127538405359 2023-01-23 15:48:52.828538: step: 1904/1541, loss: 0.12055718153715134 2023-01-23 15:48:53.775303: step: 1906/1541, loss: 0.0028935163281857967 2023-01-23 15:48:54.682565: step: 1908/1541, loss: 0.004674217198044062 2023-01-23 15:48:55.575126: step: 1910/1541, loss: 0.005140335764735937 2023-01-23 15:48:56.503821: step: 1912/1541, loss: 0.17849697172641754 2023-01-23 15:48:57.392294: step: 1914/1541, loss: 0.08322296291589737 2023-01-23 15:48:58.332352: step: 1916/1541, loss: 0.06020331382751465 2023-01-23 15:48:59.225988: step: 1918/1541, loss: 0.00178127724211663 2023-01-23 15:49:00.158789: step: 1920/1541, loss: 0.0008901756955310702 2023-01-23 15:49:01.065848: step: 1922/1541, loss: 0.04405482858419418 2023-01-23 15:49:01.985296: step: 1924/1541, loss: 0.0006174906739033759 2023-01-23 15:49:02.859928: step: 1926/1541, loss: 0.07478775829076767 2023-01-23 15:49:03.752748: step: 1928/1541, loss: 0.23277492821216583 2023-01-23 15:49:04.719501: step: 1930/1541, loss: 0.057982224971055984 2023-01-23 15:49:05.734920: step: 1932/1541, loss: 0.08809289336204529 2023-01-23 15:49:06.666901: step: 1934/1541, loss: 0.009799678809940815 2023-01-23 15:49:07.549935: step: 1936/1541, loss: 0.12498213350772858 2023-01-23 15:49:08.459192: step: 1938/1541, loss: 0.01097975205630064 2023-01-23 15:49:09.383972: step: 1940/1541, loss: 0.09995580464601517 2023-01-23 15:49:10.260476: step: 1942/1541, loss: 0.17953838407993317 2023-01-23 15:49:11.108324: step: 1944/1541, loss: 0.00031725363805890083 2023-01-23 15:49:12.052730: step: 1946/1541, loss: 0.04411599785089493 2023-01-23 15:49:12.915212: step: 1948/1541, loss: 0.15464670956134796 2023-01-23 15:49:13.788198: step: 1950/1541, loss: 0.0007151654572226107 2023-01-23 15:49:14.722398: step: 1952/1541, loss: 0.056456100195646286 2023-01-23 15:49:15.633207: step: 1954/1541, loss: 0.34957659244537354 2023-01-23 15:49:16.508494: step: 1956/1541, loss: 0.310499370098114 2023-01-23 15:49:17.401090: step: 1958/1541, loss: 0.00035573530476540327 2023-01-23 15:49:18.276478: step: 1960/1541, loss: 0.12318045645952225 2023-01-23 15:49:19.183506: step: 1962/1541, loss: 0.13065758347511292 2023-01-23 15:49:20.097596: step: 1964/1541, loss: 0.00803209189325571 2023-01-23 15:49:21.028713: step: 1966/1541, loss: 0.4560115337371826 2023-01-23 15:49:21.974059: step: 1968/1541, loss: 0.023641139268875122 2023-01-23 15:49:22.877474: step: 1970/1541, loss: 0.017471808940172195 2023-01-23 15:49:23.857170: step: 1972/1541, loss: 0.04161315783858299 2023-01-23 15:49:24.746607: step: 1974/1541, loss: 0.1443011313676834 2023-01-23 15:49:25.625194: step: 1976/1541, loss: 0.004946105182170868 2023-01-23 15:49:26.538472: step: 1978/1541, loss: 0.017066044732928276 2023-01-23 15:49:27.431738: step: 1980/1541, loss: 0.024279894307255745 2023-01-23 15:49:28.289432: step: 1982/1541, loss: 0.006811853963881731 2023-01-23 15:49:29.214648: step: 1984/1541, loss: 0.014023881405591965 2023-01-23 15:49:30.089119: step: 1986/1541, loss: 0.016815124079585075 2023-01-23 15:49:31.002457: step: 1988/1541, loss: 0.022486353293061256 2023-01-23 15:49:31.942217: step: 1990/1541, loss: 0.13306133449077606 2023-01-23 15:49:32.870175: step: 1992/1541, loss: 0.06820578128099442 2023-01-23 15:49:33.822211: step: 1994/1541, loss: 0.003288534004241228 2023-01-23 15:49:34.772547: step: 1996/1541, loss: 0.005578712094575167 2023-01-23 15:49:35.633759: step: 1998/1541, loss: 0.07351943850517273 2023-01-23 15:49:36.539196: step: 2000/1541, loss: 0.14162926375865936 2023-01-23 15:49:37.409507: step: 2002/1541, loss: 0.008509388193488121 2023-01-23 15:49:38.257813: step: 2004/1541, loss: 0.0010817254660651088 2023-01-23 15:49:39.227983: step: 2006/1541, loss: 0.222983717918396 2023-01-23 15:49:40.106327: step: 2008/1541, loss: 0.10744214057922363 2023-01-23 15:49:40.962253: step: 2010/1541, loss: 0.13395537436008453 2023-01-23 15:49:41.866382: step: 2012/1541, loss: 0.0003070459351874888 2023-01-23 15:49:42.769525: step: 2014/1541, loss: 0.04364448040723801 2023-01-23 15:49:43.632608: step: 2016/1541, loss: 0.18703734874725342 2023-01-23 15:49:44.516415: step: 2018/1541, loss: 0.008205662481486797 2023-01-23 15:49:45.363141: step: 2020/1541, loss: 0.0043577346950769424 2023-01-23 15:49:46.247043: step: 2022/1541, loss: 0.0018015409586951137 2023-01-23 15:49:47.124351: step: 2024/1541, loss: 0.23608754575252533 2023-01-23 15:49:48.021212: step: 2026/1541, loss: 0.004242768511176109 2023-01-23 15:49:48.892335: step: 2028/1541, loss: 0.13133910298347473 2023-01-23 15:49:49.795929: step: 2030/1541, loss: 0.23986946046352386 2023-01-23 15:49:50.751983: step: 2032/1541, loss: 0.1024564653635025 2023-01-23 15:49:51.630223: step: 2034/1541, loss: 0.02892334945499897 2023-01-23 15:49:52.540251: step: 2036/1541, loss: 0.006641168147325516 2023-01-23 15:49:53.437958: step: 2038/1541, loss: 0.07146058231592178 2023-01-23 15:49:54.368034: step: 2040/1541, loss: 0.02043626457452774 2023-01-23 15:49:55.236364: step: 2042/1541, loss: 0.0063544330187141895 2023-01-23 15:49:56.166297: step: 2044/1541, loss: 0.3685294985771179 2023-01-23 15:49:57.094781: step: 2046/1541, loss: 0.05837901309132576 2023-01-23 15:49:58.005468: step: 2048/1541, loss: 0.003718572435900569 2023-01-23 15:49:58.873616: step: 2050/1541, loss: 0.053010258823633194 2023-01-23 15:49:59.772515: step: 2052/1541, loss: 0.014426114037632942 2023-01-23 15:50:00.722119: step: 2054/1541, loss: 0.30193856358528137 2023-01-23 15:50:01.607152: step: 2056/1541, loss: 0.0038982387632131577 2023-01-23 15:50:02.534042: step: 2058/1541, loss: 0.3074187636375427 2023-01-23 15:50:03.429429: step: 2060/1541, loss: 0.08717895299196243 2023-01-23 15:50:04.339829: step: 2062/1541, loss: 0.018884046003222466 2023-01-23 15:50:05.269167: step: 2064/1541, loss: 0.005894866306334734 2023-01-23 15:50:06.139602: step: 2066/1541, loss: 0.02796364016830921 2023-01-23 15:50:07.028230: step: 2068/1541, loss: 0.0071340217255055904 2023-01-23 15:50:07.959859: step: 2070/1541, loss: 0.018375856801867485 2023-01-23 15:50:08.828234: step: 2072/1541, loss: 0.12574489414691925 2023-01-23 15:50:09.692123: step: 2074/1541, loss: 0.009724128060042858 2023-01-23 15:50:10.659008: step: 2076/1541, loss: 0.008690432645380497 2023-01-23 15:50:11.552879: step: 2078/1541, loss: 0.1305408626794815 2023-01-23 15:50:12.487746: step: 2080/1541, loss: 0.0051153781823813915 2023-01-23 15:50:13.406437: step: 2082/1541, loss: 0.0010831643594428897 2023-01-23 15:50:14.303336: step: 2084/1541, loss: 0.06785976141691208 2023-01-23 15:50:15.200444: step: 2086/1541, loss: 0.016721338033676147 2023-01-23 15:50:16.084509: step: 2088/1541, loss: 0.008860866539180279 2023-01-23 15:50:17.028634: step: 2090/1541, loss: 0.006426963955163956 2023-01-23 15:50:17.917818: step: 2092/1541, loss: 0.02536260522902012 2023-01-23 15:50:18.834622: step: 2094/1541, loss: 0.0716049075126648 2023-01-23 15:50:19.700201: step: 2096/1541, loss: 0.0032530748285353184 2023-01-23 15:50:20.583703: step: 2098/1541, loss: 0.10873515158891678 2023-01-23 15:50:21.470079: step: 2100/1541, loss: 0.22938881814479828 2023-01-23 15:50:22.359015: step: 2102/1541, loss: 0.02534795179963112 2023-01-23 15:50:23.327709: step: 2104/1541, loss: 0.0753784105181694 2023-01-23 15:50:24.242853: step: 2106/1541, loss: 0.0036188671365380287 2023-01-23 15:50:25.182577: step: 2108/1541, loss: 0.14002108573913574 2023-01-23 15:50:26.104259: step: 2110/1541, loss: 0.03173619881272316 2023-01-23 15:50:27.000806: step: 2112/1541, loss: 0.01097092404961586 2023-01-23 15:50:27.930929: step: 2114/1541, loss: 0.0035262631718069315 2023-01-23 15:50:28.817918: step: 2116/1541, loss: 0.00011247789370827377 2023-01-23 15:50:29.708082: step: 2118/1541, loss: 0.016467265784740448 2023-01-23 15:50:30.597283: step: 2120/1541, loss: 0.10418159514665604 2023-01-23 15:50:31.516024: step: 2122/1541, loss: 0.08209037035703659 2023-01-23 15:50:32.411162: step: 2124/1541, loss: 0.28342941403388977 2023-01-23 15:50:33.304833: step: 2126/1541, loss: 0.3232095539569855 2023-01-23 15:50:34.165847: step: 2128/1541, loss: 0.0727570503950119 2023-01-23 15:50:35.085845: step: 2130/1541, loss: 0.055556345731019974 2023-01-23 15:50:35.942969: step: 2132/1541, loss: 0.001376245985738933 2023-01-23 15:50:36.905008: step: 2134/1541, loss: 0.11092222481966019 2023-01-23 15:50:37.790256: step: 2136/1541, loss: 0.0005816790508106351 2023-01-23 15:50:38.669347: step: 2138/1541, loss: 0.008200022391974926 2023-01-23 15:50:39.553007: step: 2140/1541, loss: 0.006709340028464794 2023-01-23 15:50:40.414815: step: 2142/1541, loss: 0.06546778231859207 2023-01-23 15:50:41.319969: step: 2144/1541, loss: 0.02125387266278267 2023-01-23 15:50:42.250100: step: 2146/1541, loss: 0.22055134177207947 2023-01-23 15:50:43.187210: step: 2148/1541, loss: 0.06227228045463562 2023-01-23 15:50:44.099656: step: 2150/1541, loss: 0.022690461948513985 2023-01-23 15:50:44.999149: step: 2152/1541, loss: 0.01325985137373209 2023-01-23 15:50:45.873006: step: 2154/1541, loss: 0.4264940023422241 2023-01-23 15:50:46.772777: step: 2156/1541, loss: 0.07539597898721695 2023-01-23 15:50:47.634219: step: 2158/1541, loss: 0.009233057498931885 2023-01-23 15:50:48.542507: step: 2160/1541, loss: 0.08549073338508606 2023-01-23 15:50:49.428452: step: 2162/1541, loss: 0.005338897462934256 2023-01-23 15:50:50.318998: step: 2164/1541, loss: 0.8808006644248962 2023-01-23 15:50:51.227177: step: 2166/1541, loss: 0.025340108200907707 2023-01-23 15:50:52.093016: step: 2168/1541, loss: 0.13425679504871368 2023-01-23 15:50:52.970149: step: 2170/1541, loss: 0.3406309485435486 2023-01-23 15:50:53.877654: step: 2172/1541, loss: 0.00016991241136565804 2023-01-23 15:50:54.776298: step: 2174/1541, loss: 0.00039703905349597335 2023-01-23 15:50:55.675346: step: 2176/1541, loss: 0.05535120889544487 2023-01-23 15:50:56.607912: step: 2178/1541, loss: 0.023549793288111687 2023-01-23 15:50:57.515082: step: 2180/1541, loss: 0.14899693429470062 2023-01-23 15:50:58.462900: step: 2182/1541, loss: 0.004160136450082064 2023-01-23 15:50:59.361016: step: 2184/1541, loss: 0.010638338513672352 2023-01-23 15:51:00.271846: step: 2186/1541, loss: 0.12354391813278198 2023-01-23 15:51:01.139345: step: 2188/1541, loss: 0.002072103787213564 2023-01-23 15:51:02.065763: step: 2190/1541, loss: 0.1706451028585434 2023-01-23 15:51:02.945533: step: 2192/1541, loss: 0.015638571232557297 2023-01-23 15:51:03.879970: step: 2194/1541, loss: 0.05734306946396828 2023-01-23 15:51:04.765883: step: 2196/1541, loss: 0.24380144476890564 2023-01-23 15:51:05.677521: step: 2198/1541, loss: 0.0035738577134907246 2023-01-23 15:51:06.591477: step: 2200/1541, loss: 0.03409034386277199 2023-01-23 15:51:07.500579: step: 2202/1541, loss: 0.03738036006689072 2023-01-23 15:51:08.406543: step: 2204/1541, loss: 0.010037757456302643 2023-01-23 15:51:09.316465: step: 2206/1541, loss: 0.006541544105857611 2023-01-23 15:51:10.204792: step: 2208/1541, loss: 0.021119220182299614 2023-01-23 15:51:11.112877: step: 2210/1541, loss: 0.04032542183995247 2023-01-23 15:51:11.966650: step: 2212/1541, loss: 0.03080444224178791 2023-01-23 15:51:12.861920: step: 2214/1541, loss: 0.04399150609970093 2023-01-23 15:51:13.887385: step: 2216/1541, loss: 0.003961584530770779 2023-01-23 15:51:14.777140: step: 2218/1541, loss: 0.035527799278497696 2023-01-23 15:51:15.683514: step: 2220/1541, loss: 0.1310289055109024 2023-01-23 15:51:16.569351: step: 2222/1541, loss: 0.034014299511909485 2023-01-23 15:51:17.432617: step: 2224/1541, loss: 0.008363576605916023 2023-01-23 15:51:18.328826: step: 2226/1541, loss: 0.006905396468937397 2023-01-23 15:51:19.229167: step: 2228/1541, loss: 0.007107039913535118 2023-01-23 15:51:20.121734: step: 2230/1541, loss: 0.06546773761510849 2023-01-23 15:51:21.037659: step: 2232/1541, loss: 0.005590985994786024 2023-01-23 15:51:21.918226: step: 2234/1541, loss: 0.019660351797938347 2023-01-23 15:51:22.799398: step: 2236/1541, loss: 0.3892832100391388 2023-01-23 15:51:23.746850: step: 2238/1541, loss: 0.002016963204368949 2023-01-23 15:51:24.656113: step: 2240/1541, loss: 0.0019329090137034655 2023-01-23 15:51:25.561926: step: 2242/1541, loss: 0.004586377181112766 2023-01-23 15:51:26.440680: step: 2244/1541, loss: 0.02055497094988823 2023-01-23 15:51:27.359338: step: 2246/1541, loss: 0.0700063407421112 2023-01-23 15:51:28.266565: step: 2248/1541, loss: 1.246469497680664 2023-01-23 15:51:29.195792: step: 2250/1541, loss: 0.06301432102918625 2023-01-23 15:51:30.145544: step: 2252/1541, loss: 0.07896401733160019 2023-01-23 15:51:31.144695: step: 2254/1541, loss: 0.014718719758093357 2023-01-23 15:51:32.053888: step: 2256/1541, loss: 0.2500407099723816 2023-01-23 15:51:33.013735: step: 2258/1541, loss: 0.022090934216976166 2023-01-23 15:51:33.883023: step: 2260/1541, loss: 0.037029165774583817 2023-01-23 15:51:34.789350: step: 2262/1541, loss: 0.06569015979766846 2023-01-23 15:51:35.691463: step: 2264/1541, loss: 0.010263105854392052 2023-01-23 15:51:36.575966: step: 2266/1541, loss: 7.656206435058266e-05 2023-01-23 15:51:37.470855: step: 2268/1541, loss: 0.0036039899569004774 2023-01-23 15:51:38.375593: step: 2270/1541, loss: 0.016787197440862656 2023-01-23 15:51:39.311535: step: 2272/1541, loss: 0.03175724670290947 2023-01-23 15:51:40.204808: step: 2274/1541, loss: 0.01513952948153019 2023-01-23 15:51:41.121506: step: 2276/1541, loss: 0.002441908698529005 2023-01-23 15:51:42.043133: step: 2278/1541, loss: 0.2972191870212555 2023-01-23 15:51:42.950982: step: 2280/1541, loss: 0.031723640859127045 2023-01-23 15:51:43.821071: step: 2282/1541, loss: 0.0015870912466198206 2023-01-23 15:51:44.700110: step: 2284/1541, loss: 0.03092840686440468 2023-01-23 15:51:45.586866: step: 2286/1541, loss: 0.006540738977491856 2023-01-23 15:51:46.475894: step: 2288/1541, loss: 0.01166154257953167 2023-01-23 15:51:47.388754: step: 2290/1541, loss: 0.0006055033882148564 2023-01-23 15:51:48.364490: step: 2292/1541, loss: 0.004846737254410982 2023-01-23 15:51:49.244502: step: 2294/1541, loss: 0.04721175879240036 2023-01-23 15:51:50.138616: step: 2296/1541, loss: 0.0036449450999498367 2023-01-23 15:51:51.070517: step: 2298/1541, loss: 0.0022214397322386503 2023-01-23 15:51:51.974571: step: 2300/1541, loss: 0.12013213336467743 2023-01-23 15:51:52.850362: step: 2302/1541, loss: 0.08294101059436798 2023-01-23 15:51:53.849641: step: 2304/1541, loss: 0.004646287299692631 2023-01-23 15:51:54.758847: step: 2306/1541, loss: 0.014413110911846161 2023-01-23 15:51:55.644612: step: 2308/1541, loss: 0.001523192971944809 2023-01-23 15:51:56.550290: step: 2310/1541, loss: 0.006150290835648775 2023-01-23 15:51:57.475072: step: 2312/1541, loss: 0.04947208613157272 2023-01-23 15:51:58.430636: step: 2314/1541, loss: 0.02894105389714241 2023-01-23 15:51:59.364179: step: 2316/1541, loss: 0.03734543174505234 2023-01-23 15:52:00.232409: step: 2318/1541, loss: 0.009637598879635334 2023-01-23 15:52:01.140784: step: 2320/1541, loss: 0.1489558070898056 2023-01-23 15:52:02.054115: step: 2322/1541, loss: 0.001630665035918355 2023-01-23 15:52:02.951593: step: 2324/1541, loss: 0.03887038677930832 2023-01-23 15:52:03.873583: step: 2326/1541, loss: 0.2248384952545166 2023-01-23 15:52:04.774380: step: 2328/1541, loss: 0.022132862359285355 2023-01-23 15:52:05.649376: step: 2330/1541, loss: 2.3983713617781177e-05 2023-01-23 15:52:06.560450: step: 2332/1541, loss: 0.2660563588142395 2023-01-23 15:52:07.421923: step: 2334/1541, loss: 0.004917606711387634 2023-01-23 15:52:08.345188: step: 2336/1541, loss: 1.9868813753128052 2023-01-23 15:52:09.210716: step: 2338/1541, loss: 0.004973978269845247 2023-01-23 15:52:10.081214: step: 2340/1541, loss: 0.011041766032576561 2023-01-23 15:52:10.950340: step: 2342/1541, loss: 0.0007812106050550938 2023-01-23 15:52:11.904569: step: 2344/1541, loss: 0.0015523635083809495 2023-01-23 15:52:12.788125: step: 2346/1541, loss: 0.11327838897705078 2023-01-23 15:52:13.683338: step: 2348/1541, loss: 0.06620863825082779 2023-01-23 15:52:14.578988: step: 2350/1541, loss: 0.06185859814286232 2023-01-23 15:52:15.492863: step: 2352/1541, loss: 0.058379482477903366 2023-01-23 15:52:16.433605: step: 2354/1541, loss: 0.03490403667092323 2023-01-23 15:52:17.328674: step: 2356/1541, loss: 0.05124468728899956 2023-01-23 15:52:18.210156: step: 2358/1541, loss: 0.006362622138112783 2023-01-23 15:52:19.106127: step: 2360/1541, loss: 0.00044284193427301943 2023-01-23 15:52:19.997479: step: 2362/1541, loss: 0.008179313503205776 2023-01-23 15:52:20.928175: step: 2364/1541, loss: 0.000716887996532023 2023-01-23 15:52:21.824869: step: 2366/1541, loss: 0.001376248081214726 2023-01-23 15:52:22.717388: step: 2368/1541, loss: 0.09904377907514572 2023-01-23 15:52:23.602894: step: 2370/1541, loss: 0.00441232742741704 2023-01-23 15:52:24.515903: step: 2372/1541, loss: 0.03454246744513512 2023-01-23 15:52:25.447788: step: 2374/1541, loss: 0.000940442958381027 2023-01-23 15:52:26.359278: step: 2376/1541, loss: 0.0004288748314138502 2023-01-23 15:52:27.288235: step: 2378/1541, loss: 0.15212832391262054 2023-01-23 15:52:28.231958: step: 2380/1541, loss: 0.006279177963733673 2023-01-23 15:52:29.092897: step: 2382/1541, loss: 0.0007294082315638661 2023-01-23 15:52:30.046877: step: 2384/1541, loss: 0.08835036307573318 2023-01-23 15:52:30.898185: step: 2386/1541, loss: 0.00864869449287653 2023-01-23 15:52:31.807995: step: 2388/1541, loss: 0.04446897283196449 2023-01-23 15:52:32.760033: step: 2390/1541, loss: 0.0005326283280737698 2023-01-23 15:52:33.802826: step: 2392/1541, loss: 0.5262482762336731 2023-01-23 15:52:34.712608: step: 2394/1541, loss: 0.04526909068226814 2023-01-23 15:52:35.589278: step: 2396/1541, loss: 0.034885913133621216 2023-01-23 15:52:36.475687: step: 2398/1541, loss: 0.2222086489200592 2023-01-23 15:52:37.371996: step: 2400/1541, loss: 0.08975904434919357 2023-01-23 15:52:38.283295: step: 2402/1541, loss: 0.0016906973905861378 2023-01-23 15:52:39.179869: step: 2404/1541, loss: 0.009457274340093136 2023-01-23 15:52:40.074713: step: 2406/1541, loss: 0.2634481191635132 2023-01-23 15:52:40.925294: step: 2408/1541, loss: 0.00014281674521043897 2023-01-23 15:52:41.773786: step: 2410/1541, loss: 0.0033604989293962717 2023-01-23 15:52:42.637685: step: 2412/1541, loss: 0.046939052641391754 2023-01-23 15:52:43.559695: step: 2414/1541, loss: 0.18561838567256927 2023-01-23 15:52:44.537828: step: 2416/1541, loss: 0.007916860282421112 2023-01-23 15:52:45.405186: step: 2418/1541, loss: 0.0046182675287127495 2023-01-23 15:52:46.285405: step: 2420/1541, loss: 0.03261643275618553 2023-01-23 15:52:47.165207: step: 2422/1541, loss: 0.03070688620209694 2023-01-23 15:52:48.036491: step: 2424/1541, loss: 0.1650882214307785 2023-01-23 15:52:48.945912: step: 2426/1541, loss: 0.07908984273672104 2023-01-23 15:52:49.888022: step: 2428/1541, loss: 0.07222473621368408 2023-01-23 15:52:50.776385: step: 2430/1541, loss: 0.11374297738075256 2023-01-23 15:52:51.667609: step: 2432/1541, loss: 0.020271742716431618 2023-01-23 15:52:52.555148: step: 2434/1541, loss: 0.21182340383529663 2023-01-23 15:52:53.418933: step: 2436/1541, loss: 0.00620618648827076 2023-01-23 15:52:54.360021: step: 2438/1541, loss: 0.005203563254326582 2023-01-23 15:52:55.249215: step: 2440/1541, loss: 0.006281261797994375 2023-01-23 15:52:56.113096: step: 2442/1541, loss: 0.3869655132293701 2023-01-23 15:52:56.985170: step: 2444/1541, loss: 0.03342072665691376 2023-01-23 15:52:57.889314: step: 2446/1541, loss: 0.20854200422763824 2023-01-23 15:52:58.769775: step: 2448/1541, loss: 0.006447327323257923 2023-01-23 15:52:59.662899: step: 2450/1541, loss: 0.046766504645347595 2023-01-23 15:53:00.574383: step: 2452/1541, loss: 0.014078046195209026 2023-01-23 15:53:01.464624: step: 2454/1541, loss: 0.0012339394306764007 2023-01-23 15:53:02.351690: step: 2456/1541, loss: 0.01998007297515869 2023-01-23 15:53:03.406811: step: 2458/1541, loss: 0.031968969851732254 2023-01-23 15:53:04.291184: step: 2460/1541, loss: 0.17134888470172882 2023-01-23 15:53:05.190788: step: 2462/1541, loss: 0.5415963530540466 2023-01-23 15:53:06.050595: step: 2464/1541, loss: 0.044260941445827484 2023-01-23 15:53:06.964519: step: 2466/1541, loss: 0.0010598096996545792 2023-01-23 15:53:07.870822: step: 2468/1541, loss: 0.00858884584158659 2023-01-23 15:53:08.804842: step: 2470/1541, loss: 0.0009841832797974348 2023-01-23 15:53:09.744230: step: 2472/1541, loss: 0.03596355393528938 2023-01-23 15:53:10.650828: step: 2474/1541, loss: 0.12032439559698105 2023-01-23 15:53:11.564653: step: 2476/1541, loss: 0.029723884537816048 2023-01-23 15:53:12.470787: step: 2478/1541, loss: 0.47742959856987 2023-01-23 15:53:13.366383: step: 2480/1541, loss: 0.0007812348194420338 2023-01-23 15:53:14.240821: step: 2482/1541, loss: 0.010270928032696247 2023-01-23 15:53:15.141194: step: 2484/1541, loss: 0.008731354027986526 2023-01-23 15:53:16.029366: step: 2486/1541, loss: 0.013143813237547874 2023-01-23 15:53:16.877992: step: 2488/1541, loss: 0.07476963847875595 2023-01-23 15:53:17.741646: step: 2490/1541, loss: 1.238107681274414 2023-01-23 15:53:18.636931: step: 2492/1541, loss: 0.09965267777442932 2023-01-23 15:53:19.530144: step: 2494/1541, loss: 0.0035751303657889366 2023-01-23 15:53:20.429680: step: 2496/1541, loss: 0.00670553557574749 2023-01-23 15:53:21.303080: step: 2498/1541, loss: 0.004366633947938681 2023-01-23 15:53:22.331001: step: 2500/1541, loss: 0.04172741621732712 2023-01-23 15:53:23.203178: step: 2502/1541, loss: 0.07026679813861847 2023-01-23 15:53:24.098514: step: 2504/1541, loss: 0.013811136595904827 2023-01-23 15:53:24.996518: step: 2506/1541, loss: 0.03464687243103981 2023-01-23 15:53:25.877058: step: 2508/1541, loss: 0.003402331378310919 2023-01-23 15:53:26.831658: step: 2510/1541, loss: 0.2619856297969818 2023-01-23 15:53:27.723522: step: 2512/1541, loss: 0.0011498259846121073 2023-01-23 15:53:28.615558: step: 2514/1541, loss: 0.00022337952395901084 2023-01-23 15:53:29.539485: step: 2516/1541, loss: 0.018012741580605507 2023-01-23 15:53:30.423010: step: 2518/1541, loss: 0.0012931206729263067 2023-01-23 15:53:31.328991: step: 2520/1541, loss: 0.33636462688446045 2023-01-23 15:53:32.235421: step: 2522/1541, loss: 0.033375974744558334 2023-01-23 15:53:33.126779: step: 2524/1541, loss: 0.00018052521045319736 2023-01-23 15:53:34.045998: step: 2526/1541, loss: 0.26210692524909973 2023-01-23 15:53:34.908761: step: 2528/1541, loss: 0.38234743475914 2023-01-23 15:53:35.783676: step: 2530/1541, loss: 8.318711479660124e-06 2023-01-23 15:53:36.645199: step: 2532/1541, loss: 0.48750805854797363 2023-01-23 15:53:37.536076: step: 2534/1541, loss: 0.10069787502288818 2023-01-23 15:53:38.486631: step: 2536/1541, loss: 0.003431005170568824 2023-01-23 15:53:39.375756: step: 2538/1541, loss: 0.005121204070746899 2023-01-23 15:53:40.311998: step: 2540/1541, loss: 0.0003813771181739867 2023-01-23 15:53:41.203570: step: 2542/1541, loss: 0.3438476324081421 2023-01-23 15:53:42.094234: step: 2544/1541, loss: 0.18199416995048523 2023-01-23 15:53:43.027325: step: 2546/1541, loss: 0.17313852906227112 2023-01-23 15:53:43.995383: step: 2548/1541, loss: 0.08750051259994507 2023-01-23 15:53:44.867327: step: 2550/1541, loss: 0.051581740379333496 2023-01-23 15:53:45.745624: step: 2552/1541, loss: 0.003670129692181945 2023-01-23 15:53:46.663401: step: 2554/1541, loss: 0.07974046468734741 2023-01-23 15:53:47.616438: step: 2556/1541, loss: 0.32213667035102844 2023-01-23 15:53:48.515062: step: 2558/1541, loss: 0.4406806230545044 2023-01-23 15:53:49.401676: step: 2560/1541, loss: 0.006062396336346865 2023-01-23 15:53:50.339123: step: 2562/1541, loss: 0.4300819933414459 2023-01-23 15:53:51.225482: step: 2564/1541, loss: 0.1497039645910263 2023-01-23 15:53:52.204542: step: 2566/1541, loss: 0.07944603264331818 2023-01-23 15:53:53.080409: step: 2568/1541, loss: 0.28430506587028503 2023-01-23 15:53:53.965586: step: 2570/1541, loss: 0.021292883902788162 2023-01-23 15:53:54.840749: step: 2572/1541, loss: 0.10809705406427383 2023-01-23 15:53:55.728120: step: 2574/1541, loss: 0.006368379108607769 2023-01-23 15:53:56.606841: step: 2576/1541, loss: 0.03416803106665611 2023-01-23 15:53:57.479208: step: 2578/1541, loss: 0.0028346132021397352 2023-01-23 15:53:58.455019: step: 2580/1541, loss: 0.08843833208084106 2023-01-23 15:53:59.451277: step: 2582/1541, loss: 0.036463119089603424 2023-01-23 15:54:00.345329: step: 2584/1541, loss: 0.00012504558253567666 2023-01-23 15:54:01.252383: step: 2586/1541, loss: 0.08888693898916245 2023-01-23 15:54:02.154783: step: 2588/1541, loss: 0.022293396294116974 2023-01-23 15:54:03.031134: step: 2590/1541, loss: 0.1359359622001648 2023-01-23 15:54:03.886641: step: 2592/1541, loss: 0.12179206311702728 2023-01-23 15:54:04.824378: step: 2594/1541, loss: 0.0016689638141542673 2023-01-23 15:54:05.776294: step: 2596/1541, loss: 0.034252870827913284 2023-01-23 15:54:06.703303: step: 2598/1541, loss: 0.04974067956209183 2023-01-23 15:54:07.583851: step: 2600/1541, loss: 0.3963450491428375 2023-01-23 15:54:08.486377: step: 2602/1541, loss: 0.012854964472353458 2023-01-23 15:54:09.354890: step: 2604/1541, loss: 0.05409236252307892 2023-01-23 15:54:10.261150: step: 2606/1541, loss: 0.12419162690639496 2023-01-23 15:54:11.125419: step: 2608/1541, loss: 0.10907435417175293 2023-01-23 15:54:12.011980: step: 2610/1541, loss: 0.02920871414244175 2023-01-23 15:54:12.916652: step: 2612/1541, loss: 0.08694585412740707 2023-01-23 15:54:13.812435: step: 2614/1541, loss: 0.005155364517122507 2023-01-23 15:54:14.683336: step: 2616/1541, loss: 0.18103906512260437 2023-01-23 15:54:15.557111: step: 2618/1541, loss: 0.16466952860355377 2023-01-23 15:54:16.427193: step: 2620/1541, loss: 0.5801730751991272 2023-01-23 15:54:17.321863: step: 2622/1541, loss: 0.007581788580864668 2023-01-23 15:54:18.209066: step: 2624/1541, loss: 0.02633322775363922 2023-01-23 15:54:19.092681: step: 2626/1541, loss: 0.17720864713191986 2023-01-23 15:54:19.986464: step: 2628/1541, loss: 0.15053047239780426 2023-01-23 15:54:20.849155: step: 2630/1541, loss: 0.01558875571936369 2023-01-23 15:54:21.726845: step: 2632/1541, loss: 0.002501511247828603 2023-01-23 15:54:22.617173: step: 2634/1541, loss: 0.04119779169559479 2023-01-23 15:54:23.526128: step: 2636/1541, loss: 0.08625388890504837 2023-01-23 15:54:24.410827: step: 2638/1541, loss: 0.0031946890521794558 2023-01-23 15:54:25.286654: step: 2640/1541, loss: 0.06268365681171417 2023-01-23 15:54:26.255956: step: 2642/1541, loss: 0.013172515667974949 2023-01-23 15:54:27.111838: step: 2644/1541, loss: 0.18003897368907928 2023-01-23 15:54:28.040045: step: 2646/1541, loss: 0.01502611581236124 2023-01-23 15:54:28.928253: step: 2648/1541, loss: 0.021493468433618546 2023-01-23 15:54:29.815801: step: 2650/1541, loss: 0.009802935644984245 2023-01-23 15:54:30.694787: step: 2652/1541, loss: 0.2810271978378296 2023-01-23 15:54:31.578395: step: 2654/1541, loss: 0.01642637513577938 2023-01-23 15:54:32.517530: step: 2656/1541, loss: 0.027889788150787354 2023-01-23 15:54:33.430511: step: 2658/1541, loss: 0.0008346335962414742 2023-01-23 15:54:34.309674: step: 2660/1541, loss: 0.07244839519262314 2023-01-23 15:54:35.299957: step: 2662/1541, loss: 0.08431297540664673 2023-01-23 15:54:36.180318: step: 2664/1541, loss: 0.07564317435026169 2023-01-23 15:54:37.055005: step: 2666/1541, loss: 0.10700031369924545 2023-01-23 15:54:37.985649: step: 2668/1541, loss: 0.18256911635398865 2023-01-23 15:54:38.888875: step: 2670/1541, loss: 0.0210039634257555 2023-01-23 15:54:39.796034: step: 2672/1541, loss: 0.03444734215736389 2023-01-23 15:54:40.686790: step: 2674/1541, loss: 0.0020381691865622997 2023-01-23 15:54:41.572694: step: 2676/1541, loss: 0.04301535710692406 2023-01-23 15:54:42.459896: step: 2678/1541, loss: 0.2630396783351898 2023-01-23 15:54:43.333352: step: 2680/1541, loss: 0.06855262070894241 2023-01-23 15:54:44.196177: step: 2682/1541, loss: 0.002705136314034462 2023-01-23 15:54:45.120059: step: 2684/1541, loss: 0.035238541662693024 2023-01-23 15:54:46.029539: step: 2686/1541, loss: 0.02914995513856411 2023-01-23 15:54:46.933171: step: 2688/1541, loss: 0.0811445564031601 2023-01-23 15:54:47.853970: step: 2690/1541, loss: 0.006695534102618694 2023-01-23 15:54:48.802181: step: 2692/1541, loss: 0.26898276805877686 2023-01-23 15:54:49.718808: step: 2694/1541, loss: 0.0003144877264276147 2023-01-23 15:54:50.644803: step: 2696/1541, loss: 0.03179052099585533 2023-01-23 15:54:51.546739: step: 2698/1541, loss: 0.0007560638478025794 2023-01-23 15:54:52.428153: step: 2700/1541, loss: 0.016946865245699883 2023-01-23 15:54:53.301530: step: 2702/1541, loss: 0.23402968049049377 2023-01-23 15:54:54.218258: step: 2704/1541, loss: 0.005399795714765787 2023-01-23 15:54:55.094684: step: 2706/1541, loss: 0.003664137329906225 2023-01-23 15:54:55.977790: step: 2708/1541, loss: 0.0009698806679807603 2023-01-23 15:54:56.882368: step: 2710/1541, loss: 0.28034526109695435 2023-01-23 15:54:57.791273: step: 2712/1541, loss: 0.04023824632167816 2023-01-23 15:54:58.681831: step: 2714/1541, loss: 0.0354938879609108 2023-01-23 15:54:59.570130: step: 2716/1541, loss: 0.003234684467315674 2023-01-23 15:55:00.531425: step: 2718/1541, loss: 0.09747199714183807 2023-01-23 15:55:01.526119: step: 2720/1541, loss: 0.010437999852001667 2023-01-23 15:55:02.388380: step: 2722/1541, loss: 0.011608530767261982 2023-01-23 15:55:03.275807: step: 2724/1541, loss: 0.0011785533279180527 2023-01-23 15:55:04.199922: step: 2726/1541, loss: 0.058502111583948135 2023-01-23 15:55:05.065101: step: 2728/1541, loss: 0.0011072690831497312 2023-01-23 15:55:06.000660: step: 2730/1541, loss: 0.00178670440800488 2023-01-23 15:55:06.929947: step: 2732/1541, loss: 0.018457725644111633 2023-01-23 15:55:07.826418: step: 2734/1541, loss: 0.013171703554689884 2023-01-23 15:55:08.722803: step: 2736/1541, loss: 0.003397551132366061 2023-01-23 15:55:09.613938: step: 2738/1541, loss: 0.5528911352157593 2023-01-23 15:55:10.550291: step: 2740/1541, loss: 0.006911664269864559 2023-01-23 15:55:11.468048: step: 2742/1541, loss: 0.0042173960246145725 2023-01-23 15:55:12.337698: step: 2744/1541, loss: 0.00046379052218981087 2023-01-23 15:55:13.194501: step: 2746/1541, loss: 0.024938901886343956 2023-01-23 15:55:14.112460: step: 2748/1541, loss: 0.5050255060195923 2023-01-23 15:55:15.094774: step: 2750/1541, loss: 0.004403382074087858 2023-01-23 15:55:15.959158: step: 2752/1541, loss: 0.7875661253929138 2023-01-23 15:55:16.840453: step: 2754/1541, loss: 0.01974790170788765 2023-01-23 15:55:17.730691: step: 2756/1541, loss: 0.006564941257238388 2023-01-23 15:55:18.626401: step: 2758/1541, loss: 0.0034552651923149824 2023-01-23 15:55:19.628900: step: 2760/1541, loss: 0.19672396779060364 2023-01-23 15:55:20.543381: step: 2762/1541, loss: 0.003368377685546875 2023-01-23 15:55:21.418890: step: 2764/1541, loss: 0.02643084153532982 2023-01-23 15:55:22.302032: step: 2766/1541, loss: 0.004002790432423353 2023-01-23 15:55:23.306309: step: 2768/1541, loss: 0.05597265809774399 2023-01-23 15:55:24.310498: step: 2770/1541, loss: 0.011694769375026226 2023-01-23 15:55:25.172047: step: 2772/1541, loss: 0.002145293401554227 2023-01-23 15:55:26.089518: step: 2774/1541, loss: 0.006196903996169567 2023-01-23 15:55:26.970822: step: 2776/1541, loss: 0.02539796754717827 2023-01-23 15:55:27.899429: step: 2778/1541, loss: 0.00019108576816506684 2023-01-23 15:55:28.855655: step: 2780/1541, loss: 0.012100630439817905 2023-01-23 15:55:29.732884: step: 2782/1541, loss: 0.06172313168644905 2023-01-23 15:55:30.598108: step: 2784/1541, loss: 0.04924686253070831 2023-01-23 15:55:31.480373: step: 2786/1541, loss: 0.00017710067913867533 2023-01-23 15:55:32.383038: step: 2788/1541, loss: 0.010187527164816856 2023-01-23 15:55:33.225269: step: 2790/1541, loss: 0.2777642607688904 2023-01-23 15:55:34.141466: step: 2792/1541, loss: 0.13959920406341553 2023-01-23 15:55:34.967539: step: 2794/1541, loss: 0.043919216841459274 2023-01-23 15:55:35.850263: step: 2796/1541, loss: 0.10894639790058136 2023-01-23 15:55:36.719055: step: 2798/1541, loss: 0.00021851572091691196 2023-01-23 15:55:37.574123: step: 2800/1541, loss: 0.004390728194266558 2023-01-23 15:55:38.449095: step: 2802/1541, loss: 0.10737526416778564 2023-01-23 15:55:39.324291: step: 2804/1541, loss: 0.05285249650478363 2023-01-23 15:55:40.185551: step: 2806/1541, loss: 0.0008311847341246903 2023-01-23 15:55:41.085393: step: 2808/1541, loss: 0.008298628963530064 2023-01-23 15:55:41.981810: step: 2810/1541, loss: 0.033797282725572586 2023-01-23 15:55:42.877580: step: 2812/1541, loss: 0.19568407535552979 2023-01-23 15:55:43.760622: step: 2814/1541, loss: 0.12357411533594131 2023-01-23 15:55:44.689260: step: 2816/1541, loss: 0.11218823492527008 2023-01-23 15:55:45.603912: step: 2818/1541, loss: 0.3959500193595886 2023-01-23 15:55:46.526936: step: 2820/1541, loss: 0.0007806138601154089 2023-01-23 15:55:47.393607: step: 2822/1541, loss: 0.010018420405685902 2023-01-23 15:55:48.291978: step: 2824/1541, loss: 0.0958201214671135 2023-01-23 15:55:49.154844: step: 2826/1541, loss: 0.005230673588812351 2023-01-23 15:55:50.025682: step: 2828/1541, loss: 3.5968609154224396e-05 2023-01-23 15:55:50.892809: step: 2830/1541, loss: 0.018710151314735413 2023-01-23 15:55:51.803336: step: 2832/1541, loss: 0.04157489538192749 2023-01-23 15:55:52.680587: step: 2834/1541, loss: 0.00010836390720214695 2023-01-23 15:55:53.562329: step: 2836/1541, loss: 0.030125664547085762 2023-01-23 15:55:54.428600: step: 2838/1541, loss: 0.0007807783549651504 2023-01-23 15:55:55.367456: step: 2840/1541, loss: 0.003854892449453473 2023-01-23 15:55:56.252336: step: 2842/1541, loss: 0.0013192747719585896 2023-01-23 15:55:57.126819: step: 2844/1541, loss: 0.00103914609644562 2023-01-23 15:55:58.033136: step: 2846/1541, loss: 0.00029398826882243156 2023-01-23 15:55:58.964617: step: 2848/1541, loss: 0.0015410694759339094 2023-01-23 15:55:59.862733: step: 2850/1541, loss: 0.18512949347496033 2023-01-23 15:56:00.774173: step: 2852/1541, loss: 0.5069534778594971 2023-01-23 15:56:01.673559: step: 2854/1541, loss: 0.012563992291688919 2023-01-23 15:56:02.559105: step: 2856/1541, loss: 0.026303481310606003 2023-01-23 15:56:03.437448: step: 2858/1541, loss: 0.17289456725120544 2023-01-23 15:56:04.322691: step: 2860/1541, loss: 0.0008500578696839511 2023-01-23 15:56:05.198016: step: 2862/1541, loss: 0.09885463863611221 2023-01-23 15:56:06.087299: step: 2864/1541, loss: 0.015028588473796844 2023-01-23 15:56:06.985597: step: 2866/1541, loss: 0.005936404690146446 2023-01-23 15:56:07.877621: step: 2868/1541, loss: 0.003643559291958809 2023-01-23 15:56:08.754730: step: 2870/1541, loss: 0.05223919823765755 2023-01-23 15:56:09.662834: step: 2872/1541, loss: 0.04996166005730629 2023-01-23 15:56:10.562844: step: 2874/1541, loss: 0.012391210533678532 2023-01-23 15:56:11.430239: step: 2876/1541, loss: 0.0014161203289404511 2023-01-23 15:56:12.387434: step: 2878/1541, loss: 0.016045523807406425 2023-01-23 15:56:13.281125: step: 2880/1541, loss: 0.6414828896522522 2023-01-23 15:56:14.148678: step: 2882/1541, loss: 0.09196686744689941 2023-01-23 15:56:15.030032: step: 2884/1541, loss: 0.023182200267910957 2023-01-23 15:56:15.907406: step: 2886/1541, loss: 0.0041548339650034904 2023-01-23 15:56:16.825556: step: 2888/1541, loss: 0.22149163484573364 2023-01-23 15:56:17.744157: step: 2890/1541, loss: 0.1273849904537201 2023-01-23 15:56:18.617030: step: 2892/1541, loss: 0.01620900258421898 2023-01-23 15:56:19.572775: step: 2894/1541, loss: 0.6261981129646301 2023-01-23 15:56:20.479213: step: 2896/1541, loss: 0.15782874822616577 2023-01-23 15:56:21.380938: step: 2898/1541, loss: 0.009523909538984299 2023-01-23 15:56:22.262114: step: 2900/1541, loss: 0.04418128356337547 2023-01-23 15:56:23.141095: step: 2902/1541, loss: 0.015264321118593216 2023-01-23 15:56:24.064355: step: 2904/1541, loss: 0.024530895054340363 2023-01-23 15:56:24.916562: step: 2906/1541, loss: 0.06293885409832001 2023-01-23 15:56:25.844481: step: 2908/1541, loss: 1.3529990911483765 2023-01-23 15:56:26.744614: step: 2910/1541, loss: 0.015835048630833626 2023-01-23 15:56:27.661466: step: 2912/1541, loss: 0.010468626394867897 2023-01-23 15:56:28.575992: step: 2914/1541, loss: 0.02248901128768921 2023-01-23 15:56:29.468916: step: 2916/1541, loss: 0.030482595786452293 2023-01-23 15:56:30.404357: step: 2918/1541, loss: 0.0005906195146963 2023-01-23 15:56:31.391395: step: 2920/1541, loss: 0.002995413262397051 2023-01-23 15:56:32.255642: step: 2922/1541, loss: 0.0973128229379654 2023-01-23 15:56:33.097126: step: 2924/1541, loss: 0.07866373658180237 2023-01-23 15:56:34.062543: step: 2926/1541, loss: 0.06101684272289276 2023-01-23 15:56:34.967605: step: 2928/1541, loss: 0.08065156638622284 2023-01-23 15:56:35.904781: step: 2930/1541, loss: 0.007094845175743103 2023-01-23 15:56:36.828837: step: 2932/1541, loss: 0.001636718981899321 2023-01-23 15:56:37.711670: step: 2934/1541, loss: 0.08831894397735596 2023-01-23 15:56:38.629544: step: 2936/1541, loss: 0.007144049741327763 2023-01-23 15:56:39.529554: step: 2938/1541, loss: 0.2265322506427765 2023-01-23 15:56:40.400040: step: 2940/1541, loss: 0.11001688241958618 2023-01-23 15:56:41.263547: step: 2942/1541, loss: 0.04651531204581261 2023-01-23 15:56:42.141521: step: 2944/1541, loss: 0.2000340223312378 2023-01-23 15:56:43.035531: step: 2946/1541, loss: -1.0228701285086572e-05 2023-01-23 15:56:43.935098: step: 2948/1541, loss: 0.0198819562792778 2023-01-23 15:56:44.860685: step: 2950/1541, loss: 0.00196002796292305 2023-01-23 15:56:45.716752: step: 2952/1541, loss: 0.0033688403200358152 2023-01-23 15:56:46.599601: step: 2954/1541, loss: 0.011389588937163353 2023-01-23 15:56:47.550934: step: 2956/1541, loss: 0.05831901356577873 2023-01-23 15:56:48.431518: step: 2958/1541, loss: 0.010834434069693089 2023-01-23 15:56:49.321151: step: 2960/1541, loss: 0.004731787368655205 2023-01-23 15:56:50.228303: step: 2962/1541, loss: 0.0030451135244220495 2023-01-23 15:56:51.164443: step: 2964/1541, loss: 0.018752049654722214 2023-01-23 15:56:52.017748: step: 2966/1541, loss: 0.04199574515223503 2023-01-23 15:56:52.942944: step: 2968/1541, loss: 0.013190255500376225 2023-01-23 15:56:53.966680: step: 2970/1541, loss: 0.07642901688814163 2023-01-23 15:56:54.882595: step: 2972/1541, loss: 0.25328654050827026 2023-01-23 15:56:55.733770: step: 2974/1541, loss: 0.002212552120909095 2023-01-23 15:56:56.646580: step: 2976/1541, loss: 0.014902901835739613 2023-01-23 15:56:57.554990: step: 2978/1541, loss: 0.003885474754497409 2023-01-23 15:56:58.436510: step: 2980/1541, loss: 0.004311700351536274 2023-01-23 15:56:59.338859: step: 2982/1541, loss: 0.024256903678178787 2023-01-23 15:57:00.223317: step: 2984/1541, loss: 0.0013387973885983229 2023-01-23 15:57:01.109210: step: 2986/1541, loss: 0.1282084882259369 2023-01-23 15:57:02.025679: step: 2988/1541, loss: 0.003162747249007225 2023-01-23 15:57:02.959390: step: 2990/1541, loss: 0.02792954072356224 2023-01-23 15:57:03.853468: step: 2992/1541, loss: 0.40640413761138916 2023-01-23 15:57:04.755502: step: 2994/1541, loss: 0.03504069522023201 2023-01-23 15:57:05.636798: step: 2996/1541, loss: 0.01584884524345398 2023-01-23 15:57:06.483268: step: 2998/1541, loss: 0.08568316698074341 2023-01-23 15:57:07.370343: step: 3000/1541, loss: 0.34443923830986023 2023-01-23 15:57:08.296244: step: 3002/1541, loss: 0.03426467999815941 2023-01-23 15:57:09.239022: step: 3004/1541, loss: 0.03481053560972214 2023-01-23 15:57:10.124632: step: 3006/1541, loss: 0.006818383000791073 2023-01-23 15:57:11.008196: step: 3008/1541, loss: 0.07617966085672379 2023-01-23 15:57:11.922870: step: 3010/1541, loss: 0.0029320702888071537 2023-01-23 15:57:12.783006: step: 3012/1541, loss: 0.11073708534240723 2023-01-23 15:57:13.694696: step: 3014/1541, loss: 0.018673628568649292 2023-01-23 15:57:14.589486: step: 3016/1541, loss: 0.0520695298910141 2023-01-23 15:57:15.475079: step: 3018/1541, loss: 0.08394533395767212 2023-01-23 15:57:16.393155: step: 3020/1541, loss: 0.6776703596115112 2023-01-23 15:57:17.270512: step: 3022/1541, loss: 0.001243320293724537 2023-01-23 15:57:18.161805: step: 3024/1541, loss: 0.040850285440683365 2023-01-23 15:57:19.093484: step: 3026/1541, loss: 0.3025605380535126 2023-01-23 15:57:20.017266: step: 3028/1541, loss: 0.07940705120563507 2023-01-23 15:57:20.943155: step: 3030/1541, loss: 0.1483232080936432 2023-01-23 15:57:21.811653: step: 3032/1541, loss: 0.008587482385337353 2023-01-23 15:57:22.709953: step: 3034/1541, loss: 0.01875859685242176 2023-01-23 15:57:23.602519: step: 3036/1541, loss: 0.07201074808835983 2023-01-23 15:57:24.494693: step: 3038/1541, loss: 0.017841840162873268 2023-01-23 15:57:25.414954: step: 3040/1541, loss: 0.08868749439716339 2023-01-23 15:57:26.300631: step: 3042/1541, loss: 0.001253883820027113 2023-01-23 15:57:27.224926: step: 3044/1541, loss: 0.044090770184993744 2023-01-23 15:57:28.133509: step: 3046/1541, loss: 0.0202611293643713 2023-01-23 15:57:29.010782: step: 3048/1541, loss: 0.17195948958396912 2023-01-23 15:57:29.891383: step: 3050/1541, loss: 0.01876755617558956 2023-01-23 15:57:30.847523: step: 3052/1541, loss: 0.015040547586977482 2023-01-23 15:57:31.774625: step: 3054/1541, loss: 0.17849686741828918 2023-01-23 15:57:32.653297: step: 3056/1541, loss: 0.028739357367157936 2023-01-23 15:57:33.526698: step: 3058/1541, loss: 0.0007761581218801439 2023-01-23 15:57:34.420469: step: 3060/1541, loss: 0.02523862197995186 2023-01-23 15:57:35.341608: step: 3062/1541, loss: 0.05452106148004532 2023-01-23 15:57:36.230738: step: 3064/1541, loss: 0.036202557384967804 2023-01-23 15:57:37.128028: step: 3066/1541, loss: 0.0845109298825264 2023-01-23 15:57:37.976107: step: 3068/1541, loss: 0.03622889146208763 2023-01-23 15:57:38.867266: step: 3070/1541, loss: 0.1262214630842209 2023-01-23 15:57:39.763177: step: 3072/1541, loss: 0.0024434616789221764 2023-01-23 15:57:40.667750: step: 3074/1541, loss: 0.020285163074731827 2023-01-23 15:57:41.576519: step: 3076/1541, loss: 0.0009309180313721299 2023-01-23 15:57:42.485339: step: 3078/1541, loss: 0.008651205338537693 2023-01-23 15:57:43.375082: step: 3080/1541, loss: 0.12185567617416382 2023-01-23 15:57:44.274428: step: 3082/1541, loss: 0.02821112424135208 2023-01-23 15:57:44.888450: step: 3084/1541, loss: 0.014883562922477722 ================================================== Loss: 0.089 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6710048156535444, 'r': 0.6356065786757004, 'f1': 0.6528261991338468}, 'combined': 0.4308324034586093, 'epoch': 22} Test Chinese: {'event': {'p': 0.6140904311251314, 'r': 0.6998202516476932, 'f1': 0.6541584990198823}, 'argument': {'p': 0.7085278692903634, 'r': 0.7080118256856363, 'f1': 0.7082697534909862}, 'combined': 0.46332067884484557, 'epoch': 22} Dev Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6622286096414778, 'r': 0.6498771434645205, 'f1': 0.6559947414840996}, 'combined': 0.4326379774440589, 'epoch': 22} Test Korean: {'event': {'p': 0.6096282173498571, 'r': 0.7663271420011983, 'f1': 0.6790549508893019}, 'argument': {'p': 0.7061214899793525, 'r': 0.7237249400455513, 'f1': 0.7148148532164835}, 'combined': 0.48539856504586276, 'epoch': 22} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6676508942084495, 'r': 0.6485608009203222, 'f1': 0.6579674079565712}, 'combined': 0.4562457386833654, 'epoch': 22} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.7046691146552304, 'r': 0.7135217417237634, 'f1': 0.7090677982673104}, 'combined': 0.49278231343270057, 'epoch': 22} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6375080857839478, 'r': 0.6544330792118402, 'f1': 0.6458597200955278}, 'combined': 0.463694158017302, 'epoch': 22} Korean: {'event': {'p': 0.5625, 'r': 0.5714285714285714, 'f1': 0.5669291338582677}, 'argument': {'p': 0.6333911814681046, 'r': 0.41170426795426796, 'f1': 0.4990354763082036}, 'combined': 0.28291775034795796, 'epoch': 22} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5723667012647522, 'r': 0.612299261818107, 'f1': 0.5916599608579461}, 'combined': 0.28737769527385953, 'epoch': 22} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6868474446486486, 'r': 0.6340950032055018, 'f1': 0.6594178774763145}, 'combined': 0.43684358130043, 'epoch': 18} Test for Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7310372578834318, 'r': 0.6833493318043012, 'f1': 0.706389363899417}, 'combined': 0.4601124784178225, 'epoch': 18} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6661522633744856, 'r': 0.6366764995083579, 'f1': 0.6510809451985924}, 'combined': 0.489701907499796, 'epoch': 18} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 23 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 16:03:07.961972: step: 2/1541, loss: 0.008205832913517952 2023-01-23 16:03:08.836861: step: 4/1541, loss: 0.01076498907059431 2023-01-23 16:03:09.781595: step: 6/1541, loss: 0.11512435972690582 2023-01-23 16:03:10.671096: step: 8/1541, loss: 0.04519951343536377 2023-01-23 16:03:11.596813: step: 10/1541, loss: 0.006529401056468487 2023-01-23 16:03:12.497891: step: 12/1541, loss: 0.013989754021167755 2023-01-23 16:03:13.413936: step: 14/1541, loss: 0.2547847628593445 2023-01-23 16:03:14.313083: step: 16/1541, loss: 0.004363110288977623 2023-01-23 16:03:15.166712: step: 18/1541, loss: 0.0018070545047521591 2023-01-23 16:03:16.057001: step: 20/1541, loss: 0.22251728177070618 2023-01-23 16:03:16.963361: step: 22/1541, loss: 0.001882846001535654 2023-01-23 16:03:17.828629: step: 24/1541, loss: 0.06137992814183235 2023-01-23 16:03:18.715836: step: 26/1541, loss: 0.004965516738593578 2023-01-23 16:03:19.597418: step: 28/1541, loss: 0.034720078110694885 2023-01-23 16:03:20.481784: step: 30/1541, loss: 0.001310253399424255 2023-01-23 16:03:21.396544: step: 32/1541, loss: 0.01407424733042717 2023-01-23 16:03:22.256218: step: 34/1541, loss: 0.0016372528625652194 2023-01-23 16:03:23.153909: step: 36/1541, loss: 0.0004744830366689712 2023-01-23 16:03:24.049725: step: 38/1541, loss: 0.05508190765976906 2023-01-23 16:03:24.970479: step: 40/1541, loss: 0.3094422221183777 2023-01-23 16:03:25.875681: step: 42/1541, loss: 0.14184445142745972 2023-01-23 16:03:26.901144: step: 44/1541, loss: 0.0008655794081278145 2023-01-23 16:03:27.770969: step: 46/1541, loss: 0.3895121216773987 2023-01-23 16:03:28.690719: step: 48/1541, loss: 0.16957727074623108 2023-01-23 16:03:29.645611: step: 50/1541, loss: 0.2200416624546051 2023-01-23 16:03:30.542676: step: 52/1541, loss: 0.019049890339374542 2023-01-23 16:03:31.392422: step: 54/1541, loss: 0.00021014345111325383 2023-01-23 16:03:32.263573: step: 56/1541, loss: 0.0009198704501613975 2023-01-23 16:03:33.149654: step: 58/1541, loss: 0.055103227496147156 2023-01-23 16:03:34.068429: step: 60/1541, loss: 0.01033545657992363 2023-01-23 16:03:34.958440: step: 62/1541, loss: 0.056091610342264175 2023-01-23 16:03:35.866783: step: 64/1541, loss: 0.001243935083039105 2023-01-23 16:03:36.739322: step: 66/1541, loss: 0.0027136190328747034 2023-01-23 16:03:37.606032: step: 68/1541, loss: 0.12469106167554855 2023-01-23 16:03:38.465599: step: 70/1541, loss: 0.02207682654261589 2023-01-23 16:03:39.341910: step: 72/1541, loss: 0.0022591811139136553 2023-01-23 16:03:40.236148: step: 74/1541, loss: 0.016392312943935394 2023-01-23 16:03:41.166581: step: 76/1541, loss: 0.2701241374015808 2023-01-23 16:03:42.046013: step: 78/1541, loss: 0.00015816287486813962 2023-01-23 16:03:42.933062: step: 80/1541, loss: 0.057882484048604965 2023-01-23 16:03:43.883666: step: 82/1541, loss: 0.00029892081511206925 2023-01-23 16:03:44.785824: step: 84/1541, loss: 0.010891582816839218 2023-01-23 16:03:45.651777: step: 86/1541, loss: 0.015592743642628193 2023-01-23 16:03:46.526443: step: 88/1541, loss: 0.00029282315517775714 2023-01-23 16:03:47.397424: step: 90/1541, loss: 0.07154658436775208 2023-01-23 16:03:48.262548: step: 92/1541, loss: 0.0008904386777430773 2023-01-23 16:03:49.194689: step: 94/1541, loss: 0.003988540265709162 2023-01-23 16:03:50.103164: step: 96/1541, loss: 0.0719427689909935 2023-01-23 16:03:51.121709: step: 98/1541, loss: 0.0039458367973566055 2023-01-23 16:03:52.037420: step: 100/1541, loss: 0.015857890248298645 2023-01-23 16:03:52.986634: step: 102/1541, loss: 1.0842338800430298 2023-01-23 16:03:53.887380: step: 104/1541, loss: 0.11397887021303177 2023-01-23 16:03:54.816289: step: 106/1541, loss: 0.000140302290674299 2023-01-23 16:03:55.693676: step: 108/1541, loss: 0.0015163611387833953 2023-01-23 16:03:56.627137: step: 110/1541, loss: 0.0032612027134746313 2023-01-23 16:03:57.556094: step: 112/1541, loss: 0.08551512658596039 2023-01-23 16:03:58.487747: step: 114/1541, loss: 0.007136516273021698 2023-01-23 16:03:59.416032: step: 116/1541, loss: 0.3983885645866394 2023-01-23 16:04:00.416950: step: 118/1541, loss: 0.0007213758071884513 2023-01-23 16:04:01.329874: step: 120/1541, loss: 0.23666536808013916 2023-01-23 16:04:02.229670: step: 122/1541, loss: 0.030225805938243866 2023-01-23 16:04:03.082031: step: 124/1541, loss: 0.05823148414492607 2023-01-23 16:04:03.963730: step: 126/1541, loss: 0.013309882953763008 2023-01-23 16:04:04.877754: step: 128/1541, loss: 0.08799353986978531 2023-01-23 16:04:05.779395: step: 130/1541, loss: 0.0023097668308764696 2023-01-23 16:04:06.680805: step: 132/1541, loss: 0.0005016683717258275 2023-01-23 16:04:07.591936: step: 134/1541, loss: 4.861391062149778e-05 2023-01-23 16:04:08.483528: step: 136/1541, loss: 0.00468810461461544 2023-01-23 16:04:09.411538: step: 138/1541, loss: 0.0008476166985929012 2023-01-23 16:04:10.315051: step: 140/1541, loss: 0.0007470506825484335 2023-01-23 16:04:11.204055: step: 142/1541, loss: 0.0635710060596466 2023-01-23 16:04:12.052779: step: 144/1541, loss: 0.00024100618611555547 2023-01-23 16:04:12.962568: step: 146/1541, loss: 0.0049209436401724815 2023-01-23 16:04:13.857184: step: 148/1541, loss: 0.07661128044128418 2023-01-23 16:04:14.752925: step: 150/1541, loss: 0.004510076716542244 2023-01-23 16:04:15.671376: step: 152/1541, loss: 0.0003131759585812688 2023-01-23 16:04:16.574380: step: 154/1541, loss: 0.03865192085504532 2023-01-23 16:04:17.530579: step: 156/1541, loss: 0.01595212146639824 2023-01-23 16:04:18.375256: step: 158/1541, loss: 0.08350831270217896 2023-01-23 16:04:19.254280: step: 160/1541, loss: 0.026368526741862297 2023-01-23 16:04:20.175881: step: 162/1541, loss: 0.4365680515766144 2023-01-23 16:04:21.086001: step: 164/1541, loss: 0.031037505716085434 2023-01-23 16:04:22.008898: step: 166/1541, loss: 0.02663491852581501 2023-01-23 16:04:22.905886: step: 168/1541, loss: 0.007906516082584858 2023-01-23 16:04:23.811240: step: 170/1541, loss: 0.026947753503918648 2023-01-23 16:04:24.716089: step: 172/1541, loss: 0.0177592970430851 2023-01-23 16:04:25.590253: step: 174/1541, loss: 0.6280465126037598 2023-01-23 16:04:26.518390: step: 176/1541, loss: 0.005307917483150959 2023-01-23 16:04:27.423894: step: 178/1541, loss: 0.04221558943390846 2023-01-23 16:04:28.283217: step: 180/1541, loss: 0.0032082584220916033 2023-01-23 16:04:29.217157: step: 182/1541, loss: 0.00024560431484133005 2023-01-23 16:04:30.123851: step: 184/1541, loss: 0.12419191002845764 2023-01-23 16:04:30.997979: step: 186/1541, loss: 0.008803216740489006 2023-01-23 16:04:31.937523: step: 188/1541, loss: 0.0003091197577305138 2023-01-23 16:04:32.821018: step: 190/1541, loss: 0.0006379070109687746 2023-01-23 16:04:33.737393: step: 192/1541, loss: 0.001727774739265442 2023-01-23 16:04:34.656872: step: 194/1541, loss: 0.0032864816021174192 2023-01-23 16:04:35.548753: step: 196/1541, loss: 0.00010351603123126552 2023-01-23 16:04:36.442643: step: 198/1541, loss: 0.0011222421890124679 2023-01-23 16:04:37.323377: step: 200/1541, loss: 0.00012488063657656312 2023-01-23 16:04:38.197358: step: 202/1541, loss: 0.00014095273218117654 2023-01-23 16:04:39.162058: step: 204/1541, loss: 4.551493839244358e-05 2023-01-23 16:04:40.026024: step: 206/1541, loss: 0.27875351905822754 2023-01-23 16:04:40.933474: step: 208/1541, loss: 0.00300788227468729 2023-01-23 16:04:41.822133: step: 210/1541, loss: 0.02168113924562931 2023-01-23 16:04:42.735916: step: 212/1541, loss: 0.09148533642292023 2023-01-23 16:04:43.596078: step: 214/1541, loss: 0.012167568318545818 2023-01-23 16:04:44.484861: step: 216/1541, loss: 9.764062997419387e-05 2023-01-23 16:04:45.344918: step: 218/1541, loss: 0.3607998490333557 2023-01-23 16:04:46.214759: step: 220/1541, loss: 0.0022589548025280237 2023-01-23 16:04:47.111254: step: 222/1541, loss: 0.022958310320973396 2023-01-23 16:04:47.973924: step: 224/1541, loss: 0.027640335261821747 2023-01-23 16:04:48.857653: step: 226/1541, loss: 0.029891163110733032 2023-01-23 16:04:49.726097: step: 228/1541, loss: 0.09255477786064148 2023-01-23 16:04:50.676804: step: 230/1541, loss: 0.23009979724884033 2023-01-23 16:04:51.583054: step: 232/1541, loss: 0.1608072966337204 2023-01-23 16:04:52.451171: step: 234/1541, loss: 0.14635758101940155 2023-01-23 16:04:53.327774: step: 236/1541, loss: 0.00015000777784734964 2023-01-23 16:04:54.218929: step: 238/1541, loss: 0.018448837101459503 2023-01-23 16:04:55.158287: step: 240/1541, loss: 0.063526950776577 2023-01-23 16:04:56.018771: step: 242/1541, loss: 0.1306142956018448 2023-01-23 16:04:56.916342: step: 244/1541, loss: 0.02204579859972 2023-01-23 16:04:57.895350: step: 246/1541, loss: 0.00041478005005046725 2023-01-23 16:04:58.807411: step: 248/1541, loss: 0.010341980494558811 2023-01-23 16:04:59.733144: step: 250/1541, loss: 0.002568742260336876 2023-01-23 16:05:00.616934: step: 252/1541, loss: 0.005215104669332504 2023-01-23 16:05:01.524809: step: 254/1541, loss: 4.8571622755844146e-05 2023-01-23 16:05:02.432025: step: 256/1541, loss: 0.006186832673847675 2023-01-23 16:05:03.309118: step: 258/1541, loss: 0.0066803521476686 2023-01-23 16:05:04.171133: step: 260/1541, loss: 0.0018364943098276854 2023-01-23 16:05:05.059718: step: 262/1541, loss: 0.05385029688477516 2023-01-23 16:05:05.915061: step: 264/1541, loss: 0.00018841226119548082 2023-01-23 16:05:06.831122: step: 266/1541, loss: 0.0006154122529551387 2023-01-23 16:05:07.703466: step: 268/1541, loss: 0.001154615543782711 2023-01-23 16:05:08.574145: step: 270/1541, loss: 0.18711209297180176 2023-01-23 16:05:09.436321: step: 272/1541, loss: 0.007807488553225994 2023-01-23 16:05:10.307931: step: 274/1541, loss: 0.0012244838289916515 2023-01-23 16:05:11.182944: step: 276/1541, loss: 0.021754592657089233 2023-01-23 16:05:12.146125: step: 278/1541, loss: 0.007978103123605251 2023-01-23 16:05:13.008299: step: 280/1541, loss: 0.004712142050266266 2023-01-23 16:05:13.894370: step: 282/1541, loss: 0.0006134285358712077 2023-01-23 16:05:14.834946: step: 284/1541, loss: 0.31180432438850403 2023-01-23 16:05:15.697604: step: 286/1541, loss: 0.0005510496557690203 2023-01-23 16:05:16.611420: step: 288/1541, loss: 7.115124702453613 2023-01-23 16:05:17.521761: step: 290/1541, loss: 0.0026998750399798155 2023-01-23 16:05:18.471484: step: 292/1541, loss: 0.13980306684970856 2023-01-23 16:05:19.334957: step: 294/1541, loss: 0.00044758274452760816 2023-01-23 16:05:20.266863: step: 296/1541, loss: 0.0075658876448869705 2023-01-23 16:05:21.186711: step: 298/1541, loss: 0.020303981378674507 2023-01-23 16:05:22.074197: step: 300/1541, loss: 0.002649647183716297 2023-01-23 16:05:22.976759: step: 302/1541, loss: 0.01952623575925827 2023-01-23 16:05:23.855530: step: 304/1541, loss: 0.03974810615181923 2023-01-23 16:05:24.722858: step: 306/1541, loss: 0.0003797393583226949 2023-01-23 16:05:25.608127: step: 308/1541, loss: 0.00020953873172402382 2023-01-23 16:05:26.475549: step: 310/1541, loss: 0.006803778000175953 2023-01-23 16:05:27.429660: step: 312/1541, loss: 0.011640071868896484 2023-01-23 16:05:28.309366: step: 314/1541, loss: 0.0040674409829080105 2023-01-23 16:05:29.193649: step: 316/1541, loss: 0.01047357264906168 2023-01-23 16:05:30.090021: step: 318/1541, loss: 0.00039560996810905635 2023-01-23 16:05:30.961264: step: 320/1541, loss: 0.07200119644403458 2023-01-23 16:05:31.863140: step: 322/1541, loss: 0.12037789076566696 2023-01-23 16:05:32.744889: step: 324/1541, loss: 0.35703587532043457 2023-01-23 16:05:33.623444: step: 326/1541, loss: 0.00014273544366005808 2023-01-23 16:05:34.547591: step: 328/1541, loss: 0.3540946841239929 2023-01-23 16:05:35.424629: step: 330/1541, loss: 0.08011104166507721 2023-01-23 16:05:36.310325: step: 332/1541, loss: 0.21718864142894745 2023-01-23 16:05:37.289052: step: 334/1541, loss: 0.10458232462406158 2023-01-23 16:05:38.221734: step: 336/1541, loss: 0.07253661006689072 2023-01-23 16:05:39.110971: step: 338/1541, loss: 0.0003285135026089847 2023-01-23 16:05:40.012071: step: 340/1541, loss: 0.084368035197258 2023-01-23 16:05:40.919539: step: 342/1541, loss: 0.08800569921731949 2023-01-23 16:05:41.833184: step: 344/1541, loss: 0.0020906669087707996 2023-01-23 16:05:42.773734: step: 346/1541, loss: 0.0015907938359305263 2023-01-23 16:05:43.676144: step: 348/1541, loss: 0.003422815352678299 2023-01-23 16:05:44.529570: step: 350/1541, loss: 0.0007792926044203341 2023-01-23 16:05:45.418969: step: 352/1541, loss: 0.00038927042623981833 2023-01-23 16:05:46.306436: step: 354/1541, loss: 0.007836791686713696 2023-01-23 16:05:47.196071: step: 356/1541, loss: 0.00580071285367012 2023-01-23 16:05:48.071836: step: 358/1541, loss: 0.09986228495836258 2023-01-23 16:05:48.963191: step: 360/1541, loss: 0.0015898634446784854 2023-01-23 16:05:49.891416: step: 362/1541, loss: 0.06957769393920898 2023-01-23 16:05:50.799431: step: 364/1541, loss: 0.04284179210662842 2023-01-23 16:05:51.671852: step: 366/1541, loss: 0.01724478043615818 2023-01-23 16:05:52.513564: step: 368/1541, loss: 0.0017397220944985747 2023-01-23 16:05:53.481326: step: 370/1541, loss: 0.3203780949115753 2023-01-23 16:05:54.381612: step: 372/1541, loss: 0.0001152128170360811 2023-01-23 16:05:55.290100: step: 374/1541, loss: 0.0971042811870575 2023-01-23 16:05:56.187977: step: 376/1541, loss: 0.0646241307258606 2023-01-23 16:05:57.174646: step: 378/1541, loss: 0.0009383004507981241 2023-01-23 16:05:58.087222: step: 380/1541, loss: 0.4409390985965729 2023-01-23 16:05:59.011895: step: 382/1541, loss: 0.0009815366938710213 2023-01-23 16:05:59.930763: step: 384/1541, loss: 0.000248292344622314 2023-01-23 16:06:00.774029: step: 386/1541, loss: 0.0009083890472538769 2023-01-23 16:06:01.636463: step: 388/1541, loss: 8.417815115535632e-05 2023-01-23 16:06:02.587751: step: 390/1541, loss: 0.0018133869161829352 2023-01-23 16:06:03.478719: step: 392/1541, loss: 0.016058456152677536 2023-01-23 16:06:04.378173: step: 394/1541, loss: 0.0020475457422435284 2023-01-23 16:06:05.240951: step: 396/1541, loss: 0.0029323408380150795 2023-01-23 16:06:06.146356: step: 398/1541, loss: 0.00015413756773341447 2023-01-23 16:06:07.005166: step: 400/1541, loss: 0.21713602542877197 2023-01-23 16:06:07.881760: step: 402/1541, loss: 0.04760837182402611 2023-01-23 16:06:08.811197: step: 404/1541, loss: 0.01462325919419527 2023-01-23 16:06:09.728995: step: 406/1541, loss: 0.0005197445279918611 2023-01-23 16:06:10.615063: step: 408/1541, loss: 0.20967954397201538 2023-01-23 16:06:11.495085: step: 410/1541, loss: 0.01939748413860798 2023-01-23 16:06:12.380205: step: 412/1541, loss: 0.008606049232184887 2023-01-23 16:06:13.219949: step: 414/1541, loss: 0.0023042834363877773 2023-01-23 16:06:14.116689: step: 416/1541, loss: 0.26157814264297485 2023-01-23 16:06:15.003197: step: 418/1541, loss: 0.6615577340126038 2023-01-23 16:06:15.900988: step: 420/1541, loss: 0.0043400926515460014 2023-01-23 16:06:16.776642: step: 422/1541, loss: 0.006083513144403696 2023-01-23 16:06:17.679951: step: 424/1541, loss: 0.1913120597600937 2023-01-23 16:06:18.584382: step: 426/1541, loss: 0.1209699884057045 2023-01-23 16:06:19.461569: step: 428/1541, loss: 0.009838602505624294 2023-01-23 16:06:20.342318: step: 430/1541, loss: 0.0005183828179724514 2023-01-23 16:06:21.235202: step: 432/1541, loss: 0.16228066384792328 2023-01-23 16:06:22.238490: step: 434/1541, loss: 0.00031665642745792866 2023-01-23 16:06:23.117113: step: 436/1541, loss: 0.0024241451174020767 2023-01-23 16:06:23.975775: step: 438/1541, loss: 0.001723508001305163 2023-01-23 16:06:24.871161: step: 440/1541, loss: 0.0025039296597242355 2023-01-23 16:06:25.721530: step: 442/1541, loss: 0.0001633897190913558 2023-01-23 16:06:26.705052: step: 444/1541, loss: 0.0010657798266038299 2023-01-23 16:06:27.581346: step: 446/1541, loss: 0.4515605568885803 2023-01-23 16:06:28.432091: step: 448/1541, loss: 0.0012303448747843504 2023-01-23 16:06:29.402540: step: 450/1541, loss: 0.0009054193506017327 2023-01-23 16:06:30.271190: step: 452/1541, loss: 0.00031704167486168444 2023-01-23 16:06:31.135628: step: 454/1541, loss: 0.025148363783955574 2023-01-23 16:06:32.011067: step: 456/1541, loss: 0.00039163732435554266 2023-01-23 16:06:32.872820: step: 458/1541, loss: 0.001612979918718338 2023-01-23 16:06:33.728443: step: 460/1541, loss: 9.705322736408561e-05 2023-01-23 16:06:34.625852: step: 462/1541, loss: 0.0001404520298819989 2023-01-23 16:06:35.500154: step: 464/1541, loss: 0.26229745149612427 2023-01-23 16:06:36.400345: step: 466/1541, loss: 0.03986187279224396 2023-01-23 16:06:37.354963: step: 468/1541, loss: 0.0002199627342633903 2023-01-23 16:06:38.325891: step: 470/1541, loss: 0.0061822994612157345 2023-01-23 16:06:39.252804: step: 472/1541, loss: 0.00126964645460248 2023-01-23 16:06:40.135132: step: 474/1541, loss: 0.0004114630282856524 2023-01-23 16:06:41.000697: step: 476/1541, loss: 8.011984027689323e-05 2023-01-23 16:06:41.883652: step: 478/1541, loss: 0.10292696952819824 2023-01-23 16:06:42.888590: step: 480/1541, loss: 0.015205107629299164 2023-01-23 16:06:43.771610: step: 482/1541, loss: 0.071640245616436 2023-01-23 16:06:44.712337: step: 484/1541, loss: 0.0009304641280323267 2023-01-23 16:06:45.618163: step: 486/1541, loss: 1.05513596534729 2023-01-23 16:06:46.530916: step: 488/1541, loss: 0.059213265776634216 2023-01-23 16:06:47.418338: step: 490/1541, loss: 0.0061745247803628445 2023-01-23 16:06:48.342047: step: 492/1541, loss: 0.0004327071364969015 2023-01-23 16:06:49.211032: step: 494/1541, loss: 0.04803270101547241 2023-01-23 16:06:50.097346: step: 496/1541, loss: 0.6929053068161011 2023-01-23 16:06:50.981888: step: 498/1541, loss: 0.026013635098934174 2023-01-23 16:06:51.851614: step: 500/1541, loss: 0.0020409156568348408 2023-01-23 16:06:52.789412: step: 502/1541, loss: 0.15150567889213562 2023-01-23 16:06:53.655216: step: 504/1541, loss: 0.05044899135828018 2023-01-23 16:06:54.583156: step: 506/1541, loss: 0.05019384250044823 2023-01-23 16:06:55.475948: step: 508/1541, loss: 0.006897599436342716 2023-01-23 16:06:56.378619: step: 510/1541, loss: 0.037723932415246964 2023-01-23 16:06:57.274071: step: 512/1541, loss: 0.2451358586549759 2023-01-23 16:06:58.210405: step: 514/1541, loss: 0.05517381802201271 2023-01-23 16:06:59.078051: step: 516/1541, loss: 0.01548635121434927 2023-01-23 16:06:59.950219: step: 518/1541, loss: 0.0013292509829625487 2023-01-23 16:07:00.829666: step: 520/1541, loss: 0.8991804122924805 2023-01-23 16:07:01.718268: step: 522/1541, loss: 0.04973551258444786 2023-01-23 16:07:02.590520: step: 524/1541, loss: 0.0018785469001159072 2023-01-23 16:07:03.454440: step: 526/1541, loss: 0.03052675351500511 2023-01-23 16:07:04.352562: step: 528/1541, loss: 0.17413431406021118 2023-01-23 16:07:05.232272: step: 530/1541, loss: 0.15123647451400757 2023-01-23 16:07:06.102989: step: 532/1541, loss: 0.0004887281102128327 2023-01-23 16:07:06.972026: step: 534/1541, loss: 0.005058552138507366 2023-01-23 16:07:07.909219: step: 536/1541, loss: 0.012608327902853489 2023-01-23 16:07:08.821251: step: 538/1541, loss: 0.010685937479138374 2023-01-23 16:07:09.692116: step: 540/1541, loss: 0.00546841137111187 2023-01-23 16:07:10.615611: step: 542/1541, loss: 0.28789693117141724 2023-01-23 16:07:11.495421: step: 544/1541, loss: 0.004410295747220516 2023-01-23 16:07:12.390008: step: 546/1541, loss: 0.0006749174208380282 2023-01-23 16:07:13.278177: step: 548/1541, loss: 0.49867963790893555 2023-01-23 16:07:14.210820: step: 550/1541, loss: 0.04899149760603905 2023-01-23 16:07:15.123300: step: 552/1541, loss: 0.09986579418182373 2023-01-23 16:07:16.082684: step: 554/1541, loss: 0.03166699409484863 2023-01-23 16:07:16.994575: step: 556/1541, loss: 0.0008014383492991328 2023-01-23 16:07:17.907175: step: 558/1541, loss: 0.05036922171711922 2023-01-23 16:07:18.780691: step: 560/1541, loss: 0.0638919323682785 2023-01-23 16:07:19.639561: step: 562/1541, loss: 0.10140645503997803 2023-01-23 16:07:20.547069: step: 564/1541, loss: 0.08611052483320236 2023-01-23 16:07:21.497216: step: 566/1541, loss: 0.37631532549858093 2023-01-23 16:07:22.363493: step: 568/1541, loss: 0.09278430789709091 2023-01-23 16:07:23.259500: step: 570/1541, loss: 0.0023026312701404095 2023-01-23 16:07:24.140568: step: 572/1541, loss: 0.0004237232788000256 2023-01-23 16:07:25.017810: step: 574/1541, loss: 0.0010253542568534613 2023-01-23 16:07:25.962096: step: 576/1541, loss: 0.6030867099761963 2023-01-23 16:07:26.849452: step: 578/1541, loss: 0.04724154248833656 2023-01-23 16:07:27.751051: step: 580/1541, loss: 0.04342275857925415 2023-01-23 16:07:28.635162: step: 582/1541, loss: 0.00403648242354393 2023-01-23 16:07:29.523298: step: 584/1541, loss: 0.023838302120566368 2023-01-23 16:07:30.453612: step: 586/1541, loss: 0.0020384269300848246 2023-01-23 16:07:31.301869: step: 588/1541, loss: 0.20995552837848663 2023-01-23 16:07:32.176896: step: 590/1541, loss: 0.005232755560427904 2023-01-23 16:07:33.087310: step: 592/1541, loss: 0.010795870795845985 2023-01-23 16:07:33.965226: step: 594/1541, loss: 0.0010223608696833253 2023-01-23 16:07:34.868149: step: 596/1541, loss: 0.2873477339744568 2023-01-23 16:07:35.810193: step: 598/1541, loss: 0.0026951879262924194 2023-01-23 16:07:36.684574: step: 600/1541, loss: 0.0011618576245382428 2023-01-23 16:07:37.607480: step: 602/1541, loss: 0.0027009276673197746 2023-01-23 16:07:38.580648: step: 604/1541, loss: 0.0011260185856372118 2023-01-23 16:07:39.457691: step: 606/1541, loss: 0.1719992756843567 2023-01-23 16:07:40.375792: step: 608/1541, loss: 0.016938796266913414 2023-01-23 16:07:41.264107: step: 610/1541, loss: 0.1271817833185196 2023-01-23 16:07:42.153537: step: 612/1541, loss: 0.04726574197411537 2023-01-23 16:07:43.065199: step: 614/1541, loss: 0.025968721136450768 2023-01-23 16:07:43.931390: step: 616/1541, loss: 0.022797927260398865 2023-01-23 16:07:44.838258: step: 618/1541, loss: 0.024186842143535614 2023-01-23 16:07:45.721325: step: 620/1541, loss: 0.0032596152741461992 2023-01-23 16:07:46.593804: step: 622/1541, loss: 0.06564107537269592 2023-01-23 16:07:47.453993: step: 624/1541, loss: 0.016286995261907578 2023-01-23 16:07:48.337261: step: 626/1541, loss: 0.15561673045158386 2023-01-23 16:07:49.222547: step: 628/1541, loss: 0.4767848253250122 2023-01-23 16:07:50.103793: step: 630/1541, loss: 0.0006365125300362706 2023-01-23 16:07:51.036024: step: 632/1541, loss: 0.39229243993759155 2023-01-23 16:07:51.903779: step: 634/1541, loss: 0.034312617033720016 2023-01-23 16:07:52.853416: step: 636/1541, loss: 0.04090477526187897 2023-01-23 16:07:53.717842: step: 638/1541, loss: 0.004656099248677492 2023-01-23 16:07:54.564856: step: 640/1541, loss: 0.0030633031856268644 2023-01-23 16:07:55.576332: step: 642/1541, loss: 0.02938898466527462 2023-01-23 16:07:56.429478: step: 644/1541, loss: 0.026798970997333527 2023-01-23 16:07:57.351162: step: 646/1541, loss: 0.032244376838207245 2023-01-23 16:07:58.254080: step: 648/1541, loss: 0.0039056497626006603 2023-01-23 16:07:59.144461: step: 650/1541, loss: 0.006614347454160452 2023-01-23 16:08:00.026571: step: 652/1541, loss: 0.002151086460798979 2023-01-23 16:08:00.929939: step: 654/1541, loss: 0.06347708404064178 2023-01-23 16:08:01.905146: step: 656/1541, loss: 0.0005175817641429603 2023-01-23 16:08:02.808432: step: 658/1541, loss: 0.035860348492860794 2023-01-23 16:08:03.666208: step: 660/1541, loss: 0.009157506749033928 2023-01-23 16:08:04.569042: step: 662/1541, loss: 0.006786109413951635 2023-01-23 16:08:05.580478: step: 664/1541, loss: 0.0697312280535698 2023-01-23 16:08:06.564066: step: 666/1541, loss: 0.3278771638870239 2023-01-23 16:08:07.474610: step: 668/1541, loss: 0.011465542018413544 2023-01-23 16:08:08.377255: step: 670/1541, loss: 0.08123284578323364 2023-01-23 16:08:09.291254: step: 672/1541, loss: 0.0008636703714728355 2023-01-23 16:08:10.137770: step: 674/1541, loss: 0.14109894633293152 2023-01-23 16:08:11.061409: step: 676/1541, loss: 0.5727736949920654 2023-01-23 16:08:11.923068: step: 678/1541, loss: 0.02371404878795147 2023-01-23 16:08:12.828295: step: 680/1541, loss: 0.0016246669692918658 2023-01-23 16:08:13.730422: step: 682/1541, loss: 0.002745307283475995 2023-01-23 16:08:14.633505: step: 684/1541, loss: 0.021179834380745888 2023-01-23 16:08:15.560829: step: 686/1541, loss: 0.00016565076657570899 2023-01-23 16:08:16.467125: step: 688/1541, loss: 0.021836278960108757 2023-01-23 16:08:17.351150: step: 690/1541, loss: 0.22882825136184692 2023-01-23 16:08:18.363443: step: 692/1541, loss: 0.002546029631048441 2023-01-23 16:08:19.220476: step: 694/1541, loss: 0.04303361102938652 2023-01-23 16:08:20.105807: step: 696/1541, loss: 0.17471091449260712 2023-01-23 16:08:21.009701: step: 698/1541, loss: 0.00033148808870464563 2023-01-23 16:08:21.865253: step: 700/1541, loss: 0.0016130807343870401 2023-01-23 16:08:22.773323: step: 702/1541, loss: 0.6820102334022522 2023-01-23 16:08:23.664837: step: 704/1541, loss: 0.0036050742492079735 2023-01-23 16:08:24.612110: step: 706/1541, loss: 0.01693763956427574 2023-01-23 16:08:25.540147: step: 708/1541, loss: 0.016824953258037567 2023-01-23 16:08:26.440316: step: 710/1541, loss: 0.0012265846598893404 2023-01-23 16:08:27.368654: step: 712/1541, loss: 0.04872339218854904 2023-01-23 16:08:28.249536: step: 714/1541, loss: 0.02265581116080284 2023-01-23 16:08:29.151205: step: 716/1541, loss: 0.004177144728600979 2023-01-23 16:08:30.058855: step: 718/1541, loss: 0.022974371910095215 2023-01-23 16:08:30.940086: step: 720/1541, loss: 0.00047672668006271124 2023-01-23 16:08:31.947550: step: 722/1541, loss: 0.2129119336605072 2023-01-23 16:08:32.820761: step: 724/1541, loss: 0.8814366459846497 2023-01-23 16:08:33.788393: step: 726/1541, loss: 0.33603933453559875 2023-01-23 16:08:34.674070: step: 728/1541, loss: 0.05269956588745117 2023-01-23 16:08:35.575306: step: 730/1541, loss: 0.005815932992845774 2023-01-23 16:08:36.445988: step: 732/1541, loss: 0.04002685472369194 2023-01-23 16:08:37.377581: step: 734/1541, loss: 0.0073584094643592834 2023-01-23 16:08:38.301697: step: 736/1541, loss: 0.0009319477831013501 2023-01-23 16:08:39.158796: step: 738/1541, loss: 0.507378339767456 2023-01-23 16:08:40.066423: step: 740/1541, loss: 0.016046814620494843 2023-01-23 16:08:40.962563: step: 742/1541, loss: 0.23719681799411774 2023-01-23 16:08:41.868504: step: 744/1541, loss: 0.00019078084733337164 2023-01-23 16:08:42.812652: step: 746/1541, loss: 0.03997017443180084 2023-01-23 16:08:43.726254: step: 748/1541, loss: 0.0015412315260618925 2023-01-23 16:08:44.609246: step: 750/1541, loss: 0.04469624161720276 2023-01-23 16:08:45.516507: step: 752/1541, loss: 0.0111627746373415 2023-01-23 16:08:46.457573: step: 754/1541, loss: 0.03166851028800011 2023-01-23 16:08:47.344270: step: 756/1541, loss: 0.0003714373451657593 2023-01-23 16:08:48.195463: step: 758/1541, loss: 0.0006371339550241828 2023-01-23 16:08:49.106649: step: 760/1541, loss: 0.007182450033724308 2023-01-23 16:08:49.998466: step: 762/1541, loss: 0.008930773474276066 2023-01-23 16:08:50.891902: step: 764/1541, loss: 0.3727996349334717 2023-01-23 16:08:51.878931: step: 766/1541, loss: 0.07846932858228683 2023-01-23 16:08:52.738108: step: 768/1541, loss: 0.0018238781485706568 2023-01-23 16:08:53.632141: step: 770/1541, loss: 0.08809399604797363 2023-01-23 16:08:54.474624: step: 772/1541, loss: 0.0007255119271576405 2023-01-23 16:08:55.330517: step: 774/1541, loss: 0.10711506009101868 2023-01-23 16:08:56.246273: step: 776/1541, loss: 0.000737805210519582 2023-01-23 16:08:57.082362: step: 778/1541, loss: 0.0009216268081218004 2023-01-23 16:08:57.928117: step: 780/1541, loss: 0.4964028298854828 2023-01-23 16:08:58.841827: step: 782/1541, loss: 0.17914485931396484 2023-01-23 16:08:59.720878: step: 784/1541, loss: 0.08636319637298584 2023-01-23 16:09:00.598177: step: 786/1541, loss: 0.006624755449593067 2023-01-23 16:09:01.484885: step: 788/1541, loss: 0.25571975111961365 2023-01-23 16:09:02.422685: step: 790/1541, loss: 0.005797173362225294 2023-01-23 16:09:03.296554: step: 792/1541, loss: 0.00017509462486486882 2023-01-23 16:09:04.177035: step: 794/1541, loss: 0.000527342373970896 2023-01-23 16:09:05.092816: step: 796/1541, loss: 0.0054032327607274055 2023-01-23 16:09:06.011027: step: 798/1541, loss: 0.0007939494680613279 2023-01-23 16:09:06.890795: step: 800/1541, loss: 0.07348864525556564 2023-01-23 16:09:07.784095: step: 802/1541, loss: 0.14928558468818665 2023-01-23 16:09:08.657753: step: 804/1541, loss: 0.0003315238282084465 2023-01-23 16:09:09.556674: step: 806/1541, loss: 0.015037092380225658 2023-01-23 16:09:10.476942: step: 808/1541, loss: 0.03793199732899666 2023-01-23 16:09:11.401499: step: 810/1541, loss: 0.09994448721408844 2023-01-23 16:09:12.383797: step: 812/1541, loss: 0.05771201476454735 2023-01-23 16:09:13.277116: step: 814/1541, loss: 0.10464366525411606 2023-01-23 16:09:14.174017: step: 816/1541, loss: 0.023472405970096588 2023-01-23 16:09:15.075920: step: 818/1541, loss: 0.15454764664173126 2023-01-23 16:09:16.058920: step: 820/1541, loss: 0.09044748544692993 2023-01-23 16:09:16.956813: step: 822/1541, loss: 0.10455408692359924 2023-01-23 16:09:17.902486: step: 824/1541, loss: 0.28132107853889465 2023-01-23 16:09:18.779266: step: 826/1541, loss: 0.29177960753440857 2023-01-23 16:09:19.699569: step: 828/1541, loss: 0.019126469269394875 2023-01-23 16:09:20.575601: step: 830/1541, loss: 0.09427200257778168 2023-01-23 16:09:21.443445: step: 832/1541, loss: 0.01819387637078762 2023-01-23 16:09:22.293549: step: 834/1541, loss: 0.0005255488213151693 2023-01-23 16:09:23.279169: step: 836/1541, loss: 0.0009726972784847021 2023-01-23 16:09:24.169740: step: 838/1541, loss: 0.1789364516735077 2023-01-23 16:09:25.045921: step: 840/1541, loss: 0.0016084035160019994 2023-01-23 16:09:25.943771: step: 842/1541, loss: 1.7795978784561157 2023-01-23 16:09:26.833370: step: 844/1541, loss: 0.00730223348364234 2023-01-23 16:09:27.750384: step: 846/1541, loss: 0.023098105564713478 2023-01-23 16:09:28.635067: step: 848/1541, loss: 0.050161294639110565 2023-01-23 16:09:29.553907: step: 850/1541, loss: 0.1968904733657837 2023-01-23 16:09:30.462315: step: 852/1541, loss: 1.3527106046676636 2023-01-23 16:09:31.352179: step: 854/1541, loss: 0.0010386925423517823 2023-01-23 16:09:32.261747: step: 856/1541, loss: 0.0013178419321775436 2023-01-23 16:09:33.136085: step: 858/1541, loss: 0.04634534567594528 2023-01-23 16:09:34.045267: step: 860/1541, loss: 0.1650097817182541 2023-01-23 16:09:34.946364: step: 862/1541, loss: 0.02035929076373577 2023-01-23 16:09:35.813264: step: 864/1541, loss: 0.024898894131183624 2023-01-23 16:09:36.682883: step: 866/1541, loss: 0.00038978533120825887 2023-01-23 16:09:37.528283: step: 868/1541, loss: 0.08264736831188202 2023-01-23 16:09:38.450130: step: 870/1541, loss: 0.2495298981666565 2023-01-23 16:09:39.319892: step: 872/1541, loss: 0.005595408380031586 2023-01-23 16:09:40.214985: step: 874/1541, loss: 0.0020474863704293966 2023-01-23 16:09:41.146190: step: 876/1541, loss: 0.10940621793270111 2023-01-23 16:09:42.014396: step: 878/1541, loss: 0.032176580280065536 2023-01-23 16:09:42.866825: step: 880/1541, loss: 0.003931339830160141 2023-01-23 16:09:43.783631: step: 882/1541, loss: 0.048018619418144226 2023-01-23 16:09:44.675954: step: 884/1541, loss: 0.0009649571147747338 2023-01-23 16:09:45.658379: step: 886/1541, loss: 0.003894075518473983 2023-01-23 16:09:46.552350: step: 888/1541, loss: 0.0028965373057872057 2023-01-23 16:09:47.474947: step: 890/1541, loss: 0.06354095786809921 2023-01-23 16:09:48.382042: step: 892/1541, loss: 0.42264851927757263 2023-01-23 16:09:49.262038: step: 894/1541, loss: 0.002505144104361534 2023-01-23 16:09:50.209479: step: 896/1541, loss: 0.08328446745872498 2023-01-23 16:09:51.089390: step: 898/1541, loss: 0.0018529373919591308 2023-01-23 16:09:51.946348: step: 900/1541, loss: 0.04710143804550171 2023-01-23 16:09:52.849969: step: 902/1541, loss: 0.03851579129695892 2023-01-23 16:09:53.769774: step: 904/1541, loss: 0.021412108093500137 2023-01-23 16:09:54.674522: step: 906/1541, loss: 0.030741382390260696 2023-01-23 16:09:55.543896: step: 908/1541, loss: 0.032283276319503784 2023-01-23 16:09:56.448471: step: 910/1541, loss: 0.10284639149904251 2023-01-23 16:09:57.322183: step: 912/1541, loss: 0.00013014933210797608 2023-01-23 16:09:58.221032: step: 914/1541, loss: 0.000557565363124013 2023-01-23 16:09:59.187610: step: 916/1541, loss: 0.007792992517352104 2023-01-23 16:10:00.061687: step: 918/1541, loss: 0.0022860129829496145 2023-01-23 16:10:00.972330: step: 920/1541, loss: 0.004750717431306839 2023-01-23 16:10:01.897320: step: 922/1541, loss: 0.0014396499609574676 2023-01-23 16:10:02.764404: step: 924/1541, loss: 0.1306796371936798 2023-01-23 16:10:03.680779: step: 926/1541, loss: 0.04716840013861656 2023-01-23 16:10:04.574840: step: 928/1541, loss: 0.0045387945137917995 2023-01-23 16:10:05.442733: step: 930/1541, loss: 0.19956478476524353 2023-01-23 16:10:06.339216: step: 932/1541, loss: 0.012463107705116272 2023-01-23 16:10:07.201939: step: 934/1541, loss: 0.001212015631608665 2023-01-23 16:10:08.084335: step: 936/1541, loss: 0.0029132410418242216 2023-01-23 16:10:08.981865: step: 938/1541, loss: 0.029861584305763245 2023-01-23 16:10:09.862943: step: 940/1541, loss: 0.011669117957353592 2023-01-23 16:10:10.768546: step: 942/1541, loss: 0.41974011063575745 2023-01-23 16:10:11.696017: step: 944/1541, loss: 0.04180225357413292 2023-01-23 16:10:12.589589: step: 946/1541, loss: 0.0002797538763843477 2023-01-23 16:10:13.542258: step: 948/1541, loss: 0.04339170828461647 2023-01-23 16:10:14.488083: step: 950/1541, loss: 0.0865558534860611 2023-01-23 16:10:15.370303: step: 952/1541, loss: 0.03282415121793747 2023-01-23 16:10:16.243977: step: 954/1541, loss: 0.10377970337867737 2023-01-23 16:10:17.121912: step: 956/1541, loss: 0.0006038597784936428 2023-01-23 16:10:18.007855: step: 958/1541, loss: 0.00702089536935091 2023-01-23 16:10:18.890443: step: 960/1541, loss: 0.03888329491019249 2023-01-23 16:10:19.782940: step: 962/1541, loss: 0.10039234906435013 2023-01-23 16:10:20.714128: step: 964/1541, loss: 0.21353530883789062 2023-01-23 16:10:21.633115: step: 966/1541, loss: 0.007909219712018967 2023-01-23 16:10:22.511522: step: 968/1541, loss: 0.0036635047290474176 2023-01-23 16:10:23.412697: step: 970/1541, loss: 0.048002321273088455 2023-01-23 16:10:24.301191: step: 972/1541, loss: 0.005515719298273325 2023-01-23 16:10:25.216284: step: 974/1541, loss: 0.08011837303638458 2023-01-23 16:10:26.100321: step: 976/1541, loss: 0.0012070947559550405 2023-01-23 16:10:27.001182: step: 978/1541, loss: 0.24706557393074036 2023-01-23 16:10:28.004919: step: 980/1541, loss: 0.004960270132869482 2023-01-23 16:10:28.909749: step: 982/1541, loss: 0.25755712389945984 2023-01-23 16:10:29.810394: step: 984/1541, loss: 0.034311145544052124 2023-01-23 16:10:30.835376: step: 986/1541, loss: 0.11311235278844833 2023-01-23 16:10:31.773773: step: 988/1541, loss: 0.21397802233695984 2023-01-23 16:10:32.670262: step: 990/1541, loss: 0.003929521888494492 2023-01-23 16:10:33.567305: step: 992/1541, loss: 0.23041632771492004 2023-01-23 16:10:34.480554: step: 994/1541, loss: 0.004323937930166721 2023-01-23 16:10:35.325219: step: 996/1541, loss: 0.0015614776639267802 2023-01-23 16:10:36.326594: step: 998/1541, loss: 0.014951759949326515 2023-01-23 16:10:37.184045: step: 1000/1541, loss: 0.20140564441680908 2023-01-23 16:10:38.068286: step: 1002/1541, loss: 0.00045584383769892156 2023-01-23 16:10:38.961297: step: 1004/1541, loss: 0.004709964152425528 2023-01-23 16:10:39.824461: step: 1006/1541, loss: 0.009460070170462132 2023-01-23 16:10:40.696301: step: 1008/1541, loss: 0.11972978711128235 2023-01-23 16:10:41.594913: step: 1010/1541, loss: 0.016998691484332085 2023-01-23 16:10:42.552229: step: 1012/1541, loss: 0.0028926474042236805 2023-01-23 16:10:43.397775: step: 1014/1541, loss: 0.09806593507528305 2023-01-23 16:10:44.279754: step: 1016/1541, loss: 0.024696236476302147 2023-01-23 16:10:45.188994: step: 1018/1541, loss: 0.1030571460723877 2023-01-23 16:10:46.105106: step: 1020/1541, loss: 0.014517088420689106 2023-01-23 16:10:46.993503: step: 1022/1541, loss: 0.14303505420684814 2023-01-23 16:10:47.918062: step: 1024/1541, loss: 0.04668610170483589 2023-01-23 16:10:48.838004: step: 1026/1541, loss: 0.0061955880373716354 2023-01-23 16:10:49.721427: step: 1028/1541, loss: 0.00013898321776650846 2023-01-23 16:10:50.620039: step: 1030/1541, loss: 0.045306336134672165 2023-01-23 16:10:51.491286: step: 1032/1541, loss: 0.000996338203549385 2023-01-23 16:10:52.377493: step: 1034/1541, loss: 0.013191456906497478 2023-01-23 16:10:53.243065: step: 1036/1541, loss: 0.0061800493858754635 2023-01-23 16:10:54.201083: step: 1038/1541, loss: 0.005035314243286848 2023-01-23 16:10:55.118477: step: 1040/1541, loss: 0.5379709005355835 2023-01-23 16:10:56.025738: step: 1042/1541, loss: 0.09276384115219116 2023-01-23 16:10:56.902331: step: 1044/1541, loss: 0.016940249130129814 2023-01-23 16:10:57.800079: step: 1046/1541, loss: 0.060170963406562805 2023-01-23 16:10:58.688292: step: 1048/1541, loss: 7.785418711137027e-05 2023-01-23 16:10:59.584505: step: 1050/1541, loss: 0.13529425859451294 2023-01-23 16:11:00.451130: step: 1052/1541, loss: 0.0015669080894440413 2023-01-23 16:11:01.391837: step: 1054/1541, loss: 0.17493674159049988 2023-01-23 16:11:02.341979: step: 1056/1541, loss: 0.0008592142257839441 2023-01-23 16:11:03.298632: step: 1058/1541, loss: 0.001089815516024828 2023-01-23 16:11:04.180966: step: 1060/1541, loss: 0.1018839031457901 2023-01-23 16:11:05.096355: step: 1062/1541, loss: 0.005231994204223156 2023-01-23 16:11:06.002438: step: 1064/1541, loss: 0.0020111240446567535 2023-01-23 16:11:06.879917: step: 1066/1541, loss: 0.001522703911177814 2023-01-23 16:11:07.765328: step: 1068/1541, loss: 0.0008919976535253227 2023-01-23 16:11:08.630358: step: 1070/1541, loss: 4.302845991333015e-05 2023-01-23 16:11:09.524903: step: 1072/1541, loss: 0.029468737542629242 2023-01-23 16:11:10.433693: step: 1074/1541, loss: 0.00022484992223326117 2023-01-23 16:11:11.336265: step: 1076/1541, loss: 0.002818190725520253 2023-01-23 16:11:12.223257: step: 1078/1541, loss: 0.11336187273263931 2023-01-23 16:11:13.098051: step: 1080/1541, loss: 0.0019311945652589202 2023-01-23 16:11:14.012402: step: 1082/1541, loss: 0.0007566289277747273 2023-01-23 16:11:14.919912: step: 1084/1541, loss: 0.0005347174592316151 2023-01-23 16:11:15.802264: step: 1086/1541, loss: 0.2423081398010254 2023-01-23 16:11:16.670854: step: 1088/1541, loss: 0.021759597584605217 2023-01-23 16:11:17.547208: step: 1090/1541, loss: 0.018634017556905746 2023-01-23 16:11:18.448718: step: 1092/1541, loss: 0.11374367773532867 2023-01-23 16:11:19.345296: step: 1094/1541, loss: 0.49264124035835266 2023-01-23 16:11:20.227669: step: 1096/1541, loss: 0.003991400822997093 2023-01-23 16:11:21.153448: step: 1098/1541, loss: 0.02228745073080063 2023-01-23 16:11:22.077844: step: 1100/1541, loss: 0.004907690919935703 2023-01-23 16:11:22.994023: step: 1102/1541, loss: 0.2552306652069092 2023-01-23 16:11:23.895193: step: 1104/1541, loss: 0.33162665367126465 2023-01-23 16:11:24.825687: step: 1106/1541, loss: 0.031703654676675797 2023-01-23 16:11:25.703384: step: 1108/1541, loss: 0.0037812981754541397 2023-01-23 16:11:26.577603: step: 1110/1541, loss: 0.09553582221269608 2023-01-23 16:11:27.515432: step: 1112/1541, loss: 0.0012856508838012815 2023-01-23 16:11:28.373490: step: 1114/1541, loss: 0.007672838866710663 2023-01-23 16:11:29.222956: step: 1116/1541, loss: 0.0005444870330393314 2023-01-23 16:11:30.115066: step: 1118/1541, loss: 0.030545273795723915 2023-01-23 16:11:31.020031: step: 1120/1541, loss: 0.0010349296499043703 2023-01-23 16:11:31.904413: step: 1122/1541, loss: 0.0002669472014531493 2023-01-23 16:11:32.882660: step: 1124/1541, loss: 0.007812235504388809 2023-01-23 16:11:33.799666: step: 1126/1541, loss: 0.007534504868090153 2023-01-23 16:11:34.706075: step: 1128/1541, loss: 0.012798616662621498 2023-01-23 16:11:35.593701: step: 1130/1541, loss: 0.005874905269593 2023-01-23 16:11:36.466427: step: 1132/1541, loss: 0.007428275421261787 2023-01-23 16:11:37.354129: step: 1134/1541, loss: 0.06154071167111397 2023-01-23 16:11:38.286523: step: 1136/1541, loss: 0.002444812562316656 2023-01-23 16:11:39.131517: step: 1138/1541, loss: 0.0063141412101686 2023-01-23 16:11:40.000831: step: 1140/1541, loss: 0.16399399936199188 2023-01-23 16:11:40.887527: step: 1142/1541, loss: 0.0001464129745727405 2023-01-23 16:11:41.804682: step: 1144/1541, loss: 0.06483175605535507 2023-01-23 16:11:42.712081: step: 1146/1541, loss: 0.012531163170933723 2023-01-23 16:11:43.606928: step: 1148/1541, loss: 0.01838819682598114 2023-01-23 16:11:44.502784: step: 1150/1541, loss: 0.0055497512221336365 2023-01-23 16:11:45.428805: step: 1152/1541, loss: 0.008332585915923119 2023-01-23 16:11:46.395102: step: 1154/1541, loss: 0.005394876468926668 2023-01-23 16:11:47.297342: step: 1156/1541, loss: 0.09278267621994019 2023-01-23 16:11:48.164378: step: 1158/1541, loss: 0.007335755508393049 2023-01-23 16:11:49.148448: step: 1160/1541, loss: 0.019525310024619102 2023-01-23 16:11:50.022670: step: 1162/1541, loss: 0.0037813186645507812 2023-01-23 16:11:50.879577: step: 1164/1541, loss: 0.0022617075592279434 2023-01-23 16:11:51.772764: step: 1166/1541, loss: 0.11151943355798721 2023-01-23 16:11:52.658775: step: 1168/1541, loss: 0.032897356897592545 2023-01-23 16:11:53.555540: step: 1170/1541, loss: 0.00014249960076995194 2023-01-23 16:11:54.449484: step: 1172/1541, loss: 0.0022218655794858932 2023-01-23 16:11:55.351894: step: 1174/1541, loss: 0.13384941220283508 2023-01-23 16:11:56.219972: step: 1176/1541, loss: 0.0015795211074873805 2023-01-23 16:11:57.072578: step: 1178/1541, loss: 0.002278375905007124 2023-01-23 16:11:57.947344: step: 1180/1541, loss: 0.00010912585275946185 2023-01-23 16:11:58.933731: step: 1182/1541, loss: 0.9919047951698303 2023-01-23 16:11:59.818976: step: 1184/1541, loss: 0.24049749970436096 2023-01-23 16:12:00.684307: step: 1186/1541, loss: 0.1407022923231125 2023-01-23 16:12:01.544577: step: 1188/1541, loss: 0.000735585461370647 2023-01-23 16:12:02.450208: step: 1190/1541, loss: 1.048126459121704 2023-01-23 16:12:03.387232: step: 1192/1541, loss: 0.0031135459430515766 2023-01-23 16:12:04.272149: step: 1194/1541, loss: 0.09505169093608856 2023-01-23 16:12:05.217306: step: 1196/1541, loss: 0.05495366454124451 2023-01-23 16:12:06.088701: step: 1198/1541, loss: 0.0006575429579243064 2023-01-23 16:12:07.010664: step: 1200/1541, loss: 0.03508000448346138 2023-01-23 16:12:07.923260: step: 1202/1541, loss: 0.04215511307120323 2023-01-23 16:12:08.820543: step: 1204/1541, loss: 0.000513747741933912 2023-01-23 16:12:09.653701: step: 1206/1541, loss: 0.08627934753894806 2023-01-23 16:12:10.551916: step: 1208/1541, loss: 0.0007518015918321908 2023-01-23 16:12:11.433669: step: 1210/1541, loss: 0.0030042356811463833 2023-01-23 16:12:12.351470: step: 1212/1541, loss: 0.07010888308286667 2023-01-23 16:12:13.297705: step: 1214/1541, loss: 0.10532966256141663 2023-01-23 16:12:14.164473: step: 1216/1541, loss: 0.003330796957015991 2023-01-23 16:12:15.056847: step: 1218/1541, loss: 0.004878015257418156 2023-01-23 16:12:15.945663: step: 1220/1541, loss: 0.15508221089839935 2023-01-23 16:12:16.867038: step: 1222/1541, loss: 0.0009897429263219237 2023-01-23 16:12:17.733313: step: 1224/1541, loss: 0.00048371226876042783 2023-01-23 16:12:18.599062: step: 1226/1541, loss: 0.050592608749866486 2023-01-23 16:12:19.505621: step: 1228/1541, loss: 0.00851343758404255 2023-01-23 16:12:20.385077: step: 1230/1541, loss: 0.1817602515220642 2023-01-23 16:12:21.281963: step: 1232/1541, loss: 0.0011655696434900165 2023-01-23 16:12:22.235522: step: 1234/1541, loss: 0.33265841007232666 2023-01-23 16:12:23.103703: step: 1236/1541, loss: 0.1714312732219696 2023-01-23 16:12:24.006271: step: 1238/1541, loss: 0.013510337099432945 2023-01-23 16:12:24.907543: step: 1240/1541, loss: 0.005204334389418364 2023-01-23 16:12:25.849586: step: 1242/1541, loss: 0.0056030903942883015 2023-01-23 16:12:26.743542: step: 1244/1541, loss: 0.3609674870967865 2023-01-23 16:12:27.627034: step: 1246/1541, loss: 0.009581071324646473 2023-01-23 16:12:28.539221: step: 1248/1541, loss: 0.006014281418174505 2023-01-23 16:12:29.422321: step: 1250/1541, loss: 6.031828888808377e-05 2023-01-23 16:12:30.316761: step: 1252/1541, loss: 0.0001347097277175635 2023-01-23 16:12:31.220644: step: 1254/1541, loss: 0.0019946806132793427 2023-01-23 16:12:32.158020: step: 1256/1541, loss: 0.3312469720840454 2023-01-23 16:12:33.017468: step: 1258/1541, loss: 0.15799318253993988 2023-01-23 16:12:33.944076: step: 1260/1541, loss: 1.0173622285947204e-05 2023-01-23 16:12:34.832771: step: 1262/1541, loss: 0.46256116032600403 2023-01-23 16:12:35.695800: step: 1264/1541, loss: 0.08675694465637207 2023-01-23 16:12:36.554482: step: 1266/1541, loss: 9.0413996076677e-05 2023-01-23 16:12:37.468261: step: 1268/1541, loss: 0.18615466356277466 2023-01-23 16:12:38.418302: step: 1270/1541, loss: 0.06547695398330688 2023-01-23 16:12:39.340409: step: 1272/1541, loss: 0.0006316672661341727 2023-01-23 16:12:40.221489: step: 1274/1541, loss: 0.009825481101870537 2023-01-23 16:12:41.094315: step: 1276/1541, loss: 0.0007027123938314617 2023-01-23 16:12:41.947353: step: 1278/1541, loss: 0.10780888795852661 2023-01-23 16:12:42.870192: step: 1280/1541, loss: 0.11012709140777588 2023-01-23 16:12:43.789796: step: 1282/1541, loss: 0.04277149587869644 2023-01-23 16:12:44.675884: step: 1284/1541, loss: 0.0008536550449207425 2023-01-23 16:12:45.559438: step: 1286/1541, loss: 0.007883744314312935 2023-01-23 16:12:46.468116: step: 1288/1541, loss: 0.26773950457572937 2023-01-23 16:12:47.345736: step: 1290/1541, loss: 0.0426887646317482 2023-01-23 16:12:48.239621: step: 1292/1541, loss: 0.1817283034324646 2023-01-23 16:12:49.089954: step: 1294/1541, loss: 0.0027145310305058956 2023-01-23 16:12:50.003211: step: 1296/1541, loss: 0.4261266887187958 2023-01-23 16:12:50.944441: step: 1298/1541, loss: 0.06456069648265839 2023-01-23 16:12:51.841971: step: 1300/1541, loss: 0.0002596902777440846 2023-01-23 16:12:52.684400: step: 1302/1541, loss: 0.010355011560022831 2023-01-23 16:12:53.558161: step: 1304/1541, loss: 0.0001249109482159838 2023-01-23 16:12:54.514406: step: 1306/1541, loss: 0.0034264535643160343 2023-01-23 16:12:55.453317: step: 1308/1541, loss: 0.0013655014336109161 2023-01-23 16:12:56.424272: step: 1310/1541, loss: 0.04590776562690735 2023-01-23 16:12:57.387133: step: 1312/1541, loss: 0.037107739597558975 2023-01-23 16:12:58.262763: step: 1314/1541, loss: 0.003979336936026812 2023-01-23 16:12:59.203625: step: 1316/1541, loss: 0.0048738326877355576 2023-01-23 16:13:00.098265: step: 1318/1541, loss: 0.06300068646669388 2023-01-23 16:13:00.944323: step: 1320/1541, loss: 0.0018277987837791443 2023-01-23 16:13:01.818482: step: 1322/1541, loss: 0.001694437232799828 2023-01-23 16:13:02.740902: step: 1324/1541, loss: 0.012872934341430664 2023-01-23 16:13:03.626617: step: 1326/1541, loss: 0.10328478366136551 2023-01-23 16:13:04.544908: step: 1328/1541, loss: 0.4245944023132324 2023-01-23 16:13:05.399576: step: 1330/1541, loss: 0.0002706425730139017 2023-01-23 16:13:06.288692: step: 1332/1541, loss: 0.16558478772640228 2023-01-23 16:13:07.245266: step: 1334/1541, loss: 0.0006318551022559404 2023-01-23 16:13:08.146751: step: 1336/1541, loss: 0.041855208575725555 2023-01-23 16:13:09.087565: step: 1338/1541, loss: 0.03814348950982094 2023-01-23 16:13:09.933568: step: 1340/1541, loss: 0.030688529834151268 2023-01-23 16:13:10.808087: step: 1342/1541, loss: 0.00022487582464236766 2023-01-23 16:13:11.710086: step: 1344/1541, loss: 0.24878717958927155 2023-01-23 16:13:12.607241: step: 1346/1541, loss: 0.005938761867582798 2023-01-23 16:13:13.479398: step: 1348/1541, loss: 0.348216712474823 2023-01-23 16:13:14.370578: step: 1350/1541, loss: 0.009935996495187283 2023-01-23 16:13:15.357076: step: 1352/1541, loss: 0.18234911561012268 2023-01-23 16:13:16.223154: step: 1354/1541, loss: 0.0003501418395899236 2023-01-23 16:13:17.097431: step: 1356/1541, loss: 0.005590563639998436 2023-01-23 16:13:18.000215: step: 1358/1541, loss: 0.35026928782463074 2023-01-23 16:13:18.937343: step: 1360/1541, loss: 0.030477549880743027 2023-01-23 16:13:19.899597: step: 1362/1541, loss: 0.027088314294815063 2023-01-23 16:13:20.805920: step: 1364/1541, loss: 0.00017605577886570245 2023-01-23 16:13:21.695427: step: 1366/1541, loss: 0.002851080382242799 2023-01-23 16:13:22.643690: step: 1368/1541, loss: 0.0856199562549591 2023-01-23 16:13:23.520476: step: 1370/1541, loss: 0.001472927164286375 2023-01-23 16:13:24.413433: step: 1372/1541, loss: 0.0018981668399646878 2023-01-23 16:13:25.245744: step: 1374/1541, loss: 0.0004229082551319152 2023-01-23 16:13:26.117324: step: 1376/1541, loss: 0.24993854761123657 2023-01-23 16:13:26.962920: step: 1378/1541, loss: 0.024462658911943436 2023-01-23 16:13:28.040564: step: 1380/1541, loss: 0.08051619678735733 2023-01-23 16:13:28.921027: step: 1382/1541, loss: 0.008885577321052551 2023-01-23 16:13:29.789954: step: 1384/1541, loss: 0.0028455466963350773 2023-01-23 16:13:30.681102: step: 1386/1541, loss: 0.0008978600380942225 2023-01-23 16:13:31.532449: step: 1388/1541, loss: 0.0006771368789486587 2023-01-23 16:13:32.493034: step: 1390/1541, loss: 0.012796085327863693 2023-01-23 16:13:33.392831: step: 1392/1541, loss: 0.002127978717908263 2023-01-23 16:13:34.358663: step: 1394/1541, loss: 0.004070867784321308 2023-01-23 16:13:35.239514: step: 1396/1541, loss: 0.00046063860645517707 2023-01-23 16:13:36.129874: step: 1398/1541, loss: 0.003908259328454733 2023-01-23 16:13:37.004116: step: 1400/1541, loss: 0.002498792950063944 2023-01-23 16:13:37.871012: step: 1402/1541, loss: 0.001728883944451809 2023-01-23 16:13:38.762698: step: 1404/1541, loss: 0.0022127418778836727 2023-01-23 16:13:39.634664: step: 1406/1541, loss: 0.040435902774333954 2023-01-23 16:13:40.516464: step: 1408/1541, loss: 0.4566621780395508 2023-01-23 16:13:41.413588: step: 1410/1541, loss: 0.00021340255625545979 2023-01-23 16:13:42.341750: step: 1412/1541, loss: 0.016789376735687256 2023-01-23 16:13:43.213397: step: 1414/1541, loss: 0.003972717095166445 2023-01-23 16:13:44.090632: step: 1416/1541, loss: 0.30482375621795654 2023-01-23 16:13:45.082347: step: 1418/1541, loss: 0.020418589934706688 2023-01-23 16:13:45.950944: step: 1420/1541, loss: 3.6709392588818446e-05 2023-01-23 16:13:46.809127: step: 1422/1541, loss: 0.21335408091545105 2023-01-23 16:13:47.687815: step: 1424/1541, loss: 0.12949666380882263 2023-01-23 16:13:48.556805: step: 1426/1541, loss: 6.90963352099061e-05 2023-01-23 16:13:49.442525: step: 1428/1541, loss: 0.002733791945502162 2023-01-23 16:13:50.359289: step: 1430/1541, loss: 0.000764347380027175 2023-01-23 16:13:51.254614: step: 1432/1541, loss: 0.002450188621878624 2023-01-23 16:13:52.163282: step: 1434/1541, loss: -1.109707955038175e-05 2023-01-23 16:13:53.039202: step: 1436/1541, loss: 0.0008283150964416564 2023-01-23 16:13:53.946971: step: 1438/1541, loss: 0.006771976128220558 2023-01-23 16:13:54.865879: step: 1440/1541, loss: 0.018786905333399773 2023-01-23 16:13:55.770381: step: 1442/1541, loss: 0.0010466427775099874 2023-01-23 16:13:56.897477: step: 1444/1541, loss: 0.0020071966573596 2023-01-23 16:13:57.796353: step: 1446/1541, loss: 0.16907131671905518 2023-01-23 16:13:58.721949: step: 1448/1541, loss: 0.07544972747564316 2023-01-23 16:13:59.601874: step: 1450/1541, loss: 0.00034290156327188015 2023-01-23 16:14:00.506313: step: 1452/1541, loss: 0.04269809275865555 2023-01-23 16:14:01.381789: step: 1454/1541, loss: 0.058260589838027954 2023-01-23 16:14:02.292065: step: 1456/1541, loss: 0.001246592728421092 2023-01-23 16:14:03.179047: step: 1458/1541, loss: 0.002798772184178233 2023-01-23 16:14:04.066140: step: 1460/1541, loss: -6.903711437189486e-06 2023-01-23 16:14:05.002222: step: 1462/1541, loss: 0.0001138789375545457 2023-01-23 16:14:05.922291: step: 1464/1541, loss: 0.030344849452376366 2023-01-23 16:14:06.827292: step: 1466/1541, loss: 0.009427797980606556 2023-01-23 16:14:07.703963: step: 1468/1541, loss: 0.10924039036035538 2023-01-23 16:14:08.601799: step: 1470/1541, loss: 0.0029092724435031414 2023-01-23 16:14:09.512217: step: 1472/1541, loss: 0.06889624893665314 2023-01-23 16:14:10.356414: step: 1474/1541, loss: 0.0034886798821389675 2023-01-23 16:14:11.287352: step: 1476/1541, loss: 0.12147391587495804 2023-01-23 16:14:12.194947: step: 1478/1541, loss: 0.13286063075065613 2023-01-23 16:14:13.114021: step: 1480/1541, loss: 0.0026310954708606005 2023-01-23 16:14:14.002060: step: 1482/1541, loss: 0.2118302881717682 2023-01-23 16:14:14.867443: step: 1484/1541, loss: 0.003022467950358987 2023-01-23 16:14:15.776129: step: 1486/1541, loss: 1.0953564643859863 2023-01-23 16:14:16.681927: step: 1488/1541, loss: 0.017662320286035538 2023-01-23 16:14:17.590223: step: 1490/1541, loss: 0.7273978590965271 2023-01-23 16:14:18.476924: step: 1492/1541, loss: 0.0011918330565094948 2023-01-23 16:14:19.332689: step: 1494/1541, loss: 0.0010502400109544396 2023-01-23 16:14:20.181905: step: 1496/1541, loss: 0.35001397132873535 2023-01-23 16:14:21.061106: step: 1498/1541, loss: 0.008525111712515354 2023-01-23 16:14:21.984027: step: 1500/1541, loss: 0.003898892318829894 2023-01-23 16:14:22.853419: step: 1502/1541, loss: 0.23187030851840973 2023-01-23 16:14:23.802159: step: 1504/1541, loss: 0.0065617430955171585 2023-01-23 16:14:24.646544: step: 1506/1541, loss: 0.0250555332750082 2023-01-23 16:14:25.578402: step: 1508/1541, loss: 0.2788282632827759 2023-01-23 16:14:26.444760: step: 1510/1541, loss: 0.13651058077812195 2023-01-23 16:14:27.337626: step: 1512/1541, loss: 0.044584304094314575 2023-01-23 16:14:28.266114: step: 1514/1541, loss: 0.003936761524528265 2023-01-23 16:14:29.130122: step: 1516/1541, loss: 0.1975756287574768 2023-01-23 16:14:30.060523: step: 1518/1541, loss: 0.00115380494389683 2023-01-23 16:14:30.986991: step: 1520/1541, loss: 0.0776849240064621 2023-01-23 16:14:31.906847: step: 1522/1541, loss: 0.02756173349916935 2023-01-23 16:14:32.809530: step: 1524/1541, loss: 0.009345464408397675 2023-01-23 16:14:33.727039: step: 1526/1541, loss: 0.017725087702274323 2023-01-23 16:14:34.610335: step: 1528/1541, loss: 0.00033039358095265925 2023-01-23 16:14:35.494412: step: 1530/1541, loss: 0.02210170403122902 2023-01-23 16:14:36.372222: step: 1532/1541, loss: 0.0038755214773118496 2023-01-23 16:14:37.352192: step: 1534/1541, loss: 0.056416045874357224 2023-01-23 16:14:38.249272: step: 1536/1541, loss: 0.04665571451187134 2023-01-23 16:14:39.134512: step: 1538/1541, loss: 0.24631573259830475 2023-01-23 16:14:40.022363: step: 1540/1541, loss: 0.004532587248831987 2023-01-23 16:14:40.904192: step: 1542/1541, loss: 0.33464616537094116 2023-01-23 16:14:41.836975: step: 1544/1541, loss: 0.22955802083015442 2023-01-23 16:14:42.731449: step: 1546/1541, loss: 0.07258206605911255 2023-01-23 16:14:43.605804: step: 1548/1541, loss: 0.0037180830258876085 2023-01-23 16:14:44.495173: step: 1550/1541, loss: 0.020814621821045876 2023-01-23 16:14:45.406287: step: 1552/1541, loss: 0.0038225376047194004 2023-01-23 16:14:46.362950: step: 1554/1541, loss: 0.026757463812828064 2023-01-23 16:14:47.269711: step: 1556/1541, loss: 0.010527378879487514 2023-01-23 16:14:48.118169: step: 1558/1541, loss: 0.0029111874755471945 2023-01-23 16:14:49.008158: step: 1560/1541, loss: 0.0012252897722646594 2023-01-23 16:14:49.886641: step: 1562/1541, loss: 0.0009097559959627688 2023-01-23 16:14:50.747302: step: 1564/1541, loss: 0.10902539640665054 2023-01-23 16:14:51.628564: step: 1566/1541, loss: 0.004363501910120249 2023-01-23 16:14:52.541506: step: 1568/1541, loss: 0.001398898079060018 2023-01-23 16:14:53.438965: step: 1570/1541, loss: 0.2993890345096588 2023-01-23 16:14:54.297145: step: 1572/1541, loss: 0.13131380081176758 2023-01-23 16:14:55.171136: step: 1574/1541, loss: 0.0004725162289105356 2023-01-23 16:14:56.042271: step: 1576/1541, loss: 0.1696934700012207 2023-01-23 16:14:56.939942: step: 1578/1541, loss: 0.003925052937120199 2023-01-23 16:14:57.870376: step: 1580/1541, loss: 0.036587126553058624 2023-01-23 16:14:58.787192: step: 1582/1541, loss: 0.009249718859791756 2023-01-23 16:14:59.726634: step: 1584/1541, loss: 0.1998455673456192 2023-01-23 16:15:00.625826: step: 1586/1541, loss: 0.004275589715689421 2023-01-23 16:15:01.497298: step: 1588/1541, loss: 0.09523876011371613 2023-01-23 16:15:02.363467: step: 1590/1541, loss: 0.02292303740978241 2023-01-23 16:15:03.290866: step: 1592/1541, loss: 0.011624692939221859 2023-01-23 16:15:04.173890: step: 1594/1541, loss: 0.007743715774267912 2023-01-23 16:15:05.094891: step: 1596/1541, loss: 0.021804532036185265 2023-01-23 16:15:05.977392: step: 1598/1541, loss: 0.0016849478706717491 2023-01-23 16:15:06.870502: step: 1600/1541, loss: 0.0856471061706543 2023-01-23 16:15:07.803307: step: 1602/1541, loss: 0.004041749984025955 2023-01-23 16:15:08.714503: step: 1604/1541, loss: 0.006396267097443342 2023-01-23 16:15:09.597844: step: 1606/1541, loss: 0.04319583997130394 2023-01-23 16:15:10.475835: step: 1608/1541, loss: 0.07449787110090256 2023-01-23 16:15:11.331091: step: 1610/1541, loss: 0.004560230299830437 2023-01-23 16:15:12.249370: step: 1612/1541, loss: 0.0022643390111625195 2023-01-23 16:15:13.139111: step: 1614/1541, loss: 0.051469169557094574 2023-01-23 16:15:14.033128: step: 1616/1541, loss: 0.0021591747645288706 2023-01-23 16:15:14.912692: step: 1618/1541, loss: 0.007721188478171825 2023-01-23 16:15:15.785676: step: 1620/1541, loss: 0.06768667697906494 2023-01-23 16:15:16.661314: step: 1622/1541, loss: 0.008450121618807316 2023-01-23 16:15:17.563303: step: 1624/1541, loss: 0.0056986501440405846 2023-01-23 16:15:18.439619: step: 1626/1541, loss: 0.008438916876912117 2023-01-23 16:15:19.333055: step: 1628/1541, loss: 9.965889330487698e-05 2023-01-23 16:15:20.259616: step: 1630/1541, loss: 0.010760671459138393 2023-01-23 16:15:21.138562: step: 1632/1541, loss: 0.022444481030106544 2023-01-23 16:15:22.055346: step: 1634/1541, loss: 0.04023589566349983 2023-01-23 16:15:22.987247: step: 1636/1541, loss: 0.015187248587608337 2023-01-23 16:15:23.900455: step: 1638/1541, loss: 0.02173675037920475 2023-01-23 16:15:24.791085: step: 1640/1541, loss: 0.0039172000251710415 2023-01-23 16:15:25.655327: step: 1642/1541, loss: 0.007539187092334032 2023-01-23 16:15:26.531440: step: 1644/1541, loss: 0.0881207287311554 2023-01-23 16:15:27.443265: step: 1646/1541, loss: 0.09080790728330612 2023-01-23 16:15:28.344827: step: 1648/1541, loss: 0.0744033008813858 2023-01-23 16:15:29.227368: step: 1650/1541, loss: 0.003952489234507084 2023-01-23 16:15:30.126469: step: 1652/1541, loss: 0.007263967767357826 2023-01-23 16:15:31.010455: step: 1654/1541, loss: 0.0009430086938664317 2023-01-23 16:15:31.947267: step: 1656/1541, loss: 0.47326937317848206 2023-01-23 16:15:32.813861: step: 1658/1541, loss: 0.0033834318164736032 2023-01-23 16:15:33.740576: step: 1660/1541, loss: 0.0020680679008364677 2023-01-23 16:15:34.608965: step: 1662/1541, loss: 0.004138059448450804 2023-01-23 16:15:35.505992: step: 1664/1541, loss: 0.10117846727371216 2023-01-23 16:15:36.559095: step: 1666/1541, loss: 0.03355439752340317 2023-01-23 16:15:37.447640: step: 1668/1541, loss: 0.10758700221776962 2023-01-23 16:15:38.316717: step: 1670/1541, loss: 0.00027828849852085114 2023-01-23 16:15:39.190563: step: 1672/1541, loss: 1.018561601638794 2023-01-23 16:15:40.036664: step: 1674/1541, loss: 0.17698946595191956 2023-01-23 16:15:40.939067: step: 1676/1541, loss: 0.005080834496766329 2023-01-23 16:15:41.849471: step: 1678/1541, loss: 0.8457561135292053 2023-01-23 16:15:42.789946: step: 1680/1541, loss: 0.13271404802799225 2023-01-23 16:15:43.690686: step: 1682/1541, loss: 0.018186576664447784 2023-01-23 16:15:44.564575: step: 1684/1541, loss: 0.0017581165302544832 2023-01-23 16:15:45.420902: step: 1686/1541, loss: 0.10581322014331818 2023-01-23 16:15:46.354181: step: 1688/1541, loss: 0.3346943259239197 2023-01-23 16:15:47.213541: step: 1690/1541, loss: 0.013778549619019032 2023-01-23 16:15:48.097168: step: 1692/1541, loss: 0.03480633720755577 2023-01-23 16:15:48.970681: step: 1694/1541, loss: 0.002246582182124257 2023-01-23 16:15:49.866777: step: 1696/1541, loss: 0.0012990555260330439 2023-01-23 16:15:50.750801: step: 1698/1541, loss: 0.23720870912075043 2023-01-23 16:15:51.637685: step: 1700/1541, loss: 0.000985097954981029 2023-01-23 16:15:52.563630: step: 1702/1541, loss: 0.0025198699440807104 2023-01-23 16:15:53.462137: step: 1704/1541, loss: 0.1516149491071701 2023-01-23 16:15:54.370860: step: 1706/1541, loss: 0.010300971567630768 2023-01-23 16:15:55.258929: step: 1708/1541, loss: 0.19931933283805847 2023-01-23 16:15:56.153412: step: 1710/1541, loss: 0.0012112560216337442 2023-01-23 16:15:57.041944: step: 1712/1541, loss: 0.009040025994181633 2023-01-23 16:15:57.967193: step: 1714/1541, loss: 7.486144750146195e-05 2023-01-23 16:15:58.855702: step: 1716/1541, loss: 0.004490125924348831 2023-01-23 16:15:59.730291: step: 1718/1541, loss: 0.14837966859340668 2023-01-23 16:16:00.635813: step: 1720/1541, loss: 0.005423865281045437 2023-01-23 16:16:01.579748: step: 1722/1541, loss: 0.0014456474455073476 2023-01-23 16:16:02.445999: step: 1724/1541, loss: 0.1150556430220604 2023-01-23 16:16:03.320300: step: 1726/1541, loss: 0.0038122169207781553 2023-01-23 16:16:04.231811: step: 1728/1541, loss: 0.013342360965907574 2023-01-23 16:16:05.152196: step: 1730/1541, loss: 0.1269872784614563 2023-01-23 16:16:06.028657: step: 1732/1541, loss: 1.4832143783569336 2023-01-23 16:16:06.905253: step: 1734/1541, loss: 0.07125843316316605 2023-01-23 16:16:07.821021: step: 1736/1541, loss: 0.00011771731078624725 2023-01-23 16:16:08.720254: step: 1738/1541, loss: 0.11775586754083633 2023-01-23 16:16:09.626287: step: 1740/1541, loss: 0.05779052525758743 2023-01-23 16:16:10.547023: step: 1742/1541, loss: 0.0002941243874374777 2023-01-23 16:16:11.529040: step: 1744/1541, loss: 0.03361677750945091 2023-01-23 16:16:12.381518: step: 1746/1541, loss: 0.09224256128072739 2023-01-23 16:16:13.270737: step: 1748/1541, loss: 0.03640473634004593 2023-01-23 16:16:14.148946: step: 1750/1541, loss: 0.0034195506013929844 2023-01-23 16:16:15.084825: step: 1752/1541, loss: 0.009512179531157017 2023-01-23 16:16:15.973243: step: 1754/1541, loss: 0.11864908039569855 2023-01-23 16:16:16.968522: step: 1756/1541, loss: 0.0014287387020885944 2023-01-23 16:16:17.864521: step: 1758/1541, loss: 0.13598571717739105 2023-01-23 16:16:18.818050: step: 1760/1541, loss: 0.0006228035199455917 2023-01-23 16:16:19.720811: step: 1762/1541, loss: 0.000615469238255173 2023-01-23 16:16:20.670950: step: 1764/1541, loss: 0.0002801896189339459 2023-01-23 16:16:21.537650: step: 1766/1541, loss: 0.00013155040505807847 2023-01-23 16:16:22.394906: step: 1768/1541, loss: 0.04702894017100334 2023-01-23 16:16:23.313406: step: 1770/1541, loss: 0.22116921842098236 2023-01-23 16:16:24.247074: step: 1772/1541, loss: 0.12689001858234406 2023-01-23 16:16:25.156116: step: 1774/1541, loss: 8.324253576574847e-05 2023-01-23 16:16:26.097573: step: 1776/1541, loss: 0.3400750458240509 2023-01-23 16:16:26.999225: step: 1778/1541, loss: 0.012785175815224648 2023-01-23 16:16:27.894647: step: 1780/1541, loss: 0.0022713495418429375 2023-01-23 16:16:28.770557: step: 1782/1541, loss: 0.21145637333393097 2023-01-23 16:16:29.660865: step: 1784/1541, loss: 0.0008123167790472507 2023-01-23 16:16:30.508788: step: 1786/1541, loss: 0.00033412958146072924 2023-01-23 16:16:31.418777: step: 1788/1541, loss: 0.00010623971320455894 2023-01-23 16:16:32.361560: step: 1790/1541, loss: 0.047350380569696426 2023-01-23 16:16:33.248058: step: 1792/1541, loss: 0.17483016848564148 2023-01-23 16:16:34.140864: step: 1794/1541, loss: 0.09261057525873184 2023-01-23 16:16:35.070854: step: 1796/1541, loss: 0.20035971701145172 2023-01-23 16:16:35.982578: step: 1798/1541, loss: 0.07313676178455353 2023-01-23 16:16:36.914527: step: 1800/1541, loss: 0.10479622334241867 2023-01-23 16:16:37.806071: step: 1802/1541, loss: 0.004841729532927275 2023-01-23 16:16:38.696945: step: 1804/1541, loss: 0.051049839705228806 2023-01-23 16:16:39.636565: step: 1806/1541, loss: 0.04711150377988815 2023-01-23 16:16:40.497420: step: 1808/1541, loss: 0.06168263033032417 2023-01-23 16:16:41.393605: step: 1810/1541, loss: 0.0164113100618124 2023-01-23 16:16:42.282058: step: 1812/1541, loss: 0.006849855184555054 2023-01-23 16:16:43.189860: step: 1814/1541, loss: 0.012914864346385002 2023-01-23 16:16:44.072808: step: 1816/1541, loss: 0.0011431080056354403 2023-01-23 16:16:44.981462: step: 1818/1541, loss: 0.02393905445933342 2023-01-23 16:16:45.865896: step: 1820/1541, loss: 0.03128289058804512 2023-01-23 16:16:46.741961: step: 1822/1541, loss: 0.12009134888648987 2023-01-23 16:16:47.674297: step: 1824/1541, loss: 0.20750956237316132 2023-01-23 16:16:48.608852: step: 1826/1541, loss: 0.010653538629412651 2023-01-23 16:16:49.534251: step: 1828/1541, loss: 0.0080099543556571 2023-01-23 16:16:50.412542: step: 1830/1541, loss: 0.05140306055545807 2023-01-23 16:16:51.305021: step: 1832/1541, loss: 0.08533966541290283 2023-01-23 16:16:52.180113: step: 1834/1541, loss: 0.09030717611312866 2023-01-23 16:16:53.060176: step: 1836/1541, loss: 0.046739362180233 2023-01-23 16:16:54.043544: step: 1838/1541, loss: 0.17386429011821747 2023-01-23 16:16:54.967733: step: 1840/1541, loss: 0.08243395388126373 2023-01-23 16:16:55.879701: step: 1842/1541, loss: 0.00020221882732585073 2023-01-23 16:16:56.742749: step: 1844/1541, loss: 0.03240804746747017 2023-01-23 16:16:57.624166: step: 1846/1541, loss: 0.17442195117473602 2023-01-23 16:16:58.529304: step: 1848/1541, loss: 0.12695318460464478 2023-01-23 16:16:59.412337: step: 1850/1541, loss: 0.14963014423847198 2023-01-23 16:17:00.298968: step: 1852/1541, loss: 0.012394032441079617 2023-01-23 16:17:01.185833: step: 1854/1541, loss: 0.0003930046223104 2023-01-23 16:17:02.165045: step: 1856/1541, loss: 0.3157147467136383 2023-01-23 16:17:03.121022: step: 1858/1541, loss: 0.003248288994655013 2023-01-23 16:17:04.019683: step: 1860/1541, loss: 0.33713147044181824 2023-01-23 16:17:04.915590: step: 1862/1541, loss: 0.05483434721827507 2023-01-23 16:17:05.785923: step: 1864/1541, loss: 0.26733341813087463 2023-01-23 16:17:06.701876: step: 1866/1541, loss: 0.030624788254499435 2023-01-23 16:17:07.619718: step: 1868/1541, loss: 0.012399045750498772 2023-01-23 16:17:08.524872: step: 1870/1541, loss: 0.0048342179507017136 2023-01-23 16:17:09.458912: step: 1872/1541, loss: 0.00500990217551589 2023-01-23 16:17:10.302463: step: 1874/1541, loss: 0.028114188462495804 2023-01-23 16:17:11.225991: step: 1876/1541, loss: 0.006117434706538916 2023-01-23 16:17:12.099335: step: 1878/1541, loss: 0.0025220653042197227 2023-01-23 16:17:12.967586: step: 1880/1541, loss: 0.014881118200719357 2023-01-23 16:17:13.866316: step: 1882/1541, loss: 1.4208255379344337e-05 2023-01-23 16:17:14.782316: step: 1884/1541, loss: 0.32584235072135925 2023-01-23 16:17:15.658026: step: 1886/1541, loss: 0.060959719121456146 2023-01-23 16:17:16.514747: step: 1888/1541, loss: 0.0013391884276643395 2023-01-23 16:17:17.427281: step: 1890/1541, loss: 0.018938690423965454 2023-01-23 16:17:18.333707: step: 1892/1541, loss: 0.02693011984229088 2023-01-23 16:17:19.211892: step: 1894/1541, loss: 0.0005715157603845 2023-01-23 16:17:20.059655: step: 1896/1541, loss: 0.026119230315089226 2023-01-23 16:17:20.905219: step: 1898/1541, loss: 0.00903965812176466 2023-01-23 16:17:21.797336: step: 1900/1541, loss: 0.00026144241564907134 2023-01-23 16:17:22.726351: step: 1902/1541, loss: 0.018692132085561752 2023-01-23 16:17:23.674234: step: 1904/1541, loss: 0.0005770477582700551 2023-01-23 16:17:24.554642: step: 1906/1541, loss: 0.05090475082397461 2023-01-23 16:17:25.439544: step: 1908/1541, loss: 0.0020474118646234274 2023-01-23 16:17:26.322298: step: 1910/1541, loss: 0.006228559650480747 2023-01-23 16:17:27.289753: step: 1912/1541, loss: 0.09517434239387512 2023-01-23 16:17:28.173571: step: 1914/1541, loss: 0.07169400900602341 2023-01-23 16:17:29.122134: step: 1916/1541, loss: 0.09912168234586716 2023-01-23 16:17:29.989825: step: 1918/1541, loss: 0.02059830166399479 2023-01-23 16:17:30.874760: step: 1920/1541, loss: 0.0001754827972035855 2023-01-23 16:17:31.759437: step: 1922/1541, loss: 0.3672923743724823 2023-01-23 16:17:32.656871: step: 1924/1541, loss: 0.00010737471893662587 2023-01-23 16:17:33.549775: step: 1926/1541, loss: 0.0004090723523404449 2023-01-23 16:17:34.444133: step: 1928/1541, loss: 0.02330506592988968 2023-01-23 16:17:35.334170: step: 1930/1541, loss: 0.009716635569930077 2023-01-23 16:17:36.197947: step: 1932/1541, loss: 0.006609790027141571 2023-01-23 16:17:37.090781: step: 1934/1541, loss: 0.140544131398201 2023-01-23 16:17:37.988209: step: 1936/1541, loss: 0.20063789188861847 2023-01-23 16:17:38.879317: step: 1938/1541, loss: 0.0030554854311048985 2023-01-23 16:17:39.780271: step: 1940/1541, loss: 0.0033271508291363716 2023-01-23 16:17:40.692828: step: 1942/1541, loss: 0.27054262161254883 2023-01-23 16:17:41.618705: step: 1944/1541, loss: 0.01502346433699131 2023-01-23 16:17:42.494284: step: 1946/1541, loss: 0.0029213204979896545 2023-01-23 16:17:43.366156: step: 1948/1541, loss: 0.004365513566881418 2023-01-23 16:17:44.302234: step: 1950/1541, loss: 0.05084135755896568 2023-01-23 16:17:45.363211: step: 1952/1541, loss: 0.0012801728444173932 2023-01-23 16:17:46.243324: step: 1954/1541, loss: 0.03675076365470886 2023-01-23 16:17:47.185308: step: 1956/1541, loss: 0.01021081954240799 2023-01-23 16:17:48.062467: step: 1958/1541, loss: 0.0007243193103931844 2023-01-23 16:17:48.936235: step: 1960/1541, loss: 0.2528841197490692 2023-01-23 16:17:49.920237: step: 1962/1541, loss: 0.010754133574664593 2023-01-23 16:17:50.841639: step: 1964/1541, loss: 0.008881663903594017 2023-01-23 16:17:51.699662: step: 1966/1541, loss: 0.054775919765233994 2023-01-23 16:17:52.586070: step: 1968/1541, loss: 0.0012942742323502898 2023-01-23 16:17:53.500460: step: 1970/1541, loss: 0.0014228058280423284 2023-01-23 16:17:54.459464: step: 1972/1541, loss: 0.002162178047001362 2023-01-23 16:17:55.356077: step: 1974/1541, loss: 0.06781264394521713 2023-01-23 16:17:56.227731: step: 1976/1541, loss: 0.007778427563607693 2023-01-23 16:17:57.167399: step: 1978/1541, loss: 0.13464370369911194 2023-01-23 16:17:58.098519: step: 1980/1541, loss: 0.03923698887228966 2023-01-23 16:17:59.045299: step: 1982/1541, loss: 0.0033491712529212236 2023-01-23 16:17:59.983726: step: 1984/1541, loss: 0.16437305510044098 2023-01-23 16:18:00.828382: step: 1986/1541, loss: 0.0017276639118790627 2023-01-23 16:18:01.786055: step: 1988/1541, loss: 0.2727334201335907 2023-01-23 16:18:02.670764: step: 1990/1541, loss: 0.029305225238204002 2023-01-23 16:18:03.558853: step: 1992/1541, loss: 0.003610791638493538 2023-01-23 16:18:04.472102: step: 1994/1541, loss: 0.036173444241285324 2023-01-23 16:18:05.338139: step: 1996/1541, loss: 0.0016606334829702973 2023-01-23 16:18:06.197530: step: 1998/1541, loss: 0.0009216971811838448 2023-01-23 16:18:07.121280: step: 2000/1541, loss: 0.040129855275154114 2023-01-23 16:18:08.068556: step: 2002/1541, loss: 0.03729122132062912 2023-01-23 16:18:08.989211: step: 2004/1541, loss: 0.012001561000943184 2023-01-23 16:18:09.882874: step: 2006/1541, loss: 0.021341638639569283 2023-01-23 16:18:10.776119: step: 2008/1541, loss: 0.0071428921073675156 2023-01-23 16:18:11.626354: step: 2010/1541, loss: 0.005384420044720173 2023-01-23 16:18:12.522092: step: 2012/1541, loss: 0.004142774734646082 2023-01-23 16:18:13.379592: step: 2014/1541, loss: 0.0036005948204547167 2023-01-23 16:18:14.281331: step: 2016/1541, loss: 0.039213962852954865 2023-01-23 16:18:15.233843: step: 2018/1541, loss: 0.013449314050376415 2023-01-23 16:18:16.200094: step: 2020/1541, loss: 2.159501791000366 2023-01-23 16:18:17.142461: step: 2022/1541, loss: 0.5525521636009216 2023-01-23 16:18:18.097355: step: 2024/1541, loss: 0.006929566152393818 2023-01-23 16:18:18.981140: step: 2026/1541, loss: 0.04623976722359657 2023-01-23 16:18:19.855271: step: 2028/1541, loss: 0.10548733174800873 2023-01-23 16:18:20.739707: step: 2030/1541, loss: 0.00406384002417326 2023-01-23 16:18:21.685351: step: 2032/1541, loss: 0.02495254948735237 2023-01-23 16:18:22.575498: step: 2034/1541, loss: 0.009558668360114098 2023-01-23 16:18:23.468366: step: 2036/1541, loss: 0.0006271437741816044 2023-01-23 16:18:24.367175: step: 2038/1541, loss: 0.0007826838409528136 2023-01-23 16:18:25.307376: step: 2040/1541, loss: 0.007687271572649479 2023-01-23 16:18:26.209545: step: 2042/1541, loss: 0.004696606192737818 2023-01-23 16:18:27.196007: step: 2044/1541, loss: 0.0024834878277033567 2023-01-23 16:18:28.115730: step: 2046/1541, loss: 0.04555005952715874 2023-01-23 16:18:29.030681: step: 2048/1541, loss: 0.2213512510061264 2023-01-23 16:18:29.915179: step: 2050/1541, loss: 0.000628152396529913 2023-01-23 16:18:30.816225: step: 2052/1541, loss: 0.0021976004354655743 2023-01-23 16:18:31.719487: step: 2054/1541, loss: -1.609204446140211e-05 2023-01-23 16:18:32.613061: step: 2056/1541, loss: 0.0002105171442963183 2023-01-23 16:18:33.499653: step: 2058/1541, loss: 0.005994438659399748 2023-01-23 16:18:34.391963: step: 2060/1541, loss: 0.023106224834918976 2023-01-23 16:18:35.338761: step: 2062/1541, loss: 0.3964228332042694 2023-01-23 16:18:36.223697: step: 2064/1541, loss: 0.04736882448196411 2023-01-23 16:18:37.124113: step: 2066/1541, loss: 9.02121391845867e-05 2023-01-23 16:18:38.021606: step: 2068/1541, loss: 0.0014579003909602761 2023-01-23 16:18:38.921638: step: 2070/1541, loss: 0.022805791348218918 2023-01-23 16:18:39.811003: step: 2072/1541, loss: 0.0014558746479451656 2023-01-23 16:18:40.681349: step: 2074/1541, loss: 0.15871694684028625 2023-01-23 16:18:41.574629: step: 2076/1541, loss: 0.012597967870533466 2023-01-23 16:18:42.488458: step: 2078/1541, loss: 0.004098273813724518 2023-01-23 16:18:43.369185: step: 2080/1541, loss: 9.815796511247754e-05 2023-01-23 16:18:44.238498: step: 2082/1541, loss: 2.9335153158172034e-05 2023-01-23 16:18:45.130638: step: 2084/1541, loss: 0.01848471164703369 2023-01-23 16:18:46.011855: step: 2086/1541, loss: 0.0014448607107624412 2023-01-23 16:18:46.918847: step: 2088/1541, loss: 0.03664272278547287 2023-01-23 16:18:47.799709: step: 2090/1541, loss: 0.00826239213347435 2023-01-23 16:18:48.749428: step: 2092/1541, loss: 0.09343115985393524 2023-01-23 16:18:49.603599: step: 2094/1541, loss: 5.288077227305621e-05 2023-01-23 16:18:50.512818: step: 2096/1541, loss: 0.009341344237327576 2023-01-23 16:18:51.462393: step: 2098/1541, loss: 0.0038783052004873753 2023-01-23 16:18:52.431679: step: 2100/1541, loss: 0.0005896254442632198 2023-01-23 16:18:53.376116: step: 2102/1541, loss: 0.0007690942147746682 2023-01-23 16:18:54.360765: step: 2104/1541, loss: 0.5557559728622437 2023-01-23 16:18:55.245371: step: 2106/1541, loss: 0.001902962801977992 2023-01-23 16:18:56.175029: step: 2108/1541, loss: 0.12101547420024872 2023-01-23 16:18:57.088748: step: 2110/1541, loss: 0.06752476841211319 2023-01-23 16:18:57.959647: step: 2112/1541, loss: 0.021053919568657875 2023-01-23 16:18:58.846979: step: 2114/1541, loss: 0.00031655613565817475 2023-01-23 16:18:59.722618: step: 2116/1541, loss: 0.025837762281298637 2023-01-23 16:19:00.633754: step: 2118/1541, loss: 0.008672723546624184 2023-01-23 16:19:01.534235: step: 2120/1541, loss: 0.017177453264594078 2023-01-23 16:19:02.444377: step: 2122/1541, loss: 0.0014338798355311155 2023-01-23 16:19:03.318336: step: 2124/1541, loss: 0.0176086463034153 2023-01-23 16:19:04.212070: step: 2126/1541, loss: 0.06613974273204803 2023-01-23 16:19:05.117287: step: 2128/1541, loss: 0.12089735269546509 2023-01-23 16:19:06.041861: step: 2130/1541, loss: 0.002096258569508791 2023-01-23 16:19:06.953887: step: 2132/1541, loss: 0.002952082082629204 2023-01-23 16:19:07.815051: step: 2134/1541, loss: 0.00019758392591029406 2023-01-23 16:19:08.669590: step: 2136/1541, loss: 0.05617782846093178 2023-01-23 16:19:09.582713: step: 2138/1541, loss: 6.18605554336682e-05 2023-01-23 16:19:10.467067: step: 2140/1541, loss: 0.0030531417578458786 2023-01-23 16:19:11.352974: step: 2142/1541, loss: 0.05225945636630058 2023-01-23 16:19:12.245458: step: 2144/1541, loss: 0.0008465344435535371 2023-01-23 16:19:13.149846: step: 2146/1541, loss: 0.13909967243671417 2023-01-23 16:19:14.124711: step: 2148/1541, loss: 1.030033826828003 2023-01-23 16:19:15.022063: step: 2150/1541, loss: 0.014789166860282421 2023-01-23 16:19:15.949208: step: 2152/1541, loss: 0.0015660417266190052 2023-01-23 16:19:16.887544: step: 2154/1541, loss: 0.6544861793518066 2023-01-23 16:19:17.790802: step: 2156/1541, loss: 0.32510778307914734 2023-01-23 16:19:18.666886: step: 2158/1541, loss: 0.5586392283439636 2023-01-23 16:19:19.539446: step: 2160/1541, loss: 0.07150492072105408 2023-01-23 16:19:20.517884: step: 2162/1541, loss: 0.018854184076189995 2023-01-23 16:19:21.409679: step: 2164/1541, loss: 0.8751965761184692 2023-01-23 16:19:22.362267: step: 2166/1541, loss: 0.27821505069732666 2023-01-23 16:19:23.234853: step: 2168/1541, loss: 0.18594811856746674 2023-01-23 16:19:24.128073: step: 2170/1541, loss: 0.0002915413351729512 2023-01-23 16:19:25.029878: step: 2172/1541, loss: 0.060170453041791916 2023-01-23 16:19:25.956535: step: 2174/1541, loss: 4.488695412874222e-05 2023-01-23 16:19:26.900030: step: 2176/1541, loss: 0.05247395485639572 2023-01-23 16:19:27.854059: step: 2178/1541, loss: 4.109402289032005e-05 2023-01-23 16:19:28.742400: step: 2180/1541, loss: 0.21647140383720398 2023-01-23 16:19:29.638283: step: 2182/1541, loss: 0.004562120418995619 2023-01-23 16:19:30.515229: step: 2184/1541, loss: 0.0005933648208156228 2023-01-23 16:19:31.424588: step: 2186/1541, loss: 0.038432527333498 2023-01-23 16:19:32.291258: step: 2188/1541, loss: 0.2682119905948639 2023-01-23 16:19:33.216481: step: 2190/1541, loss: 0.00019143274403177202 2023-01-23 16:19:34.087865: step: 2192/1541, loss: 0.01625649817287922 2023-01-23 16:19:34.962819: step: 2194/1541, loss: 0.01151634007692337 2023-01-23 16:19:35.876131: step: 2196/1541, loss: 0.00742656784132123 2023-01-23 16:19:36.789347: step: 2198/1541, loss: 0.013780028559267521 2023-01-23 16:19:37.649527: step: 2200/1541, loss: 0.010790040716528893 2023-01-23 16:19:38.518087: step: 2202/1541, loss: 3.9349746657535434e-05 2023-01-23 16:19:39.387662: step: 2204/1541, loss: 0.06918448954820633 2023-01-23 16:19:40.273126: step: 2206/1541, loss: 0.061250440776348114 2023-01-23 16:19:41.255080: step: 2208/1541, loss: 0.0038516686763614416 2023-01-23 16:19:42.135850: step: 2210/1541, loss: 0.0026954293716698885 2023-01-23 16:19:43.036493: step: 2212/1541, loss: 0.7101017832756042 2023-01-23 16:19:43.896641: step: 2214/1541, loss: 0.0003050834347959608 2023-01-23 16:19:44.748486: step: 2216/1541, loss: 0.0004527743731159717 2023-01-23 16:19:45.654054: step: 2218/1541, loss: 0.00030725731630809605 2023-01-23 16:19:46.493849: step: 2220/1541, loss: 0.02821926586329937 2023-01-23 16:19:47.376659: step: 2222/1541, loss: 0.03396381437778473 2023-01-23 16:19:48.336308: step: 2224/1541, loss: 0.03950171917676926 2023-01-23 16:19:49.249154: step: 2226/1541, loss: 0.06509217619895935 2023-01-23 16:19:50.103917: step: 2228/1541, loss: 0.012284661643207073 2023-01-23 16:19:51.004899: step: 2230/1541, loss: 5.866832907486241e-06 2023-01-23 16:19:51.924488: step: 2232/1541, loss: 0.4205312430858612 2023-01-23 16:19:52.809514: step: 2234/1541, loss: 0.00105141696985811 2023-01-23 16:19:53.691506: step: 2236/1541, loss: 0.030175043269991875 2023-01-23 16:19:54.602302: step: 2238/1541, loss: 0.10049186646938324 2023-01-23 16:19:55.547979: step: 2240/1541, loss: 0.011631421744823456 2023-01-23 16:19:56.501756: step: 2242/1541, loss: 0.10151005536317825 2023-01-23 16:19:57.397778: step: 2244/1541, loss: 0.04282008111476898 2023-01-23 16:19:58.310853: step: 2246/1541, loss: 0.1276804506778717 2023-01-23 16:19:59.265664: step: 2248/1541, loss: 0.0009086555801331997 2023-01-23 16:20:00.153158: step: 2250/1541, loss: 0.0546896830201149 2023-01-23 16:20:01.078756: step: 2252/1541, loss: 0.08367660641670227 2023-01-23 16:20:01.982834: step: 2254/1541, loss: 0.007327056024223566 2023-01-23 16:20:02.877346: step: 2256/1541, loss: 0.1717769056558609 2023-01-23 16:20:03.752696: step: 2258/1541, loss: 0.02043004147708416 2023-01-23 16:20:04.628123: step: 2260/1541, loss: 0.0032082144170999527 2023-01-23 16:20:05.526913: step: 2262/1541, loss: 0.03027808852493763 2023-01-23 16:20:06.398585: step: 2264/1541, loss: 0.0007718491251580417 2023-01-23 16:20:07.270797: step: 2266/1541, loss: 0.006182988174259663 2023-01-23 16:20:08.192680: step: 2268/1541, loss: 0.12311767041683197 2023-01-23 16:20:09.058864: step: 2270/1541, loss: 0.0025365801993757486 2023-01-23 16:20:10.044037: step: 2272/1541, loss: 0.13947363197803497 2023-01-23 16:20:10.934043: step: 2274/1541, loss: 0.0015635578893125057 2023-01-23 16:20:11.792092: step: 2276/1541, loss: 0.0018501931335777044 2023-01-23 16:20:12.678667: step: 2278/1541, loss: 0.003096431726589799 2023-01-23 16:20:13.587083: step: 2280/1541, loss: 0.03934215381741524 2023-01-23 16:20:14.468846: step: 2282/1541, loss: 0.011049089953303337 2023-01-23 16:20:15.316520: step: 2284/1541, loss: 0.03455956652760506 2023-01-23 16:20:16.200080: step: 2286/1541, loss: 0.00157958606723696 2023-01-23 16:20:17.058275: step: 2288/1541, loss: 0.002898409264162183 2023-01-23 16:20:17.946459: step: 2290/1541, loss: 0.0003009262727573514 2023-01-23 16:20:18.845757: step: 2292/1541, loss: 0.2707395553588867 2023-01-23 16:20:19.794065: step: 2294/1541, loss: 0.010591389611363411 2023-01-23 16:20:20.679104: step: 2296/1541, loss: 0.037015121430158615 2023-01-23 16:20:21.580076: step: 2298/1541, loss: 1.4298086171038449e-05 2023-01-23 16:20:22.499511: step: 2300/1541, loss: 0.0015437453985214233 2023-01-23 16:20:23.422963: step: 2302/1541, loss: 0.006924621760845184 2023-01-23 16:20:24.282880: step: 2304/1541, loss: 0.1433108150959015 2023-01-23 16:20:25.189254: step: 2306/1541, loss: 0.027041716501116753 2023-01-23 16:20:26.128063: step: 2308/1541, loss: 0.018308125436306 2023-01-23 16:20:27.045332: step: 2310/1541, loss: 0.00011440963135100901 2023-01-23 16:20:27.965775: step: 2312/1541, loss: 5.2778705139644444e-05 2023-01-23 16:20:28.854204: step: 2314/1541, loss: 0.002145344391465187 2023-01-23 16:20:29.756538: step: 2316/1541, loss: 0.002041089115664363 2023-01-23 16:20:30.726045: step: 2318/1541, loss: 0.011667434126138687 2023-01-23 16:20:31.631910: step: 2320/1541, loss: 0.016470661386847496 2023-01-23 16:20:32.613665: step: 2322/1541, loss: 0.012001166120171547 2023-01-23 16:20:33.549155: step: 2324/1541, loss: 0.00013480197230819613 2023-01-23 16:20:34.410314: step: 2326/1541, loss: 0.0001070966973202303 2023-01-23 16:20:35.290330: step: 2328/1541, loss: 0.05346405878663063 2023-01-23 16:20:36.170493: step: 2330/1541, loss: 0.017426710575819016 2023-01-23 16:20:37.062823: step: 2332/1541, loss: 0.026969710364937782 2023-01-23 16:20:37.955842: step: 2334/1541, loss: 0.00013941735960543156 2023-01-23 16:20:38.850677: step: 2336/1541, loss: 0.0051104044541716576 2023-01-23 16:20:39.772555: step: 2338/1541, loss: 0.013359172269701958 2023-01-23 16:20:40.643580: step: 2340/1541, loss: 0.0023415712639689445 2023-01-23 16:20:41.597759: step: 2342/1541, loss: 0.005216128192842007 2023-01-23 16:20:42.467991: step: 2344/1541, loss: 0.0003113779239356518 2023-01-23 16:20:43.375042: step: 2346/1541, loss: 0.0009708025609143078 2023-01-23 16:20:44.246139: step: 2348/1541, loss: 0.05009683221578598 2023-01-23 16:20:45.157474: step: 2350/1541, loss: 0.021496228873729706 2023-01-23 16:20:46.034469: step: 2352/1541, loss: 0.0008562112925574183 2023-01-23 16:20:46.943577: step: 2354/1541, loss: 0.021480601280927658 2023-01-23 16:20:47.859753: step: 2356/1541, loss: 9.641046926844865e-05 2023-01-23 16:20:48.798828: step: 2358/1541, loss: 0.02820865996181965 2023-01-23 16:20:49.813760: step: 2360/1541, loss: 0.2053535133600235 2023-01-23 16:20:50.658748: step: 2362/1541, loss: 0.0004688950430136174 2023-01-23 16:20:51.547388: step: 2364/1541, loss: 0.000529897166416049 2023-01-23 16:20:52.459499: step: 2366/1541, loss: 0.21738101541996002 2023-01-23 16:20:53.368871: step: 2368/1541, loss: 0.0037744608707726 2023-01-23 16:20:54.321404: step: 2370/1541, loss: 0.0022703667636960745 2023-01-23 16:20:55.200266: step: 2372/1541, loss: 0.07572775334119797 2023-01-23 16:20:56.090652: step: 2374/1541, loss: 0.0012112994445487857 2023-01-23 16:20:56.964514: step: 2376/1541, loss: 0.0006111178081482649 2023-01-23 16:20:57.855769: step: 2378/1541, loss: 0.0015362222911790013 2023-01-23 16:20:58.762159: step: 2380/1541, loss: 0.0005181287415325642 2023-01-23 16:20:59.666928: step: 2382/1541, loss: 0.0016576850321143866 2023-01-23 16:21:00.539321: step: 2384/1541, loss: 0.0006577210151590407 2023-01-23 16:21:01.402226: step: 2386/1541, loss: 0.016824690625071526 2023-01-23 16:21:02.368741: step: 2388/1541, loss: 0.019091418012976646 2023-01-23 16:21:03.264465: step: 2390/1541, loss: 0.012906758114695549 2023-01-23 16:21:04.154685: step: 2392/1541, loss: 0.0007084361859597266 2023-01-23 16:21:05.033439: step: 2394/1541, loss: 6.857197149656713e-05 2023-01-23 16:21:05.905011: step: 2396/1541, loss: 0.21887491643428802 2023-01-23 16:21:06.861258: step: 2398/1541, loss: 0.0027156902942806482 2023-01-23 16:21:07.790287: step: 2400/1541, loss: 0.21343454718589783 2023-01-23 16:21:08.677307: step: 2402/1541, loss: 0.027579493820667267 2023-01-23 16:21:09.538468: step: 2404/1541, loss: 0.1391458511352539 2023-01-23 16:21:10.447672: step: 2406/1541, loss: 0.00034660304663702846 2023-01-23 16:21:11.354640: step: 2408/1541, loss: 0.0037385618779808283 2023-01-23 16:21:12.284927: step: 2410/1541, loss: 0.3100816011428833 2023-01-23 16:21:13.150480: step: 2412/1541, loss: 0.0007821621256880462 2023-01-23 16:21:14.030844: step: 2414/1541, loss: 0.014727620407938957 2023-01-23 16:21:14.937488: step: 2416/1541, loss: 0.024064332246780396 2023-01-23 16:21:15.847487: step: 2418/1541, loss: 0.004902430810034275 2023-01-23 16:21:16.780086: step: 2420/1541, loss: 0.0015062498860061169 2023-01-23 16:21:17.697219: step: 2422/1541, loss: 0.007447331212460995 2023-01-23 16:21:18.571922: step: 2424/1541, loss: 0.25048089027404785 2023-01-23 16:21:19.533979: step: 2426/1541, loss: 0.46433135867118835 2023-01-23 16:21:20.406615: step: 2428/1541, loss: 0.0005674045532941818 2023-01-23 16:21:21.311706: step: 2430/1541, loss: 0.22712594270706177 2023-01-23 16:21:22.202137: step: 2432/1541, loss: 0.0012536941794678569 2023-01-23 16:21:23.146896: step: 2434/1541, loss: 0.003271494759246707 2023-01-23 16:21:24.029741: step: 2436/1541, loss: 0.005099315196275711 2023-01-23 16:21:24.920348: step: 2438/1541, loss: 0.009803161956369877 2023-01-23 16:21:25.786174: step: 2440/1541, loss: 0.0063085732981562614 2023-01-23 16:21:26.741749: step: 2442/1541, loss: 0.050871651619672775 2023-01-23 16:21:27.677171: step: 2444/1541, loss: 0.0017125182785093784 2023-01-23 16:21:28.582944: step: 2446/1541, loss: 0.0007675789529457688 2023-01-23 16:21:29.582591: step: 2448/1541, loss: 0.0007911452557891607 2023-01-23 16:21:30.498375: step: 2450/1541, loss: 0.36840614676475525 2023-01-23 16:21:31.364070: step: 2452/1541, loss: 0.008779511786997318 2023-01-23 16:21:32.248508: step: 2454/1541, loss: 0.003419708227738738 2023-01-23 16:21:33.138929: step: 2456/1541, loss: 0.04582668095827103 2023-01-23 16:21:34.044095: step: 2458/1541, loss: 0.005457079969346523 2023-01-23 16:21:34.992307: step: 2460/1541, loss: 0.16272351145744324 2023-01-23 16:21:35.926345: step: 2462/1541, loss: 0.004373633302748203 2023-01-23 16:21:36.817676: step: 2464/1541, loss: 0.14531637728214264 2023-01-23 16:21:37.704457: step: 2466/1541, loss: 0.057062260806560516 2023-01-23 16:21:38.571898: step: 2468/1541, loss: 0.09725784510374069 2023-01-23 16:21:39.457307: step: 2470/1541, loss: 0.0025630281306803226 2023-01-23 16:21:40.342219: step: 2472/1541, loss: 0.08670727163553238 2023-01-23 16:21:41.249334: step: 2474/1541, loss: 0.11345506459474564 2023-01-23 16:21:42.150263: step: 2476/1541, loss: 0.008248325437307358 2023-01-23 16:21:43.089846: step: 2478/1541, loss: 0.2036525160074234 2023-01-23 16:21:43.985189: step: 2480/1541, loss: 0.21540607511997223 2023-01-23 16:21:44.877851: step: 2482/1541, loss: 0.009157232940196991 2023-01-23 16:21:45.750621: step: 2484/1541, loss: 0.3398681879043579 2023-01-23 16:21:46.670316: step: 2486/1541, loss: 0.1026960089802742 2023-01-23 16:21:47.556670: step: 2488/1541, loss: 0.01590064913034439 2023-01-23 16:21:48.420561: step: 2490/1541, loss: 0.03302622586488724 2023-01-23 16:21:49.417847: step: 2492/1541, loss: 0.0431659035384655 2023-01-23 16:21:50.265771: step: 2494/1541, loss: 0.017308887094259262 2023-01-23 16:21:51.138188: step: 2496/1541, loss: 0.14064304530620575 2023-01-23 16:21:52.013838: step: 2498/1541, loss: 0.001064665848389268 2023-01-23 16:21:52.906487: step: 2500/1541, loss: 0.1365121752023697 2023-01-23 16:21:53.787399: step: 2502/1541, loss: 0.002351417439058423 2023-01-23 16:21:54.660624: step: 2504/1541, loss: 0.0013210282195359468 2023-01-23 16:21:55.556258: step: 2506/1541, loss: 0.3484727740287781 2023-01-23 16:21:56.404215: step: 2508/1541, loss: 0.017356332391500473 2023-01-23 16:21:57.264170: step: 2510/1541, loss: 0.013333425857126713 2023-01-23 16:21:58.126525: step: 2512/1541, loss: 0.17042666673660278 2023-01-23 16:21:59.046709: step: 2514/1541, loss: 0.0008408328867517412 2023-01-23 16:21:59.928048: step: 2516/1541, loss: 0.00028581396327354014 2023-01-23 16:22:00.814574: step: 2518/1541, loss: 0.7765573859214783 2023-01-23 16:22:01.727446: step: 2520/1541, loss: 0.0017497445223852992 2023-01-23 16:22:02.662819: step: 2522/1541, loss: 0.36133673787117004 2023-01-23 16:22:03.587156: step: 2524/1541, loss: 0.007085025310516357 2023-01-23 16:22:04.476935: step: 2526/1541, loss: 0.07636682689189911 2023-01-23 16:22:05.359340: step: 2528/1541, loss: 0.1269778460264206 2023-01-23 16:22:06.243269: step: 2530/1541, loss: 0.006077586207538843 2023-01-23 16:22:07.139933: step: 2532/1541, loss: 0.13175775110721588 2023-01-23 16:22:08.027346: step: 2534/1541, loss: 0.0017940036486834288 2023-01-23 16:22:08.912929: step: 2536/1541, loss: 0.012998491525650024 2023-01-23 16:22:09.796350: step: 2538/1541, loss: 0.03365707769989967 2023-01-23 16:22:10.692476: step: 2540/1541, loss: 0.01312170084565878 2023-01-23 16:22:11.597464: step: 2542/1541, loss: 0.02220291458070278 2023-01-23 16:22:12.487013: step: 2544/1541, loss: 0.06319700181484222 2023-01-23 16:22:13.398757: step: 2546/1541, loss: 0.06279007345438004 2023-01-23 16:22:14.280928: step: 2548/1541, loss: 0.024731630459427834 2023-01-23 16:22:15.212604: step: 2550/1541, loss: 0.14930842816829681 2023-01-23 16:22:16.135970: step: 2552/1541, loss: 0.0053130825981497765 2023-01-23 16:22:17.042088: step: 2554/1541, loss: 0.059036578983068466 2023-01-23 16:22:17.939177: step: 2556/1541, loss: 0.14925585687160492 2023-01-23 16:22:18.800962: step: 2558/1541, loss: 0.08466517180204391 2023-01-23 16:22:19.725564: step: 2560/1541, loss: 0.34127750992774963 2023-01-23 16:22:20.590927: step: 2562/1541, loss: 0.1317426860332489 2023-01-23 16:22:21.484791: step: 2564/1541, loss: 0.0021366127766668797 2023-01-23 16:22:22.381031: step: 2566/1541, loss: 0.0629354864358902 2023-01-23 16:22:23.233363: step: 2568/1541, loss: 0.5585861206054688 2023-01-23 16:22:24.121182: step: 2570/1541, loss: 0.11735488474369049 2023-01-23 16:22:25.012604: step: 2572/1541, loss: 0.0027386776637285948 2023-01-23 16:22:25.896373: step: 2574/1541, loss: 0.034302327781915665 2023-01-23 16:22:26.783802: step: 2576/1541, loss: 0.08371830731630325 2023-01-23 16:22:27.637654: step: 2578/1541, loss: 0.10484229028224945 2023-01-23 16:22:28.510515: step: 2580/1541, loss: 0.0004313527897465974 2023-01-23 16:22:29.478438: step: 2582/1541, loss: 0.0019966226536780596 2023-01-23 16:22:30.387810: step: 2584/1541, loss: 0.0214128028601408 2023-01-23 16:22:31.274107: step: 2586/1541, loss: 0.07451722025871277 2023-01-23 16:22:32.232976: step: 2588/1541, loss: 0.05636616051197052 2023-01-23 16:22:33.130396: step: 2590/1541, loss: 0.012301006354391575 2023-01-23 16:22:33.997339: step: 2592/1541, loss: 0.062242958694696426 2023-01-23 16:22:34.888451: step: 2594/1541, loss: 0.028998836874961853 2023-01-23 16:22:35.807652: step: 2596/1541, loss: 0.02907281555235386 2023-01-23 16:22:36.767444: step: 2598/1541, loss: 0.0003475647827144712 2023-01-23 16:22:37.683173: step: 2600/1541, loss: 0.014509465545415878 2023-01-23 16:22:38.570929: step: 2602/1541, loss: 0.0011154450476169586 2023-01-23 16:22:39.508560: step: 2604/1541, loss: 0.07848506420850754 2023-01-23 16:22:40.415318: step: 2606/1541, loss: 0.4161917269229889 2023-01-23 16:22:41.350623: step: 2608/1541, loss: 0.47756826877593994 2023-01-23 16:22:42.237496: step: 2610/1541, loss: 0.000540989451110363 2023-01-23 16:22:43.100388: step: 2612/1541, loss: 0.003803033847361803 2023-01-23 16:22:43.991312: step: 2614/1541, loss: 0.019580258056521416 2023-01-23 16:22:44.855815: step: 2616/1541, loss: 0.004236023407429457 2023-01-23 16:22:45.718201: step: 2618/1541, loss: 0.2397088259458542 2023-01-23 16:22:46.647085: step: 2620/1541, loss: 0.007769466377794743 2023-01-23 16:22:47.572231: step: 2622/1541, loss: 0.022909214720129967 2023-01-23 16:22:48.545314: step: 2624/1541, loss: 0.00018202296632807702 2023-01-23 16:22:49.425651: step: 2626/1541, loss: 0.09259846061468124 2023-01-23 16:22:50.320406: step: 2628/1541, loss: 0.07197977602481842 2023-01-23 16:22:51.178373: step: 2630/1541, loss: 0.0007948798593133688 2023-01-23 16:22:52.034848: step: 2632/1541, loss: 0.0039186254143714905 2023-01-23 16:22:52.907149: step: 2634/1541, loss: 0.03097992204129696 2023-01-23 16:22:53.792844: step: 2636/1541, loss: 0.0016744101885706186 2023-01-23 16:22:54.704332: step: 2638/1541, loss: 0.009283630177378654 2023-01-23 16:22:55.594890: step: 2640/1541, loss: 0.01728207617998123 2023-01-23 16:22:56.498356: step: 2642/1541, loss: 0.058241453021764755 2023-01-23 16:22:57.347150: step: 2644/1541, loss: 0.10465516149997711 2023-01-23 16:22:58.271192: step: 2646/1541, loss: 0.009368274360895157 2023-01-23 16:22:59.131604: step: 2648/1541, loss: 0.0007917544571682811 2023-01-23 16:23:00.037075: step: 2650/1541, loss: 0.0605907067656517 2023-01-23 16:23:00.972401: step: 2652/1541, loss: 0.006237347144633532 2023-01-23 16:23:01.865737: step: 2654/1541, loss: 0.0011579308193176985 2023-01-23 16:23:02.801040: step: 2656/1541, loss: 0.15034322440624237 2023-01-23 16:23:03.684134: step: 2658/1541, loss: 0.005010704509913921 2023-01-23 16:23:04.559117: step: 2660/1541, loss: 0.03257323428988457 2023-01-23 16:23:05.443186: step: 2662/1541, loss: 0.07853038609027863 2023-01-23 16:23:06.269191: step: 2664/1541, loss: 0.0017557714600116014 2023-01-23 16:23:07.149145: step: 2666/1541, loss: 0.018080079928040504 2023-01-23 16:23:08.052101: step: 2668/1541, loss: 0.004149341490119696 2023-01-23 16:23:08.956167: step: 2670/1541, loss: 0.029794249683618546 2023-01-23 16:23:09.904886: step: 2672/1541, loss: 0.016691813245415688 2023-01-23 16:23:10.774344: step: 2674/1541, loss: 0.04305168241262436 2023-01-23 16:23:11.666312: step: 2676/1541, loss: 0.05492902174592018 2023-01-23 16:23:12.581086: step: 2678/1541, loss: 0.20402151346206665 2023-01-23 16:23:13.501545: step: 2680/1541, loss: 0.0824052169919014 2023-01-23 16:23:14.386969: step: 2682/1541, loss: 0.0005346478428691626 2023-01-23 16:23:15.299165: step: 2684/1541, loss: 0.000157013040734455 2023-01-23 16:23:16.228725: step: 2686/1541, loss: 0.0005730987759307027 2023-01-23 16:23:17.115503: step: 2688/1541, loss: 0.14774774014949799 2023-01-23 16:23:18.037587: step: 2690/1541, loss: 0.00045021114055998623 2023-01-23 16:23:18.929959: step: 2692/1541, loss: 0.08087455481290817 2023-01-23 16:23:19.877729: step: 2694/1541, loss: 0.0023631658405065536 2023-01-23 16:23:20.783030: step: 2696/1541, loss: 0.040542736649513245 2023-01-23 16:23:21.659150: step: 2698/1541, loss: 0.0414133220911026 2023-01-23 16:23:22.625560: step: 2700/1541, loss: 0.010038276202976704 2023-01-23 16:23:23.574307: step: 2702/1541, loss: 0.0040350910276174545 2023-01-23 16:23:24.464911: step: 2704/1541, loss: 0.00628849770873785 2023-01-23 16:23:25.383824: step: 2706/1541, loss: 0.044380221515893936 2023-01-23 16:23:26.261294: step: 2708/1541, loss: 0.012096509337425232 2023-01-23 16:23:27.136456: step: 2710/1541, loss: 0.29426679015159607 2023-01-23 16:23:28.007919: step: 2712/1541, loss: 0.009776380844414234 2023-01-23 16:23:28.965287: step: 2714/1541, loss: 0.9381173253059387 2023-01-23 16:23:29.806905: step: 2716/1541, loss: 0.26800647377967834 2023-01-23 16:23:30.709013: step: 2718/1541, loss: 0.0011534024961292744 2023-01-23 16:23:31.609228: step: 2720/1541, loss: 0.056037452071905136 2023-01-23 16:23:32.546527: step: 2722/1541, loss: 0.009985792450606823 2023-01-23 16:23:33.436783: step: 2724/1541, loss: 0.07591602206230164 2023-01-23 16:23:34.272731: step: 2726/1541, loss: 0.36831894516944885 2023-01-23 16:23:35.161888: step: 2728/1541, loss: 0.20857872068881989 2023-01-23 16:23:36.097206: step: 2730/1541, loss: 0.1602216511964798 2023-01-23 16:23:36.989819: step: 2732/1541, loss: 0.0010562858078628778 2023-01-23 16:23:37.872554: step: 2734/1541, loss: 0.1521042287349701 2023-01-23 16:23:38.766905: step: 2736/1541, loss: 0.0050084893591701984 2023-01-23 16:23:39.622503: step: 2738/1541, loss: 0.002202683361247182 2023-01-23 16:23:40.522444: step: 2740/1541, loss: 0.007858184166252613 2023-01-23 16:23:41.524216: step: 2742/1541, loss: 0.009302756749093533 2023-01-23 16:23:42.416668: step: 2744/1541, loss: 0.0008763456135056913 2023-01-23 16:23:43.356242: step: 2746/1541, loss: 0.08531492948532104 2023-01-23 16:23:44.231455: step: 2748/1541, loss: 1.0137726068496704 2023-01-23 16:23:45.137867: step: 2750/1541, loss: 0.015185961499810219 2023-01-23 16:23:46.085927: step: 2752/1541, loss: 0.017442861571907997 2023-01-23 16:23:46.973767: step: 2754/1541, loss: 0.0013566958950832486 2023-01-23 16:23:47.876946: step: 2756/1541, loss: 0.04757050424814224 2023-01-23 16:23:48.765859: step: 2758/1541, loss: 0.11412131786346436 2023-01-23 16:23:49.637034: step: 2760/1541, loss: 0.032502707093954086 2023-01-23 16:23:50.492738: step: 2762/1541, loss: 0.02771385759115219 2023-01-23 16:23:51.411992: step: 2764/1541, loss: 0.0016735868994146585 2023-01-23 16:23:52.261679: step: 2766/1541, loss: 0.028717534616589546 2023-01-23 16:23:53.143730: step: 2768/1541, loss: 0.003969424869865179 2023-01-23 16:23:54.019623: step: 2770/1541, loss: 0.05081990361213684 2023-01-23 16:23:54.901380: step: 2772/1541, loss: 2.185917854309082 2023-01-23 16:23:55.810652: step: 2774/1541, loss: 0.005456447601318359 2023-01-23 16:23:56.693093: step: 2776/1541, loss: 0.4496244788169861 2023-01-23 16:23:57.568095: step: 2778/1541, loss: 0.07681742310523987 2023-01-23 16:23:58.478165: step: 2780/1541, loss: 0.11082316935062408 2023-01-23 16:23:59.349833: step: 2782/1541, loss: 0.022153515368700027 2023-01-23 16:24:00.249598: step: 2784/1541, loss: 0.0006664287066087127 2023-01-23 16:24:01.128919: step: 2786/1541, loss: 0.00022177971550263464 2023-01-23 16:24:02.080404: step: 2788/1541, loss: 0.0009484787588007748 2023-01-23 16:24:02.946557: step: 2790/1541, loss: 0.18997523188591003 2023-01-23 16:24:03.909913: step: 2792/1541, loss: 0.018403947353363037 2023-01-23 16:24:04.823567: step: 2794/1541, loss: 0.0049403454177081585 2023-01-23 16:24:05.735691: step: 2796/1541, loss: 0.00019834539853036404 2023-01-23 16:24:06.626037: step: 2798/1541, loss: 0.132819265127182 2023-01-23 16:24:07.667625: step: 2800/1541, loss: 0.1841006726026535 2023-01-23 16:24:08.608739: step: 2802/1541, loss: 0.3125908374786377 2023-01-23 16:24:09.481811: step: 2804/1541, loss: 0.0024446502793580294 2023-01-23 16:24:10.347623: step: 2806/1541, loss: 0.0018204761436209083 2023-01-23 16:24:11.249334: step: 2808/1541, loss: 0.0051423329859972 2023-01-23 16:24:12.162900: step: 2810/1541, loss: 0.03949632868170738 2023-01-23 16:24:13.057195: step: 2812/1541, loss: 0.008433555252850056 2023-01-23 16:24:13.953246: step: 2814/1541, loss: 0.04175803065299988 2023-01-23 16:24:14.840850: step: 2816/1541, loss: 0.007083413191139698 2023-01-23 16:24:15.718921: step: 2818/1541, loss: 0.09067025035619736 2023-01-23 16:24:16.581415: step: 2820/1541, loss: 0.001128085539676249 2023-01-23 16:24:17.524705: step: 2822/1541, loss: 0.5701945424079895 2023-01-23 16:24:18.389316: step: 2824/1541, loss: 0.000462455180240795 2023-01-23 16:24:19.354420: step: 2826/1541, loss: 0.001869264175184071 2023-01-23 16:24:20.251130: step: 2828/1541, loss: 0.05702035874128342 2023-01-23 16:24:21.145139: step: 2830/1541, loss: 0.0025478412862867117 2023-01-23 16:24:22.068215: step: 2832/1541, loss: 0.0070028542540967464 2023-01-23 16:24:22.973753: step: 2834/1541, loss: 0.012790537439286709 2023-01-23 16:24:23.909309: step: 2836/1541, loss: 0.00997261144220829 2023-01-23 16:24:24.749249: step: 2838/1541, loss: 0.004643220920115709 2023-01-23 16:24:25.625343: step: 2840/1541, loss: 0.0079118050634861 2023-01-23 16:24:26.495420: step: 2842/1541, loss: 0.047550320625305176 2023-01-23 16:24:27.383610: step: 2844/1541, loss: 0.030387386679649353 2023-01-23 16:24:28.333172: step: 2846/1541, loss: 0.003544768551364541 2023-01-23 16:24:29.228717: step: 2848/1541, loss: 0.0667245015501976 2023-01-23 16:24:30.144597: step: 2850/1541, loss: 0.00021722022211179137 2023-01-23 16:24:31.063826: step: 2852/1541, loss: 0.043386150151491165 2023-01-23 16:24:31.947470: step: 2854/1541, loss: 0.27805718779563904 2023-01-23 16:24:32.797122: step: 2856/1541, loss: 0.2665315866470337 2023-01-23 16:24:33.692649: step: 2858/1541, loss: 0.0627545565366745 2023-01-23 16:24:34.598713: step: 2860/1541, loss: 0.3984932601451874 2023-01-23 16:24:35.469709: step: 2862/1541, loss: 0.012526811100542545 2023-01-23 16:24:36.359820: step: 2864/1541, loss: 0.0036756943445652723 2023-01-23 16:24:37.248410: step: 2866/1541, loss: 0.22745752334594727 2023-01-23 16:24:38.149560: step: 2868/1541, loss: 0.008841955102980137 2023-01-23 16:24:39.024038: step: 2870/1541, loss: 0.08921895176172256 2023-01-23 16:24:39.900752: step: 2872/1541, loss: 0.1791456937789917 2023-01-23 16:24:40.801782: step: 2874/1541, loss: 0.0206777174025774 2023-01-23 16:24:41.695252: step: 2876/1541, loss: 0.09800086170434952 2023-01-23 16:24:42.569360: step: 2878/1541, loss: 0.006052888929843903 2023-01-23 16:24:43.499037: step: 2880/1541, loss: 0.003062783507630229 2023-01-23 16:24:44.415216: step: 2882/1541, loss: 0.00011464818817330524 2023-01-23 16:24:45.354400: step: 2884/1541, loss: 0.2823512554168701 2023-01-23 16:24:46.229990: step: 2886/1541, loss: 0.0012793539790436625 2023-01-23 16:24:47.241890: step: 2888/1541, loss: 0.0024095301050692797 2023-01-23 16:24:48.105350: step: 2890/1541, loss: 0.25747957825660706 2023-01-23 16:24:49.000642: step: 2892/1541, loss: 0.0017349696718156338 2023-01-23 16:24:49.893805: step: 2894/1541, loss: 0.013401839882135391 2023-01-23 16:24:50.837020: step: 2896/1541, loss: 0.00044028679258190095 2023-01-23 16:24:51.726719: step: 2898/1541, loss: 0.001997571438550949 2023-01-23 16:24:52.632396: step: 2900/1541, loss: 0.0009088119841180742 2023-01-23 16:24:53.503268: step: 2902/1541, loss: 0.0038774656131863594 2023-01-23 16:24:54.423737: step: 2904/1541, loss: 0.0037030635867267847 2023-01-23 16:24:55.344845: step: 2906/1541, loss: 0.0500854067504406 2023-01-23 16:24:56.294747: step: 2908/1541, loss: 0.02547626942396164 2023-01-23 16:24:57.192435: step: 2910/1541, loss: 0.02093585580587387 2023-01-23 16:24:58.105255: step: 2912/1541, loss: 0.07404728978872299 2023-01-23 16:24:59.041709: step: 2914/1541, loss: 0.05450452119112015 2023-01-23 16:24:59.932034: step: 2916/1541, loss: 0.00011228084622416645 2023-01-23 16:25:00.869402: step: 2918/1541, loss: 0.290885865688324 2023-01-23 16:25:01.754267: step: 2920/1541, loss: 0.0011531467316672206 2023-01-23 16:25:02.686426: step: 2922/1541, loss: 6.940009188838303e-05 2023-01-23 16:25:03.605421: step: 2924/1541, loss: 0.0001687290787231177 2023-01-23 16:25:04.505691: step: 2926/1541, loss: 0.0005934654036536813 2023-01-23 16:25:05.433978: step: 2928/1541, loss: 0.006601032800972462 2023-01-23 16:25:06.380626: step: 2930/1541, loss: 0.013411743566393852 2023-01-23 16:25:07.276022: step: 2932/1541, loss: 1.945897201949265e-05 2023-01-23 16:25:08.121696: step: 2934/1541, loss: 3.138536703772843e-05 2023-01-23 16:25:09.053421: step: 2936/1541, loss: 0.034369975328445435 2023-01-23 16:25:09.972061: step: 2938/1541, loss: 0.0010481380159035325 2023-01-23 16:25:10.884850: step: 2940/1541, loss: 0.0008190827211365104 2023-01-23 16:25:11.775226: step: 2942/1541, loss: 0.00011115219967905432 2023-01-23 16:25:12.728891: step: 2944/1541, loss: 0.0008541939314454794 2023-01-23 16:25:13.611216: step: 2946/1541, loss: 0.1451205462217331 2023-01-23 16:25:14.515831: step: 2948/1541, loss: 0.014634487219154835 2023-01-23 16:25:15.393388: step: 2950/1541, loss: 0.012926552444696426 2023-01-23 16:25:16.253262: step: 2952/1541, loss: 0.0005707902600988746 2023-01-23 16:25:17.167543: step: 2954/1541, loss: 0.0011173044331371784 2023-01-23 16:25:18.099884: step: 2956/1541, loss: 0.0012238468043506145 2023-01-23 16:25:18.993797: step: 2958/1541, loss: 0.032550495117902756 2023-01-23 16:25:19.885190: step: 2960/1541, loss: 0.0810953676700592 2023-01-23 16:25:20.778965: step: 2962/1541, loss: 0.12694253027439117 2023-01-23 16:25:21.717179: step: 2964/1541, loss: 9.198130283039063e-05 2023-01-23 16:25:22.610474: step: 2966/1541, loss: 0.01814245618879795 2023-01-23 16:25:23.526202: step: 2968/1541, loss: 1.6387097275583073e-05 2023-01-23 16:25:24.432109: step: 2970/1541, loss: 0.003448747331276536 2023-01-23 16:25:25.301082: step: 2972/1541, loss: 0.00026866357075050473 2023-01-23 16:25:26.188610: step: 2974/1541, loss: 0.004874853417277336 2023-01-23 16:25:27.095045: step: 2976/1541, loss: 0.0008142408332787454 2023-01-23 16:25:28.005667: step: 2978/1541, loss: 0.11552084982395172 2023-01-23 16:25:28.892968: step: 2980/1541, loss: 0.0032004534732550383 2023-01-23 16:25:29.775963: step: 2982/1541, loss: 0.020875178277492523 2023-01-23 16:25:30.661609: step: 2984/1541, loss: 0.02195100486278534 2023-01-23 16:25:31.601020: step: 2986/1541, loss: 0.38893672823905945 2023-01-23 16:25:32.484982: step: 2988/1541, loss: 0.15293322503566742 2023-01-23 16:25:33.386808: step: 2990/1541, loss: 0.020701047033071518 2023-01-23 16:25:34.284244: step: 2992/1541, loss: 0.09925036132335663 2023-01-23 16:25:35.179412: step: 2994/1541, loss: 0.04701390117406845 2023-01-23 16:25:36.073396: step: 2996/1541, loss: 0.06373412907123566 2023-01-23 16:25:36.955865: step: 2998/1541, loss: 0.2548990547657013 2023-01-23 16:25:37.835157: step: 3000/1541, loss: 0.009600860998034477 2023-01-23 16:25:38.714410: step: 3002/1541, loss: 0.0037036167923361063 2023-01-23 16:25:39.605487: step: 3004/1541, loss: 0.0010906639508903027 2023-01-23 16:25:40.455929: step: 3006/1541, loss: 0.01817488856613636 2023-01-23 16:25:41.383805: step: 3008/1541, loss: 0.20579120516777039 2023-01-23 16:25:42.327028: step: 3010/1541, loss: 0.00844892580062151 2023-01-23 16:25:43.192238: step: 3012/1541, loss: 0.012906289659440517 2023-01-23 16:25:44.095678: step: 3014/1541, loss: 0.267122745513916 2023-01-23 16:25:44.981781: step: 3016/1541, loss: 0.019626908004283905 2023-01-23 16:25:45.847620: step: 3018/1541, loss: 0.0027235527522861958 2023-01-23 16:25:46.744459: step: 3020/1541, loss: 0.00016424476052634418 2023-01-23 16:25:47.638980: step: 3022/1541, loss: 0.006388949695974588 2023-01-23 16:25:48.582042: step: 3024/1541, loss: 0.023129794746637344 2023-01-23 16:25:49.460202: step: 3026/1541, loss: 0.036027729511260986 2023-01-23 16:25:50.457531: step: 3028/1541, loss: 0.5622988939285278 2023-01-23 16:25:51.385980: step: 3030/1541, loss: 0.0018252169247716665 2023-01-23 16:25:52.314227: step: 3032/1541, loss: 0.45110464096069336 2023-01-23 16:25:53.277235: step: 3034/1541, loss: 0.0004590660100802779 2023-01-23 16:25:54.234011: step: 3036/1541, loss: 0.015383017249405384 2023-01-23 16:25:55.118242: step: 3038/1541, loss: 0.00932838674634695 2023-01-23 16:25:55.999944: step: 3040/1541, loss: 0.0023753370624035597 2023-01-23 16:25:56.884143: step: 3042/1541, loss: 0.00027425584266893566 2023-01-23 16:25:57.792364: step: 3044/1541, loss: 0.04143612086772919 2023-01-23 16:25:58.704819: step: 3046/1541, loss: 0.14097437262535095 2023-01-23 16:25:59.618850: step: 3048/1541, loss: 0.00010564482363406569 2023-01-23 16:26:00.499599: step: 3050/1541, loss: 0.002508337376639247 2023-01-23 16:26:01.349020: step: 3052/1541, loss: 0.00012058726861141622 2023-01-23 16:26:02.286111: step: 3054/1541, loss: 0.007843307219445705 2023-01-23 16:26:03.175342: step: 3056/1541, loss: 0.256492018699646 2023-01-23 16:26:04.040272: step: 3058/1541, loss: 0.5293043851852417 2023-01-23 16:26:04.952378: step: 3060/1541, loss: 0.010059373453259468 2023-01-23 16:26:05.820297: step: 3062/1541, loss: 0.01102092582732439 2023-01-23 16:26:06.714939: step: 3064/1541, loss: 0.27898290753364563 2023-01-23 16:26:07.611726: step: 3066/1541, loss: 0.02576991729438305 2023-01-23 16:26:08.583116: step: 3068/1541, loss: 0.003206757828593254 2023-01-23 16:26:09.560716: step: 3070/1541, loss: 0.005500867497175932 2023-01-23 16:26:10.441297: step: 3072/1541, loss: 0.03293401002883911 2023-01-23 16:26:11.400017: step: 3074/1541, loss: 0.024814974516630173 2023-01-23 16:26:12.268073: step: 3076/1541, loss: 0.002447637962177396 2023-01-23 16:26:13.150363: step: 3078/1541, loss: 0.15824969112873077 2023-01-23 16:26:14.024509: step: 3080/1541, loss: 0.011999864131212234 2023-01-23 16:26:15.017996: step: 3082/1541, loss: 0.1268187314271927 2023-01-23 16:26:15.634087: step: 3084/1541, loss: 0.00397350313141942 ================================================== Loss: 0.082 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.679679224642488, 'r': 0.6459587049703025, 'f1': 0.6623900869727267}, 'combined': 0.43714408762431844, 'epoch': 23} Test Chinese: {'event': {'p': 0.6151419558359621, 'r': 0.701018573996405, 'f1': 0.6552786334360123}, 'argument': {'p': 0.720803339496083, 'r': 0.694252961199546, 'f1': 0.7072790710047019}, 'combined': 0.4634648631058534, 'epoch': 23} Dev Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.6710994412739938, 'r': 0.6556940021629696, 'f1': 0.6633072852069517}, 'combined': 0.4359625522195888, 'epoch': 23} Test Korean: {'event': {'p': 0.611534795042898, 'r': 0.768723786698622, 'f1': 0.6811786567560393}, 'argument': {'p': 0.7159306392023576, 'r': 0.7063815467875841, 'f1': 0.7111240377377481}, 'combined': 0.48440251681313024, 'epoch': 23} Dev Russian: {'event': {'p': 0.6528436018957346, 'r': 0.7336884154460719, 'f1': 0.6909090909090909}, 'argument': {'p': 0.676523549274366, 'r': 0.6541254784180288, 'f1': 0.6651360067464118}, 'combined': 0.45954851375206635, 'epoch': 23} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.7174344421937103, 'r': 0.6986291587897101, 'f1': 0.7079069336835883}, 'combined': 0.49197554497339874, 'epoch': 23} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.666347687400319, 'r': 0.6722445695897024, 'f1': 0.6692831397677212}, 'combined': 0.4919517095728549, 'epoch': 23} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5628867366155502, 'r': 0.43130282415996707, 'f1': 0.4883870214752568}, 'combined': 0.28457196526904727, 'epoch': 23} Russian: {'event': {'p': 0.47058823529411764, 'r': 0.4444444444444444, 'f1': 0.45714285714285713}, 'argument': {'p': 0.5939748501817468, 'r': 0.6216015873995024, 'f1': 0.6074742785949683}, 'combined': 0.2777025273576998, 'epoch': 23} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.679679224642488, 'r': 0.6459587049703025, 'f1': 0.6623900869727267}, 'combined': 0.43714408762431844, 'epoch': 23} Test for Chinese: {'event': {'p': 0.6151419558359621, 'r': 0.701018573996405, 'f1': 0.6552786334360123}, 'argument': {'p': 0.720803339496083, 'r': 0.694252961199546, 'f1': 0.7072790710047019}, 'combined': 0.4634648631058534, 'epoch': 23} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.666347687400319, 'r': 0.6722445695897024, 'f1': 0.6692831397677212}, 'combined': 0.4919517095728549, 'epoch': 23} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 24 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 16:31:44.309484: step: 2/1541, loss: 0.0012755778152495623 2023-01-23 16:31:45.196664: step: 4/1541, loss: 0.006059783510863781 2023-01-23 16:31:46.101533: step: 6/1541, loss: 0.0025475099682807922 2023-01-23 16:31:46.951849: step: 8/1541, loss: 0.0002745202218648046 2023-01-23 16:31:47.837686: step: 10/1541, loss: 0.0005432215984910727 2023-01-23 16:31:48.744088: step: 12/1541, loss: 0.005156044382601976 2023-01-23 16:31:49.755089: step: 14/1541, loss: 0.2901928126811981 2023-01-23 16:31:50.630803: step: 16/1541, loss: 0.0010419863974675536 2023-01-23 16:31:51.497205: step: 18/1541, loss: 0.1792575716972351 2023-01-23 16:31:52.360225: step: 20/1541, loss: 0.16154134273529053 2023-01-23 16:31:53.242950: step: 22/1541, loss: 0.06874652951955795 2023-01-23 16:31:54.119221: step: 24/1541, loss: 0.06188051402568817 2023-01-23 16:31:55.018827: step: 26/1541, loss: 0.05746160447597504 2023-01-23 16:31:55.887393: step: 28/1541, loss: 0.5184366703033447 2023-01-23 16:31:56.752057: step: 30/1541, loss: 0.0017471152823418379 2023-01-23 16:31:57.687334: step: 32/1541, loss: 0.02036900818347931 2023-01-23 16:31:58.607083: step: 34/1541, loss: 0.0003150367992930114 2023-01-23 16:31:59.498090: step: 36/1541, loss: 0.2295779585838318 2023-01-23 16:32:00.378103: step: 38/1541, loss: 0.10742587596178055 2023-01-23 16:32:01.279266: step: 40/1541, loss: 0.05402899533510208 2023-01-23 16:32:02.223792: step: 42/1541, loss: 0.02613067999482155 2023-01-23 16:32:03.092089: step: 44/1541, loss: 0.26310282945632935 2023-01-23 16:32:03.962290: step: 46/1541, loss: 0.005061349831521511 2023-01-23 16:32:04.827113: step: 48/1541, loss: 0.0002623085165396333 2023-01-23 16:32:05.722337: step: 50/1541, loss: 0.004039964638650417 2023-01-23 16:32:06.631816: step: 52/1541, loss: 0.0014743960928171873 2023-01-23 16:32:07.493031: step: 54/1541, loss: 0.03300799801945686 2023-01-23 16:32:08.369015: step: 56/1541, loss: 0.0005633039982058108 2023-01-23 16:32:09.268140: step: 58/1541, loss: 0.3277514576911926 2023-01-23 16:32:10.167497: step: 60/1541, loss: 0.00034088073880411685 2023-01-23 16:32:11.027133: step: 62/1541, loss: 0.004013936500996351 2023-01-23 16:32:11.928606: step: 64/1541, loss: 0.004119354300200939 2023-01-23 16:32:12.813970: step: 66/1541, loss: 0.0017520111287012696 2023-01-23 16:32:13.691444: step: 68/1541, loss: 0.009300806559622288 2023-01-23 16:32:14.595436: step: 70/1541, loss: 0.0006437198608182371 2023-01-23 16:32:15.515404: step: 72/1541, loss: 0.00012221797078382224 2023-01-23 16:32:16.447856: step: 74/1541, loss: 0.06325168907642365 2023-01-23 16:32:17.362878: step: 76/1541, loss: 0.010481072589755058 2023-01-23 16:32:18.224118: step: 78/1541, loss: 0.15739226341247559 2023-01-23 16:32:19.108883: step: 80/1541, loss: 0.37328481674194336 2023-01-23 16:32:20.026761: step: 82/1541, loss: 0.00033027015160769224 2023-01-23 16:32:20.904802: step: 84/1541, loss: 0.014967966824769974 2023-01-23 16:32:21.805271: step: 86/1541, loss: 0.10101593285799026 2023-01-23 16:32:22.709001: step: 88/1541, loss: 0.00012045140465488657 2023-01-23 16:32:23.616834: step: 90/1541, loss: 0.24394164979457855 2023-01-23 16:32:24.488701: step: 92/1541, loss: 0.03195561468601227 2023-01-23 16:32:25.399387: step: 94/1541, loss: 0.7469797134399414 2023-01-23 16:32:26.288514: step: 96/1541, loss: 7.350149826379493e-05 2023-01-23 16:32:27.145848: step: 98/1541, loss: 0.0827222615480423 2023-01-23 16:32:28.005511: step: 100/1541, loss: 0.018180489540100098 2023-01-23 16:32:28.886110: step: 102/1541, loss: 0.0034359856508672237 2023-01-23 16:32:29.826506: step: 104/1541, loss: 6.351572665153071e-05 2023-01-23 16:32:30.761328: step: 106/1541, loss: 0.06082650274038315 2023-01-23 16:32:31.627540: step: 108/1541, loss: 0.0020402800291776657 2023-01-23 16:32:32.564028: step: 110/1541, loss: 0.008621572516858578 2023-01-23 16:32:33.455592: step: 112/1541, loss: 0.00022998085478320718 2023-01-23 16:32:34.307270: step: 114/1541, loss: 0.02417753078043461 2023-01-23 16:32:35.197551: step: 116/1541, loss: 0.07167036831378937 2023-01-23 16:32:36.091155: step: 118/1541, loss: 0.000602614542003721 2023-01-23 16:32:36.962627: step: 120/1541, loss: 0.004074789118021727 2023-01-23 16:32:37.847170: step: 122/1541, loss: 0.0006609943229705095 2023-01-23 16:32:38.813972: step: 124/1541, loss: 0.0004277956031728536 2023-01-23 16:32:39.695790: step: 126/1541, loss: 0.10372167825698853 2023-01-23 16:32:40.590023: step: 128/1541, loss: 0.0009406688623130322 2023-01-23 16:32:41.469361: step: 130/1541, loss: 0.11877145618200302 2023-01-23 16:32:42.365975: step: 132/1541, loss: 0.02719493955373764 2023-01-23 16:32:43.317800: step: 134/1541, loss: 0.0014865989796817303 2023-01-23 16:32:44.301539: step: 136/1541, loss: 0.0009416018729098141 2023-01-23 16:32:45.234308: step: 138/1541, loss: 0.040680740028619766 2023-01-23 16:32:46.162366: step: 140/1541, loss: 0.0017290199175477028 2023-01-23 16:32:47.060178: step: 142/1541, loss: 0.00037293575587682426 2023-01-23 16:32:47.991856: step: 144/1541, loss: 0.15064865350723267 2023-01-23 16:32:48.923301: step: 146/1541, loss: 0.003232511691749096 2023-01-23 16:32:49.824564: step: 148/1541, loss: 1.0502206087112427 2023-01-23 16:32:50.728499: step: 150/1541, loss: 0.0014503825223073363 2023-01-23 16:32:51.612234: step: 152/1541, loss: 0.003313440131023526 2023-01-23 16:32:52.474198: step: 154/1541, loss: 0.021962767466902733 2023-01-23 16:32:53.328156: step: 156/1541, loss: 0.0003360130940563977 2023-01-23 16:32:54.340694: step: 158/1541, loss: 0.013031075708568096 2023-01-23 16:32:55.212629: step: 160/1541, loss: 0.0015965440543368459 2023-01-23 16:32:56.077969: step: 162/1541, loss: 0.20330600440502167 2023-01-23 16:32:57.082759: step: 164/1541, loss: 0.008768505416810513 2023-01-23 16:32:58.016292: step: 166/1541, loss: 0.0007490228163078427 2023-01-23 16:32:58.902436: step: 168/1541, loss: 0.06915771216154099 2023-01-23 16:32:59.827056: step: 170/1541, loss: 0.0038547231815755367 2023-01-23 16:33:00.708033: step: 172/1541, loss: 0.003294397611171007 2023-01-23 16:33:01.623114: step: 174/1541, loss: 0.00033904038718901575 2023-01-23 16:33:02.541319: step: 176/1541, loss: 0.0002332293806830421 2023-01-23 16:33:03.427581: step: 178/1541, loss: 0.004723507445305586 2023-01-23 16:33:04.325873: step: 180/1541, loss: 0.027823155745863914 2023-01-23 16:33:05.235992: step: 182/1541, loss: 0.001262352685444057 2023-01-23 16:33:06.198165: step: 184/1541, loss: 0.16772499680519104 2023-01-23 16:33:07.073371: step: 186/1541, loss: 0.00011095617082901299 2023-01-23 16:33:07.986840: step: 188/1541, loss: 0.0030252046417444944 2023-01-23 16:33:08.850661: step: 190/1541, loss: 0.002923327498137951 2023-01-23 16:33:09.748679: step: 192/1541, loss: 0.19928424060344696 2023-01-23 16:33:10.603962: step: 194/1541, loss: 0.1051863357424736 2023-01-23 16:33:11.496731: step: 196/1541, loss: 0.0002871831238735467 2023-01-23 16:33:12.420169: step: 198/1541, loss: 0.0003461781016085297 2023-01-23 16:33:13.287760: step: 200/1541, loss: 0.010520645417273045 2023-01-23 16:33:14.194579: step: 202/1541, loss: 0.0004320594889577478 2023-01-23 16:33:15.067596: step: 204/1541, loss: 0.01437346450984478 2023-01-23 16:33:15.977945: step: 206/1541, loss: 0.0002884305431507528 2023-01-23 16:33:16.891354: step: 208/1541, loss: 0.010990512557327747 2023-01-23 16:33:17.760197: step: 210/1541, loss: 0.007038992829620838 2023-01-23 16:33:18.623369: step: 212/1541, loss: 0.0779898539185524 2023-01-23 16:33:19.486381: step: 214/1541, loss: 0.02055945247411728 2023-01-23 16:33:20.371446: step: 216/1541, loss: 0.2563861906528473 2023-01-23 16:33:21.252706: step: 218/1541, loss: 0.004388689529150724 2023-01-23 16:33:22.165140: step: 220/1541, loss: 0.017128173261880875 2023-01-23 16:33:23.052839: step: 222/1541, loss: 0.003071740735322237 2023-01-23 16:33:23.929950: step: 224/1541, loss: 0.02293386682868004 2023-01-23 16:33:24.827752: step: 226/1541, loss: 0.10826905816793442 2023-01-23 16:33:25.699116: step: 228/1541, loss: 0.3046971559524536 2023-01-23 16:33:26.605809: step: 230/1541, loss: 0.017407147213816643 2023-01-23 16:33:27.466639: step: 232/1541, loss: 0.00026910845190286636 2023-01-23 16:33:28.373987: step: 234/1541, loss: 0.0033206839580088854 2023-01-23 16:33:29.276441: step: 236/1541, loss: 0.05271445959806442 2023-01-23 16:33:30.131420: step: 238/1541, loss: 0.26245275139808655 2023-01-23 16:33:31.013929: step: 240/1541, loss: 0.24593646824359894 2023-01-23 16:33:31.899883: step: 242/1541, loss: 0.0003122422203887254 2023-01-23 16:33:32.800084: step: 244/1541, loss: 0.016272619366645813 2023-01-23 16:33:33.762778: step: 246/1541, loss: 0.2235414832830429 2023-01-23 16:33:34.629649: step: 248/1541, loss: 0.0058016227558255196 2023-01-23 16:33:35.502025: step: 250/1541, loss: 0.0025324136950075626 2023-01-23 16:33:36.468824: step: 252/1541, loss: 0.0001722275628708303 2023-01-23 16:33:37.387332: step: 254/1541, loss: 0.0004385537758935243 2023-01-23 16:33:38.300256: step: 256/1541, loss: 0.001050422666594386 2023-01-23 16:33:39.157175: step: 258/1541, loss: 0.0005267588421702385 2023-01-23 16:33:39.996879: step: 260/1541, loss: 0.04399034380912781 2023-01-23 16:33:40.877718: step: 262/1541, loss: 0.035489220172166824 2023-01-23 16:33:41.775369: step: 264/1541, loss: 0.003738843835890293 2023-01-23 16:33:42.647763: step: 266/1541, loss: 0.050988487899303436 2023-01-23 16:33:43.548047: step: 268/1541, loss: 0.13939525187015533 2023-01-23 16:33:44.486860: step: 270/1541, loss: 0.007861741818487644 2023-01-23 16:33:45.355699: step: 272/1541, loss: 0.00038112280890345573 2023-01-23 16:33:46.258142: step: 274/1541, loss: 0.0007164754206314683 2023-01-23 16:33:47.217275: step: 276/1541, loss: 0.0009464097674936056 2023-01-23 16:33:48.111996: step: 278/1541, loss: 0.0032639349810779095 2023-01-23 16:33:48.967059: step: 280/1541, loss: 0.006235506851226091 2023-01-23 16:33:49.890468: step: 282/1541, loss: 0.001982586458325386 2023-01-23 16:33:50.768770: step: 284/1541, loss: 0.0019020794425159693 2023-01-23 16:33:51.668798: step: 286/1541, loss: 0.0028097485192120075 2023-01-23 16:33:52.603544: step: 288/1541, loss: 0.014337697066366673 2023-01-23 16:33:53.501386: step: 290/1541, loss: 1.4533388821291737e-05 2023-01-23 16:33:54.404596: step: 292/1541, loss: 0.002226521959528327 2023-01-23 16:33:55.344611: step: 294/1541, loss: 0.011159220710396767 2023-01-23 16:33:56.246529: step: 296/1541, loss: 0.6334323287010193 2023-01-23 16:33:57.190255: step: 298/1541, loss: 0.011238034814596176 2023-01-23 16:33:58.073860: step: 300/1541, loss: 0.034312985837459564 2023-01-23 16:33:58.963688: step: 302/1541, loss: 0.03837684169411659 2023-01-23 16:33:59.851136: step: 304/1541, loss: 0.0001338581641903147 2023-01-23 16:34:00.766747: step: 306/1541, loss: 0.004874016158282757 2023-01-23 16:34:01.651127: step: 308/1541, loss: 0.029661230742931366 2023-01-23 16:34:02.601360: step: 310/1541, loss: 0.032179247587919235 2023-01-23 16:34:03.456643: step: 312/1541, loss: 0.12588854134082794 2023-01-23 16:34:04.344509: step: 314/1541, loss: 0.2688031494617462 2023-01-23 16:34:05.223360: step: 316/1541, loss: 0.0737452581524849 2023-01-23 16:34:06.096175: step: 318/1541, loss: 0.04233473166823387 2023-01-23 16:34:07.031598: step: 320/1541, loss: 0.14178675413131714 2023-01-23 16:34:07.925493: step: 322/1541, loss: 4.448357503861189e-05 2023-01-23 16:34:08.811710: step: 324/1541, loss: 0.003795438911765814 2023-01-23 16:34:09.709816: step: 326/1541, loss: 0.006118211895227432 2023-01-23 16:34:10.611450: step: 328/1541, loss: 0.08378323912620544 2023-01-23 16:34:11.503770: step: 330/1541, loss: 0.10469796508550644 2023-01-23 16:34:12.359627: step: 332/1541, loss: 0.003939446993172169 2023-01-23 16:34:13.267557: step: 334/1541, loss: 0.45642536878585815 2023-01-23 16:34:14.184902: step: 336/1541, loss: 0.48136213421821594 2023-01-23 16:34:15.084721: step: 338/1541, loss: 0.10725684463977814 2023-01-23 16:34:15.974256: step: 340/1541, loss: 0.12794965505599976 2023-01-23 16:34:16.863474: step: 342/1541, loss: 0.04873956739902496 2023-01-23 16:34:17.769379: step: 344/1541, loss: 0.032074011862277985 2023-01-23 16:34:18.654894: step: 346/1541, loss: 0.032744813710451126 2023-01-23 16:34:19.557465: step: 348/1541, loss: 0.07476476579904556 2023-01-23 16:34:20.445674: step: 350/1541, loss: 0.03385088965296745 2023-01-23 16:34:21.336690: step: 352/1541, loss: 0.04156910628080368 2023-01-23 16:34:22.231785: step: 354/1541, loss: 1.1623460054397583 2023-01-23 16:34:23.196025: step: 356/1541, loss: 0.0928424820303917 2023-01-23 16:34:24.096175: step: 358/1541, loss: 0.0006249170401133597 2023-01-23 16:34:24.995306: step: 360/1541, loss: 0.12067731469869614 2023-01-23 16:34:25.918658: step: 362/1541, loss: 0.019031036645174026 2023-01-23 16:34:26.789007: step: 364/1541, loss: 0.00034210525336675346 2023-01-23 16:34:27.685562: step: 366/1541, loss: 0.09257976710796356 2023-01-23 16:34:28.569717: step: 368/1541, loss: 0.0700044184923172 2023-01-23 16:34:29.447695: step: 370/1541, loss: 0.00034605927066877484 2023-01-23 16:34:30.370436: step: 372/1541, loss: 0.0007150162127800286 2023-01-23 16:34:31.267976: step: 374/1541, loss: 0.2056220918893814 2023-01-23 16:34:32.220513: step: 376/1541, loss: 0.0015670580323785543 2023-01-23 16:34:33.088864: step: 378/1541, loss: 0.009474591352045536 2023-01-23 16:34:34.036073: step: 380/1541, loss: 0.009122923016548157 2023-01-23 16:34:34.938045: step: 382/1541, loss: 0.037838537245988846 2023-01-23 16:34:35.892522: step: 384/1541, loss: 0.0053885094821453094 2023-01-23 16:34:36.791322: step: 386/1541, loss: 0.005522522144019604 2023-01-23 16:34:37.674716: step: 388/1541, loss: 0.0021192028652876616 2023-01-23 16:34:38.588238: step: 390/1541, loss: 0.015512067824602127 2023-01-23 16:34:39.486516: step: 392/1541, loss: 0.0018679642817005515 2023-01-23 16:34:40.460882: step: 394/1541, loss: 0.015179788693785667 2023-01-23 16:34:41.352179: step: 396/1541, loss: 0.0016506932443007827 2023-01-23 16:34:42.294294: step: 398/1541, loss: 0.0002503073192201555 2023-01-23 16:34:43.212183: step: 400/1541, loss: 0.1637727916240692 2023-01-23 16:34:44.100726: step: 402/1541, loss: 0.0005901094991713762 2023-01-23 16:34:45.005325: step: 404/1541, loss: 0.08077862858772278 2023-01-23 16:34:45.886857: step: 406/1541, loss: 0.0005951955681666732 2023-01-23 16:34:46.810090: step: 408/1541, loss: 0.00023800278722774237 2023-01-23 16:34:47.715795: step: 410/1541, loss: 0.00029538606759160757 2023-01-23 16:34:48.584544: step: 412/1541, loss: 0.01715558022260666 2023-01-23 16:34:49.453533: step: 414/1541, loss: 0.006031776778399944 2023-01-23 16:34:50.316846: step: 416/1541, loss: 0.18116988241672516 2023-01-23 16:34:51.220660: step: 418/1541, loss: 0.311274915933609 2023-01-23 16:34:52.199806: step: 420/1541, loss: 0.1771896332502365 2023-01-23 16:34:53.102333: step: 422/1541, loss: 0.0010538615752011538 2023-01-23 16:34:53.987301: step: 424/1541, loss: 0.00028894771821796894 2023-01-23 16:34:54.864711: step: 426/1541, loss: 0.0025976665783673525 2023-01-23 16:34:55.760157: step: 428/1541, loss: 0.022529205307364464 2023-01-23 16:34:56.698862: step: 430/1541, loss: 0.0003647713456302881 2023-01-23 16:34:57.601053: step: 432/1541, loss: 0.010676836594939232 2023-01-23 16:34:58.470906: step: 434/1541, loss: 0.007545890286564827 2023-01-23 16:34:59.354817: step: 436/1541, loss: 0.2617137134075165 2023-01-23 16:35:00.299045: step: 438/1541, loss: 0.00032418372575193644 2023-01-23 16:35:01.174339: step: 440/1541, loss: 0.3278859555721283 2023-01-23 16:35:02.076374: step: 442/1541, loss: 0.0002780067443381995 2023-01-23 16:35:02.967015: step: 444/1541, loss: 0.005894177593290806 2023-01-23 16:35:03.858356: step: 446/1541, loss: 0.017601706087589264 2023-01-23 16:35:04.739337: step: 448/1541, loss: 0.01715470664203167 2023-01-23 16:35:05.601051: step: 450/1541, loss: 0.20000672340393066 2023-01-23 16:35:06.466221: step: 452/1541, loss: 0.002491770312190056 2023-01-23 16:35:07.375810: step: 454/1541, loss: 0.31336647272109985 2023-01-23 16:35:08.260397: step: 456/1541, loss: 0.02950304001569748 2023-01-23 16:35:09.110508: step: 458/1541, loss: 0.0023110127076506615 2023-01-23 16:35:09.970340: step: 460/1541, loss: 0.0032430400606244802 2023-01-23 16:35:10.891294: step: 462/1541, loss: 0.41652485728263855 2023-01-23 16:35:11.802424: step: 464/1541, loss: 0.01636035367846489 2023-01-23 16:35:12.726332: step: 466/1541, loss: 0.00046258149086497724 2023-01-23 16:35:13.635956: step: 468/1541, loss: 0.0002691982954274863 2023-01-23 16:35:14.522644: step: 470/1541, loss: 0.11807958781719208 2023-01-23 16:35:15.423818: step: 472/1541, loss: 0.014009243808686733 2023-01-23 16:35:16.386659: step: 474/1541, loss: 0.0014062428381294012 2023-01-23 16:35:17.242193: step: 476/1541, loss: 0.0006730027962476015 2023-01-23 16:35:18.140769: step: 478/1541, loss: 0.05939275398850441 2023-01-23 16:35:19.119830: step: 480/1541, loss: 0.22958914935588837 2023-01-23 16:35:19.991115: step: 482/1541, loss: 0.0018220957135781646 2023-01-23 16:35:20.872989: step: 484/1541, loss: 0.05681275948882103 2023-01-23 16:35:21.754416: step: 486/1541, loss: 0.07708172500133514 2023-01-23 16:35:22.643879: step: 488/1541, loss: 0.001141284708864987 2023-01-23 16:35:23.531703: step: 490/1541, loss: 0.006770928855985403 2023-01-23 16:35:24.501973: step: 492/1541, loss: 0.009858203120529652 2023-01-23 16:35:25.396587: step: 494/1541, loss: 0.007187758106738329 2023-01-23 16:35:26.263766: step: 496/1541, loss: 0.03302811086177826 2023-01-23 16:35:27.138523: step: 498/1541, loss: 0.2664012312889099 2023-01-23 16:35:28.025538: step: 500/1541, loss: 0.008073817007243633 2023-01-23 16:35:28.933009: step: 502/1541, loss: 0.1044309064745903 2023-01-23 16:35:29.807219: step: 504/1541, loss: 0.010627226904034615 2023-01-23 16:35:30.722039: step: 506/1541, loss: 0.005739713087677956 2023-01-23 16:35:31.603706: step: 508/1541, loss: 0.009846570901572704 2023-01-23 16:35:32.532920: step: 510/1541, loss: 0.0006298763328231871 2023-01-23 16:35:33.469257: step: 512/1541, loss: 0.013433612883090973 2023-01-23 16:35:34.348396: step: 514/1541, loss: 0.007891128771007061 2023-01-23 16:35:35.263112: step: 516/1541, loss: 0.0007729228236712515 2023-01-23 16:35:36.161895: step: 518/1541, loss: 0.17590981721878052 2023-01-23 16:35:37.006257: step: 520/1541, loss: 0.003692865138873458 2023-01-23 16:35:37.860266: step: 522/1541, loss: 0.0010788047220557928 2023-01-23 16:35:38.759858: step: 524/1541, loss: 0.0003267838037572801 2023-01-23 16:35:39.636201: step: 526/1541, loss: 0.0073018791154026985 2023-01-23 16:35:40.526783: step: 528/1541, loss: 0.05297531560063362 2023-01-23 16:35:41.422027: step: 530/1541, loss: 0.0015250161522999406 2023-01-23 16:35:42.337314: step: 532/1541, loss: 0.23008979856967926 2023-01-23 16:35:43.177304: step: 534/1541, loss: 0.06748063862323761 2023-01-23 16:35:44.082130: step: 536/1541, loss: 0.028110932558774948 2023-01-23 16:35:44.979592: step: 538/1541, loss: 0.0007032565772533417 2023-01-23 16:35:45.883827: step: 540/1541, loss: 0.004060680512338877 2023-01-23 16:35:46.825239: step: 542/1541, loss: 0.005607302300632 2023-01-23 16:35:47.766994: step: 544/1541, loss: 0.04840812832117081 2023-01-23 16:35:48.775768: step: 546/1541, loss: 0.08830166608095169 2023-01-23 16:35:49.646638: step: 548/1541, loss: 0.006034775171428919 2023-01-23 16:35:50.502757: step: 550/1541, loss: 0.004346185829490423 2023-01-23 16:35:51.378187: step: 552/1541, loss: 0.0027459512930363417 2023-01-23 16:35:52.273737: step: 554/1541, loss: 0.004285333212465048 2023-01-23 16:35:53.133553: step: 556/1541, loss: 0.15466713905334473 2023-01-23 16:35:54.043007: step: 558/1541, loss: 0.007206480950117111 2023-01-23 16:35:54.955848: step: 560/1541, loss: 0.003580672200769186 2023-01-23 16:35:55.854841: step: 562/1541, loss: 0.13938316702842712 2023-01-23 16:35:56.711296: step: 564/1541, loss: 0.44309401512145996 2023-01-23 16:35:57.631319: step: 566/1541, loss: 0.01832052879035473 2023-01-23 16:35:58.570723: step: 568/1541, loss: 0.0012809158070012927 2023-01-23 16:35:59.472948: step: 570/1541, loss: 0.011089926585555077 2023-01-23 16:36:00.370642: step: 572/1541, loss: 0.15065337717533112 2023-01-23 16:36:01.338595: step: 574/1541, loss: 0.0038280123844742775 2023-01-23 16:36:02.230833: step: 576/1541, loss: 0.07749196887016296 2023-01-23 16:36:03.187481: step: 578/1541, loss: 0.5098005533218384 2023-01-23 16:36:04.093573: step: 580/1541, loss: 0.013527458533644676 2023-01-23 16:36:05.018314: step: 582/1541, loss: 0.24044004082679749 2023-01-23 16:36:05.903853: step: 584/1541, loss: 0.0783698707818985 2023-01-23 16:36:06.837503: step: 586/1541, loss: 0.0015911351656541228 2023-01-23 16:36:07.732940: step: 588/1541, loss: 0.12698949873447418 2023-01-23 16:36:08.646949: step: 590/1541, loss: 0.024337580427527428 2023-01-23 16:36:09.604784: step: 592/1541, loss: 0.00899376068264246 2023-01-23 16:36:10.524944: step: 594/1541, loss: 0.07900243252515793 2023-01-23 16:36:11.395597: step: 596/1541, loss: 0.005153698846697807 2023-01-23 16:36:12.304560: step: 598/1541, loss: 0.00011811895819846541 2023-01-23 16:36:13.182964: step: 600/1541, loss: 0.05939727649092674 2023-01-23 16:36:14.083592: step: 602/1541, loss: 0.06829991936683655 2023-01-23 16:36:15.018327: step: 604/1541, loss: 0.025824198499321938 2023-01-23 16:36:15.929304: step: 606/1541, loss: 0.0005962091963738203 2023-01-23 16:36:16.839740: step: 608/1541, loss: 0.006237284746021032 2023-01-23 16:36:17.726426: step: 610/1541, loss: 0.01025342382490635 2023-01-23 16:36:18.639956: step: 612/1541, loss: 0.2532631456851959 2023-01-23 16:36:19.626698: step: 614/1541, loss: 0.1090756505727768 2023-01-23 16:36:20.567369: step: 616/1541, loss: 0.29281795024871826 2023-01-23 16:36:21.580183: step: 618/1541, loss: 0.18356554210186005 2023-01-23 16:36:22.478631: step: 620/1541, loss: 0.020569350570440292 2023-01-23 16:36:23.483333: step: 622/1541, loss: 0.025393841788172722 2023-01-23 16:36:24.386074: step: 624/1541, loss: 0.0003961769980378449 2023-01-23 16:36:25.283719: step: 626/1541, loss: 0.2769681215286255 2023-01-23 16:36:26.164224: step: 628/1541, loss: 0.005960600450634956 2023-01-23 16:36:27.094159: step: 630/1541, loss: 0.21619735658168793 2023-01-23 16:36:27.959218: step: 632/1541, loss: 0.004966884385794401 2023-01-23 16:36:28.833043: step: 634/1541, loss: 0.036817263811826706 2023-01-23 16:36:29.768058: step: 636/1541, loss: 0.007310793735086918 2023-01-23 16:36:30.642189: step: 638/1541, loss: 0.009071754291653633 2023-01-23 16:36:31.529912: step: 640/1541, loss: 0.005295434966683388 2023-01-23 16:36:32.451297: step: 642/1541, loss: 0.24216273427009583 2023-01-23 16:36:33.358737: step: 644/1541, loss: 0.009152496233582497 2023-01-23 16:36:34.233183: step: 646/1541, loss: 0.4002354145050049 2023-01-23 16:36:35.115262: step: 648/1541, loss: 0.009253068827092648 2023-01-23 16:36:36.083079: step: 650/1541, loss: 0.08795765042304993 2023-01-23 16:36:36.989586: step: 652/1541, loss: 0.04103876277804375 2023-01-23 16:36:37.859022: step: 654/1541, loss: 0.004195952322334051 2023-01-23 16:36:38.772234: step: 656/1541, loss: 0.0013462260831147432 2023-01-23 16:36:39.683205: step: 658/1541, loss: 0.0005706967785954475 2023-01-23 16:36:40.570110: step: 660/1541, loss: 0.045458581298589706 2023-01-23 16:36:41.442813: step: 662/1541, loss: 0.03716332092881203 2023-01-23 16:36:42.322217: step: 664/1541, loss: 0.1677381992340088 2023-01-23 16:36:43.222776: step: 666/1541, loss: 0.0006051656673662364 2023-01-23 16:36:44.129839: step: 668/1541, loss: 0.04908563196659088 2023-01-23 16:36:45.042367: step: 670/1541, loss: 0.01463146973401308 2023-01-23 16:36:45.937155: step: 672/1541, loss: 0.0003433425445109606 2023-01-23 16:36:46.830415: step: 674/1541, loss: 0.004583996254950762 2023-01-23 16:36:47.709555: step: 676/1541, loss: 0.017961392179131508 2023-01-23 16:36:48.586399: step: 678/1541, loss: 0.09717618674039841 2023-01-23 16:36:49.556062: step: 680/1541, loss: 0.5194641947746277 2023-01-23 16:36:50.463537: step: 682/1541, loss: 0.0010101047810167074 2023-01-23 16:36:51.352639: step: 684/1541, loss: 0.24015262722969055 2023-01-23 16:36:52.217105: step: 686/1541, loss: 0.04455508664250374 2023-01-23 16:36:53.050440: step: 688/1541, loss: 0.0566990002989769 2023-01-23 16:36:53.926148: step: 690/1541, loss: 0.003094837535172701 2023-01-23 16:36:54.809032: step: 692/1541, loss: 0.03001365438103676 2023-01-23 16:36:55.762026: step: 694/1541, loss: 0.012241321615874767 2023-01-23 16:36:56.621600: step: 696/1541, loss: 0.004299870692193508 2023-01-23 16:36:57.487826: step: 698/1541, loss: 0.0032254806719720364 2023-01-23 16:36:58.366967: step: 700/1541, loss: 0.015331544913351536 2023-01-23 16:36:59.322613: step: 702/1541, loss: 0.0037452643737196922 2023-01-23 16:37:00.258062: step: 704/1541, loss: 0.02903095632791519 2023-01-23 16:37:01.156146: step: 706/1541, loss: 0.01683325134217739 2023-01-23 16:37:02.040340: step: 708/1541, loss: 0.0022669038735330105 2023-01-23 16:37:02.928575: step: 710/1541, loss: 0.05397775024175644 2023-01-23 16:37:03.827683: step: 712/1541, loss: 0.10644359141588211 2023-01-23 16:37:04.726980: step: 714/1541, loss: 0.03468627110123634 2023-01-23 16:37:05.660281: step: 716/1541, loss: 0.003062833333387971 2023-01-23 16:37:06.561889: step: 718/1541, loss: 0.007281604688614607 2023-01-23 16:37:07.423732: step: 720/1541, loss: 0.01827489212155342 2023-01-23 16:37:08.361255: step: 722/1541, loss: 0.00536778848618269 2023-01-23 16:37:09.233904: step: 724/1541, loss: 0.2765292823314667 2023-01-23 16:37:10.070758: step: 726/1541, loss: 0.01492113433778286 2023-01-23 16:37:10.955703: step: 728/1541, loss: 0.022746212780475616 2023-01-23 16:37:11.899510: step: 730/1541, loss: 0.0026957904919981956 2023-01-23 16:37:12.813471: step: 732/1541, loss: 0.03435587137937546 2023-01-23 16:37:13.681821: step: 734/1541, loss: 0.0018749006558209658 2023-01-23 16:37:14.542991: step: 736/1541, loss: 0.11871157586574554 2023-01-23 16:37:15.404838: step: 738/1541, loss: 0.12525859475135803 2023-01-23 16:37:16.323603: step: 740/1541, loss: 0.13534417748451233 2023-01-23 16:37:17.203166: step: 742/1541, loss: 0.23576906323432922 2023-01-23 16:37:18.115964: step: 744/1541, loss: 0.019274163991212845 2023-01-23 16:37:19.070360: step: 746/1541, loss: 0.043921247124671936 2023-01-23 16:37:19.997008: step: 748/1541, loss: 0.0013151813764125109 2023-01-23 16:37:20.972849: step: 750/1541, loss: 0.29849714040756226 2023-01-23 16:37:21.876226: step: 752/1541, loss: 0.29462724924087524 2023-01-23 16:37:22.775529: step: 754/1541, loss: 0.0070139397867023945 2023-01-23 16:37:23.682184: step: 756/1541, loss: 0.0018366981530562043 2023-01-23 16:37:24.515468: step: 758/1541, loss: 0.0008579011191613972 2023-01-23 16:37:25.409201: step: 760/1541, loss: 0.005499997641891241 2023-01-23 16:37:26.360242: step: 762/1541, loss: 0.0035602031275629997 2023-01-23 16:37:27.253509: step: 764/1541, loss: 0.7281023859977722 2023-01-23 16:37:28.182852: step: 766/1541, loss: 0.24387675523757935 2023-01-23 16:37:29.108348: step: 768/1541, loss: 0.0006192871951498091 2023-01-23 16:37:30.024962: step: 770/1541, loss: 0.026226121932268143 2023-01-23 16:37:30.939614: step: 772/1541, loss: 0.04279667139053345 2023-01-23 16:37:31.860176: step: 774/1541, loss: 0.0039068893529474735 2023-01-23 16:37:32.833360: step: 776/1541, loss: 0.00980593916028738 2023-01-23 16:37:33.728701: step: 778/1541, loss: 0.0013938014162704349 2023-01-23 16:37:34.627466: step: 780/1541, loss: 0.00046565363300032914 2023-01-23 16:37:35.584205: step: 782/1541, loss: 0.057376038283109665 2023-01-23 16:37:36.503057: step: 784/1541, loss: 0.004703829996287823 2023-01-23 16:37:37.349475: step: 786/1541, loss: 0.10970663279294968 2023-01-23 16:37:38.251967: step: 788/1541, loss: 0.0011986260069534183 2023-01-23 16:37:39.144816: step: 790/1541, loss: 0.00454944372177124 2023-01-23 16:37:40.045764: step: 792/1541, loss: 0.0009828691836446524 2023-01-23 16:37:40.979041: step: 794/1541, loss: 0.0011082235723733902 2023-01-23 16:37:41.886085: step: 796/1541, loss: 0.0026215137913823128 2023-01-23 16:37:42.747780: step: 798/1541, loss: 0.023158501833677292 2023-01-23 16:37:43.640707: step: 800/1541, loss: 0.007420246955007315 2023-01-23 16:37:44.477566: step: 802/1541, loss: 3.543306956999004e-05 2023-01-23 16:37:45.389946: step: 804/1541, loss: 0.16819392144680023 2023-01-23 16:37:46.278747: step: 806/1541, loss: 0.1429135501384735 2023-01-23 16:37:47.173359: step: 808/1541, loss: 0.02040819078683853 2023-01-23 16:37:48.061783: step: 810/1541, loss: 0.0004338048165664077 2023-01-23 16:37:48.971030: step: 812/1541, loss: 0.0008410066366195679 2023-01-23 16:37:49.902820: step: 814/1541, loss: 0.001956369262188673 2023-01-23 16:37:50.832627: step: 816/1541, loss: 0.16139554977416992 2023-01-23 16:37:51.793460: step: 818/1541, loss: 0.00023639935534447432 2023-01-23 16:37:52.837508: step: 820/1541, loss: 0.0595226064324379 2023-01-23 16:37:53.740419: step: 822/1541, loss: 0.08624310791492462 2023-01-23 16:37:54.586975: step: 824/1541, loss: 0.4476704001426697 2023-01-23 16:37:55.607968: step: 826/1541, loss: 0.0002742526412475854 2023-01-23 16:37:56.583154: step: 828/1541, loss: 0.0005188885261304677 2023-01-23 16:37:57.471584: step: 830/1541, loss: 0.0001532789901830256 2023-01-23 16:37:58.359805: step: 832/1541, loss: 0.000966308347415179 2023-01-23 16:37:59.364280: step: 834/1541, loss: 0.049188680946826935 2023-01-23 16:38:00.240755: step: 836/1541, loss: 0.0040433271788060665 2023-01-23 16:38:01.140557: step: 838/1541, loss: 0.0004418098251335323 2023-01-23 16:38:02.009104: step: 840/1541, loss: 0.0011557663092389703 2023-01-23 16:38:02.936927: step: 842/1541, loss: 0.10424745827913284 2023-01-23 16:38:03.904098: step: 844/1541, loss: 0.0011858305661007762 2023-01-23 16:38:04.813207: step: 846/1541, loss: 0.07500527054071426 2023-01-23 16:38:05.688692: step: 848/1541, loss: 0.0011108387261629105 2023-01-23 16:38:06.635911: step: 850/1541, loss: 3.4674376365728676e-05 2023-01-23 16:38:07.541795: step: 852/1541, loss: 0.007677174173295498 2023-01-23 16:38:08.449703: step: 854/1541, loss: 0.001205773209221661 2023-01-23 16:38:09.343198: step: 856/1541, loss: 0.03836657479405403 2023-01-23 16:38:10.173060: step: 858/1541, loss: 0.0011713819112628698 2023-01-23 16:38:11.075756: step: 860/1541, loss: 0.0006798238609917462 2023-01-23 16:38:11.972810: step: 862/1541, loss: 0.16714121401309967 2023-01-23 16:38:12.890870: step: 864/1541, loss: 0.0006797309033572674 2023-01-23 16:38:13.788536: step: 866/1541, loss: 0.022625526413321495 2023-01-23 16:38:14.679417: step: 868/1541, loss: 0.1732037514448166 2023-01-23 16:38:15.541069: step: 870/1541, loss: 0.20852123200893402 2023-01-23 16:38:16.408811: step: 872/1541, loss: 0.1286676824092865 2023-01-23 16:38:17.283705: step: 874/1541, loss: 0.12673451006412506 2023-01-23 16:38:18.212316: step: 876/1541, loss: 0.1294827163219452 2023-01-23 16:38:19.105563: step: 878/1541, loss: 0.002217118162661791 2023-01-23 16:38:20.032744: step: 880/1541, loss: 0.002182441996410489 2023-01-23 16:38:20.923354: step: 882/1541, loss: 0.00021899183047935367 2023-01-23 16:38:21.908933: step: 884/1541, loss: 0.0003345248114783317 2023-01-23 16:38:22.808146: step: 886/1541, loss: 0.004254736006259918 2023-01-23 16:38:23.697564: step: 888/1541, loss: 0.030166897922754288 2023-01-23 16:38:24.639982: step: 890/1541, loss: 0.00013819278683513403 2023-01-23 16:38:25.576920: step: 892/1541, loss: 0.016995588317513466 2023-01-23 16:38:26.510102: step: 894/1541, loss: 0.007547643035650253 2023-01-23 16:38:27.442261: step: 896/1541, loss: 0.00015601808263454586 2023-01-23 16:38:28.357635: step: 898/1541, loss: 0.000778272224124521 2023-01-23 16:38:29.269140: step: 900/1541, loss: 0.00014118019316811115 2023-01-23 16:38:30.155318: step: 902/1541, loss: 0.4848603308200836 2023-01-23 16:38:31.049589: step: 904/1541, loss: 0.0044134631752967834 2023-01-23 16:38:31.950192: step: 906/1541, loss: 0.07064849138259888 2023-01-23 16:38:32.899876: step: 908/1541, loss: 0.326805055141449 2023-01-23 16:38:33.760233: step: 910/1541, loss: 0.26750731468200684 2023-01-23 16:38:34.656792: step: 912/1541, loss: 0.08141420781612396 2023-01-23 16:38:35.551589: step: 914/1541, loss: 0.030568638816475868 2023-01-23 16:38:36.435629: step: 916/1541, loss: 0.008095523342490196 2023-01-23 16:38:37.323007: step: 918/1541, loss: 0.016057511791586876 2023-01-23 16:38:38.227846: step: 920/1541, loss: 0.09712021797895432 2023-01-23 16:38:39.129731: step: 922/1541, loss: 2.206995850428939e-05 2023-01-23 16:38:40.046819: step: 924/1541, loss: 0.08608758449554443 2023-01-23 16:38:40.954749: step: 926/1541, loss: 0.06784924119710922 2023-01-23 16:38:41.844844: step: 928/1541, loss: 0.07539325952529907 2023-01-23 16:38:42.683779: step: 930/1541, loss: 0.03398963063955307 2023-01-23 16:38:43.569294: step: 932/1541, loss: 0.00807272456586361 2023-01-23 16:38:44.526277: step: 934/1541, loss: 4.874001751886681e-05 2023-01-23 16:38:45.399833: step: 936/1541, loss: 0.24504411220550537 2023-01-23 16:38:46.302617: step: 938/1541, loss: 0.0005994406528770924 2023-01-23 16:38:47.280823: step: 940/1541, loss: 0.30043742060661316 2023-01-23 16:38:48.182624: step: 942/1541, loss: 0.11045171320438385 2023-01-23 16:38:49.092800: step: 944/1541, loss: 0.0021599149331450462 2023-01-23 16:38:50.049881: step: 946/1541, loss: 0.18011443316936493 2023-01-23 16:38:50.941357: step: 948/1541, loss: 0.08951185643672943 2023-01-23 16:38:51.799717: step: 950/1541, loss: 0.00010104528337251395 2023-01-23 16:38:52.717093: step: 952/1541, loss: 0.0007663284195587039 2023-01-23 16:38:53.677832: step: 954/1541, loss: 0.06866961717605591 2023-01-23 16:38:54.583763: step: 956/1541, loss: 0.19790521264076233 2023-01-23 16:38:55.465799: step: 958/1541, loss: 0.020176995545625687 2023-01-23 16:38:56.356000: step: 960/1541, loss: 0.011761639267206192 2023-01-23 16:38:57.242820: step: 962/1541, loss: 0.0030562083702534437 2023-01-23 16:38:58.116483: step: 964/1541, loss: 0.0006114394636824727 2023-01-23 16:38:58.975126: step: 966/1541, loss: 0.2864038348197937 2023-01-23 16:38:59.837619: step: 968/1541, loss: 0.005853273440152407 2023-01-23 16:39:00.721552: step: 970/1541, loss: 0.1562364250421524 2023-01-23 16:39:01.567214: step: 972/1541, loss: 0.005178349558264017 2023-01-23 16:39:02.486021: step: 974/1541, loss: 0.02937413565814495 2023-01-23 16:39:03.355867: step: 976/1541, loss: 0.0027907760813832283 2023-01-23 16:39:04.224617: step: 978/1541, loss: 0.08197516947984695 2023-01-23 16:39:05.126454: step: 980/1541, loss: 0.0032885074615478516 2023-01-23 16:39:06.054516: step: 982/1541, loss: 0.02781674452126026 2023-01-23 16:39:07.056417: step: 984/1541, loss: 3.513384945108555e-05 2023-01-23 16:39:07.943626: step: 986/1541, loss: 0.11237964779138565 2023-01-23 16:39:08.812367: step: 988/1541, loss: 0.38261935114860535 2023-01-23 16:39:09.669776: step: 990/1541, loss: 0.0018995316931977868 2023-01-23 16:39:10.597799: step: 992/1541, loss: 0.000156837806571275 2023-01-23 16:39:11.465649: step: 994/1541, loss: 0.0002345453976886347 2023-01-23 16:39:12.331015: step: 996/1541, loss: 0.003451436525210738 2023-01-23 16:39:13.221635: step: 998/1541, loss: 0.00016337164561264217 2023-01-23 16:39:14.102298: step: 1000/1541, loss: 0.007632952183485031 2023-01-23 16:39:14.959190: step: 1002/1541, loss: 0.98688805103302 2023-01-23 16:39:15.888606: step: 1004/1541, loss: 0.003385161282494664 2023-01-23 16:39:16.868599: step: 1006/1541, loss: 0.000461987016024068 2023-01-23 16:39:17.801494: step: 1008/1541, loss: 0.24991439282894135 2023-01-23 16:39:18.686267: step: 1010/1541, loss: 0.020575085654854774 2023-01-23 16:39:19.595418: step: 1012/1541, loss: 0.021648671478033066 2023-01-23 16:39:20.543660: step: 1014/1541, loss: 0.07571254670619965 2023-01-23 16:39:21.443224: step: 1016/1541, loss: 0.20920546352863312 2023-01-23 16:39:22.343831: step: 1018/1541, loss: 0.0004080641083419323 2023-01-23 16:39:23.256894: step: 1020/1541, loss: 0.21499799191951752 2023-01-23 16:39:24.143121: step: 1022/1541, loss: 0.00044668943155556917 2023-01-23 16:39:25.026666: step: 1024/1541, loss: 0.043692197650671005 2023-01-23 16:39:25.914569: step: 1026/1541, loss: 7.320987060666084e-05 2023-01-23 16:39:26.807120: step: 1028/1541, loss: 0.10754271596670151 2023-01-23 16:39:27.709411: step: 1030/1541, loss: 0.00351432291790843 2023-01-23 16:39:28.666762: step: 1032/1541, loss: 0.013441421091556549 2023-01-23 16:39:29.550138: step: 1034/1541, loss: 0.013012171722948551 2023-01-23 16:39:30.444985: step: 1036/1541, loss: 0.006485351826995611 2023-01-23 16:39:31.352834: step: 1038/1541, loss: 0.6673450469970703 2023-01-23 16:39:32.243383: step: 1040/1541, loss: 0.019726179540157318 2023-01-23 16:39:33.128180: step: 1042/1541, loss: 0.14819906651973724 2023-01-23 16:39:34.126522: step: 1044/1541, loss: 0.0008211468812078238 2023-01-23 16:39:34.990558: step: 1046/1541, loss: 0.02089880406856537 2023-01-23 16:39:35.903603: step: 1048/1541, loss: 0.009287149645388126 2023-01-23 16:39:36.861439: step: 1050/1541, loss: 0.19535058736801147 2023-01-23 16:39:37.785913: step: 1052/1541, loss: 0.2676507234573364 2023-01-23 16:39:38.741947: step: 1054/1541, loss: 2.0228323936462402 2023-01-23 16:39:39.655506: step: 1056/1541, loss: 0.001546818995848298 2023-01-23 16:39:40.620545: step: 1058/1541, loss: 0.09536373615264893 2023-01-23 16:39:41.498678: step: 1060/1541, loss: 2.3277695178985596 2023-01-23 16:39:42.398372: step: 1062/1541, loss: 0.11080486327409744 2023-01-23 16:39:43.324582: step: 1064/1541, loss: 0.2162330150604248 2023-01-23 16:39:44.242288: step: 1066/1541, loss: 0.029630890116095543 2023-01-23 16:39:45.150631: step: 1068/1541, loss: 0.037934038788080215 2023-01-23 16:39:46.020439: step: 1070/1541, loss: 0.07271014153957367 2023-01-23 16:39:46.903982: step: 1072/1541, loss: 0.07689985632896423 2023-01-23 16:39:47.813466: step: 1074/1541, loss: 0.01042851060628891 2023-01-23 16:39:48.676994: step: 1076/1541, loss: 0.4771064221858978 2023-01-23 16:39:49.579361: step: 1078/1541, loss: 0.14686445891857147 2023-01-23 16:39:50.500296: step: 1080/1541, loss: 0.3794073760509491 2023-01-23 16:39:51.412548: step: 1082/1541, loss: 0.005595182999968529 2023-01-23 16:39:52.344588: step: 1084/1541, loss: 0.014684570021927357 2023-01-23 16:39:53.271361: step: 1086/1541, loss: 0.0026142969727516174 2023-01-23 16:39:54.152047: step: 1088/1541, loss: 0.0066042328253388405 2023-01-23 16:39:55.050851: step: 1090/1541, loss: 0.013139374554157257 2023-01-23 16:39:55.996330: step: 1092/1541, loss: 0.04505283012986183 2023-01-23 16:39:56.926154: step: 1094/1541, loss: 0.0013449187390506268 2023-01-23 16:39:57.821107: step: 1096/1541, loss: 1.137534499168396 2023-01-23 16:39:58.717092: step: 1098/1541, loss: 0.09317561239004135 2023-01-23 16:39:59.594876: step: 1100/1541, loss: 0.000319094950100407 2023-01-23 16:40:00.487830: step: 1102/1541, loss: 0.0029373096767812967 2023-01-23 16:40:01.364380: step: 1104/1541, loss: 0.46054452657699585 2023-01-23 16:40:02.291225: step: 1106/1541, loss: 0.04384039342403412 2023-01-23 16:40:03.205451: step: 1108/1541, loss: 0.13659276068210602 2023-01-23 16:40:04.128684: step: 1110/1541, loss: 0.0381629504263401 2023-01-23 16:40:05.045062: step: 1112/1541, loss: 0.005406643263995647 2023-01-23 16:40:05.956518: step: 1114/1541, loss: 0.031953953206539154 2023-01-23 16:40:06.935328: step: 1116/1541, loss: 0.005688961129635572 2023-01-23 16:40:07.834366: step: 1118/1541, loss: 0.0648050382733345 2023-01-23 16:40:08.724906: step: 1120/1541, loss: 0.00028922202182002366 2023-01-23 16:40:09.640395: step: 1122/1541, loss: 0.12305957823991776 2023-01-23 16:40:10.491579: step: 1124/1541, loss: 0.012297350913286209 2023-01-23 16:40:11.376302: step: 1126/1541, loss: 0.0018436359241604805 2023-01-23 16:40:12.253198: step: 1128/1541, loss: 0.0005269570974633098 2023-01-23 16:40:13.122291: step: 1130/1541, loss: 0.0010291105136275291 2023-01-23 16:40:14.125091: step: 1132/1541, loss: 0.003851449117064476 2023-01-23 16:40:15.020094: step: 1134/1541, loss: 0.023085791617631912 2023-01-23 16:40:15.889064: step: 1136/1541, loss: 0.00020231783855706453 2023-01-23 16:40:16.804439: step: 1138/1541, loss: 0.00272129918448627 2023-01-23 16:40:17.674387: step: 1140/1541, loss: 0.00807900633662939 2023-01-23 16:40:18.534836: step: 1142/1541, loss: 0.0004879531916230917 2023-01-23 16:40:19.492857: step: 1144/1541, loss: 0.09147695451974869 2023-01-23 16:40:20.348857: step: 1146/1541, loss: 0.0007911284337751567 2023-01-23 16:40:21.248947: step: 1148/1541, loss: 0.26604795455932617 2023-01-23 16:40:22.170827: step: 1150/1541, loss: 0.04643304646015167 2023-01-23 16:40:23.076803: step: 1152/1541, loss: 8.956249075708911e-05 2023-01-23 16:40:23.945710: step: 1154/1541, loss: 0.07365468889474869 2023-01-23 16:40:24.828363: step: 1156/1541, loss: 0.0008381722727790475 2023-01-23 16:40:25.735930: step: 1158/1541, loss: 0.004507175646722317 2023-01-23 16:40:26.634540: step: 1160/1541, loss: 0.0042572589591145515 2023-01-23 16:40:27.515039: step: 1162/1541, loss: 0.0008090412593446672 2023-01-23 16:40:28.406080: step: 1164/1541, loss: 0.005958168301731348 2023-01-23 16:40:29.311593: step: 1166/1541, loss: 0.015101325698196888 2023-01-23 16:40:30.192201: step: 1168/1541, loss: 0.01268846821039915 2023-01-23 16:40:31.078102: step: 1170/1541, loss: 0.005552547052502632 2023-01-23 16:40:32.014012: step: 1172/1541, loss: 0.003238109638914466 2023-01-23 16:40:32.941932: step: 1174/1541, loss: 0.06554049253463745 2023-01-23 16:40:33.988226: step: 1176/1541, loss: 0.0010072641307488084 2023-01-23 16:40:34.886821: step: 1178/1541, loss: 0.022681772708892822 2023-01-23 16:40:35.783606: step: 1180/1541, loss: 0.013299118727445602 2023-01-23 16:40:36.699268: step: 1182/1541, loss: 0.0005619896692223847 2023-01-23 16:40:37.598701: step: 1184/1541, loss: 0.0032691550441086292 2023-01-23 16:40:38.457143: step: 1186/1541, loss: 0.06754325330257416 2023-01-23 16:40:39.361666: step: 1188/1541, loss: 0.00040362041909247637 2023-01-23 16:40:40.238465: step: 1190/1541, loss: 0.012454687617719173 2023-01-23 16:40:41.142900: step: 1192/1541, loss: 0.431941956281662 2023-01-23 16:40:41.992497: step: 1194/1541, loss: 0.00026601337594911456 2023-01-23 16:40:42.843560: step: 1196/1541, loss: 0.010731217451393604 2023-01-23 16:40:43.739934: step: 1198/1541, loss: 0.02442385070025921 2023-01-23 16:40:44.656181: step: 1200/1541, loss: 0.037428464740514755 2023-01-23 16:40:45.575322: step: 1202/1541, loss: 0.000714463705662638 2023-01-23 16:40:46.446423: step: 1204/1541, loss: 0.0017817944753915071 2023-01-23 16:40:47.362834: step: 1206/1541, loss: 0.13387978076934814 2023-01-23 16:40:48.262670: step: 1208/1541, loss: 0.0029414629098027945 2023-01-23 16:40:49.127875: step: 1210/1541, loss: 0.020244287326931953 2023-01-23 16:40:50.025110: step: 1212/1541, loss: 0.00378479715436697 2023-01-23 16:40:50.916637: step: 1214/1541, loss: 0.0026129065081477165 2023-01-23 16:40:51.785383: step: 1216/1541, loss: 0.002064642496407032 2023-01-23 16:40:52.747353: step: 1218/1541, loss: 0.08603915572166443 2023-01-23 16:40:53.673441: step: 1220/1541, loss: 0.11398612707853317 2023-01-23 16:40:54.569940: step: 1222/1541, loss: 0.022452959790825844 2023-01-23 16:40:55.468407: step: 1224/1541, loss: 0.007382212672382593 2023-01-23 16:40:56.355949: step: 1226/1541, loss: 0.008776457980275154 2023-01-23 16:40:57.284984: step: 1228/1541, loss: 0.27293145656585693 2023-01-23 16:40:58.195167: step: 1230/1541, loss: 0.005669371224939823 2023-01-23 16:40:59.070470: step: 1232/1541, loss: 0.0010782908648252487 2023-01-23 16:40:59.999086: step: 1234/1541, loss: 0.009632028639316559 2023-01-23 16:41:00.885547: step: 1236/1541, loss: 0.0005010518943890929 2023-01-23 16:41:01.811499: step: 1238/1541, loss: 0.3376513719558716 2023-01-23 16:41:02.750025: step: 1240/1541, loss: 0.0030622852500528097 2023-01-23 16:41:03.668536: step: 1242/1541, loss: 2.830679178237915 2023-01-23 16:41:04.559851: step: 1244/1541, loss: 0.026066435500979424 2023-01-23 16:41:05.478108: step: 1246/1541, loss: 1.4113675355911255 2023-01-23 16:41:06.342496: step: 1248/1541, loss: 0.001902944641187787 2023-01-23 16:41:07.235541: step: 1250/1541, loss: 0.030068829655647278 2023-01-23 16:41:08.130908: step: 1252/1541, loss: 0.0071617127396166325 2023-01-23 16:41:09.011162: step: 1254/1541, loss: 0.025452803820371628 2023-01-23 16:41:09.934988: step: 1256/1541, loss: 0.07735293358564377 2023-01-23 16:41:10.841841: step: 1258/1541, loss: 0.0006060547893866897 2023-01-23 16:41:11.713482: step: 1260/1541, loss: 0.34078171849250793 2023-01-23 16:41:12.608419: step: 1262/1541, loss: 0.007140856236219406 2023-01-23 16:41:13.544777: step: 1264/1541, loss: 0.012751313857734203 2023-01-23 16:41:14.425947: step: 1266/1541, loss: 0.009995450265705585 2023-01-23 16:41:15.380192: step: 1268/1541, loss: 1.1937060356140137 2023-01-23 16:41:16.252004: step: 1270/1541, loss: 0.0004360295715741813 2023-01-23 16:41:17.122758: step: 1272/1541, loss: 0.0033031096681952477 2023-01-23 16:41:18.053234: step: 1274/1541, loss: 0.5012995600700378 2023-01-23 16:41:18.906450: step: 1276/1541, loss: 0.014168770983815193 2023-01-23 16:41:19.860154: step: 1278/1541, loss: 0.001519068842753768 2023-01-23 16:41:20.790131: step: 1280/1541, loss: 0.0779804140329361 2023-01-23 16:41:21.691455: step: 1282/1541, loss: 0.5630338788032532 2023-01-23 16:41:22.627382: step: 1284/1541, loss: 0.052147846668958664 2023-01-23 16:41:23.556748: step: 1286/1541, loss: 0.004687546286731958 2023-01-23 16:41:24.486070: step: 1288/1541, loss: 0.006263240706175566 2023-01-23 16:41:25.386429: step: 1290/1541, loss: 0.004890612792223692 2023-01-23 16:41:26.267987: step: 1292/1541, loss: 0.0668722540140152 2023-01-23 16:41:27.151539: step: 1294/1541, loss: 0.0486774779856205 2023-01-23 16:41:28.052319: step: 1296/1541, loss: 0.0006709733279421926 2023-01-23 16:41:28.987498: step: 1298/1541, loss: 0.045783914625644684 2023-01-23 16:41:29.917754: step: 1300/1541, loss: 0.19400538504123688 2023-01-23 16:41:30.794141: step: 1302/1541, loss: 0.048002395778894424 2023-01-23 16:41:31.680046: step: 1304/1541, loss: 0.001360663096420467 2023-01-23 16:41:32.621302: step: 1306/1541, loss: 0.03454286605119705 2023-01-23 16:41:33.521939: step: 1308/1541, loss: 0.002507030265405774 2023-01-23 16:41:34.435657: step: 1310/1541, loss: 0.03409726917743683 2023-01-23 16:41:35.325157: step: 1312/1541, loss: 0.09800107777118683 2023-01-23 16:41:36.243261: step: 1314/1541, loss: 0.11690922826528549 2023-01-23 16:41:37.141407: step: 1316/1541, loss: 0.00602241326123476 2023-01-23 16:41:38.033524: step: 1318/1541, loss: 0.11612454056739807 2023-01-23 16:41:38.948896: step: 1320/1541, loss: 0.0038900510407984257 2023-01-23 16:41:39.860020: step: 1322/1541, loss: 0.0511467382311821 2023-01-23 16:41:40.793776: step: 1324/1541, loss: 0.0039280173368752 2023-01-23 16:41:41.795170: step: 1326/1541, loss: 0.08424047380685806 2023-01-23 16:41:42.714207: step: 1328/1541, loss: 0.028304245322942734 2023-01-23 16:41:43.636209: step: 1330/1541, loss: 0.21988575160503387 2023-01-23 16:41:44.549339: step: 1332/1541, loss: 0.05421926826238632 2023-01-23 16:41:45.396161: step: 1334/1541, loss: 0.00017067517910618335 2023-01-23 16:41:46.288801: step: 1336/1541, loss: 0.003352418541908264 2023-01-23 16:41:47.224500: step: 1338/1541, loss: 0.15547965466976166 2023-01-23 16:41:48.099389: step: 1340/1541, loss: 0.0011248586233705282 2023-01-23 16:41:49.003387: step: 1342/1541, loss: 0.0028899717144668102 2023-01-23 16:41:49.909373: step: 1344/1541, loss: 0.008277783170342445 2023-01-23 16:41:50.772143: step: 1346/1541, loss: 0.018658800050616264 2023-01-23 16:41:51.672647: step: 1348/1541, loss: 0.0016614929772913456 2023-01-23 16:41:52.555721: step: 1350/1541, loss: 0.004519174806773663 2023-01-23 16:41:53.425696: step: 1352/1541, loss: 0.006136383395642042 2023-01-23 16:41:54.275646: step: 1354/1541, loss: 0.1357145756483078 2023-01-23 16:41:55.197680: step: 1356/1541, loss: 0.01698630303144455 2023-01-23 16:41:56.106152: step: 1358/1541, loss: 0.10808134078979492 2023-01-23 16:41:56.985240: step: 1360/1541, loss: 0.16920244693756104 2023-01-23 16:41:57.850144: step: 1362/1541, loss: 0.0025050262920558453 2023-01-23 16:41:58.780095: step: 1364/1541, loss: 0.016322318464517593 2023-01-23 16:41:59.740976: step: 1366/1541, loss: 0.009877420961856842 2023-01-23 16:42:00.668896: step: 1368/1541, loss: 0.05880644917488098 2023-01-23 16:42:01.648673: step: 1370/1541, loss: 0.00286102294921875 2023-01-23 16:42:02.600299: step: 1372/1541, loss: 0.1411971002817154 2023-01-23 16:42:03.426144: step: 1374/1541, loss: 0.057233016937971115 2023-01-23 16:42:04.331105: step: 1376/1541, loss: 0.10740025341510773 2023-01-23 16:42:05.230240: step: 1378/1541, loss: 0.03170914202928543 2023-01-23 16:42:06.175636: step: 1380/1541, loss: 0.3698243796825409 2023-01-23 16:42:07.058529: step: 1382/1541, loss: 0.00203736056573689 2023-01-23 16:42:07.949658: step: 1384/1541, loss: 0.029614374041557312 2023-01-23 16:42:08.869683: step: 1386/1541, loss: 0.07957237213850021 2023-01-23 16:42:09.789848: step: 1388/1541, loss: 0.006090671289712191 2023-01-23 16:42:10.701129: step: 1390/1541, loss: 0.0006994917057454586 2023-01-23 16:42:11.595669: step: 1392/1541, loss: 0.20993778109550476 2023-01-23 16:42:12.465691: step: 1394/1541, loss: 0.02577337436378002 2023-01-23 16:42:13.348555: step: 1396/1541, loss: 0.012169142253696918 2023-01-23 16:42:14.255384: step: 1398/1541, loss: 0.6452802419662476 2023-01-23 16:42:15.170906: step: 1400/1541, loss: 0.0016395860584452748 2023-01-23 16:42:16.039830: step: 1402/1541, loss: 0.07878682017326355 2023-01-23 16:42:16.961614: step: 1404/1541, loss: 0.0013812525430694222 2023-01-23 16:42:17.914119: step: 1406/1541, loss: 0.0015248986892402172 2023-01-23 16:42:18.838899: step: 1408/1541, loss: 0.017327498644590378 2023-01-23 16:42:19.747247: step: 1410/1541, loss: 0.0037177333142608404 2023-01-23 16:42:20.645811: step: 1412/1541, loss: 0.02293904311954975 2023-01-23 16:42:21.549882: step: 1414/1541, loss: 0.01149934995919466 2023-01-23 16:42:22.412589: step: 1416/1541, loss: 0.004289437551051378 2023-01-23 16:42:23.294854: step: 1418/1541, loss: 0.007322582416236401 2023-01-23 16:42:24.192103: step: 1420/1541, loss: 0.009338085539638996 2023-01-23 16:42:25.141911: step: 1422/1541, loss: 0.024990689009428024 2023-01-23 16:42:26.065074: step: 1424/1541, loss: 0.009325862862169743 2023-01-23 16:42:26.965408: step: 1426/1541, loss: 0.06221434101462364 2023-01-23 16:42:27.955688: step: 1428/1541, loss: 0.004668456967920065 2023-01-23 16:42:28.863934: step: 1430/1541, loss: 0.2956673502922058 2023-01-23 16:42:29.748432: step: 1432/1541, loss: 0.0517081618309021 2023-01-23 16:42:30.665691: step: 1434/1541, loss: 0.15447236597537994 2023-01-23 16:42:31.538520: step: 1436/1541, loss: 0.00018339665257371962 2023-01-23 16:42:32.463144: step: 1438/1541, loss: 0.0007877941825427115 2023-01-23 16:42:33.351178: step: 1440/1541, loss: 0.001561999204568565 2023-01-23 16:42:34.277299: step: 1442/1541, loss: 0.0038857231847941875 2023-01-23 16:42:35.222777: step: 1444/1541, loss: 0.005694888532161713 2023-01-23 16:42:36.124997: step: 1446/1541, loss: 0.038460180163383484 2023-01-23 16:42:36.983416: step: 1448/1541, loss: 0.0009573070565238595 2023-01-23 16:42:37.886754: step: 1450/1541, loss: 0.003506365465000272 2023-01-23 16:42:38.797666: step: 1452/1541, loss: 0.8060077428817749 2023-01-23 16:42:39.693943: step: 1454/1541, loss: 0.002458041300997138 2023-01-23 16:42:40.604856: step: 1456/1541, loss: 0.007591617293655872 2023-01-23 16:42:41.486669: step: 1458/1541, loss: 0.00628666253760457 2023-01-23 16:42:42.381166: step: 1460/1541, loss: 0.012981491163372993 2023-01-23 16:42:43.233531: step: 1462/1541, loss: 0.0021086144261062145 2023-01-23 16:42:44.134862: step: 1464/1541, loss: 0.1286022961139679 2023-01-23 16:42:45.023263: step: 1466/1541, loss: 0.014755772426724434 2023-01-23 16:42:45.975118: step: 1468/1541, loss: 0.032039836049079895 2023-01-23 16:42:46.936976: step: 1470/1541, loss: 0.26469191908836365 2023-01-23 16:42:47.827331: step: 1472/1541, loss: 0.25195425748825073 2023-01-23 16:42:48.749975: step: 1474/1541, loss: 0.09715509414672852 2023-01-23 16:42:49.621323: step: 1476/1541, loss: 0.0012867669574916363 2023-01-23 16:42:50.484490: step: 1478/1541, loss: 0.05972990021109581 2023-01-23 16:42:51.447395: step: 1480/1541, loss: 0.048235584050416946 2023-01-23 16:42:52.345618: step: 1482/1541, loss: 0.01433007325977087 2023-01-23 16:42:53.237938: step: 1484/1541, loss: 0.16705463826656342 2023-01-23 16:42:54.107662: step: 1486/1541, loss: 0.005875099450349808 2023-01-23 16:42:55.005280: step: 1488/1541, loss: 0.4240665137767792 2023-01-23 16:42:55.926910: step: 1490/1541, loss: 0.051298532634973526 2023-01-23 16:42:56.841339: step: 1492/1541, loss: 0.019052794203162193 2023-01-23 16:42:57.722837: step: 1494/1541, loss: 0.024263612926006317 2023-01-23 16:42:58.630224: step: 1496/1541, loss: 0.11047683656215668 2023-01-23 16:42:59.524923: step: 1498/1541, loss: 0.002606891794130206 2023-01-23 16:43:00.389795: step: 1500/1541, loss: 0.09042675048112869 2023-01-23 16:43:01.320853: step: 1502/1541, loss: 0.05509011447429657 2023-01-23 16:43:02.236720: step: 1504/1541, loss: 0.005255807191133499 2023-01-23 16:43:03.106241: step: 1506/1541, loss: 0.000855046440847218 2023-01-23 16:43:03.997946: step: 1508/1541, loss: 0.2570219933986664 2023-01-23 16:43:04.882368: step: 1510/1541, loss: 0.011673364788293839 2023-01-23 16:43:05.779822: step: 1512/1541, loss: 0.0190522950142622 2023-01-23 16:43:06.659729: step: 1514/1541, loss: 0.006158594042062759 2023-01-23 16:43:07.587711: step: 1516/1541, loss: 0.07118996977806091 2023-01-23 16:43:08.523380: step: 1518/1541, loss: 0.004121734295040369 2023-01-23 16:43:09.407287: step: 1520/1541, loss: 0.022263178601861 2023-01-23 16:43:10.317679: step: 1522/1541, loss: 0.01819310337305069 2023-01-23 16:43:11.235732: step: 1524/1541, loss: 0.01443827897310257 2023-01-23 16:43:12.142328: step: 1526/1541, loss: 0.000638660101685673 2023-01-23 16:43:13.035287: step: 1528/1541, loss: 0.005800296552479267 2023-01-23 16:43:13.933275: step: 1530/1541, loss: 0.0006626851973123848 2023-01-23 16:43:14.818528: step: 1532/1541, loss: 0.008952739648520947 2023-01-23 16:43:15.718213: step: 1534/1541, loss: 0.0003978945023845881 2023-01-23 16:43:16.612299: step: 1536/1541, loss: 0.001412464422173798 2023-01-23 16:43:17.473532: step: 1538/1541, loss: 0.18542605638504028 2023-01-23 16:43:18.351751: step: 1540/1541, loss: 0.18857397139072418 2023-01-23 16:43:19.304897: step: 1542/1541, loss: 0.00966188870370388 2023-01-23 16:43:20.210382: step: 1544/1541, loss: 7.279185228981078e-05 2023-01-23 16:43:21.145157: step: 1546/1541, loss: 0.026724431663751602 2023-01-23 16:43:21.984671: step: 1548/1541, loss: 0.0003319729003123939 2023-01-23 16:43:22.847894: step: 1550/1541, loss: 0.001722282962873578 2023-01-23 16:43:23.747907: step: 1552/1541, loss: 0.045623306185007095 2023-01-23 16:43:24.608003: step: 1554/1541, loss: 0.0018429539632052183 2023-01-23 16:43:25.532436: step: 1556/1541, loss: 0.0036285589449107647 2023-01-23 16:43:26.447562: step: 1558/1541, loss: 0.17564274370670319 2023-01-23 16:43:27.343878: step: 1560/1541, loss: 0.0010716163087636232 2023-01-23 16:43:28.245344: step: 1562/1541, loss: 0.021038370206952095 2023-01-23 16:43:29.155605: step: 1564/1541, loss: 0.14696989953517914 2023-01-23 16:43:30.027292: step: 1566/1541, loss: 0.0931316614151001 2023-01-23 16:43:30.960177: step: 1568/1541, loss: 0.00994821172207594 2023-01-23 16:43:31.851096: step: 1570/1541, loss: 0.20261792838573456 2023-01-23 16:43:32.764640: step: 1572/1541, loss: 0.009431536309421062 2023-01-23 16:43:33.605717: step: 1574/1541, loss: 0.3208799958229065 2023-01-23 16:43:34.505345: step: 1576/1541, loss: 0.08828079700469971 2023-01-23 16:43:35.390446: step: 1578/1541, loss: 0.009753754362463951 2023-01-23 16:43:36.277126: step: 1580/1541, loss: 0.25477713346481323 2023-01-23 16:43:37.175917: step: 1582/1541, loss: 0.005710138473659754 2023-01-23 16:43:38.101081: step: 1584/1541, loss: 0.06691901385784149 2023-01-23 16:43:39.001766: step: 1586/1541, loss: 0.05695275962352753 2023-01-23 16:43:39.897547: step: 1588/1541, loss: 0.03574405238032341 2023-01-23 16:43:40.803344: step: 1590/1541, loss: 0.03142710402607918 2023-01-23 16:43:41.746648: step: 1592/1541, loss: 0.0056645567528903484 2023-01-23 16:43:42.627006: step: 1594/1541, loss: 0.010349784046411514 2023-01-23 16:43:43.566674: step: 1596/1541, loss: 0.03830147534608841 2023-01-23 16:43:44.426069: step: 1598/1541, loss: 0.00312950205989182 2023-01-23 16:43:45.328060: step: 1600/1541, loss: 0.014693176373839378 2023-01-23 16:43:46.243477: step: 1602/1541, loss: 0.059448618441820145 2023-01-23 16:43:47.115392: step: 1604/1541, loss: 0.005180123262107372 2023-01-23 16:43:47.990201: step: 1606/1541, loss: 0.0013978676870465279 2023-01-23 16:43:48.903843: step: 1608/1541, loss: 0.062428198754787445 2023-01-23 16:43:49.756959: step: 1610/1541, loss: 0.0027243008371442556 2023-01-23 16:43:50.639758: step: 1612/1541, loss: 0.002802297007292509 2023-01-23 16:43:51.537018: step: 1614/1541, loss: 0.06989813596010208 2023-01-23 16:43:52.454735: step: 1616/1541, loss: 0.0008848598808981478 2023-01-23 16:43:53.307614: step: 1618/1541, loss: 0.0007309399079531431 2023-01-23 16:43:54.288298: step: 1620/1541, loss: 0.25745755434036255 2023-01-23 16:43:55.191548: step: 1622/1541, loss: 0.0016853353008627892 2023-01-23 16:43:56.091775: step: 1624/1541, loss: 0.012555615976452827 2023-01-23 16:43:57.018570: step: 1626/1541, loss: 0.06085673347115517 2023-01-23 16:43:57.944325: step: 1628/1541, loss: 0.011926013976335526 2023-01-23 16:43:58.838390: step: 1630/1541, loss: 0.09091179817914963 2023-01-23 16:43:59.713295: step: 1632/1541, loss: 0.0005296878516674042 2023-01-23 16:44:00.613080: step: 1634/1541, loss: 0.24632173776626587 2023-01-23 16:44:01.499234: step: 1636/1541, loss: 0.005700365640223026 2023-01-23 16:44:02.369949: step: 1638/1541, loss: 0.061286602169275284 2023-01-23 16:44:03.254451: step: 1640/1541, loss: 0.0007066688849590719 2023-01-23 16:44:04.116295: step: 1642/1541, loss: 0.0023786737583577633 2023-01-23 16:44:05.010507: step: 1644/1541, loss: 0.016617566347122192 2023-01-23 16:44:05.860371: step: 1646/1541, loss: 0.0020456251222640276 2023-01-23 16:44:06.741306: step: 1648/1541, loss: 0.0067211538553237915 2023-01-23 16:44:07.619568: step: 1650/1541, loss: 0.03405997157096863 2023-01-23 16:44:08.532786: step: 1652/1541, loss: 0.0658678486943245 2023-01-23 16:44:09.429583: step: 1654/1541, loss: 0.025285761803388596 2023-01-23 16:44:10.303756: step: 1656/1541, loss: 0.16899211704730988 2023-01-23 16:44:11.200401: step: 1658/1541, loss: 0.13937081396579742 2023-01-23 16:44:12.071290: step: 1660/1541, loss: 0.3826233744621277 2023-01-23 16:44:12.960153: step: 1662/1541, loss: 0.16243796050548553 2023-01-23 16:44:13.878408: step: 1664/1541, loss: 0.0004619348328560591 2023-01-23 16:44:14.757786: step: 1666/1541, loss: 0.06583324074745178 2023-01-23 16:44:15.632355: step: 1668/1541, loss: 0.17136019468307495 2023-01-23 16:44:16.513160: step: 1670/1541, loss: 0.07744644582271576 2023-01-23 16:44:17.423252: step: 1672/1541, loss: 0.28340327739715576 2023-01-23 16:44:18.297676: step: 1674/1541, loss: 0.00024831003975123167 2023-01-23 16:44:19.146363: step: 1676/1541, loss: 0.004520962946116924 2023-01-23 16:44:20.034276: step: 1678/1541, loss: 0.26786401867866516 2023-01-23 16:44:20.975734: step: 1680/1541, loss: 0.030685726553201675 2023-01-23 16:44:21.870390: step: 1682/1541, loss: 0.008136328309774399 2023-01-23 16:44:22.753915: step: 1684/1541, loss: 0.3538770079612732 2023-01-23 16:44:23.610444: step: 1686/1541, loss: 0.0008455882780253887 2023-01-23 16:44:24.516918: step: 1688/1541, loss: 0.06492985039949417 2023-01-23 16:44:25.415885: step: 1690/1541, loss: 0.03486677631735802 2023-01-23 16:44:26.307645: step: 1692/1541, loss: 0.010836480185389519 2023-01-23 16:44:27.219528: step: 1694/1541, loss: 0.004039023537188768 2023-01-23 16:44:28.121245: step: 1696/1541, loss: 0.0049410369247198105 2023-01-23 16:44:29.081882: step: 1698/1541, loss: 0.1895049810409546 2023-01-23 16:44:29.971795: step: 1700/1541, loss: 0.0010331337107345462 2023-01-23 16:44:30.867427: step: 1702/1541, loss: 0.0030975055415183306 2023-01-23 16:44:31.781890: step: 1704/1541, loss: 0.003293257672339678 2023-01-23 16:44:32.713467: step: 1706/1541, loss: 0.006046873051673174 2023-01-23 16:44:33.623075: step: 1708/1541, loss: 0.0056313565000891685 2023-01-23 16:44:34.510582: step: 1710/1541, loss: 0.07767383754253387 2023-01-23 16:44:35.385525: step: 1712/1541, loss: 0.005759466905146837 2023-01-23 16:44:36.289865: step: 1714/1541, loss: 0.40388768911361694 2023-01-23 16:44:37.197253: step: 1716/1541, loss: 0.0018105300841853023 2023-01-23 16:44:38.148785: step: 1718/1541, loss: 0.09687695652246475 2023-01-23 16:44:39.036209: step: 1720/1541, loss: 0.03733392432332039 2023-01-23 16:44:39.881394: step: 1722/1541, loss: 0.0005961950519122183 2023-01-23 16:44:40.769276: step: 1724/1541, loss: 0.7075071930885315 2023-01-23 16:44:41.646215: step: 1726/1541, loss: 0.023274701088666916 2023-01-23 16:44:42.520702: step: 1728/1541, loss: 0.14335288107395172 2023-01-23 16:44:43.406389: step: 1730/1541, loss: 0.02199385315179825 2023-01-23 16:44:44.292299: step: 1732/1541, loss: 0.025501737371087074 2023-01-23 16:44:45.197254: step: 1734/1541, loss: 0.5130980014801025 2023-01-23 16:44:46.048803: step: 1736/1541, loss: 0.0006043718894943595 2023-01-23 16:44:46.955892: step: 1738/1541, loss: 0.0031322124414145947 2023-01-23 16:44:47.831074: step: 1740/1541, loss: 0.18105074763298035 2023-01-23 16:44:48.685024: step: 1742/1541, loss: 0.0003278965887147933 2023-01-23 16:44:49.573009: step: 1744/1541, loss: 0.031067294999957085 2023-01-23 16:44:50.458814: step: 1746/1541, loss: 0.002330783288925886 2023-01-23 16:44:51.356919: step: 1748/1541, loss: 0.24710826575756073 2023-01-23 16:44:52.234389: step: 1750/1541, loss: 0.0003369693295098841 2023-01-23 16:44:53.100524: step: 1752/1541, loss: 0.0018313599284738302 2023-01-23 16:44:53.979726: step: 1754/1541, loss: 0.18129794299602509 2023-01-23 16:44:54.840619: step: 1756/1541, loss: 0.0001435188460163772 2023-01-23 16:44:55.706719: step: 1758/1541, loss: 0.49773451685905457 2023-01-23 16:44:56.674606: step: 1760/1541, loss: 0.006467688828706741 2023-01-23 16:44:57.616631: step: 1762/1541, loss: 0.009940037503838539 2023-01-23 16:44:58.496579: step: 1764/1541, loss: 0.00012794007488992065 2023-01-23 16:44:59.392485: step: 1766/1541, loss: 0.002860531909391284 2023-01-23 16:45:00.305040: step: 1768/1541, loss: 0.010631415992975235 2023-01-23 16:45:01.211994: step: 1770/1541, loss: 0.122905433177948 2023-01-23 16:45:02.154446: step: 1772/1541, loss: 0.604182779788971 2023-01-23 16:45:03.085046: step: 1774/1541, loss: 0.0009413146181032062 2023-01-23 16:45:03.965668: step: 1776/1541, loss: 0.0019625069107860327 2023-01-23 16:45:04.844754: step: 1778/1541, loss: 0.11041180044412613 2023-01-23 16:45:05.787004: step: 1780/1541, loss: 0.10163497179746628 2023-01-23 16:45:06.647389: step: 1782/1541, loss: 4.828946839552373e-05 2023-01-23 16:45:07.621960: step: 1784/1541, loss: 0.011762743815779686 2023-01-23 16:45:08.514872: step: 1786/1541, loss: 0.08964305371046066 2023-01-23 16:45:09.415230: step: 1788/1541, loss: 0.6917337775230408 2023-01-23 16:45:10.316807: step: 1790/1541, loss: 0.02512475475668907 2023-01-23 16:45:11.208543: step: 1792/1541, loss: 0.00030578786390833557 2023-01-23 16:45:12.140337: step: 1794/1541, loss: 0.16595958173274994 2023-01-23 16:45:13.019279: step: 1796/1541, loss: 0.09332289546728134 2023-01-23 16:45:13.904615: step: 1798/1541, loss: 1.0940204858779907 2023-01-23 16:45:14.787079: step: 1800/1541, loss: 0.2860771119594574 2023-01-23 16:45:15.720699: step: 1802/1541, loss: 0.07811708748340607 2023-01-23 16:45:16.680641: step: 1804/1541, loss: 0.447722464799881 2023-01-23 16:45:17.564164: step: 1806/1541, loss: 0.00014467100845649838 2023-01-23 16:45:18.502299: step: 1808/1541, loss: 0.8233395218849182 2023-01-23 16:45:19.448275: step: 1810/1541, loss: 0.027445370331406593 2023-01-23 16:45:20.382283: step: 1812/1541, loss: 0.0010785262566059828 2023-01-23 16:45:21.232165: step: 1814/1541, loss: 0.0014827025588601828 2023-01-23 16:45:22.115919: step: 1816/1541, loss: 0.21127375960350037 2023-01-23 16:45:23.043368: step: 1818/1541, loss: 0.12309280782938004 2023-01-23 16:45:23.947921: step: 1820/1541, loss: 0.2248195856809616 2023-01-23 16:45:24.936043: step: 1822/1541, loss: 0.002154644811525941 2023-01-23 16:45:25.911455: step: 1824/1541, loss: 0.0023668722715228796 2023-01-23 16:45:26.807610: step: 1826/1541, loss: 0.0007714400999248028 2023-01-23 16:45:27.713490: step: 1828/1541, loss: 0.00023931710165925324 2023-01-23 16:45:28.599362: step: 1830/1541, loss: 0.07794374972581863 2023-01-23 16:45:29.541685: step: 1832/1541, loss: 0.008727790787816048 2023-01-23 16:45:30.442122: step: 1834/1541, loss: 0.0192238949239254 2023-01-23 16:45:31.321323: step: 1836/1541, loss: 0.09079942852258682 2023-01-23 16:45:32.273133: step: 1838/1541, loss: 0.2305472195148468 2023-01-23 16:45:33.233761: step: 1840/1541, loss: 0.0032118186354637146 2023-01-23 16:45:34.129632: step: 1842/1541, loss: 0.008163834922015667 2023-01-23 16:45:35.001388: step: 1844/1541, loss: 0.3870079219341278 2023-01-23 16:45:35.924307: step: 1846/1541, loss: 0.000850302167236805 2023-01-23 16:45:36.818042: step: 1848/1541, loss: 0.00019287050236016512 2023-01-23 16:45:37.697220: step: 1850/1541, loss: 1.0880768299102783 2023-01-23 16:45:38.640228: step: 1852/1541, loss: 0.6416600942611694 2023-01-23 16:45:39.514863: step: 1854/1541, loss: 0.000658457400277257 2023-01-23 16:45:40.553693: step: 1856/1541, loss: 0.016008181497454643 2023-01-23 16:45:41.430379: step: 1858/1541, loss: 0.0013019908219575882 2023-01-23 16:45:42.347568: step: 1860/1541, loss: 0.0005821393569931388 2023-01-23 16:45:43.262194: step: 1862/1541, loss: 0.017026066780090332 2023-01-23 16:45:44.161231: step: 1864/1541, loss: 0.0030388045124709606 2023-01-23 16:45:45.034493: step: 1866/1541, loss: 0.4973873198032379 2023-01-23 16:45:45.907737: step: 1868/1541, loss: 0.005207166075706482 2023-01-23 16:45:46.807302: step: 1870/1541, loss: 0.0011664388002827764 2023-01-23 16:45:47.674830: step: 1872/1541, loss: 0.001534951850771904 2023-01-23 16:45:48.569886: step: 1874/1541, loss: 0.266693651676178 2023-01-23 16:45:49.468741: step: 1876/1541, loss: 0.003205244429409504 2023-01-23 16:45:50.342071: step: 1878/1541, loss: 0.0011947372695431113 2023-01-23 16:45:51.202377: step: 1880/1541, loss: 0.10857579857110977 2023-01-23 16:45:52.070877: step: 1882/1541, loss: 0.012949317693710327 2023-01-23 16:45:52.970583: step: 1884/1541, loss: 0.001760676153935492 2023-01-23 16:45:53.878045: step: 1886/1541, loss: 0.0014870763989165425 2023-01-23 16:45:54.793633: step: 1888/1541, loss: 0.00406648637726903 2023-01-23 16:45:55.729815: step: 1890/1541, loss: 0.05566425994038582 2023-01-23 16:45:56.621299: step: 1892/1541, loss: 0.0017311254050582647 2023-01-23 16:45:57.503788: step: 1894/1541, loss: 0.0019890274852514267 2023-01-23 16:45:58.366626: step: 1896/1541, loss: 0.05344327911734581 2023-01-23 16:45:59.292525: step: 1898/1541, loss: 0.0008121850551106036 2023-01-23 16:46:00.183246: step: 1900/1541, loss: 2.301452398300171 2023-01-23 16:46:01.112446: step: 1902/1541, loss: 0.01086399331688881 2023-01-23 16:46:02.071419: step: 1904/1541, loss: 0.12598860263824463 2023-01-23 16:46:02.981835: step: 1906/1541, loss: 0.02923492155969143 2023-01-23 16:46:03.965789: step: 1908/1541, loss: 0.07371309399604797 2023-01-23 16:46:04.903103: step: 1910/1541, loss: 0.002797057619318366 2023-01-23 16:46:05.782180: step: 1912/1541, loss: 0.07589200139045715 2023-01-23 16:46:06.654937: step: 1914/1541, loss: 0.00122243317309767 2023-01-23 16:46:07.548681: step: 1916/1541, loss: 0.005946607328951359 2023-01-23 16:46:08.407191: step: 1918/1541, loss: 0.005449843127280474 2023-01-23 16:46:09.355326: step: 1920/1541, loss: 0.002358262659981847 2023-01-23 16:46:10.237531: step: 1922/1541, loss: 0.04205726459622383 2023-01-23 16:46:11.117148: step: 1924/1541, loss: 0.05153888463973999 2023-01-23 16:46:12.026767: step: 1926/1541, loss: 0.07098264992237091 2023-01-23 16:46:12.959764: step: 1928/1541, loss: 0.009056505747139454 2023-01-23 16:46:13.856095: step: 1930/1541, loss: 0.12281707674264908 2023-01-23 16:46:14.737057: step: 1932/1541, loss: 0.005717750638723373 2023-01-23 16:46:15.622156: step: 1934/1541, loss: 0.01027930248528719 2023-01-23 16:46:16.480622: step: 1936/1541, loss: 0.0011246987851336598 2023-01-23 16:46:17.363525: step: 1938/1541, loss: 0.007576985750347376 2023-01-23 16:46:18.256072: step: 1940/1541, loss: 0.00023231239174492657 2023-01-23 16:46:19.142454: step: 1942/1541, loss: 0.0008100592531263828 2023-01-23 16:46:20.026538: step: 1944/1541, loss: 0.0009441015427000821 2023-01-23 16:46:20.958435: step: 1946/1541, loss: 0.011245694011449814 2023-01-23 16:46:21.860966: step: 1948/1541, loss: 0.36255699396133423 2023-01-23 16:46:22.728080: step: 1950/1541, loss: 0.026174090802669525 2023-01-23 16:46:23.613912: step: 1952/1541, loss: 0.008364318870007992 2023-01-23 16:46:24.519646: step: 1954/1541, loss: 0.1460975557565689 2023-01-23 16:46:25.387949: step: 1956/1541, loss: 0.01881801150739193 2023-01-23 16:46:26.267330: step: 1958/1541, loss: 0.0409172922372818 2023-01-23 16:46:27.142000: step: 1960/1541, loss: 0.17393752932548523 2023-01-23 16:46:28.054040: step: 1962/1541, loss: 0.00027801908436231315 2023-01-23 16:46:28.942556: step: 1964/1541, loss: 0.0012848691549152136 2023-01-23 16:46:29.866720: step: 1966/1541, loss: 0.002434040419757366 2023-01-23 16:46:30.902768: step: 1968/1541, loss: 0.061151061207056046 2023-01-23 16:46:31.787228: step: 1970/1541, loss: 0.3152044415473938 2023-01-23 16:46:32.643429: step: 1972/1541, loss: 0.042091552168130875 2023-01-23 16:46:33.557671: step: 1974/1541, loss: 0.0013148410944268107 2023-01-23 16:46:34.438910: step: 1976/1541, loss: 0.002621293067932129 2023-01-23 16:46:35.355568: step: 1978/1541, loss: 0.5173185467720032 2023-01-23 16:46:36.253588: step: 1980/1541, loss: 0.9885285496711731 2023-01-23 16:46:37.170763: step: 1982/1541, loss: 0.2062925547361374 2023-01-23 16:46:38.076658: step: 1984/1541, loss: 0.03304767981171608 2023-01-23 16:46:38.960833: step: 1986/1541, loss: 0.017215406522154808 2023-01-23 16:46:39.828695: step: 1988/1541, loss: 0.22614672780036926 2023-01-23 16:46:40.727874: step: 1990/1541, loss: 0.005770070478320122 2023-01-23 16:46:41.665417: step: 1992/1541, loss: 0.054716356098651886 2023-01-23 16:46:42.585913: step: 1994/1541, loss: 0.624431312084198 2023-01-23 16:46:43.487461: step: 1996/1541, loss: 0.0005821457016281784 2023-01-23 16:46:44.335763: step: 1998/1541, loss: 0.5164980888366699 2023-01-23 16:46:45.235244: step: 2000/1541, loss: 0.022241152822971344 2023-01-23 16:46:46.116472: step: 2002/1541, loss: 0.381669819355011 2023-01-23 16:46:46.973114: step: 2004/1541, loss: 0.0036522529553622007 2023-01-23 16:46:47.830195: step: 2006/1541, loss: 0.12388725578784943 2023-01-23 16:46:48.692417: step: 2008/1541, loss: 0.09526040405035019 2023-01-23 16:46:49.539988: step: 2010/1541, loss: 0.019368160516023636 2023-01-23 16:46:50.456141: step: 2012/1541, loss: 0.012738766148686409 2023-01-23 16:46:51.310744: step: 2014/1541, loss: 0.1571751981973648 2023-01-23 16:46:52.187815: step: 2016/1541, loss: 0.05207612365484238 2023-01-23 16:46:53.090889: step: 2018/1541, loss: 0.024165067821741104 2023-01-23 16:46:53.957504: step: 2020/1541, loss: 0.013858672231435776 2023-01-23 16:46:54.862684: step: 2022/1541, loss: 0.23724883794784546 2023-01-23 16:46:55.715907: step: 2024/1541, loss: 0.009977307170629501 2023-01-23 16:46:56.640866: step: 2026/1541, loss: 0.03162240609526634 2023-01-23 16:46:57.535167: step: 2028/1541, loss: 0.005611121654510498 2023-01-23 16:46:58.420301: step: 2030/1541, loss: 0.001315496163442731 2023-01-23 16:46:59.284725: step: 2032/1541, loss: 0.03707124665379524 2023-01-23 16:47:00.193519: step: 2034/1541, loss: 0.14265547692775726 2023-01-23 16:47:01.156543: step: 2036/1541, loss: 0.060314007103443146 2023-01-23 16:47:02.017094: step: 2038/1541, loss: 0.017014699056744576 2023-01-23 16:47:02.922952: step: 2040/1541, loss: 0.0006868154741823673 2023-01-23 16:47:03.809094: step: 2042/1541, loss: 0.00971588958054781 2023-01-23 16:47:04.730979: step: 2044/1541, loss: 0.0026098547969013453 2023-01-23 16:47:05.643800: step: 2046/1541, loss: 0.1625206023454666 2023-01-23 16:47:06.542193: step: 2048/1541, loss: 0.0066807870753109455 2023-01-23 16:47:07.448354: step: 2050/1541, loss: 0.12588658928871155 2023-01-23 16:47:08.346680: step: 2052/1541, loss: 0.015681026503443718 2023-01-23 16:47:09.214595: step: 2054/1541, loss: 0.010803449898958206 2023-01-23 16:47:10.117954: step: 2056/1541, loss: 0.04836598411202431 2023-01-23 16:47:10.994825: step: 2058/1541, loss: 0.026425201445817947 2023-01-23 16:47:11.958955: step: 2060/1541, loss: 0.005912075284868479 2023-01-23 16:47:12.874739: step: 2062/1541, loss: 0.019236573949456215 2023-01-23 16:47:13.740911: step: 2064/1541, loss: 0.005244735162705183 2023-01-23 16:47:14.709926: step: 2066/1541, loss: 0.1251428723335266 2023-01-23 16:47:15.580354: step: 2068/1541, loss: 0.0006380865233950317 2023-01-23 16:47:16.454670: step: 2070/1541, loss: 0.05475080758333206 2023-01-23 16:47:17.309808: step: 2072/1541, loss: 0.047526780515909195 2023-01-23 16:47:18.175395: step: 2074/1541, loss: 0.0007411256665363908 2023-01-23 16:47:19.092366: step: 2076/1541, loss: 0.12002792209386826 2023-01-23 16:47:19.973735: step: 2078/1541, loss: 0.07943146675825119 2023-01-23 16:47:20.891096: step: 2080/1541, loss: 0.0665576234459877 2023-01-23 16:47:21.805886: step: 2082/1541, loss: 0.00941532850265503 2023-01-23 16:47:22.670259: step: 2084/1541, loss: 0.0012243780074641109 2023-01-23 16:47:23.565176: step: 2086/1541, loss: 0.00491436617448926 2023-01-23 16:47:24.461801: step: 2088/1541, loss: 0.03052910417318344 2023-01-23 16:47:25.354464: step: 2090/1541, loss: 0.004389163106679916 2023-01-23 16:47:26.279852: step: 2092/1541, loss: 0.16208796203136444 2023-01-23 16:47:27.174156: step: 2094/1541, loss: 0.07619201391935349 2023-01-23 16:47:28.060033: step: 2096/1541, loss: 0.01650863140821457 2023-01-23 16:47:28.947269: step: 2098/1541, loss: 0.024579638615250587 2023-01-23 16:47:29.819158: step: 2100/1541, loss: 0.05203148350119591 2023-01-23 16:47:30.712264: step: 2102/1541, loss: 0.0009180564666166902 2023-01-23 16:47:31.595170: step: 2104/1541, loss: 0.005772263742983341 2023-01-23 16:47:32.530824: step: 2106/1541, loss: 0.0024227893445640802 2023-01-23 16:47:33.458283: step: 2108/1541, loss: 0.15827816724777222 2023-01-23 16:47:34.392165: step: 2110/1541, loss: 0.07275424897670746 2023-01-23 16:47:35.289455: step: 2112/1541, loss: 0.01034525502473116 2023-01-23 16:47:36.166288: step: 2114/1541, loss: 0.17165452241897583 2023-01-23 16:47:37.094208: step: 2116/1541, loss: 0.1961456686258316 2023-01-23 16:47:38.038499: step: 2118/1541, loss: 0.27240893244743347 2023-01-23 16:47:38.926062: step: 2120/1541, loss: 0.06381970643997192 2023-01-23 16:47:39.799670: step: 2122/1541, loss: 0.0005619334988296032 2023-01-23 16:47:40.683301: step: 2124/1541, loss: 0.0010077550541609526 2023-01-23 16:47:41.540352: step: 2126/1541, loss: 0.055597156286239624 2023-01-23 16:47:42.488958: step: 2128/1541, loss: 0.0018742969259619713 2023-01-23 16:47:43.404933: step: 2130/1541, loss: 0.04438488557934761 2023-01-23 16:47:44.447889: step: 2132/1541, loss: 0.0011382519733160734 2023-01-23 16:47:45.308603: step: 2134/1541, loss: 0.1791577786207199 2023-01-23 16:47:46.194119: step: 2136/1541, loss: 0.00080308155156672 2023-01-23 16:47:47.046977: step: 2138/1541, loss: 0.0001699081767583266 2023-01-23 16:47:47.926926: step: 2140/1541, loss: 0.03467906266450882 2023-01-23 16:47:48.844399: step: 2142/1541, loss: 0.13939239084720612 2023-01-23 16:47:49.756219: step: 2144/1541, loss: 0.099826380610466 2023-01-23 16:47:50.632906: step: 2146/1541, loss: 0.0004891512217000127 2023-01-23 16:47:51.537682: step: 2148/1541, loss: 0.0014209102373570204 2023-01-23 16:47:52.471891: step: 2150/1541, loss: 0.08048514276742935 2023-01-23 16:47:53.412448: step: 2152/1541, loss: 0.0006654658354818821 2023-01-23 16:47:54.342276: step: 2154/1541, loss: 0.0028134221211075783 2023-01-23 16:47:55.250998: step: 2156/1541, loss: 0.0009190255659632385 2023-01-23 16:47:56.114900: step: 2158/1541, loss: 0.04106007143855095 2023-01-23 16:47:56.987544: step: 2160/1541, loss: 0.08809606730937958 2023-01-23 16:47:57.855684: step: 2162/1541, loss: 0.020687401294708252 2023-01-23 16:47:58.750784: step: 2164/1541, loss: 0.009225855581462383 2023-01-23 16:47:59.676885: step: 2166/1541, loss: 0.14256413280963898 2023-01-23 16:48:00.642613: step: 2168/1541, loss: 0.003633114043623209 2023-01-23 16:48:01.525798: step: 2170/1541, loss: 0.620120644569397 2023-01-23 16:48:02.397604: step: 2172/1541, loss: 0.05874732881784439 2023-01-23 16:48:03.335253: step: 2174/1541, loss: 0.03113182820379734 2023-01-23 16:48:04.240318: step: 2176/1541, loss: 0.13720367848873138 2023-01-23 16:48:05.133633: step: 2178/1541, loss: 0.007666441611945629 2023-01-23 16:48:06.031074: step: 2180/1541, loss: 0.0005909278988838196 2023-01-23 16:48:06.929755: step: 2182/1541, loss: 0.006925535388290882 2023-01-23 16:48:07.815529: step: 2184/1541, loss: 0.6889969110488892 2023-01-23 16:48:08.751953: step: 2186/1541, loss: 0.2033953070640564 2023-01-23 16:48:09.696617: step: 2188/1541, loss: 0.0010533876484259963 2023-01-23 16:48:10.580211: step: 2190/1541, loss: 0.0016320489812642336 2023-01-23 16:48:11.492531: step: 2192/1541, loss: 0.5675832629203796 2023-01-23 16:48:12.411260: step: 2194/1541, loss: 0.01972394622862339 2023-01-23 16:48:13.274141: step: 2196/1541, loss: 0.0006659039645455778 2023-01-23 16:48:14.175278: step: 2198/1541, loss: 0.00044330471428111196 2023-01-23 16:48:15.151559: step: 2200/1541, loss: 0.003997900057584047 2023-01-23 16:48:16.047534: step: 2202/1541, loss: 0.031772688031196594 2023-01-23 16:48:16.964276: step: 2204/1541, loss: 0.004651294555515051 2023-01-23 16:48:17.844835: step: 2206/1541, loss: 0.0003211888833902776 2023-01-23 16:48:18.767162: step: 2208/1541, loss: 0.03285929188132286 2023-01-23 16:48:19.623077: step: 2210/1541, loss: 0.009951846674084663 2023-01-23 16:48:20.581499: step: 2212/1541, loss: 0.01546789612621069 2023-01-23 16:48:21.455872: step: 2214/1541, loss: 0.17580005526542664 2023-01-23 16:48:22.338233: step: 2216/1541, loss: 0.0021145574282854795 2023-01-23 16:48:23.210588: step: 2218/1541, loss: 0.0009619746124371886 2023-01-23 16:48:24.152646: step: 2220/1541, loss: 0.18481925129890442 2023-01-23 16:48:25.110680: step: 2222/1541, loss: 0.1360526829957962 2023-01-23 16:48:26.000501: step: 2224/1541, loss: 0.011038538068532944 2023-01-23 16:48:26.895190: step: 2226/1541, loss: 0.09305992722511292 2023-01-23 16:48:27.794394: step: 2228/1541, loss: 0.06729456037282944 2023-01-23 16:48:28.690634: step: 2230/1541, loss: 0.016737399622797966 2023-01-23 16:48:29.575543: step: 2232/1541, loss: 0.0016165596898645163 2023-01-23 16:48:30.467477: step: 2234/1541, loss: 2.043391704559326 2023-01-23 16:48:31.457097: step: 2236/1541, loss: 0.0057191322557628155 2023-01-23 16:48:32.323975: step: 2238/1541, loss: 0.0191339198499918 2023-01-23 16:48:33.205394: step: 2240/1541, loss: 0.0006957723526284099 2023-01-23 16:48:34.132352: step: 2242/1541, loss: 0.00604609027504921 2023-01-23 16:48:35.082485: step: 2244/1541, loss: 0.24421124160289764 2023-01-23 16:48:36.001407: step: 2246/1541, loss: 0.017053432762622833 2023-01-23 16:48:36.954183: step: 2248/1541, loss: 0.058433689177036285 2023-01-23 16:48:37.859475: step: 2250/1541, loss: 0.16011473536491394 2023-01-23 16:48:38.758210: step: 2252/1541, loss: 0.0013998812064528465 2023-01-23 16:48:39.667042: step: 2254/1541, loss: 0.0014914191560819745 2023-01-23 16:48:40.547706: step: 2256/1541, loss: 0.0074767195619642735 2023-01-23 16:48:41.441862: step: 2258/1541, loss: 0.001615299261175096 2023-01-23 16:48:42.351997: step: 2260/1541, loss: 0.00075186399044469 2023-01-23 16:48:43.217540: step: 2262/1541, loss: 0.09284749627113342 2023-01-23 16:48:44.107502: step: 2264/1541, loss: 0.00043433549581095576 2023-01-23 16:48:44.940222: step: 2266/1541, loss: 0.00044580831308849156 2023-01-23 16:48:45.850933: step: 2268/1541, loss: 0.001416560960933566 2023-01-23 16:48:46.749722: step: 2270/1541, loss: 0.005636881105601788 2023-01-23 16:48:47.621022: step: 2272/1541, loss: 0.04265018180012703 2023-01-23 16:48:48.486004: step: 2274/1541, loss: 0.0002935429511126131 2023-01-23 16:48:49.459402: step: 2276/1541, loss: 0.09014146775007248 2023-01-23 16:48:50.342501: step: 2278/1541, loss: 0.012161402963101864 2023-01-23 16:48:51.292818: step: 2280/1541, loss: 0.02454761043190956 2023-01-23 16:48:52.170183: step: 2282/1541, loss: 0.00014059460954740644 2023-01-23 16:48:53.080192: step: 2284/1541, loss: 0.031467050313949585 2023-01-23 16:48:54.074665: step: 2286/1541, loss: 0.0012076770653948188 2023-01-23 16:48:54.963404: step: 2288/1541, loss: 0.04730156809091568 2023-01-23 16:48:55.879493: step: 2290/1541, loss: 0.0001075590043910779 2023-01-23 16:48:56.748016: step: 2292/1541, loss: 0.0005732142017222941 2023-01-23 16:48:57.675156: step: 2294/1541, loss: 0.3178658187389374 2023-01-23 16:48:58.569208: step: 2296/1541, loss: 0.0034583776723593473 2023-01-23 16:48:59.468791: step: 2298/1541, loss: 0.07215695083141327 2023-01-23 16:49:00.343438: step: 2300/1541, loss: 0.03534983843564987 2023-01-23 16:49:01.298270: step: 2302/1541, loss: 0.034861184656620026 2023-01-23 16:49:02.205361: step: 2304/1541, loss: 0.002642081817612052 2023-01-23 16:49:03.136689: step: 2306/1541, loss: 0.6207960247993469 2023-01-23 16:49:04.045762: step: 2308/1541, loss: 0.0007994709303602576 2023-01-23 16:49:04.991966: step: 2310/1541, loss: 0.05875393748283386 2023-01-23 16:49:05.931219: step: 2312/1541, loss: 0.06641262769699097 2023-01-23 16:49:06.819374: step: 2314/1541, loss: 0.0199725441634655 2023-01-23 16:49:07.710626: step: 2316/1541, loss: 0.9652332067489624 2023-01-23 16:49:08.613465: step: 2318/1541, loss: 0.0037596074398607016 2023-01-23 16:49:09.500919: step: 2320/1541, loss: 0.21332935988903046 2023-01-23 16:49:10.418828: step: 2322/1541, loss: 0.002209665719419718 2023-01-23 16:49:11.354995: step: 2324/1541, loss: 0.015598224475979805 2023-01-23 16:49:12.254124: step: 2326/1541, loss: 0.02550703100860119 2023-01-23 16:49:13.132661: step: 2328/1541, loss: 0.01704166829586029 2023-01-23 16:49:14.093504: step: 2330/1541, loss: 0.03778919577598572 2023-01-23 16:49:14.998264: step: 2332/1541, loss: 0.05746840313076973 2023-01-23 16:49:15.866228: step: 2334/1541, loss: 0.42045697569847107 2023-01-23 16:49:16.750457: step: 2336/1541, loss: 0.004857051186263561 2023-01-23 16:49:17.629405: step: 2338/1541, loss: 0.0009545127395540476 2023-01-23 16:49:18.560811: step: 2340/1541, loss: 0.010268553160130978 2023-01-23 16:49:19.448008: step: 2342/1541, loss: 0.0007546536508016288 2023-01-23 16:49:20.316489: step: 2344/1541, loss: 0.0004766449856106192 2023-01-23 16:49:21.195084: step: 2346/1541, loss: 0.007216546218842268 2023-01-23 16:49:22.078572: step: 2348/1541, loss: 0.0030339069198817015 2023-01-23 16:49:22.965166: step: 2350/1541, loss: 0.007419156841933727 2023-01-23 16:49:23.838069: step: 2352/1541, loss: 0.0004930384457111359 2023-01-23 16:49:24.739612: step: 2354/1541, loss: 0.8062710165977478 2023-01-23 16:49:25.580909: step: 2356/1541, loss: 0.0022385837510228157 2023-01-23 16:49:26.470060: step: 2358/1541, loss: 0.08443843573331833 2023-01-23 16:49:27.343151: step: 2360/1541, loss: 0.014919012784957886 2023-01-23 16:49:28.251309: step: 2362/1541, loss: 0.2875508964061737 2023-01-23 16:49:29.193445: step: 2364/1541, loss: 0.09114597737789154 2023-01-23 16:49:30.074754: step: 2366/1541, loss: 0.032309554517269135 2023-01-23 16:49:30.965615: step: 2368/1541, loss: 0.0307909045368433 2023-01-23 16:49:31.885100: step: 2370/1541, loss: 0.0002427716099191457 2023-01-23 16:49:32.787428: step: 2372/1541, loss: 0.0009135011350736022 2023-01-23 16:49:33.683621: step: 2374/1541, loss: 0.0006539162714034319 2023-01-23 16:49:34.585342: step: 2376/1541, loss: 0.001132382545620203 2023-01-23 16:49:35.461799: step: 2378/1541, loss: 0.005575297866016626 2023-01-23 16:49:36.439721: step: 2380/1541, loss: 0.004700513556599617 2023-01-23 16:49:37.393470: step: 2382/1541, loss: 0.07383330166339874 2023-01-23 16:49:38.258181: step: 2384/1541, loss: 0.0005133341182954609 2023-01-23 16:49:39.197193: step: 2386/1541, loss: 0.6444395184516907 2023-01-23 16:49:40.134385: step: 2388/1541, loss: 0.0013329560169950128 2023-01-23 16:49:41.068919: step: 2390/1541, loss: 0.02520572580397129 2023-01-23 16:49:41.964517: step: 2392/1541, loss: 0.0004136345232836902 2023-01-23 16:49:42.843246: step: 2394/1541, loss: 0.048377517610788345 2023-01-23 16:49:43.696561: step: 2396/1541, loss: 0.01607453264296055 2023-01-23 16:49:44.601663: step: 2398/1541, loss: 0.0002286825911141932 2023-01-23 16:49:45.545072: step: 2400/1541, loss: 0.011380921117961407 2023-01-23 16:49:46.455649: step: 2402/1541, loss: 0.00334489275701344 2023-01-23 16:49:47.364316: step: 2404/1541, loss: 0.1678883135318756 2023-01-23 16:49:48.258050: step: 2406/1541, loss: 0.009373451583087444 2023-01-23 16:49:49.179261: step: 2408/1541, loss: 0.003144056536257267 2023-01-23 16:49:50.035055: step: 2410/1541, loss: 0.0007306768675334752 2023-01-23 16:49:50.928212: step: 2412/1541, loss: 0.003824820974841714 2023-01-23 16:49:51.831957: step: 2414/1541, loss: 0.018291590735316277 2023-01-23 16:49:52.697807: step: 2416/1541, loss: 0.025369565933942795 2023-01-23 16:49:53.620102: step: 2418/1541, loss: 0.01425715908408165 2023-01-23 16:49:54.534658: step: 2420/1541, loss: 0.3278462588787079 2023-01-23 16:49:55.404152: step: 2422/1541, loss: 0.044502142816782 2023-01-23 16:49:56.324731: step: 2424/1541, loss: 0.0020137811079621315 2023-01-23 16:49:57.291649: step: 2426/1541, loss: 0.004657102283090353 2023-01-23 16:49:58.220607: step: 2428/1541, loss: 0.026824945583939552 2023-01-23 16:49:59.088145: step: 2430/1541, loss: 0.0010011052945628762 2023-01-23 16:49:59.987752: step: 2432/1541, loss: 0.0005334201268851757 2023-01-23 16:50:00.967818: step: 2434/1541, loss: 0.00015438080299645662 2023-01-23 16:50:01.882927: step: 2436/1541, loss: 0.00037956805317662656 2023-01-23 16:50:02.890090: step: 2438/1541, loss: 0.03761952742934227 2023-01-23 16:50:03.780193: step: 2440/1541, loss: 0.43763354420661926 2023-01-23 16:50:04.744026: step: 2442/1541, loss: 0.04937673360109329 2023-01-23 16:50:05.657241: step: 2444/1541, loss: 0.02647962048649788 2023-01-23 16:50:06.571247: step: 2446/1541, loss: 0.011832505464553833 2023-01-23 16:50:07.433888: step: 2448/1541, loss: 0.0006317958468571305 2023-01-23 16:50:08.299119: step: 2450/1541, loss: 0.0006562347989529371 2023-01-23 16:50:09.259569: step: 2452/1541, loss: 0.002423187019303441 2023-01-23 16:50:10.156065: step: 2454/1541, loss: 0.6136761903762817 2023-01-23 16:50:11.038996: step: 2456/1541, loss: 0.0015426984755322337 2023-01-23 16:50:11.996515: step: 2458/1541, loss: 0.03751048445701599 2023-01-23 16:50:12.889631: step: 2460/1541, loss: 0.004515915643423796 2023-01-23 16:50:13.801178: step: 2462/1541, loss: 0.0009066573693417013 2023-01-23 16:50:14.723661: step: 2464/1541, loss: 0.02791949361562729 2023-01-23 16:50:15.592933: step: 2466/1541, loss: 0.12360482662916183 2023-01-23 16:50:16.492810: step: 2468/1541, loss: 0.17430435121059418 2023-01-23 16:50:17.377718: step: 2470/1541, loss: 0.012437419034540653 2023-01-23 16:50:18.330346: step: 2472/1541, loss: 0.018971111625432968 2023-01-23 16:50:19.235907: step: 2474/1541, loss: 0.03132326528429985 2023-01-23 16:50:20.131549: step: 2476/1541, loss: 0.00038198981201276183 2023-01-23 16:50:21.041359: step: 2478/1541, loss: 0.0045764511451125145 2023-01-23 16:50:21.911543: step: 2480/1541, loss: 0.08078157901763916 2023-01-23 16:50:22.822300: step: 2482/1541, loss: 0.03582040220499039 2023-01-23 16:50:23.706674: step: 2484/1541, loss: 0.03310023993253708 2023-01-23 16:50:24.641204: step: 2486/1541, loss: 0.001630558050237596 2023-01-23 16:50:25.560574: step: 2488/1541, loss: 0.000679841497913003 2023-01-23 16:50:26.424718: step: 2490/1541, loss: 0.0020893276669085026 2023-01-23 16:50:27.328633: step: 2492/1541, loss: 0.9552612900733948 2023-01-23 16:50:28.238513: step: 2494/1541, loss: 0.09417888522148132 2023-01-23 16:50:29.172561: step: 2496/1541, loss: 0.3174593448638916 2023-01-23 16:50:30.097619: step: 2498/1541, loss: 0.1228269413113594 2023-01-23 16:50:30.975205: step: 2500/1541, loss: 0.5275244116783142 2023-01-23 16:50:31.892463: step: 2502/1541, loss: 0.49892017245292664 2023-01-23 16:50:32.782492: step: 2504/1541, loss: 0.02287181280553341 2023-01-23 16:50:33.700046: step: 2506/1541, loss: 0.7797554731369019 2023-01-23 16:50:34.578317: step: 2508/1541, loss: 0.019698752090334892 2023-01-23 16:50:35.433288: step: 2510/1541, loss: 0.2132844626903534 2023-01-23 16:50:36.287973: step: 2512/1541, loss: 0.0023915134370326996 2023-01-23 16:50:37.197702: step: 2514/1541, loss: 0.24911534786224365 2023-01-23 16:50:38.072898: step: 2516/1541, loss: 0.007407403085380793 2023-01-23 16:50:38.973306: step: 2518/1541, loss: 0.08960495889186859 2023-01-23 16:50:39.840018: step: 2520/1541, loss: 0.01030039507895708 2023-01-23 16:50:40.764136: step: 2522/1541, loss: 0.001310264109633863 2023-01-23 16:50:41.653474: step: 2524/1541, loss: 0.0024812822230160236 2023-01-23 16:50:42.562665: step: 2526/1541, loss: 0.10743742436170578 2023-01-23 16:50:43.455342: step: 2528/1541, loss: 0.04246335104107857 2023-01-23 16:50:44.287505: step: 2530/1541, loss: 0.019447358325123787 2023-01-23 16:50:45.163208: step: 2532/1541, loss: 0.03460220620036125 2023-01-23 16:50:46.037555: step: 2534/1541, loss: 0.053952060639858246 2023-01-23 16:50:46.907385: step: 2536/1541, loss: 0.04573484882712364 2023-01-23 16:50:47.806359: step: 2538/1541, loss: 0.05817748233675957 2023-01-23 16:50:48.720829: step: 2540/1541, loss: 0.00671585975214839 2023-01-23 16:50:49.609497: step: 2542/1541, loss: 0.0037475191056728363 2023-01-23 16:50:50.507452: step: 2544/1541, loss: 0.1670503467321396 2023-01-23 16:50:51.397494: step: 2546/1541, loss: 0.005198221653699875 2023-01-23 16:50:52.327857: step: 2548/1541, loss: 0.0011840563965961337 2023-01-23 16:50:53.230707: step: 2550/1541, loss: 0.07342854142189026 2023-01-23 16:50:54.131083: step: 2552/1541, loss: 0.004708302207291126 2023-01-23 16:50:55.016895: step: 2554/1541, loss: 0.013297513127326965 2023-01-23 16:50:55.898976: step: 2556/1541, loss: 0.040902189910411835 2023-01-23 16:50:56.760628: step: 2558/1541, loss: 0.00438391137868166 2023-01-23 16:50:57.641820: step: 2560/1541, loss: 0.0022125374525785446 2023-01-23 16:50:58.539358: step: 2562/1541, loss: 0.0693119615316391 2023-01-23 16:50:59.485822: step: 2564/1541, loss: 0.050638340413570404 2023-01-23 16:51:00.342829: step: 2566/1541, loss: 0.03577306866645813 2023-01-23 16:51:01.235810: step: 2568/1541, loss: 0.005846294574439526 2023-01-23 16:51:02.147023: step: 2570/1541, loss: 0.0006805678713135421 2023-01-23 16:51:03.024555: step: 2572/1541, loss: 0.0016721051651984453 2023-01-23 16:51:03.918830: step: 2574/1541, loss: 0.001571949920617044 2023-01-23 16:51:04.766878: step: 2576/1541, loss: 0.0015172540443018079 2023-01-23 16:51:05.665549: step: 2578/1541, loss: 0.0003200221690349281 2023-01-23 16:51:06.633799: step: 2580/1541, loss: 0.286324143409729 2023-01-23 16:51:07.559009: step: 2582/1541, loss: 0.0012812891509383917 2023-01-23 16:51:08.535693: step: 2584/1541, loss: 0.0021056239493191242 2023-01-23 16:51:09.454851: step: 2586/1541, loss: 0.1291775107383728 2023-01-23 16:51:10.330964: step: 2588/1541, loss: 0.017830602824687958 2023-01-23 16:51:11.193314: step: 2590/1541, loss: 0.12640352547168732 2023-01-23 16:51:12.121492: step: 2592/1541, loss: 0.10100552439689636 2023-01-23 16:51:12.986577: step: 2594/1541, loss: 0.40714582800865173 2023-01-23 16:51:13.883492: step: 2596/1541, loss: 0.0059329853393137455 2023-01-23 16:51:14.738888: step: 2598/1541, loss: 0.16602776944637299 2023-01-23 16:51:15.632179: step: 2600/1541, loss: 0.3742655813694 2023-01-23 16:51:16.580788: step: 2602/1541, loss: 0.0034299001563340425 2023-01-23 16:51:17.488013: step: 2604/1541, loss: 0.05824281647801399 2023-01-23 16:51:18.391645: step: 2606/1541, loss: 0.022307273000478745 2023-01-23 16:51:19.326075: step: 2608/1541, loss: 0.020081503316760063 2023-01-23 16:51:20.218868: step: 2610/1541, loss: 6.034917831420898 2023-01-23 16:51:21.256661: step: 2612/1541, loss: 0.11382950842380524 2023-01-23 16:51:22.152961: step: 2614/1541, loss: 0.00263033714145422 2023-01-23 16:51:23.076061: step: 2616/1541, loss: 0.004129740409553051 2023-01-23 16:51:23.943505: step: 2618/1541, loss: 0.09263093024492264 2023-01-23 16:51:24.833113: step: 2620/1541, loss: 0.004621267318725586 2023-01-23 16:51:25.757229: step: 2622/1541, loss: 0.011523191817104816 2023-01-23 16:51:26.649607: step: 2624/1541, loss: 0.0002921027480624616 2023-01-23 16:51:27.568903: step: 2626/1541, loss: 0.0007958679343573749 2023-01-23 16:51:28.452561: step: 2628/1541, loss: 0.0012954577105119824 2023-01-23 16:51:29.366119: step: 2630/1541, loss: 0.3621709942817688 2023-01-23 16:51:30.269132: step: 2632/1541, loss: 0.04176723584532738 2023-01-23 16:51:31.179334: step: 2634/1541, loss: 0.013803991489112377 2023-01-23 16:51:32.068041: step: 2636/1541, loss: 0.02902374602854252 2023-01-23 16:51:32.993792: step: 2638/1541, loss: 0.02278134599328041 2023-01-23 16:51:33.993281: step: 2640/1541, loss: 0.011460050009191036 2023-01-23 16:51:34.870481: step: 2642/1541, loss: 0.00250784563831985 2023-01-23 16:51:35.762233: step: 2644/1541, loss: 0.11961725354194641 2023-01-23 16:51:36.657755: step: 2646/1541, loss: 0.05067138746380806 2023-01-23 16:51:37.556308: step: 2648/1541, loss: 0.010975804179906845 2023-01-23 16:51:38.421916: step: 2650/1541, loss: 0.00010571416351012886 2023-01-23 16:51:39.331786: step: 2652/1541, loss: 0.0012887096963822842 2023-01-23 16:51:40.223475: step: 2654/1541, loss: 0.0019790292717516422 2023-01-23 16:51:41.106470: step: 2656/1541, loss: 0.046282436698675156 2023-01-23 16:51:42.014018: step: 2658/1541, loss: 0.057582516223192215 2023-01-23 16:51:42.926973: step: 2660/1541, loss: 5.800797939300537 2023-01-23 16:51:43.819965: step: 2662/1541, loss: 0.0007725405157543719 2023-01-23 16:51:44.663012: step: 2664/1541, loss: 0.024327602237462997 2023-01-23 16:51:45.586900: step: 2666/1541, loss: 0.05111831799149513 2023-01-23 16:51:46.473333: step: 2668/1541, loss: 0.0004271298530511558 2023-01-23 16:51:47.396387: step: 2670/1541, loss: 0.19477300345897675 2023-01-23 16:51:48.269723: step: 2672/1541, loss: 0.009747439064085484 2023-01-23 16:51:49.178915: step: 2674/1541, loss: 0.11329422891139984 2023-01-23 16:51:50.013099: step: 2676/1541, loss: 0.04347386583685875 2023-01-23 16:51:50.901412: step: 2678/1541, loss: 0.05122005194425583 2023-01-23 16:51:51.770630: step: 2680/1541, loss: 0.012380359694361687 2023-01-23 16:51:52.693464: step: 2682/1541, loss: 0.010049487464129925 2023-01-23 16:51:53.586678: step: 2684/1541, loss: 0.0020062143448740244 2023-01-23 16:51:54.470467: step: 2686/1541, loss: 0.029423633590340614 2023-01-23 16:51:55.437116: step: 2688/1541, loss: 0.030902327969670296 2023-01-23 16:51:56.385571: step: 2690/1541, loss: 0.06046709418296814 2023-01-23 16:51:57.263886: step: 2692/1541, loss: 0.05532389134168625 2023-01-23 16:51:58.143572: step: 2694/1541, loss: 0.0014799396740272641 2023-01-23 16:51:59.052627: step: 2696/1541, loss: 0.3023439645767212 2023-01-23 16:51:59.933580: step: 2698/1541, loss: 0.03707950562238693 2023-01-23 16:52:00.827863: step: 2700/1541, loss: 0.11252858489751816 2023-01-23 16:52:01.704488: step: 2702/1541, loss: 0.0640965923666954 2023-01-23 16:52:02.625877: step: 2704/1541, loss: 0.0014686535578221083 2023-01-23 16:52:03.579057: step: 2706/1541, loss: 0.033155471086502075 2023-01-23 16:52:04.475976: step: 2708/1541, loss: 0.00043664887198247015 2023-01-23 16:52:05.393952: step: 2710/1541, loss: 0.01073499210178852 2023-01-23 16:52:06.326620: step: 2712/1541, loss: 0.6728774309158325 2023-01-23 16:52:07.207480: step: 2714/1541, loss: 0.01346665807068348 2023-01-23 16:52:08.114828: step: 2716/1541, loss: 0.07808370143175125 2023-01-23 16:52:09.005051: step: 2718/1541, loss: 0.0029322365298867226 2023-01-23 16:52:09.876056: step: 2720/1541, loss: 0.006430764216929674 2023-01-23 16:52:10.797892: step: 2722/1541, loss: 0.015591679140925407 2023-01-23 16:52:11.680432: step: 2724/1541, loss: 0.014717045240104198 2023-01-23 16:52:12.536786: step: 2726/1541, loss: 0.007288551423698664 2023-01-23 16:52:13.450946: step: 2728/1541, loss: 0.08878517150878906 2023-01-23 16:52:14.290582: step: 2730/1541, loss: 0.04554864019155502 2023-01-23 16:52:15.218287: step: 2732/1541, loss: 0.048551421612501144 2023-01-23 16:52:16.091237: step: 2734/1541, loss: 0.05576992779970169 2023-01-23 16:52:17.037757: step: 2736/1541, loss: 0.08568917959928513 2023-01-23 16:52:17.887233: step: 2738/1541, loss: 0.002491794526576996 2023-01-23 16:52:18.739030: step: 2740/1541, loss: 0.14599120616912842 2023-01-23 16:52:19.680318: step: 2742/1541, loss: 0.03901815786957741 2023-01-23 16:52:20.562015: step: 2744/1541, loss: 0.03743787482380867 2023-01-23 16:52:21.493749: step: 2746/1541, loss: 0.20714910328388214 2023-01-23 16:52:22.395852: step: 2748/1541, loss: 0.30633074045181274 2023-01-23 16:52:23.295518: step: 2750/1541, loss: 0.0005951460916548967 2023-01-23 16:52:24.190216: step: 2752/1541, loss: 0.013583301566541195 2023-01-23 16:52:25.055987: step: 2754/1541, loss: 0.13183200359344482 2023-01-23 16:52:25.962096: step: 2756/1541, loss: 0.1355353742837906 2023-01-23 16:52:26.815800: step: 2758/1541, loss: 0.04048112407326698 2023-01-23 16:52:27.737864: step: 2760/1541, loss: 0.0916023999452591 2023-01-23 16:52:28.611694: step: 2762/1541, loss: 0.0005623653996735811 2023-01-23 16:52:29.529995: step: 2764/1541, loss: 0.08596187084913254 2023-01-23 16:52:30.490420: step: 2766/1541, loss: 0.14523912966251373 2023-01-23 16:52:31.406966: step: 2768/1541, loss: 0.001193310832604766 2023-01-23 16:52:32.377494: step: 2770/1541, loss: 0.001085105468519032 2023-01-23 16:52:33.334331: step: 2772/1541, loss: 0.07605360448360443 2023-01-23 16:52:34.218998: step: 2774/1541, loss: 0.03147764503955841 2023-01-23 16:52:35.143469: step: 2776/1541, loss: 0.02437407523393631 2023-01-23 16:52:36.011075: step: 2778/1541, loss: 0.071903757750988 2023-01-23 16:52:36.925128: step: 2780/1541, loss: 0.2603035271167755 2023-01-23 16:52:37.829225: step: 2782/1541, loss: 0.001123296096920967 2023-01-23 16:52:38.724082: step: 2784/1541, loss: 0.13437524437904358 2023-01-23 16:52:39.615453: step: 2786/1541, loss: 0.0035410141572356224 2023-01-23 16:52:40.484907: step: 2788/1541, loss: 0.007034194190055132 2023-01-23 16:52:41.361940: step: 2790/1541, loss: 0.00466840248554945 2023-01-23 16:52:42.279230: step: 2792/1541, loss: 0.06057681888341904 2023-01-23 16:52:43.184987: step: 2794/1541, loss: 0.18114954233169556 2023-01-23 16:52:44.054558: step: 2796/1541, loss: 0.0056350091472268105 2023-01-23 16:52:44.987771: step: 2798/1541, loss: 0.04138258844614029 2023-01-23 16:52:45.888506: step: 2800/1541, loss: 0.005661848466843367 2023-01-23 16:52:46.788493: step: 2802/1541, loss: 0.018805842846632004 2023-01-23 16:52:47.661990: step: 2804/1541, loss: 0.00027815153589472175 2023-01-23 16:52:48.549682: step: 2806/1541, loss: 0.019625956192612648 2023-01-23 16:52:49.412706: step: 2808/1541, loss: 0.030420484021306038 2023-01-23 16:52:50.304150: step: 2810/1541, loss: 0.000246949668508023 2023-01-23 16:52:51.195316: step: 2812/1541, loss: 0.06510331481695175 2023-01-23 16:52:52.116282: step: 2814/1541, loss: 0.0164056196808815 2023-01-23 16:52:53.073352: step: 2816/1541, loss: 0.9308692216873169 2023-01-23 16:52:53.970632: step: 2818/1541, loss: 0.1050080731511116 2023-01-23 16:52:54.849487: step: 2820/1541, loss: 0.05526432767510414 2023-01-23 16:52:55.718235: step: 2822/1541, loss: 0.05991913005709648 2023-01-23 16:52:56.610250: step: 2824/1541, loss: 0.01796143874526024 2023-01-23 16:52:57.560500: step: 2826/1541, loss: 0.0022974349558353424 2023-01-23 16:52:58.449002: step: 2828/1541, loss: 0.42326557636260986 2023-01-23 16:52:59.320790: step: 2830/1541, loss: 0.00023670772498007864 2023-01-23 16:53:00.231353: step: 2832/1541, loss: 0.3110634386539459 2023-01-23 16:53:01.176795: step: 2834/1541, loss: 0.035395510494709015 2023-01-23 16:53:02.084663: step: 2836/1541, loss: 0.5388238430023193 2023-01-23 16:53:02.960974: step: 2838/1541, loss: 0.020305193960666656 2023-01-23 16:53:03.916561: step: 2840/1541, loss: 0.0913495123386383 2023-01-23 16:53:04.839897: step: 2842/1541, loss: 0.016323160380125046 2023-01-23 16:53:05.773304: step: 2844/1541, loss: 0.001707634306512773 2023-01-23 16:53:06.725748: step: 2846/1541, loss: 0.5199735164642334 2023-01-23 16:53:07.596832: step: 2848/1541, loss: 0.000532648409716785 2023-01-23 16:53:08.494715: step: 2850/1541, loss: 0.004126179497689009 2023-01-23 16:53:09.410755: step: 2852/1541, loss: 0.0003235558106098324 2023-01-23 16:53:10.288038: step: 2854/1541, loss: 0.0022007152438163757 2023-01-23 16:53:11.201995: step: 2856/1541, loss: 0.3812693953514099 2023-01-23 16:53:12.256857: step: 2858/1541, loss: 0.013883737847208977 2023-01-23 16:53:13.143311: step: 2860/1541, loss: 0.000698643212672323 2023-01-23 16:53:14.027876: step: 2862/1541, loss: 0.0007702818256802857 2023-01-23 16:53:14.957701: step: 2864/1541, loss: 0.0012608974939212203 2023-01-23 16:53:15.883184: step: 2866/1541, loss: 0.07418864965438843 2023-01-23 16:53:16.807755: step: 2868/1541, loss: 0.0051513295620679855 2023-01-23 16:53:17.768453: step: 2870/1541, loss: 0.017385724931955338 2023-01-23 16:53:18.751657: step: 2872/1541, loss: 2.643925428390503 2023-01-23 16:53:19.639386: step: 2874/1541, loss: 0.05948825180530548 2023-01-23 16:53:20.618139: step: 2876/1541, loss: 0.04403442516922951 2023-01-23 16:53:21.488242: step: 2878/1541, loss: 0.04672633111476898 2023-01-23 16:53:22.403199: step: 2880/1541, loss: 0.18495723605155945 2023-01-23 16:53:23.311597: step: 2882/1541, loss: 0.07998687028884888 2023-01-23 16:53:24.200421: step: 2884/1541, loss: 0.0003549785469658673 2023-01-23 16:53:25.137069: step: 2886/1541, loss: 0.0650876834988594 2023-01-23 16:53:26.061369: step: 2888/1541, loss: 0.0008541434071958065 2023-01-23 16:53:27.049651: step: 2890/1541, loss: 7.660056871827692e-05 2023-01-23 16:53:27.995433: step: 2892/1541, loss: 0.014231561683118343 2023-01-23 16:53:28.839148: step: 2894/1541, loss: 0.000301918393233791 2023-01-23 16:53:29.722541: step: 2896/1541, loss: 0.002258189721032977 2023-01-23 16:53:30.626824: step: 2898/1541, loss: 0.06678856164216995 2023-01-23 16:53:31.502146: step: 2900/1541, loss: 0.0006622412474825978 2023-01-23 16:53:32.432535: step: 2902/1541, loss: 0.08494129776954651 2023-01-23 16:53:33.291451: step: 2904/1541, loss: 0.15452715754508972 2023-01-23 16:53:34.126708: step: 2906/1541, loss: 0.001557346899062395 2023-01-23 16:53:35.028166: step: 2908/1541, loss: 0.07907851040363312 2023-01-23 16:53:35.910206: step: 2910/1541, loss: 0.2016446441411972 2023-01-23 16:53:36.839720: step: 2912/1541, loss: 0.011586892418563366 2023-01-23 16:53:37.802544: step: 2914/1541, loss: 0.05266231298446655 2023-01-23 16:53:38.708927: step: 2916/1541, loss: 0.20105989277362823 2023-01-23 16:53:39.616327: step: 2918/1541, loss: 0.01717824675142765 2023-01-23 16:53:40.513872: step: 2920/1541, loss: 0.23962347209453583 2023-01-23 16:53:41.444857: step: 2922/1541, loss: 0.0003540883772075176 2023-01-23 16:53:42.338031: step: 2924/1541, loss: 0.06862305849790573 2023-01-23 16:53:43.345409: step: 2926/1541, loss: 0.019321225583553314 2023-01-23 16:53:44.258123: step: 2928/1541, loss: 0.5687406659126282 2023-01-23 16:53:45.172509: step: 2930/1541, loss: 0.18920981884002686 2023-01-23 16:53:46.065562: step: 2932/1541, loss: 0.0023106480948626995 2023-01-23 16:53:46.946455: step: 2934/1541, loss: 0.0009300416568294168 2023-01-23 16:53:47.823367: step: 2936/1541, loss: 0.0004144645354244858 2023-01-23 16:53:48.742093: step: 2938/1541, loss: 0.008836641907691956 2023-01-23 16:53:49.638691: step: 2940/1541, loss: 0.03713717311620712 2023-01-23 16:53:50.575875: step: 2942/1541, loss: 0.4688044786453247 2023-01-23 16:53:51.480568: step: 2944/1541, loss: 0.02694069966673851 2023-01-23 16:53:52.335512: step: 2946/1541, loss: 0.00966192502528429 2023-01-23 16:53:53.197348: step: 2948/1541, loss: 0.00800809171050787 2023-01-23 16:53:54.107425: step: 2950/1541, loss: 0.0033735770266503096 2023-01-23 16:53:54.984182: step: 2952/1541, loss: 0.04222804307937622 2023-01-23 16:53:55.877263: step: 2954/1541, loss: 0.01755979284644127 2023-01-23 16:53:56.722893: step: 2956/1541, loss: 0.042967431247234344 2023-01-23 16:53:57.610978: step: 2958/1541, loss: 0.003928324207663536 2023-01-23 16:53:58.509773: step: 2960/1541, loss: 0.009396578185260296 2023-01-23 16:53:59.467194: step: 2962/1541, loss: 0.00040080922190099955 2023-01-23 16:54:00.369338: step: 2964/1541, loss: 0.007643580436706543 2023-01-23 16:54:01.263501: step: 2966/1541, loss: 0.12151513248682022 2023-01-23 16:54:02.183806: step: 2968/1541, loss: 0.022050641477108 2023-01-23 16:54:03.065577: step: 2970/1541, loss: 0.0013738140696659684 2023-01-23 16:54:03.986383: step: 2972/1541, loss: 0.005337964743375778 2023-01-23 16:54:04.898026: step: 2974/1541, loss: 0.014375407248735428 2023-01-23 16:54:05.835664: step: 2976/1541, loss: 0.08403509110212326 2023-01-23 16:54:06.698353: step: 2978/1541, loss: 0.0033331681042909622 2023-01-23 16:54:07.523497: step: 2980/1541, loss: 0.0007032901630736887 2023-01-23 16:54:08.434681: step: 2982/1541, loss: 0.23113320767879486 2023-01-23 16:54:09.322512: step: 2984/1541, loss: 0.0010184051934629679 2023-01-23 16:54:10.238554: step: 2986/1541, loss: 0.06667471677064896 2023-01-23 16:54:11.120098: step: 2988/1541, loss: 0.009068943560123444 2023-01-23 16:54:12.065006: step: 2990/1541, loss: 0.0073004597797989845 2023-01-23 16:54:12.923592: step: 2992/1541, loss: 0.06748609244823456 2023-01-23 16:54:13.852497: step: 2994/1541, loss: 0.0012783288257196546 2023-01-23 16:54:14.743913: step: 2996/1541, loss: 0.5226601958274841 2023-01-23 16:54:15.641190: step: 2998/1541, loss: 0.06705248355865479 2023-01-23 16:54:16.516622: step: 3000/1541, loss: 0.002232732716947794 2023-01-23 16:54:17.395554: step: 3002/1541, loss: 0.0015643889782950282 2023-01-23 16:54:18.271872: step: 3004/1541, loss: 0.01674407534301281 2023-01-23 16:54:19.157482: step: 3006/1541, loss: 0.04647661745548248 2023-01-23 16:54:20.042859: step: 3008/1541, loss: 0.0009806986199691892 2023-01-23 16:54:20.937387: step: 3010/1541, loss: 0.015720410272479057 2023-01-23 16:54:21.862903: step: 3012/1541, loss: 0.010446685366332531 2023-01-23 16:54:22.716341: step: 3014/1541, loss: 0.0425034798681736 2023-01-23 16:54:23.635067: step: 3016/1541, loss: 0.05532944202423096 2023-01-23 16:54:24.517852: step: 3018/1541, loss: 0.00019177379726897925 2023-01-23 16:54:25.420819: step: 3020/1541, loss: 0.632382869720459 2023-01-23 16:54:26.352154: step: 3022/1541, loss: 0.10538040846586227 2023-01-23 16:54:27.240878: step: 3024/1541, loss: 0.004769179504364729 2023-01-23 16:54:28.155894: step: 3026/1541, loss: 0.004263520240783691 2023-01-23 16:54:29.098244: step: 3028/1541, loss: 0.013907281681895256 2023-01-23 16:54:30.017300: step: 3030/1541, loss: 0.007084372453391552 2023-01-23 16:54:31.000429: step: 3032/1541, loss: 0.0019583546090871096 2023-01-23 16:54:31.945881: step: 3034/1541, loss: 0.014658340252935886 2023-01-23 16:54:32.819114: step: 3036/1541, loss: 0.0003067475336138159 2023-01-23 16:54:33.706645: step: 3038/1541, loss: 0.00017581373685970902 2023-01-23 16:54:34.578928: step: 3040/1541, loss: 0.0027317192871123552 2023-01-23 16:54:35.457932: step: 3042/1541, loss: 0.0016488137189298868 2023-01-23 16:54:36.378056: step: 3044/1541, loss: 0.10294975340366364 2023-01-23 16:54:37.288883: step: 3046/1541, loss: 0.0240769162774086 2023-01-23 16:54:38.156936: step: 3048/1541, loss: 0.0008914253558032215 2023-01-23 16:54:39.112444: step: 3050/1541, loss: 0.005229721777141094 2023-01-23 16:54:40.036404: step: 3052/1541, loss: 0.2663055658340454 2023-01-23 16:54:40.883922: step: 3054/1541, loss: 0.010909488424658775 2023-01-23 16:54:41.766381: step: 3056/1541, loss: 0.05238054320216179 2023-01-23 16:54:42.659680: step: 3058/1541, loss: 0.19176466763019562 2023-01-23 16:54:43.600830: step: 3060/1541, loss: 0.0076775941997766495 2023-01-23 16:54:44.470542: step: 3062/1541, loss: 0.06262975186109543 2023-01-23 16:54:45.355046: step: 3064/1541, loss: 0.0012541908072307706 2023-01-23 16:54:46.264837: step: 3066/1541, loss: 0.00016886791854631156 2023-01-23 16:54:47.178323: step: 3068/1541, loss: 0.0651036724448204 2023-01-23 16:54:48.068013: step: 3070/1541, loss: 0.0017994377994909883 2023-01-23 16:54:48.957753: step: 3072/1541, loss: 0.0021636835299432278 2023-01-23 16:54:49.853184: step: 3074/1541, loss: 0.0013457160675898194 2023-01-23 16:54:50.758330: step: 3076/1541, loss: 0.0035176114179193974 2023-01-23 16:54:51.641969: step: 3078/1541, loss: 0.005873566493391991 2023-01-23 16:54:52.574389: step: 3080/1541, loss: 7.998273940756917e-05 2023-01-23 16:54:53.448685: step: 3082/1541, loss: 0.10935770720243454 2023-01-23 16:54:54.099343: step: 3084/1541, loss: 0.7902599573135376 ================================================== Loss: 0.093 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6547812559679388, 'r': 0.6532585088610365, 'f1': 0.6540189960657643}, 'combined': 0.4316195893431493, 'epoch': 24} Test Chinese: {'event': {'p': 0.613564668769716, 'r': 0.6992210904733374, 'f1': 0.6535984318118174}, 'argument': {'p': 0.701673136441013, 'r': 0.711383107010845, 'f1': 0.7064947601633925}, 'combined': 0.4617638673260594, 'epoch': 24} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6431283877150464, 'r': 0.6629382162764698, 'f1': 0.6528830687280083}, 'combined': 0.42984848003210485, 'epoch': 24} Test Korean: {'event': {'p': 0.6124880838894184, 'r': 0.7699221090473337, 'f1': 0.6822405096894081}, 'argument': {'p': 0.6987936688487971, 'r': 0.7267454156027491, 'f1': 0.7124955054928911}, 'combined': 0.48609329681888247, 'epoch': 24} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6509545947330577, 'r': 0.6622117042660355, 'f1': 0.6565348987505234}, 'combined': 0.4544291712981679, 'epoch': 24} Test Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.6959848476747131, 'r': 0.7159628842064595, 'f1': 0.7058325285756641}, 'combined': 0.490139571519302, 'epoch': 24} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6790953008344311, 'r': 0.6911146866899078, 'f1': 0.6850522771575401}, 'combined': 0.49183240411310575, 'epoch': 24} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5879442195231669, 'r': 0.4468376068376068, 'f1': 0.5077700077700078}, 'combined': 0.2958659887793746, 'epoch': 24} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.526577365163572, 'r': 0.612299261818107, 'f1': 0.5662122206059914}, 'combined': 0.2750173642943387, 'epoch': 24} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.679679224642488, 'r': 0.6459587049703025, 'f1': 0.6623900869727267}, 'combined': 0.43714408762431844, 'epoch': 23} Test for Chinese: {'event': {'p': 0.6151419558359621, 'r': 0.701018573996405, 'f1': 0.6552786334360123}, 'argument': {'p': 0.720803339496083, 'r': 0.694252961199546, 'f1': 0.7072790710047019}, 'combined': 0.4634648631058534, 'epoch': 23} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.666347687400319, 'r': 0.6722445695897024, 'f1': 0.6692831397677212}, 'combined': 0.4919517095728549, 'epoch': 23} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 25 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 17:00:19.069276: step: 2/1541, loss: 0.0003783583524636924 2023-01-23 17:00:19.999536: step: 4/1541, loss: 0.0018623454961925745 2023-01-23 17:00:20.908776: step: 6/1541, loss: 0.15357321500778198 2023-01-23 17:00:21.773732: step: 8/1541, loss: 0.2726392149925232 2023-01-23 17:00:22.631871: step: 10/1541, loss: 0.00018087029457092285 2023-01-23 17:00:23.512758: step: 12/1541, loss: 0.04983967915177345 2023-01-23 17:00:24.402610: step: 14/1541, loss: 0.008990027010440826 2023-01-23 17:00:25.296002: step: 16/1541, loss: 0.5907168984413147 2023-01-23 17:00:26.210554: step: 18/1541, loss: 0.17865996062755585 2023-01-23 17:00:27.071677: step: 20/1541, loss: 0.0025185293052345514 2023-01-23 17:00:27.985204: step: 22/1541, loss: 0.006308277603238821 2023-01-23 17:00:28.953588: step: 24/1541, loss: 0.11319524049758911 2023-01-23 17:00:29.864782: step: 26/1541, loss: 0.0006542319897562265 2023-01-23 17:00:30.806346: step: 28/1541, loss: 0.004279728513211012 2023-01-23 17:00:31.639017: step: 30/1541, loss: 0.0015187110984697938 2023-01-23 17:00:32.530793: step: 32/1541, loss: 0.05060319975018501 2023-01-23 17:00:33.414764: step: 34/1541, loss: 0.014705819077789783 2023-01-23 17:00:34.283806: step: 36/1541, loss: 0.019589968025684357 2023-01-23 17:00:35.200392: step: 38/1541, loss: 0.0025453276466578245 2023-01-23 17:00:36.095971: step: 40/1541, loss: 0.03846823424100876 2023-01-23 17:00:37.015930: step: 42/1541, loss: 0.00823909416794777 2023-01-23 17:00:37.916149: step: 44/1541, loss: 0.057926710695028305 2023-01-23 17:00:38.808601: step: 46/1541, loss: 0.05656289681792259 2023-01-23 17:00:39.663952: step: 48/1541, loss: 0.023970676586031914 2023-01-23 17:00:40.558378: step: 50/1541, loss: 0.0007750741206109524 2023-01-23 17:00:41.517524: step: 52/1541, loss: 0.006079799495637417 2023-01-23 17:00:42.391109: step: 54/1541, loss: 0.02034439891576767 2023-01-23 17:00:43.299824: step: 56/1541, loss: 0.013890785165131092 2023-01-23 17:00:44.264024: step: 58/1541, loss: 0.22331206500530243 2023-01-23 17:00:45.221410: step: 60/1541, loss: 0.0021788461599498987 2023-01-23 17:00:46.120160: step: 62/1541, loss: 0.0870589017868042 2023-01-23 17:00:46.999476: step: 64/1541, loss: 0.047400906682014465 2023-01-23 17:00:47.904237: step: 66/1541, loss: 0.0033587173093110323 2023-01-23 17:00:48.782967: step: 68/1541, loss: 0.0001265244500245899 2023-01-23 17:00:49.669143: step: 70/1541, loss: 0.0009141078917309642 2023-01-23 17:00:50.524319: step: 72/1541, loss: 0.024019485339522362 2023-01-23 17:00:51.415818: step: 74/1541, loss: 0.21441341936588287 2023-01-23 17:00:52.294523: step: 76/1541, loss: 0.004480357747524977 2023-01-23 17:00:53.143083: step: 78/1541, loss: 0.008822847157716751 2023-01-23 17:00:54.023107: step: 80/1541, loss: 0.022926855832338333 2023-01-23 17:00:54.933723: step: 82/1541, loss: 0.051246751099824905 2023-01-23 17:00:55.842958: step: 84/1541, loss: 0.0005282012280076742 2023-01-23 17:00:56.738943: step: 86/1541, loss: 0.07585419714450836 2023-01-23 17:00:57.661558: step: 88/1541, loss: 0.017878185957670212 2023-01-23 17:00:58.543629: step: 90/1541, loss: 0.025505272671580315 2023-01-23 17:00:59.425005: step: 92/1541, loss: 0.0002586142218206078 2023-01-23 17:01:00.344086: step: 94/1541, loss: 9.28951776586473e-05 2023-01-23 17:01:01.236490: step: 96/1541, loss: 8.530765626346692e-05 2023-01-23 17:01:02.153356: step: 98/1541, loss: 0.02162836119532585 2023-01-23 17:01:03.103381: step: 100/1541, loss: 0.04167190566658974 2023-01-23 17:01:03.985242: step: 102/1541, loss: 0.0006969044334255159 2023-01-23 17:01:04.903428: step: 104/1541, loss: 0.0029984430875629187 2023-01-23 17:01:05.790267: step: 106/1541, loss: 0.41456300020217896 2023-01-23 17:01:06.668560: step: 108/1541, loss: 0.0008503337157890201 2023-01-23 17:01:07.624788: step: 110/1541, loss: 0.12498064339160919 2023-01-23 17:01:08.486647: step: 112/1541, loss: 0.0010891275014728308 2023-01-23 17:01:09.400147: step: 114/1541, loss: 0.23973770439624786 2023-01-23 17:01:10.332688: step: 116/1541, loss: 0.08784594386816025 2023-01-23 17:01:11.268888: step: 118/1541, loss: 0.0676150768995285 2023-01-23 17:01:12.149570: step: 120/1541, loss: 0.38759544491767883 2023-01-23 17:01:13.004311: step: 122/1541, loss: 0.04122134670615196 2023-01-23 17:01:13.914163: step: 124/1541, loss: 0.18859225511550903 2023-01-23 17:01:14.816331: step: 126/1541, loss: 0.00026164576411247253 2023-01-23 17:01:15.736525: step: 128/1541, loss: 0.00012880410940852016 2023-01-23 17:01:16.661919: step: 130/1541, loss: 0.08550067991018295 2023-01-23 17:01:17.652502: step: 132/1541, loss: 0.0010346309281885624 2023-01-23 17:01:18.581392: step: 134/1541, loss: 0.00045824440894648433 2023-01-23 17:01:19.446013: step: 136/1541, loss: 0.07105319947004318 2023-01-23 17:01:20.322804: step: 138/1541, loss: 0.10267388075590134 2023-01-23 17:01:21.230717: step: 140/1541, loss: 0.014656292274594307 2023-01-23 17:01:22.113669: step: 142/1541, loss: 0.263631135225296 2023-01-23 17:01:23.037522: step: 144/1541, loss: 0.001104844966903329 2023-01-23 17:01:23.928592: step: 146/1541, loss: 0.001429818570613861 2023-01-23 17:01:24.827833: step: 148/1541, loss: 0.03776904195547104 2023-01-23 17:01:25.714595: step: 150/1541, loss: 0.030956506729125977 2023-01-23 17:01:26.629586: step: 152/1541, loss: 0.0019517067121341825 2023-01-23 17:01:27.538225: step: 154/1541, loss: 0.16111527383327484 2023-01-23 17:01:28.402900: step: 156/1541, loss: 0.01237509772181511 2023-01-23 17:01:29.259092: step: 158/1541, loss: 0.10875825583934784 2023-01-23 17:01:30.107673: step: 160/1541, loss: 0.006582172587513924 2023-01-23 17:01:31.010069: step: 162/1541, loss: 0.0288633331656456 2023-01-23 17:01:31.938606: step: 164/1541, loss: 0.0013208322925493121 2023-01-23 17:01:32.805485: step: 166/1541, loss: 0.05342217534780502 2023-01-23 17:01:33.717057: step: 168/1541, loss: 0.0015428982442244887 2023-01-23 17:01:34.619252: step: 170/1541, loss: 0.022920582443475723 2023-01-23 17:01:35.504688: step: 172/1541, loss: 0.000418320472817868 2023-01-23 17:01:36.416037: step: 174/1541, loss: 0.0004574523773044348 2023-01-23 17:01:37.305129: step: 176/1541, loss: 0.06978371739387512 2023-01-23 17:01:38.191531: step: 178/1541, loss: 0.0006942630279809237 2023-01-23 17:01:39.108641: step: 180/1541, loss: 0.0465424582362175 2023-01-23 17:01:40.010168: step: 182/1541, loss: 0.006051789969205856 2023-01-23 17:01:40.871368: step: 184/1541, loss: 0.15948231518268585 2023-01-23 17:01:41.760468: step: 186/1541, loss: 0.0023395197931677103 2023-01-23 17:01:42.652852: step: 188/1541, loss: 0.0015399008989334106 2023-01-23 17:01:43.582638: step: 190/1541, loss: 0.002438613912090659 2023-01-23 17:01:44.469100: step: 192/1541, loss: 0.0362667553126812 2023-01-23 17:01:45.375839: step: 194/1541, loss: 0.2296576201915741 2023-01-23 17:01:46.328603: step: 196/1541, loss: 8.702322520548478e-05 2023-01-23 17:01:47.300408: step: 198/1541, loss: 0.014764882624149323 2023-01-23 17:01:48.206862: step: 200/1541, loss: 0.00029471947345882654 2023-01-23 17:01:49.126776: step: 202/1541, loss: 0.00010215848305961117 2023-01-23 17:01:50.001639: step: 204/1541, loss: 0.001621446805074811 2023-01-23 17:01:50.926078: step: 206/1541, loss: 0.0383215956389904 2023-01-23 17:01:51.806658: step: 208/1541, loss: 0.11763828992843628 2023-01-23 17:01:52.745591: step: 210/1541, loss: 0.00037147116381675005 2023-01-23 17:01:53.638353: step: 212/1541, loss: 0.10125932097434998 2023-01-23 17:01:54.509401: step: 214/1541, loss: 0.05203818902373314 2023-01-23 17:01:55.396385: step: 216/1541, loss: 0.15284623205661774 2023-01-23 17:01:56.299338: step: 218/1541, loss: 0.00048507677274756134 2023-01-23 17:01:57.177310: step: 220/1541, loss: 0.006823941599577665 2023-01-23 17:01:58.094659: step: 222/1541, loss: 0.2148112654685974 2023-01-23 17:01:59.001507: step: 224/1541, loss: 0.002624376444146037 2023-01-23 17:01:59.997021: step: 226/1541, loss: 0.00045496210805140436 2023-01-23 17:02:00.880683: step: 228/1541, loss: 0.0017591649666428566 2023-01-23 17:02:01.752543: step: 230/1541, loss: 0.08245251327753067 2023-01-23 17:02:02.639589: step: 232/1541, loss: 0.0004125272680539638 2023-01-23 17:02:03.523521: step: 234/1541, loss: 0.10185066610574722 2023-01-23 17:02:04.465921: step: 236/1541, loss: 0.00046043237671256065 2023-01-23 17:02:05.369780: step: 238/1541, loss: 0.019308457151055336 2023-01-23 17:02:06.305237: step: 240/1541, loss: 0.0003523822233546525 2023-01-23 17:02:07.218996: step: 242/1541, loss: 0.11658502370119095 2023-01-23 17:02:08.107897: step: 244/1541, loss: 0.0024159031454473734 2023-01-23 17:02:09.000509: step: 246/1541, loss: 0.00042252492858096957 2023-01-23 17:02:09.883699: step: 248/1541, loss: 0.05053813010454178 2023-01-23 17:02:10.778774: step: 250/1541, loss: 0.006292772013694048 2023-01-23 17:02:11.671072: step: 252/1541, loss: 0.016314812004566193 2023-01-23 17:02:12.565755: step: 254/1541, loss: 0.00029104272834956646 2023-01-23 17:02:13.505429: step: 256/1541, loss: 0.07544568181037903 2023-01-23 17:02:14.381872: step: 258/1541, loss: 0.0005078139947727323 2023-01-23 17:02:15.286702: step: 260/1541, loss: 0.06274508684873581 2023-01-23 17:02:16.180462: step: 262/1541, loss: 0.011803828179836273 2023-01-23 17:02:17.099248: step: 264/1541, loss: 0.00045067898463457823 2023-01-23 17:02:18.001935: step: 266/1541, loss: 0.01204299833625555 2023-01-23 17:02:18.905975: step: 268/1541, loss: 0.11440776288509369 2023-01-23 17:02:19.811239: step: 270/1541, loss: 0.02680211514234543 2023-01-23 17:02:20.703800: step: 272/1541, loss: 0.07221489399671555 2023-01-23 17:02:21.583901: step: 274/1541, loss: 3.931785613531247e-05 2023-01-23 17:02:22.489557: step: 276/1541, loss: 0.08905874937772751 2023-01-23 17:02:23.401705: step: 278/1541, loss: 0.02073686011135578 2023-01-23 17:02:24.374176: step: 280/1541, loss: 0.23940393328666687 2023-01-23 17:02:25.279759: step: 282/1541, loss: 0.001315878122113645 2023-01-23 17:02:26.161522: step: 284/1541, loss: 0.00075399165507406 2023-01-23 17:02:27.083246: step: 286/1541, loss: 0.01617913879454136 2023-01-23 17:02:28.020994: step: 288/1541, loss: 0.025619840249419212 2023-01-23 17:02:28.976274: step: 290/1541, loss: 0.006105935666710138 2023-01-23 17:02:29.874441: step: 292/1541, loss: 0.03384476900100708 2023-01-23 17:02:30.752067: step: 294/1541, loss: 0.012690131552517414 2023-01-23 17:02:31.618955: step: 296/1541, loss: 0.009147836826741695 2023-01-23 17:02:32.546954: step: 298/1541, loss: 0.15372246503829956 2023-01-23 17:02:33.464849: step: 300/1541, loss: 0.00711430050432682 2023-01-23 17:02:34.381195: step: 302/1541, loss: 0.00011252034164499491 2023-01-23 17:02:35.268101: step: 304/1541, loss: 0.022350365296006203 2023-01-23 17:02:36.182849: step: 306/1541, loss: 0.007493768353015184 2023-01-23 17:02:37.031920: step: 308/1541, loss: 0.001867803861387074 2023-01-23 17:02:37.920702: step: 310/1541, loss: 0.015316099859774113 2023-01-23 17:02:38.848148: step: 312/1541, loss: 0.033673569560050964 2023-01-23 17:02:39.735626: step: 314/1541, loss: 0.03760828450322151 2023-01-23 17:02:40.637179: step: 316/1541, loss: -1.7591602954780683e-05 2023-01-23 17:02:41.542107: step: 318/1541, loss: 0.003345077158883214 2023-01-23 17:02:42.474334: step: 320/1541, loss: 0.04866628721356392 2023-01-23 17:02:43.373599: step: 322/1541, loss: 0.0345684215426445 2023-01-23 17:02:44.225260: step: 324/1541, loss: 0.3023178279399872 2023-01-23 17:02:45.102630: step: 326/1541, loss: 0.18756069242954254 2023-01-23 17:02:45.988092: step: 328/1541, loss: 0.0017325482331216335 2023-01-23 17:02:46.891914: step: 330/1541, loss: 0.007252660114318132 2023-01-23 17:02:47.863710: step: 332/1541, loss: 0.0014867017744109035 2023-01-23 17:02:48.735631: step: 334/1541, loss: 0.000107255567854736 2023-01-23 17:02:49.667068: step: 336/1541, loss: 0.01258915662765503 2023-01-23 17:02:50.586496: step: 338/1541, loss: 0.000242748370510526 2023-01-23 17:02:51.493003: step: 340/1541, loss: 0.00407223729416728 2023-01-23 17:02:52.399742: step: 342/1541, loss: 0.05480795353651047 2023-01-23 17:02:53.303896: step: 344/1541, loss: 0.010864868760108948 2023-01-23 17:02:54.266805: step: 346/1541, loss: 0.0040165698155760765 2023-01-23 17:02:55.154135: step: 348/1541, loss: 0.00013428146485239267 2023-01-23 17:02:56.032852: step: 350/1541, loss: 0.03149287402629852 2023-01-23 17:02:56.948115: step: 352/1541, loss: 0.03276216983795166 2023-01-23 17:02:57.856820: step: 354/1541, loss: 0.0026837405748665333 2023-01-23 17:02:58.715962: step: 356/1541, loss: 0.2789282500743866 2023-01-23 17:02:59.590340: step: 358/1541, loss: 0.012342491187155247 2023-01-23 17:03:00.498485: step: 360/1541, loss: 0.0014144277665764093 2023-01-23 17:03:01.433590: step: 362/1541, loss: 0.00012070181401213631 2023-01-23 17:03:02.299628: step: 364/1541, loss: 0.0001995874335989356 2023-01-23 17:03:03.180397: step: 366/1541, loss: 0.015503394417464733 2023-01-23 17:03:04.086436: step: 368/1541, loss: 0.061737559735774994 2023-01-23 17:03:05.033218: step: 370/1541, loss: 0.0008074851939454675 2023-01-23 17:03:05.933045: step: 372/1541, loss: 0.13086877763271332 2023-01-23 17:03:06.785124: step: 374/1541, loss: 0.13909828662872314 2023-01-23 17:03:07.658695: step: 376/1541, loss: 0.0043509905226528645 2023-01-23 17:03:08.535324: step: 378/1541, loss: 0.000104340651887469 2023-01-23 17:03:09.461554: step: 380/1541, loss: 9.03618783922866e-05 2023-01-23 17:03:10.424601: step: 382/1541, loss: 0.22006480395793915 2023-01-23 17:03:11.337364: step: 384/1541, loss: 8.992209768621251e-05 2023-01-23 17:03:12.216895: step: 386/1541, loss: 9.592009882908314e-05 2023-01-23 17:03:13.072552: step: 388/1541, loss: 0.004977799486368895 2023-01-23 17:03:13.986468: step: 390/1541, loss: 0.10614094138145447 2023-01-23 17:03:14.998003: step: 392/1541, loss: 0.07456891983747482 2023-01-23 17:03:15.901721: step: 394/1541, loss: 0.002360598649829626 2023-01-23 17:03:16.797652: step: 396/1541, loss: 0.0011029021115973592 2023-01-23 17:03:17.680355: step: 398/1541, loss: 0.3973184823989868 2023-01-23 17:03:18.639081: step: 400/1541, loss: 0.1660519689321518 2023-01-23 17:03:19.520811: step: 402/1541, loss: 0.0003627676051110029 2023-01-23 17:03:20.438052: step: 404/1541, loss: 0.000533397076651454 2023-01-23 17:03:21.316845: step: 406/1541, loss: 0.06662249565124512 2023-01-23 17:03:22.205311: step: 408/1541, loss: 0.02474486455321312 2023-01-23 17:03:23.077572: step: 410/1541, loss: 0.0005487122107297182 2023-01-23 17:03:23.974219: step: 412/1541, loss: 0.093269944190979 2023-01-23 17:03:24.902889: step: 414/1541, loss: 0.02960943803191185 2023-01-23 17:03:25.768633: step: 416/1541, loss: 0.0006112052360549569 2023-01-23 17:03:26.631797: step: 418/1541, loss: 0.002746363403275609 2023-01-23 17:03:27.518157: step: 420/1541, loss: 0.00016109028365463018 2023-01-23 17:03:28.337827: step: 422/1541, loss: 0.01584661565721035 2023-01-23 17:03:29.242714: step: 424/1541, loss: 0.000925055705010891 2023-01-23 17:03:30.130722: step: 426/1541, loss: 0.23500405251979828 2023-01-23 17:03:31.022740: step: 428/1541, loss: 0.009255999699234962 2023-01-23 17:03:31.947776: step: 430/1541, loss: 0.09806931018829346 2023-01-23 17:03:32.878679: step: 432/1541, loss: 0.0006891153170727193 2023-01-23 17:03:33.786082: step: 434/1541, loss: 0.15305104851722717 2023-01-23 17:03:34.715419: step: 436/1541, loss: 0.001377300126478076 2023-01-23 17:03:35.598829: step: 438/1541, loss: 0.33406323194503784 2023-01-23 17:03:36.461251: step: 440/1541, loss: 0.00015644787345081568 2023-01-23 17:03:37.353483: step: 442/1541, loss: 0.19861772656440735 2023-01-23 17:03:38.236780: step: 444/1541, loss: 0.00019310141215100884 2023-01-23 17:03:39.179974: step: 446/1541, loss: 0.0029067734722048044 2023-01-23 17:03:40.087092: step: 448/1541, loss: 0.0055365776643157005 2023-01-23 17:03:40.962185: step: 450/1541, loss: 0.00029676847043447196 2023-01-23 17:03:41.884802: step: 452/1541, loss: 0.06259805709123611 2023-01-23 17:03:42.751912: step: 454/1541, loss: 0.00037966022500768304 2023-01-23 17:03:43.628209: step: 456/1541, loss: 0.04351133480668068 2023-01-23 17:03:44.550427: step: 458/1541, loss: 0.5794202089309692 2023-01-23 17:03:45.435183: step: 460/1541, loss: 0.014889485202729702 2023-01-23 17:03:46.424734: step: 462/1541, loss: 0.00406012125313282 2023-01-23 17:03:47.310751: step: 464/1541, loss: 0.002541686873883009 2023-01-23 17:03:48.208340: step: 466/1541, loss: 0.0006905526970513165 2023-01-23 17:03:49.070111: step: 468/1541, loss: 0.0022190124727785587 2023-01-23 17:03:49.948924: step: 470/1541, loss: 0.16278880834579468 2023-01-23 17:03:50.879866: step: 472/1541, loss: 0.04773644357919693 2023-01-23 17:03:51.738632: step: 474/1541, loss: 0.0010087299160659313 2023-01-23 17:03:52.615658: step: 476/1541, loss: 0.022599969059228897 2023-01-23 17:03:53.512160: step: 478/1541, loss: 0.019790414720773697 2023-01-23 17:03:54.427195: step: 480/1541, loss: 0.0052987392991781235 2023-01-23 17:03:55.299962: step: 482/1541, loss: 0.02446458674967289 2023-01-23 17:03:56.285843: step: 484/1541, loss: 0.004862937610596418 2023-01-23 17:03:57.151380: step: 486/1541, loss: 0.007654859684407711 2023-01-23 17:03:58.056322: step: 488/1541, loss: 0.011316723190248013 2023-01-23 17:03:58.981932: step: 490/1541, loss: 0.0007859939942136407 2023-01-23 17:03:59.844024: step: 492/1541, loss: 2.1389781977632083e-05 2023-01-23 17:04:00.717561: step: 494/1541, loss: 9.746372234076262e-05 2023-01-23 17:04:01.571887: step: 496/1541, loss: 0.13018301129341125 2023-01-23 17:04:02.447241: step: 498/1541, loss: 0.06393683701753616 2023-01-23 17:04:03.296575: step: 500/1541, loss: 0.023668989539146423 2023-01-23 17:04:04.179303: step: 502/1541, loss: 0.03553786128759384 2023-01-23 17:04:05.111818: step: 504/1541, loss: 0.00923857931047678 2023-01-23 17:04:06.018228: step: 506/1541, loss: 0.0894179493188858 2023-01-23 17:04:06.866499: step: 508/1541, loss: 0.06747051328420639 2023-01-23 17:04:07.740984: step: 510/1541, loss: 0.2122970074415207 2023-01-23 17:04:08.647751: step: 512/1541, loss: 0.5458452701568604 2023-01-23 17:04:09.537421: step: 514/1541, loss: 0.04279547929763794 2023-01-23 17:04:10.481025: step: 516/1541, loss: 0.004105870611965656 2023-01-23 17:04:11.386503: step: 518/1541, loss: 0.0006260580266825855 2023-01-23 17:04:12.321299: step: 520/1541, loss: 0.08263470232486725 2023-01-23 17:04:13.221587: step: 522/1541, loss: 0.11414670944213867 2023-01-23 17:04:14.159346: step: 524/1541, loss: 0.2747301459312439 2023-01-23 17:04:15.037590: step: 526/1541, loss: 0.00016565719852223992 2023-01-23 17:04:15.948376: step: 528/1541, loss: 0.006030371878296137 2023-01-23 17:04:16.804121: step: 530/1541, loss: 0.028598669916391373 2023-01-23 17:04:17.675168: step: 532/1541, loss: 0.0031585823744535446 2023-01-23 17:04:18.551288: step: 534/1541, loss: 0.006507818587124348 2023-01-23 17:04:19.485303: step: 536/1541, loss: 0.0005600858712568879 2023-01-23 17:04:20.351861: step: 538/1541, loss: 0.00015317682118620723 2023-01-23 17:04:21.258261: step: 540/1541, loss: 9.14433621801436e-05 2023-01-23 17:04:22.152853: step: 542/1541, loss: 0.023274095728993416 2023-01-23 17:04:23.054359: step: 544/1541, loss: 0.011893963441252708 2023-01-23 17:04:23.964153: step: 546/1541, loss: 0.04474298655986786 2023-01-23 17:04:24.922836: step: 548/1541, loss: 0.00010209984611719847 2023-01-23 17:04:25.797057: step: 550/1541, loss: 0.002224985510110855 2023-01-23 17:04:26.708701: step: 552/1541, loss: 0.011717995628714561 2023-01-23 17:04:27.590190: step: 554/1541, loss: 0.0015772228362038732 2023-01-23 17:04:28.507340: step: 556/1541, loss: 0.0033210485707968473 2023-01-23 17:04:29.391133: step: 558/1541, loss: 0.09040245413780212 2023-01-23 17:04:30.282662: step: 560/1541, loss: 0.008765438571572304 2023-01-23 17:04:31.188762: step: 562/1541, loss: 0.3869897425174713 2023-01-23 17:04:32.127399: step: 564/1541, loss: 0.001351348590105772 2023-01-23 17:04:33.061855: step: 566/1541, loss: 0.5769901871681213 2023-01-23 17:04:33.947789: step: 568/1541, loss: 0.2213781625032425 2023-01-23 17:04:34.812974: step: 570/1541, loss: 1.2323925495147705 2023-01-23 17:04:35.710347: step: 572/1541, loss: 0.000312946445774287 2023-01-23 17:04:36.604992: step: 574/1541, loss: 0.0007139468798413873 2023-01-23 17:04:37.512659: step: 576/1541, loss: 0.027170710265636444 2023-01-23 17:04:38.401191: step: 578/1541, loss: 1.425292339263251e-05 2023-01-23 17:04:39.265926: step: 580/1541, loss: 0.024242345243692398 2023-01-23 17:04:40.140324: step: 582/1541, loss: 0.000788695877417922 2023-01-23 17:04:41.011559: step: 584/1541, loss: 0.0007064969977363944 2023-01-23 17:04:41.926992: step: 586/1541, loss: 0.0003793729410972446 2023-01-23 17:04:42.796451: step: 588/1541, loss: 0.011988095007836819 2023-01-23 17:04:43.732954: step: 590/1541, loss: 0.3811056315898895 2023-01-23 17:04:44.624720: step: 592/1541, loss: 0.0016714079538360238 2023-01-23 17:04:45.502344: step: 594/1541, loss: 0.12478533387184143 2023-01-23 17:04:46.406163: step: 596/1541, loss: 0.34161919355392456 2023-01-23 17:04:47.337774: step: 598/1541, loss: 0.002024205168709159 2023-01-23 17:04:48.272170: step: 600/1541, loss: 0.002022536937147379 2023-01-23 17:04:49.152974: step: 602/1541, loss: 0.06869419664144516 2023-01-23 17:04:50.049374: step: 604/1541, loss: 0.12984715402126312 2023-01-23 17:04:50.929774: step: 606/1541, loss: 0.00216683535836637 2023-01-23 17:04:51.795569: step: 608/1541, loss: 0.0158788301050663 2023-01-23 17:04:52.660428: step: 610/1541, loss: 0.00022128754062578082 2023-01-23 17:04:53.598124: step: 612/1541, loss: 0.00034621538361534476 2023-01-23 17:04:54.478252: step: 614/1541, loss: 0.06516607850790024 2023-01-23 17:04:55.347466: step: 616/1541, loss: 0.0012146709486842155 2023-01-23 17:04:56.250446: step: 618/1541, loss: 0.0023803566582500935 2023-01-23 17:04:57.168437: step: 620/1541, loss: 0.03528718277812004 2023-01-23 17:04:58.068348: step: 622/1541, loss: 0.001961890608072281 2023-01-23 17:04:59.087173: step: 624/1541, loss: 0.018393900245428085 2023-01-23 17:05:00.028066: step: 626/1541, loss: 0.013398155570030212 2023-01-23 17:05:00.894511: step: 628/1541, loss: 0.018724635243415833 2023-01-23 17:05:01.807564: step: 630/1541, loss: 0.0011116639943793416 2023-01-23 17:05:02.715663: step: 632/1541, loss: 0.15350402891635895 2023-01-23 17:05:03.651396: step: 634/1541, loss: 0.0011318889446556568 2023-01-23 17:05:04.603412: step: 636/1541, loss: 0.005399349145591259 2023-01-23 17:05:05.492223: step: 638/1541, loss: 0.10659746825695038 2023-01-23 17:05:06.378500: step: 640/1541, loss: 0.43762755393981934 2023-01-23 17:05:07.261057: step: 642/1541, loss: 0.0009116354631260037 2023-01-23 17:05:08.147876: step: 644/1541, loss: 0.03436025604605675 2023-01-23 17:05:09.068532: step: 646/1541, loss: 0.1451844573020935 2023-01-23 17:05:09.939821: step: 648/1541, loss: 0.000382530182832852 2023-01-23 17:05:10.787243: step: 650/1541, loss: 0.00191210163757205 2023-01-23 17:05:11.643144: step: 652/1541, loss: 0.00025785667821764946 2023-01-23 17:05:12.539468: step: 654/1541, loss: 0.03667900711297989 2023-01-23 17:05:13.446796: step: 656/1541, loss: 0.43257150053977966 2023-01-23 17:05:14.341411: step: 658/1541, loss: 0.006619419902563095 2023-01-23 17:05:15.246742: step: 660/1541, loss: 0.045947495847940445 2023-01-23 17:05:16.174966: step: 662/1541, loss: 0.02967718429863453 2023-01-23 17:05:17.067050: step: 664/1541, loss: 0.04263123497366905 2023-01-23 17:05:17.952746: step: 666/1541, loss: 0.0041784862987697124 2023-01-23 17:05:18.873615: step: 668/1541, loss: 0.0022972896695137024 2023-01-23 17:05:19.741627: step: 670/1541, loss: 0.12465342879295349 2023-01-23 17:05:20.652846: step: 672/1541, loss: 0.032209113240242004 2023-01-23 17:05:21.524571: step: 674/1541, loss: 0.0034674308262765408 2023-01-23 17:05:22.393372: step: 676/1541, loss: 0.016248004510998726 2023-01-23 17:05:23.278787: step: 678/1541, loss: 0.00031197271891869605 2023-01-23 17:05:24.175456: step: 680/1541, loss: 0.003877928014844656 2023-01-23 17:05:25.095701: step: 682/1541, loss: 0.0028125976677984 2023-01-23 17:05:25.993851: step: 684/1541, loss: 0.24513497948646545 2023-01-23 17:05:26.925928: step: 686/1541, loss: 0.005115031730383635 2023-01-23 17:05:27.937862: step: 688/1541, loss: 0.1483663022518158 2023-01-23 17:05:28.860776: step: 690/1541, loss: 0.08410069346427917 2023-01-23 17:05:29.789078: step: 692/1541, loss: 0.4511849880218506 2023-01-23 17:05:30.681330: step: 694/1541, loss: 0.0696168839931488 2023-01-23 17:05:31.566047: step: 696/1541, loss: 0.0005960798007436097 2023-01-23 17:05:32.474143: step: 698/1541, loss: 0.0012962162727490067 2023-01-23 17:05:33.500256: step: 700/1541, loss: 0.0013412358239293098 2023-01-23 17:05:34.402218: step: 702/1541, loss: 0.3529190719127655 2023-01-23 17:05:35.269281: step: 704/1541, loss: 0.034530334174633026 2023-01-23 17:05:36.191307: step: 706/1541, loss: 0.0077073038555681705 2023-01-23 17:05:37.090535: step: 708/1541, loss: 0.011830468662083149 2023-01-23 17:05:37.950004: step: 710/1541, loss: 0.0008359226631000638 2023-01-23 17:05:38.821101: step: 712/1541, loss: 0.0007920563803054392 2023-01-23 17:05:39.705636: step: 714/1541, loss: 0.49177658557891846 2023-01-23 17:05:40.630981: step: 716/1541, loss: 0.004185529891401529 2023-01-23 17:05:41.542541: step: 718/1541, loss: 6.752718036295846e-05 2023-01-23 17:05:42.426608: step: 720/1541, loss: 0.0027529315557330847 2023-01-23 17:05:43.344335: step: 722/1541, loss: 0.0012776072835549712 2023-01-23 17:05:44.214154: step: 724/1541, loss: 0.0009375690133310854 2023-01-23 17:05:45.110329: step: 726/1541, loss: 0.07481340318918228 2023-01-23 17:05:45.992019: step: 728/1541, loss: 0.000580829509999603 2023-01-23 17:05:46.907292: step: 730/1541, loss: 0.03642081469297409 2023-01-23 17:05:47.820518: step: 732/1541, loss: 0.006533415988087654 2023-01-23 17:05:48.750484: step: 734/1541, loss: 0.2052232325077057 2023-01-23 17:05:49.651843: step: 736/1541, loss: 0.0008058113744482398 2023-01-23 17:05:50.606121: step: 738/1541, loss: 0.028174452483654022 2023-01-23 17:05:51.508333: step: 740/1541, loss: 3.512563489493914e-05 2023-01-23 17:05:52.346082: step: 742/1541, loss: 0.00015816098311915994 2023-01-23 17:05:53.214805: step: 744/1541, loss: 0.03906307741999626 2023-01-23 17:05:54.090558: step: 746/1541, loss: 0.11144234985113144 2023-01-23 17:05:54.978584: step: 748/1541, loss: 0.024998705834150314 2023-01-23 17:05:55.823925: step: 750/1541, loss: 0.02400185912847519 2023-01-23 17:05:56.718342: step: 752/1541, loss: 0.08489851653575897 2023-01-23 17:05:57.599858: step: 754/1541, loss: 0.0009188094991259277 2023-01-23 17:05:58.466879: step: 756/1541, loss: 0.005098456516861916 2023-01-23 17:05:59.394347: step: 758/1541, loss: 0.001988882664591074 2023-01-23 17:06:00.290591: step: 760/1541, loss: 0.0042216358706355095 2023-01-23 17:06:01.165682: step: 762/1541, loss: 0.20727822184562683 2023-01-23 17:06:02.075169: step: 764/1541, loss: 0.2532865107059479 2023-01-23 17:06:02.988780: step: 766/1541, loss: 0.0003434932150412351 2023-01-23 17:06:03.901970: step: 768/1541, loss: 0.0002130858920281753 2023-01-23 17:06:04.769791: step: 770/1541, loss: 0.00029546732548624277 2023-01-23 17:06:05.695654: step: 772/1541, loss: 0.021178193390369415 2023-01-23 17:06:06.596099: step: 774/1541, loss: 0.26272889971733093 2023-01-23 17:06:07.459263: step: 776/1541, loss: 0.0013331131776794791 2023-01-23 17:06:08.413152: step: 778/1541, loss: 0.0035092015750706196 2023-01-23 17:06:09.351234: step: 780/1541, loss: 0.00042584093171171844 2023-01-23 17:06:10.323347: step: 782/1541, loss: 0.007547616492956877 2023-01-23 17:06:11.186021: step: 784/1541, loss: 0.014332088641822338 2023-01-23 17:06:12.059397: step: 786/1541, loss: 0.06718095391988754 2023-01-23 17:06:12.950761: step: 788/1541, loss: 0.20746630430221558 2023-01-23 17:06:13.879945: step: 790/1541, loss: 0.004753566347062588 2023-01-23 17:06:14.769473: step: 792/1541, loss: 0.0068072001449763775 2023-01-23 17:06:15.661416: step: 794/1541, loss: 0.08147747814655304 2023-01-23 17:06:16.539633: step: 796/1541, loss: 0.060260653495788574 2023-01-23 17:06:17.450783: step: 798/1541, loss: 0.03430810570716858 2023-01-23 17:06:18.346105: step: 800/1541, loss: 0.00424968870356679 2023-01-23 17:06:19.220062: step: 802/1541, loss: 0.0007482982473447919 2023-01-23 17:06:20.180582: step: 804/1541, loss: 0.020157791674137115 2023-01-23 17:06:21.019198: step: 806/1541, loss: 0.04084713011980057 2023-01-23 17:06:21.933080: step: 808/1541, loss: 0.39787349104881287 2023-01-23 17:06:22.810158: step: 810/1541, loss: 0.03889156132936478 2023-01-23 17:06:23.740129: step: 812/1541, loss: 0.1052522137761116 2023-01-23 17:06:24.652465: step: 814/1541, loss: 0.014092565514147282 2023-01-23 17:06:25.606232: step: 816/1541, loss: 0.1794384866952896 2023-01-23 17:06:26.535961: step: 818/1541, loss: 0.0007621488184668124 2023-01-23 17:06:27.486851: step: 820/1541, loss: 0.08874303102493286 2023-01-23 17:06:28.411394: step: 822/1541, loss: 0.05450074374675751 2023-01-23 17:06:29.328925: step: 824/1541, loss: 0.0003472281387075782 2023-01-23 17:06:30.191574: step: 826/1541, loss: 0.0022330181673169136 2023-01-23 17:06:31.067001: step: 828/1541, loss: 0.00036683405051007867 2023-01-23 17:06:32.001352: step: 830/1541, loss: 0.008758390322327614 2023-01-23 17:06:32.845972: step: 832/1541, loss: 0.002026219153776765 2023-01-23 17:06:33.728254: step: 834/1541, loss: 0.00023544921714346856 2023-01-23 17:06:34.656129: step: 836/1541, loss: 0.026270408183336258 2023-01-23 17:06:35.520477: step: 838/1541, loss: 0.08234167844057083 2023-01-23 17:06:36.429637: step: 840/1541, loss: 0.00032017691410146654 2023-01-23 17:06:37.307512: step: 842/1541, loss: 0.0009908397914841771 2023-01-23 17:06:38.184599: step: 844/1541, loss: 0.030147138983011246 2023-01-23 17:06:39.063167: step: 846/1541, loss: 0.16465239226818085 2023-01-23 17:06:39.941904: step: 848/1541, loss: 0.013932893052697182 2023-01-23 17:06:40.901196: step: 850/1541, loss: 0.03805963695049286 2023-01-23 17:06:41.814933: step: 852/1541, loss: 0.11103500425815582 2023-01-23 17:06:42.715922: step: 854/1541, loss: 0.39527377486228943 2023-01-23 17:06:43.628196: step: 856/1541, loss: 0.0028866613283753395 2023-01-23 17:06:44.537108: step: 858/1541, loss: 0.04155995696783066 2023-01-23 17:06:45.430281: step: 860/1541, loss: 0.09238781780004501 2023-01-23 17:06:46.340471: step: 862/1541, loss: 0.0036763770040124655 2023-01-23 17:06:47.298496: step: 864/1541, loss: 0.004592909011989832 2023-01-23 17:06:48.184621: step: 866/1541, loss: 0.03522081300616264 2023-01-23 17:06:49.024378: step: 868/1541, loss: 0.0003388436743989587 2023-01-23 17:06:49.920511: step: 870/1541, loss: 0.0009960797615349293 2023-01-23 17:06:50.789025: step: 872/1541, loss: 0.002213866915553808 2023-01-23 17:06:51.741553: step: 874/1541, loss: 0.29739102721214294 2023-01-23 17:06:52.596831: step: 876/1541, loss: 0.0005624319310300052 2023-01-23 17:06:53.504439: step: 878/1541, loss: 0.15502126514911652 2023-01-23 17:06:54.450805: step: 880/1541, loss: 5.915463043493219e-05 2023-01-23 17:06:55.371960: step: 882/1541, loss: 1.2362210750579834 2023-01-23 17:06:56.268128: step: 884/1541, loss: 0.044239070266485214 2023-01-23 17:06:57.161018: step: 886/1541, loss: 0.016184132546186447 2023-01-23 17:06:58.180422: step: 888/1541, loss: 8.878907829057425e-05 2023-01-23 17:06:59.080425: step: 890/1541, loss: 0.00038162883720360696 2023-01-23 17:06:59.972961: step: 892/1541, loss: 0.017248088493943214 2023-01-23 17:07:00.860475: step: 894/1541, loss: 0.004395094700157642 2023-01-23 17:07:01.729664: step: 896/1541, loss: 0.0003369584446772933 2023-01-23 17:07:02.619946: step: 898/1541, loss: 0.0010794472182169557 2023-01-23 17:07:03.512737: step: 900/1541, loss: 0.4572143256664276 2023-01-23 17:07:04.417947: step: 902/1541, loss: 0.6034805774688721 2023-01-23 17:07:05.377831: step: 904/1541, loss: 0.006136818323284388 2023-01-23 17:07:06.260719: step: 906/1541, loss: 0.008064360357820988 2023-01-23 17:07:07.196145: step: 908/1541, loss: -4.7090306907193735e-06 2023-01-23 17:07:08.164896: step: 910/1541, loss: 0.0027369291055947542 2023-01-23 17:07:09.075234: step: 912/1541, loss: 0.47181975841522217 2023-01-23 17:07:09.980218: step: 914/1541, loss: 0.48792338371276855 2023-01-23 17:07:10.872108: step: 916/1541, loss: 0.0015800939872860909 2023-01-23 17:07:11.899066: step: 918/1541, loss: 0.011018220335245132 2023-01-23 17:07:12.817388: step: 920/1541, loss: 0.0018927704077214003 2023-01-23 17:07:13.711509: step: 922/1541, loss: 0.0024699096102267504 2023-01-23 17:07:14.662156: step: 924/1541, loss: 0.0006455270922742784 2023-01-23 17:07:15.533402: step: 926/1541, loss: 0.09527816623449326 2023-01-23 17:07:16.464642: step: 928/1541, loss: 0.09736799448728561 2023-01-23 17:07:17.380935: step: 930/1541, loss: 0.15559738874435425 2023-01-23 17:07:18.260082: step: 932/1541, loss: 0.00014017612556926906 2023-01-23 17:07:19.140567: step: 934/1541, loss: 0.0003055195847991854 2023-01-23 17:07:20.014919: step: 936/1541, loss: 0.5898765325546265 2023-01-23 17:07:20.973773: step: 938/1541, loss: 0.09739630669355392 2023-01-23 17:07:21.847501: step: 940/1541, loss: 0.00033517874544486403 2023-01-23 17:07:22.721037: step: 942/1541, loss: 0.0014778821496292949 2023-01-23 17:07:23.626960: step: 944/1541, loss: 0.00018407647439744323 2023-01-23 17:07:24.517035: step: 946/1541, loss: 0.0003994582511950284 2023-01-23 17:07:25.478125: step: 948/1541, loss: 0.0003034265246242285 2023-01-23 17:07:26.507250: step: 950/1541, loss: 0.0011632718378677964 2023-01-23 17:07:27.410050: step: 952/1541, loss: 0.013656619936227798 2023-01-23 17:07:28.309831: step: 954/1541, loss: 0.0003003640449605882 2023-01-23 17:07:29.179615: step: 956/1541, loss: 0.218220055103302 2023-01-23 17:07:30.034976: step: 958/1541, loss: 0.00014424820255953819 2023-01-23 17:07:30.976769: step: 960/1541, loss: 0.0007928216946311295 2023-01-23 17:07:31.877672: step: 962/1541, loss: 0.02894287370145321 2023-01-23 17:07:32.776788: step: 964/1541, loss: 0.47240307927131653 2023-01-23 17:07:33.649358: step: 966/1541, loss: 0.33062466979026794 2023-01-23 17:07:34.554850: step: 968/1541, loss: 0.2597869038581848 2023-01-23 17:07:35.444396: step: 970/1541, loss: 0.03022507205605507 2023-01-23 17:07:36.328589: step: 972/1541, loss: 0.5381840467453003 2023-01-23 17:07:37.217169: step: 974/1541, loss: 0.08356763422489166 2023-01-23 17:07:38.090849: step: 976/1541, loss: 0.008808592334389687 2023-01-23 17:07:39.014854: step: 978/1541, loss: 0.1297139674425125 2023-01-23 17:07:39.889626: step: 980/1541, loss: 0.04509643092751503 2023-01-23 17:07:40.783164: step: 982/1541, loss: 0.0011504950234666467 2023-01-23 17:07:41.705354: step: 984/1541, loss: 0.036199938505887985 2023-01-23 17:07:42.594682: step: 986/1541, loss: 0.0034929318353533745 2023-01-23 17:07:43.562237: step: 988/1541, loss: 0.17463548481464386 2023-01-23 17:07:44.494959: step: 990/1541, loss: 0.1231488436460495 2023-01-23 17:07:45.384432: step: 992/1541, loss: 0.004786649718880653 2023-01-23 17:07:46.294273: step: 994/1541, loss: 0.0009574749856255949 2023-01-23 17:07:47.186872: step: 996/1541, loss: 0.2648865282535553 2023-01-23 17:07:48.131570: step: 998/1541, loss: 0.0021913249511271715 2023-01-23 17:07:49.032240: step: 1000/1541, loss: 0.0009030459332279861 2023-01-23 17:07:49.968917: step: 1002/1541, loss: 0.09153751283884048 2023-01-23 17:07:50.851826: step: 1004/1541, loss: 0.002623272128403187 2023-01-23 17:07:51.739360: step: 1006/1541, loss: 0.0003255514020565897 2023-01-23 17:07:52.614000: step: 1008/1541, loss: 1.3689870834350586 2023-01-23 17:07:53.493350: step: 1010/1541, loss: 0.2497362494468689 2023-01-23 17:07:54.393325: step: 1012/1541, loss: 0.7738041281700134 2023-01-23 17:07:55.256696: step: 1014/1541, loss: 0.009949360974133015 2023-01-23 17:07:56.172488: step: 1016/1541, loss: 0.0004513350431807339 2023-01-23 17:07:57.072432: step: 1018/1541, loss: 0.000783382507506758 2023-01-23 17:07:57.963631: step: 1020/1541, loss: 0.0021769318263977766 2023-01-23 17:07:58.946259: step: 1022/1541, loss: 0.0005904954741708934 2023-01-23 17:07:59.808808: step: 1024/1541, loss: 0.09134450554847717 2023-01-23 17:08:00.757221: step: 1026/1541, loss: 1.4179476499557495 2023-01-23 17:08:01.691942: step: 1028/1541, loss: 0.0453750379383564 2023-01-23 17:08:02.654971: step: 1030/1541, loss: 0.1555854231119156 2023-01-23 17:08:03.521973: step: 1032/1541, loss: 0.0037468471564352512 2023-01-23 17:08:04.385973: step: 1034/1541, loss: 0.009282751008868217 2023-01-23 17:08:05.278065: step: 1036/1541, loss: 0.35208719968795776 2023-01-23 17:08:06.117903: step: 1038/1541, loss: 0.07740478217601776 2023-01-23 17:08:06.974126: step: 1040/1541, loss: 0.009965543635189533 2023-01-23 17:08:07.885311: step: 1042/1541, loss: 0.3079136610031128 2023-01-23 17:08:08.745047: step: 1044/1541, loss: 0.07108455896377563 2023-01-23 17:08:09.698245: step: 1046/1541, loss: 0.017016980797052383 2023-01-23 17:08:10.608063: step: 1048/1541, loss: 0.021965261548757553 2023-01-23 17:08:11.479997: step: 1050/1541, loss: 0.0002979933051392436 2023-01-23 17:08:12.388506: step: 1052/1541, loss: 0.0027674930170178413 2023-01-23 17:08:13.261134: step: 1054/1541, loss: 0.016461288556456566 2023-01-23 17:08:14.156137: step: 1056/1541, loss: 0.03219176456332207 2023-01-23 17:08:15.071913: step: 1058/1541, loss: 0.028309080749750137 2023-01-23 17:08:16.034059: step: 1060/1541, loss: 0.004040006082504988 2023-01-23 17:08:16.965909: step: 1062/1541, loss: 0.025911841541528702 2023-01-23 17:08:17.956653: step: 1064/1541, loss: 0.005848179571330547 2023-01-23 17:08:18.892706: step: 1066/1541, loss: 0.05013977736234665 2023-01-23 17:08:19.783141: step: 1068/1541, loss: 0.0033953150268644094 2023-01-23 17:08:20.638931: step: 1070/1541, loss: 0.0032327473163604736 2023-01-23 17:08:21.531371: step: 1072/1541, loss: 0.001044316217303276 2023-01-23 17:08:22.524485: step: 1074/1541, loss: 0.05904532968997955 2023-01-23 17:08:23.428346: step: 1076/1541, loss: 0.0002842434332706034 2023-01-23 17:08:24.336224: step: 1078/1541, loss: 0.17226170003414154 2023-01-23 17:08:25.220725: step: 1080/1541, loss: 0.006180042400956154 2023-01-23 17:08:26.186692: step: 1082/1541, loss: 0.0005132812075316906 2023-01-23 17:08:27.043471: step: 1084/1541, loss: 0.00011309704859741032 2023-01-23 17:08:27.946969: step: 1086/1541, loss: 0.003725511720404029 2023-01-23 17:08:28.851501: step: 1088/1541, loss: 0.004272944759577513 2023-01-23 17:08:29.759012: step: 1090/1541, loss: 0.22128218412399292 2023-01-23 17:08:30.650690: step: 1092/1541, loss: 0.13702358305454254 2023-01-23 17:08:31.559427: step: 1094/1541, loss: 0.06011774390935898 2023-01-23 17:08:32.467093: step: 1096/1541, loss: 0.0029443465173244476 2023-01-23 17:08:33.339851: step: 1098/1541, loss: 0.28936100006103516 2023-01-23 17:08:34.255591: step: 1100/1541, loss: 0.0012440166901797056 2023-01-23 17:08:35.132266: step: 1102/1541, loss: 1.2345874309539795 2023-01-23 17:08:36.054970: step: 1104/1541, loss: 0.0017692269757390022 2023-01-23 17:08:36.945394: step: 1106/1541, loss: 0.022110480815172195 2023-01-23 17:08:37.826147: step: 1108/1541, loss: 0.000691669643856585 2023-01-23 17:08:38.698878: step: 1110/1541, loss: 0.0004601788823492825 2023-01-23 17:08:39.573362: step: 1112/1541, loss: 0.0002518592227715999 2023-01-23 17:08:40.581098: step: 1114/1541, loss: 0.028832852840423584 2023-01-23 17:08:41.506855: step: 1116/1541, loss: 0.0010923831723630428 2023-01-23 17:08:42.397424: step: 1118/1541, loss: 0.04534095525741577 2023-01-23 17:08:43.366452: step: 1120/1541, loss: 0.013741308823227882 2023-01-23 17:08:44.333905: step: 1122/1541, loss: 0.06248645484447479 2023-01-23 17:08:45.222399: step: 1124/1541, loss: 0.0003472274402156472 2023-01-23 17:08:46.167252: step: 1126/1541, loss: 0.15639711916446686 2023-01-23 17:08:47.080254: step: 1128/1541, loss: 0.24560660123825073 2023-01-23 17:08:48.080478: step: 1130/1541, loss: 0.0006129321991465986 2023-01-23 17:08:48.938200: step: 1132/1541, loss: 0.00032051288872025907 2023-01-23 17:08:49.834202: step: 1134/1541, loss: 0.04040738195180893 2023-01-23 17:08:50.750122: step: 1136/1541, loss: 0.6909326314926147 2023-01-23 17:08:51.698917: step: 1138/1541, loss: 9.908380889100954e-05 2023-01-23 17:08:52.677080: step: 1140/1541, loss: 8.815937326289713e-05 2023-01-23 17:08:53.551040: step: 1142/1541, loss: 0.0239795483648777 2023-01-23 17:08:54.443196: step: 1144/1541, loss: 0.0004304326721467078 2023-01-23 17:08:55.335540: step: 1146/1541, loss: 0.0024003498256206512 2023-01-23 17:08:56.216923: step: 1148/1541, loss: 0.0009523184271529317 2023-01-23 17:08:57.065729: step: 1150/1541, loss: 0.0005901118856854737 2023-01-23 17:08:57.942906: step: 1152/1541, loss: 0.005542958155274391 2023-01-23 17:08:58.797650: step: 1154/1541, loss: 0.0012604841031134129 2023-01-23 17:08:59.681188: step: 1156/1541, loss: 0.0006464349571615458 2023-01-23 17:09:00.653636: step: 1158/1541, loss: 0.00011079792602686211 2023-01-23 17:09:01.530252: step: 1160/1541, loss: 0.00010833427950274199 2023-01-23 17:09:02.463539: step: 1162/1541, loss: 0.0005696006119251251 2023-01-23 17:09:03.332259: step: 1164/1541, loss: 0.010755710303783417 2023-01-23 17:09:04.205318: step: 1166/1541, loss: 0.009763631969690323 2023-01-23 17:09:05.115403: step: 1168/1541, loss: 0.0014884683769196272 2023-01-23 17:09:06.002292: step: 1170/1541, loss: 0.0003470857336651534 2023-01-23 17:09:06.857004: step: 1172/1541, loss: 0.000806037918664515 2023-01-23 17:09:07.807831: step: 1174/1541, loss: 0.017098402604460716 2023-01-23 17:09:08.689278: step: 1176/1541, loss: 0.00020762185158673674 2023-01-23 17:09:09.585586: step: 1178/1541, loss: 0.0042701102793216705 2023-01-23 17:09:10.477325: step: 1180/1541, loss: 0.7268374562263489 2023-01-23 17:09:11.336212: step: 1182/1541, loss: 0.00012663498637266457 2023-01-23 17:09:12.235489: step: 1184/1541, loss: 0.00041486372356303036 2023-01-23 17:09:13.110141: step: 1186/1541, loss: 0.0794811025261879 2023-01-23 17:09:13.964828: step: 1188/1541, loss: 0.002010463969781995 2023-01-23 17:09:14.866223: step: 1190/1541, loss: 0.00010024998482549563 2023-01-23 17:09:15.819649: step: 1192/1541, loss: 0.44886523485183716 2023-01-23 17:09:16.741809: step: 1194/1541, loss: 0.16819161176681519 2023-01-23 17:09:17.613177: step: 1196/1541, loss: 0.05213456228375435 2023-01-23 17:09:18.494829: step: 1198/1541, loss: 0.49328452348709106 2023-01-23 17:09:19.434845: step: 1200/1541, loss: 0.030437221750617027 2023-01-23 17:09:20.344967: step: 1202/1541, loss: 0.017906516790390015 2023-01-23 17:09:21.210629: step: 1204/1541, loss: 0.002885902766138315 2023-01-23 17:09:22.114539: step: 1206/1541, loss: 0.10684384405612946 2023-01-23 17:09:23.001594: step: 1208/1541, loss: 0.001095841871574521 2023-01-23 17:09:23.921913: step: 1210/1541, loss: 0.3054278492927551 2023-01-23 17:09:24.834501: step: 1212/1541, loss: 1.429979681968689 2023-01-23 17:09:25.729405: step: 1214/1541, loss: 0.0022896677255630493 2023-01-23 17:09:26.614857: step: 1216/1541, loss: 0.051301028579473495 2023-01-23 17:09:27.565548: step: 1218/1541, loss: 0.02899741381406784 2023-01-23 17:09:28.423737: step: 1220/1541, loss: 0.03815879672765732 2023-01-23 17:09:29.347736: step: 1222/1541, loss: 0.0003998707397840917 2023-01-23 17:09:30.228505: step: 1224/1541, loss: 0.09648828208446503 2023-01-23 17:09:31.086901: step: 1226/1541, loss: 0.0017947485903277993 2023-01-23 17:09:31.953348: step: 1228/1541, loss: 0.00026861432706937194 2023-01-23 17:09:32.829859: step: 1230/1541, loss: 0.007202246692031622 2023-01-23 17:09:33.766562: step: 1232/1541, loss: 0.00029865207034163177 2023-01-23 17:09:34.642446: step: 1234/1541, loss: 0.013261341489851475 2023-01-23 17:09:35.493074: step: 1236/1541, loss: 0.00023000253713689744 2023-01-23 17:09:36.418791: step: 1238/1541, loss: 0.004979168996214867 2023-01-23 17:09:37.294588: step: 1240/1541, loss: 0.02878263033926487 2023-01-23 17:09:38.233496: step: 1242/1541, loss: 0.024354800581932068 2023-01-23 17:09:39.108914: step: 1244/1541, loss: 0.0009253235184587538 2023-01-23 17:09:40.007141: step: 1246/1541, loss: 0.0009324104175902903 2023-01-23 17:09:40.912159: step: 1248/1541, loss: 0.02407209947705269 2023-01-23 17:09:41.844923: step: 1250/1541, loss: 0.02082066796720028 2023-01-23 17:09:42.742554: step: 1252/1541, loss: 0.00023067256552167237 2023-01-23 17:09:43.710935: step: 1254/1541, loss: 0.003785247216001153 2023-01-23 17:09:44.580108: step: 1256/1541, loss: 0.01651635766029358 2023-01-23 17:09:45.539965: step: 1258/1541, loss: 0.009974371641874313 2023-01-23 17:09:46.412178: step: 1260/1541, loss: 0.02189009264111519 2023-01-23 17:09:47.357243: step: 1262/1541, loss: 0.24993301928043365 2023-01-23 17:09:48.260831: step: 1264/1541, loss: 0.03711336478590965 2023-01-23 17:09:49.148631: step: 1266/1541, loss: 0.010278301313519478 2023-01-23 17:09:50.028243: step: 1268/1541, loss: 7.901963545009494e-05 2023-01-23 17:09:51.020655: step: 1270/1541, loss: 0.01774212159216404 2023-01-23 17:09:51.904886: step: 1272/1541, loss: 0.0011595595860853791 2023-01-23 17:09:52.823433: step: 1274/1541, loss: 0.29365718364715576 2023-01-23 17:09:53.699907: step: 1276/1541, loss: 0.00046646641567349434 2023-01-23 17:09:54.597019: step: 1278/1541, loss: 0.08379911631345749 2023-01-23 17:09:55.565132: step: 1280/1541, loss: 0.00624519819393754 2023-01-23 17:09:56.511207: step: 1282/1541, loss: 9.110265091294423e-05 2023-01-23 17:09:57.396399: step: 1284/1541, loss: 0.3709786832332611 2023-01-23 17:09:58.324566: step: 1286/1541, loss: 0.001787033281289041 2023-01-23 17:09:59.219028: step: 1288/1541, loss: 0.3029506802558899 2023-01-23 17:10:00.114502: step: 1290/1541, loss: 0.0869981124997139 2023-01-23 17:10:01.008362: step: 1292/1541, loss: 0.1735081970691681 2023-01-23 17:10:01.948871: step: 1294/1541, loss: 0.000966300256550312 2023-01-23 17:10:02.888513: step: 1296/1541, loss: 0.0013893635477870703 2023-01-23 17:10:03.756669: step: 1298/1541, loss: 0.0031906827352941036 2023-01-23 17:10:04.645765: step: 1300/1541, loss: 0.0033762026578187943 2023-01-23 17:10:05.599516: step: 1302/1541, loss: 0.0019617662765085697 2023-01-23 17:10:06.536163: step: 1304/1541, loss: 0.0007713902159593999 2023-01-23 17:10:07.443514: step: 1306/1541, loss: 0.15676406025886536 2023-01-23 17:10:08.402349: step: 1308/1541, loss: 0.013199751265347004 2023-01-23 17:10:09.272500: step: 1310/1541, loss: 0.0004174764617346227 2023-01-23 17:10:10.157535: step: 1312/1541, loss: 0.028482060879468918 2023-01-23 17:10:11.062049: step: 1314/1541, loss: 0.007813194766640663 2023-01-23 17:10:11.923167: step: 1316/1541, loss: 0.003416449297219515 2023-01-23 17:10:12.796610: step: 1318/1541, loss: 0.12076474726200104 2023-01-23 17:10:13.672899: step: 1320/1541, loss: 0.24676519632339478 2023-01-23 17:10:14.602845: step: 1322/1541, loss: 0.008879224769771099 2023-01-23 17:10:15.491281: step: 1324/1541, loss: 0.004565782379359007 2023-01-23 17:10:16.436307: step: 1326/1541, loss: 0.000794190913438797 2023-01-23 17:10:17.338277: step: 1328/1541, loss: 0.0004351310199126601 2023-01-23 17:10:18.217010: step: 1330/1541, loss: 0.0847153589129448 2023-01-23 17:10:19.131320: step: 1332/1541, loss: 0.029500704258680344 2023-01-23 17:10:20.080503: step: 1334/1541, loss: 0.040000513195991516 2023-01-23 17:10:20.968064: step: 1336/1541, loss: 0.0008789162384346128 2023-01-23 17:10:21.816380: step: 1338/1541, loss: 0.004277859814465046 2023-01-23 17:10:22.666775: step: 1340/1541, loss: 0.000215682593989186 2023-01-23 17:10:23.538822: step: 1342/1541, loss: 0.01610513962805271 2023-01-23 17:10:24.414508: step: 1344/1541, loss: 0.008918952196836472 2023-01-23 17:10:25.295285: step: 1346/1541, loss: 0.017588844522833824 2023-01-23 17:10:26.210056: step: 1348/1541, loss: 0.002454598667100072 2023-01-23 17:10:27.106932: step: 1350/1541, loss: 0.0018980999011546373 2023-01-23 17:10:27.966105: step: 1352/1541, loss: 0.02259282022714615 2023-01-23 17:10:28.870178: step: 1354/1541, loss: 0.16808190941810608 2023-01-23 17:10:29.761236: step: 1356/1541, loss: 0.0008217529975809157 2023-01-23 17:10:30.719221: step: 1358/1541, loss: 0.0022891154512763023 2023-01-23 17:10:31.625260: step: 1360/1541, loss: 0.20041733980178833 2023-01-23 17:10:32.541320: step: 1362/1541, loss: 0.04571106284856796 2023-01-23 17:10:33.459282: step: 1364/1541, loss: 0.2834804654121399 2023-01-23 17:10:34.343839: step: 1366/1541, loss: 0.0049079363234341145 2023-01-23 17:10:35.234439: step: 1368/1541, loss: 0.0006954895216040313 2023-01-23 17:10:36.100135: step: 1370/1541, loss: 0.0072464169934391975 2023-01-23 17:10:36.990792: step: 1372/1541, loss: 0.007720367982983589 2023-01-23 17:10:37.918938: step: 1374/1541, loss: 0.0010187271982431412 2023-01-23 17:10:38.771267: step: 1376/1541, loss: 0.004848489072173834 2023-01-23 17:10:39.630426: step: 1378/1541, loss: 0.19729427993297577 2023-01-23 17:10:40.521758: step: 1380/1541, loss: 0.10842276364564896 2023-01-23 17:10:41.399100: step: 1382/1541, loss: 0.01361838635057211 2023-01-23 17:10:42.254845: step: 1384/1541, loss: 0.00026674268883652985 2023-01-23 17:10:43.177740: step: 1386/1541, loss: 0.030490368604660034 2023-01-23 17:10:44.138891: step: 1388/1541, loss: 0.0011717742308974266 2023-01-23 17:10:45.049255: step: 1390/1541, loss: 0.0033954863902181387 2023-01-23 17:10:45.933289: step: 1392/1541, loss: 0.002364734187722206 2023-01-23 17:10:46.797337: step: 1394/1541, loss: 0.000455403293017298 2023-01-23 17:10:47.730347: step: 1396/1541, loss: 0.009351006709039211 2023-01-23 17:10:48.639199: step: 1398/1541, loss: 0.01071165781468153 2023-01-23 17:10:49.517702: step: 1400/1541, loss: 0.12113545089960098 2023-01-23 17:10:50.377146: step: 1402/1541, loss: 0.36160093545913696 2023-01-23 17:10:51.262455: step: 1404/1541, loss: 0.022868899628520012 2023-01-23 17:10:52.172049: step: 1406/1541, loss: 0.0028891353867948055 2023-01-23 17:10:53.048953: step: 1408/1541, loss: 0.006730395834892988 2023-01-23 17:10:53.971777: step: 1410/1541, loss: 0.00014155010285321623 2023-01-23 17:10:54.924915: step: 1412/1541, loss: 0.08207177370786667 2023-01-23 17:10:55.836202: step: 1414/1541, loss: 0.0012270513689145446 2023-01-23 17:10:56.772757: step: 1416/1541, loss: 0.00019257204257883132 2023-01-23 17:10:57.655137: step: 1418/1541, loss: 0.21942339837551117 2023-01-23 17:10:58.563528: step: 1420/1541, loss: 0.0042364164255559444 2023-01-23 17:10:59.444093: step: 1422/1541, loss: 0.0023858153726905584 2023-01-23 17:11:00.348965: step: 1424/1541, loss: 1.1437867879867554 2023-01-23 17:11:01.278006: step: 1426/1541, loss: 0.0010808876249939203 2023-01-23 17:11:02.189039: step: 1428/1541, loss: 0.0005942240823060274 2023-01-23 17:11:03.141127: step: 1430/1541, loss: 0.06633415818214417 2023-01-23 17:11:04.137648: step: 1432/1541, loss: 0.27178484201431274 2023-01-23 17:11:05.039711: step: 1434/1541, loss: 0.34889349341392517 2023-01-23 17:11:06.000481: step: 1436/1541, loss: 0.0038826086092740297 2023-01-23 17:11:06.868189: step: 1438/1541, loss: 0.05882398039102554 2023-01-23 17:11:07.748014: step: 1440/1541, loss: 0.13312512636184692 2023-01-23 17:11:08.635949: step: 1442/1541, loss: 0.09344597905874252 2023-01-23 17:11:09.523939: step: 1444/1541, loss: 0.05149148032069206 2023-01-23 17:11:10.402788: step: 1446/1541, loss: 0.0038467475678771734 2023-01-23 17:11:11.289081: step: 1448/1541, loss: 0.008538097143173218 2023-01-23 17:11:12.182883: step: 1450/1541, loss: 0.011230843141674995 2023-01-23 17:11:13.066598: step: 1452/1541, loss: 0.012572279199957848 2023-01-23 17:11:13.944079: step: 1454/1541, loss: 0.008687619119882584 2023-01-23 17:11:14.892074: step: 1456/1541, loss: 0.2330601066350937 2023-01-23 17:11:15.762089: step: 1458/1541, loss: 0.0521264411509037 2023-01-23 17:11:16.715646: step: 1460/1541, loss: 0.0007517733029089868 2023-01-23 17:11:17.592301: step: 1462/1541, loss: 0.005630109459161758 2023-01-23 17:11:18.517273: step: 1464/1541, loss: 0.0012201901990920305 2023-01-23 17:11:19.534968: step: 1466/1541, loss: 0.08881684392690659 2023-01-23 17:11:20.428832: step: 1468/1541, loss: 0.19601453840732574 2023-01-23 17:11:21.323671: step: 1470/1541, loss: 0.011212234385311604 2023-01-23 17:11:22.254707: step: 1472/1541, loss: 0.014004445634782314 2023-01-23 17:11:23.142590: step: 1474/1541, loss: 0.05010060966014862 2023-01-23 17:11:24.047858: step: 1476/1541, loss: 0.054339416325092316 2023-01-23 17:11:24.944919: step: 1478/1541, loss: 0.11025850474834442 2023-01-23 17:11:25.889474: step: 1480/1541, loss: 0.1092166006565094 2023-01-23 17:11:26.779582: step: 1482/1541, loss: 0.0078615453094244 2023-01-23 17:11:27.674527: step: 1484/1541, loss: 0.19460417330265045 2023-01-23 17:11:28.606966: step: 1486/1541, loss: 0.0016876960871741176 2023-01-23 17:11:29.556630: step: 1488/1541, loss: 0.0007748905336484313 2023-01-23 17:11:30.438270: step: 1490/1541, loss: 0.07418835163116455 2023-01-23 17:11:31.374790: step: 1492/1541, loss: 0.0018133493140339851 2023-01-23 17:11:32.248635: step: 1494/1541, loss: 0.004738382063806057 2023-01-23 17:11:33.191965: step: 1496/1541, loss: 0.010378179140388966 2023-01-23 17:11:34.117021: step: 1498/1541, loss: 0.006413000635802746 2023-01-23 17:11:34.978364: step: 1500/1541, loss: 0.07275707274675369 2023-01-23 17:11:35.871724: step: 1502/1541, loss: 0.03991598263382912 2023-01-23 17:11:36.768057: step: 1504/1541, loss: 0.008610174991190434 2023-01-23 17:11:37.684174: step: 1506/1541, loss: 0.21797466278076172 2023-01-23 17:11:38.549615: step: 1508/1541, loss: 0.01941707916557789 2023-01-23 17:11:39.457644: step: 1510/1541, loss: 0.020531976595520973 2023-01-23 17:11:40.378451: step: 1512/1541, loss: 0.0009837312391027808 2023-01-23 17:11:41.254265: step: 1514/1541, loss: 0.13877147436141968 2023-01-23 17:11:42.170565: step: 1516/1541, loss: 0.0008572214865125716 2023-01-23 17:11:43.081420: step: 1518/1541, loss: 0.0011401241645216942 2023-01-23 17:11:43.965155: step: 1520/1541, loss: 0.01062605157494545 2023-01-23 17:11:44.835223: step: 1522/1541, loss: 0.020981628447771072 2023-01-23 17:11:45.730368: step: 1524/1541, loss: 0.00012469783541746438 2023-01-23 17:11:46.660934: step: 1526/1541, loss: 0.0016209816094487906 2023-01-23 17:11:47.598983: step: 1528/1541, loss: 0.4435597062110901 2023-01-23 17:11:48.482799: step: 1530/1541, loss: 0.0010710557689890265 2023-01-23 17:11:49.407230: step: 1532/1541, loss: 0.0008474751957692206 2023-01-23 17:11:50.299928: step: 1534/1541, loss: 0.0016363967442885041 2023-01-23 17:11:51.217673: step: 1536/1541, loss: 0.0005260683828964829 2023-01-23 17:11:52.117487: step: 1538/1541, loss: 0.00022357591660693288 2023-01-23 17:11:53.072859: step: 1540/1541, loss: 0.00042535661486908793 2023-01-23 17:11:53.956685: step: 1542/1541, loss: 0.1674632877111435 2023-01-23 17:11:54.856665: step: 1544/1541, loss: 0.0009006599429994822 2023-01-23 17:11:55.735043: step: 1546/1541, loss: 0.0014877335634082556 2023-01-23 17:11:56.638951: step: 1548/1541, loss: 0.0015699754003435373 2023-01-23 17:11:57.546576: step: 1550/1541, loss: 0.01638798788189888 2023-01-23 17:11:58.441498: step: 1552/1541, loss: 0.0178244449198246 2023-01-23 17:11:59.345668: step: 1554/1541, loss: 0.1354156881570816 2023-01-23 17:12:00.250000: step: 1556/1541, loss: 0.01500389352440834 2023-01-23 17:12:01.126293: step: 1558/1541, loss: 0.15128856897354126 2023-01-23 17:12:02.046527: step: 1560/1541, loss: 0.17711900174617767 2023-01-23 17:12:02.927508: step: 1562/1541, loss: 0.00014600470603909343 2023-01-23 17:12:03.839675: step: 1564/1541, loss: 0.03879139572381973 2023-01-23 17:12:04.739672: step: 1566/1541, loss: 0.00062275753589347 2023-01-23 17:12:05.684955: step: 1568/1541, loss: 0.004331247881054878 2023-01-23 17:12:06.566093: step: 1570/1541, loss: 0.011226052418351173 2023-01-23 17:12:07.554859: step: 1572/1541, loss: 0.3033248484134674 2023-01-23 17:12:08.497732: step: 1574/1541, loss: 2.070990562438965 2023-01-23 17:12:09.356061: step: 1576/1541, loss: 0.03320540487766266 2023-01-23 17:12:10.302066: step: 1578/1541, loss: 0.0014611524529755116 2023-01-23 17:12:11.199549: step: 1580/1541, loss: 0.025169624015688896 2023-01-23 17:12:12.113163: step: 1582/1541, loss: 0.001736232079565525 2023-01-23 17:12:13.024452: step: 1584/1541, loss: 0.04007308930158615 2023-01-23 17:12:13.908389: step: 1586/1541, loss: 0.005748586729168892 2023-01-23 17:12:14.795254: step: 1588/1541, loss: 0.0009352442575618625 2023-01-23 17:12:15.691962: step: 1590/1541, loss: 0.06318832188844681 2023-01-23 17:12:16.559832: step: 1592/1541, loss: 0.05787975341081619 2023-01-23 17:12:17.460517: step: 1594/1541, loss: 0.006951896473765373 2023-01-23 17:12:18.325960: step: 1596/1541, loss: 0.04128723964095116 2023-01-23 17:12:19.240118: step: 1598/1541, loss: 0.01924068294465542 2023-01-23 17:12:20.112681: step: 1600/1541, loss: 0.00033865825389511883 2023-01-23 17:12:20.972237: step: 1602/1541, loss: 0.11916303634643555 2023-01-23 17:12:21.858713: step: 1604/1541, loss: 0.0952364057302475 2023-01-23 17:12:22.770155: step: 1606/1541, loss: 0.009037462994456291 2023-01-23 17:12:23.692359: step: 1608/1541, loss: 0.03526301309466362 2023-01-23 17:12:24.556161: step: 1610/1541, loss: 0.12028728425502777 2023-01-23 17:12:25.542970: step: 1612/1541, loss: 0.017144903540611267 2023-01-23 17:12:26.432978: step: 1614/1541, loss: 0.07159308344125748 2023-01-23 17:12:27.298896: step: 1616/1541, loss: 0.0009295867057517171 2023-01-23 17:12:28.165473: step: 1618/1541, loss: 0.006710881367325783 2023-01-23 17:12:29.061736: step: 1620/1541, loss: 0.055688973516225815 2023-01-23 17:12:29.944036: step: 1622/1541, loss: 0.023697013035416603 2023-01-23 17:12:30.826725: step: 1624/1541, loss: 0.025815699249505997 2023-01-23 17:12:31.741232: step: 1626/1541, loss: 0.001992968376725912 2023-01-23 17:12:32.668065: step: 1628/1541, loss: 0.2195441573858261 2023-01-23 17:12:33.498024: step: 1630/1541, loss: 0.05068429186940193 2023-01-23 17:12:34.481019: step: 1632/1541, loss: 0.0012045636540278792 2023-01-23 17:12:35.374826: step: 1634/1541, loss: 0.08549770712852478 2023-01-23 17:12:36.285866: step: 1636/1541, loss: 0.15165936946868896 2023-01-23 17:12:37.204815: step: 1638/1541, loss: 0.005693231709301472 2023-01-23 17:12:38.106043: step: 1640/1541, loss: 0.009075944311916828 2023-01-23 17:12:38.999994: step: 1642/1541, loss: 0.000412640132708475 2023-01-23 17:12:39.872645: step: 1644/1541, loss: 0.000705409562215209 2023-01-23 17:12:40.762624: step: 1646/1541, loss: 0.0936199352145195 2023-01-23 17:12:41.635474: step: 1648/1541, loss: 0.001615302637219429 2023-01-23 17:12:42.561518: step: 1650/1541, loss: 0.04082091897726059 2023-01-23 17:12:43.405717: step: 1652/1541, loss: 0.00018613692373037338 2023-01-23 17:12:44.330964: step: 1654/1541, loss: 0.07573897391557693 2023-01-23 17:12:45.236906: step: 1656/1541, loss: 0.7246133089065552 2023-01-23 17:12:46.254573: step: 1658/1541, loss: 0.6795738935470581 2023-01-23 17:12:47.098323: step: 1660/1541, loss: 0.010869761928915977 2023-01-23 17:12:48.021235: step: 1662/1541, loss: 0.5017357468605042 2023-01-23 17:12:48.898335: step: 1664/1541, loss: 0.04831044748425484 2023-01-23 17:12:49.810323: step: 1666/1541, loss: 0.5421145558357239 2023-01-23 17:12:50.685359: step: 1668/1541, loss: 0.014517314732074738 2023-01-23 17:12:51.578820: step: 1670/1541, loss: 0.2519895136356354 2023-01-23 17:12:52.434913: step: 1672/1541, loss: 0.0005972236394882202 2023-01-23 17:12:53.338703: step: 1674/1541, loss: 0.11251227557659149 2023-01-23 17:12:54.277221: step: 1676/1541, loss: 0.0012927221832796931 2023-01-23 17:12:55.174202: step: 1678/1541, loss: 0.06895953416824341 2023-01-23 17:12:56.069464: step: 1680/1541, loss: 0.0016640903195366263 2023-01-23 17:12:56.955460: step: 1682/1541, loss: 0.003156038699671626 2023-01-23 17:12:57.901656: step: 1684/1541, loss: 0.00483715021982789 2023-01-23 17:12:58.750837: step: 1686/1541, loss: 0.1681271195411682 2023-01-23 17:12:59.662693: step: 1688/1541, loss: 0.22104579210281372 2023-01-23 17:13:00.581821: step: 1690/1541, loss: 0.001129703363403678 2023-01-23 17:13:01.455073: step: 1692/1541, loss: 0.010114110074937344 2023-01-23 17:13:02.397686: step: 1694/1541, loss: 0.006282222457230091 2023-01-23 17:13:03.268511: step: 1696/1541, loss: 0.8071218132972717 2023-01-23 17:13:04.130389: step: 1698/1541, loss: 0.022244960069656372 2023-01-23 17:13:05.017519: step: 1700/1541, loss: 0.0057671209797263145 2023-01-23 17:13:05.893342: step: 1702/1541, loss: 0.20737190544605255 2023-01-23 17:13:06.773696: step: 1704/1541, loss: 0.020357822999358177 2023-01-23 17:13:07.686788: step: 1706/1541, loss: 0.0012091828975826502 2023-01-23 17:13:08.554311: step: 1708/1541, loss: 0.00298026273958385 2023-01-23 17:13:09.406668: step: 1710/1541, loss: 0.027968594804406166 2023-01-23 17:13:10.312769: step: 1712/1541, loss: 0.28867992758750916 2023-01-23 17:13:11.246592: step: 1714/1541, loss: 0.09892234206199646 2023-01-23 17:13:12.143885: step: 1716/1541, loss: 0.016414334997534752 2023-01-23 17:13:13.025557: step: 1718/1541, loss: 0.1700042486190796 2023-01-23 17:13:13.922980: step: 1720/1541, loss: 0.0008849324658513069 2023-01-23 17:13:14.832430: step: 1722/1541, loss: 0.03324247524142265 2023-01-23 17:13:15.689072: step: 1724/1541, loss: 0.0007794490084052086 2023-01-23 17:13:16.561791: step: 1726/1541, loss: 0.02817932702600956 2023-01-23 17:13:17.450744: step: 1728/1541, loss: 0.05802442133426666 2023-01-23 17:13:18.319097: step: 1730/1541, loss: 0.0014592137886211276 2023-01-23 17:13:19.202774: step: 1732/1541, loss: 0.006372534204274416 2023-01-23 17:13:20.091838: step: 1734/1541, loss: 0.03195913881063461 2023-01-23 17:13:21.000197: step: 1736/1541, loss: 0.013966117985546589 2023-01-23 17:13:21.905746: step: 1738/1541, loss: 0.22375856339931488 2023-01-23 17:13:22.843792: step: 1740/1541, loss: 0.0004031491407658905 2023-01-23 17:13:23.755535: step: 1742/1541, loss: 0.05559016764163971 2023-01-23 17:13:24.631413: step: 1744/1541, loss: 0.005461955443024635 2023-01-23 17:13:25.546062: step: 1746/1541, loss: 0.11168690025806427 2023-01-23 17:13:26.400733: step: 1748/1541, loss: 0.02574058063328266 2023-01-23 17:13:27.277004: step: 1750/1541, loss: 0.0012682807864621282 2023-01-23 17:13:28.185787: step: 1752/1541, loss: 0.0003480295417830348 2023-01-23 17:13:29.082025: step: 1754/1541, loss: 0.374481737613678 2023-01-23 17:13:29.953138: step: 1756/1541, loss: 0.1757606416940689 2023-01-23 17:13:30.828371: step: 1758/1541, loss: 0.005108896177262068 2023-01-23 17:13:31.744267: step: 1760/1541, loss: 0.03650178760290146 2023-01-23 17:13:32.643823: step: 1762/1541, loss: 0.0012097132857888937 2023-01-23 17:13:33.565270: step: 1764/1541, loss: 0.0015012016519904137 2023-01-23 17:13:34.492674: step: 1766/1541, loss: 0.1649026870727539 2023-01-23 17:13:35.419806: step: 1768/1541, loss: 0.034368112683296204 2023-01-23 17:13:36.340457: step: 1770/1541, loss: 0.008962197229266167 2023-01-23 17:13:37.235631: step: 1772/1541, loss: 0.00035634663072414696 2023-01-23 17:13:38.113528: step: 1774/1541, loss: 0.0008282577618956566 2023-01-23 17:13:38.980367: step: 1776/1541, loss: 0.0001585787977091968 2023-01-23 17:13:39.845822: step: 1778/1541, loss: 0.0018669291166588664 2023-01-23 17:13:40.692718: step: 1780/1541, loss: 0.0006216445472091436 2023-01-23 17:13:41.588424: step: 1782/1541, loss: 0.1291273534297943 2023-01-23 17:13:42.492912: step: 1784/1541, loss: 0.010266386903822422 2023-01-23 17:13:43.446125: step: 1786/1541, loss: 0.26524612307548523 2023-01-23 17:13:44.404280: step: 1788/1541, loss: 0.014675314538180828 2023-01-23 17:13:45.294796: step: 1790/1541, loss: 2.888269955292344e-05 2023-01-23 17:13:46.193260: step: 1792/1541, loss: 0.0006758318631909788 2023-01-23 17:13:47.074135: step: 1794/1541, loss: 0.005692522972822189 2023-01-23 17:13:47.951857: step: 1796/1541, loss: 0.027106845751404762 2023-01-23 17:13:48.830232: step: 1798/1541, loss: 0.019849443808197975 2023-01-23 17:13:49.723793: step: 1800/1541, loss: 0.2164902538061142 2023-01-23 17:13:50.609610: step: 1802/1541, loss: 0.6170259714126587 2023-01-23 17:13:51.536038: step: 1804/1541, loss: 0.0014711596304550767 2023-01-23 17:13:52.416269: step: 1806/1541, loss: 0.10018528252840042 2023-01-23 17:13:53.299095: step: 1808/1541, loss: 0.2747137248516083 2023-01-23 17:13:54.193055: step: 1810/1541, loss: 0.00026759912725538015 2023-01-23 17:13:55.143626: step: 1812/1541, loss: 0.014800718054175377 2023-01-23 17:13:56.017920: step: 1814/1541, loss: 0.05681832134723663 2023-01-23 17:13:56.964719: step: 1816/1541, loss: 0.359090119600296 2023-01-23 17:13:57.852638: step: 1818/1541, loss: 0.07053482532501221 2023-01-23 17:13:58.750510: step: 1820/1541, loss: 0.0012941320892423391 2023-01-23 17:13:59.694566: step: 1822/1541, loss: 0.09796801954507828 2023-01-23 17:14:00.559625: step: 1824/1541, loss: 0.0006385073065757751 2023-01-23 17:14:01.453558: step: 1826/1541, loss: 0.015127654187381268 2023-01-23 17:14:02.397273: step: 1828/1541, loss: 0.02428949810564518 2023-01-23 17:14:03.394610: step: 1830/1541, loss: 0.0664471983909607 2023-01-23 17:14:04.292404: step: 1832/1541, loss: 0.00040525232907384634 2023-01-23 17:14:05.177675: step: 1834/1541, loss: 0.0003251030866522342 2023-01-23 17:14:06.080859: step: 1836/1541, loss: 0.001976588973775506 2023-01-23 17:14:06.970263: step: 1838/1541, loss: 0.009443117305636406 2023-01-23 17:14:07.869088: step: 1840/1541, loss: 0.0006550695397891104 2023-01-23 17:14:08.781435: step: 1842/1541, loss: 0.005376857239753008 2023-01-23 17:14:09.657423: step: 1844/1541, loss: 0.06711948662996292 2023-01-23 17:14:10.556765: step: 1846/1541, loss: 0.02115531824529171 2023-01-23 17:14:11.445606: step: 1848/1541, loss: 0.024091584607958794 2023-01-23 17:14:12.352763: step: 1850/1541, loss: 0.003439268795773387 2023-01-23 17:14:13.258702: step: 1852/1541, loss: 0.0012914893450215459 2023-01-23 17:14:14.138928: step: 1854/1541, loss: 0.12855583429336548 2023-01-23 17:14:15.040233: step: 1856/1541, loss: 1.1943580830120482e-05 2023-01-23 17:14:15.908414: step: 1858/1541, loss: 0.02364223450422287 2023-01-23 17:14:16.767585: step: 1860/1541, loss: 0.007570660673081875 2023-01-23 17:14:17.715605: step: 1862/1541, loss: 0.029480479657649994 2023-01-23 17:14:18.595113: step: 1864/1541, loss: 0.0027115570846945047 2023-01-23 17:14:19.521556: step: 1866/1541, loss: 0.0020598263945430517 2023-01-23 17:14:20.423182: step: 1868/1541, loss: 0.053588174283504486 2023-01-23 17:14:21.297996: step: 1870/1541, loss: 0.02260213904082775 2023-01-23 17:14:22.184449: step: 1872/1541, loss: 0.10457953065633774 2023-01-23 17:14:23.057339: step: 1874/1541, loss: 0.03945840895175934 2023-01-23 17:14:23.920958: step: 1876/1541, loss: 0.000264670088654384 2023-01-23 17:14:24.827163: step: 1878/1541, loss: 0.00649844016879797 2023-01-23 17:14:25.732877: step: 1880/1541, loss: 0.004755969159305096 2023-01-23 17:14:26.630298: step: 1882/1541, loss: 0.0001291240332648158 2023-01-23 17:14:27.552150: step: 1884/1541, loss: 0.02508533000946045 2023-01-23 17:14:28.445096: step: 1886/1541, loss: 0.00011563111183932051 2023-01-23 17:14:29.419629: step: 1888/1541, loss: 0.2215261310338974 2023-01-23 17:14:30.314422: step: 1890/1541, loss: 0.661745548248291 2023-01-23 17:14:31.219926: step: 1892/1541, loss: 0.074222631752491 2023-01-23 17:14:32.090862: step: 1894/1541, loss: 0.08984450250864029 2023-01-23 17:14:33.012901: step: 1896/1541, loss: 0.11107213795185089 2023-01-23 17:14:33.914406: step: 1898/1541, loss: 0.08020297437906265 2023-01-23 17:14:34.763748: step: 1900/1541, loss: 0.055903058499097824 2023-01-23 17:14:35.677298: step: 1902/1541, loss: 0.44041866064071655 2023-01-23 17:14:36.568255: step: 1904/1541, loss: 0.0043486133217811584 2023-01-23 17:14:37.450052: step: 1906/1541, loss: 0.0048314244486391544 2023-01-23 17:14:38.393011: step: 1908/1541, loss: 0.0007791031966917217 2023-01-23 17:14:39.307433: step: 1910/1541, loss: 0.038562506437301636 2023-01-23 17:14:40.207612: step: 1912/1541, loss: 0.03429247811436653 2023-01-23 17:14:41.154366: step: 1914/1541, loss: 0.0022846590727567673 2023-01-23 17:14:42.027700: step: 1916/1541, loss: 0.0020970518235117197 2023-01-23 17:14:42.914164: step: 1918/1541, loss: 0.07805391401052475 2023-01-23 17:14:43.845176: step: 1920/1541, loss: 0.012916826643049717 2023-01-23 17:14:44.725632: step: 1922/1541, loss: 0.03565385192632675 2023-01-23 17:14:45.631814: step: 1924/1541, loss: 0.00028842396568506956 2023-01-23 17:14:46.499029: step: 1926/1541, loss: 0.021242311224341393 2023-01-23 17:14:47.376161: step: 1928/1541, loss: 0.013867711648344994 2023-01-23 17:14:48.261666: step: 1930/1541, loss: 0.005922094918787479 2023-01-23 17:14:49.227835: step: 1932/1541, loss: 0.0025735758244991302 2023-01-23 17:14:50.158791: step: 1934/1541, loss: 0.020709794014692307 2023-01-23 17:14:51.052701: step: 1936/1541, loss: 0.03366429731249809 2023-01-23 17:14:51.997814: step: 1938/1541, loss: 0.9029256701469421 2023-01-23 17:14:52.864834: step: 1940/1541, loss: 0.06662473827600479 2023-01-23 17:14:53.747749: step: 1942/1541, loss: 0.0006365127628669143 2023-01-23 17:14:54.632346: step: 1944/1541, loss: 0.07052217423915863 2023-01-23 17:14:55.549387: step: 1946/1541, loss: 0.1752673238515854 2023-01-23 17:14:56.499816: step: 1948/1541, loss: 0.019300345331430435 2023-01-23 17:14:57.380804: step: 1950/1541, loss: 0.01621413603425026 2023-01-23 17:14:58.287793: step: 1952/1541, loss: 0.03623638302087784 2023-01-23 17:14:59.220001: step: 1954/1541, loss: 0.03735889121890068 2023-01-23 17:15:00.120321: step: 1956/1541, loss: 0.07997407764196396 2023-01-23 17:15:01.001833: step: 1958/1541, loss: 0.02977338433265686 2023-01-23 17:15:01.939356: step: 1960/1541, loss: 0.007108880206942558 2023-01-23 17:15:02.844945: step: 1962/1541, loss: 0.006970853079110384 2023-01-23 17:15:03.751129: step: 1964/1541, loss: 0.4966067969799042 2023-01-23 17:15:04.637538: step: 1966/1541, loss: 0.01806461252272129 2023-01-23 17:15:05.496145: step: 1968/1541, loss: 0.07415596395730972 2023-01-23 17:15:06.415191: step: 1970/1541, loss: 0.02623755857348442 2023-01-23 17:15:07.350539: step: 1972/1541, loss: 0.39110907912254333 2023-01-23 17:15:08.215481: step: 1974/1541, loss: 0.0016645763535052538 2023-01-23 17:15:09.144053: step: 1976/1541, loss: 0.03677257150411606 2023-01-23 17:15:10.008694: step: 1978/1541, loss: 0.0900641530752182 2023-01-23 17:15:10.894711: step: 1980/1541, loss: 0.04428214207291603 2023-01-23 17:15:11.821629: step: 1982/1541, loss: 0.004277598578482866 2023-01-23 17:15:12.749037: step: 1984/1541, loss: 0.08351384103298187 2023-01-23 17:15:13.646523: step: 1986/1541, loss: 0.00282940361648798 2023-01-23 17:15:14.547127: step: 1988/1541, loss: 0.09972167760133743 2023-01-23 17:15:15.421928: step: 1990/1541, loss: 0.10979456454515457 2023-01-23 17:15:16.342734: step: 1992/1541, loss: 0.02833193726837635 2023-01-23 17:15:17.206074: step: 1994/1541, loss: 0.29020705819129944 2023-01-23 17:15:18.089234: step: 1996/1541, loss: 0.006388519424945116 2023-01-23 17:15:18.988207: step: 1998/1541, loss: 7.400727918138728e-05 2023-01-23 17:15:19.889697: step: 2000/1541, loss: 0.0033788816072046757 2023-01-23 17:15:20.775172: step: 2002/1541, loss: 0.004023434594273567 2023-01-23 17:15:21.744481: step: 2004/1541, loss: 0.09517332166433334 2023-01-23 17:15:22.599825: step: 2006/1541, loss: 0.009918212890625 2023-01-23 17:15:23.508072: step: 2008/1541, loss: 0.007438980042934418 2023-01-23 17:15:24.410362: step: 2010/1541, loss: 0.0996701568365097 2023-01-23 17:15:25.279524: step: 2012/1541, loss: 0.002303118584677577 2023-01-23 17:15:26.140505: step: 2014/1541, loss: 0.00022669161262456328 2023-01-23 17:15:27.015467: step: 2016/1541, loss: 0.0037691991310566664 2023-01-23 17:15:27.891699: step: 2018/1541, loss: 0.007352788932621479 2023-01-23 17:15:28.798746: step: 2020/1541, loss: 0.015579781495034695 2023-01-23 17:15:29.659827: step: 2022/1541, loss: 0.054032109677791595 2023-01-23 17:15:30.564073: step: 2024/1541, loss: 0.001781249069608748 2023-01-23 17:15:31.501414: step: 2026/1541, loss: 0.007370717357844114 2023-01-23 17:15:32.419893: step: 2028/1541, loss: 0.002224629046395421 2023-01-23 17:15:33.282058: step: 2030/1541, loss: 0.06746536493301392 2023-01-23 17:15:34.205307: step: 2032/1541, loss: 7.730013749096543e-05 2023-01-23 17:15:35.095902: step: 2034/1541, loss: 0.014715137891471386 2023-01-23 17:15:36.065558: step: 2036/1541, loss: 0.1718999594449997 2023-01-23 17:15:36.970742: step: 2038/1541, loss: 0.1937200129032135 2023-01-23 17:15:37.837930: step: 2040/1541, loss: 0.05457797273993492 2023-01-23 17:15:38.774191: step: 2042/1541, loss: 0.05837633088231087 2023-01-23 17:15:39.717681: step: 2044/1541, loss: 0.37236377596855164 2023-01-23 17:15:40.599923: step: 2046/1541, loss: 0.13257594406604767 2023-01-23 17:15:41.452882: step: 2048/1541, loss: 0.001569946063682437 2023-01-23 17:15:42.349900: step: 2050/1541, loss: 0.04088861867785454 2023-01-23 17:15:43.202626: step: 2052/1541, loss: 0.0006057284190319479 2023-01-23 17:15:44.087213: step: 2054/1541, loss: 0.0004993146285414696 2023-01-23 17:15:44.994185: step: 2056/1541, loss: 0.05284916236996651 2023-01-23 17:15:45.891266: step: 2058/1541, loss: 9.5512303232681e-05 2023-01-23 17:15:46.752768: step: 2060/1541, loss: 0.12106527388095856 2023-01-23 17:15:47.649963: step: 2062/1541, loss: 0.013451723381876945 2023-01-23 17:15:48.589557: step: 2064/1541, loss: 0.2078215628862381 2023-01-23 17:15:49.439976: step: 2066/1541, loss: 4.002639252576046e-05 2023-01-23 17:15:50.404676: step: 2068/1541, loss: 0.03791225329041481 2023-01-23 17:15:51.307352: step: 2070/1541, loss: 0.00018329820886719972 2023-01-23 17:15:52.176559: step: 2072/1541, loss: 0.00036406650906428695 2023-01-23 17:15:53.123860: step: 2074/1541, loss: 0.04994313418865204 2023-01-23 17:15:54.091231: step: 2076/1541, loss: 0.0020723615307360888 2023-01-23 17:15:54.992718: step: 2078/1541, loss: 0.06457992643117905 2023-01-23 17:15:55.896358: step: 2080/1541, loss: 0.014951810240745544 2023-01-23 17:15:56.863166: step: 2082/1541, loss: 0.1290832906961441 2023-01-23 17:15:57.792519: step: 2084/1541, loss: 0.036391738802194595 2023-01-23 17:15:58.647799: step: 2086/1541, loss: 0.07358771562576294 2023-01-23 17:15:59.521097: step: 2088/1541, loss: 0.0004524586256593466 2023-01-23 17:16:00.399502: step: 2090/1541, loss: 0.10344724357128143 2023-01-23 17:16:01.307377: step: 2092/1541, loss: 0.024605171754956245 2023-01-23 17:16:02.228288: step: 2094/1541, loss: 0.007884055376052856 2023-01-23 17:16:03.113635: step: 2096/1541, loss: 0.0008132277871482074 2023-01-23 17:16:04.015407: step: 2098/1541, loss: 0.02734973467886448 2023-01-23 17:16:04.967692: step: 2100/1541, loss: 0.04502229392528534 2023-01-23 17:16:05.895397: step: 2102/1541, loss: 0.024061379954218864 2023-01-23 17:16:06.789187: step: 2104/1541, loss: 0.02906639687716961 2023-01-23 17:16:07.654581: step: 2106/1541, loss: 0.014458185993134975 2023-01-23 17:16:08.583314: step: 2108/1541, loss: 0.07738525420427322 2023-01-23 17:16:09.481411: step: 2110/1541, loss: 0.011076275259256363 2023-01-23 17:16:10.400592: step: 2112/1541, loss: 0.18804384768009186 2023-01-23 17:16:11.286278: step: 2114/1541, loss: 0.0039240578189492226 2023-01-23 17:16:12.197875: step: 2116/1541, loss: 0.01064714603126049 2023-01-23 17:16:13.074160: step: 2118/1541, loss: 0.006891186349093914 2023-01-23 17:16:13.961870: step: 2120/1541, loss: 0.0912867859005928 2023-01-23 17:16:14.909989: step: 2122/1541, loss: 0.005728662945330143 2023-01-23 17:16:15.814774: step: 2124/1541, loss: 0.001395129831507802 2023-01-23 17:16:16.718654: step: 2126/1541, loss: 0.010049646720290184 2023-01-23 17:16:17.637424: step: 2128/1541, loss: 0.0027845920994877815 2023-01-23 17:16:18.509091: step: 2130/1541, loss: 0.0006692283786833286 2023-01-23 17:16:19.415424: step: 2132/1541, loss: 0.01699705794453621 2023-01-23 17:16:20.287566: step: 2134/1541, loss: 0.02576233074069023 2023-01-23 17:16:21.231736: step: 2136/1541, loss: 0.005247904919087887 2023-01-23 17:16:22.094039: step: 2138/1541, loss: 0.06460873037576675 2023-01-23 17:16:22.983167: step: 2140/1541, loss: 0.0003957459994126111 2023-01-23 17:16:23.861425: step: 2142/1541, loss: 0.0008594175451435149 2023-01-23 17:16:24.732712: step: 2144/1541, loss: 0.00010248801845591515 2023-01-23 17:16:25.598408: step: 2146/1541, loss: 0.00987743679434061 2023-01-23 17:16:26.526502: step: 2148/1541, loss: 0.005570963025093079 2023-01-23 17:16:27.433235: step: 2150/1541, loss: 0.1717238575220108 2023-01-23 17:16:28.382123: step: 2152/1541, loss: 0.0038377062883228064 2023-01-23 17:16:29.290133: step: 2154/1541, loss: 0.00015163460921030492 2023-01-23 17:16:30.203003: step: 2156/1541, loss: 0.002440558047965169 2023-01-23 17:16:31.130022: step: 2158/1541, loss: 0.0038519264198839664 2023-01-23 17:16:32.018266: step: 2160/1541, loss: 0.022152304649353027 2023-01-23 17:16:32.912922: step: 2162/1541, loss: 0.013338680379092693 2023-01-23 17:16:33.774356: step: 2164/1541, loss: 0.0010029016993939877 2023-01-23 17:16:34.756472: step: 2166/1541, loss: 0.03939913213253021 2023-01-23 17:16:35.649525: step: 2168/1541, loss: 0.00012202389916637912 2023-01-23 17:16:36.549238: step: 2170/1541, loss: 0.00989721529185772 2023-01-23 17:16:37.493254: step: 2172/1541, loss: 0.022478152066469193 2023-01-23 17:16:38.371092: step: 2174/1541, loss: 0.0050196461379528046 2023-01-23 17:16:39.290053: step: 2176/1541, loss: 0.272283673286438 2023-01-23 17:16:40.178629: step: 2178/1541, loss: 0.02170180156826973 2023-01-23 17:16:41.021111: step: 2180/1541, loss: 0.0239033792167902 2023-01-23 17:16:41.923823: step: 2182/1541, loss: 0.00032181452843360603 2023-01-23 17:16:42.829470: step: 2184/1541, loss: 0.032331038266420364 2023-01-23 17:16:43.723608: step: 2186/1541, loss: 0.1205039992928505 2023-01-23 17:16:44.630225: step: 2188/1541, loss: 0.04875946789979935 2023-01-23 17:16:45.531879: step: 2190/1541, loss: 0.016902398318052292 2023-01-23 17:16:46.415844: step: 2192/1541, loss: 0.00020966878219041973 2023-01-23 17:16:47.321060: step: 2194/1541, loss: 6.673510506516322e-05 2023-01-23 17:16:48.209222: step: 2196/1541, loss: 0.14563293755054474 2023-01-23 17:16:49.088135: step: 2198/1541, loss: 0.002496576402336359 2023-01-23 17:16:50.063305: step: 2200/1541, loss: 0.00019059391343034804 2023-01-23 17:16:50.959554: step: 2202/1541, loss: 0.008992386981844902 2023-01-23 17:16:51.829697: step: 2204/1541, loss: 0.0033291163854300976 2023-01-23 17:16:52.728023: step: 2206/1541, loss: 0.0003133227291982621 2023-01-23 17:16:53.614619: step: 2208/1541, loss: 0.00570899760350585 2023-01-23 17:16:54.516732: step: 2210/1541, loss: 0.00021716015180572867 2023-01-23 17:16:55.422863: step: 2212/1541, loss: 9.167832467937842e-05 2023-01-23 17:16:56.275427: step: 2214/1541, loss: 0.12092026323080063 2023-01-23 17:16:57.169697: step: 2216/1541, loss: 0.0003739129169844091 2023-01-23 17:16:58.052159: step: 2218/1541, loss: 0.1343585103750229 2023-01-23 17:16:58.931244: step: 2220/1541, loss: 0.00019395086565054953 2023-01-23 17:16:59.818442: step: 2222/1541, loss: 0.0009750880999490619 2023-01-23 17:17:00.687870: step: 2224/1541, loss: 6.040681910235435e-05 2023-01-23 17:17:01.569516: step: 2226/1541, loss: 0.00046432478120550513 2023-01-23 17:17:02.482742: step: 2228/1541, loss: 0.15904544293880463 2023-01-23 17:17:03.357831: step: 2230/1541, loss: 0.0005521088023670018 2023-01-23 17:17:04.276215: step: 2232/1541, loss: 0.1919008046388626 2023-01-23 17:17:05.163507: step: 2234/1541, loss: 0.020835325121879578 2023-01-23 17:17:06.088870: step: 2236/1541, loss: 0.015720069408416748 2023-01-23 17:17:07.004474: step: 2238/1541, loss: 0.4722934663295746 2023-01-23 17:17:07.877560: step: 2240/1541, loss: 0.0005176828126423061 2023-01-23 17:17:08.804054: step: 2242/1541, loss: 3.4166202545166016 2023-01-23 17:17:09.775993: step: 2244/1541, loss: 0.0011889992747455835 2023-01-23 17:17:10.659560: step: 2246/1541, loss: 0.05698032304644585 2023-01-23 17:17:11.548957: step: 2248/1541, loss: 0.24992798268795013 2023-01-23 17:17:12.518398: step: 2250/1541, loss: 0.028490690514445305 2023-01-23 17:17:13.426463: step: 2252/1541, loss: 0.00019222394621465355 2023-01-23 17:17:14.303123: step: 2254/1541, loss: 0.022860605269670486 2023-01-23 17:17:15.190208: step: 2256/1541, loss: 0.2863273024559021 2023-01-23 17:17:16.095842: step: 2258/1541, loss: 0.4436624050140381 2023-01-23 17:17:16.997784: step: 2260/1541, loss: 0.06292691826820374 2023-01-23 17:17:17.909515: step: 2262/1541, loss: 0.004587073344737291 2023-01-23 17:17:18.753647: step: 2264/1541, loss: 0.0005697644082829356 2023-01-23 17:17:19.644459: step: 2266/1541, loss: 0.06126859039068222 2023-01-23 17:17:20.534382: step: 2268/1541, loss: 0.024363156408071518 2023-01-23 17:17:21.476121: step: 2270/1541, loss: 2.646344728418626e-05 2023-01-23 17:17:22.430919: step: 2272/1541, loss: 0.00263908295892179 2023-01-23 17:17:23.308788: step: 2274/1541, loss: 0.035172708332538605 2023-01-23 17:17:24.202025: step: 2276/1541, loss: 0.08083495497703552 2023-01-23 17:17:25.078261: step: 2278/1541, loss: 0.04188980162143707 2023-01-23 17:17:26.005027: step: 2280/1541, loss: 0.0008037814404815435 2023-01-23 17:17:26.904193: step: 2282/1541, loss: 0.6609233617782593 2023-01-23 17:17:27.863321: step: 2284/1541, loss: 0.2862927317619324 2023-01-23 17:17:28.712494: step: 2286/1541, loss: 0.00024108114303089678 2023-01-23 17:17:29.547935: step: 2288/1541, loss: 0.03704702481627464 2023-01-23 17:17:30.411486: step: 2290/1541, loss: 0.01213194988667965 2023-01-23 17:17:31.279901: step: 2292/1541, loss: 0.0001705551112536341 2023-01-23 17:17:32.242914: step: 2294/1541, loss: 0.0099838487803936 2023-01-23 17:17:33.095246: step: 2296/1541, loss: 0.00022284567239694297 2023-01-23 17:17:33.973319: step: 2298/1541, loss: 0.0010610909666866064 2023-01-23 17:17:34.912012: step: 2300/1541, loss: 0.02424733340740204 2023-01-23 17:17:35.856359: step: 2302/1541, loss: 0.0011495104990899563 2023-01-23 17:17:36.725857: step: 2304/1541, loss: 0.005758173298090696 2023-01-23 17:17:37.595252: step: 2306/1541, loss: 0.0004616093647200614 2023-01-23 17:17:38.486777: step: 2308/1541, loss: 0.0027346336282789707 2023-01-23 17:17:39.424915: step: 2310/1541, loss: 0.005113519728183746 2023-01-23 17:17:40.304059: step: 2312/1541, loss: 0.0019319128477945924 2023-01-23 17:17:41.158969: step: 2314/1541, loss: 0.000183097057742998 2023-01-23 17:17:42.083420: step: 2316/1541, loss: 0.0013544079847633839 2023-01-23 17:17:42.960821: step: 2318/1541, loss: 0.002579697407782078 2023-01-23 17:17:43.880378: step: 2320/1541, loss: 0.03788481280207634 2023-01-23 17:17:44.749153: step: 2322/1541, loss: 0.12901881337165833 2023-01-23 17:17:45.634987: step: 2324/1541, loss: 0.005918480921536684 2023-01-23 17:17:46.481869: step: 2326/1541, loss: 0.061088405549526215 2023-01-23 17:17:47.443677: step: 2328/1541, loss: 0.02775581367313862 2023-01-23 17:17:48.373398: step: 2330/1541, loss: 0.00976890791207552 2023-01-23 17:17:49.275221: step: 2332/1541, loss: 0.0026595252566039562 2023-01-23 17:17:50.215336: step: 2334/1541, loss: 8.301766501972452e-05 2023-01-23 17:17:51.063341: step: 2336/1541, loss: 0.013690837658941746 2023-01-23 17:17:51.979896: step: 2338/1541, loss: 0.0016903224168345332 2023-01-23 17:17:52.889086: step: 2340/1541, loss: 0.04419058933854103 2023-01-23 17:17:53.791055: step: 2342/1541, loss: 0.04112616181373596 2023-01-23 17:17:54.669980: step: 2344/1541, loss: 0.27699631452560425 2023-01-23 17:17:55.558010: step: 2346/1541, loss: 0.027552394196391106 2023-01-23 17:17:56.440492: step: 2348/1541, loss: 0.16156348586082458 2023-01-23 17:17:57.383983: step: 2350/1541, loss: 0.1640416383743286 2023-01-23 17:17:58.249679: step: 2352/1541, loss: 0.00837174616754055 2023-01-23 17:17:59.103781: step: 2354/1541, loss: 0.0012320957612246275 2023-01-23 17:18:00.024844: step: 2356/1541, loss: 0.003385790390893817 2023-01-23 17:18:00.962180: step: 2358/1541, loss: 0.012581313028931618 2023-01-23 17:18:01.892121: step: 2360/1541, loss: 0.061309345066547394 2023-01-23 17:18:02.784861: step: 2362/1541, loss: 0.013088997453451157 2023-01-23 17:18:03.711556: step: 2364/1541, loss: 0.0013289491180330515 2023-01-23 17:18:04.602833: step: 2366/1541, loss: 0.0077750347554683685 2023-01-23 17:18:05.512303: step: 2368/1541, loss: 0.00020902842516079545 2023-01-23 17:18:06.444102: step: 2370/1541, loss: 0.018928162753582 2023-01-23 17:18:07.321973: step: 2372/1541, loss: 0.006797782611101866 2023-01-23 17:18:08.180664: step: 2374/1541, loss: 0.0002682057092897594 2023-01-23 17:18:09.075868: step: 2376/1541, loss: 0.2812514901161194 2023-01-23 17:18:09.932131: step: 2378/1541, loss: 0.00015820955741219223 2023-01-23 17:18:10.839979: step: 2380/1541, loss: 0.3136056065559387 2023-01-23 17:18:11.727468: step: 2382/1541, loss: 0.207949697971344 2023-01-23 17:18:12.677436: step: 2384/1541, loss: 0.00028057408053427935 2023-01-23 17:18:13.624625: step: 2386/1541, loss: 0.00021627024398185313 2023-01-23 17:18:14.530128: step: 2388/1541, loss: 0.04384322091937065 2023-01-23 17:18:15.388131: step: 2390/1541, loss: 0.00987885519862175 2023-01-23 17:18:16.359820: step: 2392/1541, loss: 1.999000960495323e-05 2023-01-23 17:18:17.298396: step: 2394/1541, loss: 0.07928363233804703 2023-01-23 17:18:18.197558: step: 2396/1541, loss: 0.4077514111995697 2023-01-23 17:18:19.094022: step: 2398/1541, loss: 0.0001321585150435567 2023-01-23 17:18:19.991136: step: 2400/1541, loss: 0.8485201597213745 2023-01-23 17:18:20.856004: step: 2402/1541, loss: 0.05075416341423988 2023-01-23 17:18:21.781757: step: 2404/1541, loss: 0.01736791804432869 2023-01-23 17:18:22.671273: step: 2406/1541, loss: 0.0004465613455977291 2023-01-23 17:18:23.562184: step: 2408/1541, loss: 0.12008529901504517 2023-01-23 17:18:24.437037: step: 2410/1541, loss: 0.23470713198184967 2023-01-23 17:18:25.319022: step: 2412/1541, loss: 0.00151268660556525 2023-01-23 17:18:26.224539: step: 2414/1541, loss: 0.00038426349055953324 2023-01-23 17:18:27.094968: step: 2416/1541, loss: 0.011445102281868458 2023-01-23 17:18:27.971251: step: 2418/1541, loss: 0.002550610573962331 2023-01-23 17:18:28.837649: step: 2420/1541, loss: 0.08184938877820969 2023-01-23 17:18:29.714574: step: 2422/1541, loss: 0.03342067822813988 2023-01-23 17:18:30.673547: step: 2424/1541, loss: 0.000651631795335561 2023-01-23 17:18:31.684544: step: 2426/1541, loss: -1.0578324690868612e-05 2023-01-23 17:18:32.709912: step: 2428/1541, loss: 0.021167904138565063 2023-01-23 17:18:33.577259: step: 2430/1541, loss: 2.4441320420010015e-05 2023-01-23 17:18:34.526879: step: 2432/1541, loss: 0.009167256765067577 2023-01-23 17:18:35.452357: step: 2434/1541, loss: 0.025632590055465698 2023-01-23 17:18:36.332821: step: 2436/1541, loss: 0.049440257251262665 2023-01-23 17:18:37.215109: step: 2438/1541, loss: 1.8304991722106934 2023-01-23 17:18:38.097518: step: 2440/1541, loss: 0.2554200291633606 2023-01-23 17:18:38.971792: step: 2442/1541, loss: 0.21945345401763916 2023-01-23 17:18:39.923278: step: 2444/1541, loss: 0.009326118975877762 2023-01-23 17:18:40.800777: step: 2446/1541, loss: 0.20230582356452942 2023-01-23 17:18:41.676865: step: 2448/1541, loss: 0.015793126076459885 2023-01-23 17:18:42.531606: step: 2450/1541, loss: 0.009332071989774704 2023-01-23 17:18:43.416878: step: 2452/1541, loss: 0.0028582413215190172 2023-01-23 17:18:44.291030: step: 2454/1541, loss: 0.000632158131338656 2023-01-23 17:18:45.143114: step: 2456/1541, loss: 0.010780895128846169 2023-01-23 17:18:46.036032: step: 2458/1541, loss: 0.009399522095918655 2023-01-23 17:18:46.923519: step: 2460/1541, loss: 0.13370151817798615 2023-01-23 17:18:47.859759: step: 2462/1541, loss: 0.007957291789352894 2023-01-23 17:18:48.759905: step: 2464/1541, loss: 0.000396436924347654 2023-01-23 17:18:49.651439: step: 2466/1541, loss: 0.0011951987398788333 2023-01-23 17:18:50.528545: step: 2468/1541, loss: 0.39754748344421387 2023-01-23 17:18:51.408511: step: 2470/1541, loss: 1.3112705945968628 2023-01-23 17:18:52.294697: step: 2472/1541, loss: 0.0002653947740327567 2023-01-23 17:18:53.257044: step: 2474/1541, loss: 0.22513167560100555 2023-01-23 17:18:54.148506: step: 2476/1541, loss: 0.21311156451702118 2023-01-23 17:18:55.026847: step: 2478/1541, loss: 7.008969259914011e-05 2023-01-23 17:18:55.958462: step: 2480/1541, loss: 0.0007911816355772316 2023-01-23 17:18:56.868930: step: 2482/1541, loss: 0.029754744842648506 2023-01-23 17:18:57.755240: step: 2484/1541, loss: 0.021105201914906502 2023-01-23 17:18:58.667385: step: 2486/1541, loss: 0.4510086476802826 2023-01-23 17:18:59.566953: step: 2488/1541, loss: 0.004440469201654196 2023-01-23 17:19:00.459716: step: 2490/1541, loss: 0.0002826258132699877 2023-01-23 17:19:01.353949: step: 2492/1541, loss: 0.07057005912065506 2023-01-23 17:19:02.270185: step: 2494/1541, loss: 0.002017698250710964 2023-01-23 17:19:03.131309: step: 2496/1541, loss: 0.009805644862353802 2023-01-23 17:19:04.015124: step: 2498/1541, loss: 0.0040627229027450085 2023-01-23 17:19:04.933134: step: 2500/1541, loss: 0.07191310077905655 2023-01-23 17:19:05.852892: step: 2502/1541, loss: 0.0071645816788077354 2023-01-23 17:19:06.804396: step: 2504/1541, loss: 0.002601210493594408 2023-01-23 17:19:07.690080: step: 2506/1541, loss: 0.03633997216820717 2023-01-23 17:19:08.560773: step: 2508/1541, loss: 0.000611236144322902 2023-01-23 17:19:09.450417: step: 2510/1541, loss: 0.010963182896375656 2023-01-23 17:19:10.371440: step: 2512/1541, loss: 0.02632812038064003 2023-01-23 17:19:11.231611: step: 2514/1541, loss: 0.0017794170416891575 2023-01-23 17:19:12.138527: step: 2516/1541, loss: 0.014438482001423836 2023-01-23 17:19:13.064567: step: 2518/1541, loss: 0.0016532037407159805 2023-01-23 17:19:13.942098: step: 2520/1541, loss: 0.0036448233295232058 2023-01-23 17:19:14.810686: step: 2522/1541, loss: 0.02090594545006752 2023-01-23 17:19:15.703207: step: 2524/1541, loss: 0.28053298592567444 2023-01-23 17:19:16.574545: step: 2526/1541, loss: 0.3679579198360443 2023-01-23 17:19:17.424687: step: 2528/1541, loss: 0.002076260047033429 2023-01-23 17:19:18.313213: step: 2530/1541, loss: 0.28326287865638733 2023-01-23 17:19:19.189078: step: 2532/1541, loss: 0.22606384754180908 2023-01-23 17:19:20.044178: step: 2534/1541, loss: 0.16643542051315308 2023-01-23 17:19:20.956003: step: 2536/1541, loss: 0.0026473409961909056 2023-01-23 17:19:21.868870: step: 2538/1541, loss: 0.04348321259021759 2023-01-23 17:19:22.798143: step: 2540/1541, loss: 0.03886524960398674 2023-01-23 17:19:23.667597: step: 2542/1541, loss: 0.0046698059886693954 2023-01-23 17:19:24.550357: step: 2544/1541, loss: 0.0005193860270082951 2023-01-23 17:19:25.488564: step: 2546/1541, loss: 0.004268113523721695 2023-01-23 17:19:26.403314: step: 2548/1541, loss: 0.0001605615980224684 2023-01-23 17:19:27.258512: step: 2550/1541, loss: 0.0005281027988530695 2023-01-23 17:19:28.133386: step: 2552/1541, loss: 0.002562836278229952 2023-01-23 17:19:29.085016: step: 2554/1541, loss: 0.2269108146429062 2023-01-23 17:19:29.963715: step: 2556/1541, loss: 0.13997991383075714 2023-01-23 17:19:30.852504: step: 2558/1541, loss: 0.35764336585998535 2023-01-23 17:19:31.730611: step: 2560/1541, loss: 0.002211327664554119 2023-01-23 17:19:32.676074: step: 2562/1541, loss: 0.00018561114848125726 2023-01-23 17:19:33.588883: step: 2564/1541, loss: 0.003269774839282036 2023-01-23 17:19:34.490321: step: 2566/1541, loss: 0.0007138425717130303 2023-01-23 17:19:35.409185: step: 2568/1541, loss: 0.01721678115427494 2023-01-23 17:19:36.296029: step: 2570/1541, loss: 0.012297239154577255 2023-01-23 17:19:37.249338: step: 2572/1541, loss: 0.002638676203787327 2023-01-23 17:19:38.149074: step: 2574/1541, loss: 0.16876929998397827 2023-01-23 17:19:39.037590: step: 2576/1541, loss: 0.11389585584402084 2023-01-23 17:19:39.948005: step: 2578/1541, loss: 0.4326440393924713 2023-01-23 17:19:40.835580: step: 2580/1541, loss: 0.29954826831817627 2023-01-23 17:19:41.748092: step: 2582/1541, loss: 0.012812596745789051 2023-01-23 17:19:42.628422: step: 2584/1541, loss: 0.1340222805738449 2023-01-23 17:19:43.506451: step: 2586/1541, loss: 0.0030223883222788572 2023-01-23 17:19:44.355702: step: 2588/1541, loss: 0.05442260205745697 2023-01-23 17:19:45.301751: step: 2590/1541, loss: 0.004589094780385494 2023-01-23 17:19:46.195841: step: 2592/1541, loss: 0.0023913006298244 2023-01-23 17:19:47.188217: step: 2594/1541, loss: 0.002469735685735941 2023-01-23 17:19:48.123400: step: 2596/1541, loss: 0.13159748911857605 2023-01-23 17:19:49.056869: step: 2598/1541, loss: 0.31756100058555603 2023-01-23 17:19:49.980235: step: 2600/1541, loss: 0.027657704427838326 2023-01-23 17:19:50.909467: step: 2602/1541, loss: 0.30361172556877136 2023-01-23 17:19:51.792022: step: 2604/1541, loss: 0.08632156997919083 2023-01-23 17:19:52.717323: step: 2606/1541, loss: 0.001754791708663106 2023-01-23 17:19:53.578262: step: 2608/1541, loss: 0.0025410838425159454 2023-01-23 17:19:54.457321: step: 2610/1541, loss: 0.21379615366458893 2023-01-23 17:19:55.369838: step: 2612/1541, loss: 0.08194923400878906 2023-01-23 17:19:56.276087: step: 2614/1541, loss: 0.08862755447626114 2023-01-23 17:19:57.156199: step: 2616/1541, loss: 0.0018792635528370738 2023-01-23 17:19:58.031096: step: 2618/1541, loss: 0.020654235035181046 2023-01-23 17:19:58.905414: step: 2620/1541, loss: 0.10655627399682999 2023-01-23 17:19:59.801354: step: 2622/1541, loss: 0.05471744388341904 2023-01-23 17:20:00.713804: step: 2624/1541, loss: 0.05726008117198944 2023-01-23 17:20:01.601517: step: 2626/1541, loss: 0.004268140532076359 2023-01-23 17:20:02.459715: step: 2628/1541, loss: 0.004037933889776468 2023-01-23 17:20:03.349362: step: 2630/1541, loss: 0.061266571283340454 2023-01-23 17:20:04.254086: step: 2632/1541, loss: 0.12226961553096771 2023-01-23 17:20:05.187231: step: 2634/1541, loss: 0.0797082856297493 2023-01-23 17:20:06.114069: step: 2636/1541, loss: 0.037629928439855576 2023-01-23 17:20:07.004019: step: 2638/1541, loss: 0.002410070737823844 2023-01-23 17:20:07.888942: step: 2640/1541, loss: 0.008814530447125435 2023-01-23 17:20:08.784018: step: 2642/1541, loss: 0.13475963473320007 2023-01-23 17:20:09.700408: step: 2644/1541, loss: 0.4534603953361511 2023-01-23 17:20:10.625174: step: 2646/1541, loss: 0.05175337195396423 2023-01-23 17:20:11.469154: step: 2648/1541, loss: 0.002552845049649477 2023-01-23 17:20:12.345364: step: 2650/1541, loss: 0.03947041183710098 2023-01-23 17:20:13.386145: step: 2652/1541, loss: 0.008573574014008045 2023-01-23 17:20:14.281797: step: 2654/1541, loss: 0.0028187325224280357 2023-01-23 17:20:15.140880: step: 2656/1541, loss: 0.000698826159350574 2023-01-23 17:20:16.032542: step: 2658/1541, loss: 0.09182801842689514 2023-01-23 17:20:16.910971: step: 2660/1541, loss: 0.019271856173872948 2023-01-23 17:20:17.792194: step: 2662/1541, loss: 0.0002677026786841452 2023-01-23 17:20:18.703349: step: 2664/1541, loss: 0.007753545884042978 2023-01-23 17:20:19.587989: step: 2666/1541, loss: 0.0011253084521740675 2023-01-23 17:20:20.462474: step: 2668/1541, loss: 0.06960826367139816 2023-01-23 17:20:21.450951: step: 2670/1541, loss: 0.05205884575843811 2023-01-23 17:20:22.388248: step: 2672/1541, loss: 0.014530431479215622 2023-01-23 17:20:23.294719: step: 2674/1541, loss: 0.08622201532125473 2023-01-23 17:20:24.145174: step: 2676/1541, loss: 0.0002121502038789913 2023-01-23 17:20:25.157442: step: 2678/1541, loss: 0.001438242383301258 2023-01-23 17:20:26.042497: step: 2680/1541, loss: 0.013805960305035114 2023-01-23 17:20:26.892500: step: 2682/1541, loss: 6.169982225401327e-05 2023-01-23 17:20:27.819800: step: 2684/1541, loss: 0.06813132762908936 2023-01-23 17:20:28.747463: step: 2686/1541, loss: 0.007503743749111891 2023-01-23 17:20:29.637157: step: 2688/1541, loss: 0.00016575441986788064 2023-01-23 17:20:30.574292: step: 2690/1541, loss: 0.004612206481397152 2023-01-23 17:20:31.461254: step: 2692/1541, loss: 2.196689456468448e-05 2023-01-23 17:20:32.364136: step: 2694/1541, loss: 0.30049118399620056 2023-01-23 17:20:33.245443: step: 2696/1541, loss: 0.1422959566116333 2023-01-23 17:20:34.181001: step: 2698/1541, loss: 0.011267660185694695 2023-01-23 17:20:35.056870: step: 2700/1541, loss: 0.00026274382253177464 2023-01-23 17:20:35.971298: step: 2702/1541, loss: 0.04588615521788597 2023-01-23 17:20:36.847615: step: 2704/1541, loss: 0.052904509007930756 2023-01-23 17:20:37.795783: step: 2706/1541, loss: 0.005531533621251583 2023-01-23 17:20:38.692179: step: 2708/1541, loss: 0.003868132596835494 2023-01-23 17:20:39.639080: step: 2710/1541, loss: 0.00043159606866538525 2023-01-23 17:20:40.528283: step: 2712/1541, loss: 0.005419457796961069 2023-01-23 17:20:41.436809: step: 2714/1541, loss: 0.0019590093288570642 2023-01-23 17:20:42.371226: step: 2716/1541, loss: 0.011420940980315208 2023-01-23 17:20:43.296824: step: 2718/1541, loss: 0.026368124410510063 2023-01-23 17:20:44.162683: step: 2720/1541, loss: 0.0012977231526747346 2023-01-23 17:20:45.067345: step: 2722/1541, loss: 0.0003799052210524678 2023-01-23 17:20:45.970424: step: 2724/1541, loss: 0.18481528759002686 2023-01-23 17:20:46.899307: step: 2726/1541, loss: 0.001998331630602479 2023-01-23 17:20:47.780267: step: 2728/1541, loss: 0.000961343408562243 2023-01-23 17:20:48.809236: step: 2730/1541, loss: 0.19625578820705414 2023-01-23 17:20:49.675304: step: 2732/1541, loss: 0.0016829726519063115 2023-01-23 17:20:50.588162: step: 2734/1541, loss: 0.0004519314388744533 2023-01-23 17:20:51.539504: step: 2736/1541, loss: 0.059990204870700836 2023-01-23 17:20:52.412681: step: 2738/1541, loss: 0.0013570445589721203 2023-01-23 17:20:53.388853: step: 2740/1541, loss: 0.000717468443326652 2023-01-23 17:20:54.284529: step: 2742/1541, loss: 0.14153656363487244 2023-01-23 17:20:55.151396: step: 2744/1541, loss: 0.14470691978931427 2023-01-23 17:20:56.048987: step: 2746/1541, loss: 0.027201883494853973 2023-01-23 17:20:56.918518: step: 2748/1541, loss: 0.0009065382764674723 2023-01-23 17:20:57.826426: step: 2750/1541, loss: 0.16970257461071014 2023-01-23 17:20:58.786382: step: 2752/1541, loss: 0.0017589664785191417 2023-01-23 17:20:59.666994: step: 2754/1541, loss: 0.11314097046852112 2023-01-23 17:21:00.544161: step: 2756/1541, loss: 0.00045662588672712445 2023-01-23 17:21:01.454539: step: 2758/1541, loss: 0.009458557702600956 2023-01-23 17:21:02.357205: step: 2760/1541, loss: 0.025902878493070602 2023-01-23 17:21:03.257965: step: 2762/1541, loss: 0.0069469185546040535 2023-01-23 17:21:04.146207: step: 2764/1541, loss: 0.13932116329669952 2023-01-23 17:21:05.046653: step: 2766/1541, loss: 0.1051105484366417 2023-01-23 17:21:05.911365: step: 2768/1541, loss: 0.15726196765899658 2023-01-23 17:21:06.830938: step: 2770/1541, loss: 0.04445300251245499 2023-01-23 17:21:07.707400: step: 2772/1541, loss: 0.21655087172985077 2023-01-23 17:21:08.643994: step: 2774/1541, loss: 0.10606545209884644 2023-01-23 17:21:09.539857: step: 2776/1541, loss: 0.0010652242926880717 2023-01-23 17:21:10.403805: step: 2778/1541, loss: 0.0014773437287658453 2023-01-23 17:21:11.308986: step: 2780/1541, loss: 0.01437379326671362 2023-01-23 17:21:12.181616: step: 2782/1541, loss: 0.0037812639493495226 2023-01-23 17:21:13.085144: step: 2784/1541, loss: 0.023549232631921768 2023-01-23 17:21:14.009567: step: 2786/1541, loss: 0.12960608303546906 2023-01-23 17:21:14.912733: step: 2788/1541, loss: 0.3580659031867981 2023-01-23 17:21:15.803009: step: 2790/1541, loss: 0.007922082208096981 2023-01-23 17:21:16.696393: step: 2792/1541, loss: 0.009421140886843204 2023-01-23 17:21:17.588595: step: 2794/1541, loss: 0.0005771318101324141 2023-01-23 17:21:18.529187: step: 2796/1541, loss: 0.09064322710037231 2023-01-23 17:21:19.454985: step: 2798/1541, loss: 0.0025149895809590816 2023-01-23 17:21:20.494652: step: 2800/1541, loss: 0.21514572203159332 2023-01-23 17:21:21.416487: step: 2802/1541, loss: 0.00033028129837475717 2023-01-23 17:21:22.307648: step: 2804/1541, loss: 0.3078533411026001 2023-01-23 17:21:23.209232: step: 2806/1541, loss: 0.11829771101474762 2023-01-23 17:21:24.138851: step: 2808/1541, loss: 0.04367951676249504 2023-01-23 17:21:25.012137: step: 2810/1541, loss: 0.08487848192453384 2023-01-23 17:21:25.867209: step: 2812/1541, loss: 0.00016942918591666967 2023-01-23 17:21:26.795682: step: 2814/1541, loss: 0.00327637093141675 2023-01-23 17:21:27.675374: step: 2816/1541, loss: 0.007852674461901188 2023-01-23 17:21:28.564211: step: 2818/1541, loss: 0.0761398896574974 2023-01-23 17:21:29.472107: step: 2820/1541, loss: 0.08502127230167389 2023-01-23 17:21:30.383005: step: 2822/1541, loss: 0.19921237230300903 2023-01-23 17:21:31.317616: step: 2824/1541, loss: 0.023575900122523308 2023-01-23 17:21:32.241500: step: 2826/1541, loss: 0.001741998828947544 2023-01-23 17:21:33.153752: step: 2828/1541, loss: 0.03295888379216194 2023-01-23 17:21:34.022134: step: 2830/1541, loss: 0.16263188421726227 2023-01-23 17:21:34.903934: step: 2832/1541, loss: 0.08448212593793869 2023-01-23 17:21:35.781455: step: 2834/1541, loss: 0.10380850732326508 2023-01-23 17:21:36.696859: step: 2836/1541, loss: 5.210860399529338e-05 2023-01-23 17:21:37.606550: step: 2838/1541, loss: 0.07963988929986954 2023-01-23 17:21:38.501714: step: 2840/1541, loss: 0.22655977308750153 2023-01-23 17:21:39.395278: step: 2842/1541, loss: 0.0022227519657462835 2023-01-23 17:21:40.290155: step: 2844/1541, loss: 0.0012114824494346976 2023-01-23 17:21:41.226883: step: 2846/1541, loss: 0.005384880118072033 2023-01-23 17:21:42.154221: step: 2848/1541, loss: 0.09898459911346436 2023-01-23 17:21:43.044961: step: 2850/1541, loss: 0.004136436153203249 2023-01-23 17:21:43.994548: step: 2852/1541, loss: 0.08245447278022766 2023-01-23 17:21:44.902422: step: 2854/1541, loss: 0.09971477836370468 2023-01-23 17:21:45.785291: step: 2856/1541, loss: 0.0017903267871588469 2023-01-23 17:21:46.687149: step: 2858/1541, loss: 0.19096703827381134 2023-01-23 17:21:47.569717: step: 2860/1541, loss: 0.0009469944052398205 2023-01-23 17:21:48.435047: step: 2862/1541, loss: 0.00010529032442718744 2023-01-23 17:21:49.328639: step: 2864/1541, loss: 0.014064512215554714 2023-01-23 17:21:50.264429: step: 2866/1541, loss: 0.01674852892756462 2023-01-23 17:21:51.136265: step: 2868/1541, loss: 0.0014342907816171646 2023-01-23 17:21:52.181851: step: 2870/1541, loss: 0.05735025554895401 2023-01-23 17:21:53.052864: step: 2872/1541, loss: 0.15490788221359253 2023-01-23 17:21:53.984963: step: 2874/1541, loss: 0.01849713921546936 2023-01-23 17:21:54.883875: step: 2876/1541, loss: 0.1803613305091858 2023-01-23 17:21:55.787417: step: 2878/1541, loss: 0.0024187949020415545 2023-01-23 17:21:56.683192: step: 2880/1541, loss: 0.002117451513186097 2023-01-23 17:21:57.526374: step: 2882/1541, loss: 0.0003171579446643591 2023-01-23 17:21:58.433202: step: 2884/1541, loss: 0.0048717535100877285 2023-01-23 17:21:59.315358: step: 2886/1541, loss: 0.17174257338047028 2023-01-23 17:22:00.224917: step: 2888/1541, loss: 0.24143382906913757 2023-01-23 17:22:01.168166: step: 2890/1541, loss: 0.0008642640896141529 2023-01-23 17:22:02.174717: step: 2892/1541, loss: 0.08378349244594574 2023-01-23 17:22:03.100580: step: 2894/1541, loss: 0.003402936737984419 2023-01-23 17:22:03.990217: step: 2896/1541, loss: 0.37598860263824463 2023-01-23 17:22:04.897078: step: 2898/1541, loss: 0.21666578948497772 2023-01-23 17:22:05.788282: step: 2900/1541, loss: 0.005339363124221563 2023-01-23 17:22:06.707353: step: 2902/1541, loss: 0.055409446358680725 2023-01-23 17:22:07.624514: step: 2904/1541, loss: 0.24445804953575134 2023-01-23 17:22:08.510061: step: 2906/1541, loss: 0.05123898759484291 2023-01-23 17:22:09.397233: step: 2908/1541, loss: 0.005474940873682499 2023-01-23 17:22:10.309469: step: 2910/1541, loss: 0.03154817596077919 2023-01-23 17:22:11.227077: step: 2912/1541, loss: 0.0066941832192242146 2023-01-23 17:22:12.111533: step: 2914/1541, loss: 0.06275531649589539 2023-01-23 17:22:13.053239: step: 2916/1541, loss: 0.005901767406612635 2023-01-23 17:22:13.958544: step: 2918/1541, loss: 0.0011989868944510818 2023-01-23 17:22:14.878993: step: 2920/1541, loss: 0.00044741862802766263 2023-01-23 17:22:15.797440: step: 2922/1541, loss: 0.004181693773716688 2023-01-23 17:22:16.639633: step: 2924/1541, loss: 9.087965965270996 2023-01-23 17:22:17.543226: step: 2926/1541, loss: 0.004974257666617632 2023-01-23 17:22:18.457210: step: 2928/1541, loss: 0.3206561803817749 2023-01-23 17:22:19.347048: step: 2930/1541, loss: 0.0017546098679304123 2023-01-23 17:22:20.242904: step: 2932/1541, loss: 0.08253275603055954 2023-01-23 17:22:21.175224: step: 2934/1541, loss: 0.24880513548851013 2023-01-23 17:22:22.170951: step: 2936/1541, loss: 0.06870082765817642 2023-01-23 17:22:23.048810: step: 2938/1541, loss: 0.0008096725214272738 2023-01-23 17:22:23.951968: step: 2940/1541, loss: 0.01335899531841278 2023-01-23 17:22:24.831837: step: 2942/1541, loss: 0.0004333677643444389 2023-01-23 17:22:25.739819: step: 2944/1541, loss: 0.0007971763843670487 2023-01-23 17:22:26.700839: step: 2946/1541, loss: 0.010259030386805534 2023-01-23 17:22:27.604870: step: 2948/1541, loss: 0.0005966806784272194 2023-01-23 17:22:28.516068: step: 2950/1541, loss: 0.0002531518111936748 2023-01-23 17:22:29.469062: step: 2952/1541, loss: 0.005958755500614643 2023-01-23 17:22:30.428844: step: 2954/1541, loss: 0.28638821840286255 2023-01-23 17:22:31.324233: step: 2956/1541, loss: 0.15130826830863953 2023-01-23 17:22:32.233071: step: 2958/1541, loss: 0.0009403994772583246 2023-01-23 17:22:33.173683: step: 2960/1541, loss: 0.13494469225406647 2023-01-23 17:22:34.067041: step: 2962/1541, loss: 0.0006669907015748322 2023-01-23 17:22:34.950197: step: 2964/1541, loss: 0.12177790701389313 2023-01-23 17:22:35.835423: step: 2966/1541, loss: 0.021316519007086754 2023-01-23 17:22:36.701533: step: 2968/1541, loss: 0.0012185316300019622 2023-01-23 17:22:37.656145: step: 2970/1541, loss: 0.0071175131015479565 2023-01-23 17:22:38.549830: step: 2972/1541, loss: 0.02172246389091015 2023-01-23 17:22:39.557303: step: 2974/1541, loss: 0.01295074075460434 2023-01-23 17:22:40.450741: step: 2976/1541, loss: 0.08402504026889801 2023-01-23 17:22:41.370678: step: 2978/1541, loss: 0.013186803087592125 2023-01-23 17:22:42.276577: step: 2980/1541, loss: 0.3013596534729004 2023-01-23 17:22:43.179695: step: 2982/1541, loss: 0.007827762514352798 2023-01-23 17:22:44.093138: step: 2984/1541, loss: 0.01147311832755804 2023-01-23 17:22:45.007771: step: 2986/1541, loss: 0.01278086006641388 2023-01-23 17:22:45.958623: step: 2988/1541, loss: 0.10038353502750397 2023-01-23 17:22:46.826315: step: 2990/1541, loss: 0.2186928540468216 2023-01-23 17:22:47.767830: step: 2992/1541, loss: 0.1092069149017334 2023-01-23 17:22:48.632039: step: 2994/1541, loss: 0.09676552563905716 2023-01-23 17:22:49.517225: step: 2996/1541, loss: 0.03690134733915329 2023-01-23 17:22:50.381209: step: 2998/1541, loss: 0.005170505028218031 2023-01-23 17:22:51.298655: step: 3000/1541, loss: 0.011037451215088367 2023-01-23 17:22:52.164730: step: 3002/1541, loss: 0.18295718729496002 2023-01-23 17:22:53.118429: step: 3004/1541, loss: 0.01764056086540222 2023-01-23 17:22:53.989646: step: 3006/1541, loss: 0.08123290538787842 2023-01-23 17:22:54.912428: step: 3008/1541, loss: 0.34328579902648926 2023-01-23 17:22:55.815327: step: 3010/1541, loss: 0.03229101002216339 2023-01-23 17:22:56.675198: step: 3012/1541, loss: 0.0010827133664861321 2023-01-23 17:22:57.550892: step: 3014/1541, loss: 0.02449031174182892 2023-01-23 17:22:58.485984: step: 3016/1541, loss: 0.040000494569540024 2023-01-23 17:22:59.367253: step: 3018/1541, loss: 0.0031573683954775333 2023-01-23 17:23:00.264164: step: 3020/1541, loss: 0.14618971943855286 2023-01-23 17:23:01.170145: step: 3022/1541, loss: 0.0014637098647654057 2023-01-23 17:23:02.080596: step: 3024/1541, loss: 0.025013793259859085 2023-01-23 17:23:02.960465: step: 3026/1541, loss: 0.48725783824920654 2023-01-23 17:23:03.865012: step: 3028/1541, loss: 0.11209611594676971 2023-01-23 17:23:04.760923: step: 3030/1541, loss: 0.17363214492797852 2023-01-23 17:23:05.649521: step: 3032/1541, loss: 0.003163662739098072 2023-01-23 17:23:06.502490: step: 3034/1541, loss: 0.7829481363296509 2023-01-23 17:23:07.399869: step: 3036/1541, loss: 0.003483027685433626 2023-01-23 17:23:08.291065: step: 3038/1541, loss: 0.011670948937535286 2023-01-23 17:23:09.167551: step: 3040/1541, loss: 0.013355255126953125 2023-01-23 17:23:10.044027: step: 3042/1541, loss: 0.0011032449547201395 2023-01-23 17:23:10.926681: step: 3044/1541, loss: 0.012667224742472172 2023-01-23 17:23:11.860233: step: 3046/1541, loss: 0.00499608414247632 2023-01-23 17:23:12.729530: step: 3048/1541, loss: 0.037906941026449203 2023-01-23 17:23:13.629247: step: 3050/1541, loss: 0.001664096605964005 2023-01-23 17:23:14.518617: step: 3052/1541, loss: 0.05766548961400986 2023-01-23 17:23:15.427234: step: 3054/1541, loss: 0.044556472450494766 2023-01-23 17:23:16.366930: step: 3056/1541, loss: 0.00010961879161186516 2023-01-23 17:23:17.289421: step: 3058/1541, loss: 0.012562703341245651 2023-01-23 17:23:18.148792: step: 3060/1541, loss: 0.00582300266250968 2023-01-23 17:23:19.014696: step: 3062/1541, loss: 0.014249532483518124 2023-01-23 17:23:19.985031: step: 3064/1541, loss: 0.06552017480134964 2023-01-23 17:23:20.879618: step: 3066/1541, loss: 0.479879766702652 2023-01-23 17:23:21.783895: step: 3068/1541, loss: 6.755909271305427e-05 2023-01-23 17:23:22.663499: step: 3070/1541, loss: 1.6077961921691895 2023-01-23 17:23:23.578395: step: 3072/1541, loss: 0.0374022014439106 2023-01-23 17:23:24.460079: step: 3074/1541, loss: 0.0007857019663788378 2023-01-23 17:23:25.365631: step: 3076/1541, loss: 0.01611666940152645 2023-01-23 17:23:26.233204: step: 3078/1541, loss: 1.1937618255615234 2023-01-23 17:23:27.096819: step: 3080/1541, loss: 0.0053702169097959995 2023-01-23 17:23:28.006391: step: 3082/1541, loss: 0.03688746318221092 2023-01-23 17:23:28.589345: step: 3084/1541, loss: 0.006555307190865278 ================================================== Loss: 0.082 -------------------- Dev Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6748395028808764, 'r': 0.6385747632400816, 'f1': 0.6562064793919685}, 'combined': 0.4371955007535911, 'epoch': 25} Test Chinese: {'event': {'p': 0.6109358569926393, 'r': 0.6962252846015579, 'f1': 0.6507980957714926}, 'argument': {'p': 0.7154126636351134, 'r': 0.6940648330608487, 'f1': 0.7045770821024057}, 'combined': 0.45853742335648023, 'epoch': 25} Dev Korean: {'event': {'p': 0.5774509803921568, 'r': 0.7842876165113183, 'f1': 0.6651609260304913}, 'argument': {'p': 0.6698169538960423, 'r': 0.6525065372842279, 'f1': 0.6610484410450397}, 'combined': 0.4397035931965312, 'epoch': 25} Test Korean: {'event': {'p': 0.611534795042898, 'r': 0.768723786698622, 'f1': 0.6811786567560393}, 'argument': {'p': 0.7157954795955861, 'r': 0.7080069011935282, 'f1': 0.711879887594975}, 'combined': 0.4849173856035853, 'epoch': 25} Dev Russian: {'event': {'p': 0.6563981042654028, 'r': 0.7376830892143809, 'f1': 0.6946708463949843}, 'argument': {'p': 0.6773132344375045, 'r': 0.6472444001020548, 'f1': 0.6619375203814012}, 'combined': 0.4598286975439452, 'epoch': 25} Test Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7165382631843054, 'r': 0.6985347892349509, 'f1': 0.7074220002939199}, 'combined': 0.4912433219918114, 'epoch': 25} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6613345847859121, 'r': 0.7050076234038498, 'f1': 0.6824731331580647}, 'combined': 0.5133131257940999, 'epoch': 25} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5668613357292602, 'r': 0.3901772830344259, 'f1': 0.4622100122100122}, 'combined': 0.26931921971292044, 'epoch': 25} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5875966625966627, 'r': 0.573931623931624, 'f1': 0.5806837606837608}, 'combined': 0.29863736263736274, 'epoch': 25} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6748395028808764, 'r': 0.6385747632400816, 'f1': 0.6562064793919685}, 'combined': 0.4371955007535911, 'epoch': 25} Test for Chinese: {'event': {'p': 0.6109358569926393, 'r': 0.6962252846015579, 'f1': 0.6507980957714926}, 'argument': {'p': 0.7154126636351134, 'r': 0.6940648330608487, 'f1': 0.7045770821024057}, 'combined': 0.45853742335648023, 'epoch': 25} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6613345847859121, 'r': 0.7050076234038498, 'f1': 0.6824731331580647}, 'combined': 0.5133131257940999, 'epoch': 25} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 26 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 17:28:55.993888: step: 2/1541, loss: 0.00401120213791728 2023-01-23 17:28:56.880930: step: 4/1541, loss: 0.0010135489283129573 2023-01-23 17:28:57.814132: step: 6/1541, loss: 0.0004984920378774405 2023-01-23 17:28:58.667380: step: 8/1541, loss: 0.00021210710110608488 2023-01-23 17:28:59.538584: step: 10/1541, loss: 0.16781795024871826 2023-01-23 17:29:00.425796: step: 12/1541, loss: 0.057517342269420624 2023-01-23 17:29:01.352860: step: 14/1541, loss: 0.034966640174388885 2023-01-23 17:29:02.233281: step: 16/1541, loss: 0.000978513271547854 2023-01-23 17:29:03.157813: step: 18/1541, loss: 0.018645504489541054 2023-01-23 17:29:04.041346: step: 20/1541, loss: 0.00458922004327178 2023-01-23 17:29:04.904402: step: 22/1541, loss: 0.0019141464726999402 2023-01-23 17:29:05.822180: step: 24/1541, loss: 0.1320715993642807 2023-01-23 17:29:06.734785: step: 26/1541, loss: 0.004668141715228558 2023-01-23 17:29:07.626938: step: 28/1541, loss: 0.02626211754977703 2023-01-23 17:29:08.492432: step: 30/1541, loss: 0.02574007585644722 2023-01-23 17:29:09.385770: step: 32/1541, loss: 0.0008135272655636072 2023-01-23 17:29:10.255330: step: 34/1541, loss: 0.0012855723034590483 2023-01-23 17:29:11.137185: step: 36/1541, loss: 0.011521203443408012 2023-01-23 17:29:12.012683: step: 38/1541, loss: 0.045992787927389145 2023-01-23 17:29:12.881048: step: 40/1541, loss: 0.0010191568871960044 2023-01-23 17:29:13.822834: step: 42/1541, loss: 0.008569210767745972 2023-01-23 17:29:14.704342: step: 44/1541, loss: 0.028271883726119995 2023-01-23 17:29:15.564901: step: 46/1541, loss: 0.0002544703602325171 2023-01-23 17:29:16.459518: step: 48/1541, loss: 6.245687109185383e-05 2023-01-23 17:29:17.342769: step: 50/1541, loss: 0.0003251680755056441 2023-01-23 17:29:18.246018: step: 52/1541, loss: 0.05148226395249367 2023-01-23 17:29:19.115880: step: 54/1541, loss: 0.03433370962738991 2023-01-23 17:29:19.987634: step: 56/1541, loss: 0.05613824725151062 2023-01-23 17:29:20.868333: step: 58/1541, loss: 0.0011278269812464714 2023-01-23 17:29:21.752200: step: 60/1541, loss: 0.006111787166446447 2023-01-23 17:29:22.626622: step: 62/1541, loss: 0.023114003241062164 2023-01-23 17:29:23.480093: step: 64/1541, loss: 0.033055417239665985 2023-01-23 17:29:24.394204: step: 66/1541, loss: 0.04021983966231346 2023-01-23 17:29:25.290755: step: 68/1541, loss: 0.10177158564329147 2023-01-23 17:29:26.201891: step: 70/1541, loss: 0.0002971911453641951 2023-01-23 17:29:27.113064: step: 72/1541, loss: 0.046450622379779816 2023-01-23 17:29:27.989826: step: 74/1541, loss: 0.13485823571681976 2023-01-23 17:29:28.932264: step: 76/1541, loss: 0.6364850997924805 2023-01-23 17:29:29.824504: step: 78/1541, loss: 9.083363693207502e-05 2023-01-23 17:29:30.795527: step: 80/1541, loss: 0.011843699030578136 2023-01-23 17:29:31.698736: step: 82/1541, loss: 0.049694377928972244 2023-01-23 17:29:32.581663: step: 84/1541, loss: 0.04862583056092262 2023-01-23 17:29:33.512157: step: 86/1541, loss: 0.002659125719219446 2023-01-23 17:29:34.444077: step: 88/1541, loss: 0.0055660055950284 2023-01-23 17:29:35.316593: step: 90/1541, loss: 0.12402797490358353 2023-01-23 17:29:36.260442: step: 92/1541, loss: 0.009506622329354286 2023-01-23 17:29:37.209458: step: 94/1541, loss: 0.10543501377105713 2023-01-23 17:29:38.097605: step: 96/1541, loss: 0.00010721814760472625 2023-01-23 17:29:38.977879: step: 98/1541, loss: 0.0004480578936636448 2023-01-23 17:29:39.843134: step: 100/1541, loss: 0.0010902585927397013 2023-01-23 17:29:40.756468: step: 102/1541, loss: 0.2699539065361023 2023-01-23 17:29:41.673133: step: 104/1541, loss: 0.0850018560886383 2023-01-23 17:29:42.575676: step: 106/1541, loss: 0.02265254408121109 2023-01-23 17:29:43.484587: step: 108/1541, loss: 0.0035628643818199635 2023-01-23 17:29:44.427624: step: 110/1541, loss: 0.062366075813770294 2023-01-23 17:29:45.349640: step: 112/1541, loss: 0.00045165300252847373 2023-01-23 17:29:46.236559: step: 114/1541, loss: 0.10214326530694962 2023-01-23 17:29:47.123040: step: 116/1541, loss: 0.014692059718072414 2023-01-23 17:29:47.986606: step: 118/1541, loss: 0.0003192488511558622 2023-01-23 17:29:48.874160: step: 120/1541, loss: 0.007526181172579527 2023-01-23 17:29:49.737321: step: 122/1541, loss: 0.07622459530830383 2023-01-23 17:29:50.604788: step: 124/1541, loss: 0.18562090396881104 2023-01-23 17:29:51.517337: step: 126/1541, loss: 0.002738791285082698 2023-01-23 17:29:52.468030: step: 128/1541, loss: 0.0005781205254606903 2023-01-23 17:29:53.377895: step: 130/1541, loss: 0.0026950889732688665 2023-01-23 17:29:54.276771: step: 132/1541, loss: 0.02023504301905632 2023-01-23 17:29:55.183489: step: 134/1541, loss: 0.0003825342864729464 2023-01-23 17:29:56.094810: step: 136/1541, loss: 0.002765616402029991 2023-01-23 17:29:56.970394: step: 138/1541, loss: 0.00022544550301972777 2023-01-23 17:29:57.849473: step: 140/1541, loss: 0.225179985165596 2023-01-23 17:29:58.736516: step: 142/1541, loss: 0.007971406914293766 2023-01-23 17:29:59.634963: step: 144/1541, loss: 0.00022751596407033503 2023-01-23 17:30:00.559322: step: 146/1541, loss: 9.328527085017413e-05 2023-01-23 17:30:01.454905: step: 148/1541, loss: 0.13826915621757507 2023-01-23 17:30:02.362598: step: 150/1541, loss: 0.0015474112005904317 2023-01-23 17:30:03.230359: step: 152/1541, loss: 0.0014562022406607866 2023-01-23 17:30:04.121751: step: 154/1541, loss: 0.120822012424469 2023-01-23 17:30:05.035414: step: 156/1541, loss: 0.00012278703798074275 2023-01-23 17:30:05.917345: step: 158/1541, loss: 0.22528088092803955 2023-01-23 17:30:06.775038: step: 160/1541, loss: 0.0017896387726068497 2023-01-23 17:30:07.659119: step: 162/1541, loss: 0.12645640969276428 2023-01-23 17:30:08.546707: step: 164/1541, loss: 0.02658686414361 2023-01-23 17:30:09.456623: step: 166/1541, loss: 0.003732598852366209 2023-01-23 17:30:10.326646: step: 168/1541, loss: 0.00031010943348519504 2023-01-23 17:30:11.224517: step: 170/1541, loss: 0.00024011445930227637 2023-01-23 17:30:12.065905: step: 172/1541, loss: 0.00021205938537605107 2023-01-23 17:30:12.949164: step: 174/1541, loss: 0.08815600723028183 2023-01-23 17:30:13.857510: step: 176/1541, loss: 0.007104738615453243 2023-01-23 17:30:14.744705: step: 178/1541, loss: 0.011671408079564571 2023-01-23 17:30:15.624903: step: 180/1541, loss: 0.0001315923291258514 2023-01-23 17:30:16.507596: step: 182/1541, loss: 0.4769648611545563 2023-01-23 17:30:17.469845: step: 184/1541, loss: 0.10706216096878052 2023-01-23 17:30:18.384310: step: 186/1541, loss: 0.005152874626219273 2023-01-23 17:30:19.348781: step: 188/1541, loss: 0.1015864759683609 2023-01-23 17:30:20.268793: step: 190/1541, loss: 0.024348415434360504 2023-01-23 17:30:21.165001: step: 192/1541, loss: 0.01733611710369587 2023-01-23 17:30:22.017120: step: 194/1541, loss: 0.014649861492216587 2023-01-23 17:30:22.982485: step: 196/1541, loss: 0.0018818010576069355 2023-01-23 17:30:23.891043: step: 198/1541, loss: 0.01038581132888794 2023-01-23 17:30:24.763379: step: 200/1541, loss: 0.3031465709209442 2023-01-23 17:30:25.632341: step: 202/1541, loss: 0.001189058180898428 2023-01-23 17:30:26.576473: step: 204/1541, loss: 0.017211884260177612 2023-01-23 17:30:27.583152: step: 206/1541, loss: 0.04504112899303436 2023-01-23 17:30:28.437603: step: 208/1541, loss: 0.00869370624423027 2023-01-23 17:30:29.301703: step: 210/1541, loss: 0.39489370584487915 2023-01-23 17:30:30.222245: step: 212/1541, loss: 0.01165157463401556 2023-01-23 17:30:31.147772: step: 214/1541, loss: 0.13277366757392883 2023-01-23 17:30:32.008266: step: 216/1541, loss: 0.00723331468179822 2023-01-23 17:30:32.925412: step: 218/1541, loss: 0.004273855127394199 2023-01-23 17:30:33.775797: step: 220/1541, loss: 0.04314164072275162 2023-01-23 17:30:34.653960: step: 222/1541, loss: 0.08349571377038956 2023-01-23 17:30:35.536622: step: 224/1541, loss: 0.06410456448793411 2023-01-23 17:30:36.391083: step: 226/1541, loss: 0.0005843473481945693 2023-01-23 17:30:37.324028: step: 228/1541, loss: 0.00016260759730357677 2023-01-23 17:30:38.230957: step: 230/1541, loss: 0.042903803288936615 2023-01-23 17:30:39.202357: step: 232/1541, loss: 0.000551815377548337 2023-01-23 17:30:40.115049: step: 234/1541, loss: 0.5471183061599731 2023-01-23 17:30:41.005443: step: 236/1541, loss: 3.3636119042057544e-05 2023-01-23 17:30:41.953047: step: 238/1541, loss: 0.0015415200032293797 2023-01-23 17:30:42.855217: step: 240/1541, loss: 0.0005416450439952314 2023-01-23 17:30:43.789024: step: 242/1541, loss: 7.43508426239714e-05 2023-01-23 17:30:44.699045: step: 244/1541, loss: 0.013142191804945469 2023-01-23 17:30:45.583081: step: 246/1541, loss: 0.00033186166547238827 2023-01-23 17:30:46.526235: step: 248/1541, loss: 0.4223596751689911 2023-01-23 17:30:47.435072: step: 250/1541, loss: 0.0008109091431833804 2023-01-23 17:30:48.315208: step: 252/1541, loss: 0.20774225890636444 2023-01-23 17:30:49.210528: step: 254/1541, loss: 0.009410814382135868 2023-01-23 17:30:50.068517: step: 256/1541, loss: 3.0295148462755606e-05 2023-01-23 17:30:50.896550: step: 258/1541, loss: 0.000822004396468401 2023-01-23 17:30:51.771508: step: 260/1541, loss: 0.00010396756988484412 2023-01-23 17:30:52.625718: step: 262/1541, loss: 0.34293195605278015 2023-01-23 17:30:53.557827: step: 264/1541, loss: 0.006027680821716785 2023-01-23 17:30:54.449750: step: 266/1541, loss: 0.005622449330985546 2023-01-23 17:30:55.320602: step: 268/1541, loss: 0.15971754491329193 2023-01-23 17:30:56.214446: step: 270/1541, loss: 0.35487100481987 2023-01-23 17:30:57.126159: step: 272/1541, loss: 0.00025103340158239007 2023-01-23 17:30:58.034031: step: 274/1541, loss: 0.17860612273216248 2023-01-23 17:30:59.020222: step: 276/1541, loss: 0.0005296823219396174 2023-01-23 17:30:59.860302: step: 278/1541, loss: 0.0014510073233395815 2023-01-23 17:31:00.730073: step: 280/1541, loss: 0.009667456150054932 2023-01-23 17:31:01.650195: step: 282/1541, loss: 0.011165887117385864 2023-01-23 17:31:02.562564: step: 284/1541, loss: 0.20742163062095642 2023-01-23 17:31:03.473428: step: 286/1541, loss: 0.00013190681056585163 2023-01-23 17:31:04.389655: step: 288/1541, loss: 0.001627886900678277 2023-01-23 17:31:05.299571: step: 290/1541, loss: 0.0010263776639476418 2023-01-23 17:31:06.171855: step: 292/1541, loss: 0.002109107794240117 2023-01-23 17:31:07.031966: step: 294/1541, loss: 0.007780379615724087 2023-01-23 17:31:07.931902: step: 296/1541, loss: 0.04988076537847519 2023-01-23 17:31:08.806916: step: 298/1541, loss: 0.01557240542024374 2023-01-23 17:31:09.699858: step: 300/1541, loss: 0.00018929973884951323 2023-01-23 17:31:10.592455: step: 302/1541, loss: 0.0032380667980760336 2023-01-23 17:31:11.498943: step: 304/1541, loss: 0.007237069308757782 2023-01-23 17:31:12.460944: step: 306/1541, loss: 0.008534278720617294 2023-01-23 17:31:13.402840: step: 308/1541, loss: 0.0006715047638863325 2023-01-23 17:31:14.302217: step: 310/1541, loss: 0.039938975125551224 2023-01-23 17:31:15.232639: step: 312/1541, loss: 0.020891409367322922 2023-01-23 17:31:16.143953: step: 314/1541, loss: 0.007578189950436354 2023-01-23 17:31:17.085617: step: 316/1541, loss: 0.1443951427936554 2023-01-23 17:31:18.048239: step: 318/1541, loss: 0.000462741416413337 2023-01-23 17:31:18.917156: step: 320/1541, loss: 0.02999832294881344 2023-01-23 17:31:19.822205: step: 322/1541, loss: 0.00018406880553811789 2023-01-23 17:31:20.678424: step: 324/1541, loss: 0.05176481977105141 2023-01-23 17:31:21.555117: step: 326/1541, loss: 0.0020413920283317566 2023-01-23 17:31:22.459018: step: 328/1541, loss: 0.001840456505306065 2023-01-23 17:31:23.390151: step: 330/1541, loss: 0.002314443001523614 2023-01-23 17:31:24.342119: step: 332/1541, loss: 0.005632978398352861 2023-01-23 17:31:25.273054: step: 334/1541, loss: 0.044772177934646606 2023-01-23 17:31:26.202052: step: 336/1541, loss: 0.0006152429268695414 2023-01-23 17:31:27.117988: step: 338/1541, loss: 0.0051577407866716385 2023-01-23 17:31:27.978231: step: 340/1541, loss: 0.03028446063399315 2023-01-23 17:31:28.888858: step: 342/1541, loss: 0.1539977788925171 2023-01-23 17:31:29.770764: step: 344/1541, loss: 0.19609633088111877 2023-01-23 17:31:30.716718: step: 346/1541, loss: 0.17526304721832275 2023-01-23 17:31:31.646162: step: 348/1541, loss: 0.0037055888678878546 2023-01-23 17:31:32.574896: step: 350/1541, loss: 0.09729211032390594 2023-01-23 17:31:33.431036: step: 352/1541, loss: 0.0015138856833800673 2023-01-23 17:31:34.345893: step: 354/1541, loss: 0.007091935258358717 2023-01-23 17:31:35.228702: step: 356/1541, loss: 0.00023619607964064926 2023-01-23 17:31:36.094778: step: 358/1541, loss: 0.00018019639537669718 2023-01-23 17:31:36.988147: step: 360/1541, loss: 0.03714178502559662 2023-01-23 17:31:37.911355: step: 362/1541, loss: 0.01643664948642254 2023-01-23 17:31:38.809798: step: 364/1541, loss: 0.0005651481915265322 2023-01-23 17:31:39.726479: step: 366/1541, loss: 0.005450605880469084 2023-01-23 17:31:40.641809: step: 368/1541, loss: 0.0016273267101496458 2023-01-23 17:31:41.570760: step: 370/1541, loss: 0.002771329367533326 2023-01-23 17:31:42.453417: step: 372/1541, loss: 8.772475848672912e-05 2023-01-23 17:31:43.323065: step: 374/1541, loss: 6.516973371617496e-05 2023-01-23 17:31:44.225531: step: 376/1541, loss: 0.030556613579392433 2023-01-23 17:31:45.082971: step: 378/1541, loss: 0.00011593548697419465 2023-01-23 17:31:45.979725: step: 380/1541, loss: 0.001967966789379716 2023-01-23 17:31:46.897564: step: 382/1541, loss: 0.00032985134748741984 2023-01-23 17:31:47.819613: step: 384/1541, loss: 0.001969611272215843 2023-01-23 17:31:48.748426: step: 386/1541, loss: 0.016619820147752762 2023-01-23 17:31:49.668501: step: 388/1541, loss: 5.43337082490325e-05 2023-01-23 17:31:50.585128: step: 390/1541, loss: 0.0001318075374001637 2023-01-23 17:31:51.470346: step: 392/1541, loss: 0.0025005631614476442 2023-01-23 17:31:52.369625: step: 394/1541, loss: 0.07263505458831787 2023-01-23 17:31:53.270767: step: 396/1541, loss: 0.13053680956363678 2023-01-23 17:31:54.143911: step: 398/1541, loss: 0.00028038679738529027 2023-01-23 17:31:55.077184: step: 400/1541, loss: 0.004299825057387352 2023-01-23 17:31:56.004628: step: 402/1541, loss: 0.007530106231570244 2023-01-23 17:31:56.895776: step: 404/1541, loss: 0.03694280609488487 2023-01-23 17:31:57.746669: step: 406/1541, loss: 5.689287354471162e-05 2023-01-23 17:31:58.709962: step: 408/1541, loss: 0.0004833271959796548 2023-01-23 17:31:59.559558: step: 410/1541, loss: 4.626697045750916e-05 2023-01-23 17:32:00.483542: step: 412/1541, loss: 0.0001520430960226804 2023-01-23 17:32:01.431110: step: 414/1541, loss: 0.004734792746603489 2023-01-23 17:32:02.324311: step: 416/1541, loss: 0.16214685142040253 2023-01-23 17:32:03.187355: step: 418/1541, loss: 3.567321255104616e-05 2023-01-23 17:32:04.072105: step: 420/1541, loss: 0.00021491969528142363 2023-01-23 17:32:04.990883: step: 422/1541, loss: 0.000941118982154876 2023-01-23 17:32:05.889931: step: 424/1541, loss: 0.000596729398239404 2023-01-23 17:32:06.764010: step: 426/1541, loss: 0.003391280071809888 2023-01-23 17:32:07.684235: step: 428/1541, loss: 0.20937424898147583 2023-01-23 17:32:08.568787: step: 430/1541, loss: 0.003462341846898198 2023-01-23 17:32:09.496635: step: 432/1541, loss: 0.601220428943634 2023-01-23 17:32:10.384540: step: 434/1541, loss: 0.002253335202112794 2023-01-23 17:32:11.254138: step: 436/1541, loss: 0.00262727914378047 2023-01-23 17:32:12.112702: step: 438/1541, loss: 0.0005083588766865432 2023-01-23 17:32:12.997950: step: 440/1541, loss: 0.007830715738236904 2023-01-23 17:32:13.933976: step: 442/1541, loss: 0.012794713489711285 2023-01-23 17:32:14.795983: step: 444/1541, loss: 0.0030719239730387926 2023-01-23 17:32:15.757059: step: 446/1541, loss: 0.18635554611682892 2023-01-23 17:32:16.650304: step: 448/1541, loss: 0.06899826228618622 2023-01-23 17:32:17.558399: step: 450/1541, loss: 0.0071840472519397736 2023-01-23 17:32:18.448601: step: 452/1541, loss: 0.0672239363193512 2023-01-23 17:32:19.315854: step: 454/1541, loss: 0.02468043752014637 2023-01-23 17:32:20.160258: step: 456/1541, loss: 0.0018033471424132586 2023-01-23 17:32:21.020051: step: 458/1541, loss: 0.0003688119468279183 2023-01-23 17:32:21.888085: step: 460/1541, loss: 0.014284765347838402 2023-01-23 17:32:22.771514: step: 462/1541, loss: 0.010126964189112186 2023-01-23 17:32:23.704136: step: 464/1541, loss: 0.158375546336174 2023-01-23 17:32:24.588674: step: 466/1541, loss: 0.08437390625476837 2023-01-23 17:32:25.452926: step: 468/1541, loss: 0.003073807805776596 2023-01-23 17:32:26.303884: step: 470/1541, loss: 0.0017020991072058678 2023-01-23 17:32:27.211933: step: 472/1541, loss: 3.5771963666775264e-06 2023-01-23 17:32:28.122104: step: 474/1541, loss: 0.0020564482547342777 2023-01-23 17:32:29.041774: step: 476/1541, loss: 7.942500815261155e-05 2023-01-23 17:32:29.970565: step: 478/1541, loss: 0.06711086630821228 2023-01-23 17:32:30.905827: step: 480/1541, loss: 0.0022609715815633535 2023-01-23 17:32:31.873509: step: 482/1541, loss: 0.00039676736923865974 2023-01-23 17:32:32.811924: step: 484/1541, loss: 0.05774492770433426 2023-01-23 17:32:33.682351: step: 486/1541, loss: 0.0008715770673006773 2023-01-23 17:32:34.573011: step: 488/1541, loss: 0.030917750671505928 2023-01-23 17:32:35.496298: step: 490/1541, loss: 0.0021401518024504185 2023-01-23 17:32:36.456631: step: 492/1541, loss: 0.12162595987319946 2023-01-23 17:32:37.348008: step: 494/1541, loss: 0.0015329020097851753 2023-01-23 17:32:38.290515: step: 496/1541, loss: 0.001980891451239586 2023-01-23 17:32:39.167215: step: 498/1541, loss: 4.000937406090088e-05 2023-01-23 17:32:40.096259: step: 500/1541, loss: 0.00038391686393879354 2023-01-23 17:32:40.984713: step: 502/1541, loss: 0.0012013837695121765 2023-01-23 17:32:41.915422: step: 504/1541, loss: 0.00042331102304160595 2023-01-23 17:32:42.811173: step: 506/1541, loss: 0.009925664402544498 2023-01-23 17:32:43.640435: step: 508/1541, loss: 0.0002603757311590016 2023-01-23 17:32:44.578755: step: 510/1541, loss: 1.6447691917419434 2023-01-23 17:32:45.505235: step: 512/1541, loss: 0.01206746231764555 2023-01-23 17:32:46.425625: step: 514/1541, loss: 0.10639763623476028 2023-01-23 17:32:47.305331: step: 516/1541, loss: 0.04000266268849373 2023-01-23 17:32:48.277073: step: 518/1541, loss: 0.004236126784235239 2023-01-23 17:32:49.195731: step: 520/1541, loss: 0.00014818519412074238 2023-01-23 17:32:50.135539: step: 522/1541, loss: 0.0052007208578288555 2023-01-23 17:32:51.037969: step: 524/1541, loss: 0.0031033973209559917 2023-01-23 17:32:51.995341: step: 526/1541, loss: 0.3800598680973053 2023-01-23 17:32:52.864851: step: 528/1541, loss: 0.13969576358795166 2023-01-23 17:32:53.708122: step: 530/1541, loss: 0.064781054854393 2023-01-23 17:32:54.612138: step: 532/1541, loss: 0.0005680330796167254 2023-01-23 17:32:55.516234: step: 534/1541, loss: 0.0010804925113916397 2023-01-23 17:32:56.468001: step: 536/1541, loss: 0.04007064178586006 2023-01-23 17:32:57.351781: step: 538/1541, loss: 0.1409166157245636 2023-01-23 17:32:58.231699: step: 540/1541, loss: 0.003409169614315033 2023-01-23 17:32:59.160654: step: 542/1541, loss: 0.09634570777416229 2023-01-23 17:33:00.096180: step: 544/1541, loss: 0.0033616763539612293 2023-01-23 17:33:00.954826: step: 546/1541, loss: 0.4861772060394287 2023-01-23 17:33:01.824103: step: 548/1541, loss: 0.005052981432527304 2023-01-23 17:33:02.724989: step: 550/1541, loss: 0.2437165379524231 2023-01-23 17:33:03.626476: step: 552/1541, loss: 0.009127918630838394 2023-01-23 17:33:04.497450: step: 554/1541, loss: 0.05546809360384941 2023-01-23 17:33:05.390784: step: 556/1541, loss: 0.00836234912276268 2023-01-23 17:33:06.290330: step: 558/1541, loss: 0.0005296564195305109 2023-01-23 17:33:07.176975: step: 560/1541, loss: 0.00024912410299293697 2023-01-23 17:33:08.073935: step: 562/1541, loss: 0.39497557282447815 2023-01-23 17:33:08.983031: step: 564/1541, loss: 0.1333947479724884 2023-01-23 17:33:09.882711: step: 566/1541, loss: 0.1979508399963379 2023-01-23 17:33:10.756750: step: 568/1541, loss: 0.014920027926564217 2023-01-23 17:33:11.640697: step: 570/1541, loss: 0.03549804538488388 2023-01-23 17:33:12.535581: step: 572/1541, loss: 0.00045756695908494294 2023-01-23 17:33:13.413526: step: 574/1541, loss: 0.02196403220295906 2023-01-23 17:33:14.298842: step: 576/1541, loss: 0.00316318916156888 2023-01-23 17:33:15.262816: step: 578/1541, loss: 0.004834212362766266 2023-01-23 17:33:16.129419: step: 580/1541, loss: 0.07300879806280136 2023-01-23 17:33:16.990801: step: 582/1541, loss: 0.027611752972006798 2023-01-23 17:33:17.885357: step: 584/1541, loss: 0.0010820006718859076 2023-01-23 17:33:18.813291: step: 586/1541, loss: 0.004439425189048052 2023-01-23 17:33:19.694408: step: 588/1541, loss: 0.6586123108863831 2023-01-23 17:33:20.592012: step: 590/1541, loss: 0.004270058590918779 2023-01-23 17:33:21.462762: step: 592/1541, loss: 0.0018676978070288897 2023-01-23 17:33:22.348654: step: 594/1541, loss: 0.2956749498844147 2023-01-23 17:33:23.237602: step: 596/1541, loss: 0.22672802209854126 2023-01-23 17:33:24.118620: step: 598/1541, loss: 0.011690856888890266 2023-01-23 17:33:24.998606: step: 600/1541, loss: 0.0003580644552130252 2023-01-23 17:33:25.860887: step: 602/1541, loss: 0.09819266945123672 2023-01-23 17:33:26.766503: step: 604/1541, loss: 0.0018558282172307372 2023-01-23 17:33:27.701833: step: 606/1541, loss: 0.1582762598991394 2023-01-23 17:33:28.618353: step: 608/1541, loss: 0.0007678780821152031 2023-01-23 17:33:29.495502: step: 610/1541, loss: 8.779166819294915e-05 2023-01-23 17:33:30.408929: step: 612/1541, loss: 0.006049741059541702 2023-01-23 17:33:31.285870: step: 614/1541, loss: 0.0003237351484131068 2023-01-23 17:33:32.183500: step: 616/1541, loss: 0.005893843714147806 2023-01-23 17:33:33.110843: step: 618/1541, loss: 0.12252379208803177 2023-01-23 17:33:34.047079: step: 620/1541, loss: 0.16985397040843964 2023-01-23 17:33:34.981717: step: 622/1541, loss: 0.0032018288038671017 2023-01-23 17:33:35.878531: step: 624/1541, loss: 0.007797255180776119 2023-01-23 17:33:36.808165: step: 626/1541, loss: 0.006771296728402376 2023-01-23 17:33:37.719575: step: 628/1541, loss: 0.047782883048057556 2023-01-23 17:33:38.605231: step: 630/1541, loss: 0.0003282743855379522 2023-01-23 17:33:39.511698: step: 632/1541, loss: 0.00034226960269734263 2023-01-23 17:33:40.449895: step: 634/1541, loss: 0.0010430653346702456 2023-01-23 17:33:41.361978: step: 636/1541, loss: 0.0005920342518948019 2023-01-23 17:33:42.233989: step: 638/1541, loss: 0.04636702314019203 2023-01-23 17:33:43.125103: step: 640/1541, loss: 0.0006078130681999028 2023-01-23 17:33:44.015814: step: 642/1541, loss: 0.6256701946258545 2023-01-23 17:33:44.994763: step: 644/1541, loss: 0.00024348213628400117 2023-01-23 17:33:45.909883: step: 646/1541, loss: 0.05757911875844002 2023-01-23 17:33:46.806374: step: 648/1541, loss: -5.349156253942056e-06 2023-01-23 17:33:47.672706: step: 650/1541, loss: 0.407284140586853 2023-01-23 17:33:48.577093: step: 652/1541, loss: 0.10091233998537064 2023-01-23 17:33:49.476700: step: 654/1541, loss: 0.23873849213123322 2023-01-23 17:33:50.349349: step: 656/1541, loss: 0.02862226963043213 2023-01-23 17:33:51.283943: step: 658/1541, loss: 0.05733821913599968 2023-01-23 17:33:52.158557: step: 660/1541, loss: 0.12496265769004822 2023-01-23 17:33:53.027603: step: 662/1541, loss: 0.0006872522062622011 2023-01-23 17:33:53.891450: step: 664/1541, loss: 0.0013984262477606535 2023-01-23 17:33:54.782100: step: 666/1541, loss: 0.12668299674987793 2023-01-23 17:33:55.752706: step: 668/1541, loss: 0.010590774938464165 2023-01-23 17:33:56.654575: step: 670/1541, loss: 0.10347922146320343 2023-01-23 17:33:57.562652: step: 672/1541, loss: 0.00011031625035684556 2023-01-23 17:33:58.487566: step: 674/1541, loss: 0.03789270296692848 2023-01-23 17:33:59.391800: step: 676/1541, loss: 0.06426627188920975 2023-01-23 17:34:00.297064: step: 678/1541, loss: 0.5836150050163269 2023-01-23 17:34:01.223372: step: 680/1541, loss: 0.28369760513305664 2023-01-23 17:34:02.188016: step: 682/1541, loss: 0.2658870816230774 2023-01-23 17:34:03.067026: step: 684/1541, loss: 0.00015353331400547177 2023-01-23 17:34:03.925621: step: 686/1541, loss: 0.33741819858551025 2023-01-23 17:34:04.856020: step: 688/1541, loss: 0.37151840329170227 2023-01-23 17:34:05.743721: step: 690/1541, loss: 0.20303258299827576 2023-01-23 17:34:06.621060: step: 692/1541, loss: 0.18925794959068298 2023-01-23 17:34:07.556180: step: 694/1541, loss: 0.02127736061811447 2023-01-23 17:34:08.419502: step: 696/1541, loss: 0.04342465102672577 2023-01-23 17:34:09.281536: step: 698/1541, loss: 0.00047997161163948476 2023-01-23 17:34:10.150598: step: 700/1541, loss: 0.0010578730143606663 2023-01-23 17:34:11.094451: step: 702/1541, loss: 0.0009642053628340364 2023-01-23 17:34:11.990725: step: 704/1541, loss: 0.009535183198750019 2023-01-23 17:34:12.967921: step: 706/1541, loss: 0.0400376133620739 2023-01-23 17:34:13.833051: step: 708/1541, loss: 0.06428224593400955 2023-01-23 17:34:14.720524: step: 710/1541, loss: 0.4005647301673889 2023-01-23 17:34:15.638859: step: 712/1541, loss: 0.0003644576936494559 2023-01-23 17:34:16.522957: step: 714/1541, loss: 0.0916934385895729 2023-01-23 17:34:17.442889: step: 716/1541, loss: 0.11606790125370026 2023-01-23 17:34:18.357734: step: 718/1541, loss: 0.0006182899232953787 2023-01-23 17:34:19.225574: step: 720/1541, loss: 0.06051209568977356 2023-01-23 17:34:20.098103: step: 722/1541, loss: 0.0003891478118021041 2023-01-23 17:34:20.983060: step: 724/1541, loss: 0.1779547780752182 2023-01-23 17:34:21.856121: step: 726/1541, loss: 0.08652909845113754 2023-01-23 17:34:22.782994: step: 728/1541, loss: 0.008173437789082527 2023-01-23 17:34:23.678293: step: 730/1541, loss: 0.007658097427338362 2023-01-23 17:34:24.560985: step: 732/1541, loss: 0.10828260332345963 2023-01-23 17:34:25.441058: step: 734/1541, loss: 0.0020261332392692566 2023-01-23 17:34:26.317400: step: 736/1541, loss: 0.0010388041846454144 2023-01-23 17:34:27.220200: step: 738/1541, loss: 0.0012787212617695332 2023-01-23 17:34:28.142469: step: 740/1541, loss: 0.008317936211824417 2023-01-23 17:34:29.078073: step: 742/1541, loss: 0.29761943221092224 2023-01-23 17:34:30.016275: step: 744/1541, loss: 0.0010782924946397543 2023-01-23 17:34:30.930049: step: 746/1541, loss: 0.09781554341316223 2023-01-23 17:34:31.868885: step: 748/1541, loss: 0.0007031478453427553 2023-01-23 17:34:32.755608: step: 750/1541, loss: 0.007506408728659153 2023-01-23 17:34:33.658290: step: 752/1541, loss: 0.016022589057683945 2023-01-23 17:34:34.529180: step: 754/1541, loss: 0.09065038710832596 2023-01-23 17:34:35.443127: step: 756/1541, loss: 0.0003743863489944488 2023-01-23 17:34:36.366245: step: 758/1541, loss: 0.014228876680135727 2023-01-23 17:34:37.282837: step: 760/1541, loss: 0.0014677810249850154 2023-01-23 17:34:38.221547: step: 762/1541, loss: 0.5785027146339417 2023-01-23 17:34:39.140239: step: 764/1541, loss: 0.0013942563673481345 2023-01-23 17:34:40.031948: step: 766/1541, loss: 0.11565521359443665 2023-01-23 17:34:40.903670: step: 768/1541, loss: 0.05112439766526222 2023-01-23 17:34:41.802667: step: 770/1541, loss: 0.2789290249347687 2023-01-23 17:34:42.779835: step: 772/1541, loss: 0.1245003491640091 2023-01-23 17:34:43.654439: step: 774/1541, loss: 1.6548627172596753e-05 2023-01-23 17:34:44.505081: step: 776/1541, loss: 0.010318887419998646 2023-01-23 17:34:45.406371: step: 778/1541, loss: 0.0007592850597575307 2023-01-23 17:34:46.248923: step: 780/1541, loss: 0.008045041002333164 2023-01-23 17:34:47.118699: step: 782/1541, loss: 0.12054125964641571 2023-01-23 17:34:47.974023: step: 784/1541, loss: 0.01172991469502449 2023-01-23 17:34:48.864071: step: 786/1541, loss: 0.00013965331891085953 2023-01-23 17:34:49.714154: step: 788/1541, loss: 0.00798849854618311 2023-01-23 17:34:50.583806: step: 790/1541, loss: 0.004470342304557562 2023-01-23 17:34:51.576080: step: 792/1541, loss: 0.19652211666107178 2023-01-23 17:34:52.471928: step: 794/1541, loss: 7.941407966427505e-05 2023-01-23 17:34:53.374623: step: 796/1541, loss: 0.17146913707256317 2023-01-23 17:34:54.298296: step: 798/1541, loss: 0.35245969891548157 2023-01-23 17:34:55.244527: step: 800/1541, loss: 0.15341275930404663 2023-01-23 17:34:56.144499: step: 802/1541, loss: 0.0010894151637330651 2023-01-23 17:34:57.059742: step: 804/1541, loss: 0.0006488664075732231 2023-01-23 17:34:57.966741: step: 806/1541, loss: 0.12098581343889236 2023-01-23 17:34:58.873647: step: 808/1541, loss: 0.01004605833441019 2023-01-23 17:34:59.766759: step: 810/1541, loss: 0.0008370914729312062 2023-01-23 17:35:00.667404: step: 812/1541, loss: 0.0005689823301509023 2023-01-23 17:35:01.575121: step: 814/1541, loss: 0.0009752950281836092 2023-01-23 17:35:02.484281: step: 816/1541, loss: 0.010214561596512794 2023-01-23 17:35:03.325413: step: 818/1541, loss: 0.0013718090485781431 2023-01-23 17:35:04.204244: step: 820/1541, loss: 0.01798785850405693 2023-01-23 17:35:05.193365: step: 822/1541, loss: 0.0024161296896636486 2023-01-23 17:35:06.167479: step: 824/1541, loss: 0.0001579618256073445 2023-01-23 17:35:07.068372: step: 826/1541, loss: 0.008004467934370041 2023-01-23 17:35:07.997793: step: 828/1541, loss: 0.013991611078381538 2023-01-23 17:35:08.842633: step: 830/1541, loss: 0.01904761977493763 2023-01-23 17:35:09.765902: step: 832/1541, loss: 0.0006983205676078796 2023-01-23 17:35:10.705201: step: 834/1541, loss: 0.9388204216957092 2023-01-23 17:35:11.619348: step: 836/1541, loss: 0.0009186596143990755 2023-01-23 17:35:12.538098: step: 838/1541, loss: 0.002083207480609417 2023-01-23 17:35:13.425347: step: 840/1541, loss: 0.08670410513877869 2023-01-23 17:35:14.327064: step: 842/1541, loss: 0.02529131807386875 2023-01-23 17:35:15.222616: step: 844/1541, loss: 9.148156823357567e-05 2023-01-23 17:35:16.090496: step: 846/1541, loss: 0.046699170023202896 2023-01-23 17:35:16.935247: step: 848/1541, loss: 0.01499793492257595 2023-01-23 17:35:17.898609: step: 850/1541, loss: 0.006017576437443495 2023-01-23 17:35:18.802223: step: 852/1541, loss: 0.010076616890728474 2023-01-23 17:35:19.716485: step: 854/1541, loss: 0.34097781777381897 2023-01-23 17:35:20.590370: step: 856/1541, loss: 0.02734159119427204 2023-01-23 17:35:21.480930: step: 858/1541, loss: 0.7158618569374084 2023-01-23 17:35:22.365164: step: 860/1541, loss: 0.0006236156332306564 2023-01-23 17:35:23.269740: step: 862/1541, loss: 0.01684856414794922 2023-01-23 17:35:24.189922: step: 864/1541, loss: 0.0026136848609894514 2023-01-23 17:35:25.136779: step: 866/1541, loss: 0.03689352795481682 2023-01-23 17:35:26.027150: step: 868/1541, loss: 0.007217314559966326 2023-01-23 17:35:26.933621: step: 870/1541, loss: 0.00013806768401991576 2023-01-23 17:35:27.848508: step: 872/1541, loss: 0.6344402432441711 2023-01-23 17:35:28.804806: step: 874/1541, loss: 0.12335656583309174 2023-01-23 17:35:29.666654: step: 876/1541, loss: 0.0018639806658029556 2023-01-23 17:35:30.538861: step: 878/1541, loss: 0.3062137961387634 2023-01-23 17:35:31.431255: step: 880/1541, loss: 0.8990775346755981 2023-01-23 17:35:32.360270: step: 882/1541, loss: 0.0571305975317955 2023-01-23 17:35:33.245469: step: 884/1541, loss: 0.0003004761238116771 2023-01-23 17:35:34.135197: step: 886/1541, loss: 0.002654748037457466 2023-01-23 17:35:35.024214: step: 888/1541, loss: 0.00013072976435068995 2023-01-23 17:35:35.912828: step: 890/1541, loss: 0.02611757442355156 2023-01-23 17:35:36.830151: step: 892/1541, loss: 0.01972995698451996 2023-01-23 17:35:37.726769: step: 894/1541, loss: 0.023560835048556328 2023-01-23 17:35:38.674846: step: 896/1541, loss: 0.0007010354893282056 2023-01-23 17:35:39.627799: step: 898/1541, loss: 0.3374880254268646 2023-01-23 17:35:40.523816: step: 900/1541, loss: 0.2477399706840515 2023-01-23 17:35:41.511976: step: 902/1541, loss: 0.16441938281059265 2023-01-23 17:35:42.406446: step: 904/1541, loss: 0.0317772701382637 2023-01-23 17:35:43.295741: step: 906/1541, loss: 0.06258537620306015 2023-01-23 17:35:44.159871: step: 908/1541, loss: 0.1326252967119217 2023-01-23 17:35:45.029216: step: 910/1541, loss: 0.14352574944496155 2023-01-23 17:35:45.925365: step: 912/1541, loss: 0.07156786322593689 2023-01-23 17:35:46.819696: step: 914/1541, loss: 0.11578898876905441 2023-01-23 17:35:47.723283: step: 916/1541, loss: 0.0009878486162051558 2023-01-23 17:35:48.603872: step: 918/1541, loss: 0.003480044426396489 2023-01-23 17:35:49.523043: step: 920/1541, loss: 0.0034711367916315794 2023-01-23 17:35:50.415157: step: 922/1541, loss: 0.009418101981282234 2023-01-23 17:35:51.303769: step: 924/1541, loss: 0.007646926678717136 2023-01-23 17:35:52.198713: step: 926/1541, loss: 0.0028876778669655323 2023-01-23 17:35:53.066485: step: 928/1541, loss: 0.333319753408432 2023-01-23 17:35:53.966000: step: 930/1541, loss: 0.0011000167578458786 2023-01-23 17:35:54.976170: step: 932/1541, loss: 0.024781011044979095 2023-01-23 17:35:55.871368: step: 934/1541, loss: 0.0037267813459038734 2023-01-23 17:35:56.768734: step: 936/1541, loss: 0.012770650908350945 2023-01-23 17:35:57.681463: step: 938/1541, loss: 0.025803428143262863 2023-01-23 17:35:58.569519: step: 940/1541, loss: 0.026410242542624474 2023-01-23 17:35:59.441058: step: 942/1541, loss: 0.005586936604231596 2023-01-23 17:36:00.362565: step: 944/1541, loss: 0.0022097809705883265 2023-01-23 17:36:01.239326: step: 946/1541, loss: 0.000808753480669111 2023-01-23 17:36:02.160532: step: 948/1541, loss: 0.003965216688811779 2023-01-23 17:36:03.039155: step: 950/1541, loss: 0.0001572799519635737 2023-01-23 17:36:03.895595: step: 952/1541, loss: 0.001000461052171886 2023-01-23 17:36:04.742018: step: 954/1541, loss: 0.12320695072412491 2023-01-23 17:36:05.588162: step: 956/1541, loss: 0.0037717244122177362 2023-01-23 17:36:06.543045: step: 958/1541, loss: 0.0443769246339798 2023-01-23 17:36:07.424651: step: 960/1541, loss: 0.0011951439082622528 2023-01-23 17:36:08.320318: step: 962/1541, loss: 0.0009005199535749853 2023-01-23 17:36:09.220198: step: 964/1541, loss: 0.07971120625734329 2023-01-23 17:36:10.110881: step: 966/1541, loss: 1.5223474502563477 2023-01-23 17:36:10.994065: step: 968/1541, loss: 0.001263688551262021 2023-01-23 17:36:11.905688: step: 970/1541, loss: 0.0017781952628865838 2023-01-23 17:36:12.781416: step: 972/1541, loss: 0.18361686170101166 2023-01-23 17:36:13.678888: step: 974/1541, loss: 0.09348548948764801 2023-01-23 17:36:14.598157: step: 976/1541, loss: 0.0008188260835595429 2023-01-23 17:36:15.554377: step: 978/1541, loss: 0.3260839879512787 2023-01-23 17:36:16.484813: step: 980/1541, loss: 0.0023607846815139055 2023-01-23 17:36:17.432464: step: 982/1541, loss: 0.08120125532150269 2023-01-23 17:36:18.326036: step: 984/1541, loss: 0.017861686646938324 2023-01-23 17:36:19.236751: step: 986/1541, loss: 0.006273673847317696 2023-01-23 17:36:20.130856: step: 988/1541, loss: 0.09427618980407715 2023-01-23 17:36:21.008836: step: 990/1541, loss: 0.0017972554778680205 2023-01-23 17:36:21.907802: step: 992/1541, loss: 0.04449013993144035 2023-01-23 17:36:22.801387: step: 994/1541, loss: 0.008678803220391273 2023-01-23 17:36:23.692656: step: 996/1541, loss: 0.07243458181619644 2023-01-23 17:36:24.581771: step: 998/1541, loss: 0.004590278957039118 2023-01-23 17:36:25.509010: step: 1000/1541, loss: 0.2412199079990387 2023-01-23 17:36:26.405064: step: 1002/1541, loss: 0.006801976356655359 2023-01-23 17:36:27.296504: step: 1004/1541, loss: 0.00840606726706028 2023-01-23 17:36:28.190130: step: 1006/1541, loss: 0.011894267052412033 2023-01-23 17:36:29.059860: step: 1008/1541, loss: 0.0026932149194180965 2023-01-23 17:36:29.959234: step: 1010/1541, loss: 0.002690328750759363 2023-01-23 17:36:30.873616: step: 1012/1541, loss: 0.002901877975091338 2023-01-23 17:36:31.795486: step: 1014/1541, loss: 0.0013556736521422863 2023-01-23 17:36:32.650316: step: 1016/1541, loss: 0.0059016491286456585 2023-01-23 17:36:33.560799: step: 1018/1541, loss: 0.030499309301376343 2023-01-23 17:36:34.547546: step: 1020/1541, loss: 0.02324068546295166 2023-01-23 17:36:35.433948: step: 1022/1541, loss: 0.028673242777585983 2023-01-23 17:36:36.298802: step: 1024/1541, loss: 0.031006112694740295 2023-01-23 17:36:37.183046: step: 1026/1541, loss: 0.001024499419145286 2023-01-23 17:36:38.026753: step: 1028/1541, loss: 0.08664188534021378 2023-01-23 17:36:38.927131: step: 1030/1541, loss: 0.000956765259616077 2023-01-23 17:36:39.803194: step: 1032/1541, loss: 0.1069774404168129 2023-01-23 17:36:40.672345: step: 1034/1541, loss: 0.0058104367926716805 2023-01-23 17:36:41.533228: step: 1036/1541, loss: 0.09860307723283768 2023-01-23 17:36:42.428633: step: 1038/1541, loss: 0.18407869338989258 2023-01-23 17:36:43.285427: step: 1040/1541, loss: 0.006579836364835501 2023-01-23 17:36:44.211044: step: 1042/1541, loss: 0.2638927698135376 2023-01-23 17:36:45.069588: step: 1044/1541, loss: 0.004911514930427074 2023-01-23 17:36:45.924587: step: 1046/1541, loss: 0.18181239068508148 2023-01-23 17:36:46.770931: step: 1048/1541, loss: 0.002280206186696887 2023-01-23 17:36:47.668445: step: 1050/1541, loss: 0.08442752808332443 2023-01-23 17:36:48.512527: step: 1052/1541, loss: 0.00015191324928309768 2023-01-23 17:36:49.455768: step: 1054/1541, loss: 0.0327872671186924 2023-01-23 17:36:50.342733: step: 1056/1541, loss: 0.07203377038240433 2023-01-23 17:36:51.260872: step: 1058/1541, loss: 0.09995625913143158 2023-01-23 17:36:52.125388: step: 1060/1541, loss: 0.009711562655866146 2023-01-23 17:36:52.984330: step: 1062/1541, loss: 0.003185961162671447 2023-01-23 17:36:53.847887: step: 1064/1541, loss: 0.00038910467992536724 2023-01-23 17:36:54.715828: step: 1066/1541, loss: 0.0012843336444348097 2023-01-23 17:36:55.579200: step: 1068/1541, loss: 0.007798801176249981 2023-01-23 17:36:56.485603: step: 1070/1541, loss: 0.15586519241333008 2023-01-23 17:36:57.363496: step: 1072/1541, loss: 0.001250864122994244 2023-01-23 17:36:58.228458: step: 1074/1541, loss: 0.010172364301979542 2023-01-23 17:36:59.160181: step: 1076/1541, loss: 0.00022766453912481666 2023-01-23 17:37:00.024517: step: 1078/1541, loss: 0.0008385662222281098 2023-01-23 17:37:00.907253: step: 1080/1541, loss: 0.005541845690459013 2023-01-23 17:37:01.781694: step: 1082/1541, loss: 0.009329051710665226 2023-01-23 17:37:02.680081: step: 1084/1541, loss: 0.0007096846820786595 2023-01-23 17:37:03.554628: step: 1086/1541, loss: 0.10880439728498459 2023-01-23 17:37:04.409798: step: 1088/1541, loss: 0.0017190418438985944 2023-01-23 17:37:05.293091: step: 1090/1541, loss: 0.0006591157289221883 2023-01-23 17:37:06.221446: step: 1092/1541, loss: 0.01753312535583973 2023-01-23 17:37:07.135854: step: 1094/1541, loss: 0.010138905607163906 2023-01-23 17:37:07.989834: step: 1096/1541, loss: 0.05747392028570175 2023-01-23 17:37:08.858132: step: 1098/1541, loss: 0.11226112395524979 2023-01-23 17:37:09.735751: step: 1100/1541, loss: 0.1666209101676941 2023-01-23 17:37:10.626211: step: 1102/1541, loss: 0.0007448947289958596 2023-01-23 17:37:11.532590: step: 1104/1541, loss: 0.2431858479976654 2023-01-23 17:37:12.420150: step: 1106/1541, loss: 0.15975995361804962 2023-01-23 17:37:13.311385: step: 1108/1541, loss: 0.015113186091184616 2023-01-23 17:37:14.249302: step: 1110/1541, loss: 0.0003254014882259071 2023-01-23 17:37:15.151922: step: 1112/1541, loss: 0.01136186346411705 2023-01-23 17:37:16.052298: step: 1114/1541, loss: 0.0002609938383102417 2023-01-23 17:37:16.921302: step: 1116/1541, loss: 0.1481303572654724 2023-01-23 17:37:17.756483: step: 1118/1541, loss: 0.0005450129974633455 2023-01-23 17:37:18.674258: step: 1120/1541, loss: 0.0004261935828253627 2023-01-23 17:37:19.559298: step: 1122/1541, loss: 0.0005198426661081612 2023-01-23 17:37:20.412268: step: 1124/1541, loss: 0.0008657379075884819 2023-01-23 17:37:21.296033: step: 1126/1541, loss: 0.005508897826075554 2023-01-23 17:37:22.196474: step: 1128/1541, loss: 0.09116290509700775 2023-01-23 17:37:23.065403: step: 1130/1541, loss: 0.025994136929512024 2023-01-23 17:37:23.945106: step: 1132/1541, loss: 0.0001706887996988371 2023-01-23 17:37:24.861007: step: 1134/1541, loss: 0.1139926016330719 2023-01-23 17:37:25.731311: step: 1136/1541, loss: 0.000276528880931437 2023-01-23 17:37:26.636748: step: 1138/1541, loss: 0.042101070284843445 2023-01-23 17:37:27.557356: step: 1140/1541, loss: 0.18758061528205872 2023-01-23 17:37:28.405029: step: 1142/1541, loss: 0.005478256847709417 2023-01-23 17:37:29.311059: step: 1144/1541, loss: 0.9374191164970398 2023-01-23 17:37:30.248320: step: 1146/1541, loss: 0.3433851897716522 2023-01-23 17:37:31.191858: step: 1148/1541, loss: 0.3374623656272888 2023-01-23 17:37:32.087913: step: 1150/1541, loss: 0.25735345482826233 2023-01-23 17:37:33.009500: step: 1152/1541, loss: 0.129957914352417 2023-01-23 17:37:33.957993: step: 1154/1541, loss: 0.010853199288249016 2023-01-23 17:37:34.880873: step: 1156/1541, loss: 0.03038831613957882 2023-01-23 17:37:35.769556: step: 1158/1541, loss: 0.0006213272572495043 2023-01-23 17:37:36.624953: step: 1160/1541, loss: 0.012904858216643333 2023-01-23 17:37:37.512232: step: 1162/1541, loss: 0.00041324549238197505 2023-01-23 17:37:38.397248: step: 1164/1541, loss: 0.0005497575039044023 2023-01-23 17:37:39.322935: step: 1166/1541, loss: 0.028824057430028915 2023-01-23 17:37:40.248834: step: 1168/1541, loss: 0.0023419149219989777 2023-01-23 17:37:41.144598: step: 1170/1541, loss: 0.13535985350608826 2023-01-23 17:37:42.084061: step: 1172/1541, loss: 0.0011963291326537728 2023-01-23 17:37:42.940771: step: 1174/1541, loss: 0.003149135271087289 2023-01-23 17:37:43.810049: step: 1176/1541, loss: 0.0011900125537067652 2023-01-23 17:37:44.724095: step: 1178/1541, loss: 0.07227448374032974 2023-01-23 17:37:45.601062: step: 1180/1541, loss: 0.00012709182919934392 2023-01-23 17:37:46.496030: step: 1182/1541, loss: 0.002067083027213812 2023-01-23 17:37:47.399622: step: 1184/1541, loss: 0.017918474972248077 2023-01-23 17:37:48.228670: step: 1186/1541, loss: 0.0005240959580987692 2023-01-23 17:37:49.091085: step: 1188/1541, loss: 0.0028723881114274263 2023-01-23 17:37:49.987201: step: 1190/1541, loss: 0.005076469387859106 2023-01-23 17:37:50.877069: step: 1192/1541, loss: 0.455203652381897 2023-01-23 17:37:51.765637: step: 1194/1541, loss: 0.011217300780117512 2023-01-23 17:37:52.651909: step: 1196/1541, loss: 0.0002720402553677559 2023-01-23 17:37:53.554498: step: 1198/1541, loss: 0.0005237803561612964 2023-01-23 17:37:54.470930: step: 1200/1541, loss: 0.015439396724104881 2023-01-23 17:37:55.348421: step: 1202/1541, loss: 0.02727161906659603 2023-01-23 17:37:56.247528: step: 1204/1541, loss: 0.014318503439426422 2023-01-23 17:37:57.136602: step: 1206/1541, loss: 0.0014034794876351953 2023-01-23 17:37:58.123560: step: 1208/1541, loss: 0.00012055748811690137 2023-01-23 17:37:59.011618: step: 1210/1541, loss: 0.01877453923225403 2023-01-23 17:37:59.895837: step: 1212/1541, loss: 0.06708429008722305 2023-01-23 17:38:00.766732: step: 1214/1541, loss: 0.0011850794544443488 2023-01-23 17:38:01.655465: step: 1216/1541, loss: 0.002164227422326803 2023-01-23 17:38:02.571155: step: 1218/1541, loss: 0.004348460119217634 2023-01-23 17:38:03.458181: step: 1220/1541, loss: 0.42885395884513855 2023-01-23 17:38:04.329679: step: 1222/1541, loss: 0.02184230089187622 2023-01-23 17:38:05.284138: step: 1224/1541, loss: 0.003947252407670021 2023-01-23 17:38:06.143680: step: 1226/1541, loss: 0.050772011280059814 2023-01-23 17:38:07.072332: step: 1228/1541, loss: 0.012437902390956879 2023-01-23 17:38:07.979708: step: 1230/1541, loss: 0.014536192640662193 2023-01-23 17:38:08.878878: step: 1232/1541, loss: 0.01972973719239235 2023-01-23 17:38:09.790012: step: 1234/1541, loss: 0.03777521476149559 2023-01-23 17:38:10.704877: step: 1236/1541, loss: 0.0035366613883525133 2023-01-23 17:38:11.610267: step: 1238/1541, loss: 0.00782732293009758 2023-01-23 17:38:12.570959: step: 1240/1541, loss: 0.018998440355062485 2023-01-23 17:38:13.554610: step: 1242/1541, loss: 0.007404295727610588 2023-01-23 17:38:14.446172: step: 1244/1541, loss: 0.00047237263061106205 2023-01-23 17:38:15.320916: step: 1246/1541, loss: 0.0013592271134257317 2023-01-23 17:38:16.201054: step: 1248/1541, loss: 4.477964830584824e-05 2023-01-23 17:38:17.079881: step: 1250/1541, loss: 0.07886794954538345 2023-01-23 17:38:17.948959: step: 1252/1541, loss: 0.0003707635623868555 2023-01-23 17:38:18.830683: step: 1254/1541, loss: 0.12192513793706894 2023-01-23 17:38:19.732640: step: 1256/1541, loss: 0.34195300936698914 2023-01-23 17:38:20.721567: step: 1258/1541, loss: 0.28095442056655884 2023-01-23 17:38:21.649414: step: 1260/1541, loss: 0.006172524765133858 2023-01-23 17:38:22.510759: step: 1262/1541, loss: 0.13946473598480225 2023-01-23 17:38:23.485106: step: 1264/1541, loss: 0.2283753603696823 2023-01-23 17:38:24.493751: step: 1266/1541, loss: 0.0005042488337494433 2023-01-23 17:38:25.391671: step: 1268/1541, loss: 0.0035675521939992905 2023-01-23 17:38:26.298037: step: 1270/1541, loss: 1.158466100692749 2023-01-23 17:38:27.210506: step: 1272/1541, loss: 0.0034867615904659033 2023-01-23 17:38:28.088138: step: 1274/1541, loss: 0.00010137429489986971 2023-01-23 17:38:28.912688: step: 1276/1541, loss: 0.03624419495463371 2023-01-23 17:38:29.843452: step: 1278/1541, loss: 0.0005133129889145494 2023-01-23 17:38:30.708095: step: 1280/1541, loss: 0.0018511428497731686 2023-01-23 17:38:31.562591: step: 1282/1541, loss: 0.0012511075474321842 2023-01-23 17:38:32.492949: step: 1284/1541, loss: 0.01223923359066248 2023-01-23 17:38:33.344267: step: 1286/1541, loss: 0.0012959259329363704 2023-01-23 17:38:34.240308: step: 1288/1541, loss: 0.02417759783565998 2023-01-23 17:38:35.105207: step: 1290/1541, loss: 0.11765077710151672 2023-01-23 17:38:36.042102: step: 1292/1541, loss: 0.04882558807730675 2023-01-23 17:38:36.962605: step: 1294/1541, loss: 0.019893454387784004 2023-01-23 17:38:37.817362: step: 1296/1541, loss: 0.026350609958171844 2023-01-23 17:38:38.683880: step: 1298/1541, loss: 0.042041100561618805 2023-01-23 17:38:39.642501: step: 1300/1541, loss: 0.00035453171585686505 2023-01-23 17:38:40.531949: step: 1302/1541, loss: 0.0012883740710094571 2023-01-23 17:38:41.448558: step: 1304/1541, loss: 0.003943231888115406 2023-01-23 17:38:42.357118: step: 1306/1541, loss: 0.6399024724960327 2023-01-23 17:38:43.285012: step: 1308/1541, loss: 0.028799433261156082 2023-01-23 17:38:44.213045: step: 1310/1541, loss: 0.023457298055291176 2023-01-23 17:38:45.113935: step: 1312/1541, loss: 0.03463777154684067 2023-01-23 17:38:46.073346: step: 1314/1541, loss: 0.015345285646617413 2023-01-23 17:38:46.942724: step: 1316/1541, loss: 0.09076423943042755 2023-01-23 17:38:47.806895: step: 1318/1541, loss: 0.0027762041427195072 2023-01-23 17:38:48.659809: step: 1320/1541, loss: 0.001420388463884592 2023-01-23 17:38:49.539770: step: 1322/1541, loss: 0.0017800919013097882 2023-01-23 17:38:50.534540: step: 1324/1541, loss: 0.10159193724393845 2023-01-23 17:38:51.544435: step: 1326/1541, loss: 0.001884405966848135 2023-01-23 17:38:52.462027: step: 1328/1541, loss: 0.023493383079767227 2023-01-23 17:38:53.344021: step: 1330/1541, loss: 0.43954214453697205 2023-01-23 17:38:54.243965: step: 1332/1541, loss: 0.07759658992290497 2023-01-23 17:38:55.125617: step: 1334/1541, loss: 0.007739721797406673 2023-01-23 17:38:55.994758: step: 1336/1541, loss: 0.1662352830171585 2023-01-23 17:38:56.889873: step: 1338/1541, loss: 0.003309226129204035 2023-01-23 17:38:57.755804: step: 1340/1541, loss: 0.0019117231713607907 2023-01-23 17:38:58.722533: step: 1342/1541, loss: 0.14744292199611664 2023-01-23 17:38:59.603636: step: 1344/1541, loss: 0.002951147733256221 2023-01-23 17:39:00.593004: step: 1346/1541, loss: 0.11522242426872253 2023-01-23 17:39:01.544177: step: 1348/1541, loss: 0.004112422466278076 2023-01-23 17:39:02.428274: step: 1350/1541, loss: 0.006685703992843628 2023-01-23 17:39:03.300605: step: 1352/1541, loss: 0.13875968754291534 2023-01-23 17:39:04.269744: step: 1354/1541, loss: 0.48607388138771057 2023-01-23 17:39:05.183885: step: 1356/1541, loss: 0.1267903596162796 2023-01-23 17:39:06.089542: step: 1358/1541, loss: 0.01274242252111435 2023-01-23 17:39:06.950033: step: 1360/1541, loss: 0.006445291917771101 2023-01-23 17:39:07.886592: step: 1362/1541, loss: 0.23288460075855255 2023-01-23 17:39:08.809535: step: 1364/1541, loss: 0.001080637681297958 2023-01-23 17:39:09.736944: step: 1366/1541, loss: 0.029429348185658455 2023-01-23 17:39:10.626658: step: 1368/1541, loss: 0.01645708456635475 2023-01-23 17:39:11.517109: step: 1370/1541, loss: 0.011121841147542 2023-01-23 17:39:12.419372: step: 1372/1541, loss: 0.028908174484968185 2023-01-23 17:39:13.288726: step: 1374/1541, loss: 0.010859576985239983 2023-01-23 17:39:14.192794: step: 1376/1541, loss: 0.015499457716941833 2023-01-23 17:39:15.120441: step: 1378/1541, loss: 0.5343545079231262 2023-01-23 17:39:16.021336: step: 1380/1541, loss: 0.0011661341413855553 2023-01-23 17:39:16.907470: step: 1382/1541, loss: 0.0010874569416046143 2023-01-23 17:39:17.849978: step: 1384/1541, loss: 0.17551998794078827 2023-01-23 17:39:18.762747: step: 1386/1541, loss: 0.0001410218101227656 2023-01-23 17:39:19.713068: step: 1388/1541, loss: 0.0001761933817761019 2023-01-23 17:39:20.611469: step: 1390/1541, loss: 0.00522350100800395 2023-01-23 17:39:21.455258: step: 1392/1541, loss: 0.5185355544090271 2023-01-23 17:39:22.391338: step: 1394/1541, loss: 1.0432049036026 2023-01-23 17:39:23.318216: step: 1396/1541, loss: 0.2605507969856262 2023-01-23 17:39:24.198785: step: 1398/1541, loss: 0.062231022864580154 2023-01-23 17:39:25.188224: step: 1400/1541, loss: 0.1453421413898468 2023-01-23 17:39:26.091545: step: 1402/1541, loss: 0.2492469698190689 2023-01-23 17:39:26.970122: step: 1404/1541, loss: 0.012593033723533154 2023-01-23 17:39:27.873620: step: 1406/1541, loss: 0.0051287938840687275 2023-01-23 17:39:28.737141: step: 1408/1541, loss: 0.0003613104345276952 2023-01-23 17:39:29.633860: step: 1410/1541, loss: 0.0015544164925813675 2023-01-23 17:39:30.517839: step: 1412/1541, loss: 0.0007111758459359407 2023-01-23 17:39:31.397415: step: 1414/1541, loss: 0.00285525294020772 2023-01-23 17:39:32.284753: step: 1416/1541, loss: 0.02768578939139843 2023-01-23 17:39:33.164386: step: 1418/1541, loss: 0.003632889362052083 2023-01-23 17:39:34.038961: step: 1420/1541, loss: 0.00017535535153001547 2023-01-23 17:39:34.906461: step: 1422/1541, loss: 0.00387173960916698 2023-01-23 17:39:35.764997: step: 1424/1541, loss: 0.0178569033741951 2023-01-23 17:39:36.687785: step: 1426/1541, loss: 0.2836681604385376 2023-01-23 17:39:37.607864: step: 1428/1541, loss: 0.047084741294384 2023-01-23 17:39:38.495164: step: 1430/1541, loss: 0.00011828159040305763 2023-01-23 17:39:39.376082: step: 1432/1541, loss: 0.5040891170501709 2023-01-23 17:39:40.291042: step: 1434/1541, loss: 0.004185100551694632 2023-01-23 17:39:41.199073: step: 1436/1541, loss: 0.011199750937521458 2023-01-23 17:39:42.139229: step: 1438/1541, loss: 0.08812855184078217 2023-01-23 17:39:43.047644: step: 1440/1541, loss: 0.0016668556490913033 2023-01-23 17:39:43.900798: step: 1442/1541, loss: 0.0008385778055526316 2023-01-23 17:39:44.710633: step: 1444/1541, loss: 0.0026822732761502266 2023-01-23 17:39:45.626302: step: 1446/1541, loss: 0.03902897238731384 2023-01-23 17:39:46.545498: step: 1448/1541, loss: 0.0037355702370405197 2023-01-23 17:39:47.471639: step: 1450/1541, loss: 0.5323036909103394 2023-01-23 17:39:48.362432: step: 1452/1541, loss: 0.0013641641708090901 2023-01-23 17:39:49.257889: step: 1454/1541, loss: 0.07272741943597794 2023-01-23 17:39:50.130644: step: 1456/1541, loss: 0.001297790091484785 2023-01-23 17:39:51.089242: step: 1458/1541, loss: 0.07799084484577179 2023-01-23 17:39:51.949712: step: 1460/1541, loss: 0.0005478973034769297 2023-01-23 17:39:52.861336: step: 1462/1541, loss: 0.05056750401854515 2023-01-23 17:39:53.730315: step: 1464/1541, loss: 0.00015209645789582282 2023-01-23 17:39:54.654125: step: 1466/1541, loss: 0.0007873381255194545 2023-01-23 17:39:55.583575: step: 1468/1541, loss: 0.01348490547388792 2023-01-23 17:39:56.487208: step: 1470/1541, loss: 0.0006142326164990664 2023-01-23 17:39:57.390015: step: 1472/1541, loss: 0.019956102594733238 2023-01-23 17:39:58.283990: step: 1474/1541, loss: 0.0005618563154712319 2023-01-23 17:39:59.294009: step: 1476/1541, loss: 0.1602841168642044 2023-01-23 17:40:00.207510: step: 1478/1541, loss: 0.002987489104270935 2023-01-23 17:40:01.130993: step: 1480/1541, loss: 0.002579920692369342 2023-01-23 17:40:02.014471: step: 1482/1541, loss: 0.2744229733943939 2023-01-23 17:40:02.895347: step: 1484/1541, loss: 0.05449976772069931 2023-01-23 17:40:03.792092: step: 1486/1541, loss: 0.029777981340885162 2023-01-23 17:40:04.686939: step: 1488/1541, loss: 0.000503272342029959 2023-01-23 17:40:05.556590: step: 1490/1541, loss: 0.02343248762190342 2023-01-23 17:40:06.463315: step: 1492/1541, loss: 0.0011292058043181896 2023-01-23 17:40:07.422527: step: 1494/1541, loss: 0.006944969762116671 2023-01-23 17:40:08.337560: step: 1496/1541, loss: 0.028724655508995056 2023-01-23 17:40:09.235697: step: 1498/1541, loss: 0.0005180565058253706 2023-01-23 17:40:10.177682: step: 1500/1541, loss: 0.31250321865081787 2023-01-23 17:40:11.074536: step: 1502/1541, loss: 0.20015867054462433 2023-01-23 17:40:11.942713: step: 1504/1541, loss: 0.0002950712514575571 2023-01-23 17:40:12.841840: step: 1506/1541, loss: 0.016816068440675735 2023-01-23 17:40:13.734252: step: 1508/1541, loss: 0.0008035619975998998 2023-01-23 17:40:14.669284: step: 1510/1541, loss: 0.01315980777144432 2023-01-23 17:40:15.589225: step: 1512/1541, loss: 0.0704517588019371 2023-01-23 17:40:16.500026: step: 1514/1541, loss: 0.17534960806369781 2023-01-23 17:40:17.387361: step: 1516/1541, loss: 0.04424697905778885 2023-01-23 17:40:18.278182: step: 1518/1541, loss: 0.00040713814087212086 2023-01-23 17:40:19.132281: step: 1520/1541, loss: 0.001564292935654521 2023-01-23 17:40:19.979900: step: 1522/1541, loss: 0.013735909014940262 2023-01-23 17:40:20.922079: step: 1524/1541, loss: 0.01891440711915493 2023-01-23 17:40:21.792146: step: 1526/1541, loss: 0.0008841300150379539 2023-01-23 17:40:22.689882: step: 1528/1541, loss: 0.03706826642155647 2023-01-23 17:40:23.601182: step: 1530/1541, loss: 0.010747138410806656 2023-01-23 17:40:24.469387: step: 1532/1541, loss: 0.017534848302602768 2023-01-23 17:40:25.335361: step: 1534/1541, loss: 0.1030116081237793 2023-01-23 17:40:26.226921: step: 1536/1541, loss: 0.3871493339538574 2023-01-23 17:40:27.117680: step: 1538/1541, loss: 0.006960293278098106 2023-01-23 17:40:28.018476: step: 1540/1541, loss: 0.0005673357518389821 2023-01-23 17:40:28.927057: step: 1542/1541, loss: 0.0007414433639496565 2023-01-23 17:40:29.798686: step: 1544/1541, loss: 0.003103084396570921 2023-01-23 17:40:30.735979: step: 1546/1541, loss: 0.04006018117070198 2023-01-23 17:40:31.623792: step: 1548/1541, loss: 0.024340659379959106 2023-01-23 17:40:32.548235: step: 1550/1541, loss: 0.11917268484830856 2023-01-23 17:40:33.429741: step: 1552/1541, loss: 0.05770478397607803 2023-01-23 17:40:34.365435: step: 1554/1541, loss: 0.0017892515752464533 2023-01-23 17:40:35.294392: step: 1556/1541, loss: 0.0006240211660042405 2023-01-23 17:40:36.176015: step: 1558/1541, loss: 0.0008049341849982738 2023-01-23 17:40:37.037433: step: 1560/1541, loss: 0.008596287108957767 2023-01-23 17:40:37.913045: step: 1562/1541, loss: 0.0019023652421310544 2023-01-23 17:40:38.790676: step: 1564/1541, loss: 0.0004392931587062776 2023-01-23 17:40:39.689505: step: 1566/1541, loss: 0.009041477926075459 2023-01-23 17:40:40.669256: step: 1568/1541, loss: 0.09003575146198273 2023-01-23 17:40:41.587317: step: 1570/1541, loss: 0.07375357300043106 2023-01-23 17:40:42.537189: step: 1572/1541, loss: 0.0010883822105824947 2023-01-23 17:40:43.424707: step: 1574/1541, loss: 0.016794400289654732 2023-01-23 17:40:44.417139: step: 1576/1541, loss: 0.027224330231547356 2023-01-23 17:40:45.400004: step: 1578/1541, loss: 0.00022228533634915948 2023-01-23 17:40:46.252797: step: 1580/1541, loss: 0.02126045897603035 2023-01-23 17:40:47.139262: step: 1582/1541, loss: 0.00026598048862069845 2023-01-23 17:40:48.040258: step: 1584/1541, loss: 0.0007210051990114152 2023-01-23 17:40:48.980814: step: 1586/1541, loss: 0.002713924739509821 2023-01-23 17:40:49.836189: step: 1588/1541, loss: 0.06416664272546768 2023-01-23 17:40:50.719159: step: 1590/1541, loss: 0.006250117905437946 2023-01-23 17:40:51.588823: step: 1592/1541, loss: 0.001514172530733049 2023-01-23 17:40:52.463692: step: 1594/1541, loss: 0.0008795084431767464 2023-01-23 17:40:53.352429: step: 1596/1541, loss: 0.20144973695278168 2023-01-23 17:40:54.229211: step: 1598/1541, loss: 0.0523979589343071 2023-01-23 17:40:55.146177: step: 1600/1541, loss: 9.304546983912587e-05 2023-01-23 17:40:56.002764: step: 1602/1541, loss: 0.07361829280853271 2023-01-23 17:40:56.889857: step: 1604/1541, loss: 0.2512845993041992 2023-01-23 17:40:57.795320: step: 1606/1541, loss: 0.014544224366545677 2023-01-23 17:40:58.670416: step: 1608/1541, loss: 0.0037321061827242374 2023-01-23 17:40:59.559130: step: 1610/1541, loss: 0.00041925665573216975 2023-01-23 17:41:00.439705: step: 1612/1541, loss: 5.1460479880915955e-05 2023-01-23 17:41:01.309602: step: 1614/1541, loss: 0.00015020646969787776 2023-01-23 17:41:02.205412: step: 1616/1541, loss: 0.15346159040927887 2023-01-23 17:41:03.065577: step: 1618/1541, loss: 0.00011608003114815801 2023-01-23 17:41:04.068836: step: 1620/1541, loss: 0.05053433030843735 2023-01-23 17:41:05.006787: step: 1622/1541, loss: 0.03024410270154476 2023-01-23 17:41:05.921248: step: 1624/1541, loss: 0.000577528786379844 2023-01-23 17:41:06.850230: step: 1626/1541, loss: 0.0003061549796257168 2023-01-23 17:41:07.813402: step: 1628/1541, loss: 0.0007806341745890677 2023-01-23 17:41:08.744470: step: 1630/1541, loss: 0.03187025710940361 2023-01-23 17:41:09.580013: step: 1632/1541, loss: 0.0769181028008461 2023-01-23 17:41:10.557489: step: 1634/1541, loss: 0.00484620314091444 2023-01-23 17:41:11.451508: step: 1636/1541, loss: 0.01448018942028284 2023-01-23 17:41:12.326382: step: 1638/1541, loss: 0.0020188656635582447 2023-01-23 17:41:13.251069: step: 1640/1541, loss: 0.00025237660156562924 2023-01-23 17:41:14.131168: step: 1642/1541, loss: 0.00015572734992019832 2023-01-23 17:41:15.048945: step: 1644/1541, loss: 0.0927693098783493 2023-01-23 17:41:15.933430: step: 1646/1541, loss: 0.21207676827907562 2023-01-23 17:41:16.796711: step: 1648/1541, loss: 0.0014731001574546099 2023-01-23 17:41:17.722237: step: 1650/1541, loss: 0.0014344645896926522 2023-01-23 17:41:18.623849: step: 1652/1541, loss: 0.07592194527387619 2023-01-23 17:41:19.535145: step: 1654/1541, loss: 0.0004776694404426962 2023-01-23 17:41:20.461337: step: 1656/1541, loss: 0.23806330561637878 2023-01-23 17:41:21.333559: step: 1658/1541, loss: 0.11018406599760056 2023-01-23 17:41:22.234448: step: 1660/1541, loss: 0.005775084253400564 2023-01-23 17:41:23.162400: step: 1662/1541, loss: 0.00045961729483678937 2023-01-23 17:41:24.049842: step: 1664/1541, loss: 0.04046996682882309 2023-01-23 17:41:24.974289: step: 1666/1541, loss: 0.0009783967398107052 2023-01-23 17:41:25.866018: step: 1668/1541, loss: 0.08311254531145096 2023-01-23 17:41:26.794832: step: 1670/1541, loss: 0.0005713864229619503 2023-01-23 17:41:27.648986: step: 1672/1541, loss: 0.0019383059116080403 2023-01-23 17:41:28.514964: step: 1674/1541, loss: 0.04185371845960617 2023-01-23 17:41:29.406756: step: 1676/1541, loss: 0.0012948918156325817 2023-01-23 17:41:30.370232: step: 1678/1541, loss: 0.002256861887872219 2023-01-23 17:41:31.309499: step: 1680/1541, loss: 0.0006920630694366992 2023-01-23 17:41:32.232651: step: 1682/1541, loss: 0.022650253027677536 2023-01-23 17:41:33.121228: step: 1684/1541, loss: 0.0011432173196226358 2023-01-23 17:41:33.976979: step: 1686/1541, loss: 1.3355189561843872 2023-01-23 17:41:34.895954: step: 1688/1541, loss: 0.7860860228538513 2023-01-23 17:41:35.862599: step: 1690/1541, loss: 0.01740727759897709 2023-01-23 17:41:36.825217: step: 1692/1541, loss: 0.0017978789983317256 2023-01-23 17:41:37.658311: step: 1694/1541, loss: 0.0009323915583081543 2023-01-23 17:41:38.548076: step: 1696/1541, loss: 0.023018531501293182 2023-01-23 17:41:39.442870: step: 1698/1541, loss: 0.057402633130550385 2023-01-23 17:41:40.339196: step: 1700/1541, loss: 0.0010157356737181544 2023-01-23 17:41:41.228357: step: 1702/1541, loss: 0.02420433796942234 2023-01-23 17:41:42.176231: step: 1704/1541, loss: 0.013014714233577251 2023-01-23 17:41:43.115494: step: 1706/1541, loss: 0.016469385474920273 2023-01-23 17:41:44.030337: step: 1708/1541, loss: 0.0010402093175798655 2023-01-23 17:41:44.953918: step: 1710/1541, loss: 0.00017267806106247008 2023-01-23 17:41:45.853038: step: 1712/1541, loss: 0.21967096626758575 2023-01-23 17:41:46.765572: step: 1714/1541, loss: 0.13346238434314728 2023-01-23 17:41:47.646131: step: 1716/1541, loss: 0.005509517155587673 2023-01-23 17:41:48.542669: step: 1718/1541, loss: 0.14249691367149353 2023-01-23 17:41:49.431010: step: 1720/1541, loss: 0.0024343091063201427 2023-01-23 17:41:50.306261: step: 1722/1541, loss: 0.007272918242961168 2023-01-23 17:41:51.231488: step: 1724/1541, loss: 0.004485410172492266 2023-01-23 17:41:52.089016: step: 1726/1541, loss: 0.14968359470367432 2023-01-23 17:41:52.968010: step: 1728/1541, loss: 0.007375416811555624 2023-01-23 17:41:53.849223: step: 1730/1541, loss: 0.0009274188196286559 2023-01-23 17:41:54.756784: step: 1732/1541, loss: 0.036867570132017136 2023-01-23 17:41:55.644481: step: 1734/1541, loss: 0.0708356574177742 2023-01-23 17:41:56.527631: step: 1736/1541, loss: 0.003203450469300151 2023-01-23 17:41:57.375097: step: 1738/1541, loss: 0.1943860799074173 2023-01-23 17:41:58.265696: step: 1740/1541, loss: 0.0014469083398580551 2023-01-23 17:41:59.186309: step: 1742/1541, loss: 0.0005969650228507817 2023-01-23 17:42:00.108863: step: 1744/1541, loss: 0.0009352375636808574 2023-01-23 17:42:01.008337: step: 1746/1541, loss: 0.10110455006361008 2023-01-23 17:42:01.956340: step: 1748/1541, loss: 0.13105930387973785 2023-01-23 17:42:02.845157: step: 1750/1541, loss: 0.0002803568495437503 2023-01-23 17:42:03.746170: step: 1752/1541, loss: 0.04994764178991318 2023-01-23 17:42:04.616579: step: 1754/1541, loss: 0.026865394786000252 2023-01-23 17:42:05.476941: step: 1756/1541, loss: 0.00013912063150200993 2023-01-23 17:42:06.315678: step: 1758/1541, loss: 0.07375282794237137 2023-01-23 17:42:07.322859: step: 1760/1541, loss: 0.00027636217419058084 2023-01-23 17:42:08.247844: step: 1762/1541, loss: 0.0030026433523744345 2023-01-23 17:42:09.131229: step: 1764/1541, loss: 0.2528144121170044 2023-01-23 17:42:10.016388: step: 1766/1541, loss: 0.001834280788898468 2023-01-23 17:42:10.884249: step: 1768/1541, loss: 0.0002549690252635628 2023-01-23 17:42:11.767509: step: 1770/1541, loss: 0.06515076756477356 2023-01-23 17:42:12.736123: step: 1772/1541, loss: 0.20075727999210358 2023-01-23 17:42:13.646194: step: 1774/1541, loss: 0.00428284564986825 2023-01-23 17:42:14.563910: step: 1776/1541, loss: 0.4392217695713043 2023-01-23 17:42:15.471363: step: 1778/1541, loss: 0.007963289506733418 2023-01-23 17:42:16.389210: step: 1780/1541, loss: 0.1717434972524643 2023-01-23 17:42:17.377285: step: 1782/1541, loss: 0.0038683924358338118 2023-01-23 17:42:18.377524: step: 1784/1541, loss: 0.005273730959743261 2023-01-23 17:42:19.357111: step: 1786/1541, loss: 0.005751175805926323 2023-01-23 17:42:20.201990: step: 1788/1541, loss: 0.011530866846442223 2023-01-23 17:42:21.088899: step: 1790/1541, loss: 0.024094006046652794 2023-01-23 17:42:21.947448: step: 1792/1541, loss: 0.033687666058540344 2023-01-23 17:42:22.833039: step: 1794/1541, loss: 0.004124343860894442 2023-01-23 17:42:23.787659: step: 1796/1541, loss: 0.033417463302612305 2023-01-23 17:42:24.734217: step: 1798/1541, loss: 0.06277541071176529 2023-01-23 17:42:25.594620: step: 1800/1541, loss: 0.008069654926657677 2023-01-23 17:42:26.471109: step: 1802/1541, loss: 0.00028559297788888216 2023-01-23 17:42:27.310642: step: 1804/1541, loss: 0.0074972729198634624 2023-01-23 17:42:28.165134: step: 1806/1541, loss: 0.0016290958737954497 2023-01-23 17:42:29.045271: step: 1808/1541, loss: 0.009297419339418411 2023-01-23 17:42:29.964591: step: 1810/1541, loss: 0.15189746022224426 2023-01-23 17:42:30.821326: step: 1812/1541, loss: 0.0008468716987408698 2023-01-23 17:42:31.726550: step: 1814/1541, loss: 0.16837379336357117 2023-01-23 17:42:32.644877: step: 1816/1541, loss: 0.009922757744789124 2023-01-23 17:42:33.520176: step: 1818/1541, loss: 0.0936301201581955 2023-01-23 17:42:34.416018: step: 1820/1541, loss: 0.030554167926311493 2023-01-23 17:42:35.281708: step: 1822/1541, loss: 0.003403231268748641 2023-01-23 17:42:36.182766: step: 1824/1541, loss: 0.0007437534513883293 2023-01-23 17:42:37.128119: step: 1826/1541, loss: 0.00023358376347459853 2023-01-23 17:42:38.014567: step: 1828/1541, loss: 0.14435960352420807 2023-01-23 17:42:38.888126: step: 1830/1541, loss: 0.037617217749357224 2023-01-23 17:42:39.806332: step: 1832/1541, loss: 0.0010099451756104827 2023-01-23 17:42:40.735924: step: 1834/1541, loss: 0.0031219872180372477 2023-01-23 17:42:41.674849: step: 1836/1541, loss: 0.25698554515838623 2023-01-23 17:42:42.547827: step: 1838/1541, loss: 0.007989700883626938 2023-01-23 17:42:43.422008: step: 1840/1541, loss: 0.027271956205368042 2023-01-23 17:42:44.294495: step: 1842/1541, loss: 0.04096487537026405 2023-01-23 17:42:45.210972: step: 1844/1541, loss: 0.0038445398677140474 2023-01-23 17:42:46.100597: step: 1846/1541, loss: 0.015538625419139862 2023-01-23 17:42:47.011911: step: 1848/1541, loss: 0.05112683027982712 2023-01-23 17:42:48.018420: step: 1850/1541, loss: 0.001240154029801488 2023-01-23 17:42:48.926895: step: 1852/1541, loss: 0.02529461495578289 2023-01-23 17:42:49.902298: step: 1854/1541, loss: 0.13167701661586761 2023-01-23 17:42:50.842612: step: 1856/1541, loss: 0.1313573718070984 2023-01-23 17:42:51.756845: step: 1858/1541, loss: 0.001513928291387856 2023-01-23 17:42:52.659511: step: 1860/1541, loss: 0.004827334079891443 2023-01-23 17:42:53.546764: step: 1862/1541, loss: 0.0022019946482032537 2023-01-23 17:42:54.456810: step: 1864/1541, loss: 0.0009293971816077828 2023-01-23 17:42:55.361029: step: 1866/1541, loss: 0.02852283976972103 2023-01-23 17:42:56.245707: step: 1868/1541, loss: 0.00018165665096603334 2023-01-23 17:42:57.165505: step: 1870/1541, loss: 0.16576090455055237 2023-01-23 17:42:58.082435: step: 1872/1541, loss: 0.006624814122915268 2023-01-23 17:42:58.994489: step: 1874/1541, loss: 0.003959638066589832 2023-01-23 17:42:59.850684: step: 1876/1541, loss: 0.003929770085960627 2023-01-23 17:43:00.763399: step: 1878/1541, loss: 0.01185477152466774 2023-01-23 17:43:01.642688: step: 1880/1541, loss: 0.06148187443614006 2023-01-23 17:43:02.585386: step: 1882/1541, loss: 0.0007010999252088368 2023-01-23 17:43:03.473636: step: 1884/1541, loss: 0.2244524210691452 2023-01-23 17:43:04.362585: step: 1886/1541, loss: 0.0013449465623125434 2023-01-23 17:43:05.318675: step: 1888/1541, loss: 0.0011315320152789354 2023-01-23 17:43:06.187998: step: 1890/1541, loss: 0.001103049493394792 2023-01-23 17:43:07.088978: step: 1892/1541, loss: 0.029201213270425797 2023-01-23 17:43:07.999766: step: 1894/1541, loss: 0.053584836423397064 2023-01-23 17:43:08.908611: step: 1896/1541, loss: 2.6409894417156465e-05 2023-01-23 17:43:09.841831: step: 1898/1541, loss: 0.3308333158493042 2023-01-23 17:43:10.704891: step: 1900/1541, loss: 0.001562755205668509 2023-01-23 17:43:11.635875: step: 1902/1541, loss: 0.18030886352062225 2023-01-23 17:43:12.514599: step: 1904/1541, loss: 0.01363530382514 2023-01-23 17:43:13.427847: step: 1906/1541, loss: 0.19370044767856598 2023-01-23 17:43:14.371380: step: 1908/1541, loss: 0.008442853577435017 2023-01-23 17:43:15.279747: step: 1910/1541, loss: 0.004520133137702942 2023-01-23 17:43:16.168203: step: 1912/1541, loss: 0.0026047038845717907 2023-01-23 17:43:17.012021: step: 1914/1541, loss: 0.00829007476568222 2023-01-23 17:43:17.921898: step: 1916/1541, loss: 0.0480266734957695 2023-01-23 17:43:18.781216: step: 1918/1541, loss: 0.021356021985411644 2023-01-23 17:43:19.669498: step: 1920/1541, loss: 0.0003785394364967942 2023-01-23 17:43:20.558543: step: 1922/1541, loss: 0.0011284539941698313 2023-01-23 17:43:21.498504: step: 1924/1541, loss: 0.0008809752762317657 2023-01-23 17:43:22.488123: step: 1926/1541, loss: 0.0064326683059334755 2023-01-23 17:43:23.459672: step: 1928/1541, loss: 0.0031726208981126547 2023-01-23 17:43:24.354850: step: 1930/1541, loss: 0.037767332047224045 2023-01-23 17:43:25.254514: step: 1932/1541, loss: 0.2962922155857086 2023-01-23 17:43:26.170153: step: 1934/1541, loss: 0.5105782747268677 2023-01-23 17:43:27.055604: step: 1936/1541, loss: 0.0013564301189035177 2023-01-23 17:43:27.928865: step: 1938/1541, loss: 0.0008617712301202118 2023-01-23 17:43:28.799066: step: 1940/1541, loss: 0.1337917000055313 2023-01-23 17:43:29.656638: step: 1942/1541, loss: 0.26621830463409424 2023-01-23 17:43:30.552657: step: 1944/1541, loss: 0.15665315091609955 2023-01-23 17:43:31.517284: step: 1946/1541, loss: 0.002625585999339819 2023-01-23 17:43:32.503836: step: 1948/1541, loss: 0.014730115421116352 2023-01-23 17:43:33.369124: step: 1950/1541, loss: 0.05335370451211929 2023-01-23 17:43:34.269320: step: 1952/1541, loss: 0.000587573682423681 2023-01-23 17:43:35.139841: step: 1954/1541, loss: 0.022397711873054504 2023-01-23 17:43:36.063590: step: 1956/1541, loss: 0.00077998818596825 2023-01-23 17:43:36.952407: step: 1958/1541, loss: 0.027656985446810722 2023-01-23 17:43:37.880477: step: 1960/1541, loss: 0.3370313048362732 2023-01-23 17:43:38.765208: step: 1962/1541, loss: 0.008633844554424286 2023-01-23 17:43:39.635198: step: 1964/1541, loss: 0.2895766794681549 2023-01-23 17:43:40.599806: step: 1966/1541, loss: 0.006564954295754433 2023-01-23 17:43:41.530534: step: 1968/1541, loss: 0.3490530848503113 2023-01-23 17:43:42.435458: step: 1970/1541, loss: 0.0011811397271230817 2023-01-23 17:43:43.387332: step: 1972/1541, loss: 0.007501889951527119 2023-01-23 17:43:44.288197: step: 1974/1541, loss: 0.006019266322255135 2023-01-23 17:43:45.156044: step: 1976/1541, loss: 0.005821153521537781 2023-01-23 17:43:46.049579: step: 1978/1541, loss: 8.49132047733292e-05 2023-01-23 17:43:46.978368: step: 1980/1541, loss: 0.11998534947633743 2023-01-23 17:43:47.966157: step: 1982/1541, loss: 0.006566658616065979 2023-01-23 17:43:48.884127: step: 1984/1541, loss: 0.0015514657134190202 2023-01-23 17:43:49.787578: step: 1986/1541, loss: 0.0001961961534107104 2023-01-23 17:43:50.682108: step: 1988/1541, loss: 0.003329658415168524 2023-01-23 17:43:51.559525: step: 1990/1541, loss: 0.1579500138759613 2023-01-23 17:43:52.467333: step: 1992/1541, loss: 0.18378733098506927 2023-01-23 17:43:53.480472: step: 1994/1541, loss: 0.06209946423768997 2023-01-23 17:43:54.360719: step: 1996/1541, loss: 0.020976366475224495 2023-01-23 17:43:55.291149: step: 1998/1541, loss: 0.021746709942817688 2023-01-23 17:43:56.183104: step: 2000/1541, loss: 0.0705597847700119 2023-01-23 17:43:57.177346: step: 2002/1541, loss: 0.021962810307741165 2023-01-23 17:43:58.068593: step: 2004/1541, loss: 7.22107506589964e-05 2023-01-23 17:43:58.994052: step: 2006/1541, loss: 0.0010682319989427924 2023-01-23 17:43:59.937472: step: 2008/1541, loss: 0.4877607226371765 2023-01-23 17:44:00.800278: step: 2010/1541, loss: 9.033361857291311e-05 2023-01-23 17:44:01.658633: step: 2012/1541, loss: 0.0629412978887558 2023-01-23 17:44:02.599094: step: 2014/1541, loss: 0.13795728981494904 2023-01-23 17:44:03.482630: step: 2016/1541, loss: 0.0009662163793109357 2023-01-23 17:44:04.429917: step: 2018/1541, loss: 0.03609991446137428 2023-01-23 17:44:05.319417: step: 2020/1541, loss: 0.0009447860647924244 2023-01-23 17:44:06.263296: step: 2022/1541, loss: 0.00030305268592201173 2023-01-23 17:44:07.169620: step: 2024/1541, loss: 0.06684715300798416 2023-01-23 17:44:08.111025: step: 2026/1541, loss: 0.020833011716604233 2023-01-23 17:44:09.017120: step: 2028/1541, loss: 0.0007087379926815629 2023-01-23 17:44:09.902034: step: 2030/1541, loss: 0.00019171705935150385 2023-01-23 17:44:10.817120: step: 2032/1541, loss: 9.314523049397394e-05 2023-01-23 17:44:11.768819: step: 2034/1541, loss: 0.009125231765210629 2023-01-23 17:44:12.674713: step: 2036/1541, loss: 0.0005415641353465617 2023-01-23 17:44:13.600998: step: 2038/1541, loss: 0.023718081414699554 2023-01-23 17:44:14.494784: step: 2040/1541, loss: 0.023636968806385994 2023-01-23 17:44:15.381132: step: 2042/1541, loss: 0.00048695673467591405 2023-01-23 17:44:16.255975: step: 2044/1541, loss: 0.0011857980862259865 2023-01-23 17:44:17.223425: step: 2046/1541, loss: 0.14694948494434357 2023-01-23 17:44:18.176481: step: 2048/1541, loss: 0.0014070161851122975 2023-01-23 17:44:19.112473: step: 2050/1541, loss: 0.01632433943450451 2023-01-23 17:44:19.979429: step: 2052/1541, loss: 0.00253892270848155 2023-01-23 17:44:20.886874: step: 2054/1541, loss: 0.0016587846912443638 2023-01-23 17:44:21.840568: step: 2056/1541, loss: 0.0015322467079386115 2023-01-23 17:44:22.716429: step: 2058/1541, loss: 0.023957479745149612 2023-01-23 17:44:23.635196: step: 2060/1541, loss: 0.08895127475261688 2023-01-23 17:44:24.559398: step: 2062/1541, loss: 0.20910102128982544 2023-01-23 17:44:25.492312: step: 2064/1541, loss: 0.022211601957678795 2023-01-23 17:44:26.401954: step: 2066/1541, loss: 0.00732091348618269 2023-01-23 17:44:27.259258: step: 2068/1541, loss: 0.015235812403261662 2023-01-23 17:44:28.195238: step: 2070/1541, loss: 0.014107814058661461 2023-01-23 17:44:29.144151: step: 2072/1541, loss: 0.2946197986602783 2023-01-23 17:44:30.169260: step: 2074/1541, loss: 0.34089168906211853 2023-01-23 17:44:31.057975: step: 2076/1541, loss: 0.03596680238842964 2023-01-23 17:44:31.973350: step: 2078/1541, loss: 0.003678149078041315 2023-01-23 17:44:32.848057: step: 2080/1541, loss: 0.0016565878177061677 2023-01-23 17:44:33.758495: step: 2082/1541, loss: 0.0013849488459527493 2023-01-23 17:44:34.655065: step: 2084/1541, loss: 0.013269199058413506 2023-01-23 17:44:35.595050: step: 2086/1541, loss: 1.6678218841552734 2023-01-23 17:44:36.483414: step: 2088/1541, loss: 0.0007702029542997479 2023-01-23 17:44:37.371040: step: 2090/1541, loss: 0.0026123563293367624 2023-01-23 17:44:38.255104: step: 2092/1541, loss: 0.2288728803396225 2023-01-23 17:44:39.126409: step: 2094/1541, loss: 0.02738415077328682 2023-01-23 17:44:40.017133: step: 2096/1541, loss: 0.26055940985679626 2023-01-23 17:44:40.910427: step: 2098/1541, loss: 0.006370556075125933 2023-01-23 17:44:41.807825: step: 2100/1541, loss: 0.01950313150882721 2023-01-23 17:44:42.692576: step: 2102/1541, loss: 0.03604569658637047 2023-01-23 17:44:43.594392: step: 2104/1541, loss: 0.013858089223504066 2023-01-23 17:44:44.502138: step: 2106/1541, loss: 0.013178747147321701 2023-01-23 17:44:45.419547: step: 2108/1541, loss: 0.007846090011298656 2023-01-23 17:44:46.278423: step: 2110/1541, loss: 0.0007903524092398584 2023-01-23 17:44:47.197674: step: 2112/1541, loss: 0.02431618608534336 2023-01-23 17:44:48.121150: step: 2114/1541, loss: 0.001365098636597395 2023-01-23 17:44:49.072658: step: 2116/1541, loss: 0.3312780261039734 2023-01-23 17:44:50.064370: step: 2118/1541, loss: 0.0013011705595999956 2023-01-23 17:44:50.937426: step: 2120/1541, loss: 0.0011430999729782343 2023-01-23 17:44:51.855766: step: 2122/1541, loss: 0.0006396850803866982 2023-01-23 17:44:52.738640: step: 2124/1541, loss: 0.08105860650539398 2023-01-23 17:44:53.649443: step: 2126/1541, loss: 7.311900844797492e-05 2023-01-23 17:44:54.530958: step: 2128/1541, loss: 0.0009358953684568405 2023-01-23 17:44:55.393823: step: 2130/1541, loss: 0.29308614134788513 2023-01-23 17:44:56.284345: step: 2132/1541, loss: 0.008232706226408482 2023-01-23 17:44:57.145028: step: 2134/1541, loss: 0.029702335596084595 2023-01-23 17:44:58.057141: step: 2136/1541, loss: 0.019656797870993614 2023-01-23 17:44:58.972342: step: 2138/1541, loss: 0.010098292492330074 2023-01-23 17:44:59.873619: step: 2140/1541, loss: 0.005227129440754652 2023-01-23 17:45:00.774796: step: 2142/1541, loss: 0.7388927936553955 2023-01-23 17:45:01.720478: step: 2144/1541, loss: 0.15045343339443207 2023-01-23 17:45:02.670672: step: 2146/1541, loss: 0.008703473955392838 2023-01-23 17:45:03.598308: step: 2148/1541, loss: 0.10347095131874084 2023-01-23 17:45:04.480230: step: 2150/1541, loss: 0.003925363998860121 2023-01-23 17:45:05.333261: step: 2152/1541, loss: 0.0005121338181197643 2023-01-23 17:45:06.215458: step: 2154/1541, loss: 0.09043799340724945 2023-01-23 17:45:07.164663: step: 2156/1541, loss: 0.015444732271134853 2023-01-23 17:45:08.136382: step: 2158/1541, loss: 0.010404285043478012 2023-01-23 17:45:08.992267: step: 2160/1541, loss: 0.006623984780162573 2023-01-23 17:45:09.877228: step: 2162/1541, loss: 9.40079044084996e-05 2023-01-23 17:45:10.732043: step: 2164/1541, loss: 0.0011277050944045186 2023-01-23 17:45:11.587273: step: 2166/1541, loss: 0.00016410581883974373 2023-01-23 17:45:12.470673: step: 2168/1541, loss: 0.045334719121456146 2023-01-23 17:45:13.363274: step: 2170/1541, loss: 0.00014331217971630394 2023-01-23 17:45:14.219248: step: 2172/1541, loss: 0.06522634625434875 2023-01-23 17:45:15.100226: step: 2174/1541, loss: 0.38273653388023376 2023-01-23 17:45:16.009522: step: 2176/1541, loss: 2.0011069864267483e-05 2023-01-23 17:45:16.972470: step: 2178/1541, loss: 0.8867189288139343 2023-01-23 17:45:17.927736: step: 2180/1541, loss: 0.9155216217041016 2023-01-23 17:45:18.814309: step: 2182/1541, loss: 0.5367528796195984 2023-01-23 17:45:19.725554: step: 2184/1541, loss: 0.3716152310371399 2023-01-23 17:45:20.616871: step: 2186/1541, loss: 0.1504332572221756 2023-01-23 17:45:21.513194: step: 2188/1541, loss: 0.0002899601822718978 2023-01-23 17:45:22.367599: step: 2190/1541, loss: 0.0015153770800679922 2023-01-23 17:45:23.250467: step: 2192/1541, loss: 9.832709474721923e-05 2023-01-23 17:45:24.133871: step: 2194/1541, loss: 0.010325418785214424 2023-01-23 17:45:25.063879: step: 2196/1541, loss: 1.0614749193191528 2023-01-23 17:45:25.954381: step: 2198/1541, loss: 0.3021203577518463 2023-01-23 17:45:26.855316: step: 2200/1541, loss: 0.3317778408527374 2023-01-23 17:45:27.759995: step: 2202/1541, loss: 0.31998348236083984 2023-01-23 17:45:28.663288: step: 2204/1541, loss: 0.06351961940526962 2023-01-23 17:45:29.562919: step: 2206/1541, loss: 0.4872811734676361 2023-01-23 17:45:30.478629: step: 2208/1541, loss: 0.03795713558793068 2023-01-23 17:45:31.339643: step: 2210/1541, loss: 0.00848053116351366 2023-01-23 17:45:32.234929: step: 2212/1541, loss: 0.0004225042066536844 2023-01-23 17:45:33.104159: step: 2214/1541, loss: 0.005309028550982475 2023-01-23 17:45:33.967781: step: 2216/1541, loss: 0.00034104479709640145 2023-01-23 17:45:34.836759: step: 2218/1541, loss: 0.01802544854581356 2023-01-23 17:45:35.787407: step: 2220/1541, loss: 0.0008310318225994706 2023-01-23 17:45:36.685149: step: 2222/1541, loss: 0.05997087433934212 2023-01-23 17:45:37.637330: step: 2224/1541, loss: 0.0038008028641343117 2023-01-23 17:45:38.594955: step: 2226/1541, loss: 0.0030443351715803146 2023-01-23 17:45:39.509477: step: 2228/1541, loss: 0.1698857843875885 2023-01-23 17:45:40.390740: step: 2230/1541, loss: 0.025660132989287376 2023-01-23 17:45:41.274694: step: 2232/1541, loss: 0.05997781828045845 2023-01-23 17:45:42.227773: step: 2234/1541, loss: 0.06518981605768204 2023-01-23 17:45:43.093882: step: 2236/1541, loss: 0.1620057076215744 2023-01-23 17:45:44.008832: step: 2238/1541, loss: 0.0023858314380049706 2023-01-23 17:45:44.909890: step: 2240/1541, loss: 0.5108574628829956 2023-01-23 17:45:45.802668: step: 2242/1541, loss: 0.017691539600491524 2023-01-23 17:45:46.718955: step: 2244/1541, loss: 0.0010067822877317667 2023-01-23 17:45:47.685946: step: 2246/1541, loss: 0.00013526524708140641 2023-01-23 17:45:48.569465: step: 2248/1541, loss: 0.0037282854318618774 2023-01-23 17:45:49.477370: step: 2250/1541, loss: 0.0005717988824471831 2023-01-23 17:45:50.348298: step: 2252/1541, loss: 0.0009264555992558599 2023-01-23 17:45:51.281899: step: 2254/1541, loss: 0.0014295360306277871 2023-01-23 17:45:52.168383: step: 2256/1541, loss: 0.0003798257384914905 2023-01-23 17:45:53.087289: step: 2258/1541, loss: 0.00957209337502718 2023-01-23 17:45:53.987656: step: 2260/1541, loss: 1.801915168762207 2023-01-23 17:45:54.926042: step: 2262/1541, loss: 0.001561698503792286 2023-01-23 17:45:55.819102: step: 2264/1541, loss: 0.014170670881867409 2023-01-23 17:45:56.750530: step: 2266/1541, loss: 0.16335168480873108 2023-01-23 17:45:57.643287: step: 2268/1541, loss: 0.002921181730926037 2023-01-23 17:45:58.572646: step: 2270/1541, loss: 0.025790337473154068 2023-01-23 17:45:59.481644: step: 2272/1541, loss: 0.17959126830101013 2023-01-23 17:46:00.360654: step: 2274/1541, loss: 0.012478053569793701 2023-01-23 17:46:01.263321: step: 2276/1541, loss: 0.0007444218499585986 2023-01-23 17:46:02.195117: step: 2278/1541, loss: 0.005304368678480387 2023-01-23 17:46:03.118827: step: 2280/1541, loss: 0.002972446382045746 2023-01-23 17:46:04.005476: step: 2282/1541, loss: 2.065239667892456 2023-01-23 17:46:04.890541: step: 2284/1541, loss: 0.02415155991911888 2023-01-23 17:46:05.780783: step: 2286/1541, loss: 0.135675311088562 2023-01-23 17:46:06.680301: step: 2288/1541, loss: 0.01997608318924904 2023-01-23 17:46:07.571161: step: 2290/1541, loss: 0.0031300431583076715 2023-01-23 17:46:08.439356: step: 2292/1541, loss: 0.007519140839576721 2023-01-23 17:46:09.311860: step: 2294/1541, loss: 0.0044660912826657295 2023-01-23 17:46:10.152388: step: 2296/1541, loss: 0.03491087630391121 2023-01-23 17:46:11.033389: step: 2298/1541, loss: 0.01101276371628046 2023-01-23 17:46:11.940662: step: 2300/1541, loss: 0.015165707096457481 2023-01-23 17:46:12.916042: step: 2302/1541, loss: 0.0019520299974828959 2023-01-23 17:46:13.833343: step: 2304/1541, loss: 0.08165812492370605 2023-01-23 17:46:14.771196: step: 2306/1541, loss: 0.07755342125892639 2023-01-23 17:46:15.707954: step: 2308/1541, loss: 0.0008012678590603173 2023-01-23 17:46:16.578110: step: 2310/1541, loss: 0.001126195304095745 2023-01-23 17:46:17.486550: step: 2312/1541, loss: 0.4443376660346985 2023-01-23 17:46:18.336578: step: 2314/1541, loss: 0.0007578245713375509 2023-01-23 17:46:19.169039: step: 2316/1541, loss: 0.0007082755910232663 2023-01-23 17:46:20.090871: step: 2318/1541, loss: 0.07996934652328491 2023-01-23 17:46:21.041521: step: 2320/1541, loss: 0.00019466070807538927 2023-01-23 17:46:21.917465: step: 2322/1541, loss: 0.29540538787841797 2023-01-23 17:46:22.817854: step: 2324/1541, loss: 0.09732549637556076 2023-01-23 17:46:23.704071: step: 2326/1541, loss: 0.004086923785507679 2023-01-23 17:46:24.670217: step: 2328/1541, loss: 0.0024716665502637625 2023-01-23 17:46:25.538798: step: 2330/1541, loss: 3.1959283660398796e-05 2023-01-23 17:46:26.480642: step: 2332/1541, loss: 0.030773799866437912 2023-01-23 17:46:27.413905: step: 2334/1541, loss: 0.04632768779993057 2023-01-23 17:46:28.326651: step: 2336/1541, loss: 0.0002743793302215636 2023-01-23 17:46:29.301678: step: 2338/1541, loss: 0.06331189721822739 2023-01-23 17:46:30.174758: step: 2340/1541, loss: 0.003409450175240636 2023-01-23 17:46:31.066595: step: 2342/1541, loss: 0.0951118990778923 2023-01-23 17:46:31.971364: step: 2344/1541, loss: 0.00041315637645311654 2023-01-23 17:46:32.860197: step: 2346/1541, loss: 0.07151962071657181 2023-01-23 17:46:33.736495: step: 2348/1541, loss: 0.013402975164353848 2023-01-23 17:46:34.634392: step: 2350/1541, loss: 8.548505138605833e-05 2023-01-23 17:46:35.517976: step: 2352/1541, loss: 0.001579717150889337 2023-01-23 17:46:36.417376: step: 2354/1541, loss: 0.039921484887599945 2023-01-23 17:46:37.364611: step: 2356/1541, loss: 0.1913338601589203 2023-01-23 17:46:38.274356: step: 2358/1541, loss: 0.3473508656024933 2023-01-23 17:46:39.180427: step: 2360/1541, loss: 0.060919515788555145 2023-01-23 17:46:40.061935: step: 2362/1541, loss: 0.017802931368350983 2023-01-23 17:46:40.969893: step: 2364/1541, loss: 0.0009048653300851583 2023-01-23 17:46:41.842005: step: 2366/1541, loss: 8.289275137940422e-05 2023-01-23 17:46:42.747271: step: 2368/1541, loss: 0.193874791264534 2023-01-23 17:46:43.610773: step: 2370/1541, loss: 0.5036941766738892 2023-01-23 17:46:44.471641: step: 2372/1541, loss: 0.0008115286473184824 2023-01-23 17:46:45.365254: step: 2374/1541, loss: 0.0011286335065960884 2023-01-23 17:46:46.254614: step: 2376/1541, loss: 0.00042923138244077563 2023-01-23 17:46:47.154476: step: 2378/1541, loss: 0.0001300478761550039 2023-01-23 17:46:48.051519: step: 2380/1541, loss: 0.0045661781914532185 2023-01-23 17:46:48.947820: step: 2382/1541, loss: 0.0012205932289361954 2023-01-23 17:46:49.797752: step: 2384/1541, loss: 0.011434757150709629 2023-01-23 17:46:50.709379: step: 2386/1541, loss: 0.013805046677589417 2023-01-23 17:46:51.598243: step: 2388/1541, loss: 0.0006059985607862473 2023-01-23 17:46:52.501006: step: 2390/1541, loss: 0.04854225739836693 2023-01-23 17:46:53.414713: step: 2392/1541, loss: 0.3287629783153534 2023-01-23 17:46:54.308289: step: 2394/1541, loss: 0.03084406442940235 2023-01-23 17:46:55.282951: step: 2396/1541, loss: 0.051196858286857605 2023-01-23 17:46:56.170070: step: 2398/1541, loss: 0.03555160388350487 2023-01-23 17:46:57.048770: step: 2400/1541, loss: 0.002049949485808611 2023-01-23 17:46:57.927107: step: 2402/1541, loss: 0.0010102081578224897 2023-01-23 17:46:58.830293: step: 2404/1541, loss: 0.0007837697630748153 2023-01-23 17:46:59.733878: step: 2406/1541, loss: 0.1584843546152115 2023-01-23 17:47:00.725282: step: 2408/1541, loss: 0.002256353385746479 2023-01-23 17:47:01.574618: step: 2410/1541, loss: 0.001068792655132711 2023-01-23 17:47:02.591492: step: 2412/1541, loss: 0.0022111604921519756 2023-01-23 17:47:03.568999: step: 2414/1541, loss: 0.00047511811135336757 2023-01-23 17:47:04.474366: step: 2416/1541, loss: 0.04175515100359917 2023-01-23 17:47:05.347025: step: 2418/1541, loss: 0.014165339060127735 2023-01-23 17:47:06.304593: step: 2420/1541, loss: 0.019765619188547134 2023-01-23 17:47:07.195170: step: 2422/1541, loss: 0.00026638800045475364 2023-01-23 17:47:08.073247: step: 2424/1541, loss: 0.001302809570915997 2023-01-23 17:47:08.922845: step: 2426/1541, loss: 0.0029753854032605886 2023-01-23 17:47:09.818955: step: 2428/1541, loss: 0.06713197380304337 2023-01-23 17:47:10.698607: step: 2430/1541, loss: 0.007915513589978218 2023-01-23 17:47:11.673752: step: 2432/1541, loss: 0.03565952926874161 2023-01-23 17:47:12.565779: step: 2434/1541, loss: 0.07431720197200775 2023-01-23 17:47:13.492043: step: 2436/1541, loss: 0.016200922429561615 2023-01-23 17:47:14.391148: step: 2438/1541, loss: 0.011298516765236855 2023-01-23 17:47:15.299752: step: 2440/1541, loss: 0.00794160459190607 2023-01-23 17:47:16.238876: step: 2442/1541, loss: 0.005664667580276728 2023-01-23 17:47:17.155118: step: 2444/1541, loss: 0.0034753368236124516 2023-01-23 17:47:18.042329: step: 2446/1541, loss: 0.0022199631202965975 2023-01-23 17:47:18.955121: step: 2448/1541, loss: 0.31798791885375977 2023-01-23 17:47:19.857174: step: 2450/1541, loss: 0.3236621022224426 2023-01-23 17:47:20.763409: step: 2452/1541, loss: 9.36442447709851e-05 2023-01-23 17:47:21.659086: step: 2454/1541, loss: 0.008535858243703842 2023-01-23 17:47:22.550953: step: 2456/1541, loss: 0.0007174389902502298 2023-01-23 17:47:23.437103: step: 2458/1541, loss: 0.0030092361848801374 2023-01-23 17:47:24.321017: step: 2460/1541, loss: 0.003548250300809741 2023-01-23 17:47:25.241181: step: 2462/1541, loss: 0.0121307997033 2023-01-23 17:47:26.153465: step: 2464/1541, loss: 0.015307697467505932 2023-01-23 17:47:27.023686: step: 2466/1541, loss: 0.2981744110584259 2023-01-23 17:47:27.899388: step: 2468/1541, loss: 0.08175604045391083 2023-01-23 17:47:28.776645: step: 2470/1541, loss: 0.0039045549929142 2023-01-23 17:47:29.710670: step: 2472/1541, loss: 0.002348930574953556 2023-01-23 17:47:30.630580: step: 2474/1541, loss: 0.001028026919811964 2023-01-23 17:47:31.564303: step: 2476/1541, loss: 0.12587134540081024 2023-01-23 17:47:32.454721: step: 2478/1541, loss: 0.3729793131351471 2023-01-23 17:47:33.373810: step: 2480/1541, loss: 0.1979614645242691 2023-01-23 17:47:34.242298: step: 2482/1541, loss: 0.22368036210536957 2023-01-23 17:47:35.098435: step: 2484/1541, loss: 0.034278515726327896 2023-01-23 17:47:35.993088: step: 2486/1541, loss: 0.0020408814307302237 2023-01-23 17:47:36.871700: step: 2488/1541, loss: 0.00046267613652162254 2023-01-23 17:47:37.772765: step: 2490/1541, loss: 0.058724213391542435 2023-01-23 17:47:38.669071: step: 2492/1541, loss: 0.13721053302288055 2023-01-23 17:47:39.552341: step: 2494/1541, loss: 0.008006591349840164 2023-01-23 17:47:40.567096: step: 2496/1541, loss: 0.1318301260471344 2023-01-23 17:47:41.474869: step: 2498/1541, loss: 0.15439742803573608 2023-01-23 17:47:42.371825: step: 2500/1541, loss: 0.006558570545166731 2023-01-23 17:47:43.275637: step: 2502/1541, loss: 0.0042110770009458065 2023-01-23 17:47:44.202526: step: 2504/1541, loss: 0.019173407927155495 2023-01-23 17:47:45.070539: step: 2506/1541, loss: 0.18427152931690216 2023-01-23 17:47:45.954617: step: 2508/1541, loss: 0.1485726684331894 2023-01-23 17:47:46.879618: step: 2510/1541, loss: 0.0927119329571724 2023-01-23 17:47:47.775793: step: 2512/1541, loss: 0.014642526395618916 2023-01-23 17:47:48.805028: step: 2514/1541, loss: 0.16922074556350708 2023-01-23 17:47:49.685623: step: 2516/1541, loss: 0.0011492242338135839 2023-01-23 17:47:50.525540: step: 2518/1541, loss: 0.007077702321112156 2023-01-23 17:47:51.403771: step: 2520/1541, loss: 0.05663745105266571 2023-01-23 17:47:52.306331: step: 2522/1541, loss: 0.004578548017889261 2023-01-23 17:47:53.163288: step: 2524/1541, loss: 0.031148815527558327 2023-01-23 17:47:54.061906: step: 2526/1541, loss: 0.17420759797096252 2023-01-23 17:47:54.941929: step: 2528/1541, loss: 0.05045849084854126 2023-01-23 17:47:55.865173: step: 2530/1541, loss: 0.0022274740040302277 2023-01-23 17:47:56.838451: step: 2532/1541, loss: 0.04863951727747917 2023-01-23 17:47:57.706649: step: 2534/1541, loss: 0.09014235436916351 2023-01-23 17:47:58.602996: step: 2536/1541, loss: 0.0007236662204377353 2023-01-23 17:47:59.482157: step: 2538/1541, loss: 0.02707001194357872 2023-01-23 17:48:00.399850: step: 2540/1541, loss: 0.010495192371308804 2023-01-23 17:48:01.311945: step: 2542/1541, loss: 0.3280615508556366 2023-01-23 17:48:02.229842: step: 2544/1541, loss: 0.0024217788595706224 2023-01-23 17:48:03.100323: step: 2546/1541, loss: 0.0014009812148287892 2023-01-23 17:48:04.129532: step: 2548/1541, loss: 0.005177939776331186 2023-01-23 17:48:04.987996: step: 2550/1541, loss: 0.07682844251394272 2023-01-23 17:48:05.868297: step: 2552/1541, loss: 0.0015734824119135737 2023-01-23 17:48:06.765705: step: 2554/1541, loss: 0.03534873202443123 2023-01-23 17:48:07.661808: step: 2556/1541, loss: 0.002394195180386305 2023-01-23 17:48:08.604054: step: 2558/1541, loss: 0.0041078911162912846 2023-01-23 17:48:09.482255: step: 2560/1541, loss: 0.018208097666502 2023-01-23 17:48:10.413467: step: 2562/1541, loss: 0.01984676532447338 2023-01-23 17:48:11.310580: step: 2564/1541, loss: 0.0011145527241751552 2023-01-23 17:48:12.196685: step: 2566/1541, loss: 0.0014394219033420086 2023-01-23 17:48:13.082149: step: 2568/1541, loss: 0.11374159157276154 2023-01-23 17:48:13.989418: step: 2570/1541, loss: 0.34505271911621094 2023-01-23 17:48:14.897456: step: 2572/1541, loss: 0.028773125261068344 2023-01-23 17:48:15.797011: step: 2574/1541, loss: 0.008449374698102474 2023-01-23 17:48:16.729313: step: 2576/1541, loss: 2.2026824808563106e-05 2023-01-23 17:48:17.604904: step: 2578/1541, loss: 0.584883987903595 2023-01-23 17:48:18.496539: step: 2580/1541, loss: 0.06589766591787338 2023-01-23 17:48:19.385434: step: 2582/1541, loss: 0.03435415402054787 2023-01-23 17:48:20.300642: step: 2584/1541, loss: 0.0168627817183733 2023-01-23 17:48:21.217449: step: 2586/1541, loss: 0.04169579967856407 2023-01-23 17:48:22.097468: step: 2588/1541, loss: 0.019077980890870094 2023-01-23 17:48:22.982451: step: 2590/1541, loss: 0.001744171604514122 2023-01-23 17:48:23.823813: step: 2592/1541, loss: 0.044533900916576385 2023-01-23 17:48:24.714278: step: 2594/1541, loss: 0.0023688827641308308 2023-01-23 17:48:25.621106: step: 2596/1541, loss: 0.0009224996319971979 2023-01-23 17:48:26.502820: step: 2598/1541, loss: 0.0017480008536949754 2023-01-23 17:48:27.355257: step: 2600/1541, loss: 0.00017460854724049568 2023-01-23 17:48:28.242516: step: 2602/1541, loss: 0.004750351887196302 2023-01-23 17:48:29.231595: step: 2604/1541, loss: 0.007546375971287489 2023-01-23 17:48:30.114913: step: 2606/1541, loss: 0.0008017569198273122 2023-01-23 17:48:30.986927: step: 2608/1541, loss: 0.035278622061014175 2023-01-23 17:48:31.892578: step: 2610/1541, loss: 0.00476294057443738 2023-01-23 17:48:32.798638: step: 2612/1541, loss: 0.00014620114234276116 2023-01-23 17:48:33.696430: step: 2614/1541, loss: 0.0004508973506744951 2023-01-23 17:48:34.614308: step: 2616/1541, loss: 0.05170406773686409 2023-01-23 17:48:35.482747: step: 2618/1541, loss: 0.0004568019649013877 2023-01-23 17:48:36.381257: step: 2620/1541, loss: 0.08147820830345154 2023-01-23 17:48:37.288733: step: 2622/1541, loss: 0.35077765583992004 2023-01-23 17:48:38.148530: step: 2624/1541, loss: 0.0071319942362606525 2023-01-23 17:48:39.054859: step: 2626/1541, loss: 0.002894115401431918 2023-01-23 17:48:39.962450: step: 2628/1541, loss: 4.213106512906961e-05 2023-01-23 17:48:40.972581: step: 2630/1541, loss: 0.07892894744873047 2023-01-23 17:48:41.914170: step: 2632/1541, loss: 0.1841679811477661 2023-01-23 17:48:42.798263: step: 2634/1541, loss: 0.1543443351984024 2023-01-23 17:48:43.781179: step: 2636/1541, loss: 0.0623285248875618 2023-01-23 17:48:44.679606: step: 2638/1541, loss: 0.04826541244983673 2023-01-23 17:48:45.594518: step: 2640/1541, loss: 0.10164955258369446 2023-01-23 17:48:46.551098: step: 2642/1541, loss: 0.015631603077054024 2023-01-23 17:48:47.469105: step: 2644/1541, loss: 0.06074410304427147 2023-01-23 17:48:48.371134: step: 2646/1541, loss: 0.35215556621551514 2023-01-23 17:48:49.288950: step: 2648/1541, loss: 0.00703853415325284 2023-01-23 17:48:50.171349: step: 2650/1541, loss: 0.21409635245800018 2023-01-23 17:48:51.120878: step: 2652/1541, loss: 0.009624115191400051 2023-01-23 17:48:52.078361: step: 2654/1541, loss: 0.4278639554977417 2023-01-23 17:48:52.966586: step: 2656/1541, loss: 0.0036215793807059526 2023-01-23 17:48:53.848706: step: 2658/1541, loss: 0.026980403810739517 2023-01-23 17:48:54.749416: step: 2660/1541, loss: 0.002494052518159151 2023-01-23 17:48:55.664210: step: 2662/1541, loss: 0.0003207832050975412 2023-01-23 17:48:56.581441: step: 2664/1541, loss: 0.4125002920627594 2023-01-23 17:48:57.472423: step: 2666/1541, loss: 0.0031397875864058733 2023-01-23 17:48:58.411203: step: 2668/1541, loss: 5.032945773564279e-05 2023-01-23 17:48:59.278917: step: 2670/1541, loss: 0.004348320420831442 2023-01-23 17:49:00.182290: step: 2672/1541, loss: 0.08487051725387573 2023-01-23 17:49:01.069568: step: 2674/1541, loss: 0.01901141367852688 2023-01-23 17:49:01.967676: step: 2676/1541, loss: 0.0032156461384147406 2023-01-23 17:49:02.862464: step: 2678/1541, loss: 0.00013404157652985305 2023-01-23 17:49:03.761279: step: 2680/1541, loss: 0.0027319358196109533 2023-01-23 17:49:04.650110: step: 2682/1541, loss: 0.0015850873896852136 2023-01-23 17:49:05.555450: step: 2684/1541, loss: 0.15923906862735748 2023-01-23 17:49:06.424267: step: 2686/1541, loss: 0.00835230853408575 2023-01-23 17:49:07.324481: step: 2688/1541, loss: 0.002187373349443078 2023-01-23 17:49:08.231719: step: 2690/1541, loss: 0.061294108629226685 2023-01-23 17:49:09.113426: step: 2692/1541, loss: 0.10602133721113205 2023-01-23 17:49:10.007917: step: 2694/1541, loss: 0.016451820731163025 2023-01-23 17:49:10.862198: step: 2696/1541, loss: 0.001974421786144376 2023-01-23 17:49:11.761921: step: 2698/1541, loss: 0.00021447068138513714 2023-01-23 17:49:12.642266: step: 2700/1541, loss: 0.05643882974982262 2023-01-23 17:49:13.554996: step: 2702/1541, loss: 0.0003560068435035646 2023-01-23 17:49:14.481243: step: 2704/1541, loss: 0.18734458088874817 2023-01-23 17:49:15.382892: step: 2706/1541, loss: 0.0814763754606247 2023-01-23 17:49:16.396648: step: 2708/1541, loss: 0.0004925358225591481 2023-01-23 17:49:17.330848: step: 2710/1541, loss: 0.1547364741563797 2023-01-23 17:49:18.195622: step: 2712/1541, loss: 0.0015154541470110416 2023-01-23 17:49:19.107921: step: 2714/1541, loss: 0.00011292370618321002 2023-01-23 17:49:20.012467: step: 2716/1541, loss: 0.0015857663238421082 2023-01-23 17:49:20.907148: step: 2718/1541, loss: 0.0005666188080795109 2023-01-23 17:49:21.861367: step: 2720/1541, loss: 0.004200703464448452 2023-01-23 17:49:22.769829: step: 2722/1541, loss: 0.22434091567993164 2023-01-23 17:49:23.644254: step: 2724/1541, loss: 0.04740498214960098 2023-01-23 17:49:24.503700: step: 2726/1541, loss: 0.033499643206596375 2023-01-23 17:49:25.357876: step: 2728/1541, loss: 0.005281072109937668 2023-01-23 17:49:26.253865: step: 2730/1541, loss: 0.0005661110626533628 2023-01-23 17:49:27.119177: step: 2732/1541, loss: 0.0002838966902345419 2023-01-23 17:49:28.016488: step: 2734/1541, loss: 0.00011677144357236102 2023-01-23 17:49:28.858039: step: 2736/1541, loss: 0.013570389710366726 2023-01-23 17:49:29.748084: step: 2738/1541, loss: 0.018606243655085564 2023-01-23 17:49:30.606802: step: 2740/1541, loss: 0.00015424980665557086 2023-01-23 17:49:31.563015: step: 2742/1541, loss: 0.00450114207342267 2023-01-23 17:49:32.449463: step: 2744/1541, loss: 0.14241746068000793 2023-01-23 17:49:33.375141: step: 2746/1541, loss: 0.0035168644972145557 2023-01-23 17:49:34.303233: step: 2748/1541, loss: 0.002681834390386939 2023-01-23 17:49:35.197682: step: 2750/1541, loss: 0.11392880976200104 2023-01-23 17:49:36.069545: step: 2752/1541, loss: 0.018653349950909615 2023-01-23 17:49:36.960530: step: 2754/1541, loss: 0.010745245032012463 2023-01-23 17:49:37.887617: step: 2756/1541, loss: 0.009388906881213188 2023-01-23 17:49:38.795984: step: 2758/1541, loss: 0.014700807631015778 2023-01-23 17:49:39.701030: step: 2760/1541, loss: 0.128401517868042 2023-01-23 17:49:40.580265: step: 2762/1541, loss: 0.04796696826815605 2023-01-23 17:49:41.452126: step: 2764/1541, loss: 0.1788967400789261 2023-01-23 17:49:42.397062: step: 2766/1541, loss: 0.0012894284445792437 2023-01-23 17:49:43.333876: step: 2768/1541, loss: 0.17486463487148285 2023-01-23 17:49:44.223289: step: 2770/1541, loss: 0.007496447302401066 2023-01-23 17:49:45.125554: step: 2772/1541, loss: 0.07344318181276321 2023-01-23 17:49:46.065751: step: 2774/1541, loss: 0.00040987570537254214 2023-01-23 17:49:46.949072: step: 2776/1541, loss: 0.06494468450546265 2023-01-23 17:49:47.869559: step: 2778/1541, loss: 0.0049169473350048065 2023-01-23 17:49:48.841950: step: 2780/1541, loss: 0.0012072697281837463 2023-01-23 17:49:49.768975: step: 2782/1541, loss: 0.010968266054987907 2023-01-23 17:49:50.726486: step: 2784/1541, loss: 0.023616133257746696 2023-01-23 17:49:51.593311: step: 2786/1541, loss: 0.0024698874913156033 2023-01-23 17:49:52.488654: step: 2788/1541, loss: 0.007809215225279331 2023-01-23 17:49:53.396707: step: 2790/1541, loss: 0.07622916251420975 2023-01-23 17:49:54.231523: step: 2792/1541, loss: 0.010049140080809593 2023-01-23 17:49:55.235621: step: 2794/1541, loss: 0.009340393356978893 2023-01-23 17:49:56.147159: step: 2796/1541, loss: 0.09521002322435379 2023-01-23 17:49:57.011445: step: 2798/1541, loss: 0.11681286990642548 2023-01-23 17:49:57.901697: step: 2800/1541, loss: 0.07114296406507492 2023-01-23 17:49:58.828396: step: 2802/1541, loss: 0.01651686616241932 2023-01-23 17:49:59.757243: step: 2804/1541, loss: 0.1150626614689827 2023-01-23 17:50:00.708188: step: 2806/1541, loss: 1.6618006229400635 2023-01-23 17:50:01.672522: step: 2808/1541, loss: 0.0011535113444551826 2023-01-23 17:50:02.608444: step: 2810/1541, loss: 0.004409173037856817 2023-01-23 17:50:03.461286: step: 2812/1541, loss: 0.947961688041687 2023-01-23 17:50:04.307588: step: 2814/1541, loss: 0.3818747401237488 2023-01-23 17:50:05.235508: step: 2816/1541, loss: 0.030300475656986237 2023-01-23 17:50:06.172160: step: 2818/1541, loss: 0.028247199952602386 2023-01-23 17:50:07.068323: step: 2820/1541, loss: 0.05519480258226395 2023-01-23 17:50:07.946594: step: 2822/1541, loss: 0.011319166980683804 2023-01-23 17:50:08.862038: step: 2824/1541, loss: 0.03378576412796974 2023-01-23 17:50:09.758846: step: 2826/1541, loss: 0.06541011482477188 2023-01-23 17:50:10.648091: step: 2828/1541, loss: 0.005678283050656319 2023-01-23 17:50:11.577335: step: 2830/1541, loss: 0.07159431278705597 2023-01-23 17:50:12.440589: step: 2832/1541, loss: 0.05191764608025551 2023-01-23 17:50:13.317589: step: 2834/1541, loss: 0.021729083731770515 2023-01-23 17:50:14.200689: step: 2836/1541, loss: 0.0022549021523445845 2023-01-23 17:50:15.120345: step: 2838/1541, loss: 0.002510567894205451 2023-01-23 17:50:16.080874: step: 2840/1541, loss: 0.010431697592139244 2023-01-23 17:50:16.972735: step: 2842/1541, loss: 0.16166724264621735 2023-01-23 17:50:17.917530: step: 2844/1541, loss: 0.030802225694060326 2023-01-23 17:50:18.776189: step: 2846/1541, loss: 0.0005814876058138907 2023-01-23 17:50:19.674083: step: 2848/1541, loss: 0.0035405317321419716 2023-01-23 17:50:20.591269: step: 2850/1541, loss: 0.03159680962562561 2023-01-23 17:50:21.527324: step: 2852/1541, loss: 0.0017909130547195673 2023-01-23 17:50:22.403517: step: 2854/1541, loss: 0.002428669948130846 2023-01-23 17:50:23.309183: step: 2856/1541, loss: 0.016075311228632927 2023-01-23 17:50:24.176682: step: 2858/1541, loss: 0.0008912817575037479 2023-01-23 17:50:25.169069: step: 2860/1541, loss: 0.04546457901597023 2023-01-23 17:50:26.055976: step: 2862/1541, loss: 0.023913517594337463 2023-01-23 17:50:26.945939: step: 2864/1541, loss: 0.0014443928375840187 2023-01-23 17:50:27.822343: step: 2866/1541, loss: 0.004317378159612417 2023-01-23 17:50:28.739119: step: 2868/1541, loss: 0.006408238783478737 2023-01-23 17:50:29.658653: step: 2870/1541, loss: 0.10288780927658081 2023-01-23 17:50:30.539297: step: 2872/1541, loss: 0.024197766557335854 2023-01-23 17:50:31.437839: step: 2874/1541, loss: 0.2242688685655594 2023-01-23 17:50:32.388450: step: 2876/1541, loss: 0.00011797322076745331 2023-01-23 17:50:33.274431: step: 2878/1541, loss: 0.0006561552872881293 2023-01-23 17:50:34.209359: step: 2880/1541, loss: 0.17273540794849396 2023-01-23 17:50:35.118652: step: 2882/1541, loss: 0.35843104124069214 2023-01-23 17:50:35.999740: step: 2884/1541, loss: 0.1427791714668274 2023-01-23 17:50:36.850227: step: 2886/1541, loss: 0.018364684656262398 2023-01-23 17:50:37.709606: step: 2888/1541, loss: 0.00044293139944784343 2023-01-23 17:50:38.579519: step: 2890/1541, loss: 0.10505188256502151 2023-01-23 17:50:39.481700: step: 2892/1541, loss: 0.08222459256649017 2023-01-23 17:50:40.404469: step: 2894/1541, loss: 0.05194244161248207 2023-01-23 17:50:41.341852: step: 2896/1541, loss: 0.00054686184739694 2023-01-23 17:50:42.315707: step: 2898/1541, loss: 0.0011506772134453058 2023-01-23 17:50:43.231040: step: 2900/1541, loss: 0.00032968982122838497 2023-01-23 17:50:44.110589: step: 2902/1541, loss: 0.37267759442329407 2023-01-23 17:50:45.029656: step: 2904/1541, loss: 0.0721164420247078 2023-01-23 17:50:45.936524: step: 2906/1541, loss: 0.0002611158415675163 2023-01-23 17:50:46.839684: step: 2908/1541, loss: 0.0813191831111908 2023-01-23 17:50:47.696576: step: 2910/1541, loss: 0.01742221787571907 2023-01-23 17:50:48.569609: step: 2912/1541, loss: 0.15463384985923767 2023-01-23 17:50:49.493544: step: 2914/1541, loss: 0.0016397001454606652 2023-01-23 17:50:50.372208: step: 2916/1541, loss: 0.0009377422393299639 2023-01-23 17:50:51.268870: step: 2918/1541, loss: 0.931181013584137 2023-01-23 17:50:52.132220: step: 2920/1541, loss: 0.20597627758979797 2023-01-23 17:50:52.982939: step: 2922/1541, loss: 0.0005581713048741221 2023-01-23 17:50:53.859987: step: 2924/1541, loss: 0.024001838639378548 2023-01-23 17:50:54.757702: step: 2926/1541, loss: 0.009011968038976192 2023-01-23 17:50:55.647360: step: 2928/1541, loss: 0.1230221763253212 2023-01-23 17:50:56.523061: step: 2930/1541, loss: 0.007669690065085888 2023-01-23 17:50:57.388910: step: 2932/1541, loss: 0.002974555129185319 2023-01-23 17:50:58.293177: step: 2934/1541, loss: 0.004047964233905077 2023-01-23 17:50:59.183216: step: 2936/1541, loss: 0.0408557653427124 2023-01-23 17:51:00.111603: step: 2938/1541, loss: 0.13497279584407806 2023-01-23 17:51:00.980002: step: 2940/1541, loss: 0.00043835624819621444 2023-01-23 17:51:01.947956: step: 2942/1541, loss: 0.004963363520801067 2023-01-23 17:51:02.832224: step: 2944/1541, loss: 0.831804633140564 2023-01-23 17:51:03.704249: step: 2946/1541, loss: 0.07121330499649048 2023-01-23 17:51:04.629038: step: 2948/1541, loss: 0.0007236337987706065 2023-01-23 17:51:05.511782: step: 2950/1541, loss: 0.002559191547334194 2023-01-23 17:51:06.425328: step: 2952/1541, loss: 0.23446723818778992 2023-01-23 17:51:07.291086: step: 2954/1541, loss: 0.050512295216321945 2023-01-23 17:51:08.189594: step: 2956/1541, loss: 9.86531304079108e-05 2023-01-23 17:51:09.043313: step: 2958/1541, loss: 0.0006645945832133293 2023-01-23 17:51:09.924215: step: 2960/1541, loss: 0.0009313852060586214 2023-01-23 17:51:10.832738: step: 2962/1541, loss: 0.13835301995277405 2023-01-23 17:51:11.719516: step: 2964/1541, loss: 0.0017152581131085753 2023-01-23 17:51:12.644067: step: 2966/1541, loss: 0.00044712936505675316 2023-01-23 17:51:13.530620: step: 2968/1541, loss: 0.007446460425853729 2023-01-23 17:51:14.434223: step: 2970/1541, loss: 0.02441893331706524 2023-01-23 17:51:15.279624: step: 2972/1541, loss: 9.637093899073079e-05 2023-01-23 17:51:16.228971: step: 2974/1541, loss: 0.23524844646453857 2023-01-23 17:51:17.120714: step: 2976/1541, loss: 0.03699636459350586 2023-01-23 17:51:18.034154: step: 2978/1541, loss: 0.0018232834991067648 2023-01-23 17:51:18.926064: step: 2980/1541, loss: 0.00019583576067816466 2023-01-23 17:51:19.818419: step: 2982/1541, loss: 0.00022044780780561268 2023-01-23 17:51:20.688775: step: 2984/1541, loss: 0.0007202315609902143 2023-01-23 17:51:21.638050: step: 2986/1541, loss: 0.32904690504074097 2023-01-23 17:51:22.538445: step: 2988/1541, loss: 0.029740504920482635 2023-01-23 17:51:23.422184: step: 2990/1541, loss: 0.04110078141093254 2023-01-23 17:51:24.315525: step: 2992/1541, loss: 0.0007418295135721564 2023-01-23 17:51:25.254849: step: 2994/1541, loss: 0.8598541617393494 2023-01-23 17:51:26.154591: step: 2996/1541, loss: 0.16831909120082855 2023-01-23 17:51:27.029651: step: 2998/1541, loss: 0.015667149797081947 2023-01-23 17:51:27.906471: step: 3000/1541, loss: 6.767898594262078e-05 2023-01-23 17:51:28.811042: step: 3002/1541, loss: 0.0007584011764265597 2023-01-23 17:51:29.692574: step: 3004/1541, loss: 0.0018943165196105838 2023-01-23 17:51:30.570074: step: 3006/1541, loss: 0.27783119678497314 2023-01-23 17:51:31.448518: step: 3008/1541, loss: 0.14346949756145477 2023-01-23 17:51:32.339640: step: 3010/1541, loss: 0.4643014967441559 2023-01-23 17:51:33.286809: step: 3012/1541, loss: 0.013239575549960136 2023-01-23 17:51:34.196407: step: 3014/1541, loss: 0.09006498008966446 2023-01-23 17:51:35.090752: step: 3016/1541, loss: 0.004224561620503664 2023-01-23 17:51:35.987869: step: 3018/1541, loss: 0.22515395283699036 2023-01-23 17:51:36.876370: step: 3020/1541, loss: 0.0018219028133898973 2023-01-23 17:51:37.807966: step: 3022/1541, loss: 0.15560567378997803 2023-01-23 17:51:38.811339: step: 3024/1541, loss: 0.016224753111600876 2023-01-23 17:51:39.693505: step: 3026/1541, loss: 0.29661938548088074 2023-01-23 17:51:40.555525: step: 3028/1541, loss: 0.04540763422846794 2023-01-23 17:51:41.419432: step: 3030/1541, loss: 0.37755486369132996 2023-01-23 17:51:42.279351: step: 3032/1541, loss: 0.003918008878827095 2023-01-23 17:51:43.132334: step: 3034/1541, loss: 0.001330291386693716 2023-01-23 17:51:44.028922: step: 3036/1541, loss: 0.10408704727888107 2023-01-23 17:51:44.907039: step: 3038/1541, loss: 0.09624572098255157 2023-01-23 17:51:45.829221: step: 3040/1541, loss: 0.03762468695640564 2023-01-23 17:51:46.737531: step: 3042/1541, loss: 0.06486016511917114 2023-01-23 17:51:47.613865: step: 3044/1541, loss: 0.010095479898154736 2023-01-23 17:51:48.516130: step: 3046/1541, loss: 0.0010537970811128616 2023-01-23 17:51:49.399975: step: 3048/1541, loss: 0.008355822414159775 2023-01-23 17:51:50.270511: step: 3050/1541, loss: 0.006105473265051842 2023-01-23 17:51:51.159708: step: 3052/1541, loss: 0.00018566797371022403 2023-01-23 17:51:52.077634: step: 3054/1541, loss: 0.021358998492360115 2023-01-23 17:51:52.980255: step: 3056/1541, loss: 0.01902640052139759 2023-01-23 17:51:53.859987: step: 3058/1541, loss: 0.00018233933951705694 2023-01-23 17:51:54.762489: step: 3060/1541, loss: 0.028611626476049423 2023-01-23 17:51:55.679499: step: 3062/1541, loss: 0.002723496872931719 2023-01-23 17:51:56.533420: step: 3064/1541, loss: 0.008842415176331997 2023-01-23 17:51:57.413668: step: 3066/1541, loss: 0.3393429219722748 2023-01-23 17:51:58.343721: step: 3068/1541, loss: 0.07681061327457428 2023-01-23 17:51:59.263321: step: 3070/1541, loss: 0.0034958624746650457 2023-01-23 17:52:00.175876: step: 3072/1541, loss: 0.4610866904258728 2023-01-23 17:52:01.089839: step: 3074/1541, loss: 0.01534364651888609 2023-01-23 17:52:02.015896: step: 3076/1541, loss: 6.282328104134649e-05 2023-01-23 17:52:02.963106: step: 3078/1541, loss: 0.0006471217493526638 2023-01-23 17:52:03.849466: step: 3080/1541, loss: 0.10365581512451172 2023-01-23 17:52:04.750219: step: 3082/1541, loss: 0.002873768797144294 2023-01-23 17:52:05.376180: step: 3084/1541, loss: 0.012348378077149391 ================================================== Loss: 0.075 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6810820418882222, 'r': 0.6435670496662952, 'f1': 0.6617933203189905}, 'combined': 0.43675025169666387, 'epoch': 26} Test Chinese: {'event': {'p': 0.6077812828601472, 'r': 0.6926303175554224, 'f1': 0.6474376925231027}, 'argument': {'p': 0.7213274123810446, 'r': 0.7003127754580717, 'f1': 0.7106647750952937}, 'combined': 0.4601111621451467, 'epoch': 26} Dev Korean: {'event': {'p': 0.5686274509803921, 'r': 0.7723035952063915, 'f1': 0.6549971767363071}, 'argument': {'p': 0.6699095436188013, 'r': 0.6554925520057711, 'f1': 0.6626226377781327}, 'combined': 0.43401595698624157, 'epoch': 26} Test Korean: {'event': {'p': 0.6086749285033365, 'r': 0.7651288196524865, 'f1': 0.6779930979559331}, 'argument': {'p': 0.719367683224033, 'r': 0.715075211965764, 'f1': 0.7172150251330616}, 'combined': 0.4862668367905068, 'epoch': 26} Dev Russian: {'event': {'p': 0.6504739336492891, 'r': 0.7310252996005326, 'f1': 0.6884012539184953}, 'argument': {'p': 0.6782118548747985, 'r': 0.649634079199111, 'f1': 0.6636154429328351}, 'combined': 0.4568337030346413, 'epoch': 26} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.7201609983745072, 'r': 0.7043929363139024, 'f1': 0.7121897008058545}, 'combined': 0.4937583344693104, 'epoch': 26} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6636363636363637, 'r': 0.7012006861063464, 'f1': 0.6819015846538783}, 'combined': 0.5128832431584726, 'epoch': 26} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5598387503149408, 'r': 0.45804988662131524, 'f1': 0.5038548752834466}, 'combined': 0.2935847304801185, 'epoch': 26} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5638570649208947, 'r': 0.6163088849135361, 'f1': 0.5889173789173789}, 'combined': 0.28604558404558406, 'epoch': 26} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6748395028808764, 'r': 0.6385747632400816, 'f1': 0.6562064793919685}, 'combined': 0.4371955007535911, 'epoch': 25} Test for Chinese: {'event': {'p': 0.6109358569926393, 'r': 0.6962252846015579, 'f1': 0.6507980957714926}, 'argument': {'p': 0.7154126636351134, 'r': 0.6940648330608487, 'f1': 0.7045770821024057}, 'combined': 0.45853742335648023, 'epoch': 25} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6613345847859121, 'r': 0.7050076234038498, 'f1': 0.6824731331580647}, 'combined': 0.5133131257940999, 'epoch': 25} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 27 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 17:57:27.944399: step: 2/1541, loss: 0.07283709198236465 2023-01-23 17:57:28.858117: step: 4/1541, loss: 0.015200561843812466 2023-01-23 17:57:29.802403: step: 6/1541, loss: 0.1604229062795639 2023-01-23 17:57:30.710900: step: 8/1541, loss: 0.0003460958250798285 2023-01-23 17:57:31.609446: step: 10/1541, loss: 0.0002640494203660637 2023-01-23 17:57:32.479405: step: 12/1541, loss: 0.0008807028061710298 2023-01-23 17:57:33.375015: step: 14/1541, loss: 0.0030832148622721434 2023-01-23 17:57:34.266732: step: 16/1541, loss: 0.00011614768300205469 2023-01-23 17:57:35.121177: step: 18/1541, loss: 0.26672494411468506 2023-01-23 17:57:36.009256: step: 20/1541, loss: 0.006004920229315758 2023-01-23 17:57:36.904116: step: 22/1541, loss: 0.00244910828769207 2023-01-23 17:57:37.844066: step: 24/1541, loss: 0.000568314571864903 2023-01-23 17:57:38.785039: step: 26/1541, loss: 0.0034795929677784443 2023-01-23 17:57:39.688783: step: 28/1541, loss: 0.0010099747451022267 2023-01-23 17:57:40.590589: step: 30/1541, loss: 0.00010077083425130695 2023-01-23 17:57:41.527901: step: 32/1541, loss: 0.5522806644439697 2023-01-23 17:57:42.460015: step: 34/1541, loss: 0.0002947974717244506 2023-01-23 17:57:43.333867: step: 36/1541, loss: 0.0001157509905169718 2023-01-23 17:57:44.205145: step: 38/1541, loss: 0.02526974305510521 2023-01-23 17:57:45.059054: step: 40/1541, loss: 0.00044543229159899056 2023-01-23 17:57:45.908240: step: 42/1541, loss: 4.0101298509398475e-05 2023-01-23 17:57:46.905093: step: 44/1541, loss: 0.0016571212327107787 2023-01-23 17:57:47.810594: step: 46/1541, loss: 0.025844257324934006 2023-01-23 17:57:48.696896: step: 48/1541, loss: 0.0007706863107159734 2023-01-23 17:57:49.674488: step: 50/1541, loss: 0.03433513641357422 2023-01-23 17:57:50.588288: step: 52/1541, loss: 0.0202409066259861 2023-01-23 17:57:51.522944: step: 54/1541, loss: 0.003350481390953064 2023-01-23 17:57:52.403656: step: 56/1541, loss: 0.22065168619155884 2023-01-23 17:57:53.317027: step: 58/1541, loss: 0.0017876578494906425 2023-01-23 17:57:54.275483: step: 60/1541, loss: 0.00011031172471120954 2023-01-23 17:57:55.185127: step: 62/1541, loss: 0.003936642315238714 2023-01-23 17:57:56.056655: step: 64/1541, loss: 0.004814852960407734 2023-01-23 17:57:56.929280: step: 66/1541, loss: 0.009551108814775944 2023-01-23 17:57:57.799843: step: 68/1541, loss: 0.0006688829744234681 2023-01-23 17:57:58.684709: step: 70/1541, loss: 0.0012907723430544138 2023-01-23 17:57:59.549859: step: 72/1541, loss: 0.3278496563434601 2023-01-23 17:58:00.507511: step: 74/1541, loss: 0.06469602882862091 2023-01-23 17:58:01.455241: step: 76/1541, loss: 0.05497293546795845 2023-01-23 17:58:02.352292: step: 78/1541, loss: 0.0008292000857181847 2023-01-23 17:58:03.222538: step: 80/1541, loss: 0.01639399491250515 2023-01-23 17:58:04.106377: step: 82/1541, loss: 0.00972603727132082 2023-01-23 17:58:04.985151: step: 84/1541, loss: 0.00530648697167635 2023-01-23 17:58:05.889633: step: 86/1541, loss: 0.004954563453793526 2023-01-23 17:58:06.844544: step: 88/1541, loss: 1.330230712890625 2023-01-23 17:58:07.721218: step: 90/1541, loss: 0.0023209552746266127 2023-01-23 17:58:08.612863: step: 92/1541, loss: 0.012316975742578506 2023-01-23 17:58:09.483071: step: 94/1541, loss: 0.03093576803803444 2023-01-23 17:58:10.362905: step: 96/1541, loss: 0.0002592662931419909 2023-01-23 17:58:11.298535: step: 98/1541, loss: 0.0139645766466856 2023-01-23 17:58:12.225893: step: 100/1541, loss: 0.0058931224048137665 2023-01-23 17:58:13.126169: step: 102/1541, loss: 0.1305747926235199 2023-01-23 17:58:14.046191: step: 104/1541, loss: 0.2852785885334015 2023-01-23 17:58:14.943222: step: 106/1541, loss: 0.04078283905982971 2023-01-23 17:58:15.844833: step: 108/1541, loss: 0.027525754645466805 2023-01-23 17:58:16.771818: step: 110/1541, loss: 0.3429343104362488 2023-01-23 17:58:17.677561: step: 112/1541, loss: 0.00022968201665207744 2023-01-23 17:58:18.567354: step: 114/1541, loss: 4.104375693714246e-05 2023-01-23 17:58:19.475801: step: 116/1541, loss: 0.0014364932430908084 2023-01-23 17:58:20.414957: step: 118/1541, loss: 0.0009201444918289781 2023-01-23 17:58:21.295977: step: 120/1541, loss: 0.3879695236682892 2023-01-23 17:58:22.162878: step: 122/1541, loss: 0.09017079323530197 2023-01-23 17:58:23.071412: step: 124/1541, loss: 0.6563896536827087 2023-01-23 17:58:23.957661: step: 126/1541, loss: 0.00015733588952571154 2023-01-23 17:58:24.830310: step: 128/1541, loss: 0.00043859906145371497 2023-01-23 17:58:25.735403: step: 130/1541, loss: 0.024129308760166168 2023-01-23 17:58:26.600341: step: 132/1541, loss: 0.4005509614944458 2023-01-23 17:58:27.485547: step: 134/1541, loss: 0.15371915698051453 2023-01-23 17:58:28.420487: step: 136/1541, loss: 0.0010298880515620112 2023-01-23 17:58:29.389159: step: 138/1541, loss: 0.02334531396627426 2023-01-23 17:58:30.293698: step: 140/1541, loss: 0.00048067717580124736 2023-01-23 17:58:31.181564: step: 142/1541, loss: 0.009390237741172314 2023-01-23 17:58:32.040114: step: 144/1541, loss: 0.5397969484329224 2023-01-23 17:58:32.920426: step: 146/1541, loss: 0.3719683885574341 2023-01-23 17:58:33.828767: step: 148/1541, loss: 0.06723015010356903 2023-01-23 17:58:34.686254: step: 150/1541, loss: 0.047367461025714874 2023-01-23 17:58:35.582365: step: 152/1541, loss: 0.0005913583445362747 2023-01-23 17:58:36.499033: step: 154/1541, loss: 0.037965912371873856 2023-01-23 17:58:37.367173: step: 156/1541, loss: 0.03613697737455368 2023-01-23 17:58:38.256654: step: 158/1541, loss: 0.11240153759717941 2023-01-23 17:58:39.142822: step: 160/1541, loss: 0.0010147374123334885 2023-01-23 17:58:40.036916: step: 162/1541, loss: 0.1765008419752121 2023-01-23 17:58:41.065828: step: 164/1541, loss: 0.09623755514621735 2023-01-23 17:58:41.970788: step: 166/1541, loss: 0.0004837347660213709 2023-01-23 17:58:42.817339: step: 168/1541, loss: 0.07107651978731155 2023-01-23 17:58:43.720672: step: 170/1541, loss: 0.09371251612901688 2023-01-23 17:58:44.629382: step: 172/1541, loss: 0.008168065920472145 2023-01-23 17:58:45.559362: step: 174/1541, loss: 0.0044967252761125565 2023-01-23 17:58:46.412640: step: 176/1541, loss: 0.002844666363671422 2023-01-23 17:58:47.312112: step: 178/1541, loss: 0.0003455121477600187 2023-01-23 17:58:48.223413: step: 180/1541, loss: 0.04317326843738556 2023-01-23 17:58:49.096906: step: 182/1541, loss: 0.00329953758046031 2023-01-23 17:58:49.966994: step: 184/1541, loss: 0.01276694517582655 2023-01-23 17:58:50.822967: step: 186/1541, loss: 0.004421725869178772 2023-01-23 17:58:51.727891: step: 188/1541, loss: 0.0383797325193882 2023-01-23 17:58:52.693528: step: 190/1541, loss: 0.0403301827609539 2023-01-23 17:58:53.650625: step: 192/1541, loss: 0.0002836535859387368 2023-01-23 17:58:54.553662: step: 194/1541, loss: 0.0018853741930797696 2023-01-23 17:58:55.481833: step: 196/1541, loss: 0.006863745395094156 2023-01-23 17:58:56.391612: step: 198/1541, loss: 0.017429249361157417 2023-01-23 17:58:57.310370: step: 200/1541, loss: 0.00012562495248857886 2023-01-23 17:58:58.204082: step: 202/1541, loss: 0.07516290247440338 2023-01-23 17:58:59.102919: step: 204/1541, loss: 0.0010751786176115274 2023-01-23 17:59:00.060386: step: 206/1541, loss: 0.008516889065504074 2023-01-23 17:59:00.950765: step: 208/1541, loss: 0.0010553458705544472 2023-01-23 17:59:01.809904: step: 210/1541, loss: 0.0828784927725792 2023-01-23 17:59:02.722247: step: 212/1541, loss: 0.0004622564883902669 2023-01-23 17:59:03.616691: step: 214/1541, loss: 0.003980512730777264 2023-01-23 17:59:04.494513: step: 216/1541, loss: 0.01466457825154066 2023-01-23 17:59:05.390737: step: 218/1541, loss: 0.8950121998786926 2023-01-23 17:59:06.294437: step: 220/1541, loss: 0.0001912835577968508 2023-01-23 17:59:07.194042: step: 222/1541, loss: 0.09842263907194138 2023-01-23 17:59:08.078522: step: 224/1541, loss: 9.684912220109254e-05 2023-01-23 17:59:08.955882: step: 226/1541, loss: 0.17371879518032074 2023-01-23 17:59:09.955625: step: 228/1541, loss: 0.003970833495259285 2023-01-23 17:59:10.822775: step: 230/1541, loss: 0.0002001599787035957 2023-01-23 17:59:11.681267: step: 232/1541, loss: 0.0001584272540640086 2023-01-23 17:59:12.643902: step: 234/1541, loss: 0.03239699453115463 2023-01-23 17:59:13.560353: step: 236/1541, loss: 0.010088596493005753 2023-01-23 17:59:14.482218: step: 238/1541, loss: 0.26873624324798584 2023-01-23 17:59:15.344635: step: 240/1541, loss: 0.05309980735182762 2023-01-23 17:59:16.237667: step: 242/1541, loss: 0.020588543266057968 2023-01-23 17:59:17.107552: step: 244/1541, loss: 0.01301724836230278 2023-01-23 17:59:18.082086: step: 246/1541, loss: 0.01682693511247635 2023-01-23 17:59:18.970264: step: 248/1541, loss: 0.003911660052835941 2023-01-23 17:59:19.862646: step: 250/1541, loss: 0.013673562556505203 2023-01-23 17:59:20.752683: step: 252/1541, loss: 0.001131583470851183 2023-01-23 17:59:21.623139: step: 254/1541, loss: 0.00023098917154129595 2023-01-23 17:59:22.493377: step: 256/1541, loss: 0.010178491473197937 2023-01-23 17:59:23.430852: step: 258/1541, loss: 0.013513032346963882 2023-01-23 17:59:24.279663: step: 260/1541, loss: 4.671896022045985e-05 2023-01-23 17:59:25.217220: step: 262/1541, loss: 0.003769803326576948 2023-01-23 17:59:26.098148: step: 264/1541, loss: 0.003655648324638605 2023-01-23 17:59:26.950567: step: 266/1541, loss: 0.0002186257333960384 2023-01-23 17:59:27.820532: step: 268/1541, loss: 0.001516859163530171 2023-01-23 17:59:28.699687: step: 270/1541, loss: 0.003295780625194311 2023-01-23 17:59:29.562088: step: 272/1541, loss: 2.614482946228236e-05 2023-01-23 17:59:30.494560: step: 274/1541, loss: 0.001069219782948494 2023-01-23 17:59:31.370855: step: 276/1541, loss: 0.07124155759811401 2023-01-23 17:59:32.255464: step: 278/1541, loss: 0.00129249831661582 2023-01-23 17:59:33.197244: step: 280/1541, loss: 0.005940701346844435 2023-01-23 17:59:34.106438: step: 282/1541, loss: 0.01132931374013424 2023-01-23 17:59:35.007182: step: 284/1541, loss: 0.004763514269143343 2023-01-23 17:59:35.878085: step: 286/1541, loss: 0.000810268975328654 2023-01-23 17:59:36.756986: step: 288/1541, loss: 0.199727401137352 2023-01-23 17:59:37.695061: step: 290/1541, loss: 0.11185810714960098 2023-01-23 17:59:38.613641: step: 292/1541, loss: 0.2968308627605438 2023-01-23 17:59:39.589606: step: 294/1541, loss: 0.161673903465271 2023-01-23 17:59:40.466994: step: 296/1541, loss: 0.25941383838653564 2023-01-23 17:59:41.386862: step: 298/1541, loss: 0.0009470876539126039 2023-01-23 17:59:42.355563: step: 300/1541, loss: 0.000353877927409485 2023-01-23 17:59:43.235191: step: 302/1541, loss: 0.003488925751298666 2023-01-23 17:59:44.165571: step: 304/1541, loss: 0.0007850160473026335 2023-01-23 17:59:45.155354: step: 306/1541, loss: 0.08349671959877014 2023-01-23 17:59:46.090970: step: 308/1541, loss: 0.009394793771207333 2023-01-23 17:59:47.048129: step: 310/1541, loss: 0.10068286955356598 2023-01-23 17:59:47.976668: step: 312/1541, loss: 0.00021089412621222436 2023-01-23 17:59:48.889710: step: 314/1541, loss: 0.01702660322189331 2023-01-23 17:59:49.770950: step: 316/1541, loss: 0.001251551671884954 2023-01-23 17:59:50.664001: step: 318/1541, loss: 0.04852648824453354 2023-01-23 17:59:51.526018: step: 320/1541, loss: 0.25437045097351074 2023-01-23 17:59:52.398173: step: 322/1541, loss: 0.0012469522189348936 2023-01-23 17:59:53.307363: step: 324/1541, loss: 0.00011252229160163552 2023-01-23 17:59:54.221776: step: 326/1541, loss: 0.00039929698687046766 2023-01-23 17:59:55.248794: step: 328/1541, loss: 0.06592962890863419 2023-01-23 17:59:56.131149: step: 330/1541, loss: 0.0032867789268493652 2023-01-23 17:59:57.019401: step: 332/1541, loss: 0.006042896769940853 2023-01-23 17:59:57.898938: step: 334/1541, loss: 0.0011577941477298737 2023-01-23 17:59:58.796639: step: 336/1541, loss: 0.00025699648540467024 2023-01-23 17:59:59.666209: step: 338/1541, loss: 0.00016324811440426856 2023-01-23 18:00:00.552270: step: 340/1541, loss: 0.0026530781760811806 2023-01-23 18:00:01.434793: step: 342/1541, loss: 0.27085399627685547 2023-01-23 18:00:02.327720: step: 344/1541, loss: 0.0006283913389779627 2023-01-23 18:00:03.210061: step: 346/1541, loss: 0.20729653537273407 2023-01-23 18:00:04.161474: step: 348/1541, loss: 0.0005452589248307049 2023-01-23 18:00:05.058873: step: 350/1541, loss: 0.00020014568872284144 2023-01-23 18:00:06.030953: step: 352/1541, loss: 0.003629665821790695 2023-01-23 18:00:06.879441: step: 354/1541, loss: 0.0006490212399512529 2023-01-23 18:00:07.760599: step: 356/1541, loss: 0.0013075845781713724 2023-01-23 18:00:08.640502: step: 358/1541, loss: 0.0009340292890556157 2023-01-23 18:00:09.566254: step: 360/1541, loss: 0.0016185181448236108 2023-01-23 18:00:10.494042: step: 362/1541, loss: 0.00019114544556941837 2023-01-23 18:00:11.414711: step: 364/1541, loss: 0.0013756239786744118 2023-01-23 18:00:12.352302: step: 366/1541, loss: 0.0005839166115038097 2023-01-23 18:00:13.275426: step: 368/1541, loss: 0.16135968267917633 2023-01-23 18:00:14.218683: step: 370/1541, loss: 0.12822021543979645 2023-01-23 18:00:15.137054: step: 372/1541, loss: 0.05938100442290306 2023-01-23 18:00:16.009949: step: 374/1541, loss: 0.0009262322564609349 2023-01-23 18:00:16.925670: step: 376/1541, loss: 0.0016923939110711217 2023-01-23 18:00:17.822152: step: 378/1541, loss: 0.06810467690229416 2023-01-23 18:00:18.740800: step: 380/1541, loss: 0.12835147976875305 2023-01-23 18:00:19.671142: step: 382/1541, loss: 0.03280533850193024 2023-01-23 18:00:20.595862: step: 384/1541, loss: 0.0005280638579279184 2023-01-23 18:00:21.569314: step: 386/1541, loss: 0.07565031200647354 2023-01-23 18:00:22.480598: step: 388/1541, loss: 0.005925037898123264 2023-01-23 18:00:23.364191: step: 390/1541, loss: 0.0019166412530466914 2023-01-23 18:00:24.260623: step: 392/1541, loss: 0.00015137951413635164 2023-01-23 18:00:25.119217: step: 394/1541, loss: 0.04664111137390137 2023-01-23 18:00:26.008505: step: 396/1541, loss: 0.20409439504146576 2023-01-23 18:00:26.965497: step: 398/1541, loss: 0.000354024232365191 2023-01-23 18:00:27.828957: step: 400/1541, loss: 0.16616424918174744 2023-01-23 18:00:28.726123: step: 402/1541, loss: 0.0012973870616406202 2023-01-23 18:00:29.595398: step: 404/1541, loss: 0.026329338550567627 2023-01-23 18:00:30.547259: step: 406/1541, loss: 0.006738945376127958 2023-01-23 18:00:31.442366: step: 408/1541, loss: 0.016260815784335136 2023-01-23 18:00:32.437625: step: 410/1541, loss: 0.00027535847038961947 2023-01-23 18:00:33.278924: step: 412/1541, loss: 0.00010300932626705617 2023-01-23 18:00:34.128345: step: 414/1541, loss: 0.11363080888986588 2023-01-23 18:00:35.025751: step: 416/1541, loss: 0.00012210311251692474 2023-01-23 18:00:35.965276: step: 418/1541, loss: 0.0008193327812477946 2023-01-23 18:00:36.847860: step: 420/1541, loss: 0.011651730164885521 2023-01-23 18:00:37.740700: step: 422/1541, loss: 6.835607200628147e-05 2023-01-23 18:00:38.666355: step: 424/1541, loss: 0.01683942601084709 2023-01-23 18:00:39.525117: step: 426/1541, loss: 0.0010049233678728342 2023-01-23 18:00:40.393593: step: 428/1541, loss: 4.1537434299243614e-05 2023-01-23 18:00:41.268715: step: 430/1541, loss: 0.0016917845932766795 2023-01-23 18:00:42.152525: step: 432/1541, loss: 0.00018344743875786662 2023-01-23 18:00:43.029127: step: 434/1541, loss: 0.09024136513471603 2023-01-23 18:00:43.908865: step: 436/1541, loss: 0.0022013175766915083 2023-01-23 18:00:44.831239: step: 438/1541, loss: 0.15722759068012238 2023-01-23 18:00:45.730656: step: 440/1541, loss: 0.026748383417725563 2023-01-23 18:00:46.599308: step: 442/1541, loss: 0.000249574426561594 2023-01-23 18:00:47.497485: step: 444/1541, loss: 0.04802611842751503 2023-01-23 18:00:48.395238: step: 446/1541, loss: 0.0021073161624372005 2023-01-23 18:00:49.362160: step: 448/1541, loss: 0.000534337421413511 2023-01-23 18:00:50.294694: step: 450/1541, loss: 0.325612872838974 2023-01-23 18:00:51.227866: step: 452/1541, loss: 0.10989665985107422 2023-01-23 18:00:52.082062: step: 454/1541, loss: 0.1337105631828308 2023-01-23 18:00:52.927496: step: 456/1541, loss: 9.70621986198239e-05 2023-01-23 18:00:53.812750: step: 458/1541, loss: 0.2825363874435425 2023-01-23 18:00:54.677830: step: 460/1541, loss: 0.0049674841575324535 2023-01-23 18:00:55.540250: step: 462/1541, loss: 0.037314996123313904 2023-01-23 18:00:56.432700: step: 464/1541, loss: 5.231381510384381e-05 2023-01-23 18:00:57.302549: step: 466/1541, loss: 0.0008243987103924155 2023-01-23 18:00:58.207176: step: 468/1541, loss: 0.011534999124705791 2023-01-23 18:00:59.090514: step: 470/1541, loss: 0.002158394316211343 2023-01-23 18:00:59.989018: step: 472/1541, loss: 0.1849549561738968 2023-01-23 18:01:00.894526: step: 474/1541, loss: 0.35437798500061035 2023-01-23 18:01:01.860866: step: 476/1541, loss: 0.03154651075601578 2023-01-23 18:01:02.773848: step: 478/1541, loss: 4.403445200296119e-05 2023-01-23 18:01:03.693064: step: 480/1541, loss: 0.0008374430472031236 2023-01-23 18:01:04.591428: step: 482/1541, loss: 0.32782289385795593 2023-01-23 18:01:05.463931: step: 484/1541, loss: 0.00920526497066021 2023-01-23 18:01:06.472067: step: 486/1541, loss: 0.0010014506988227367 2023-01-23 18:01:07.372644: step: 488/1541, loss: 0.00015675649046897888 2023-01-23 18:01:08.272132: step: 490/1541, loss: 0.009960196912288666 2023-01-23 18:01:09.135383: step: 492/1541, loss: 0.0005221511237323284 2023-01-23 18:01:10.020911: step: 494/1541, loss: 0.13340209424495697 2023-01-23 18:01:10.927131: step: 496/1541, loss: 0.003717210376635194 2023-01-23 18:01:11.857109: step: 498/1541, loss: 0.0011470529716461897 2023-01-23 18:01:12.738982: step: 500/1541, loss: 0.007702245842665434 2023-01-23 18:01:13.683762: step: 502/1541, loss: 0.26209962368011475 2023-01-23 18:01:14.554142: step: 504/1541, loss: 0.00015225446259137243 2023-01-23 18:01:15.421062: step: 506/1541, loss: 0.049495629966259 2023-01-23 18:01:16.321601: step: 508/1541, loss: 0.000492056249640882 2023-01-23 18:01:17.234015: step: 510/1541, loss: 0.0025157760828733444 2023-01-23 18:01:18.135516: step: 512/1541, loss: 0.044030413031578064 2023-01-23 18:01:19.021269: step: 514/1541, loss: 0.0005072755157016218 2023-01-23 18:01:19.951634: step: 516/1541, loss: 0.11772962659597397 2023-01-23 18:01:20.821989: step: 518/1541, loss: 0.06533557176589966 2023-01-23 18:01:21.722663: step: 520/1541, loss: 0.0005023585399612784 2023-01-23 18:01:22.604920: step: 522/1541, loss: 1.6089881683001295e-05 2023-01-23 18:01:23.477922: step: 524/1541, loss: 0.002885179128497839 2023-01-23 18:01:24.349129: step: 526/1541, loss: 0.01000415813177824 2023-01-23 18:01:25.245422: step: 528/1541, loss: 0.11674176901578903 2023-01-23 18:01:26.145383: step: 530/1541, loss: 0.00026160324341617525 2023-01-23 18:01:27.032655: step: 532/1541, loss: 0.00467086723074317 2023-01-23 18:01:27.938677: step: 534/1541, loss: 0.03441203013062477 2023-01-23 18:01:28.887922: step: 536/1541, loss: 0.001151318079791963 2023-01-23 18:01:29.801577: step: 538/1541, loss: 0.0003952666884288192 2023-01-23 18:01:30.695819: step: 540/1541, loss: 0.3164379596710205 2023-01-23 18:01:31.560886: step: 542/1541, loss: 0.00027069711359217763 2023-01-23 18:01:32.480699: step: 544/1541, loss: 0.23448951542377472 2023-01-23 18:01:33.474519: step: 546/1541, loss: 0.0003075978602282703 2023-01-23 18:01:34.361051: step: 548/1541, loss: 0.0032498640939593315 2023-01-23 18:01:35.255992: step: 550/1541, loss: 7.607426960021257e-05 2023-01-23 18:01:36.169255: step: 552/1541, loss: 0.15283848345279694 2023-01-23 18:01:37.033679: step: 554/1541, loss: 0.042067110538482666 2023-01-23 18:01:37.917692: step: 556/1541, loss: 0.011895086616277695 2023-01-23 18:01:38.816882: step: 558/1541, loss: 8.784547389950603e-05 2023-01-23 18:01:39.708634: step: 560/1541, loss: 4.8110803618328646e-05 2023-01-23 18:01:40.598307: step: 562/1541, loss: 0.10816147923469543 2023-01-23 18:01:41.520741: step: 564/1541, loss: 0.009651250205934048 2023-01-23 18:01:42.462704: step: 566/1541, loss: 0.05143225938081741 2023-01-23 18:01:43.347492: step: 568/1541, loss: 0.06050761789083481 2023-01-23 18:01:44.240506: step: 570/1541, loss: 0.0003252389724366367 2023-01-23 18:01:45.139738: step: 572/1541, loss: 0.001033929642289877 2023-01-23 18:01:46.050101: step: 574/1541, loss: 0.1432420164346695 2023-01-23 18:01:46.934856: step: 576/1541, loss: 0.17816461622714996 2023-01-23 18:01:47.859420: step: 578/1541, loss: 0.020475121214985847 2023-01-23 18:01:48.752935: step: 580/1541, loss: 0.0003714760532602668 2023-01-23 18:01:49.717833: step: 582/1541, loss: 0.12443475425243378 2023-01-23 18:01:50.637808: step: 584/1541, loss: 0.005674114916473627 2023-01-23 18:01:51.572710: step: 586/1541, loss: 0.023614710196852684 2023-01-23 18:01:52.483992: step: 588/1541, loss: 0.08165384083986282 2023-01-23 18:01:53.378305: step: 590/1541, loss: 0.09462262690067291 2023-01-23 18:01:54.262215: step: 592/1541, loss: 0.06865577399730682 2023-01-23 18:01:55.132360: step: 594/1541, loss: 0.0009861115831881762 2023-01-23 18:01:56.013093: step: 596/1541, loss: 0.005961399991065264 2023-01-23 18:01:56.874012: step: 598/1541, loss: 0.0021451017819344997 2023-01-23 18:01:57.771332: step: 600/1541, loss: 0.03259604051709175 2023-01-23 18:01:58.681957: step: 602/1541, loss: 0.0009297914220951498 2023-01-23 18:01:59.575005: step: 604/1541, loss: 0.0008968605543486774 2023-01-23 18:02:00.481442: step: 606/1541, loss: 6.63386017549783e-05 2023-01-23 18:02:01.415687: step: 608/1541, loss: 0.002717583440244198 2023-01-23 18:02:02.389862: step: 610/1541, loss: 0.0014847536804154515 2023-01-23 18:02:03.291636: step: 612/1541, loss: 0.023650700226426125 2023-01-23 18:02:04.146807: step: 614/1541, loss: 0.09619182348251343 2023-01-23 18:02:05.025005: step: 616/1541, loss: 0.0066593848168849945 2023-01-23 18:02:05.914060: step: 618/1541, loss: 0.0067407311871647835 2023-01-23 18:02:06.763811: step: 620/1541, loss: 0.0007040647906251252 2023-01-23 18:02:07.631490: step: 622/1541, loss: 0.034689176827669144 2023-01-23 18:02:08.534432: step: 624/1541, loss: 0.000728944200091064 2023-01-23 18:02:09.433810: step: 626/1541, loss: 0.1034860908985138 2023-01-23 18:02:10.303488: step: 628/1541, loss: 0.0005947193712927401 2023-01-23 18:02:11.191924: step: 630/1541, loss: 0.0382598415017128 2023-01-23 18:02:12.104643: step: 632/1541, loss: 0.00821090117096901 2023-01-23 18:02:13.024886: step: 634/1541, loss: 0.006921982858330011 2023-01-23 18:02:13.903499: step: 636/1541, loss: 0.003113884711638093 2023-01-23 18:02:14.807076: step: 638/1541, loss: 0.0056877750903368 2023-01-23 18:02:15.695128: step: 640/1541, loss: 9.978083107853308e-05 2023-01-23 18:02:16.573712: step: 642/1541, loss: 0.007272801827639341 2023-01-23 18:02:17.450419: step: 644/1541, loss: 0.018440021201968193 2023-01-23 18:02:18.305314: step: 646/1541, loss: 0.22874373197555542 2023-01-23 18:02:19.188188: step: 648/1541, loss: 0.4702325463294983 2023-01-23 18:02:20.143612: step: 650/1541, loss: 0.039185088127851486 2023-01-23 18:02:21.055798: step: 652/1541, loss: 0.0011894898489117622 2023-01-23 18:02:21.933966: step: 654/1541, loss: 0.0836523175239563 2023-01-23 18:02:22.807187: step: 656/1541, loss: 0.0002577352279331535 2023-01-23 18:02:23.642898: step: 658/1541, loss: 0.007581545505672693 2023-01-23 18:02:24.569317: step: 660/1541, loss: 0.06585086137056351 2023-01-23 18:02:25.495587: step: 662/1541, loss: 0.00040847313357517123 2023-01-23 18:02:26.429017: step: 664/1541, loss: 0.06428927183151245 2023-01-23 18:02:27.309645: step: 666/1541, loss: 0.00023511146719101816 2023-01-23 18:02:28.268219: step: 668/1541, loss: 0.004230995196849108 2023-01-23 18:02:29.296673: step: 670/1541, loss: 0.07155057787895203 2023-01-23 18:02:30.304392: step: 672/1541, loss: 0.007267941255122423 2023-01-23 18:02:31.209857: step: 674/1541, loss: 0.0037863438483327627 2023-01-23 18:02:32.135652: step: 676/1541, loss: 0.028117051348090172 2023-01-23 18:02:33.008894: step: 678/1541, loss: 0.005632256157696247 2023-01-23 18:02:33.882838: step: 680/1541, loss: 0.06265348196029663 2023-01-23 18:02:34.789869: step: 682/1541, loss: 0.0023947800509631634 2023-01-23 18:02:35.702058: step: 684/1541, loss: 0.002478761365637183 2023-01-23 18:02:36.575070: step: 686/1541, loss: 0.00043163742520846426 2023-01-23 18:02:37.450151: step: 688/1541, loss: 0.0062207854352891445 2023-01-23 18:02:38.315635: step: 690/1541, loss: 0.02029033564031124 2023-01-23 18:02:39.199797: step: 692/1541, loss: 0.0016346992924809456 2023-01-23 18:02:40.112219: step: 694/1541, loss: 0.5332521796226501 2023-01-23 18:02:41.012056: step: 696/1541, loss: 0.22450433671474457 2023-01-23 18:02:41.935354: step: 698/1541, loss: 0.06318198889493942 2023-01-23 18:02:42.792958: step: 700/1541, loss: 0.006612657103687525 2023-01-23 18:02:43.679033: step: 702/1541, loss: 0.09300645440816879 2023-01-23 18:02:44.582303: step: 704/1541, loss: 0.012224039062857628 2023-01-23 18:02:45.454124: step: 706/1541, loss: 0.0015893547097221017 2023-01-23 18:02:46.340297: step: 708/1541, loss: 0.0009725545533001423 2023-01-23 18:02:47.300095: step: 710/1541, loss: 0.00019767071353271604 2023-01-23 18:02:48.280422: step: 712/1541, loss: 0.0056334505788981915 2023-01-23 18:02:49.194770: step: 714/1541, loss: 0.0009692453313618898 2023-01-23 18:02:50.077015: step: 716/1541, loss: 0.0028294518124312162 2023-01-23 18:02:50.993970: step: 718/1541, loss: 0.039223819971084595 2023-01-23 18:02:51.939804: step: 720/1541, loss: 0.0015393593348562717 2023-01-23 18:02:52.840188: step: 722/1541, loss: 0.00012110341776860878 2023-01-23 18:02:53.694497: step: 724/1541, loss: 0.0002790489816106856 2023-01-23 18:02:54.568032: step: 726/1541, loss: 0.07400023192167282 2023-01-23 18:02:55.449418: step: 728/1541, loss: 0.09176111221313477 2023-01-23 18:02:56.364077: step: 730/1541, loss: 0.003940935246646404 2023-01-23 18:02:57.217669: step: 732/1541, loss: 0.0017404932295903563 2023-01-23 18:02:58.104586: step: 734/1541, loss: 0.017721669748425484 2023-01-23 18:02:58.993002: step: 736/1541, loss: 0.00012167681416030973 2023-01-23 18:02:59.887012: step: 738/1541, loss: 0.0008625767659395933 2023-01-23 18:03:00.803885: step: 740/1541, loss: 0.000792907434515655 2023-01-23 18:03:01.699796: step: 742/1541, loss: 0.657971203327179 2023-01-23 18:03:02.620785: step: 744/1541, loss: 0.06913617998361588 2023-01-23 18:03:03.494416: step: 746/1541, loss: 0.005024665966629982 2023-01-23 18:03:04.384502: step: 748/1541, loss: 0.0017707733204588294 2023-01-23 18:03:05.312471: step: 750/1541, loss: 0.0005204521003179252 2023-01-23 18:03:06.198640: step: 752/1541, loss: 0.22472445666790009 2023-01-23 18:03:07.061012: step: 754/1541, loss: 0.2016354650259018 2023-01-23 18:03:07.945909: step: 756/1541, loss: 0.03662300482392311 2023-01-23 18:03:08.835903: step: 758/1541, loss: 0.006987826898694038 2023-01-23 18:03:09.669483: step: 760/1541, loss: 0.00023659855651203543 2023-01-23 18:03:10.581164: step: 762/1541, loss: 0.00024071089865174145 2023-01-23 18:03:11.459675: step: 764/1541, loss: 0.008977671153843403 2023-01-23 18:03:12.368194: step: 766/1541, loss: 0.05015706643462181 2023-01-23 18:03:13.322665: step: 768/1541, loss: 0.00037968679680489004 2023-01-23 18:03:14.238673: step: 770/1541, loss: 0.04106549173593521 2023-01-23 18:03:15.138968: step: 772/1541, loss: 6.676172779407352e-05 2023-01-23 18:03:16.045907: step: 774/1541, loss: 0.08417996019124985 2023-01-23 18:03:16.921903: step: 776/1541, loss: 0.000244302413193509 2023-01-23 18:03:17.805186: step: 778/1541, loss: 0.008333787322044373 2023-01-23 18:03:18.707480: step: 780/1541, loss: 0.08468125760555267 2023-01-23 18:03:19.601706: step: 782/1541, loss: 0.016865191981196404 2023-01-23 18:03:20.518943: step: 784/1541, loss: 0.15510134398937225 2023-01-23 18:03:21.432449: step: 786/1541, loss: 0.011859357357025146 2023-01-23 18:03:22.294083: step: 788/1541, loss: 0.01609036698937416 2023-01-23 18:03:23.140816: step: 790/1541, loss: 0.022086497396230698 2023-01-23 18:03:24.140304: step: 792/1541, loss: 0.0005674487329088151 2023-01-23 18:03:25.065462: step: 794/1541, loss: 0.12404868751764297 2023-01-23 18:03:25.980798: step: 796/1541, loss: 0.09193741530179977 2023-01-23 18:03:26.866062: step: 798/1541, loss: 0.0005801658844575286 2023-01-23 18:03:27.758475: step: 800/1541, loss: 0.011654546484351158 2023-01-23 18:03:28.671254: step: 802/1541, loss: 0.00013239748659543693 2023-01-23 18:03:29.553961: step: 804/1541, loss: 0.05161243677139282 2023-01-23 18:03:30.455699: step: 806/1541, loss: 0.19151648879051208 2023-01-23 18:03:31.352747: step: 808/1541, loss: 0.02707555890083313 2023-01-23 18:03:32.283494: step: 810/1541, loss: 4.7118228394538164e-05 2023-01-23 18:03:33.206613: step: 812/1541, loss: 0.2702816426753998 2023-01-23 18:03:34.119928: step: 814/1541, loss: 0.0004259222187101841 2023-01-23 18:03:34.964451: step: 816/1541, loss: 0.0009046352352015674 2023-01-23 18:03:35.852053: step: 818/1541, loss: 0.14198829233646393 2023-01-23 18:03:36.746926: step: 820/1541, loss: 0.0019376665586605668 2023-01-23 18:03:37.603222: step: 822/1541, loss: 0.0017962572164833546 2023-01-23 18:03:38.474927: step: 824/1541, loss: 2.6755758881336078e-05 2023-01-23 18:03:39.353418: step: 826/1541, loss: 0.004692000336945057 2023-01-23 18:03:40.259698: step: 828/1541, loss: 0.4798257648944855 2023-01-23 18:03:41.123634: step: 830/1541, loss: 0.0034426848869770765 2023-01-23 18:03:42.022914: step: 832/1541, loss: 0.02797512523829937 2023-01-23 18:03:42.873947: step: 834/1541, loss: 0.00011960166011704132 2023-01-23 18:03:43.804642: step: 836/1541, loss: 0.10295069217681885 2023-01-23 18:03:44.668004: step: 838/1541, loss: 0.3728184998035431 2023-01-23 18:03:45.530455: step: 840/1541, loss: 0.0002626032510306686 2023-01-23 18:03:46.386616: step: 842/1541, loss: 0.004977541975677013 2023-01-23 18:03:47.254738: step: 844/1541, loss: 0.03760391101241112 2023-01-23 18:03:48.137003: step: 846/1541, loss: 0.12548235058784485 2023-01-23 18:03:49.024481: step: 848/1541, loss: 0.0004085801192559302 2023-01-23 18:03:49.888047: step: 850/1541, loss: 0.00019112438894808292 2023-01-23 18:03:50.774946: step: 852/1541, loss: 0.03203263506293297 2023-01-23 18:03:51.638529: step: 854/1541, loss: 0.00017468516307417303 2023-01-23 18:03:52.555386: step: 856/1541, loss: 0.049579933285713196 2023-01-23 18:03:53.430807: step: 858/1541, loss: 0.004128508269786835 2023-01-23 18:03:54.339899: step: 860/1541, loss: 0.0014692529803141952 2023-01-23 18:03:55.251658: step: 862/1541, loss: 0.0002488211030140519 2023-01-23 18:03:56.147194: step: 864/1541, loss: 0.11853273957967758 2023-01-23 18:03:57.032491: step: 866/1541, loss: 0.0032136798836290836 2023-01-23 18:03:57.933909: step: 868/1541, loss: 0.0005046084988862276 2023-01-23 18:03:58.828713: step: 870/1541, loss: 0.0061806570738554 2023-01-23 18:03:59.736596: step: 872/1541, loss: 0.00016351049998775125 2023-01-23 18:04:00.665965: step: 874/1541, loss: 0.09169107675552368 2023-01-23 18:04:01.551385: step: 876/1541, loss: 0.00031668145675212145 2023-01-23 18:04:02.480784: step: 878/1541, loss: 0.0003273326437920332 2023-01-23 18:04:03.337070: step: 880/1541, loss: 0.020964380353689194 2023-01-23 18:04:04.268618: step: 882/1541, loss: 0.00010931403085123748 2023-01-23 18:04:05.180878: step: 884/1541, loss: 0.01628521829843521 2023-01-23 18:04:06.065037: step: 886/1541, loss: 0.000528748263604939 2023-01-23 18:04:06.930262: step: 888/1541, loss: 0.022423939779400826 2023-01-23 18:04:07.827290: step: 890/1541, loss: 0.04285341873764992 2023-01-23 18:04:08.767901: step: 892/1541, loss: 7.813020783942193e-05 2023-01-23 18:04:09.690566: step: 894/1541, loss: 0.0008976270910352468 2023-01-23 18:04:10.572778: step: 896/1541, loss: 0.0005973702645860612 2023-01-23 18:04:11.490191: step: 898/1541, loss: 0.0041030398570001125 2023-01-23 18:04:12.386970: step: 900/1541, loss: 1.7759740352630615 2023-01-23 18:04:13.290974: step: 902/1541, loss: 0.0483875647187233 2023-01-23 18:04:14.172689: step: 904/1541, loss: 0.13768620789051056 2023-01-23 18:04:15.089501: step: 906/1541, loss: 0.00893334299325943 2023-01-23 18:04:15.998516: step: 908/1541, loss: 0.0003157333703711629 2023-01-23 18:04:16.897157: step: 910/1541, loss: 0.0016260219272226095 2023-01-23 18:04:17.825406: step: 912/1541, loss: 0.2600429058074951 2023-01-23 18:04:18.706300: step: 914/1541, loss: 0.09834174811840057 2023-01-23 18:04:19.659624: step: 916/1541, loss: 0.001677669701166451 2023-01-23 18:04:20.635556: step: 918/1541, loss: 0.01661919802427292 2023-01-23 18:04:21.532907: step: 920/1541, loss: 0.004865002352744341 2023-01-23 18:04:22.394214: step: 922/1541, loss: 0.0064811669290065765 2023-01-23 18:04:23.268995: step: 924/1541, loss: 0.08120278269052505 2023-01-23 18:04:24.142088: step: 926/1541, loss: 0.009267129935324192 2023-01-23 18:04:25.083048: step: 928/1541, loss: 0.0028448929078876972 2023-01-23 18:04:25.958957: step: 930/1541, loss: 0.0003571149136405438 2023-01-23 18:04:26.802539: step: 932/1541, loss: 0.17611193656921387 2023-01-23 18:04:27.631765: step: 934/1541, loss: 0.0032423005905002356 2023-01-23 18:04:28.566450: step: 936/1541, loss: 0.11018069088459015 2023-01-23 18:04:29.461861: step: 938/1541, loss: 0.0055410913191735744 2023-01-23 18:04:30.358009: step: 940/1541, loss: 0.04224662855267525 2023-01-23 18:04:31.231229: step: 942/1541, loss: 0.14229469001293182 2023-01-23 18:04:32.163258: step: 944/1541, loss: 9.916328417602926e-05 2023-01-23 18:04:33.031460: step: 946/1541, loss: 0.0003788807080127299 2023-01-23 18:04:33.904470: step: 948/1541, loss: 0.0011255299905315042 2023-01-23 18:04:34.769254: step: 950/1541, loss: 0.0010677911341190338 2023-01-23 18:04:35.630024: step: 952/1541, loss: 6.830608617747203e-05 2023-01-23 18:04:36.511056: step: 954/1541, loss: 0.004047640133649111 2023-01-23 18:04:37.394395: step: 956/1541, loss: 0.0006896753329783678 2023-01-23 18:04:38.265242: step: 958/1541, loss: 0.007210678420960903 2023-01-23 18:04:39.134864: step: 960/1541, loss: 4.4822711060987785e-05 2023-01-23 18:04:40.018718: step: 962/1541, loss: 0.014955095946788788 2023-01-23 18:04:40.969400: step: 964/1541, loss: 0.000320344784995541 2023-01-23 18:04:41.856042: step: 966/1541, loss: 0.05218656361103058 2023-01-23 18:04:42.757864: step: 968/1541, loss: 0.09300991147756577 2023-01-23 18:04:43.694920: step: 970/1541, loss: 0.11110693961381912 2023-01-23 18:04:44.561478: step: 972/1541, loss: 0.0053191776387393475 2023-01-23 18:04:45.552760: step: 974/1541, loss: 0.22277198731899261 2023-01-23 18:04:46.441248: step: 976/1541, loss: 0.00014807026309426874 2023-01-23 18:04:47.347021: step: 978/1541, loss: 0.0132143534719944 2023-01-23 18:04:48.341231: step: 980/1541, loss: 0.08707223832607269 2023-01-23 18:04:49.203543: step: 982/1541, loss: 0.009194506332278252 2023-01-23 18:04:50.081662: step: 984/1541, loss: 0.1835903376340866 2023-01-23 18:04:50.980311: step: 986/1541, loss: 0.026411859318614006 2023-01-23 18:04:51.863174: step: 988/1541, loss: 0.021955400705337524 2023-01-23 18:04:52.746449: step: 990/1541, loss: 0.0004309914365876466 2023-01-23 18:04:53.673700: step: 992/1541, loss: 0.017039474099874496 2023-01-23 18:04:54.567797: step: 994/1541, loss: 0.000983741250820458 2023-01-23 18:04:55.471929: step: 996/1541, loss: 0.4232889413833618 2023-01-23 18:04:56.397655: step: 998/1541, loss: 0.0003699549415614456 2023-01-23 18:04:57.261300: step: 1000/1541, loss: 0.035164158791303635 2023-01-23 18:04:58.265358: step: 1002/1541, loss: 0.010098562575876713 2023-01-23 18:04:59.160302: step: 1004/1541, loss: 0.04782905429601669 2023-01-23 18:05:00.063884: step: 1006/1541, loss: 0.00024302949896082282 2023-01-23 18:05:00.967224: step: 1008/1541, loss: 0.40208378434181213 2023-01-23 18:05:01.883245: step: 1010/1541, loss: 0.0002655113348737359 2023-01-23 18:05:02.757303: step: 1012/1541, loss: 0.00029039522632956505 2023-01-23 18:05:03.610573: step: 1014/1541, loss: 0.2318417876958847 2023-01-23 18:05:04.477280: step: 1016/1541, loss: 0.0002036391233559698 2023-01-23 18:05:05.403663: step: 1018/1541, loss: 0.0012809677282348275 2023-01-23 18:05:06.270520: step: 1020/1541, loss: 0.00026383635122328997 2023-01-23 18:05:07.168362: step: 1022/1541, loss: 0.00036473190994001925 2023-01-23 18:05:08.096210: step: 1024/1541, loss: 0.0009313205373473465 2023-01-23 18:05:08.972022: step: 1026/1541, loss: 5.520402919501066e-05 2023-01-23 18:05:09.808029: step: 1028/1541, loss: 0.00042879150714725256 2023-01-23 18:05:10.701301: step: 1030/1541, loss: 0.0003256766067352146 2023-01-23 18:05:11.585766: step: 1032/1541, loss: 0.00018355547217652202 2023-01-23 18:05:12.458297: step: 1034/1541, loss: 0.004948724061250687 2023-01-23 18:05:13.349973: step: 1036/1541, loss: 0.05255608260631561 2023-01-23 18:05:14.180695: step: 1038/1541, loss: 0.0013953727902844548 2023-01-23 18:05:15.086022: step: 1040/1541, loss: 0.0005466061411425471 2023-01-23 18:05:15.953787: step: 1042/1541, loss: 0.0009820088744163513 2023-01-23 18:05:16.824701: step: 1044/1541, loss: 0.028765350580215454 2023-01-23 18:05:17.711860: step: 1046/1541, loss: 0.03865562006831169 2023-01-23 18:05:18.645062: step: 1048/1541, loss: 0.0008149492787197232 2023-01-23 18:05:19.530994: step: 1050/1541, loss: 0.06961632519960403 2023-01-23 18:05:20.377250: step: 1052/1541, loss: 0.05210712179541588 2023-01-23 18:05:21.237371: step: 1054/1541, loss: 0.002205176744610071 2023-01-23 18:05:22.124716: step: 1056/1541, loss: 0.004450452048331499 2023-01-23 18:05:22.984621: step: 1058/1541, loss: 0.046277254819869995 2023-01-23 18:05:23.943301: step: 1060/1541, loss: 0.015776995569467545 2023-01-23 18:05:24.821629: step: 1062/1541, loss: 0.0035305682104080915 2023-01-23 18:05:25.711672: step: 1064/1541, loss: 0.003638716647401452 2023-01-23 18:05:26.569227: step: 1066/1541, loss: -9.83491372608114e-06 2023-01-23 18:05:27.401857: step: 1068/1541, loss: -3.4124519743272685e-07 2023-01-23 18:05:28.322793: step: 1070/1541, loss: 0.02927556075155735 2023-01-23 18:05:29.219734: step: 1072/1541, loss: 0.0005777303595095873 2023-01-23 18:05:30.082779: step: 1074/1541, loss: 0.026766104623675346 2023-01-23 18:05:30.986355: step: 1076/1541, loss: 0.000768456025980413 2023-01-23 18:05:31.939013: step: 1078/1541, loss: 0.003332558088004589 2023-01-23 18:05:32.846002: step: 1080/1541, loss: 0.022349826991558075 2023-01-23 18:05:33.741783: step: 1082/1541, loss: 5.8388519391883165e-05 2023-01-23 18:05:34.632224: step: 1084/1541, loss: 0.11416687071323395 2023-01-23 18:05:35.588015: step: 1086/1541, loss: 0.0010607101721689105 2023-01-23 18:05:36.538017: step: 1088/1541, loss: 0.08305294066667557 2023-01-23 18:05:37.411767: step: 1090/1541, loss: 0.0016497953329235315 2023-01-23 18:05:38.332156: step: 1092/1541, loss: 0.035953588783741 2023-01-23 18:05:39.253011: step: 1094/1541, loss: 0.09492509812116623 2023-01-23 18:05:40.102671: step: 1096/1541, loss: 0.00016294904344249517 2023-01-23 18:05:41.006029: step: 1098/1541, loss: 0.0255209282040596 2023-01-23 18:05:41.923352: step: 1100/1541, loss: 0.22004643082618713 2023-01-23 18:05:42.893764: step: 1102/1541, loss: 0.00012255215551704168 2023-01-23 18:05:43.801520: step: 1104/1541, loss: 0.00680004246532917 2023-01-23 18:05:44.733924: step: 1106/1541, loss: 0.0008162890444509685 2023-01-23 18:05:45.617817: step: 1108/1541, loss: 0.0008053603814914823 2023-01-23 18:05:46.501660: step: 1110/1541, loss: 0.027931489050388336 2023-01-23 18:05:47.408906: step: 1112/1541, loss: 9.711132588563487e-05 2023-01-23 18:05:48.268519: step: 1114/1541, loss: 0.015267214737832546 2023-01-23 18:05:49.150189: step: 1116/1541, loss: 0.03473985195159912 2023-01-23 18:05:50.043077: step: 1118/1541, loss: 0.6016972661018372 2023-01-23 18:05:50.905329: step: 1120/1541, loss: 0.0004767312784679234 2023-01-23 18:05:51.776425: step: 1122/1541, loss: 0.000989201245829463 2023-01-23 18:05:52.652119: step: 1124/1541, loss: 0.002279613632708788 2023-01-23 18:05:53.552439: step: 1126/1541, loss: 0.0016346012707799673 2023-01-23 18:05:54.426745: step: 1128/1541, loss: 0.09958838671445847 2023-01-23 18:05:55.300411: step: 1130/1541, loss: 0.003559565870091319 2023-01-23 18:05:56.218602: step: 1132/1541, loss: 0.03484702482819557 2023-01-23 18:05:57.131549: step: 1134/1541, loss: 0.0011130907805636525 2023-01-23 18:05:57.995337: step: 1136/1541, loss: 8.424468251178041e-05 2023-01-23 18:05:58.874100: step: 1138/1541, loss: 0.08271709829568863 2023-01-23 18:05:59.797811: step: 1140/1541, loss: 0.00044858368346467614 2023-01-23 18:06:00.615528: step: 1142/1541, loss: 0.0022484848741441965 2023-01-23 18:06:01.519449: step: 1144/1541, loss: 0.16757145524024963 2023-01-23 18:06:02.476573: step: 1146/1541, loss: 0.03518151864409447 2023-01-23 18:06:03.374465: step: 1148/1541, loss: 0.07967031002044678 2023-01-23 18:06:04.235311: step: 1150/1541, loss: 0.2710338532924652 2023-01-23 18:06:05.112946: step: 1152/1541, loss: 0.018323814496397972 2023-01-23 18:06:05.998000: step: 1154/1541, loss: 0.00037894552224315703 2023-01-23 18:06:06.907457: step: 1156/1541, loss: 0.009429240599274635 2023-01-23 18:06:07.825623: step: 1158/1541, loss: 0.0008784724050201476 2023-01-23 18:06:08.782260: step: 1160/1541, loss: 0.1445206105709076 2023-01-23 18:06:09.657133: step: 1162/1541, loss: 0.00010998627112712711 2023-01-23 18:06:10.531616: step: 1164/1541, loss: 0.0009447004995308816 2023-01-23 18:06:11.391502: step: 1166/1541, loss: 0.000724065292160958 2023-01-23 18:06:12.257601: step: 1168/1541, loss: 0.00249271746724844 2023-01-23 18:06:13.269778: step: 1170/1541, loss: 0.05844223499298096 2023-01-23 18:06:14.162513: step: 1172/1541, loss: 0.0029706410132348537 2023-01-23 18:06:15.043622: step: 1174/1541, loss: 0.019777575507760048 2023-01-23 18:06:15.966406: step: 1176/1541, loss: 0.009015889838337898 2023-01-23 18:06:16.855574: step: 1178/1541, loss: 0.0004903128719888628 2023-01-23 18:06:17.765875: step: 1180/1541, loss: 0.01115107536315918 2023-01-23 18:06:18.683014: step: 1182/1541, loss: 0.36886805295944214 2023-01-23 18:06:19.548872: step: 1184/1541, loss: 0.0017509415047243237 2023-01-23 18:06:20.432619: step: 1186/1541, loss: 0.11059583723545074 2023-01-23 18:06:21.368389: step: 1188/1541, loss: 0.1179577112197876 2023-01-23 18:06:22.265026: step: 1190/1541, loss: 0.0017023951513692737 2023-01-23 18:06:23.143943: step: 1192/1541, loss: 0.0012984279310330749 2023-01-23 18:06:24.033289: step: 1194/1541, loss: 5.1974013331346214e-05 2023-01-23 18:06:24.936870: step: 1196/1541, loss: 0.02642470970749855 2023-01-23 18:06:25.868036: step: 1198/1541, loss: 0.007192032411694527 2023-01-23 18:06:26.733852: step: 1200/1541, loss: 0.010500965639948845 2023-01-23 18:06:27.657296: step: 1202/1541, loss: 0.0003446925838943571 2023-01-23 18:06:28.527020: step: 1204/1541, loss: 0.0008015433559194207 2023-01-23 18:06:29.463775: step: 1206/1541, loss: 5.570256325881928e-05 2023-01-23 18:06:30.403354: step: 1208/1541, loss: 0.004395774565637112 2023-01-23 18:06:31.280906: step: 1210/1541, loss: 0.000734786328393966 2023-01-23 18:06:32.229057: step: 1212/1541, loss: 6.230220606084913e-05 2023-01-23 18:06:33.104127: step: 1214/1541, loss: 0.8056137561798096 2023-01-23 18:06:34.009479: step: 1216/1541, loss: 0.008541599847376347 2023-01-23 18:06:34.940075: step: 1218/1541, loss: 0.0017757292371243238 2023-01-23 18:06:35.845919: step: 1220/1541, loss: 0.012400605715811253 2023-01-23 18:06:36.729695: step: 1222/1541, loss: 0.0003669857978820801 2023-01-23 18:06:37.620840: step: 1224/1541, loss: 0.001133125158958137 2023-01-23 18:06:38.480956: step: 1226/1541, loss: 0.1884223073720932 2023-01-23 18:06:39.453248: step: 1228/1541, loss: 0.05695591866970062 2023-01-23 18:06:40.383726: step: 1230/1541, loss: 0.0002461111871525645 2023-01-23 18:06:41.297676: step: 1232/1541, loss: 0.09734038263559341 2023-01-23 18:06:42.178322: step: 1234/1541, loss: 0.25219303369522095 2023-01-23 18:06:43.023649: step: 1236/1541, loss: 0.003619173541665077 2023-01-23 18:06:43.931745: step: 1238/1541, loss: 0.007386699318885803 2023-01-23 18:06:44.804224: step: 1240/1541, loss: 0.0004971115267835557 2023-01-23 18:06:45.714885: step: 1242/1541, loss: 0.00017238639702554792 2023-01-23 18:06:46.611818: step: 1244/1541, loss: 0.005500345956534147 2023-01-23 18:06:47.485938: step: 1246/1541, loss: 0.24633796513080597 2023-01-23 18:06:48.480955: step: 1248/1541, loss: 0.2585676610469818 2023-01-23 18:06:49.414544: step: 1250/1541, loss: 0.0607101246714592 2023-01-23 18:06:50.307969: step: 1252/1541, loss: 0.00019624651758931577 2023-01-23 18:06:51.195075: step: 1254/1541, loss: 0.0019677304662764072 2023-01-23 18:06:52.056650: step: 1256/1541, loss: 0.006649977061897516 2023-01-23 18:06:52.985546: step: 1258/1541, loss: 0.0001294621906708926 2023-01-23 18:06:53.928465: step: 1260/1541, loss: 0.08223327994346619 2023-01-23 18:06:54.832480: step: 1262/1541, loss: 0.561098039150238 2023-01-23 18:06:55.728931: step: 1264/1541, loss: 0.0009946690406650305 2023-01-23 18:06:56.736675: step: 1266/1541, loss: 1.1295260190963745 2023-01-23 18:06:57.648760: step: 1268/1541, loss: 0.01542128250002861 2023-01-23 18:06:58.516182: step: 1270/1541, loss: 0.0016107449773699045 2023-01-23 18:06:59.387362: step: 1272/1541, loss: 0.10351352393627167 2023-01-23 18:07:00.288425: step: 1274/1541, loss: 0.009687778539955616 2023-01-23 18:07:01.175811: step: 1276/1541, loss: 0.5390536189079285 2023-01-23 18:07:02.059186: step: 1278/1541, loss: 0.07071732729673386 2023-01-23 18:07:02.934234: step: 1280/1541, loss: 0.0004092457238584757 2023-01-23 18:07:03.796830: step: 1282/1541, loss: 7.201807420642581e-06 2023-01-23 18:07:04.690836: step: 1284/1541, loss: 0.020528623834252357 2023-01-23 18:07:05.608534: step: 1286/1541, loss: 0.44756945967674255 2023-01-23 18:07:06.519551: step: 1288/1541, loss: 0.028654461726546288 2023-01-23 18:07:07.401615: step: 1290/1541, loss: 0.09336689114570618 2023-01-23 18:07:08.268428: step: 1292/1541, loss: 0.004355344921350479 2023-01-23 18:07:09.206888: step: 1294/1541, loss: 0.022255294024944305 2023-01-23 18:07:10.098364: step: 1296/1541, loss: 0.05730894207954407 2023-01-23 18:07:10.969274: step: 1298/1541, loss: 0.0020423531532287598 2023-01-23 18:07:11.899228: step: 1300/1541, loss: 0.002007388276979327 2023-01-23 18:07:12.780855: step: 1302/1541, loss: 0.029318513348698616 2023-01-23 18:07:13.659540: step: 1304/1541, loss: 0.015509824268519878 2023-01-23 18:07:14.571037: step: 1306/1541, loss: 0.03560793027281761 2023-01-23 18:07:15.438876: step: 1308/1541, loss: 0.0016474591102451086 2023-01-23 18:07:16.345408: step: 1310/1541, loss: 0.003148226300254464 2023-01-23 18:07:17.247687: step: 1312/1541, loss: 0.030043654143810272 2023-01-23 18:07:18.109007: step: 1314/1541, loss: 0.0004920635838061571 2023-01-23 18:07:19.011495: step: 1316/1541, loss: 0.09913169592618942 2023-01-23 18:07:19.900624: step: 1318/1541, loss: 0.00016093596059363335 2023-01-23 18:07:20.777364: step: 1320/1541, loss: 0.020617956295609474 2023-01-23 18:07:21.654951: step: 1322/1541, loss: 0.00046985200606286526 2023-01-23 18:07:22.574088: step: 1324/1541, loss: 0.03658950701355934 2023-01-23 18:07:23.474420: step: 1326/1541, loss: 0.003590310225263238 2023-01-23 18:07:24.355922: step: 1328/1541, loss: 0.0002751030551735312 2023-01-23 18:07:25.241002: step: 1330/1541, loss: 0.004734113346785307 2023-01-23 18:07:26.156411: step: 1332/1541, loss: 0.002847780706360936 2023-01-23 18:07:27.062161: step: 1334/1541, loss: 0.21810106933116913 2023-01-23 18:07:28.022278: step: 1336/1541, loss: 0.00010786388156702742 2023-01-23 18:07:28.991189: step: 1338/1541, loss: 0.006444484926760197 2023-01-23 18:07:29.939128: step: 1340/1541, loss: 0.0004572336911223829 2023-01-23 18:07:30.830247: step: 1342/1541, loss: 0.07678332924842834 2023-01-23 18:07:31.749445: step: 1344/1541, loss: 0.00136326823849231 2023-01-23 18:07:32.610437: step: 1346/1541, loss: 0.00015388666361104697 2023-01-23 18:07:33.542557: step: 1348/1541, loss: 0.0001811387191992253 2023-01-23 18:07:34.436037: step: 1350/1541, loss: 0.00024078309070318937 2023-01-23 18:07:35.331025: step: 1352/1541, loss: 0.13624094426631927 2023-01-23 18:07:36.229768: step: 1354/1541, loss: 0.43146035075187683 2023-01-23 18:07:37.146091: step: 1356/1541, loss: 1.129845380783081 2023-01-23 18:07:38.065940: step: 1358/1541, loss: 0.9437760710716248 2023-01-23 18:07:38.944586: step: 1360/1541, loss: 0.0002947468892671168 2023-01-23 18:07:39.833558: step: 1362/1541, loss: 1.6782509192125872e-05 2023-01-23 18:07:40.713413: step: 1364/1541, loss: 0.08503146469593048 2023-01-23 18:07:41.568898: step: 1366/1541, loss: 0.001509802183136344 2023-01-23 18:07:42.473467: step: 1368/1541, loss: 0.008793383836746216 2023-01-23 18:07:43.343689: step: 1370/1541, loss: 0.006990802939981222 2023-01-23 18:07:44.276716: step: 1372/1541, loss: 0.03507024794816971 2023-01-23 18:07:45.163102: step: 1374/1541, loss: 0.06696155667304993 2023-01-23 18:07:46.041371: step: 1376/1541, loss: 0.012660201638936996 2023-01-23 18:07:46.953268: step: 1378/1541, loss: 0.0038054478354752064 2023-01-23 18:07:47.835819: step: 1380/1541, loss: 0.0001311897940468043 2023-01-23 18:07:48.752997: step: 1382/1541, loss: 0.0009671750594861805 2023-01-23 18:07:49.642120: step: 1384/1541, loss: 0.002820607740432024 2023-01-23 18:07:50.586917: step: 1386/1541, loss: 0.3637692332267761 2023-01-23 18:07:51.448863: step: 1388/1541, loss: 0.030728571116924286 2023-01-23 18:07:52.320696: step: 1390/1541, loss: 0.00019701328710652888 2023-01-23 18:07:53.231262: step: 1392/1541, loss: 0.0018362163100391626 2023-01-23 18:07:54.092564: step: 1394/1541, loss: 0.0023423642851412296 2023-01-23 18:07:55.019314: step: 1396/1541, loss: 0.8551843166351318 2023-01-23 18:07:55.935159: step: 1398/1541, loss: 0.0341833159327507 2023-01-23 18:07:56.805408: step: 1400/1541, loss: 0.005364479962736368 2023-01-23 18:07:57.744718: step: 1402/1541, loss: 0.0006873844540677965 2023-01-23 18:07:58.613570: step: 1404/1541, loss: 0.08763013780117035 2023-01-23 18:07:59.472938: step: 1406/1541, loss: 0.0015329316956922412 2023-01-23 18:08:00.327304: step: 1408/1541, loss: 0.00016091160068754107 2023-01-23 18:08:01.211207: step: 1410/1541, loss: 0.0003292854526080191 2023-01-23 18:08:02.095031: step: 1412/1541, loss: 0.0014846548438072205 2023-01-23 18:08:02.966582: step: 1414/1541, loss: 0.08035828173160553 2023-01-23 18:08:03.907126: step: 1416/1541, loss: 0.012561745010316372 2023-01-23 18:08:04.813746: step: 1418/1541, loss: 0.48422160744667053 2023-01-23 18:08:05.722026: step: 1420/1541, loss: 0.0015783201670274138 2023-01-23 18:08:06.698835: step: 1422/1541, loss: 0.0008096444071270525 2023-01-23 18:08:07.594450: step: 1424/1541, loss: 0.002300953958183527 2023-01-23 18:08:08.520249: step: 1426/1541, loss: 0.010426908731460571 2023-01-23 18:08:09.404879: step: 1428/1541, loss: 0.005757657345384359 2023-01-23 18:08:10.299181: step: 1430/1541, loss: 0.0010709459893405437 2023-01-23 18:08:11.212080: step: 1432/1541, loss: 0.18250803649425507 2023-01-23 18:08:12.074062: step: 1434/1541, loss: 0.0003857340198010206 2023-01-23 18:08:13.004552: step: 1436/1541, loss: 0.06514520198106766 2023-01-23 18:08:13.918689: step: 1438/1541, loss: 0.011234440840780735 2023-01-23 18:08:14.861588: step: 1440/1541, loss: 0.060444917529821396 2023-01-23 18:08:15.748264: step: 1442/1541, loss: 0.0008176093106158078 2023-01-23 18:08:16.656656: step: 1444/1541, loss: 0.0005140351131558418 2023-01-23 18:08:17.592935: step: 1446/1541, loss: 0.0035194160882383585 2023-01-23 18:08:18.468117: step: 1448/1541, loss: 0.0047545162960886955 2023-01-23 18:08:19.374263: step: 1450/1541, loss: 0.1139678880572319 2023-01-23 18:08:20.292159: step: 1452/1541, loss: 0.13685403764247894 2023-01-23 18:08:21.172254: step: 1454/1541, loss: 0.06177398934960365 2023-01-23 18:08:22.087136: step: 1456/1541, loss: 0.08603740483522415 2023-01-23 18:08:22.959066: step: 1458/1541, loss: 0.4707818031311035 2023-01-23 18:08:23.857029: step: 1460/1541, loss: 6.851132638985291e-05 2023-01-23 18:08:24.778224: step: 1462/1541, loss: 0.006689415778964758 2023-01-23 18:08:25.795358: step: 1464/1541, loss: 0.11040103435516357 2023-01-23 18:08:26.671851: step: 1466/1541, loss: 6.64143226458691e-05 2023-01-23 18:08:27.568737: step: 1468/1541, loss: 0.007906744256615639 2023-01-23 18:08:28.399623: step: 1470/1541, loss: 0.019748084247112274 2023-01-23 18:08:29.289049: step: 1472/1541, loss: 0.09609241038560867 2023-01-23 18:08:30.182759: step: 1474/1541, loss: 0.00034811079967767 2023-01-23 18:08:31.066609: step: 1476/1541, loss: 0.0005840006051585078 2023-01-23 18:08:31.963400: step: 1478/1541, loss: 0.0010698391124606133 2023-01-23 18:08:32.838035: step: 1480/1541, loss: 0.0005451907636597753 2023-01-23 18:08:33.771148: step: 1482/1541, loss: 8.217854337999597e-05 2023-01-23 18:08:34.687573: step: 1484/1541, loss: 0.0002477490925230086 2023-01-23 18:08:35.577042: step: 1486/1541, loss: 0.05192507058382034 2023-01-23 18:08:36.453143: step: 1488/1541, loss: 0.005326913669705391 2023-01-23 18:08:37.418329: step: 1490/1541, loss: 0.013826964423060417 2023-01-23 18:08:38.290214: step: 1492/1541, loss: 0.002657443517819047 2023-01-23 18:08:39.174237: step: 1494/1541, loss: 0.03431977331638336 2023-01-23 18:08:40.061022: step: 1496/1541, loss: 0.0019371751695871353 2023-01-23 18:08:40.922571: step: 1498/1541, loss: 0.0032146000303328037 2023-01-23 18:08:41.858061: step: 1500/1541, loss: 0.04319760575890541 2023-01-23 18:08:42.748800: step: 1502/1541, loss: -1.1001549864886329e-05 2023-01-23 18:08:43.628757: step: 1504/1541, loss: 0.0010497367475181818 2023-01-23 18:08:44.532185: step: 1506/1541, loss: 0.0002482445852365345 2023-01-23 18:08:45.441846: step: 1508/1541, loss: 0.03111601248383522 2023-01-23 18:08:46.326218: step: 1510/1541, loss: 0.15555471181869507 2023-01-23 18:08:47.216681: step: 1512/1541, loss: 0.0008288726676255465 2023-01-23 18:08:48.089450: step: 1514/1541, loss: 0.07695981860160828 2023-01-23 18:08:48.972205: step: 1516/1541, loss: 0.2142041027545929 2023-01-23 18:08:49.877693: step: 1518/1541, loss: 0.00256509892642498 2023-01-23 18:08:50.757047: step: 1520/1541, loss: 0.027763990685343742 2023-01-23 18:08:51.618398: step: 1522/1541, loss: 0.002219155663624406 2023-01-23 18:08:52.546849: step: 1524/1541, loss: 0.02961801551282406 2023-01-23 18:08:53.466279: step: 1526/1541, loss: 0.06563322991132736 2023-01-23 18:08:54.379832: step: 1528/1541, loss: 0.038354791700839996 2023-01-23 18:08:55.260936: step: 1530/1541, loss: 0.009196115657687187 2023-01-23 18:08:56.182721: step: 1532/1541, loss: 2.2271335124969482 2023-01-23 18:08:57.115232: step: 1534/1541, loss: 0.00034498487366363406 2023-01-23 18:08:58.015817: step: 1536/1541, loss: 7.284778985194862e-05 2023-01-23 18:08:58.876461: step: 1538/1541, loss: 0.03455822914838791 2023-01-23 18:08:59.774416: step: 1540/1541, loss: 0.006010747980326414 2023-01-23 18:09:00.673643: step: 1542/1541, loss: 0.0021603796631097794 2023-01-23 18:09:01.582951: step: 1544/1541, loss: 0.0007201727712526917 2023-01-23 18:09:02.458501: step: 1546/1541, loss: 0.004183616954833269 2023-01-23 18:09:03.344919: step: 1548/1541, loss: 0.20508408546447754 2023-01-23 18:09:04.235604: step: 1550/1541, loss: 0.010987000539898872 2023-01-23 18:09:05.083114: step: 1552/1541, loss: 0.056321654468774796 2023-01-23 18:09:05.974889: step: 1554/1541, loss: 0.30840128660202026 2023-01-23 18:09:06.898481: step: 1556/1541, loss: 0.0581563338637352 2023-01-23 18:09:07.834809: step: 1558/1541, loss: 0.0010711891809478402 2023-01-23 18:09:08.734680: step: 1560/1541, loss: 0.04391020908951759 2023-01-23 18:09:09.598807: step: 1562/1541, loss: 0.003066526260226965 2023-01-23 18:09:10.503684: step: 1564/1541, loss: 0.2892928421497345 2023-01-23 18:09:11.532650: step: 1566/1541, loss: 0.007998084649443626 2023-01-23 18:09:12.406991: step: 1568/1541, loss: 0.0005694454885087907 2023-01-23 18:09:13.287270: step: 1570/1541, loss: 0.009333280846476555 2023-01-23 18:09:14.212809: step: 1572/1541, loss: 0.045881204307079315 2023-01-23 18:09:15.128342: step: 1574/1541, loss: 9.940074960468337e-06 2023-01-23 18:09:16.007680: step: 1576/1541, loss: 0.24623069167137146 2023-01-23 18:09:16.904303: step: 1578/1541, loss: 0.00016699641128070652 2023-01-23 18:09:17.752202: step: 1580/1541, loss: 0.0005862066755071282 2023-01-23 18:09:18.636711: step: 1582/1541, loss: 0.02655692771077156 2023-01-23 18:09:19.530450: step: 1584/1541, loss: 0.04341554269194603 2023-01-23 18:09:20.481255: step: 1586/1541, loss: 0.031086117029190063 2023-01-23 18:09:21.371832: step: 1588/1541, loss: 0.022222984582185745 2023-01-23 18:09:22.300675: step: 1590/1541, loss: 0.0868268832564354 2023-01-23 18:09:23.209974: step: 1592/1541, loss: 0.0017936615040525794 2023-01-23 18:09:24.141714: step: 1594/1541, loss: 0.005598028190433979 2023-01-23 18:09:25.036412: step: 1596/1541, loss: 0.06995315849781036 2023-01-23 18:09:25.936717: step: 1598/1541, loss: 0.0926317498087883 2023-01-23 18:09:26.848478: step: 1600/1541, loss: 0.00959932804107666 2023-01-23 18:09:27.707103: step: 1602/1541, loss: 0.0011564163723960519 2023-01-23 18:09:28.584121: step: 1604/1541, loss: 0.00026767048984766006 2023-01-23 18:09:29.449034: step: 1606/1541, loss: 0.07065402716398239 2023-01-23 18:09:30.350264: step: 1608/1541, loss: 0.25535520911216736 2023-01-23 18:09:31.253252: step: 1610/1541, loss: 0.0037865901831537485 2023-01-23 18:09:32.204035: step: 1612/1541, loss: 0.006906297989189625 2023-01-23 18:09:33.082547: step: 1614/1541, loss: 0.03123619593679905 2023-01-23 18:09:33.963189: step: 1616/1541, loss: 0.007719515822827816 2023-01-23 18:09:34.879164: step: 1618/1541, loss: 0.30815619230270386 2023-01-23 18:09:35.788299: step: 1620/1541, loss: 0.022463858127593994 2023-01-23 18:09:36.701207: step: 1622/1541, loss: 0.23170137405395508 2023-01-23 18:09:37.553489: step: 1624/1541, loss: 0.0028179679065942764 2023-01-23 18:09:38.455161: step: 1626/1541, loss: 0.06473864614963531 2023-01-23 18:09:39.334226: step: 1628/1541, loss: 0.016655931249260902 2023-01-23 18:09:40.217297: step: 1630/1541, loss: 0.040195781737565994 2023-01-23 18:09:41.133834: step: 1632/1541, loss: 0.07107111811637878 2023-01-23 18:09:42.035025: step: 1634/1541, loss: 0.02066994085907936 2023-01-23 18:09:42.941328: step: 1636/1541, loss: 0.1338619440793991 2023-01-23 18:09:43.843768: step: 1638/1541, loss: 0.5167928338050842 2023-01-23 18:09:44.802563: step: 1640/1541, loss: 0.016344349831342697 2023-01-23 18:09:45.670891: step: 1642/1541, loss: 0.011404043063521385 2023-01-23 18:09:46.584609: step: 1644/1541, loss: 0.0008112806244753301 2023-01-23 18:09:47.485560: step: 1646/1541, loss: 0.02859037183225155 2023-01-23 18:09:48.370924: step: 1648/1541, loss: 0.001630171900615096 2023-01-23 18:09:49.235371: step: 1650/1541, loss: 0.006121886428445578 2023-01-23 18:09:50.202934: step: 1652/1541, loss: 0.0002313627628609538 2023-01-23 18:09:51.048752: step: 1654/1541, loss: 0.008432356640696526 2023-01-23 18:09:51.909526: step: 1656/1541, loss: 0.013656620867550373 2023-01-23 18:09:52.790367: step: 1658/1541, loss: 0.11813082545995712 2023-01-23 18:09:53.680017: step: 1660/1541, loss: 0.008268271572887897 2023-01-23 18:09:54.590474: step: 1662/1541, loss: 0.002277936553582549 2023-01-23 18:09:55.506043: step: 1664/1541, loss: 0.02408611588180065 2023-01-23 18:09:56.414251: step: 1666/1541, loss: 0.0382218137383461 2023-01-23 18:09:57.325125: step: 1668/1541, loss: 0.12477504462003708 2023-01-23 18:09:58.217265: step: 1670/1541, loss: 0.00023398407211061567 2023-01-23 18:09:59.091264: step: 1672/1541, loss: 0.17564056813716888 2023-01-23 18:10:00.016016: step: 1674/1541, loss: 0.08208814263343811 2023-01-23 18:10:01.005703: step: 1676/1541, loss: 0.13249008357524872 2023-01-23 18:10:01.948148: step: 1678/1541, loss: 0.0005404255934990942 2023-01-23 18:10:02.826610: step: 1680/1541, loss: 0.01572316139936447 2023-01-23 18:10:03.750175: step: 1682/1541, loss: 0.17859908938407898 2023-01-23 18:10:04.679493: step: 1684/1541, loss: 0.002427362371236086 2023-01-23 18:10:05.562196: step: 1686/1541, loss: 0.0019752003718167543 2023-01-23 18:10:06.435946: step: 1688/1541, loss: 0.0006752380868420005 2023-01-23 18:10:07.310718: step: 1690/1541, loss: 0.0024852335918694735 2023-01-23 18:10:08.305015: step: 1692/1541, loss: 0.004663566127419472 2023-01-23 18:10:09.177869: step: 1694/1541, loss: 0.0010031788842752576 2023-01-23 18:10:10.101089: step: 1696/1541, loss: 0.00021743858815170825 2023-01-23 18:10:10.988066: step: 1698/1541, loss: 0.0022468678653240204 2023-01-23 18:10:11.915039: step: 1700/1541, loss: 0.0002157563139917329 2023-01-23 18:10:12.864671: step: 1702/1541, loss: 0.011448699980974197 2023-01-23 18:10:13.770672: step: 1704/1541, loss: 0.20118802785873413 2023-01-23 18:10:14.660547: step: 1706/1541, loss: 0.0007605870487168431 2023-01-23 18:10:15.564462: step: 1708/1541, loss: 0.007784541230648756 2023-01-23 18:10:16.480563: step: 1710/1541, loss: 0.05981840193271637 2023-01-23 18:10:17.385942: step: 1712/1541, loss: 0.016986742615699768 2023-01-23 18:10:18.325054: step: 1714/1541, loss: 0.0003229582798667252 2023-01-23 18:10:19.308021: step: 1716/1541, loss: 0.0003009136999025941 2023-01-23 18:10:20.210830: step: 1718/1541, loss: 0.09942114353179932 2023-01-23 18:10:21.118516: step: 1720/1541, loss: 0.1972118318080902 2023-01-23 18:10:22.016646: step: 1722/1541, loss: 0.011799145489931107 2023-01-23 18:10:22.931356: step: 1724/1541, loss: 0.017480218783020973 2023-01-23 18:10:23.860117: step: 1726/1541, loss: 0.17582203447818756 2023-01-23 18:10:24.816070: step: 1728/1541, loss: 0.33083999156951904 2023-01-23 18:10:25.711467: step: 1730/1541, loss: 0.49085062742233276 2023-01-23 18:10:26.608827: step: 1732/1541, loss: 0.006904331035912037 2023-01-23 18:10:27.440066: step: 1734/1541, loss: 0.00011698353773681447 2023-01-23 18:10:28.431629: step: 1736/1541, loss: 0.019269170239567757 2023-01-23 18:10:29.336817: step: 1738/1541, loss: 0.20717674493789673 2023-01-23 18:10:30.233699: step: 1740/1541, loss: 0.12908300757408142 2023-01-23 18:10:31.100027: step: 1742/1541, loss: 0.0002334842865820974 2023-01-23 18:10:31.985445: step: 1744/1541, loss: 0.019319668412208557 2023-01-23 18:10:32.880160: step: 1746/1541, loss: 0.0003449548385106027 2023-01-23 18:10:33.799000: step: 1748/1541, loss: 0.0023813778534531593 2023-01-23 18:10:34.696967: step: 1750/1541, loss: 0.0031894734129309654 2023-01-23 18:10:35.600162: step: 1752/1541, loss: 0.03997449204325676 2023-01-23 18:10:36.467625: step: 1754/1541, loss: 0.0006607082905247808 2023-01-23 18:10:37.341792: step: 1756/1541, loss: 0.027049526572227478 2023-01-23 18:10:38.338317: step: 1758/1541, loss: 0.08381082862615585 2023-01-23 18:10:39.215111: step: 1760/1541, loss: 0.06755110621452332 2023-01-23 18:10:40.126025: step: 1762/1541, loss: 0.0036634260322898626 2023-01-23 18:10:40.983802: step: 1764/1541, loss: 0.002044986467808485 2023-01-23 18:10:41.886783: step: 1766/1541, loss: 0.10468649864196777 2023-01-23 18:10:42.784346: step: 1768/1541, loss: 0.0009844242595136166 2023-01-23 18:10:43.693396: step: 1770/1541, loss: 0.021448541432619095 2023-01-23 18:10:44.581591: step: 1772/1541, loss: 0.048793449997901917 2023-01-23 18:10:45.491468: step: 1774/1541, loss: 0.089656300842762 2023-01-23 18:10:46.339264: step: 1776/1541, loss: 0.03485666960477829 2023-01-23 18:10:47.253528: step: 1778/1541, loss: 0.0012031354708597064 2023-01-23 18:10:48.162561: step: 1780/1541, loss: 0.17989027500152588 2023-01-23 18:10:49.012743: step: 1782/1541, loss: 0.00020440362277440727 2023-01-23 18:10:49.945933: step: 1784/1541, loss: 0.0006913394317962229 2023-01-23 18:10:50.868506: step: 1786/1541, loss: 0.0008943334105424583 2023-01-23 18:10:51.773883: step: 1788/1541, loss: 0.00156494474504143 2023-01-23 18:10:52.633887: step: 1790/1541, loss: 0.003023663302883506 2023-01-23 18:10:53.549742: step: 1792/1541, loss: 0.01887362450361252 2023-01-23 18:10:54.435442: step: 1794/1541, loss: 0.0004726998449768871 2023-01-23 18:10:55.306051: step: 1796/1541, loss: 1.1944214747927617e-05 2023-01-23 18:10:56.158046: step: 1798/1541, loss: 0.0013181845424696803 2023-01-23 18:10:57.032653: step: 1800/1541, loss: 0.0062447465024888515 2023-01-23 18:10:57.908624: step: 1802/1541, loss: 0.08644216507673264 2023-01-23 18:10:58.826112: step: 1804/1541, loss: 0.004776030778884888 2023-01-23 18:10:59.825635: step: 1806/1541, loss: 0.0008601301233284175 2023-01-23 18:11:00.789856: step: 1808/1541, loss: 0.0004334970435593277 2023-01-23 18:11:01.711171: step: 1810/1541, loss: 0.0005178871215321124 2023-01-23 18:11:02.644231: step: 1812/1541, loss: 8.219356823246926e-05 2023-01-23 18:11:03.535630: step: 1814/1541, loss: 0.11494307219982147 2023-01-23 18:11:04.435251: step: 1816/1541, loss: 0.10471523553133011 2023-01-23 18:11:05.308871: step: 1818/1541, loss: 0.010805428959429264 2023-01-23 18:11:06.227657: step: 1820/1541, loss: 0.0346485935151577 2023-01-23 18:11:07.134553: step: 1822/1541, loss: 0.3886757791042328 2023-01-23 18:11:08.040209: step: 1824/1541, loss: 0.013046403415501118 2023-01-23 18:11:08.914397: step: 1826/1541, loss: 0.19698357582092285 2023-01-23 18:11:09.796513: step: 1828/1541, loss: 0.008480205200612545 2023-01-23 18:11:10.736472: step: 1830/1541, loss: 0.07970651239156723 2023-01-23 18:11:11.664617: step: 1832/1541, loss: 0.025851808488368988 2023-01-23 18:11:12.556000: step: 1834/1541, loss: 0.0022868113592267036 2023-01-23 18:11:13.450784: step: 1836/1541, loss: 0.0004206479643471539 2023-01-23 18:11:14.347846: step: 1838/1541, loss: 0.00017236018902622163 2023-01-23 18:11:15.249903: step: 1840/1541, loss: 0.24532513320446014 2023-01-23 18:11:16.119594: step: 1842/1541, loss: 0.00937502458691597 2023-01-23 18:11:17.006574: step: 1844/1541, loss: 0.00034075917210429907 2023-01-23 18:11:17.944399: step: 1846/1541, loss: 0.0028711955528706312 2023-01-23 18:11:18.830769: step: 1848/1541, loss: 0.33708009123802185 2023-01-23 18:11:19.696994: step: 1850/1541, loss: 0.004454371985048056 2023-01-23 18:11:20.624682: step: 1852/1541, loss: 0.004120710305869579 2023-01-23 18:11:21.478479: step: 1854/1541, loss: 0.0002867320436052978 2023-01-23 18:11:22.371447: step: 1856/1541, loss: 0.006310540717095137 2023-01-23 18:11:23.273905: step: 1858/1541, loss: 0.017249401658773422 2023-01-23 18:11:24.140067: step: 1860/1541, loss: 0.0006020005093887448 2023-01-23 18:11:25.077857: step: 1862/1541, loss: 0.0009175211889669299 2023-01-23 18:11:26.013699: step: 1864/1541, loss: 0.5285756587982178 2023-01-23 18:11:26.963121: step: 1866/1541, loss: 0.12917903065681458 2023-01-23 18:11:27.839567: step: 1868/1541, loss: 0.0051902225241065025 2023-01-23 18:11:28.746301: step: 1870/1541, loss: 0.0009586201049387455 2023-01-23 18:11:29.648960: step: 1872/1541, loss: 0.0053427801467478275 2023-01-23 18:11:30.570771: step: 1874/1541, loss: 0.004061682615429163 2023-01-23 18:11:31.493831: step: 1876/1541, loss: 0.20635835826396942 2023-01-23 18:11:32.386739: step: 1878/1541, loss: 0.004339421633630991 2023-01-23 18:11:33.274763: step: 1880/1541, loss: 0.0006343177519738674 2023-01-23 18:11:34.215641: step: 1882/1541, loss: 0.00040377123514190316 2023-01-23 18:11:35.128943: step: 1884/1541, loss: 0.01770339161157608 2023-01-23 18:11:36.084331: step: 1886/1541, loss: 0.2975214123725891 2023-01-23 18:11:37.005218: step: 1888/1541, loss: 0.007107188459485769 2023-01-23 18:11:37.889214: step: 1890/1541, loss: 0.000643803970888257 2023-01-23 18:11:38.775357: step: 1892/1541, loss: 0.0005853007896803319 2023-01-23 18:11:39.677477: step: 1894/1541, loss: 0.00010633537021931261 2023-01-23 18:11:40.567914: step: 1896/1541, loss: 0.01962308958172798 2023-01-23 18:11:41.446810: step: 1898/1541, loss: 0.0007006798987276852 2023-01-23 18:11:42.410436: step: 1900/1541, loss: 0.07584288716316223 2023-01-23 18:11:43.329224: step: 1902/1541, loss: 0.002122380305081606 2023-01-23 18:11:44.237985: step: 1904/1541, loss: 0.03429434075951576 2023-01-23 18:11:45.096117: step: 1906/1541, loss: 0.0027912775985896587 2023-01-23 18:11:45.954834: step: 1908/1541, loss: 0.025581281632184982 2023-01-23 18:11:46.875779: step: 1910/1541, loss: 0.0009715604246594012 2023-01-23 18:11:47.788894: step: 1912/1541, loss: 0.00015718437498435378 2023-01-23 18:11:48.710743: step: 1914/1541, loss: 0.0014504272257909179 2023-01-23 18:11:49.615737: step: 1916/1541, loss: 0.00044283774332143366 2023-01-23 18:11:50.530852: step: 1918/1541, loss: 0.0464196503162384 2023-01-23 18:11:51.415295: step: 1920/1541, loss: 0.0003973890852648765 2023-01-23 18:11:52.306373: step: 1922/1541, loss: 0.19226078689098358 2023-01-23 18:11:53.195143: step: 1924/1541, loss: 0.0006519260932691395 2023-01-23 18:11:54.072364: step: 1926/1541, loss: 0.05312201753258705 2023-01-23 18:11:54.940373: step: 1928/1541, loss: 0.0022634370252490044 2023-01-23 18:11:55.831126: step: 1930/1541, loss: 0.0003040932642761618 2023-01-23 18:11:56.748670: step: 1932/1541, loss: 0.027463510632514954 2023-01-23 18:11:57.671674: step: 1934/1541, loss: 0.022005388513207436 2023-01-23 18:11:58.701007: step: 1936/1541, loss: 0.0013179662637412548 2023-01-23 18:11:59.581850: step: 1938/1541, loss: 0.00015087216161191463 2023-01-23 18:12:00.554895: step: 1940/1541, loss: 0.057231880724430084 2023-01-23 18:12:01.492708: step: 1942/1541, loss: -3.814764568232931e-06 2023-01-23 18:12:02.349229: step: 1944/1541, loss: 0.0005998459528200328 2023-01-23 18:12:03.240254: step: 1946/1541, loss: 0.00015732257452327758 2023-01-23 18:12:04.154079: step: 1948/1541, loss: 0.00010320216097170487 2023-01-23 18:12:05.071749: step: 1950/1541, loss: 0.09640910476446152 2023-01-23 18:12:05.999974: step: 1952/1541, loss: 0.07175610959529877 2023-01-23 18:12:06.879932: step: 1954/1541, loss: 0.2897128760814667 2023-01-23 18:12:07.759603: step: 1956/1541, loss: 0.08497476577758789 2023-01-23 18:12:08.631117: step: 1958/1541, loss: 0.006855321582406759 2023-01-23 18:12:09.602337: step: 1960/1541, loss: 0.15419892966747284 2023-01-23 18:12:10.488518: step: 1962/1541, loss: 0.040945276618003845 2023-01-23 18:12:11.364402: step: 1964/1541, loss: 0.08201431483030319 2023-01-23 18:12:12.230390: step: 1966/1541, loss: 0.01126929558813572 2023-01-23 18:12:13.106392: step: 1968/1541, loss: 0.009581646881997585 2023-01-23 18:12:14.033430: step: 1970/1541, loss: 0.010596594773232937 2023-01-23 18:12:14.963826: step: 1972/1541, loss: 0.007599834818392992 2023-01-23 18:12:15.870405: step: 1974/1541, loss: 0.06066419556736946 2023-01-23 18:12:16.771211: step: 1976/1541, loss: 0.006515652406960726 2023-01-23 18:12:17.699290: step: 1978/1541, loss: 0.042950212955474854 2023-01-23 18:12:18.599476: step: 1980/1541, loss: 0.1812545210123062 2023-01-23 18:12:19.468545: step: 1982/1541, loss: 0.21801821887493134 2023-01-23 18:12:20.436546: step: 1984/1541, loss: 3.767695670831017e-05 2023-01-23 18:12:21.365591: step: 1986/1541, loss: 0.2782556414604187 2023-01-23 18:12:22.288538: step: 1988/1541, loss: 0.1693250834941864 2023-01-23 18:12:23.203961: step: 1990/1541, loss: 0.003363337367773056 2023-01-23 18:12:24.114462: step: 1992/1541, loss: 0.12432508915662766 2023-01-23 18:12:25.020836: step: 1994/1541, loss: 0.00037906435318291187 2023-01-23 18:12:25.938342: step: 1996/1541, loss: 6.653399032074958e-05 2023-01-23 18:12:26.835776: step: 1998/1541, loss: 0.3562543988227844 2023-01-23 18:12:27.787102: step: 2000/1541, loss: 0.013602659106254578 2023-01-23 18:12:28.726472: step: 2002/1541, loss: 0.011515368707478046 2023-01-23 18:12:29.616630: step: 2004/1541, loss: 0.0015630921116098762 2023-01-23 18:12:30.554504: step: 2006/1541, loss: 0.0002885697758756578 2023-01-23 18:12:31.461513: step: 2008/1541, loss: 0.0052816192619502544 2023-01-23 18:12:32.373189: step: 2010/1541, loss: 0.0017050276510417461 2023-01-23 18:12:33.231297: step: 2012/1541, loss: 0.003464616136625409 2023-01-23 18:12:34.164364: step: 2014/1541, loss: 0.01878969557583332 2023-01-23 18:12:35.084875: step: 2016/1541, loss: 0.07367933541536331 2023-01-23 18:12:36.043360: step: 2018/1541, loss: 0.0004644588043447584 2023-01-23 18:12:36.919255: step: 2020/1541, loss: 0.28444787859916687 2023-01-23 18:12:37.821808: step: 2022/1541, loss: 0.8217397332191467 2023-01-23 18:12:38.770023: step: 2024/1541, loss: 0.0003537094744388014 2023-01-23 18:12:39.635822: step: 2026/1541, loss: 0.016239216551184654 2023-01-23 18:12:40.508274: step: 2028/1541, loss: 0.021111559122800827 2023-01-23 18:12:41.406993: step: 2030/1541, loss: 0.01172546111047268 2023-01-23 18:12:42.298949: step: 2032/1541, loss: 0.0007670692284591496 2023-01-23 18:12:43.197049: step: 2034/1541, loss: 0.05070085823535919 2023-01-23 18:12:44.216251: step: 2036/1541, loss: 0.1503676027059555 2023-01-23 18:12:45.086396: step: 2038/1541, loss: 0.008908127434551716 2023-01-23 18:12:45.957600: step: 2040/1541, loss: 0.05859752371907234 2023-01-23 18:12:46.839110: step: 2042/1541, loss: 0.4116576910018921 2023-01-23 18:12:47.734515: step: 2044/1541, loss: 0.0882120132446289 2023-01-23 18:12:48.620444: step: 2046/1541, loss: 0.0016882955096662045 2023-01-23 18:12:49.535281: step: 2048/1541, loss: 0.00029423984233289957 2023-01-23 18:12:50.503153: step: 2050/1541, loss: 0.06967988610267639 2023-01-23 18:12:51.425665: step: 2052/1541, loss: 0.10030999034643173 2023-01-23 18:12:52.378553: step: 2054/1541, loss: 0.0010138388024643064 2023-01-23 18:12:53.253685: step: 2056/1541, loss: 0.07975471019744873 2023-01-23 18:12:54.147113: step: 2058/1541, loss: 0.5726033449172974 2023-01-23 18:12:55.025045: step: 2060/1541, loss: 0.09812993556261063 2023-01-23 18:12:55.926311: step: 2062/1541, loss: 0.10084138810634613 2023-01-23 18:12:56.839373: step: 2064/1541, loss: 0.0031728779431432486 2023-01-23 18:12:57.712878: step: 2066/1541, loss: 0.058111246675252914 2023-01-23 18:12:58.591345: step: 2068/1541, loss: 0.0001478155463701114 2023-01-23 18:12:59.498306: step: 2070/1541, loss: 0.03162414953112602 2023-01-23 18:13:00.390777: step: 2072/1541, loss: 0.032793644815683365 2023-01-23 18:13:01.330123: step: 2074/1541, loss: 0.011169077828526497 2023-01-23 18:13:02.232968: step: 2076/1541, loss: 0.032905302941799164 2023-01-23 18:13:03.191661: step: 2078/1541, loss: 0.0020027298014611006 2023-01-23 18:13:04.079421: step: 2080/1541, loss: 0.0034937229938805103 2023-01-23 18:13:04.999304: step: 2082/1541, loss: 0.20012491941452026 2023-01-23 18:13:05.914471: step: 2084/1541, loss: 0.0157973300665617 2023-01-23 18:13:06.794921: step: 2086/1541, loss: 0.0013010312104597688 2023-01-23 18:13:07.671846: step: 2088/1541, loss: 0.03589301556348801 2023-01-23 18:13:08.569397: step: 2090/1541, loss: 0.04942098632454872 2023-01-23 18:13:09.426720: step: 2092/1541, loss: 0.0007058686460368335 2023-01-23 18:13:10.358623: step: 2094/1541, loss: 0.002912141615524888 2023-01-23 18:13:11.281885: step: 2096/1541, loss: 0.0813068076968193 2023-01-23 18:13:12.191004: step: 2098/1541, loss: 0.0016939161578193307 2023-01-23 18:13:13.076765: step: 2100/1541, loss: 0.04576407000422478 2023-01-23 18:13:13.979383: step: 2102/1541, loss: 0.011441493406891823 2023-01-23 18:13:14.882701: step: 2104/1541, loss: 0.0005600740551017225 2023-01-23 18:13:15.737181: step: 2106/1541, loss: 0.00015069960500113666 2023-01-23 18:13:16.677598: step: 2108/1541, loss: 2.9603341317852028e-05 2023-01-23 18:13:17.556161: step: 2110/1541, loss: 0.23106510937213898 2023-01-23 18:13:18.452365: step: 2112/1541, loss: 0.2746286988258362 2023-01-23 18:13:19.339191: step: 2114/1541, loss: 0.000668401422444731 2023-01-23 18:13:20.263034: step: 2116/1541, loss: 0.0045879557728767395 2023-01-23 18:13:21.129095: step: 2118/1541, loss: 0.0004387862572912127 2023-01-23 18:13:22.018701: step: 2120/1541, loss: 0.030625484883785248 2023-01-23 18:13:22.894161: step: 2122/1541, loss: 0.0008098557009361684 2023-01-23 18:13:23.795972: step: 2124/1541, loss: 0.0026755831204354763 2023-01-23 18:13:24.758960: step: 2126/1541, loss: 0.010462518781423569 2023-01-23 18:13:25.705801: step: 2128/1541, loss: 0.09437929093837738 2023-01-23 18:13:26.650633: step: 2130/1541, loss: 0.0007009065011516213 2023-01-23 18:13:27.567651: step: 2132/1541, loss: 0.09063167870044708 2023-01-23 18:13:28.464371: step: 2134/1541, loss: 0.00011686504876706749 2023-01-23 18:13:29.348718: step: 2136/1541, loss: 3.196180477971211e-05 2023-01-23 18:13:30.253840: step: 2138/1541, loss: 0.2209702581167221 2023-01-23 18:13:31.152342: step: 2140/1541, loss: 0.001456225523725152 2023-01-23 18:13:32.083590: step: 2142/1541, loss: 0.006095658056437969 2023-01-23 18:13:32.985344: step: 2144/1541, loss: 0.224302738904953 2023-01-23 18:13:33.870224: step: 2146/1541, loss: 0.007373901084065437 2023-01-23 18:13:34.801208: step: 2148/1541, loss: 0.002499524038285017 2023-01-23 18:13:35.680471: step: 2150/1541, loss: 0.00011895864008693025 2023-01-23 18:13:36.590787: step: 2152/1541, loss: 8.695227734278888e-05 2023-01-23 18:13:37.478490: step: 2154/1541, loss: 0.005102814175188541 2023-01-23 18:13:38.394581: step: 2156/1541, loss: 0.0017676895949989557 2023-01-23 18:13:39.291659: step: 2158/1541, loss: 0.00025718213873915374 2023-01-23 18:13:40.173303: step: 2160/1541, loss: 0.0008543035364709795 2023-01-23 18:13:41.057915: step: 2162/1541, loss: 0.015940561890602112 2023-01-23 18:13:41.939152: step: 2164/1541, loss: 0.09479404985904694 2023-01-23 18:13:42.853895: step: 2166/1541, loss: 0.037702858448028564 2023-01-23 18:13:43.748048: step: 2168/1541, loss: 0.002434241585433483 2023-01-23 18:13:44.620768: step: 2170/1541, loss: 0.10230700671672821 2023-01-23 18:13:45.536132: step: 2172/1541, loss: 0.15930557250976562 2023-01-23 18:13:46.404904: step: 2174/1541, loss: 0.6925547122955322 2023-01-23 18:13:47.267908: step: 2176/1541, loss: 0.000695912865921855 2023-01-23 18:13:48.195291: step: 2178/1541, loss: 0.0004184640711173415 2023-01-23 18:13:49.119714: step: 2180/1541, loss: 0.028089361265301704 2023-01-23 18:13:49.999693: step: 2182/1541, loss: 0.0006906273774802685 2023-01-23 18:13:50.892633: step: 2184/1541, loss: 0.0034716317895799875 2023-01-23 18:13:51.742147: step: 2186/1541, loss: 0.0005062692798674107 2023-01-23 18:13:52.654971: step: 2188/1541, loss: 0.0038592100609093904 2023-01-23 18:13:53.611676: step: 2190/1541, loss: 0.0005354949971660972 2023-01-23 18:13:54.476370: step: 2192/1541, loss: 0.0010868821991607547 2023-01-23 18:13:55.356424: step: 2194/1541, loss: 0.05701693519949913 2023-01-23 18:13:56.255630: step: 2196/1541, loss: 0.0012697404017671943 2023-01-23 18:13:57.144931: step: 2198/1541, loss: 0.05309101194143295 2023-01-23 18:13:58.023711: step: 2200/1541, loss: 0.00013580468657892197 2023-01-23 18:13:58.960973: step: 2202/1541, loss: 0.018033307045698166 2023-01-23 18:13:59.843661: step: 2204/1541, loss: 0.00011139816342620179 2023-01-23 18:14:00.771360: step: 2206/1541, loss: 0.02626320905983448 2023-01-23 18:14:01.696459: step: 2208/1541, loss: 0.0005806786939501762 2023-01-23 18:14:02.616566: step: 2210/1541, loss: 0.061358384788036346 2023-01-23 18:14:03.580964: step: 2212/1541, loss: 0.19656242430210114 2023-01-23 18:14:04.495340: step: 2214/1541, loss: 0.1670876294374466 2023-01-23 18:14:05.330122: step: 2216/1541, loss: 0.0006587943062186241 2023-01-23 18:14:06.259554: step: 2218/1541, loss: 0.1919236034154892 2023-01-23 18:14:07.196610: step: 2220/1541, loss: 0.007075579836964607 2023-01-23 18:14:08.084955: step: 2222/1541, loss: 0.12448073923587799 2023-01-23 18:14:08.967730: step: 2224/1541, loss: 0.003405404044315219 2023-01-23 18:14:09.863900: step: 2226/1541, loss: 0.00024617131566628814 2023-01-23 18:14:10.765865: step: 2228/1541, loss: 0.842531144618988 2023-01-23 18:14:11.653516: step: 2230/1541, loss: 0.03578600287437439 2023-01-23 18:14:12.583864: step: 2232/1541, loss: 0.0619698129594326 2023-01-23 18:14:13.497755: step: 2234/1541, loss: 0.0022883578203618526 2023-01-23 18:14:14.409447: step: 2236/1541, loss: 0.0008101777639240026 2023-01-23 18:14:15.294419: step: 2238/1541, loss: 0.0009915699483826756 2023-01-23 18:14:16.191326: step: 2240/1541, loss: 0.00242483033798635 2023-01-23 18:14:17.063162: step: 2242/1541, loss: 1.4705027341842651 2023-01-23 18:14:17.959764: step: 2244/1541, loss: 0.01434313040226698 2023-01-23 18:14:18.817822: step: 2246/1541, loss: 5.07219519931823e-05 2023-01-23 18:14:19.671646: step: 2248/1541, loss: 0.0005651850369758904 2023-01-23 18:14:20.561320: step: 2250/1541, loss: 0.0013607708970084786 2023-01-23 18:14:21.514020: step: 2252/1541, loss: 0.0741920918226242 2023-01-23 18:14:22.434109: step: 2254/1541, loss: 0.000181040886673145 2023-01-23 18:14:23.350723: step: 2256/1541, loss: 0.041135355830192566 2023-01-23 18:14:24.181800: step: 2258/1541, loss: 0.00013220476103015244 2023-01-23 18:14:25.058260: step: 2260/1541, loss: 0.18870489299297333 2023-01-23 18:14:25.935567: step: 2262/1541, loss: 0.0013806151691824198 2023-01-23 18:14:26.839016: step: 2264/1541, loss: 0.000769465696066618 2023-01-23 18:14:27.723227: step: 2266/1541, loss: 0.001346109202131629 2023-01-23 18:14:28.596439: step: 2268/1541, loss: 0.29526957869529724 2023-01-23 18:14:29.492094: step: 2270/1541, loss: 0.0009219835628755391 2023-01-23 18:14:30.364360: step: 2272/1541, loss: 0.009830167517066002 2023-01-23 18:14:31.256159: step: 2274/1541, loss: 0.1046195924282074 2023-01-23 18:14:32.237520: step: 2276/1541, loss: 0.0008115672972053289 2023-01-23 18:14:33.160472: step: 2278/1541, loss: 0.0007638094248250127 2023-01-23 18:14:34.075591: step: 2280/1541, loss: 0.00013652608322445303 2023-01-23 18:14:35.013081: step: 2282/1541, loss: 0.025853093713521957 2023-01-23 18:14:35.942323: step: 2284/1541, loss: 0.006385211832821369 2023-01-23 18:14:36.845137: step: 2286/1541, loss: 0.005818115547299385 2023-01-23 18:14:37.762871: step: 2288/1541, loss: 0.0019103179220110178 2023-01-23 18:14:38.637678: step: 2290/1541, loss: 0.31159618496894836 2023-01-23 18:14:39.522872: step: 2292/1541, loss: 0.0006595840095542371 2023-01-23 18:14:40.462535: step: 2294/1541, loss: 0.01644640415906906 2023-01-23 18:14:41.352218: step: 2296/1541, loss: 0.0025672968477010727 2023-01-23 18:14:42.297217: step: 2298/1541, loss: 0.0003403241280466318 2023-01-23 18:14:43.145000: step: 2300/1541, loss: 9.874928218778223e-05 2023-01-23 18:14:44.025296: step: 2302/1541, loss: 1.055396556854248 2023-01-23 18:14:44.928309: step: 2304/1541, loss: 0.03297737240791321 2023-01-23 18:14:45.841323: step: 2306/1541, loss: 0.0009578409953974187 2023-01-23 18:14:46.732297: step: 2308/1541, loss: 0.015832703560590744 2023-01-23 18:14:47.657622: step: 2310/1541, loss: 0.0029071669559925795 2023-01-23 18:14:48.539207: step: 2312/1541, loss: 0.324531227350235 2023-01-23 18:14:49.555477: step: 2314/1541, loss: 0.0019952065777033567 2023-01-23 18:14:50.449052: step: 2316/1541, loss: 0.0027589919045567513 2023-01-23 18:14:51.352128: step: 2318/1541, loss: 0.0014315582811832428 2023-01-23 18:14:52.358726: step: 2320/1541, loss: 0.0001381195179419592 2023-01-23 18:14:53.325667: step: 2322/1541, loss: 0.0062641422264277935 2023-01-23 18:14:54.197024: step: 2324/1541, loss: 2.2619862647843547e-05 2023-01-23 18:14:55.095170: step: 2326/1541, loss: 0.1501186639070511 2023-01-23 18:14:56.015685: step: 2328/1541, loss: -4.670405360229779e-06 2023-01-23 18:14:56.987742: step: 2330/1541, loss: 0.005278830416500568 2023-01-23 18:14:57.866724: step: 2332/1541, loss: 0.02781089022755623 2023-01-23 18:14:58.741594: step: 2334/1541, loss: 0.0006154695875011384 2023-01-23 18:14:59.675601: step: 2336/1541, loss: 0.0005012602196075022 2023-01-23 18:15:00.547677: step: 2338/1541, loss: 4.0152372093871236e-05 2023-01-23 18:15:01.446117: step: 2340/1541, loss: 0.0010874142171815038 2023-01-23 18:15:02.347518: step: 2342/1541, loss: 0.00021710195869673043 2023-01-23 18:15:03.198195: step: 2344/1541, loss: 0.0025206191930919886 2023-01-23 18:15:04.158076: step: 2346/1541, loss: 0.026718497276306152 2023-01-23 18:15:05.050921: step: 2348/1541, loss: 0.055223122239112854 2023-01-23 18:15:05.963375: step: 2350/1541, loss: 0.00039250755799002945 2023-01-23 18:15:06.874225: step: 2352/1541, loss: 9.554425196256489e-05 2023-01-23 18:15:07.765774: step: 2354/1541, loss: 5.117478940519504e-05 2023-01-23 18:15:08.712747: step: 2356/1541, loss: 6.518003647215664e-05 2023-01-23 18:15:09.641685: step: 2358/1541, loss: 0.15193647146224976 2023-01-23 18:15:10.504362: step: 2360/1541, loss: 0.01739172637462616 2023-01-23 18:15:11.424268: step: 2362/1541, loss: 0.040216315537691116 2023-01-23 18:15:12.285543: step: 2364/1541, loss: 0.018409155309200287 2023-01-23 18:15:13.188695: step: 2366/1541, loss: 0.15885359048843384 2023-01-23 18:15:14.077388: step: 2368/1541, loss: 0.03833479806780815 2023-01-23 18:15:14.951191: step: 2370/1541, loss: 4.2938641854561865e-05 2023-01-23 18:15:15.859735: step: 2372/1541, loss: 0.07530826330184937 2023-01-23 18:15:16.800757: step: 2374/1541, loss: 1.3511402606964111 2023-01-23 18:15:17.721359: step: 2376/1541, loss: 0.0032757744193077087 2023-01-23 18:15:18.629685: step: 2378/1541, loss: 0.1624983698129654 2023-01-23 18:15:19.542458: step: 2380/1541, loss: 0.020851684734225273 2023-01-23 18:15:20.458248: step: 2382/1541, loss: 0.00918244943022728 2023-01-23 18:15:21.392870: step: 2384/1541, loss: 0.07280849665403366 2023-01-23 18:15:22.276488: step: 2386/1541, loss: 0.0036601300816982985 2023-01-23 18:15:23.176495: step: 2388/1541, loss: 6.665595719823614e-05 2023-01-23 18:15:24.069499: step: 2390/1541, loss: 0.008424720726907253 2023-01-23 18:15:25.004881: step: 2392/1541, loss: 0.2556552588939667 2023-01-23 18:15:25.926653: step: 2394/1541, loss: 1.4093421668803785e-05 2023-01-23 18:15:26.789987: step: 2396/1541, loss: 0.22249384224414825 2023-01-23 18:15:27.679352: step: 2398/1541, loss: 1.9880028048646636e-05 2023-01-23 18:15:28.563922: step: 2400/1541, loss: 0.0002010466268984601 2023-01-23 18:15:29.480648: step: 2402/1541, loss: 0.00010194922651862726 2023-01-23 18:15:30.354022: step: 2404/1541, loss: 0.0028809839859604836 2023-01-23 18:15:31.215688: step: 2406/1541, loss: 0.08835531026124954 2023-01-23 18:15:32.072609: step: 2408/1541, loss: 0.0010119066573679447 2023-01-23 18:15:32.988619: step: 2410/1541, loss: 0.00038403156213462353 2023-01-23 18:15:33.862413: step: 2412/1541, loss: 0.12183995544910431 2023-01-23 18:15:34.746109: step: 2414/1541, loss: 0.0006202144431881607 2023-01-23 18:15:35.637224: step: 2416/1541, loss: 0.04998067766427994 2023-01-23 18:15:36.503919: step: 2418/1541, loss: 0.00028748708427883685 2023-01-23 18:15:37.421555: step: 2420/1541, loss: 0.0002963546721730381 2023-01-23 18:15:38.248900: step: 2422/1541, loss: 3.784165528486483e-05 2023-01-23 18:15:39.141519: step: 2424/1541, loss: 0.014983449131250381 2023-01-23 18:15:40.047614: step: 2426/1541, loss: 0.002483648480847478 2023-01-23 18:15:40.990440: step: 2428/1541, loss: 0.0003531498950906098 2023-01-23 18:15:41.886030: step: 2430/1541, loss: 0.0719388946890831 2023-01-23 18:15:42.773708: step: 2432/1541, loss: 0.008320293389260769 2023-01-23 18:15:43.670459: step: 2434/1541, loss: 0.0016220881370827556 2023-01-23 18:15:44.711630: step: 2436/1541, loss: 0.00397009402513504 2023-01-23 18:15:45.599510: step: 2438/1541, loss: 0.00046869803918525577 2023-01-23 18:15:46.513083: step: 2440/1541, loss: 0.22609467804431915 2023-01-23 18:15:47.404676: step: 2442/1541, loss: 0.023000825196504593 2023-01-23 18:15:48.293274: step: 2444/1541, loss: 1.3434469110507052e-06 2023-01-23 18:15:49.153999: step: 2446/1541, loss: 4.1308179788757116e-05 2023-01-23 18:15:50.037750: step: 2448/1541, loss: 0.04226816073060036 2023-01-23 18:15:50.948272: step: 2450/1541, loss: 0.00836515799164772 2023-01-23 18:15:51.885962: step: 2452/1541, loss: 0.0012371486518532038 2023-01-23 18:15:52.789965: step: 2454/1541, loss: 6.21393191977404e-05 2023-01-23 18:15:53.643455: step: 2456/1541, loss: 0.0009167725220322609 2023-01-23 18:15:54.480610: step: 2458/1541, loss: 0.23938317596912384 2023-01-23 18:15:55.346005: step: 2460/1541, loss: 0.00011027356231352314 2023-01-23 18:15:56.224270: step: 2462/1541, loss: 0.00031996911275200546 2023-01-23 18:15:57.134992: step: 2464/1541, loss: 8.96900673978962e-05 2023-01-23 18:15:58.097335: step: 2466/1541, loss: 0.0005524064181372523 2023-01-23 18:15:59.014604: step: 2468/1541, loss: 0.1136452928185463 2023-01-23 18:15:59.893068: step: 2470/1541, loss: 0.018361041322350502 2023-01-23 18:16:00.788610: step: 2472/1541, loss: 0.0005641893367283046 2023-01-23 18:16:01.668314: step: 2474/1541, loss: 0.00015202070062514395 2023-01-23 18:16:02.634664: step: 2476/1541, loss: 0.029269956052303314 2023-01-23 18:16:03.607417: step: 2478/1541, loss: 0.007080934010446072 2023-01-23 18:16:04.475337: step: 2480/1541, loss: 0.29278427362442017 2023-01-23 18:16:05.366889: step: 2482/1541, loss: 0.0004226568271405995 2023-01-23 18:16:06.235974: step: 2484/1541, loss: 0.00045886519365012646 2023-01-23 18:16:07.134347: step: 2486/1541, loss: 0.27330076694488525 2023-01-23 18:16:08.049537: step: 2488/1541, loss: 0.00036020364495925605 2023-01-23 18:16:08.920298: step: 2490/1541, loss: 0.1215142235159874 2023-01-23 18:16:09.812405: step: 2492/1541, loss: 0.0001534590555820614 2023-01-23 18:16:10.688299: step: 2494/1541, loss: 3.1450970709556714e-05 2023-01-23 18:16:11.604344: step: 2496/1541, loss: 0.9248500466346741 2023-01-23 18:16:12.492762: step: 2498/1541, loss: 4.899590203422122e-05 2023-01-23 18:16:13.365851: step: 2500/1541, loss: 0.003414566395804286 2023-01-23 18:16:14.237501: step: 2502/1541, loss: 0.013755854219198227 2023-01-23 18:16:15.141745: step: 2504/1541, loss: 0.06473945081233978 2023-01-23 18:16:16.075749: step: 2506/1541, loss: 0.05215095728635788 2023-01-23 18:16:17.066310: step: 2508/1541, loss: 0.7049499154090881 2023-01-23 18:16:17.983967: step: 2510/1541, loss: 0.03586241602897644 2023-01-23 18:16:18.869127: step: 2512/1541, loss: 0.09391242265701294 2023-01-23 18:16:19.740278: step: 2514/1541, loss: 0.056086257100105286 2023-01-23 18:16:20.631197: step: 2516/1541, loss: 0.009614570997655392 2023-01-23 18:16:21.595005: step: 2518/1541, loss: 0.1393987089395523 2023-01-23 18:16:22.529885: step: 2520/1541, loss: 0.0092454319819808 2023-01-23 18:16:23.388277: step: 2522/1541, loss: 0.009040690958499908 2023-01-23 18:16:24.293540: step: 2524/1541, loss: 0.008891330100595951 2023-01-23 18:16:25.218785: step: 2526/1541, loss: 0.0032027519773691893 2023-01-23 18:16:26.065970: step: 2528/1541, loss: 0.045544590801000595 2023-01-23 18:16:26.941122: step: 2530/1541, loss: 0.007327056489884853 2023-01-23 18:16:27.815382: step: 2532/1541, loss: 0.001745646120980382 2023-01-23 18:16:28.783146: step: 2534/1541, loss: 0.001535756979137659 2023-01-23 18:16:29.688824: step: 2536/1541, loss: 0.002522167982533574 2023-01-23 18:16:30.646635: step: 2538/1541, loss: 0.0012190289562568069 2023-01-23 18:16:31.575133: step: 2540/1541, loss: 0.03879240155220032 2023-01-23 18:16:32.464667: step: 2542/1541, loss: 0.01257464848458767 2023-01-23 18:16:33.423204: step: 2544/1541, loss: 0.0005364675307646394 2023-01-23 18:16:34.304098: step: 2546/1541, loss: 0.0015974562847986817 2023-01-23 18:16:35.211813: step: 2548/1541, loss: 0.0003425382310524583 2023-01-23 18:16:36.097477: step: 2550/1541, loss: 0.007122629322111607 2023-01-23 18:16:37.044024: step: 2552/1541, loss: 0.0006037611747160554 2023-01-23 18:16:37.954091: step: 2554/1541, loss: 0.06559842824935913 2023-01-23 18:16:38.836340: step: 2556/1541, loss: 0.006414417643100023 2023-01-23 18:16:39.717697: step: 2558/1541, loss: 0.0043794275261461735 2023-01-23 18:16:40.554223: step: 2560/1541, loss: 0.0021180901676416397 2023-01-23 18:16:41.459767: step: 2562/1541, loss: 0.21857784688472748 2023-01-23 18:16:42.341644: step: 2564/1541, loss: 0.0008940503466874361 2023-01-23 18:16:43.229149: step: 2566/1541, loss: 0.21974501013755798 2023-01-23 18:16:44.168923: step: 2568/1541, loss: 0.10452856868505478 2023-01-23 18:16:45.069321: step: 2570/1541, loss: 0.0786261036992073 2023-01-23 18:16:45.982856: step: 2572/1541, loss: 0.0003426424227654934 2023-01-23 18:16:46.867012: step: 2574/1541, loss: 0.0016647088341414928 2023-01-23 18:16:47.777544: step: 2576/1541, loss: 0.0011707018129527569 2023-01-23 18:16:48.665544: step: 2578/1541, loss: 0.010342567227780819 2023-01-23 18:16:49.548814: step: 2580/1541, loss: 0.011874302290380001 2023-01-23 18:16:50.410309: step: 2582/1541, loss: 0.10403170436620712 2023-01-23 18:16:51.300328: step: 2584/1541, loss: 0.0005629365332424641 2023-01-23 18:16:52.237584: step: 2586/1541, loss: 0.2375704050064087 2023-01-23 18:16:53.137820: step: 2588/1541, loss: 0.0010072404984384775 2023-01-23 18:16:54.043402: step: 2590/1541, loss: 0.004357419908046722 2023-01-23 18:16:54.971599: step: 2592/1541, loss: 0.004888735245913267 2023-01-23 18:16:55.878086: step: 2594/1541, loss: 0.00069518166128546 2023-01-23 18:16:56.782708: step: 2596/1541, loss: 0.03825124353170395 2023-01-23 18:16:57.693391: step: 2598/1541, loss: 0.026955684646964073 2023-01-23 18:16:58.619876: step: 2600/1541, loss: 0.08725572377443314 2023-01-23 18:16:59.475284: step: 2602/1541, loss: 0.0037984568625688553 2023-01-23 18:17:00.338599: step: 2604/1541, loss: 0.23606020212173462 2023-01-23 18:17:01.219620: step: 2606/1541, loss: 0.03829061612486839 2023-01-23 18:17:02.095332: step: 2608/1541, loss: 0.009048440493643284 2023-01-23 18:17:02.971416: step: 2610/1541, loss: 0.5466181039810181 2023-01-23 18:17:03.879031: step: 2612/1541, loss: 0.020030617713928223 2023-01-23 18:17:04.747832: step: 2614/1541, loss: 0.012203460559248924 2023-01-23 18:17:05.679164: step: 2616/1541, loss: 0.002729101339355111 2023-01-23 18:17:06.543706: step: 2618/1541, loss: 0.005758671555668116 2023-01-23 18:17:07.436611: step: 2620/1541, loss: 0.00021200146875344217 2023-01-23 18:17:08.310875: step: 2622/1541, loss: 0.016413666307926178 2023-01-23 18:17:09.175770: step: 2624/1541, loss: 0.004156805574893951 2023-01-23 18:17:10.063081: step: 2626/1541, loss: 0.11917614191770554 2023-01-23 18:17:10.933047: step: 2628/1541, loss: 0.0015289302682504058 2023-01-23 18:17:11.903428: step: 2630/1541, loss: 0.02570534311234951 2023-01-23 18:17:12.786010: step: 2632/1541, loss: 0.651662290096283 2023-01-23 18:17:13.742027: step: 2634/1541, loss: 0.00030013994546607137 2023-01-23 18:17:14.658660: step: 2636/1541, loss: 0.06689676642417908 2023-01-23 18:17:15.532773: step: 2638/1541, loss: 0.005613873712718487 2023-01-23 18:17:16.410688: step: 2640/1541, loss: 0.16160108149051666 2023-01-23 18:17:17.297180: step: 2642/1541, loss: 0.41953033208847046 2023-01-23 18:17:18.192213: step: 2644/1541, loss: 0.18865202367305756 2023-01-23 18:17:19.107417: step: 2646/1541, loss: 0.0010870159603655338 2023-01-23 18:17:20.017618: step: 2648/1541, loss: 0.0013511752476915717 2023-01-23 18:17:20.933386: step: 2650/1541, loss: 0.2184346467256546 2023-01-23 18:17:21.788207: step: 2652/1541, loss: 0.003311872947961092 2023-01-23 18:17:22.659716: step: 2654/1541, loss: 0.002043725224211812 2023-01-23 18:17:23.547669: step: 2656/1541, loss: 0.04176713526248932 2023-01-23 18:17:24.447053: step: 2658/1541, loss: 0.0018707241397351027 2023-01-23 18:17:25.295977: step: 2660/1541, loss: 6.500598829006776e-05 2023-01-23 18:17:26.246765: step: 2662/1541, loss: 0.10820292681455612 2023-01-23 18:17:27.138491: step: 2664/1541, loss: 0.000648569199256599 2023-01-23 18:17:28.022514: step: 2666/1541, loss: 0.0023672026582062244 2023-01-23 18:17:28.911958: step: 2668/1541, loss: 0.0007031039567664266 2023-01-23 18:17:29.861871: step: 2670/1541, loss: 0.0028513376601040363 2023-01-23 18:17:30.786292: step: 2672/1541, loss: 4.184249337413348e-05 2023-01-23 18:17:31.648387: step: 2674/1541, loss: 0.1003432646393776 2023-01-23 18:17:32.607278: step: 2676/1541, loss: 0.00021235839813016355 2023-01-23 18:17:33.473109: step: 2678/1541, loss: 0.0004080817161593586 2023-01-23 18:17:34.370792: step: 2680/1541, loss: 0.003554908325895667 2023-01-23 18:17:35.330723: step: 2682/1541, loss: 4.436239032656886e-05 2023-01-23 18:17:36.289060: step: 2684/1541, loss: 0.09766042232513428 2023-01-23 18:17:37.200444: step: 2686/1541, loss: 0.07969141751527786 2023-01-23 18:17:38.073394: step: 2688/1541, loss: 0.004001772031188011 2023-01-23 18:17:38.999612: step: 2690/1541, loss: 0.0004303109599277377 2023-01-23 18:17:39.897117: step: 2692/1541, loss: 0.0003554046561475843 2023-01-23 18:17:40.921625: step: 2694/1541, loss: 0.13719600439071655 2023-01-23 18:17:41.824066: step: 2696/1541, loss: 0.003129606833681464 2023-01-23 18:17:42.713001: step: 2698/1541, loss: 0.0008894460625015199 2023-01-23 18:17:43.615860: step: 2700/1541, loss: 0.005137926898896694 2023-01-23 18:17:44.516673: step: 2702/1541, loss: 0.00792921707034111 2023-01-23 18:17:45.391381: step: 2704/1541, loss: 0.0024583658669143915 2023-01-23 18:17:46.277470: step: 2706/1541, loss: 0.06579519063234329 2023-01-23 18:17:47.189276: step: 2708/1541, loss: 0.09499124437570572 2023-01-23 18:17:48.100063: step: 2710/1541, loss: 0.00012121610052417964 2023-01-23 18:17:49.000394: step: 2712/1541, loss: 0.0004980720113962889 2023-01-23 18:17:49.895353: step: 2714/1541, loss: 0.040345028042793274 2023-01-23 18:17:50.805258: step: 2716/1541, loss: 0.01075106393545866 2023-01-23 18:17:51.703467: step: 2718/1541, loss: 0.05044209584593773 2023-01-23 18:17:52.705332: step: 2720/1541, loss: 0.008586497977375984 2023-01-23 18:17:53.563912: step: 2722/1541, loss: 0.015155664645135403 2023-01-23 18:17:54.444737: step: 2724/1541, loss: 0.008933315984904766 2023-01-23 18:17:55.355045: step: 2726/1541, loss: 0.003255494637414813 2023-01-23 18:17:56.216209: step: 2728/1541, loss: 0.6979281902313232 2023-01-23 18:17:57.080947: step: 2730/1541, loss: 0.004076776560395956 2023-01-23 18:17:57.961384: step: 2732/1541, loss: 0.02900507301092148 2023-01-23 18:17:58.876401: step: 2734/1541, loss: 0.0009384449222125113 2023-01-23 18:17:59.785963: step: 2736/1541, loss: 0.013885396532714367 2023-01-23 18:18:00.670713: step: 2738/1541, loss: 0.000553069869056344 2023-01-23 18:18:01.569941: step: 2740/1541, loss: 0.0015345056308433414 2023-01-23 18:18:02.430971: step: 2742/1541, loss: 0.0032522231340408325 2023-01-23 18:18:03.286691: step: 2744/1541, loss: 0.015650035813450813 2023-01-23 18:18:04.226903: step: 2746/1541, loss: 0.001803691266104579 2023-01-23 18:18:05.115970: step: 2748/1541, loss: 0.018320942297577858 2023-01-23 18:18:06.002640: step: 2750/1541, loss: 0.06047479063272476 2023-01-23 18:18:06.890296: step: 2752/1541, loss: 0.006079499144107103 2023-01-23 18:18:07.721348: step: 2754/1541, loss: 0.0026605483144521713 2023-01-23 18:18:08.733461: step: 2756/1541, loss: 0.019125651568174362 2023-01-23 18:18:09.615632: step: 2758/1541, loss: 0.04823699966073036 2023-01-23 18:18:10.509071: step: 2760/1541, loss: 0.07200809568166733 2023-01-23 18:18:11.434948: step: 2762/1541, loss: 1.1127272844314575 2023-01-23 18:18:12.314640: step: 2764/1541, loss: 0.0006568913231603801 2023-01-23 18:18:13.221803: step: 2766/1541, loss: 0.015815965831279755 2023-01-23 18:18:14.122134: step: 2768/1541, loss: 0.00019532554142642766 2023-01-23 18:18:14.991753: step: 2770/1541, loss: 0.12565487623214722 2023-01-23 18:18:15.901000: step: 2772/1541, loss: 0.0006168716354295611 2023-01-23 18:18:16.836782: step: 2774/1541, loss: 0.00048298161709681153 2023-01-23 18:18:17.770799: step: 2776/1541, loss: 0.06651955842971802 2023-01-23 18:18:18.662447: step: 2778/1541, loss: 0.09024415165185928 2023-01-23 18:18:19.515797: step: 2780/1541, loss: 0.0017597574042156339 2023-01-23 18:18:20.400712: step: 2782/1541, loss: 0.000209248362807557 2023-01-23 18:18:21.279403: step: 2784/1541, loss: 0.0006859056884422898 2023-01-23 18:18:22.129519: step: 2786/1541, loss: 0.0005219293525442481 2023-01-23 18:18:23.054427: step: 2788/1541, loss: 0.0028042527846992016 2023-01-23 18:18:23.933191: step: 2790/1541, loss: 0.24144579470157623 2023-01-23 18:18:24.841106: step: 2792/1541, loss: 3.2242965971818194e-05 2023-01-23 18:18:25.731866: step: 2794/1541, loss: 0.044703010469675064 2023-01-23 18:18:26.722277: step: 2796/1541, loss: 0.6074207425117493 2023-01-23 18:18:27.636541: step: 2798/1541, loss: 0.02990422397851944 2023-01-23 18:18:28.513858: step: 2800/1541, loss: 0.0003078911977354437 2023-01-23 18:18:29.445959: step: 2802/1541, loss: 0.02192990481853485 2023-01-23 18:18:30.384763: step: 2804/1541, loss: 0.00012948564835824072 2023-01-23 18:18:31.279112: step: 2806/1541, loss: 0.0008821450173854828 2023-01-23 18:18:32.205874: step: 2808/1541, loss: 0.012872571125626564 2023-01-23 18:18:33.121274: step: 2810/1541, loss: 0.0006790048792026937 2023-01-23 18:18:34.034656: step: 2812/1541, loss: 0.0006753510679118335 2023-01-23 18:18:34.940480: step: 2814/1541, loss: 0.0007061639917083085 2023-01-23 18:18:35.824067: step: 2816/1541, loss: 0.1641165167093277 2023-01-23 18:18:36.743182: step: 2818/1541, loss: 0.3649086654186249 2023-01-23 18:18:37.689366: step: 2820/1541, loss: 8.329324919031933e-05 2023-01-23 18:18:38.591043: step: 2822/1541, loss: 0.0007397215231321752 2023-01-23 18:18:39.485715: step: 2824/1541, loss: 0.08475132286548615 2023-01-23 18:18:40.355415: step: 2826/1541, loss: 0.09249228984117508 2023-01-23 18:18:41.277061: step: 2828/1541, loss: 0.038713835179805756 2023-01-23 18:18:42.117010: step: 2830/1541, loss: 0.0037692436017096043 2023-01-23 18:18:43.002032: step: 2832/1541, loss: 0.6127044558525085 2023-01-23 18:18:43.912619: step: 2834/1541, loss: 0.006161608267575502 2023-01-23 18:18:44.801032: step: 2836/1541, loss: 0.0012521737953647971 2023-01-23 18:18:45.724891: step: 2838/1541, loss: 0.001046743942424655 2023-01-23 18:18:46.586912: step: 2840/1541, loss: 0.06560095399618149 2023-01-23 18:18:47.509126: step: 2842/1541, loss: 0.02902309037744999 2023-01-23 18:18:48.396982: step: 2844/1541, loss: 0.06819217652082443 2023-01-23 18:18:49.294898: step: 2846/1541, loss: 0.010368051007390022 2023-01-23 18:18:50.184510: step: 2848/1541, loss: 0.0792199969291687 2023-01-23 18:18:51.085793: step: 2850/1541, loss: 0.01689170114696026 2023-01-23 18:18:51.968106: step: 2852/1541, loss: 0.00012810523912776262 2023-01-23 18:18:52.860151: step: 2854/1541, loss: 0.5171281099319458 2023-01-23 18:18:53.765404: step: 2856/1541, loss: 0.023312075063586235 2023-01-23 18:18:54.668681: step: 2858/1541, loss: 0.10564836859703064 2023-01-23 18:18:55.556318: step: 2860/1541, loss: 0.0052646454423666 2023-01-23 18:18:56.421347: step: 2862/1541, loss: 0.06874158978462219 2023-01-23 18:18:57.344598: step: 2864/1541, loss: 0.08903548866510391 2023-01-23 18:18:58.300442: step: 2866/1541, loss: -3.62038690582267e-06 2023-01-23 18:18:59.198135: step: 2868/1541, loss: 0.00816419254988432 2023-01-23 18:19:00.074375: step: 2870/1541, loss: 0.20190243422985077 2023-01-23 18:19:00.941632: step: 2872/1541, loss: 0.38653433322906494 2023-01-23 18:19:01.820648: step: 2874/1541, loss: 0.06078965216875076 2023-01-23 18:19:02.689499: step: 2876/1541, loss: 0.2800281345844269 2023-01-23 18:19:03.571130: step: 2878/1541, loss: 0.0001628049067221582 2023-01-23 18:19:04.506677: step: 2880/1541, loss: 0.037828583270311356 2023-01-23 18:19:05.387856: step: 2882/1541, loss: 0.2596006393432617 2023-01-23 18:19:06.225572: step: 2884/1541, loss: 0.0018525623017922044 2023-01-23 18:19:07.120144: step: 2886/1541, loss: 0.017554830759763718 2023-01-23 18:19:08.038129: step: 2888/1541, loss: 0.022773226723074913 2023-01-23 18:19:08.984550: step: 2890/1541, loss: 0.007124120835214853 2023-01-23 18:19:09.930344: step: 2892/1541, loss: 0.21724483370780945 2023-01-23 18:19:10.809338: step: 2894/1541, loss: 0.01022691186517477 2023-01-23 18:19:11.713526: step: 2896/1541, loss: 3.992723213741556e-05 2023-01-23 18:19:12.600216: step: 2898/1541, loss: 0.02160360850393772 2023-01-23 18:19:13.530925: step: 2900/1541, loss: 0.0089067742228508 2023-01-23 18:19:14.453610: step: 2902/1541, loss: 0.053172241896390915 2023-01-23 18:19:15.353400: step: 2904/1541, loss: 9.013586532091722e-05 2023-01-23 18:19:16.267016: step: 2906/1541, loss: 0.002667824039235711 2023-01-23 18:19:17.148344: step: 2908/1541, loss: 0.004208644386380911 2023-01-23 18:19:17.986273: step: 2910/1541, loss: 0.0005923857679590583 2023-01-23 18:19:18.894307: step: 2912/1541, loss: 1.876552778412588e-05 2023-01-23 18:19:19.771360: step: 2914/1541, loss: 0.0609026700258255 2023-01-23 18:19:20.620571: step: 2916/1541, loss: 0.0339287593960762 2023-01-23 18:19:21.532716: step: 2918/1541, loss: 0.20890486240386963 2023-01-23 18:19:22.443930: step: 2920/1541, loss: 0.0003459189319983125 2023-01-23 18:19:23.362774: step: 2922/1541, loss: 0.0009375306544825435 2023-01-23 18:19:24.210175: step: 2924/1541, loss: 0.003113220678642392 2023-01-23 18:19:25.092422: step: 2926/1541, loss: 0.008981599472463131 2023-01-23 18:19:26.012743: step: 2928/1541, loss: 0.0016999704530462623 2023-01-23 18:19:26.892267: step: 2930/1541, loss: 0.0003974839055445045 2023-01-23 18:19:27.856529: step: 2932/1541, loss: 0.038603492081165314 2023-01-23 18:19:28.741114: step: 2934/1541, loss: 0.035079509019851685 2023-01-23 18:19:29.624563: step: 2936/1541, loss: 0.09247660636901855 2023-01-23 18:19:30.525004: step: 2938/1541, loss: 0.09056585282087326 2023-01-23 18:19:31.520282: step: 2940/1541, loss: 0.0013671178603544831 2023-01-23 18:19:32.432450: step: 2942/1541, loss: 0.0015177848981693387 2023-01-23 18:19:33.352201: step: 2944/1541, loss: 0.04465962201356888 2023-01-23 18:19:34.255988: step: 2946/1541, loss: 0.08684637397527695 2023-01-23 18:19:35.128430: step: 2948/1541, loss: 0.000324016873491928 2023-01-23 18:19:35.962175: step: 2950/1541, loss: 0.00016690079064574093 2023-01-23 18:19:36.834431: step: 2952/1541, loss: 0.0001362270413665101 2023-01-23 18:19:37.769633: step: 2954/1541, loss: 0.5511447787284851 2023-01-23 18:19:38.698611: step: 2956/1541, loss: 0.005178993567824364 2023-01-23 18:19:39.573651: step: 2958/1541, loss: 0.0002984481689054519 2023-01-23 18:19:40.489164: step: 2960/1541, loss: 0.22277727723121643 2023-01-23 18:19:41.373563: step: 2962/1541, loss: 0.01591629348695278 2023-01-23 18:19:42.291334: step: 2964/1541, loss: 0.013706314377486706 2023-01-23 18:19:43.216180: step: 2966/1541, loss: 0.006458706222474575 2023-01-23 18:19:44.096760: step: 2968/1541, loss: 0.09746159613132477 2023-01-23 18:19:44.986606: step: 2970/1541, loss: 0.02888890728354454 2023-01-23 18:19:45.879558: step: 2972/1541, loss: 0.0022215889766812325 2023-01-23 18:19:46.771744: step: 2974/1541, loss: 0.00018365858704783022 2023-01-23 18:19:47.659144: step: 2976/1541, loss: 0.00014144554734230042 2023-01-23 18:19:48.566705: step: 2978/1541, loss: 0.000656538235489279 2023-01-23 18:19:49.441947: step: 2980/1541, loss: 0.5322413444519043 2023-01-23 18:19:50.329248: step: 2982/1541, loss: 0.0004460570926312357 2023-01-23 18:19:51.208686: step: 2984/1541, loss: 0.28268465399742126 2023-01-23 18:19:52.106043: step: 2986/1541, loss: 0.043647680431604385 2023-01-23 18:19:52.989642: step: 2988/1541, loss: 0.003301961813122034 2023-01-23 18:19:53.950824: step: 2990/1541, loss: 0.014025850221514702 2023-01-23 18:19:54.817633: step: 2992/1541, loss: 0.011642820201814175 2023-01-23 18:19:55.756665: step: 2994/1541, loss: 0.02980806864798069 2023-01-23 18:19:56.686563: step: 2996/1541, loss: 0.01619611121714115 2023-01-23 18:19:57.579628: step: 2998/1541, loss: 0.0007826011860743165 2023-01-23 18:19:58.503763: step: 3000/1541, loss: 0.009462973102927208 2023-01-23 18:19:59.400030: step: 3002/1541, loss: 0.015774531289935112 2023-01-23 18:20:00.283963: step: 3004/1541, loss: 0.0002930402697529644 2023-01-23 18:20:01.204383: step: 3006/1541, loss: 0.11678000539541245 2023-01-23 18:20:02.154870: step: 3008/1541, loss: 0.10388985276222229 2023-01-23 18:20:03.001864: step: 3010/1541, loss: 0.20538848638534546 2023-01-23 18:20:03.903820: step: 3012/1541, loss: 0.0036645045038312674 2023-01-23 18:20:04.787745: step: 3014/1541, loss: 0.0009136637672781944 2023-01-23 18:20:05.683422: step: 3016/1541, loss: 0.029775647446513176 2023-01-23 18:20:06.579241: step: 3018/1541, loss: 0.04022267833352089 2023-01-23 18:20:07.471732: step: 3020/1541, loss: 0.009112052619457245 2023-01-23 18:20:08.322346: step: 3022/1541, loss: 0.006355213932693005 2023-01-23 18:20:09.230399: step: 3024/1541, loss: 0.11667486280202866 2023-01-23 18:20:10.152364: step: 3026/1541, loss: 0.0008615987608209252 2023-01-23 18:20:11.037326: step: 3028/1541, loss: 0.035062026232481 2023-01-23 18:20:11.936676: step: 3030/1541, loss: 0.0658852607011795 2023-01-23 18:20:12.867856: step: 3032/1541, loss: 0.0018640620401129127 2023-01-23 18:20:13.828469: step: 3034/1541, loss: 0.040410805493593216 2023-01-23 18:20:14.685531: step: 3036/1541, loss: 0.000391306821256876 2023-01-23 18:20:15.643487: step: 3038/1541, loss: 0.0010377101134508848 2023-01-23 18:20:16.556852: step: 3040/1541, loss: 0.09030163288116455 2023-01-23 18:20:17.406854: step: 3042/1541, loss: 0.20517560839653015 2023-01-23 18:20:18.275140: step: 3044/1541, loss: 0.023219367489218712 2023-01-23 18:20:19.197343: step: 3046/1541, loss: 0.011365324258804321 2023-01-23 18:20:20.106583: step: 3048/1541, loss: 0.005002985708415508 2023-01-23 18:20:20.999436: step: 3050/1541, loss: 0.0062294406816363335 2023-01-23 18:20:21.904371: step: 3052/1541, loss: 0.026503896340727806 2023-01-23 18:20:22.941324: step: 3054/1541, loss: 0.004310510586947203 2023-01-23 18:20:23.892251: step: 3056/1541, loss: 0.38135045766830444 2023-01-23 18:20:24.793435: step: 3058/1541, loss: 0.007570417132228613 2023-01-23 18:20:25.740288: step: 3060/1541, loss: 0.05812457948923111 2023-01-23 18:20:26.596458: step: 3062/1541, loss: 0.0025805002078413963 2023-01-23 18:20:27.497275: step: 3064/1541, loss: 0.005278031807392836 2023-01-23 18:20:28.385457: step: 3066/1541, loss: 0.013660772703588009 2023-01-23 18:20:29.264046: step: 3068/1541, loss: 0.05164366960525513 2023-01-23 18:20:30.169156: step: 3070/1541, loss: 0.12941685318946838 2023-01-23 18:20:31.075187: step: 3072/1541, loss: 0.16394682228565216 2023-01-23 18:20:31.973353: step: 3074/1541, loss: 0.009620307013392448 2023-01-23 18:20:32.829663: step: 3076/1541, loss: 0.0017609071219339967 2023-01-23 18:20:33.726691: step: 3078/1541, loss: 0.0026438869535923004 2023-01-23 18:20:34.657585: step: 3080/1541, loss: 0.003511868417263031 2023-01-23 18:20:35.523863: step: 3082/1541, loss: 0.01386902667582035 2023-01-23 18:20:36.190489: step: 3084/1541, loss: 0.2414318472146988 ================================================== Loss: 0.063 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6800046033188152, 'r': 0.644549110383126, 'f1': 0.6618023226866295}, 'combined': 0.43758969699052963, 'epoch': 27} Test Chinese: {'event': {'p': 0.6098843322818086, 'r': 0.695026962252846, 'f1': 0.6496779613553626}, 'argument': {'p': 0.7224781577231381, 'r': 0.6800104748485761, 'f1': 0.7006013506151025}, 'combined': 0.45516525719043344, 'epoch': 27} Dev Korean: {'event': {'p': 0.5705882352941176, 'r': 0.7749667110519307, 'f1': 0.6572557876905702}, 'argument': {'p': 0.673183849651783, 'r': 0.6519355789310668, 'f1': 0.6623893564357923}, 'combined': 0.4353592382220566, 'epoch': 27} Test Korean: {'event': {'p': 0.607721639656816, 'r': 0.7639304973037747, 'f1': 0.6769312450225644}, 'argument': {'p': 0.7239969439454942, 'r': 0.6947625167847737, 'f1': 0.7090785335255458}, 'combined': 0.4799974145182219, 'epoch': 27} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6790052558154206, 'r': 0.6514158774752906, 'f1': 0.6649245016625662}, 'combined': 0.4610699052280867, 'epoch': 27} Test Russian: {'event': {'p': 0.6488749345892203, 'r': 0.7429598562013181, 'f1': 0.6927374301675977}, 'argument': {'p': 0.72095499117386, 'r': 0.6841292326318859, 'f1': 0.7020595299803036}, 'combined': 0.48634291462322704, 'epoch': 27} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6734006734006733, 'r': 0.6436041834271922, 'f1': 0.658165364047717}, 'combined': 0.48377966930003125, 'epoch': 27} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5715783168613356, 'r': 0.3934240362811791, 'f1': 0.466056166056166}, 'combined': 0.27156028573351404, 'epoch': 27} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.554375696767001, 'r': 0.5930530709600477, 'f1': 0.5730625180063381}, 'combined': 0.27834465160307853, 'epoch': 27} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.6748395028808764, 'r': 0.6385747632400816, 'f1': 0.6562064793919685}, 'combined': 0.4371955007535911, 'epoch': 25} Test for Chinese: {'event': {'p': 0.6109358569926393, 'r': 0.6962252846015579, 'f1': 0.6507980957714926}, 'argument': {'p': 0.7154126636351134, 'r': 0.6940648330608487, 'f1': 0.7045770821024057}, 'combined': 0.45853742335648023, 'epoch': 25} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6613345847859121, 'r': 0.7050076234038498, 'f1': 0.6824731331580647}, 'combined': 0.5133131257940999, 'epoch': 25} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 28 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 18:25:57.351893: step: 2/1541, loss: 0.000560737622436136 2023-01-23 18:25:58.280842: step: 4/1541, loss: 0.12026713043451309 2023-01-23 18:25:59.186312: step: 6/1541, loss: 0.00843876414000988 2023-01-23 18:26:00.058576: step: 8/1541, loss: 0.007188974414020777 2023-01-23 18:26:00.948550: step: 10/1541, loss: 0.019944047555327415 2023-01-23 18:26:01.936775: step: 12/1541, loss: 0.00240871449932456 2023-01-23 18:26:02.805718: step: 14/1541, loss: 0.009339043870568275 2023-01-23 18:26:03.677492: step: 16/1541, loss: 0.00024503940949216485 2023-01-23 18:26:04.549318: step: 18/1541, loss: 0.00011440635717008263 2023-01-23 18:26:05.414107: step: 20/1541, loss: 0.005838128738105297 2023-01-23 18:26:06.305785: step: 22/1541, loss: 0.07099093496799469 2023-01-23 18:26:07.267908: step: 24/1541, loss: 0.0001945921394508332 2023-01-23 18:26:08.140988: step: 26/1541, loss: 0.002515004249289632 2023-01-23 18:26:09.028667: step: 28/1541, loss: 0.00013078682241030037 2023-01-23 18:26:09.889787: step: 30/1541, loss: 0.00021415673836600035 2023-01-23 18:26:10.785259: step: 32/1541, loss: 0.2801644802093506 2023-01-23 18:26:11.619893: step: 34/1541, loss: 0.015387952327728271 2023-01-23 18:26:12.535219: step: 36/1541, loss: 0.8690601587295532 2023-01-23 18:26:13.418030: step: 38/1541, loss: 0.0010572135215625167 2023-01-23 18:26:14.302022: step: 40/1541, loss: 0.43333834409713745 2023-01-23 18:26:15.182930: step: 42/1541, loss: 0.003728017443791032 2023-01-23 18:26:16.065159: step: 44/1541, loss: 0.0008507370948791504 2023-01-23 18:26:16.954822: step: 46/1541, loss: -1.895017885544803e-05 2023-01-23 18:26:17.876414: step: 48/1541, loss: 0.008301468566060066 2023-01-23 18:26:18.865068: step: 50/1541, loss: 0.004112527705729008 2023-01-23 18:26:19.771668: step: 52/1541, loss: 0.14759016036987305 2023-01-23 18:26:20.710162: step: 54/1541, loss: 0.00027552072424441576 2023-01-23 18:26:21.573769: step: 56/1541, loss: 0.00028374025714583695 2023-01-23 18:26:22.490309: step: 58/1541, loss: 0.0010099022183567286 2023-01-23 18:26:23.332846: step: 60/1541, loss: 0.0013944595120847225 2023-01-23 18:26:24.260428: step: 62/1541, loss: 0.0008052501943893731 2023-01-23 18:26:25.127091: step: 64/1541, loss: 0.036226969212293625 2023-01-23 18:26:26.008407: step: 66/1541, loss: 2.8538906917674467e-05 2023-01-23 18:26:26.855989: step: 68/1541, loss: 7.337338320212439e-05 2023-01-23 18:26:27.740093: step: 70/1541, loss: 3.380365524208173e-05 2023-01-23 18:26:28.611183: step: 72/1541, loss: 0.005177992861717939 2023-01-23 18:26:29.490173: step: 74/1541, loss: 0.010264303535223007 2023-01-23 18:26:30.426010: step: 76/1541, loss: 0.003795550437644124 2023-01-23 18:26:31.325089: step: 78/1541, loss: 0.0004122188547626138 2023-01-23 18:26:32.244532: step: 80/1541, loss: 0.18943801522254944 2023-01-23 18:26:33.103494: step: 82/1541, loss: 0.0007287967600859702 2023-01-23 18:26:33.960979: step: 84/1541, loss: 3.8067482819315046e-05 2023-01-23 18:26:34.841822: step: 86/1541, loss: 0.08645591884851456 2023-01-23 18:26:35.728026: step: 88/1541, loss: 0.0003759139508474618 2023-01-23 18:26:36.610718: step: 90/1541, loss: 0.0006435590330511332 2023-01-23 18:26:37.599443: step: 92/1541, loss: 0.010853763669729233 2023-01-23 18:26:38.473036: step: 94/1541, loss: 0.00016922243230510503 2023-01-23 18:26:39.381867: step: 96/1541, loss: -3.5301106890983647e-06 2023-01-23 18:26:40.249492: step: 98/1541, loss: 0.0002608326030895114 2023-01-23 18:26:41.176258: step: 100/1541, loss: 0.008329191245138645 2023-01-23 18:26:42.081193: step: 102/1541, loss: 0.0008511252235621214 2023-01-23 18:26:42.947867: step: 104/1541, loss: 0.42162927985191345 2023-01-23 18:26:43.848901: step: 106/1541, loss: 0.00013742950977757573 2023-01-23 18:26:44.737409: step: 108/1541, loss: 0.04750152304768562 2023-01-23 18:26:45.616335: step: 110/1541, loss: 0.0006387382745742798 2023-01-23 18:26:46.516806: step: 112/1541, loss: 0.0008995072566904128 2023-01-23 18:26:47.436272: step: 114/1541, loss: 0.0002729846746660769 2023-01-23 18:26:48.348311: step: 116/1541, loss: 0.00026157498359680176 2023-01-23 18:26:49.229969: step: 118/1541, loss: -1.3715062777919229e-05 2023-01-23 18:26:50.133005: step: 120/1541, loss: 0.31106558442115784 2023-01-23 18:26:51.003450: step: 122/1541, loss: 0.0032196270767599344 2023-01-23 18:26:51.894127: step: 124/1541, loss: 0.002914806129410863 2023-01-23 18:26:52.788961: step: 126/1541, loss: 0.08523622900247574 2023-01-23 18:26:53.672749: step: 128/1541, loss: 0.0007141410605981946 2023-01-23 18:26:54.552566: step: 130/1541, loss: 0.0008545774035155773 2023-01-23 18:26:55.429114: step: 132/1541, loss: 0.003970908001065254 2023-01-23 18:26:56.304959: step: 134/1541, loss: 0.08153198659420013 2023-01-23 18:26:57.168224: step: 136/1541, loss: 0.029085449874401093 2023-01-23 18:26:58.084422: step: 138/1541, loss: 0.0004632804775610566 2023-01-23 18:26:58.984894: step: 140/1541, loss: 0.08620870113372803 2023-01-23 18:26:59.915588: step: 142/1541, loss: 0.14108823239803314 2023-01-23 18:27:00.870097: step: 144/1541, loss: 0.01912713795900345 2023-01-23 18:27:01.775685: step: 146/1541, loss: 0.0004455114249140024 2023-01-23 18:27:02.677614: step: 148/1541, loss: 0.34555941820144653 2023-01-23 18:27:03.566361: step: 150/1541, loss: 0.027257611975073814 2023-01-23 18:27:04.436088: step: 152/1541, loss: 0.0006754425703547895 2023-01-23 18:27:05.309450: step: 154/1541, loss: 0.0054785870015621185 2023-01-23 18:27:06.181173: step: 156/1541, loss: 0.00244394619949162 2023-01-23 18:27:07.046660: step: 158/1541, loss: 5.929503095103428e-05 2023-01-23 18:27:07.935350: step: 160/1541, loss: 0.0704255923628807 2023-01-23 18:27:08.840796: step: 162/1541, loss: 4.2793773900484666e-05 2023-01-23 18:27:09.720506: step: 164/1541, loss: 0.00016969068383332342 2023-01-23 18:27:10.622154: step: 166/1541, loss: 0.001404040725901723 2023-01-23 18:27:11.565559: step: 168/1541, loss: 0.04784909635782242 2023-01-23 18:27:12.427252: step: 170/1541, loss: 0.005081280134618282 2023-01-23 18:27:13.341910: step: 172/1541, loss: 0.002890327712520957 2023-01-23 18:27:14.236204: step: 174/1541, loss: 0.001581933698616922 2023-01-23 18:27:15.127043: step: 176/1541, loss: 0.027739044278860092 2023-01-23 18:27:16.051553: step: 178/1541, loss: 1.8469429051037878e-05 2023-01-23 18:27:16.966950: step: 180/1541, loss: 0.0002844980626832694 2023-01-23 18:27:17.893452: step: 182/1541, loss: 7.364949124166742e-05 2023-01-23 18:27:18.798245: step: 184/1541, loss: 0.013544304296374321 2023-01-23 18:27:19.671541: step: 186/1541, loss: 0.0009375689551234245 2023-01-23 18:27:20.518863: step: 188/1541, loss: 2.51303608820308e-05 2023-01-23 18:27:21.359872: step: 190/1541, loss: 0.07140854746103287 2023-01-23 18:27:22.208305: step: 192/1541, loss: 0.001051182160153985 2023-01-23 18:27:23.158813: step: 194/1541, loss: 0.4323181211948395 2023-01-23 18:27:24.074147: step: 196/1541, loss: 0.14913544058799744 2023-01-23 18:27:24.991654: step: 198/1541, loss: 0.18638822436332703 2023-01-23 18:27:25.879729: step: 200/1541, loss: 0.008071291260421276 2023-01-23 18:27:26.744025: step: 202/1541, loss: 0.0013807018985971808 2023-01-23 18:27:27.574082: step: 204/1541, loss: 0.0003468003123998642 2023-01-23 18:27:28.516830: step: 206/1541, loss: 0.13553528487682343 2023-01-23 18:27:29.431136: step: 208/1541, loss: 0.013335278257727623 2023-01-23 18:27:30.296371: step: 210/1541, loss: 0.0006209178827702999 2023-01-23 18:27:31.161788: step: 212/1541, loss: 1.750178125803359e-05 2023-01-23 18:27:32.155360: step: 214/1541, loss: 0.0010577710345387459 2023-01-23 18:27:33.027862: step: 216/1541, loss: 0.000248217984335497 2023-01-23 18:27:33.925466: step: 218/1541, loss: 0.03971337899565697 2023-01-23 18:27:34.860942: step: 220/1541, loss: 0.0001497863995609805 2023-01-23 18:27:35.709790: step: 222/1541, loss: 0.010738974437117577 2023-01-23 18:27:36.637602: step: 224/1541, loss: 0.00022203494154382497 2023-01-23 18:27:37.587845: step: 226/1541, loss: 0.10058844834566116 2023-01-23 18:27:38.443443: step: 228/1541, loss: 0.0008865779964253306 2023-01-23 18:27:39.323408: step: 230/1541, loss: 0.0002792969171423465 2023-01-23 18:27:40.195750: step: 232/1541, loss: 0.00010320236469851807 2023-01-23 18:27:41.103389: step: 234/1541, loss: 0.07715078443288803 2023-01-23 18:27:42.028396: step: 236/1541, loss: 0.36780956387519836 2023-01-23 18:27:42.914778: step: 238/1541, loss: 0.00015945994528010488 2023-01-23 18:27:43.798574: step: 240/1541, loss: 0.007109515368938446 2023-01-23 18:27:44.712400: step: 242/1541, loss: 0.1723427176475525 2023-01-23 18:27:45.623576: step: 244/1541, loss: 0.007135307416319847 2023-01-23 18:27:46.571435: step: 246/1541, loss: 0.00655859149992466 2023-01-23 18:27:47.486190: step: 248/1541, loss: 0.05659959092736244 2023-01-23 18:27:48.366896: step: 250/1541, loss: 0.0006253933534026146 2023-01-23 18:27:49.222380: step: 252/1541, loss: 0.028729671612381935 2023-01-23 18:27:50.133874: step: 254/1541, loss: 0.01668953336775303 2023-01-23 18:27:51.065378: step: 256/1541, loss: 0.2767428159713745 2023-01-23 18:27:51.978979: step: 258/1541, loss: 0.00037185545079410076 2023-01-23 18:27:52.866973: step: 260/1541, loss: 0.016557546332478523 2023-01-23 18:27:53.743252: step: 262/1541, loss: 0.0003664087562356144 2023-01-23 18:27:54.626690: step: 264/1541, loss: 0.10498341172933578 2023-01-23 18:27:55.563259: step: 266/1541, loss: 0.0072004650719463825 2023-01-23 18:27:56.441583: step: 268/1541, loss: 0.007842177525162697 2023-01-23 18:27:57.343713: step: 270/1541, loss: 0.0031393924728035927 2023-01-23 18:27:58.212929: step: 272/1541, loss: 0.001544842729344964 2023-01-23 18:27:59.083222: step: 274/1541, loss: 0.000215087435208261 2023-01-23 18:27:59.977608: step: 276/1541, loss: 0.013144426047801971 2023-01-23 18:28:00.882569: step: 278/1541, loss: 0.005983320996165276 2023-01-23 18:28:01.748692: step: 280/1541, loss: 0.003419845597818494 2023-01-23 18:28:02.608663: step: 282/1541, loss: 0.01247961726039648 2023-01-23 18:28:03.518303: step: 284/1541, loss: 0.0012594264699146152 2023-01-23 18:28:04.367446: step: 286/1541, loss: 0.00047695773537270725 2023-01-23 18:28:05.237020: step: 288/1541, loss: 0.0005761050269939005 2023-01-23 18:28:06.118899: step: 290/1541, loss: 8.506779704475775e-05 2023-01-23 18:28:07.079900: step: 292/1541, loss: 0.011744927614927292 2023-01-23 18:28:07.961283: step: 294/1541, loss: 4.753221946884878e-05 2023-01-23 18:28:08.865861: step: 296/1541, loss: 0.0007155478815548122 2023-01-23 18:28:09.782755: step: 298/1541, loss: 0.0022435763385146856 2023-01-23 18:28:10.682910: step: 300/1541, loss: 0.022134900093078613 2023-01-23 18:28:11.558581: step: 302/1541, loss: 0.006889904849231243 2023-01-23 18:28:12.404362: step: 304/1541, loss: 0.001254080911166966 2023-01-23 18:28:13.341324: step: 306/1541, loss: 0.7724735736846924 2023-01-23 18:28:14.245511: step: 308/1541, loss: 0.00045390913146547973 2023-01-23 18:28:15.127099: step: 310/1541, loss: 0.01770387776196003 2023-01-23 18:28:15.998540: step: 312/1541, loss: 0.00043198009370826185 2023-01-23 18:28:16.893072: step: 314/1541, loss: 0.00017305590154137462 2023-01-23 18:28:17.790611: step: 316/1541, loss: 0.00011275819997536018 2023-01-23 18:28:18.719257: step: 318/1541, loss: 0.0008148857741616666 2023-01-23 18:28:19.669203: step: 320/1541, loss: 0.006291989237070084 2023-01-23 18:28:20.549367: step: 322/1541, loss: 0.025214670225977898 2023-01-23 18:28:21.469404: step: 324/1541, loss: 0.0003817934775725007 2023-01-23 18:28:22.332759: step: 326/1541, loss: 0.010855240747332573 2023-01-23 18:28:23.254995: step: 328/1541, loss: 0.009537982754409313 2023-01-23 18:28:24.134947: step: 330/1541, loss: 0.00019312271615490317 2023-01-23 18:28:25.034280: step: 332/1541, loss: 0.00016354979015886784 2023-01-23 18:28:25.897555: step: 334/1541, loss: 0.06338396668434143 2023-01-23 18:28:26.792309: step: 336/1541, loss: 0.0005164144095033407 2023-01-23 18:28:27.697952: step: 338/1541, loss: 0.0024917712435126305 2023-01-23 18:28:28.590271: step: 340/1541, loss: 7.155984349083155e-05 2023-01-23 18:28:29.506067: step: 342/1541, loss: 0.0008855867199599743 2023-01-23 18:28:30.387942: step: 344/1541, loss: 5.261788464849815e-05 2023-01-23 18:28:31.294117: step: 346/1541, loss: 0.0004280167049728334 2023-01-23 18:28:32.222955: step: 348/1541, loss: 0.0010453133145347238 2023-01-23 18:28:33.110323: step: 350/1541, loss: 0.005044063553214073 2023-01-23 18:28:34.024200: step: 352/1541, loss: 0.26770442724227905 2023-01-23 18:28:34.911657: step: 354/1541, loss: 0.010622221045196056 2023-01-23 18:28:35.805278: step: 356/1541, loss: 0.0001234890951309353 2023-01-23 18:28:36.701970: step: 358/1541, loss: 0.08045612275600433 2023-01-23 18:28:37.582423: step: 360/1541, loss: 0.0016067244578152895 2023-01-23 18:28:38.542360: step: 362/1541, loss: 0.0017373692244291306 2023-01-23 18:28:39.466663: step: 364/1541, loss: 0.0011890958994626999 2023-01-23 18:28:40.344974: step: 366/1541, loss: 1.2459601163864136 2023-01-23 18:28:41.303211: step: 368/1541, loss: 0.03568814694881439 2023-01-23 18:28:42.250658: step: 370/1541, loss: 0.9561830163002014 2023-01-23 18:28:43.131464: step: 372/1541, loss: 0.002029197057709098 2023-01-23 18:28:44.029262: step: 374/1541, loss: 0.003396255662664771 2023-01-23 18:28:44.897496: step: 376/1541, loss: 0.0013620599638670683 2023-01-23 18:28:45.793597: step: 378/1541, loss: 0.03410262614488602 2023-01-23 18:28:46.658487: step: 380/1541, loss: 0.10239071398973465 2023-01-23 18:28:47.575069: step: 382/1541, loss: 0.0008292261627502739 2023-01-23 18:28:48.455428: step: 384/1541, loss: 0.000694511691108346 2023-01-23 18:28:49.348339: step: 386/1541, loss: 0.02795681357383728 2023-01-23 18:28:50.216947: step: 388/1541, loss: 0.00185583031270653 2023-01-23 18:28:51.169454: step: 390/1541, loss: 0.0006948674563318491 2023-01-23 18:28:52.066448: step: 392/1541, loss: 0.0012948081130161881 2023-01-23 18:28:52.986087: step: 394/1541, loss: 0.000694877642672509 2023-01-23 18:28:53.916607: step: 396/1541, loss: 0.005610211752355099 2023-01-23 18:28:54.844740: step: 398/1541, loss: 0.00015775448991917074 2023-01-23 18:28:55.770103: step: 400/1541, loss: 0.2741051912307739 2023-01-23 18:28:56.657147: step: 402/1541, loss: 0.04891873896121979 2023-01-23 18:28:57.556477: step: 404/1541, loss: 0.018841523677110672 2023-01-23 18:28:58.407831: step: 406/1541, loss: 0.0008922203560359776 2023-01-23 18:28:59.309943: step: 408/1541, loss: 9.77537056314759e-05 2023-01-23 18:29:00.168072: step: 410/1541, loss: 0.0003189962881151587 2023-01-23 18:29:01.050037: step: 412/1541, loss: 3.604738230933435e-05 2023-01-23 18:29:01.947345: step: 414/1541, loss: 0.05589703470468521 2023-01-23 18:29:02.853033: step: 416/1541, loss: 0.0009106069337576628 2023-01-23 18:29:03.807525: step: 418/1541, loss: 0.00019736928516067564 2023-01-23 18:29:04.691520: step: 420/1541, loss: 0.022692808881402016 2023-01-23 18:29:05.577470: step: 422/1541, loss: 0.6616010069847107 2023-01-23 18:29:06.502217: step: 424/1541, loss: 0.010679787956178188 2023-01-23 18:29:07.399545: step: 426/1541, loss: 4.8116569814737886e-05 2023-01-23 18:29:08.330177: step: 428/1541, loss: 0.002338338876143098 2023-01-23 18:29:09.241302: step: 430/1541, loss: 0.038980334997177124 2023-01-23 18:29:10.117617: step: 432/1541, loss: 7.078611088218167e-05 2023-01-23 18:29:10.995266: step: 434/1541, loss: 0.2322903275489807 2023-01-23 18:29:11.909912: step: 436/1541, loss: 0.04210308939218521 2023-01-23 18:29:12.796574: step: 438/1541, loss: 0.09478960931301117 2023-01-23 18:29:13.720877: step: 440/1541, loss: 0.06633918732404709 2023-01-23 18:29:14.642177: step: 442/1541, loss: 0.29811686277389526 2023-01-23 18:29:15.582705: step: 444/1541, loss: 0.015347841195762157 2023-01-23 18:29:16.504181: step: 446/1541, loss: 0.022250957787036896 2023-01-23 18:29:17.403003: step: 448/1541, loss: 0.00010619754903018475 2023-01-23 18:29:18.297834: step: 450/1541, loss: 0.08964791148900986 2023-01-23 18:29:19.187198: step: 452/1541, loss: 0.013840599916875362 2023-01-23 18:29:20.056855: step: 454/1541, loss: 0.0005200490704737604 2023-01-23 18:29:20.993513: step: 456/1541, loss: 0.0041815415024757385 2023-01-23 18:29:21.864988: step: 458/1541, loss: 0.0013066381216049194 2023-01-23 18:29:22.730660: step: 460/1541, loss: 0.0008922404376789927 2023-01-23 18:29:23.648919: step: 462/1541, loss: 0.0006921435706317425 2023-01-23 18:29:24.552521: step: 464/1541, loss: 0.0388532429933548 2023-01-23 18:29:25.430437: step: 466/1541, loss: 0.03031868487596512 2023-01-23 18:29:26.347456: step: 468/1541, loss: 0.0017271671677008271 2023-01-23 18:29:27.265822: step: 470/1541, loss: 0.07452814280986786 2023-01-23 18:29:28.155045: step: 472/1541, loss: 0.0012394406367093325 2023-01-23 18:29:29.104034: step: 474/1541, loss: 0.002122939331457019 2023-01-23 18:29:29.986247: step: 476/1541, loss: 0.000490283127874136 2023-01-23 18:29:30.877359: step: 478/1541, loss: 0.03991233557462692 2023-01-23 18:29:31.737586: step: 480/1541, loss: 0.01049106940627098 2023-01-23 18:29:32.666247: step: 482/1541, loss: 0.1578516662120819 2023-01-23 18:29:33.554069: step: 484/1541, loss: 0.00398009829223156 2023-01-23 18:29:34.443151: step: 486/1541, loss: 0.17411789298057556 2023-01-23 18:29:35.378540: step: 488/1541, loss: 0.08609737455844879 2023-01-23 18:29:36.243350: step: 490/1541, loss: 0.055610015988349915 2023-01-23 18:29:37.169282: step: 492/1541, loss: 0.0713215172290802 2023-01-23 18:29:38.035464: step: 494/1541, loss: 0.00033311365405097604 2023-01-23 18:29:38.894873: step: 496/1541, loss: 0.07790514826774597 2023-01-23 18:29:39.868784: step: 498/1541, loss: 0.07054169476032257 2023-01-23 18:29:40.754391: step: 500/1541, loss: 0.021228211000561714 2023-01-23 18:29:41.625612: step: 502/1541, loss: 0.005546597298234701 2023-01-23 18:29:42.484438: step: 504/1541, loss: 0.00042136991396546364 2023-01-23 18:29:43.451165: step: 506/1541, loss: 0.003978627268224955 2023-01-23 18:29:44.407720: step: 508/1541, loss: 0.060072481632232666 2023-01-23 18:29:45.274324: step: 510/1541, loss: 0.026158589869737625 2023-01-23 18:29:46.136799: step: 512/1541, loss: 0.008348122239112854 2023-01-23 18:29:47.100873: step: 514/1541, loss: 0.003538422752171755 2023-01-23 18:29:48.006938: step: 516/1541, loss: 0.24396641552448273 2023-01-23 18:29:48.865901: step: 518/1541, loss: 0.052409037947654724 2023-01-23 18:29:49.728989: step: 520/1541, loss: 0.013061687350273132 2023-01-23 18:29:50.637375: step: 522/1541, loss: 0.0010758511489257216 2023-01-23 18:29:51.518153: step: 524/1541, loss: 0.10539422184228897 2023-01-23 18:29:52.362326: step: 526/1541, loss: 0.0014973076758906245 2023-01-23 18:29:53.237754: step: 528/1541, loss: 0.13855570554733276 2023-01-23 18:29:54.135553: step: 530/1541, loss: 0.033247265964746475 2023-01-23 18:29:55.031502: step: 532/1541, loss: 0.00012837853864766657 2023-01-23 18:29:55.942089: step: 534/1541, loss: 0.0225359033793211 2023-01-23 18:29:56.858366: step: 536/1541, loss: 4.343412001617253e-05 2023-01-23 18:29:57.712174: step: 538/1541, loss: 0.0004438038740772754 2023-01-23 18:29:58.638062: step: 540/1541, loss: 0.1301889717578888 2023-01-23 18:29:59.499723: step: 542/1541, loss: 0.13882143795490265 2023-01-23 18:30:00.395513: step: 544/1541, loss: 0.002565496601164341 2023-01-23 18:30:01.285312: step: 546/1541, loss: 0.0026613075751811266 2023-01-23 18:30:02.216749: step: 548/1541, loss: 0.0473967045545578 2023-01-23 18:30:03.090402: step: 550/1541, loss: 0.00046663335524499416 2023-01-23 18:30:03.956551: step: 552/1541, loss: 0.00028826965717598796 2023-01-23 18:30:04.839307: step: 554/1541, loss: 0.004387336317449808 2023-01-23 18:30:05.718059: step: 556/1541, loss: 0.0006477212882600725 2023-01-23 18:30:06.620321: step: 558/1541, loss: 0.6453854441642761 2023-01-23 18:30:07.506266: step: 560/1541, loss: 0.000484868127387017 2023-01-23 18:30:08.432940: step: 562/1541, loss: 0.0013373936526477337 2023-01-23 18:30:09.401476: step: 564/1541, loss: 0.0013769102515652776 2023-01-23 18:30:10.255749: step: 566/1541, loss: 0.0022848534863442183 2023-01-23 18:30:11.168509: step: 568/1541, loss: 0.04273230582475662 2023-01-23 18:30:12.050902: step: 570/1541, loss: 0.2707805335521698 2023-01-23 18:30:12.958882: step: 572/1541, loss: 2.9937942599644884e-05 2023-01-23 18:30:13.988300: step: 574/1541, loss: 0.0713958889245987 2023-01-23 18:30:14.915736: step: 576/1541, loss: 0.00011580361751839519 2023-01-23 18:30:15.827654: step: 578/1541, loss: 0.002409194130450487 2023-01-23 18:30:16.711042: step: 580/1541, loss: 0.15242397785186768 2023-01-23 18:30:17.548751: step: 582/1541, loss: 6.740834942320362e-05 2023-01-23 18:30:18.406563: step: 584/1541, loss: 0.00019091824651695788 2023-01-23 18:30:19.237511: step: 586/1541, loss: 1.0341287634219043e-05 2023-01-23 18:30:20.128972: step: 588/1541, loss: 0.00013465080701280385 2023-01-23 18:30:21.045157: step: 590/1541, loss: 0.20176072418689728 2023-01-23 18:30:21.968313: step: 592/1541, loss: 0.0010314668761566281 2023-01-23 18:30:22.875724: step: 594/1541, loss: 0.05060373619198799 2023-01-23 18:30:23.740707: step: 596/1541, loss: 0.013730790466070175 2023-01-23 18:30:24.592772: step: 598/1541, loss: 0.00038239164859987795 2023-01-23 18:30:25.514392: step: 600/1541, loss: 0.016005586832761765 2023-01-23 18:30:26.469316: step: 602/1541, loss: 0.042864736169576645 2023-01-23 18:30:27.355374: step: 604/1541, loss: 0.0006515458226203918 2023-01-23 18:30:28.250049: step: 606/1541, loss: 0.00015778013039380312 2023-01-23 18:30:29.191732: step: 608/1541, loss: 0.2077101171016693 2023-01-23 18:30:30.111542: step: 610/1541, loss: 0.0007028537220321596 2023-01-23 18:30:31.044015: step: 612/1541, loss: 0.006852603051811457 2023-01-23 18:30:31.938755: step: 614/1541, loss: 0.0022657220251858234 2023-01-23 18:30:32.888204: step: 616/1541, loss: 0.1780829280614853 2023-01-23 18:30:33.743656: step: 618/1541, loss: 0.10164710134267807 2023-01-23 18:30:34.683075: step: 620/1541, loss: 0.08139854669570923 2023-01-23 18:30:35.602839: step: 622/1541, loss: 0.003004611935466528 2023-01-23 18:30:36.473065: step: 624/1541, loss: 0.00010371450480306521 2023-01-23 18:30:37.379908: step: 626/1541, loss: 0.07738897949457169 2023-01-23 18:30:38.355924: step: 628/1541, loss: 6.249453872442245e-05 2023-01-23 18:30:39.235043: step: 630/1541, loss: 0.001546566723845899 2023-01-23 18:30:40.113629: step: 632/1541, loss: 2.7886002499144524e-05 2023-01-23 18:30:40.989437: step: 634/1541, loss: 0.0007722901063971221 2023-01-23 18:30:41.931967: step: 636/1541, loss: 0.19695937633514404 2023-01-23 18:30:42.820888: step: 638/1541, loss: 0.00020763222710229456 2023-01-23 18:30:43.781993: step: 640/1541, loss: 0.6280604600906372 2023-01-23 18:30:44.665317: step: 642/1541, loss: 0.07570876181125641 2023-01-23 18:30:45.536107: step: 644/1541, loss: 0.17814785242080688 2023-01-23 18:30:46.499069: step: 646/1541, loss: 0.0930863544344902 2023-01-23 18:30:47.396232: step: 648/1541, loss: 0.0003357137320563197 2023-01-23 18:30:48.285279: step: 650/1541, loss: 0.00014161737635731697 2023-01-23 18:30:49.205279: step: 652/1541, loss: 0.11429539322853088 2023-01-23 18:30:50.136322: step: 654/1541, loss: 0.00294082285836339 2023-01-23 18:30:51.067376: step: 656/1541, loss: 0.02020624652504921 2023-01-23 18:30:51.941844: step: 658/1541, loss: 0.008722454309463501 2023-01-23 18:30:52.917969: step: 660/1541, loss: 0.07401624321937561 2023-01-23 18:30:53.834031: step: 662/1541, loss: 0.3866901397705078 2023-01-23 18:30:54.734947: step: 664/1541, loss: 0.018901431933045387 2023-01-23 18:30:55.636622: step: 666/1541, loss: 0.029354311525821686 2023-01-23 18:30:56.512257: step: 668/1541, loss: 0.00022509093105327338 2023-01-23 18:30:57.427430: step: 670/1541, loss: 0.04739103093743324 2023-01-23 18:30:58.307870: step: 672/1541, loss: 0.012087889946997166 2023-01-23 18:30:59.155557: step: 674/1541, loss: 0.0005976957036182284 2023-01-23 18:31:00.034224: step: 676/1541, loss: 0.2564072906970978 2023-01-23 18:31:00.941923: step: 678/1541, loss: 0.0006862831069156528 2023-01-23 18:31:01.872861: step: 680/1541, loss: 0.00020125447190366685 2023-01-23 18:31:02.747095: step: 682/1541, loss: 0.03894485905766487 2023-01-23 18:31:03.613082: step: 684/1541, loss: 0.00455365190282464 2023-01-23 18:31:04.512492: step: 686/1541, loss: 1.1295256614685059 2023-01-23 18:31:05.497881: step: 688/1541, loss: 0.021864455193281174 2023-01-23 18:31:06.367517: step: 690/1541, loss: 0.0008952039061114192 2023-01-23 18:31:07.210238: step: 692/1541, loss: 9.862409206107259e-05 2023-01-23 18:31:08.115542: step: 694/1541, loss: 0.0008152705268003047 2023-01-23 18:31:08.993897: step: 696/1541, loss: 0.009786244481801987 2023-01-23 18:31:09.903643: step: 698/1541, loss: 0.0002392929745838046 2023-01-23 18:31:10.784379: step: 700/1541, loss: 0.042336348444223404 2023-01-23 18:31:11.684052: step: 702/1541, loss: 0.09532447159290314 2023-01-23 18:31:12.576900: step: 704/1541, loss: 0.5923126339912415 2023-01-23 18:31:13.460178: step: 706/1541, loss: 0.00018400906992610544 2023-01-23 18:31:14.361932: step: 708/1541, loss: 0.1366436928510666 2023-01-23 18:31:15.218429: step: 710/1541, loss: 0.0009375408408232033 2023-01-23 18:31:16.134774: step: 712/1541, loss: 0.00024411422782577574 2023-01-23 18:31:17.000623: step: 714/1541, loss: 0.002963940380141139 2023-01-23 18:31:17.902683: step: 716/1541, loss: 0.0004766710626427084 2023-01-23 18:31:18.783434: step: 718/1541, loss: 0.11029252409934998 2023-01-23 18:31:19.668614: step: 720/1541, loss: 1.029180884361267 2023-01-23 18:31:20.586995: step: 722/1541, loss: 0.0025924642104655504 2023-01-23 18:31:21.576299: step: 724/1541, loss: 0.0024673768784850836 2023-01-23 18:31:22.475248: step: 726/1541, loss: 0.0002780741488095373 2023-01-23 18:31:23.386429: step: 728/1541, loss: 0.020576143637299538 2023-01-23 18:31:24.300894: step: 730/1541, loss: 0.033212751150131226 2023-01-23 18:31:25.175486: step: 732/1541, loss: 0.04144524037837982 2023-01-23 18:31:26.117157: step: 734/1541, loss: 0.03766733035445213 2023-01-23 18:31:27.045075: step: 736/1541, loss: 0.08768666535615921 2023-01-23 18:31:27.906301: step: 738/1541, loss: 0.0008547852048650384 2023-01-23 18:31:28.887162: step: 740/1541, loss: 0.10054557025432587 2023-01-23 18:31:29.873546: step: 742/1541, loss: 0.0008183708414435387 2023-01-23 18:31:30.778832: step: 744/1541, loss: 0.0004259400302544236 2023-01-23 18:31:31.638799: step: 746/1541, loss: 0.0004071072908118367 2023-01-23 18:31:32.558269: step: 748/1541, loss: 0.0006737482035532594 2023-01-23 18:31:33.427794: step: 750/1541, loss: 0.0005535252275876701 2023-01-23 18:31:34.351688: step: 752/1541, loss: 0.04799406975507736 2023-01-23 18:31:35.308378: step: 754/1541, loss: 0.01276666484773159 2023-01-23 18:31:36.176889: step: 756/1541, loss: 0.013659871183335781 2023-01-23 18:31:37.034930: step: 758/1541, loss: 0.008824997581541538 2023-01-23 18:31:37.962557: step: 760/1541, loss: 0.08262001723051071 2023-01-23 18:31:38.843931: step: 762/1541, loss: 0.0003510269452817738 2023-01-23 18:31:39.724205: step: 764/1541, loss: 0.0015672302106395364 2023-01-23 18:31:40.609440: step: 766/1541, loss: 0.2642131745815277 2023-01-23 18:31:41.511811: step: 768/1541, loss: 0.008770450949668884 2023-01-23 18:31:42.528792: step: 770/1541, loss: 0.10141615569591522 2023-01-23 18:31:43.357288: step: 772/1541, loss: 0.021464619785547256 2023-01-23 18:31:44.235238: step: 774/1541, loss: 0.006611498538404703 2023-01-23 18:31:45.091014: step: 776/1541, loss: 0.002030926290899515 2023-01-23 18:31:46.004863: step: 778/1541, loss: 0.007084494456648827 2023-01-23 18:31:46.892444: step: 780/1541, loss: 0.003311683889478445 2023-01-23 18:31:47.772534: step: 782/1541, loss: 0.09543313831090927 2023-01-23 18:31:48.640509: step: 784/1541, loss: 0.2291882336139679 2023-01-23 18:31:49.521492: step: 786/1541, loss: 0.0004942366504110396 2023-01-23 18:31:50.405783: step: 788/1541, loss: 0.00020174267410766333 2023-01-23 18:31:51.240386: step: 790/1541, loss: 0.001508191111497581 2023-01-23 18:31:52.134281: step: 792/1541, loss: 0.03091309778392315 2023-01-23 18:31:53.057775: step: 794/1541, loss: 0.027061166241765022 2023-01-23 18:31:54.025104: step: 796/1541, loss: 0.2120538055896759 2023-01-23 18:31:54.943080: step: 798/1541, loss: 0.0742291510105133 2023-01-23 18:31:55.847835: step: 800/1541, loss: 0.0011190050281584263 2023-01-23 18:31:56.760750: step: 802/1541, loss: 0.0003255074843764305 2023-01-23 18:31:57.640044: step: 804/1541, loss: 0.004672880284488201 2023-01-23 18:31:58.580562: step: 806/1541, loss: 0.002679572906345129 2023-01-23 18:31:59.471502: step: 808/1541, loss: 0.017152149230241776 2023-01-23 18:32:00.347885: step: 810/1541, loss: 0.03076045587658882 2023-01-23 18:32:01.242842: step: 812/1541, loss: 0.005303008947521448 2023-01-23 18:32:02.202790: step: 814/1541, loss: 0.07909611612558365 2023-01-23 18:32:03.102059: step: 816/1541, loss: 0.0040581063367426395 2023-01-23 18:32:04.048585: step: 818/1541, loss: 0.004024209454655647 2023-01-23 18:32:04.946892: step: 820/1541, loss: 0.003374508349224925 2023-01-23 18:32:05.875954: step: 822/1541, loss: 0.14635659754276276 2023-01-23 18:32:06.730519: step: 824/1541, loss: 0.005153413861989975 2023-01-23 18:32:07.653061: step: 826/1541, loss: 0.00016245862934738398 2023-01-23 18:32:08.560690: step: 828/1541, loss: 0.23875299096107483 2023-01-23 18:32:09.486288: step: 830/1541, loss: 0.0065600634552538395 2023-01-23 18:32:10.353479: step: 832/1541, loss: 0.006028102245181799 2023-01-23 18:32:11.248816: step: 834/1541, loss: 0.08395050466060638 2023-01-23 18:32:12.133393: step: 836/1541, loss: 0.04194490984082222 2023-01-23 18:32:13.039417: step: 838/1541, loss: 0.040702078491449356 2023-01-23 18:32:13.943916: step: 840/1541, loss: 0.04683607071638107 2023-01-23 18:32:14.805697: step: 842/1541, loss: 0.18096262216567993 2023-01-23 18:32:15.792247: step: 844/1541, loss: 0.00025430068490095437 2023-01-23 18:32:16.631080: step: 846/1541, loss: 0.32877105474472046 2023-01-23 18:32:17.531122: step: 848/1541, loss: 0.1906566172838211 2023-01-23 18:32:18.431563: step: 850/1541, loss: 0.575947642326355 2023-01-23 18:32:19.323671: step: 852/1541, loss: 0.014429804868996143 2023-01-23 18:32:20.234944: step: 854/1541, loss: 0.00014769405242986977 2023-01-23 18:32:21.191460: step: 856/1541, loss: 2.470161962264683e-05 2023-01-23 18:32:22.062756: step: 858/1541, loss: 0.008039368316531181 2023-01-23 18:32:23.020484: step: 860/1541, loss: 0.1273920089006424 2023-01-23 18:32:23.908685: step: 862/1541, loss: 0.0016733369557186961 2023-01-23 18:32:24.809623: step: 864/1541, loss: 0.0003343689313624054 2023-01-23 18:32:25.688485: step: 866/1541, loss: 0.0007101839291863143 2023-01-23 18:32:26.629032: step: 868/1541, loss: 0.0007743935566395521 2023-01-23 18:32:27.545710: step: 870/1541, loss: 0.0008470473112538457 2023-01-23 18:32:28.445993: step: 872/1541, loss: 0.18672122061252594 2023-01-23 18:32:29.344174: step: 874/1541, loss: 0.004105246160179377 2023-01-23 18:32:30.223597: step: 876/1541, loss: 0.0007880930206738412 2023-01-23 18:32:31.095786: step: 878/1541, loss: 0.006235281936824322 2023-01-23 18:32:32.008274: step: 880/1541, loss: 0.0001087581185856834 2023-01-23 18:32:32.892787: step: 882/1541, loss: 0.007970857433974743 2023-01-23 18:32:33.825615: step: 884/1541, loss: 0.002771912142634392 2023-01-23 18:32:34.736520: step: 886/1541, loss: 0.0001793791598174721 2023-01-23 18:32:35.623743: step: 888/1541, loss: 0.0001594496425241232 2023-01-23 18:32:36.523393: step: 890/1541, loss: 0.03899802640080452 2023-01-23 18:32:37.414358: step: 892/1541, loss: 0.06769724190235138 2023-01-23 18:32:38.278627: step: 894/1541, loss: 0.06595747917890549 2023-01-23 18:32:39.151712: step: 896/1541, loss: 0.0002762594085652381 2023-01-23 18:32:40.052235: step: 898/1541, loss: 0.004337851889431477 2023-01-23 18:32:40.924843: step: 900/1541, loss: 0.0006392514915205538 2023-01-23 18:32:41.812612: step: 902/1541, loss: 0.00031011400278657675 2023-01-23 18:32:42.800896: step: 904/1541, loss: 0.3615485429763794 2023-01-23 18:32:43.740119: step: 906/1541, loss: 0.754432737827301 2023-01-23 18:32:44.607266: step: 908/1541, loss: 0.00024065042089205235 2023-01-23 18:32:45.569982: step: 910/1541, loss: 0.049627210944890976 2023-01-23 18:32:46.503065: step: 912/1541, loss: 0.00014446594286710024 2023-01-23 18:32:47.383512: step: 914/1541, loss: 0.003139292122796178 2023-01-23 18:32:48.300979: step: 916/1541, loss: 0.29825353622436523 2023-01-23 18:32:49.209998: step: 918/1541, loss: 0.00033647261443547904 2023-01-23 18:32:50.124312: step: 920/1541, loss: 0.0003285211860202253 2023-01-23 18:32:51.137686: step: 922/1541, loss: 0.01720307022333145 2023-01-23 18:32:52.032293: step: 924/1541, loss: 0.24417981505393982 2023-01-23 18:32:52.922948: step: 926/1541, loss: 0.008528709411621094 2023-01-23 18:32:53.881941: step: 928/1541, loss: 8.251901454059407e-05 2023-01-23 18:32:54.855292: step: 930/1541, loss: 0.13942061364650726 2023-01-23 18:32:55.739630: step: 932/1541, loss: 0.0022351054940372705 2023-01-23 18:32:56.608027: step: 934/1541, loss: 0.2573232650756836 2023-01-23 18:32:57.480278: step: 936/1541, loss: 0.026154683902859688 2023-01-23 18:32:58.379724: step: 938/1541, loss: 0.9729883074760437 2023-01-23 18:32:59.285765: step: 940/1541, loss: 0.03235050290822983 2023-01-23 18:33:00.132189: step: 942/1541, loss: 0.0021356549113988876 2023-01-23 18:33:00.999842: step: 944/1541, loss: 0.009603592567145824 2023-01-23 18:33:01.892231: step: 946/1541, loss: 0.00220996281132102 2023-01-23 18:33:02.819839: step: 948/1541, loss: 0.2646619379520416 2023-01-23 18:33:03.672311: step: 950/1541, loss: 0.0003762739070225507 2023-01-23 18:33:04.550168: step: 952/1541, loss: 0.013926247134804726 2023-01-23 18:33:05.470584: step: 954/1541, loss: 0.0006262824172154069 2023-01-23 18:33:06.403402: step: 956/1541, loss: 0.0009075960260815918 2023-01-23 18:33:07.317086: step: 958/1541, loss: 0.002127670217305422 2023-01-23 18:33:08.222006: step: 960/1541, loss: 0.0012118211016058922 2023-01-23 18:33:09.163563: step: 962/1541, loss: 0.001166346250101924 2023-01-23 18:33:10.062678: step: 964/1541, loss: 8.975229866337031e-05 2023-01-23 18:33:10.963326: step: 966/1541, loss: 0.00043758773244917393 2023-01-23 18:33:11.839553: step: 968/1541, loss: 0.2522099018096924 2023-01-23 18:33:12.730742: step: 970/1541, loss: 0.0007350408122874796 2023-01-23 18:33:13.600245: step: 972/1541, loss: 0.9120892286300659 2023-01-23 18:33:14.496875: step: 974/1541, loss: 0.08057833462953568 2023-01-23 18:33:15.376397: step: 976/1541, loss: 0.2503592371940613 2023-01-23 18:33:16.273814: step: 978/1541, loss: 0.10196007043123245 2023-01-23 18:33:17.193737: step: 980/1541, loss: 0.007905839011073112 2023-01-23 18:33:18.094301: step: 982/1541, loss: 8.790067658992484e-05 2023-01-23 18:33:18.959054: step: 984/1541, loss: 0.0019142255187034607 2023-01-23 18:33:19.823574: step: 986/1541, loss: 0.00839295331388712 2023-01-23 18:33:20.783030: step: 988/1541, loss: 0.04496431350708008 2023-01-23 18:33:21.708246: step: 990/1541, loss: 0.0031969437841326 2023-01-23 18:33:22.629415: step: 992/1541, loss: 0.0002739084593486041 2023-01-23 18:33:23.537062: step: 994/1541, loss: 0.26363059878349304 2023-01-23 18:33:24.435928: step: 996/1541, loss: 0.03532680124044418 2023-01-23 18:33:25.355107: step: 998/1541, loss: 0.001191756222397089 2023-01-23 18:33:26.245942: step: 1000/1541, loss: 0.0003382558934390545 2023-01-23 18:33:27.108715: step: 1002/1541, loss: 0.15446391701698303 2023-01-23 18:33:27.998245: step: 1004/1541, loss: 0.004143576603382826 2023-01-23 18:33:28.931625: step: 1006/1541, loss: 0.00037887587677687407 2023-01-23 18:33:29.819460: step: 1008/1541, loss: 0.013889992609620094 2023-01-23 18:33:30.762673: step: 1010/1541, loss: 0.09936697036027908 2023-01-23 18:33:31.672925: step: 1012/1541, loss: 0.03545970469713211 2023-01-23 18:33:32.664359: step: 1014/1541, loss: 0.13445137441158295 2023-01-23 18:33:33.560785: step: 1016/1541, loss: 0.18303200602531433 2023-01-23 18:33:34.441575: step: 1018/1541, loss: 0.0013051379937678576 2023-01-23 18:33:35.369461: step: 1020/1541, loss: 0.04373924061655998 2023-01-23 18:33:36.259755: step: 1022/1541, loss: 0.00041525281267240644 2023-01-23 18:33:37.208151: step: 1024/1541, loss: 0.05896700918674469 2023-01-23 18:33:38.083922: step: 1026/1541, loss: 0.0044064451940357685 2023-01-23 18:33:38.982194: step: 1028/1541, loss: 0.11562097072601318 2023-01-23 18:33:39.884432: step: 1030/1541, loss: 0.0026773654390126467 2023-01-23 18:33:40.784767: step: 1032/1541, loss: 0.033693019300699234 2023-01-23 18:33:41.695348: step: 1034/1541, loss: 0.06540220230817795 2023-01-23 18:33:42.618507: step: 1036/1541, loss: 0.05493145436048508 2023-01-23 18:33:43.483671: step: 1038/1541, loss: 0.3774963617324829 2023-01-23 18:33:44.342410: step: 1040/1541, loss: 0.0020207250490784645 2023-01-23 18:33:45.256753: step: 1042/1541, loss: 0.0020855972543358803 2023-01-23 18:33:46.245144: step: 1044/1541, loss: 0.05275467783212662 2023-01-23 18:33:47.154105: step: 1046/1541, loss: 0.009177474305033684 2023-01-23 18:33:48.020539: step: 1048/1541, loss: 0.0001405993098160252 2023-01-23 18:33:48.933842: step: 1050/1541, loss: 0.0035508940927684307 2023-01-23 18:33:49.823659: step: 1052/1541, loss: 0.028404347598552704 2023-01-23 18:33:50.696963: step: 1054/1541, loss: 0.10402963310480118 2023-01-23 18:33:51.576880: step: 1056/1541, loss: 0.0007099430076777935 2023-01-23 18:33:52.463132: step: 1058/1541, loss: 0.057602912187576294 2023-01-23 18:33:53.366860: step: 1060/1541, loss: 0.0047285351902246475 2023-01-23 18:33:54.231742: step: 1062/1541, loss: 0.11474250257015228 2023-01-23 18:33:55.191613: step: 1064/1541, loss: 0.007085460238158703 2023-01-23 18:33:56.070389: step: 1066/1541, loss: 0.0001152875047409907 2023-01-23 18:33:56.947398: step: 1068/1541, loss: 0.00041949161095544696 2023-01-23 18:33:57.818291: step: 1070/1541, loss: 0.019604511559009552 2023-01-23 18:33:58.689378: step: 1072/1541, loss: 0.00044770792010240257 2023-01-23 18:33:59.585169: step: 1074/1541, loss: 0.004860677756369114 2023-01-23 18:34:00.537209: step: 1076/1541, loss: 0.0028881041798740625 2023-01-23 18:34:01.505692: step: 1078/1541, loss: 0.0014555092202499509 2023-01-23 18:34:02.416625: step: 1080/1541, loss: 0.03790031000971794 2023-01-23 18:34:03.256656: step: 1082/1541, loss: 0.0759812444448471 2023-01-23 18:34:04.218441: step: 1084/1541, loss: 0.00013797525025438517 2023-01-23 18:34:05.172528: step: 1086/1541, loss: 0.08599801361560822 2023-01-23 18:34:06.061962: step: 1088/1541, loss: 0.00043323702993802726 2023-01-23 18:34:06.950163: step: 1090/1541, loss: 0.09110701084136963 2023-01-23 18:34:07.823733: step: 1092/1541, loss: 0.08419515192508698 2023-01-23 18:34:08.744020: step: 1094/1541, loss: 8.743802027311176e-05 2023-01-23 18:34:09.647004: step: 1096/1541, loss: 0.0029209183994680643 2023-01-23 18:34:10.527413: step: 1098/1541, loss: 0.03225964680314064 2023-01-23 18:34:11.450955: step: 1100/1541, loss: 0.2596701979637146 2023-01-23 18:34:12.324299: step: 1102/1541, loss: 0.688955545425415 2023-01-23 18:34:13.277026: step: 1104/1541, loss: 0.02132692001760006 2023-01-23 18:34:14.220104: step: 1106/1541, loss: 0.09136263281106949 2023-01-23 18:34:15.091180: step: 1108/1541, loss: 0.005687041208148003 2023-01-23 18:34:15.975146: step: 1110/1541, loss: 0.00038497638888657093 2023-01-23 18:34:16.883625: step: 1112/1541, loss: 0.004540789872407913 2023-01-23 18:34:17.756088: step: 1114/1541, loss: 0.013820542953908443 2023-01-23 18:34:18.651302: step: 1116/1541, loss: 0.08847863227128983 2023-01-23 18:34:19.527043: step: 1118/1541, loss: 0.005387568846344948 2023-01-23 18:34:20.374955: step: 1120/1541, loss: 0.000714882044121623 2023-01-23 18:34:21.297221: step: 1122/1541, loss: 0.026676315814256668 2023-01-23 18:34:22.194209: step: 1124/1541, loss: 0.00038464879617094994 2023-01-23 18:34:23.071164: step: 1126/1541, loss: 0.0006465225596912205 2023-01-23 18:34:23.953862: step: 1128/1541, loss: 0.0074380128644406796 2023-01-23 18:34:24.850109: step: 1130/1541, loss: 0.0005627292557619512 2023-01-23 18:34:25.718786: step: 1132/1541, loss: 0.021132508292794228 2023-01-23 18:34:26.624177: step: 1134/1541, loss: 0.022070415318012238 2023-01-23 18:34:27.509993: step: 1136/1541, loss: 0.0026095088105648756 2023-01-23 18:34:28.417740: step: 1138/1541, loss: 0.14622066915035248 2023-01-23 18:34:29.366855: step: 1140/1541, loss: 0.0003609820269048214 2023-01-23 18:34:30.273865: step: 1142/1541, loss: 0.3829137086868286 2023-01-23 18:34:31.175532: step: 1144/1541, loss: 0.6640048027038574 2023-01-23 18:34:32.062155: step: 1146/1541, loss: 0.04835051670670509 2023-01-23 18:34:32.970113: step: 1148/1541, loss: 0.0016602296382188797 2023-01-23 18:34:33.918040: step: 1150/1541, loss: 0.0003167208924423903 2023-01-23 18:34:34.802348: step: 1152/1541, loss: 0.018132032826542854 2023-01-23 18:34:35.732868: step: 1154/1541, loss: 0.0011513065546751022 2023-01-23 18:34:36.623945: step: 1156/1541, loss: 0.007220556028187275 2023-01-23 18:34:37.509261: step: 1158/1541, loss: 0.08171363919973373 2023-01-23 18:34:38.389786: step: 1160/1541, loss: 0.0007230093469843268 2023-01-23 18:34:39.270249: step: 1162/1541, loss: 0.8156329393386841 2023-01-23 18:34:40.149223: step: 1164/1541, loss: 0.002328435191884637 2023-01-23 18:34:41.104144: step: 1166/1541, loss: 0.01222945936024189 2023-01-23 18:34:41.984102: step: 1168/1541, loss: 0.16121676564216614 2023-01-23 18:34:42.843102: step: 1170/1541, loss: 0.0020207262132316828 2023-01-23 18:34:43.777408: step: 1172/1541, loss: 0.5142428874969482 2023-01-23 18:34:44.663996: step: 1174/1541, loss: 0.0034814891405403614 2023-01-23 18:34:45.538131: step: 1176/1541, loss: 0.0033496320247650146 2023-01-23 18:34:46.477419: step: 1178/1541, loss: 0.0005354670574888587 2023-01-23 18:34:47.391621: step: 1180/1541, loss: 0.0013987733982503414 2023-01-23 18:34:48.276756: step: 1182/1541, loss: 0.006537191104143858 2023-01-23 18:34:49.176725: step: 1184/1541, loss: 0.2633938491344452 2023-01-23 18:34:50.055858: step: 1186/1541, loss: 0.02516201324760914 2023-01-23 18:34:50.948742: step: 1188/1541, loss: 0.004623127169907093 2023-01-23 18:34:51.818948: step: 1190/1541, loss: 0.0010169987799599767 2023-01-23 18:34:52.772593: step: 1192/1541, loss: 0.0006393191288225353 2023-01-23 18:34:53.692032: step: 1194/1541, loss: 0.0033348624128848314 2023-01-23 18:34:54.563182: step: 1196/1541, loss: 0.0011216058628633618 2023-01-23 18:34:55.486653: step: 1198/1541, loss: 0.04400966688990593 2023-01-23 18:34:56.376376: step: 1200/1541, loss: 0.0053372569382190704 2023-01-23 18:34:57.267542: step: 1202/1541, loss: 0.02881317026913166 2023-01-23 18:34:58.130608: step: 1204/1541, loss: 0.11413050442934036 2023-01-23 18:34:59.037370: step: 1206/1541, loss: 0.00024363469856325537 2023-01-23 18:34:59.938890: step: 1208/1541, loss: 0.0003635520697571337 2023-01-23 18:35:00.850552: step: 1210/1541, loss: 0.0003772758645936847 2023-01-23 18:35:01.748441: step: 1212/1541, loss: 0.07181130349636078 2023-01-23 18:35:02.645325: step: 1214/1541, loss: 0.004638098180294037 2023-01-23 18:35:03.549473: step: 1216/1541, loss: 0.18800237774848938 2023-01-23 18:35:04.482515: step: 1218/1541, loss: 0.005803467705845833 2023-01-23 18:35:05.359853: step: 1220/1541, loss: 0.030058065429329872 2023-01-23 18:35:06.294951: step: 1222/1541, loss: 0.006048867478966713 2023-01-23 18:35:07.185695: step: 1224/1541, loss: 0.029324062168598175 2023-01-23 18:35:08.074486: step: 1226/1541, loss: 0.09983278065919876 2023-01-23 18:35:08.955199: step: 1228/1541, loss: 0.011698516085743904 2023-01-23 18:35:09.875033: step: 1230/1541, loss: 0.0021289039868861437 2023-01-23 18:35:10.835950: step: 1232/1541, loss: 0.0015328795416280627 2023-01-23 18:35:11.710584: step: 1234/1541, loss: 2.0249764929758385e-05 2023-01-23 18:35:12.617391: step: 1236/1541, loss: 0.0016323892632499337 2023-01-23 18:35:13.498497: step: 1238/1541, loss: 0.018680335953831673 2023-01-23 18:35:14.449342: step: 1240/1541, loss: 0.000664460938423872 2023-01-23 18:35:15.355877: step: 1242/1541, loss: 0.00029517480288632214 2023-01-23 18:35:16.242459: step: 1244/1541, loss: 0.0049734534695744514 2023-01-23 18:35:17.203192: step: 1246/1541, loss: 0.4456152617931366 2023-01-23 18:35:18.091383: step: 1248/1541, loss: 0.0015286774141713977 2023-01-23 18:35:18.990954: step: 1250/1541, loss: 0.0002061462146230042 2023-01-23 18:35:19.879433: step: 1252/1541, loss: 0.001396896899677813 2023-01-23 18:35:20.755561: step: 1254/1541, loss: 0.0005375048494897783 2023-01-23 18:35:21.651611: step: 1256/1541, loss: 0.22675952315330505 2023-01-23 18:35:22.533666: step: 1258/1541, loss: 0.016022522002458572 2023-01-23 18:35:23.432768: step: 1260/1541, loss: 0.005478229373693466 2023-01-23 18:35:24.342866: step: 1262/1541, loss: 0.009758873842656612 2023-01-23 18:35:25.228649: step: 1264/1541, loss: 0.002886389847844839 2023-01-23 18:35:26.112471: step: 1266/1541, loss: 0.00021579075837507844 2023-01-23 18:35:27.067003: step: 1268/1541, loss: 0.0006614456069655716 2023-01-23 18:35:27.938516: step: 1270/1541, loss: 0.5515654683113098 2023-01-23 18:35:28.783748: step: 1272/1541, loss: 0.0025689234025776386 2023-01-23 18:35:29.672787: step: 1274/1541, loss: 0.02870737761259079 2023-01-23 18:35:30.615825: step: 1276/1541, loss: 2.7220072746276855 2023-01-23 18:35:31.512862: step: 1278/1541, loss: 0.043870046734809875 2023-01-23 18:35:32.399541: step: 1280/1541, loss: 0.15234632790088654 2023-01-23 18:35:33.280813: step: 1282/1541, loss: 0.0024836817756295204 2023-01-23 18:35:34.173712: step: 1284/1541, loss: 0.1423044204711914 2023-01-23 18:35:35.081432: step: 1286/1541, loss: 0.0017886464484035969 2023-01-23 18:35:35.973739: step: 1288/1541, loss: 0.004951875191181898 2023-01-23 18:35:36.847167: step: 1290/1541, loss: 0.09015563130378723 2023-01-23 18:35:37.701999: step: 1292/1541, loss: 8.731057459954172e-05 2023-01-23 18:35:38.583104: step: 1294/1541, loss: 0.0011276538716629148 2023-01-23 18:35:39.466477: step: 1296/1541, loss: 0.005349237471818924 2023-01-23 18:35:40.341188: step: 1298/1541, loss: 9.450865036342293e-05 2023-01-23 18:35:41.221236: step: 1300/1541, loss: 0.11557196080684662 2023-01-23 18:35:42.139797: step: 1302/1541, loss: 0.000759228365495801 2023-01-23 18:35:43.005185: step: 1304/1541, loss: 0.0021174969151616096 2023-01-23 18:35:43.894869: step: 1306/1541, loss: 0.09092172980308533 2023-01-23 18:35:44.814341: step: 1308/1541, loss: 0.07568895071744919 2023-01-23 18:35:45.758280: step: 1310/1541, loss: 0.0007735935505479574 2023-01-23 18:35:46.652371: step: 1312/1541, loss: 0.09961234033107758 2023-01-23 18:35:47.528687: step: 1314/1541, loss: 0.002515062689781189 2023-01-23 18:35:48.404139: step: 1316/1541, loss: 0.0009265574626624584 2023-01-23 18:35:49.269919: step: 1318/1541, loss: 5.1521838031476364e-05 2023-01-23 18:35:50.196825: step: 1320/1541, loss: 0.012946589849889278 2023-01-23 18:35:51.105896: step: 1322/1541, loss: 0.009909655898809433 2023-01-23 18:35:51.986426: step: 1324/1541, loss: 0.006509741768240929 2023-01-23 18:35:52.860329: step: 1326/1541, loss: 6.273035978665575e-05 2023-01-23 18:35:53.745839: step: 1328/1541, loss: 0.030066004022955894 2023-01-23 18:35:54.639548: step: 1330/1541, loss: 0.0020730383694171906 2023-01-23 18:35:55.484562: step: 1332/1541, loss: 0.009178526699543 2023-01-23 18:35:56.362025: step: 1334/1541, loss: 0.0003285598650109023 2023-01-23 18:35:57.251983: step: 1336/1541, loss: 0.00023949748720042408 2023-01-23 18:35:58.123132: step: 1338/1541, loss: 0.0004278788692317903 2023-01-23 18:35:58.961425: step: 1340/1541, loss: 0.05243270844221115 2023-01-23 18:35:59.845185: step: 1342/1541, loss: 0.1856975555419922 2023-01-23 18:36:00.827278: step: 1344/1541, loss: 0.0016032513231039047 2023-01-23 18:36:01.789042: step: 1346/1541, loss: 0.09887164831161499 2023-01-23 18:36:02.682853: step: 1348/1541, loss: 1.3378592729568481 2023-01-23 18:36:03.556873: step: 1350/1541, loss: 0.0021671680733561516 2023-01-23 18:36:04.475126: step: 1352/1541, loss: 0.007731545250862837 2023-01-23 18:36:05.375190: step: 1354/1541, loss: 0.000505596399307251 2023-01-23 18:36:06.248842: step: 1356/1541, loss: 0.2369929403066635 2023-01-23 18:36:07.123591: step: 1358/1541, loss: 0.14611786603927612 2023-01-23 18:36:08.052612: step: 1360/1541, loss: 0.036762069910764694 2023-01-23 18:36:08.932853: step: 1362/1541, loss: 0.23965299129486084 2023-01-23 18:36:09.811554: step: 1364/1541, loss: 0.003026035148650408 2023-01-23 18:36:10.676769: step: 1366/1541, loss: 0.003938100766390562 2023-01-23 18:36:11.543111: step: 1368/1541, loss: 0.006977391429245472 2023-01-23 18:36:12.437809: step: 1370/1541, loss: 0.10919887572526932 2023-01-23 18:36:13.394163: step: 1372/1541, loss: 0.03530018404126167 2023-01-23 18:36:14.318526: step: 1374/1541, loss: 0.00042275807936675847 2023-01-23 18:36:15.236567: step: 1376/1541, loss: 0.0020109033212065697 2023-01-23 18:36:16.172958: step: 1378/1541, loss: 6.344691064441577e-05 2023-01-23 18:36:17.050459: step: 1380/1541, loss: 0.48627156019210815 2023-01-23 18:36:17.915581: step: 1382/1541, loss: 0.06492997705936432 2023-01-23 18:36:18.828842: step: 1384/1541, loss: 0.11045565456151962 2023-01-23 18:36:19.687986: step: 1386/1541, loss: 0.01310120802372694 2023-01-23 18:36:20.605898: step: 1388/1541, loss: 0.0032936586067080498 2023-01-23 18:36:21.632476: step: 1390/1541, loss: 0.001689239521510899 2023-01-23 18:36:22.527913: step: 1392/1541, loss: 0.016272559762001038 2023-01-23 18:36:23.408125: step: 1394/1541, loss: 0.0004260234418325126 2023-01-23 18:36:24.329231: step: 1396/1541, loss: 0.16167527437210083 2023-01-23 18:36:25.190574: step: 1398/1541, loss: 0.0001240137789864093 2023-01-23 18:36:26.058517: step: 1400/1541, loss: 0.0006137460004538298 2023-01-23 18:36:26.990129: step: 1402/1541, loss: 5.672126280842349e-05 2023-01-23 18:36:27.868604: step: 1404/1541, loss: 0.0003758296661544591 2023-01-23 18:36:28.739026: step: 1406/1541, loss: 0.000333671661792323 2023-01-23 18:36:29.625570: step: 1408/1541, loss: 0.07658278197050095 2023-01-23 18:36:30.524434: step: 1410/1541, loss: 4.604876085068099e-05 2023-01-23 18:36:31.415461: step: 1412/1541, loss: 0.0003302149707451463 2023-01-23 18:36:32.343572: step: 1414/1541, loss: 0.0026656335685402155 2023-01-23 18:36:33.314565: step: 1416/1541, loss: 0.004883022978901863 2023-01-23 18:36:34.204473: step: 1418/1541, loss: 0.09886103868484497 2023-01-23 18:36:35.063640: step: 1420/1541, loss: 0.04558298736810684 2023-01-23 18:36:35.929522: step: 1422/1541, loss: 0.0010800898307934403 2023-01-23 18:36:36.840422: step: 1424/1541, loss: 0.14055639505386353 2023-01-23 18:36:37.782501: step: 1426/1541, loss: 0.0014966302551329136 2023-01-23 18:36:38.650013: step: 1428/1541, loss: 0.002386734588071704 2023-01-23 18:36:39.540840: step: 1430/1541, loss: 0.014854209497570992 2023-01-23 18:36:40.440007: step: 1432/1541, loss: 0.0007217118400149047 2023-01-23 18:36:41.277540: step: 1434/1541, loss: 0.00010377315629739314 2023-01-23 18:36:42.277558: step: 1436/1541, loss: 0.09949243813753128 2023-01-23 18:36:43.230048: step: 1438/1541, loss: 0.0033785447012633085 2023-01-23 18:36:44.103162: step: 1440/1541, loss: 0.0009690954466350377 2023-01-23 18:36:45.077801: step: 1442/1541, loss: 0.01725723221898079 2023-01-23 18:36:45.925923: step: 1444/1541, loss: 0.0008518173708580434 2023-01-23 18:36:46.762825: step: 1446/1541, loss: 0.005704888142645359 2023-01-23 18:36:47.654323: step: 1448/1541, loss: 0.08573175221681595 2023-01-23 18:36:48.519360: step: 1450/1541, loss: 0.0014376717153936625 2023-01-23 18:36:49.397703: step: 1452/1541, loss: 0.0066039348021149635 2023-01-23 18:36:50.310817: step: 1454/1541, loss: 0.0995139628648758 2023-01-23 18:36:51.191692: step: 1456/1541, loss: 0.0009935232810676098 2023-01-23 18:36:52.160281: step: 1458/1541, loss: 0.012458342127501965 2023-01-23 18:36:53.041251: step: 1460/1541, loss: 0.04614356905221939 2023-01-23 18:36:53.908015: step: 1462/1541, loss: 0.011025974527001381 2023-01-23 18:36:54.889715: step: 1464/1541, loss: 0.02535507269203663 2023-01-23 18:36:55.774589: step: 1466/1541, loss: 0.0001892825384857133 2023-01-23 18:36:56.685706: step: 1468/1541, loss: 0.038323622196912766 2023-01-23 18:36:57.566672: step: 1470/1541, loss: 0.07277994602918625 2023-01-23 18:36:58.506965: step: 1472/1541, loss: 0.0073679909110069275 2023-01-23 18:36:59.498228: step: 1474/1541, loss: 0.007254000753164291 2023-01-23 18:37:00.391557: step: 1476/1541, loss: 0.0025243042036890984 2023-01-23 18:37:01.294112: step: 1478/1541, loss: 0.007168468553572893 2023-01-23 18:37:02.180715: step: 1480/1541, loss: 0.013066116720438004 2023-01-23 18:37:03.087152: step: 1482/1541, loss: 0.00016911076090764254 2023-01-23 18:37:03.992624: step: 1484/1541, loss: 0.4913274049758911 2023-01-23 18:37:04.875143: step: 1486/1541, loss: 0.001120762899518013 2023-01-23 18:37:05.827172: step: 1488/1541, loss: 0.0679938793182373 2023-01-23 18:37:06.789605: step: 1490/1541, loss: 0.000911921204533428 2023-01-23 18:37:07.703865: step: 1492/1541, loss: 0.20166093111038208 2023-01-23 18:37:08.592316: step: 1494/1541, loss: 8.794862515060231e-05 2023-01-23 18:37:09.506630: step: 1496/1541, loss: 0.0021696938201785088 2023-01-23 18:37:10.443407: step: 1498/1541, loss: 0.4374047517776489 2023-01-23 18:37:11.381878: step: 1500/1541, loss: 0.10643049329519272 2023-01-23 18:37:12.285896: step: 1502/1541, loss: 0.5037903785705566 2023-01-23 18:37:13.192778: step: 1504/1541, loss: 0.08800362050533295 2023-01-23 18:37:14.095818: step: 1506/1541, loss: 0.0005562273436225951 2023-01-23 18:37:14.969297: step: 1508/1541, loss: 0.00018539537268225104 2023-01-23 18:37:15.898463: step: 1510/1541, loss: 0.005681038834154606 2023-01-23 18:37:16.790544: step: 1512/1541, loss: 0.11535076051950455 2023-01-23 18:37:17.680353: step: 1514/1541, loss: 0.26258888840675354 2023-01-23 18:37:18.578096: step: 1516/1541, loss: 0.21263907849788666 2023-01-23 18:37:19.475367: step: 1518/1541, loss: 0.0007241339772008359 2023-01-23 18:37:20.375077: step: 1520/1541, loss: 0.04057789966464043 2023-01-23 18:37:21.269942: step: 1522/1541, loss: 0.00031818068237043917 2023-01-23 18:37:22.156208: step: 1524/1541, loss: 0.009444048628211021 2023-01-23 18:37:23.056827: step: 1526/1541, loss: 0.009549995884299278 2023-01-23 18:37:23.947401: step: 1528/1541, loss: 0.0067940326407551765 2023-01-23 18:37:24.882053: step: 1530/1541, loss: 0.002067694440484047 2023-01-23 18:37:25.762839: step: 1532/1541, loss: 0.013092946261167526 2023-01-23 18:37:26.700423: step: 1534/1541, loss: 0.03459765017032623 2023-01-23 18:37:27.579794: step: 1536/1541, loss: 0.1823272556066513 2023-01-23 18:37:28.480881: step: 1538/1541, loss: 0.0018328443402424455 2023-01-23 18:37:29.433361: step: 1540/1541, loss: 0.033122718334198 2023-01-23 18:37:30.360338: step: 1542/1541, loss: 0.0801306888461113 2023-01-23 18:37:31.362284: step: 1544/1541, loss: 0.023072045296430588 2023-01-23 18:37:32.283010: step: 1546/1541, loss: 0.0008726532687433064 2023-01-23 18:37:33.194012: step: 1548/1541, loss: 0.04146651551127434 2023-01-23 18:37:34.085058: step: 1550/1541, loss: 0.041915569454431534 2023-01-23 18:37:34.996194: step: 1552/1541, loss: 0.04056614264845848 2023-01-23 18:37:35.879546: step: 1554/1541, loss: 0.10620860010385513 2023-01-23 18:37:36.748357: step: 1556/1541, loss: 0.2740974426269531 2023-01-23 18:37:37.666967: step: 1558/1541, loss: 0.001212189788930118 2023-01-23 18:37:38.588660: step: 1560/1541, loss: 0.0009568184614181519 2023-01-23 18:37:39.490089: step: 1562/1541, loss: 0.0016868605744093657 2023-01-23 18:37:40.423365: step: 1564/1541, loss: 0.0007473021396435797 2023-01-23 18:37:41.327376: step: 1566/1541, loss: 0.1307002156972885 2023-01-23 18:37:42.228876: step: 1568/1541, loss: 0.0006989941466599703 2023-01-23 18:37:43.121665: step: 1570/1541, loss: 0.0006791444611735642 2023-01-23 18:37:43.985319: step: 1572/1541, loss: 0.00045378616778180003 2023-01-23 18:37:44.845216: step: 1574/1541, loss: 0.0026738778688013554 2023-01-23 18:37:45.752200: step: 1576/1541, loss: 0.13205300271511078 2023-01-23 18:37:46.673006: step: 1578/1541, loss: 0.0013706794707104564 2023-01-23 18:37:47.597748: step: 1580/1541, loss: 0.001001476077362895 2023-01-23 18:37:48.486055: step: 1582/1541, loss: 0.026975981891155243 2023-01-23 18:37:49.394213: step: 1584/1541, loss: 0.007721924688667059 2023-01-23 18:37:50.293782: step: 1586/1541, loss: 0.00027070671785622835 2023-01-23 18:37:51.202951: step: 1588/1541, loss: 0.13081994652748108 2023-01-23 18:37:52.083818: step: 1590/1541, loss: 0.21741820871829987 2023-01-23 18:37:52.997016: step: 1592/1541, loss: 0.005239877849817276 2023-01-23 18:37:53.879247: step: 1594/1541, loss: 0.0002299715270055458 2023-01-23 18:37:54.775739: step: 1596/1541, loss: 0.0006389754707925022 2023-01-23 18:37:55.669375: step: 1598/1541, loss: 0.0017469166778028011 2023-01-23 18:37:56.542474: step: 1600/1541, loss: 0.007440538145601749 2023-01-23 18:37:57.419426: step: 1602/1541, loss: 0.00011749268014682457 2023-01-23 18:37:58.260911: step: 1604/1541, loss: 0.0024325998965650797 2023-01-23 18:37:59.234516: step: 1606/1541, loss: 0.14793360233306885 2023-01-23 18:38:00.136284: step: 1608/1541, loss: 0.13911548256874084 2023-01-23 18:38:01.080124: step: 1610/1541, loss: 0.052226871252059937 2023-01-23 18:38:01.947397: step: 1612/1541, loss: 0.014271171763539314 2023-01-23 18:38:02.853604: step: 1614/1541, loss: 0.05052836239337921 2023-01-23 18:38:03.743706: step: 1616/1541, loss: 0.021929442882537842 2023-01-23 18:38:04.632594: step: 1618/1541, loss: 0.004678803961724043 2023-01-23 18:38:05.541022: step: 1620/1541, loss: 0.05651096627116203 2023-01-23 18:38:06.444693: step: 1622/1541, loss: 0.0014849748695269227 2023-01-23 18:38:07.340377: step: 1624/1541, loss: 0.006037917919456959 2023-01-23 18:38:08.266630: step: 1626/1541, loss: 0.17497359216213226 2023-01-23 18:38:09.201521: step: 1628/1541, loss: 0.0007311664521694183 2023-01-23 18:38:10.142699: step: 1630/1541, loss: 2.614445686340332 2023-01-23 18:38:11.045378: step: 1632/1541, loss: 0.0026278963778167963 2023-01-23 18:38:11.980293: step: 1634/1541, loss: 0.12090032547712326 2023-01-23 18:38:12.907175: step: 1636/1541, loss: 0.0021547297947108746 2023-01-23 18:38:13.772441: step: 1638/1541, loss: 0.00964913796633482 2023-01-23 18:38:14.649384: step: 1640/1541, loss: 0.0031683912966400385 2023-01-23 18:38:15.536239: step: 1642/1541, loss: 0.014560859650373459 2023-01-23 18:38:16.440628: step: 1644/1541, loss: 0.01575576141476631 2023-01-23 18:38:17.326265: step: 1646/1541, loss: 0.03385806456208229 2023-01-23 18:38:18.206335: step: 1648/1541, loss: 0.012277382425963879 2023-01-23 18:38:19.079139: step: 1650/1541, loss: 0.0005647151847369969 2023-01-23 18:38:19.938187: step: 1652/1541, loss: 0.0010113307507708669 2023-01-23 18:38:20.795117: step: 1654/1541, loss: 0.0010224722791463137 2023-01-23 18:38:21.690828: step: 1656/1541, loss: 0.002596881240606308 2023-01-23 18:38:22.579870: step: 1658/1541, loss: 0.019773779436945915 2023-01-23 18:38:23.442882: step: 1660/1541, loss: 0.0002182696625823155 2023-01-23 18:38:24.349863: step: 1662/1541, loss: 0.13560782372951508 2023-01-23 18:38:25.237670: step: 1664/1541, loss: 0.00012868657358922064 2023-01-23 18:38:26.103885: step: 1666/1541, loss: 0.08970849215984344 2023-01-23 18:38:27.006366: step: 1668/1541, loss: 0.022787155583500862 2023-01-23 18:38:27.904089: step: 1670/1541, loss: 0.006563135422766209 2023-01-23 18:38:28.788647: step: 1672/1541, loss: 0.052235547453165054 2023-01-23 18:38:29.645658: step: 1674/1541, loss: 0.0033060875721275806 2023-01-23 18:38:30.551772: step: 1676/1541, loss: 0.05210649594664574 2023-01-23 18:38:31.439860: step: 1678/1541, loss: 4.505211836658418e-05 2023-01-23 18:38:32.370940: step: 1680/1541, loss: 0.010419830679893494 2023-01-23 18:38:33.291648: step: 1682/1541, loss: 0.01315898448228836 2023-01-23 18:38:34.179224: step: 1684/1541, loss: 0.0022917483001947403 2023-01-23 18:38:35.073522: step: 1686/1541, loss: 0.00999024324119091 2023-01-23 18:38:35.926545: step: 1688/1541, loss: 0.061040010303258896 2023-01-23 18:38:36.803310: step: 1690/1541, loss: 0.04821214824914932 2023-01-23 18:38:37.880080: step: 1692/1541, loss: 0.0005431314348243177 2023-01-23 18:38:38.783373: step: 1694/1541, loss: 0.02159849926829338 2023-01-23 18:38:39.649858: step: 1696/1541, loss: 0.0030244537629187107 2023-01-23 18:38:40.586158: step: 1698/1541, loss: 0.0019891737028956413 2023-01-23 18:38:41.531875: step: 1700/1541, loss: 0.0004154724592808634 2023-01-23 18:38:42.448909: step: 1702/1541, loss: 0.007024294231086969 2023-01-23 18:38:43.336183: step: 1704/1541, loss: 0.023190990090370178 2023-01-23 18:38:44.247390: step: 1706/1541, loss: 0.015825118869543076 2023-01-23 18:38:45.206290: step: 1708/1541, loss: 0.0004978934302926064 2023-01-23 18:38:46.098356: step: 1710/1541, loss: 0.0009330411558039486 2023-01-23 18:38:47.010810: step: 1712/1541, loss: 0.012929065153002739 2023-01-23 18:38:47.879753: step: 1714/1541, loss: 0.00012068937940057367 2023-01-23 18:38:48.837563: step: 1716/1541, loss: 0.19110892713069916 2023-01-23 18:38:49.732188: step: 1718/1541, loss: 0.0010067675029858947 2023-01-23 18:38:50.642842: step: 1720/1541, loss: 0.0002947327448055148 2023-01-23 18:38:51.579390: step: 1722/1541, loss: 0.05531615763902664 2023-01-23 18:38:52.489901: step: 1724/1541, loss: 0.03461383655667305 2023-01-23 18:38:53.418919: step: 1726/1541, loss: 0.00045311779831536114 2023-01-23 18:38:54.401697: step: 1728/1541, loss: 0.0025923321954905987 2023-01-23 18:38:55.272644: step: 1730/1541, loss: 0.040683384984731674 2023-01-23 18:38:56.252665: step: 1732/1541, loss: 0.000202852301299572 2023-01-23 18:38:57.158451: step: 1734/1541, loss: 0.039443954825401306 2023-01-23 18:38:58.036049: step: 1736/1541, loss: 0.0016907009994611144 2023-01-23 18:38:58.937663: step: 1738/1541, loss: 0.19063088297843933 2023-01-23 18:38:59.818461: step: 1740/1541, loss: 0.014179637655615807 2023-01-23 18:39:00.718313: step: 1742/1541, loss: 0.0008222992764785886 2023-01-23 18:39:01.621038: step: 1744/1541, loss: 0.16122151911258698 2023-01-23 18:39:02.471655: step: 1746/1541, loss: 0.0123096639290452 2023-01-23 18:39:03.331766: step: 1748/1541, loss: 0.000599354156292975 2023-01-23 18:39:04.232882: step: 1750/1541, loss: 0.16565556824207306 2023-01-23 18:39:05.120360: step: 1752/1541, loss: 0.001583558158017695 2023-01-23 18:39:06.015159: step: 1754/1541, loss: 0.1550486981868744 2023-01-23 18:39:06.911830: step: 1756/1541, loss: 0.0036365343257784843 2023-01-23 18:39:07.817247: step: 1758/1541, loss: 0.007865546271204948 2023-01-23 18:39:08.693272: step: 1760/1541, loss: 0.1608966737985611 2023-01-23 18:39:09.616985: step: 1762/1541, loss: 0.007909988053143024 2023-01-23 18:39:10.518356: step: 1764/1541, loss: 0.00044170732144266367 2023-01-23 18:39:11.435264: step: 1766/1541, loss: 0.01953660510480404 2023-01-23 18:39:12.311716: step: 1768/1541, loss: 0.0016445154324173927 2023-01-23 18:39:13.204322: step: 1770/1541, loss: 0.0002324179222341627 2023-01-23 18:39:14.106649: step: 1772/1541, loss: 0.03796854242682457 2023-01-23 18:39:15.027940: step: 1774/1541, loss: 0.0038445615209639072 2023-01-23 18:39:15.949625: step: 1776/1541, loss: 0.007851897738873959 2023-01-23 18:39:16.859922: step: 1778/1541, loss: -2.6053190595121123e-05 2023-01-23 18:39:17.727589: step: 1780/1541, loss: -2.469130777171813e-06 2023-01-23 18:39:18.657133: step: 1782/1541, loss: 0.00030898029217496514 2023-01-23 18:39:19.549148: step: 1784/1541, loss: 0.01449813786894083 2023-01-23 18:39:20.405893: step: 1786/1541, loss: 0.00134307483676821 2023-01-23 18:39:21.270941: step: 1788/1541, loss: 0.11719749122858047 2023-01-23 18:39:22.159483: step: 1790/1541, loss: 0.01603494957089424 2023-01-23 18:39:23.029188: step: 1792/1541, loss: 0.0006645936518907547 2023-01-23 18:39:23.904086: step: 1794/1541, loss: 0.0005075386143289506 2023-01-23 18:39:24.778887: step: 1796/1541, loss: 0.00026412622537463903 2023-01-23 18:39:25.617361: step: 1798/1541, loss: 2.819036672008224e-05 2023-01-23 18:39:26.594008: step: 1800/1541, loss: 0.14230139553546906 2023-01-23 18:39:27.572617: step: 1802/1541, loss: 0.3182142674922943 2023-01-23 18:39:28.463984: step: 1804/1541, loss: 0.573965847492218 2023-01-23 18:39:29.364316: step: 1806/1541, loss: 3.15704382956028e-05 2023-01-23 18:39:30.250951: step: 1808/1541, loss: 0.47859853506088257 2023-01-23 18:39:31.197401: step: 1810/1541, loss: 0.011238520964980125 2023-01-23 18:39:32.084982: step: 1812/1541, loss: 0.06531288474798203 2023-01-23 18:39:33.034224: step: 1814/1541, loss: 2.091746500809677e-05 2023-01-23 18:39:33.985610: step: 1816/1541, loss: 0.0022199181839823723 2023-01-23 18:39:34.892686: step: 1818/1541, loss: 0.0063179414719343185 2023-01-23 18:39:35.783230: step: 1820/1541, loss: 1.4186313152313232 2023-01-23 18:39:36.644482: step: 1822/1541, loss: 0.23125335574150085 2023-01-23 18:39:37.515257: step: 1824/1541, loss: 0.1187363788485527 2023-01-23 18:39:38.452362: step: 1826/1541, loss: 0.00016721889551263303 2023-01-23 18:39:39.332826: step: 1828/1541, loss: 0.17936211824417114 2023-01-23 18:39:40.238782: step: 1830/1541, loss: 0.015661535784602165 2023-01-23 18:39:41.159388: step: 1832/1541, loss: 0.09989429265260696 2023-01-23 18:39:42.055083: step: 1834/1541, loss: 1.768210768699646 2023-01-23 18:39:42.998125: step: 1836/1541, loss: 0.0049904994666576385 2023-01-23 18:39:43.907439: step: 1838/1541, loss: 1.6877424716949463 2023-01-23 18:39:44.803364: step: 1840/1541, loss: 0.12868860363960266 2023-01-23 18:39:45.700486: step: 1842/1541, loss: 0.0006284620030783117 2023-01-23 18:39:46.610389: step: 1844/1541, loss: -6.0286747611826286e-06 2023-01-23 18:39:47.459001: step: 1846/1541, loss: 0.0014100518310442567 2023-01-23 18:39:48.394442: step: 1848/1541, loss: 0.10964111238718033 2023-01-23 18:39:49.289671: step: 1850/1541, loss: 0.06373509019613266 2023-01-23 18:39:50.235644: step: 1852/1541, loss: 0.001992157893255353 2023-01-23 18:39:51.101239: step: 1854/1541, loss: 0.303913414478302 2023-01-23 18:39:52.018497: step: 1856/1541, loss: 0.0009966075886040926 2023-01-23 18:39:52.900303: step: 1858/1541, loss: 0.026331141591072083 2023-01-23 18:39:53.772421: step: 1860/1541, loss: 0.0006525017670355737 2023-01-23 18:39:54.687907: step: 1862/1541, loss: 0.0030786790885031223 2023-01-23 18:39:55.619470: step: 1864/1541, loss: 0.006762898527085781 2023-01-23 18:39:56.530119: step: 1866/1541, loss: 0.038886286318302155 2023-01-23 18:39:57.419928: step: 1868/1541, loss: 0.01607680134475231 2023-01-23 18:39:58.325334: step: 1870/1541, loss: 0.005279873497784138 2023-01-23 18:39:59.241483: step: 1872/1541, loss: 0.015382083132863045 2023-01-23 18:40:00.166343: step: 1874/1541, loss: 0.025695830583572388 2023-01-23 18:40:01.082226: step: 1876/1541, loss: 1.2027522325515747 2023-01-23 18:40:02.028084: step: 1878/1541, loss: 0.007993974722921848 2023-01-23 18:40:02.881027: step: 1880/1541, loss: 0.00024297456548083574 2023-01-23 18:40:03.814380: step: 1882/1541, loss: 0.012747135013341904 2023-01-23 18:40:04.744277: step: 1884/1541, loss: 0.01171894371509552 2023-01-23 18:40:05.679747: step: 1886/1541, loss: 0.000930746435187757 2023-01-23 18:40:06.601855: step: 1888/1541, loss: 0.001953727100044489 2023-01-23 18:40:07.559801: step: 1890/1541, loss: 0.02061721682548523 2023-01-23 18:40:08.392470: step: 1892/1541, loss: 0.0007971228915266693 2023-01-23 18:40:09.250592: step: 1894/1541, loss: 0.10909100621938705 2023-01-23 18:40:10.159391: step: 1896/1541, loss: 0.18221989274024963 2023-01-23 18:40:11.035656: step: 1898/1541, loss: 0.2544681131839752 2023-01-23 18:40:11.988803: step: 1900/1541, loss: 0.22125598788261414 2023-01-23 18:40:12.868220: step: 1902/1541, loss: 0.0001809810200938955 2023-01-23 18:40:13.838099: step: 1904/1541, loss: 0.9515969157218933 2023-01-23 18:40:14.760939: step: 1906/1541, loss: 0.05964545160531998 2023-01-23 18:40:15.705537: step: 1908/1541, loss: 0.38619354367256165 2023-01-23 18:40:16.619526: step: 1910/1541, loss: 0.0009211087017320096 2023-01-23 18:40:17.505642: step: 1912/1541, loss: 0.00036360142985358834 2023-01-23 18:40:18.385835: step: 1914/1541, loss: 0.0013594243209809065 2023-01-23 18:40:19.269448: step: 1916/1541, loss: 0.06510619819164276 2023-01-23 18:40:20.196881: step: 1918/1541, loss: 0.0004061410727445036 2023-01-23 18:40:21.063329: step: 1920/1541, loss: 0.0028772493824362755 2023-01-23 18:40:21.963503: step: 1922/1541, loss: 0.003847710555419326 2023-01-23 18:40:22.892865: step: 1924/1541, loss: 0.025114433839917183 2023-01-23 18:40:23.764599: step: 1926/1541, loss: 0.00016833608970046043 2023-01-23 18:40:24.671929: step: 1928/1541, loss: 0.004298894666135311 2023-01-23 18:40:25.546946: step: 1930/1541, loss: 0.019564606249332428 2023-01-23 18:40:26.457448: step: 1932/1541, loss: 0.022715851664543152 2023-01-23 18:40:27.367321: step: 1934/1541, loss: 0.12602561712265015 2023-01-23 18:40:28.281076: step: 1936/1541, loss: 0.00018247764091938734 2023-01-23 18:40:29.239430: step: 1938/1541, loss: 0.18341368436813354 2023-01-23 18:40:30.161106: step: 1940/1541, loss: 0.07787472009658813 2023-01-23 18:40:31.097271: step: 1942/1541, loss: 0.0005825788248330355 2023-01-23 18:40:32.004217: step: 1944/1541, loss: 0.039735715836286545 2023-01-23 18:40:32.934825: step: 1946/1541, loss: 0.00516046117991209 2023-01-23 18:40:33.788269: step: 1948/1541, loss: 0.002900778315961361 2023-01-23 18:40:34.646123: step: 1950/1541, loss: 0.0025233321357518435 2023-01-23 18:40:35.597039: step: 1952/1541, loss: 0.006924319546669722 2023-01-23 18:40:36.544034: step: 1954/1541, loss: 0.008678983896970749 2023-01-23 18:40:37.523988: step: 1956/1541, loss: 0.012837231159210205 2023-01-23 18:40:38.441978: step: 1958/1541, loss: 1.4955343008041382 2023-01-23 18:40:39.310665: step: 1960/1541, loss: 0.059775739908218384 2023-01-23 18:40:40.195749: step: 1962/1541, loss: 0.022169796749949455 2023-01-23 18:40:41.151072: step: 1964/1541, loss: 0.2685748040676117 2023-01-23 18:40:42.022459: step: 1966/1541, loss: 0.004767311736941338 2023-01-23 18:40:42.905627: step: 1968/1541, loss: 0.6908244490623474 2023-01-23 18:40:43.816619: step: 1970/1541, loss: 0.11225620657205582 2023-01-23 18:40:44.722995: step: 1972/1541, loss: 0.01033670175820589 2023-01-23 18:40:45.615311: step: 1974/1541, loss: 0.04683741182088852 2023-01-23 18:40:46.477686: step: 1976/1541, loss: 0.0006081450846977532 2023-01-23 18:40:47.475004: step: 1978/1541, loss: 0.07582683861255646 2023-01-23 18:40:48.351371: step: 1980/1541, loss: 0.0016639649402350187 2023-01-23 18:40:49.278010: step: 1982/1541, loss: 0.0008537899702787399 2023-01-23 18:40:50.159500: step: 1984/1541, loss: 0.0004954121541231871 2023-01-23 18:40:51.103668: step: 1986/1541, loss: 0.2019149512052536 2023-01-23 18:40:52.007143: step: 1988/1541, loss: 0.00025276155793108046 2023-01-23 18:40:52.888351: step: 1990/1541, loss: 0.01582612283527851 2023-01-23 18:40:53.761618: step: 1992/1541, loss: 0.000159799717948772 2023-01-23 18:40:54.622871: step: 1994/1541, loss: 0.000250504381256178 2023-01-23 18:40:55.581781: step: 1996/1541, loss: 0.00010397034930065274 2023-01-23 18:40:56.434491: step: 1998/1541, loss: 0.0018685010727494955 2023-01-23 18:40:57.345070: step: 2000/1541, loss: 0.001612493535503745 2023-01-23 18:40:58.221629: step: 2002/1541, loss: 0.0001812715781852603 2023-01-23 18:40:59.108549: step: 2004/1541, loss: 0.0015660502249374986 2023-01-23 18:40:59.980276: step: 2006/1541, loss: 0.04509972035884857 2023-01-23 18:41:00.875620: step: 2008/1541, loss: 0.0403679683804512 2023-01-23 18:41:01.789962: step: 2010/1541, loss: 0.1615227907896042 2023-01-23 18:41:02.713717: step: 2012/1541, loss: 0.013069918379187584 2023-01-23 18:41:03.674250: step: 2014/1541, loss: 0.0019874467980116606 2023-01-23 18:41:04.609519: step: 2016/1541, loss: 0.0025150359142571688 2023-01-23 18:41:05.520606: step: 2018/1541, loss: 0.5285612344741821 2023-01-23 18:41:06.380158: step: 2020/1541, loss: 0.30716633796691895 2023-01-23 18:41:07.284325: step: 2022/1541, loss: 0.25396350026130676 2023-01-23 18:41:08.169570: step: 2024/1541, loss: 0.015547706745564938 2023-01-23 18:41:09.048805: step: 2026/1541, loss: 0.0009492578683421016 2023-01-23 18:41:09.985029: step: 2028/1541, loss: 0.05530804395675659 2023-01-23 18:41:10.875997: step: 2030/1541, loss: 0.23353226482868195 2023-01-23 18:41:11.783797: step: 2032/1541, loss: 0.000652479357086122 2023-01-23 18:41:12.682460: step: 2034/1541, loss: 0.3425303101539612 2023-01-23 18:41:13.552916: step: 2036/1541, loss: 0.0744989886879921 2023-01-23 18:41:14.470574: step: 2038/1541, loss: 0.17594914138317108 2023-01-23 18:41:15.442321: step: 2040/1541, loss: 0.033685244619846344 2023-01-23 18:41:16.332572: step: 2042/1541, loss: 0.01088764425367117 2023-01-23 18:41:17.278400: step: 2044/1541, loss: 0.012679838575422764 2023-01-23 18:41:18.170455: step: 2046/1541, loss: 0.0014759695623070002 2023-01-23 18:41:19.064862: step: 2048/1541, loss: 0.010079344734549522 2023-01-23 18:41:19.990454: step: 2050/1541, loss: 0.004812918603420258 2023-01-23 18:41:20.918580: step: 2052/1541, loss: 0.0431143157184124 2023-01-23 18:41:21.787767: step: 2054/1541, loss: 0.1763744354248047 2023-01-23 18:41:22.717739: step: 2056/1541, loss: 0.0010627657175064087 2023-01-23 18:41:23.597851: step: 2058/1541, loss: 0.21359333395957947 2023-01-23 18:41:24.484092: step: 2060/1541, loss: 0.020594164729118347 2023-01-23 18:41:25.403131: step: 2062/1541, loss: 0.001721794600598514 2023-01-23 18:41:26.341723: step: 2064/1541, loss: 0.011606226675212383 2023-01-23 18:41:27.213694: step: 2066/1541, loss: 0.008776161819696426 2023-01-23 18:41:28.125043: step: 2068/1541, loss: 0.04972223564982414 2023-01-23 18:41:29.024560: step: 2070/1541, loss: 0.0034546377137303352 2023-01-23 18:41:29.913137: step: 2072/1541, loss: 0.5505915880203247 2023-01-23 18:41:30.806300: step: 2074/1541, loss: 0.015938395634293556 2023-01-23 18:41:31.696295: step: 2076/1541, loss: 0.019570330157876015 2023-01-23 18:41:32.598920: step: 2078/1541, loss: 0.002089453861117363 2023-01-23 18:41:33.507374: step: 2080/1541, loss: 0.003423162270337343 2023-01-23 18:41:34.434159: step: 2082/1541, loss: 0.054865702986717224 2023-01-23 18:41:35.312340: step: 2084/1541, loss: 0.001423809677362442 2023-01-23 18:41:36.161323: step: 2086/1541, loss: 0.005884143523871899 2023-01-23 18:41:37.066944: step: 2088/1541, loss: 0.002001652494072914 2023-01-23 18:41:37.959502: step: 2090/1541, loss: 0.000765472068451345 2023-01-23 18:41:38.827907: step: 2092/1541, loss: 0.000654963543638587 2023-01-23 18:41:39.704489: step: 2094/1541, loss: 0.00043878788710571826 2023-01-23 18:41:40.670691: step: 2096/1541, loss: 0.0023157414980232716 2023-01-23 18:41:41.574040: step: 2098/1541, loss: 0.0009685468394309282 2023-01-23 18:41:42.516073: step: 2100/1541, loss: 0.01203247345983982 2023-01-23 18:41:43.397379: step: 2102/1541, loss: 0.0003120720502920449 2023-01-23 18:41:44.318168: step: 2104/1541, loss: 0.0856909453868866 2023-01-23 18:41:45.211068: step: 2106/1541, loss: 0.029164133593440056 2023-01-23 18:41:46.110146: step: 2108/1541, loss: 4.7314704715972766e-05 2023-01-23 18:41:47.003033: step: 2110/1541, loss: 0.028159109875559807 2023-01-23 18:41:47.878751: step: 2112/1541, loss: 0.0009946419158950448 2023-01-23 18:41:48.779929: step: 2114/1541, loss: 0.0005888700834475458 2023-01-23 18:41:49.702188: step: 2116/1541, loss: 0.0028985305689275265 2023-01-23 18:41:50.587525: step: 2118/1541, loss: 0.0030620074830949306 2023-01-23 18:41:51.491131: step: 2120/1541, loss: 0.08267149329185486 2023-01-23 18:41:52.381634: step: 2122/1541, loss: 1.652145504951477 2023-01-23 18:41:53.282937: step: 2124/1541, loss: 0.2080840766429901 2023-01-23 18:41:54.202110: step: 2126/1541, loss: 0.005525955930352211 2023-01-23 18:41:55.103144: step: 2128/1541, loss: 0.04865765571594238 2023-01-23 18:41:55.988596: step: 2130/1541, loss: 0.11332307755947113 2023-01-23 18:41:56.884769: step: 2132/1541, loss: 0.006581615656614304 2023-01-23 18:41:57.809572: step: 2134/1541, loss: 6.720989767927676e-05 2023-01-23 18:41:58.676474: step: 2136/1541, loss: -6.826913249824429e-06 2023-01-23 18:41:59.571121: step: 2138/1541, loss: 0.04224443435668945 2023-01-23 18:42:00.464479: step: 2140/1541, loss: 0.05781477317214012 2023-01-23 18:42:01.338563: step: 2142/1541, loss: 0.015173731371760368 2023-01-23 18:42:02.220493: step: 2144/1541, loss: 0.0004994067130610347 2023-01-23 18:42:03.109324: step: 2146/1541, loss: 0.002940686885267496 2023-01-23 18:42:03.995308: step: 2148/1541, loss: 0.0005570980720221996 2023-01-23 18:42:04.868774: step: 2150/1541, loss: 0.0008413186296820641 2023-01-23 18:42:05.770420: step: 2152/1541, loss: 0.020029671490192413 2023-01-23 18:42:06.680301: step: 2154/1541, loss: 0.0035476537887007 2023-01-23 18:42:07.547543: step: 2156/1541, loss: 7.907791587058455e-05 2023-01-23 18:42:08.443862: step: 2158/1541, loss: 0.403458833694458 2023-01-23 18:42:09.385424: step: 2160/1541, loss: 0.005567634478211403 2023-01-23 18:42:10.284022: step: 2162/1541, loss: 0.29720479249954224 2023-01-23 18:42:11.193701: step: 2164/1541, loss: 0.046364348381757736 2023-01-23 18:42:12.084352: step: 2166/1541, loss: 0.540464460849762 2023-01-23 18:42:12.994242: step: 2168/1541, loss: 0.05348394438624382 2023-01-23 18:42:13.947409: step: 2170/1541, loss: 0.0016478076577186584 2023-01-23 18:42:14.846710: step: 2172/1541, loss: 0.10370122641324997 2023-01-23 18:42:15.723654: step: 2174/1541, loss: 0.01586941070854664 2023-01-23 18:42:16.648485: step: 2176/1541, loss: 0.00044408102985471487 2023-01-23 18:42:17.545402: step: 2178/1541, loss: 0.003032466396689415 2023-01-23 18:42:18.462424: step: 2180/1541, loss: 0.0038112415932118893 2023-01-23 18:42:19.450496: step: 2182/1541, loss: 0.00015564380737487227 2023-01-23 18:42:20.342706: step: 2184/1541, loss: 0.12247129529714584 2023-01-23 18:42:21.224899: step: 2186/1541, loss: 0.0033469446934759617 2023-01-23 18:42:22.111706: step: 2188/1541, loss: 3.821746213361621e-05 2023-01-23 18:42:23.007840: step: 2190/1541, loss: 0.002176070585846901 2023-01-23 18:42:23.935243: step: 2192/1541, loss: 0.0027545911725610495 2023-01-23 18:42:24.823527: step: 2194/1541, loss: 0.016972705721855164 2023-01-23 18:42:25.722972: step: 2196/1541, loss: 0.09282699227333069 2023-01-23 18:42:26.595053: step: 2198/1541, loss: 0.021121609956026077 2023-01-23 18:42:27.526393: step: 2200/1541, loss: 0.010517050512135029 2023-01-23 18:42:28.521973: step: 2202/1541, loss: 0.001078461529687047 2023-01-23 18:42:29.455393: step: 2204/1541, loss: 0.008945976383984089 2023-01-23 18:42:30.373116: step: 2206/1541, loss: 0.23817245662212372 2023-01-23 18:42:31.261034: step: 2208/1541, loss: 0.0009386498713865876 2023-01-23 18:42:32.155788: step: 2210/1541, loss: 0.0013460268964990973 2023-01-23 18:42:33.020276: step: 2212/1541, loss: 0.006216393318027258 2023-01-23 18:42:33.946670: step: 2214/1541, loss: 0.3665751814842224 2023-01-23 18:42:34.873293: step: 2216/1541, loss: 0.0024162193294614553 2023-01-23 18:42:35.720305: step: 2218/1541, loss: 0.00020153261721134186 2023-01-23 18:42:36.636122: step: 2220/1541, loss: 0.0009169945260509849 2023-01-23 18:42:37.514703: step: 2222/1541, loss: 0.17051757872104645 2023-01-23 18:42:38.384770: step: 2224/1541, loss: 0.0019110538996756077 2023-01-23 18:42:39.276935: step: 2226/1541, loss: 0.00038135109934955835 2023-01-23 18:42:40.176047: step: 2228/1541, loss: 0.033929672092199326 2023-01-23 18:42:41.070467: step: 2230/1541, loss: 0.0017560611013323069 2023-01-23 18:42:41.946557: step: 2232/1541, loss: 0.3356791138648987 2023-01-23 18:42:42.852242: step: 2234/1541, loss: 0.06829545646905899 2023-01-23 18:42:43.732038: step: 2236/1541, loss: 0.00243758806027472 2023-01-23 18:42:44.613746: step: 2238/1541, loss: 0.4504743218421936 2023-01-23 18:42:45.497744: step: 2240/1541, loss: 0.02411394566297531 2023-01-23 18:42:46.368363: step: 2242/1541, loss: 0.0045373160392045975 2023-01-23 18:42:47.237036: step: 2244/1541, loss: 0.0035470975562930107 2023-01-23 18:42:48.097287: step: 2246/1541, loss: 0.00045012839836999774 2023-01-23 18:42:49.139850: step: 2248/1541, loss: 0.022615980356931686 2023-01-23 18:42:50.045335: step: 2250/1541, loss: 0.2883734405040741 2023-01-23 18:42:50.927269: step: 2252/1541, loss: 0.22267675399780273 2023-01-23 18:42:51.815786: step: 2254/1541, loss: 0.08861605823040009 2023-01-23 18:42:52.703819: step: 2256/1541, loss: 0.0008332077995873988 2023-01-23 18:42:53.646170: step: 2258/1541, loss: 0.20414769649505615 2023-01-23 18:42:54.553123: step: 2260/1541, loss: 0.0003042453317902982 2023-01-23 18:42:55.459544: step: 2262/1541, loss: 0.021786199882626534 2023-01-23 18:42:56.365588: step: 2264/1541, loss: 0.004797571338713169 2023-01-23 18:42:57.259953: step: 2266/1541, loss: 0.15233823657035828 2023-01-23 18:42:58.169142: step: 2268/1541, loss: 0.00086463603656739 2023-01-23 18:42:59.097342: step: 2270/1541, loss: 0.002195556415244937 2023-01-23 18:42:59.934681: step: 2272/1541, loss: 0.02928842045366764 2023-01-23 18:43:00.813873: step: 2274/1541, loss: 0.0033885142765939236 2023-01-23 18:43:01.698197: step: 2276/1541, loss: 0.39314183592796326 2023-01-23 18:43:02.590780: step: 2278/1541, loss: 0.008556719869375229 2023-01-23 18:43:03.512608: step: 2280/1541, loss: 0.0021627340465784073 2023-01-23 18:43:04.460827: step: 2282/1541, loss: 0.13333597779273987 2023-01-23 18:43:05.299917: step: 2284/1541, loss: 0.0010163129772990942 2023-01-23 18:43:06.249823: step: 2286/1541, loss: 0.0007208820898085833 2023-01-23 18:43:07.229905: step: 2288/1541, loss: 0.19128409028053284 2023-01-23 18:43:08.182520: step: 2290/1541, loss: 0.01003981288522482 2023-01-23 18:43:09.168154: step: 2292/1541, loss: 0.35505279898643494 2023-01-23 18:43:10.051078: step: 2294/1541, loss: 0.005931749474257231 2023-01-23 18:43:10.942019: step: 2296/1541, loss: 0.0011605829931795597 2023-01-23 18:43:11.856360: step: 2298/1541, loss: 0.18309524655342102 2023-01-23 18:43:12.768499: step: 2300/1541, loss: 0.008084697648882866 2023-01-23 18:43:13.659439: step: 2302/1541, loss: 0.31734877824783325 2023-01-23 18:43:14.651921: step: 2304/1541, loss: 0.05070757493376732 2023-01-23 18:43:15.588998: step: 2306/1541, loss: 0.05885964632034302 2023-01-23 18:43:16.492808: step: 2308/1541, loss: 0.0030400254763662815 2023-01-23 18:43:17.373757: step: 2310/1541, loss: 0.00037807979970239103 2023-01-23 18:43:18.251712: step: 2312/1541, loss: 0.01915048435330391 2023-01-23 18:43:19.133428: step: 2314/1541, loss: 0.00865861400961876 2023-01-23 18:43:20.040620: step: 2316/1541, loss: 0.0021019475534558296 2023-01-23 18:43:20.914764: step: 2318/1541, loss: 0.001576457405462861 2023-01-23 18:43:21.843904: step: 2320/1541, loss: 0.0008246887591667473 2023-01-23 18:43:22.732757: step: 2322/1541, loss: 0.024322381243109703 2023-01-23 18:43:23.653166: step: 2324/1541, loss: 0.0003631889703683555 2023-01-23 18:43:24.576476: step: 2326/1541, loss: 0.8388743996620178 2023-01-23 18:43:25.482157: step: 2328/1541, loss: 0.002284785732626915 2023-01-23 18:43:26.425535: step: 2330/1541, loss: 0.0007562446407973766 2023-01-23 18:43:27.299354: step: 2332/1541, loss: 0.0012847500620409846 2023-01-23 18:43:28.181417: step: 2334/1541, loss: 0.0017784088850021362 2023-01-23 18:43:29.076749: step: 2336/1541, loss: 0.0011510852491483092 2023-01-23 18:43:30.023336: step: 2338/1541, loss: 0.03148895129561424 2023-01-23 18:43:30.930459: step: 2340/1541, loss: 0.012898627668619156 2023-01-23 18:43:31.884848: step: 2342/1541, loss: 0.0017320207552984357 2023-01-23 18:43:32.766640: step: 2344/1541, loss: 0.0003734538331627846 2023-01-23 18:43:33.642307: step: 2346/1541, loss: 1.2503098249435425 2023-01-23 18:43:34.572908: step: 2348/1541, loss: 4.281541259842925e-05 2023-01-23 18:43:35.436740: step: 2350/1541, loss: 0.11214423179626465 2023-01-23 18:43:36.341700: step: 2352/1541, loss: 0.05536068230867386 2023-01-23 18:43:37.213148: step: 2354/1541, loss: 0.01246988121420145 2023-01-23 18:43:38.116733: step: 2356/1541, loss: 0.0007133105536922812 2023-01-23 18:43:39.071298: step: 2358/1541, loss: 0.02025093138217926 2023-01-23 18:43:40.037214: step: 2360/1541, loss: 0.25003254413604736 2023-01-23 18:43:40.908709: step: 2362/1541, loss: 1.6600866729277186e-05 2023-01-23 18:43:41.864205: step: 2364/1541, loss: 0.09289440512657166 2023-01-23 18:43:42.779700: step: 2366/1541, loss: 0.00020815448078792542 2023-01-23 18:43:43.676876: step: 2368/1541, loss: 1.1311230659484863 2023-01-23 18:43:44.537168: step: 2370/1541, loss: 0.0005627265782095492 2023-01-23 18:43:45.393627: step: 2372/1541, loss: 0.0003238217905163765 2023-01-23 18:43:46.272808: step: 2374/1541, loss: 0.001587921055033803 2023-01-23 18:43:47.180848: step: 2376/1541, loss: 0.00011584975436562672 2023-01-23 18:43:48.068365: step: 2378/1541, loss: 0.0015378714306280017 2023-01-23 18:43:48.957810: step: 2380/1541, loss: 0.00015898339916020632 2023-01-23 18:43:49.855244: step: 2382/1541, loss: 0.3558677136898041 2023-01-23 18:43:50.754924: step: 2384/1541, loss: 0.00011393729801056907 2023-01-23 18:43:51.648613: step: 2386/1541, loss: 0.03067122958600521 2023-01-23 18:43:52.521668: step: 2388/1541, loss: 0.00224513141438365 2023-01-23 18:43:53.412939: step: 2390/1541, loss: 0.007942378520965576 2023-01-23 18:43:54.416692: step: 2392/1541, loss: 0.0491512306034565 2023-01-23 18:43:55.295844: step: 2394/1541, loss: 0.001547279884107411 2023-01-23 18:43:56.245937: step: 2396/1541, loss: 0.333822101354599 2023-01-23 18:43:57.111060: step: 2398/1541, loss: 0.16794176399707794 2023-01-23 18:43:58.032417: step: 2400/1541, loss: 0.13338354229927063 2023-01-23 18:43:58.904177: step: 2402/1541, loss: 1.9341525330673903e-05 2023-01-23 18:43:59.792948: step: 2404/1541, loss: 0.174157053232193 2023-01-23 18:44:00.688004: step: 2406/1541, loss: 0.0010336649138480425 2023-01-23 18:44:01.550001: step: 2408/1541, loss: 0.00597273837774992 2023-01-23 18:44:02.410708: step: 2410/1541, loss: 0.00013826304348185658 2023-01-23 18:44:03.306398: step: 2412/1541, loss: 0.0005087839672341943 2023-01-23 18:44:04.199027: step: 2414/1541, loss: 0.0011178413406014442 2023-01-23 18:44:05.092233: step: 2416/1541, loss: 0.00011553614604054019 2023-01-23 18:44:05.978552: step: 2418/1541, loss: 0.3180798888206482 2023-01-23 18:44:06.905053: step: 2420/1541, loss: 0.0009315373608842492 2023-01-23 18:44:07.808854: step: 2422/1541, loss: 0.009591151028871536 2023-01-23 18:44:08.691161: step: 2424/1541, loss: 0.0008962826686911285 2023-01-23 18:44:09.569742: step: 2426/1541, loss: 0.02592441253364086 2023-01-23 18:44:10.437340: step: 2428/1541, loss: 0.00018016657850239426 2023-01-23 18:44:11.340934: step: 2430/1541, loss: 0.07345977425575256 2023-01-23 18:44:12.253954: step: 2432/1541, loss: 0.009300383739173412 2023-01-23 18:44:13.142989: step: 2434/1541, loss: 0.01498451828956604 2023-01-23 18:44:14.190686: step: 2436/1541, loss: 0.032409485429525375 2023-01-23 18:44:15.097405: step: 2438/1541, loss: 0.000698840944096446 2023-01-23 18:44:16.078617: step: 2440/1541, loss: 0.002022397704422474 2023-01-23 18:44:16.953958: step: 2442/1541, loss: 0.00012257398338988423 2023-01-23 18:44:17.856818: step: 2444/1541, loss: 0.0024463930167257786 2023-01-23 18:44:18.744038: step: 2446/1541, loss: 0.000902986153960228 2023-01-23 18:44:19.713804: step: 2448/1541, loss: 1.143438696861267 2023-01-23 18:44:20.566024: step: 2450/1541, loss: 0.00019407649233471602 2023-01-23 18:44:21.440076: step: 2452/1541, loss: 0.007306785322725773 2023-01-23 18:44:22.386026: step: 2454/1541, loss: 0.005487162619829178 2023-01-23 18:44:23.244684: step: 2456/1541, loss: 0.7964544892311096 2023-01-23 18:44:24.145675: step: 2458/1541, loss: 0.03449976444244385 2023-01-23 18:44:25.048894: step: 2460/1541, loss: 0.0035480225924402475 2023-01-23 18:44:25.917391: step: 2462/1541, loss: 0.00010327538620913401 2023-01-23 18:44:26.847902: step: 2464/1541, loss: 0.17807941138744354 2023-01-23 18:44:27.739021: step: 2466/1541, loss: 0.12190910428762436 2023-01-23 18:44:28.635568: step: 2468/1541, loss: 0.0028830126393586397 2023-01-23 18:44:29.527768: step: 2470/1541, loss: 0.002401102799922228 2023-01-23 18:44:30.446922: step: 2472/1541, loss: 0.018185200169682503 2023-01-23 18:44:31.330715: step: 2474/1541, loss: 0.00016480230260640383 2023-01-23 18:44:32.271129: step: 2476/1541, loss: 0.2866147458553314 2023-01-23 18:44:33.193933: step: 2478/1541, loss: 0.013473445549607277 2023-01-23 18:44:34.070693: step: 2480/1541, loss: 0.013091444969177246 2023-01-23 18:44:35.009035: step: 2482/1541, loss: 0.24931193888187408 2023-01-23 18:44:35.929854: step: 2484/1541, loss: 0.5517493486404419 2023-01-23 18:44:36.862068: step: 2486/1541, loss: 0.19009383022785187 2023-01-23 18:44:37.787793: step: 2488/1541, loss: 0.00020597146067302674 2023-01-23 18:44:38.691571: step: 2490/1541, loss: 0.0024190691765397787 2023-01-23 18:44:39.606099: step: 2492/1541, loss: 0.00022226168948691338 2023-01-23 18:44:40.502940: step: 2494/1541, loss: 0.012124106287956238 2023-01-23 18:44:41.413465: step: 2496/1541, loss: 0.03502662852406502 2023-01-23 18:44:42.314843: step: 2498/1541, loss: 0.0005728319520130754 2023-01-23 18:44:43.281536: step: 2500/1541, loss: 0.30157843232154846 2023-01-23 18:44:44.228924: step: 2502/1541, loss: 0.08167483657598495 2023-01-23 18:44:45.134830: step: 2504/1541, loss: 0.08109927922487259 2023-01-23 18:44:46.042754: step: 2506/1541, loss: 0.00022415585408452898 2023-01-23 18:44:46.896146: step: 2508/1541, loss: 0.0001794601121218875 2023-01-23 18:44:47.781276: step: 2510/1541, loss: 0.01512281782925129 2023-01-23 18:44:48.668748: step: 2512/1541, loss: 2.0296852588653564 2023-01-23 18:44:49.595067: step: 2514/1541, loss: 0.17851172387599945 2023-01-23 18:44:50.512594: step: 2516/1541, loss: 0.03333611041307449 2023-01-23 18:44:51.431474: step: 2518/1541, loss: 0.0020155012607574463 2023-01-23 18:44:52.325852: step: 2520/1541, loss: 0.0008221916505135596 2023-01-23 18:44:53.184292: step: 2522/1541, loss: 0.00023602653527632356 2023-01-23 18:44:54.076226: step: 2524/1541, loss: 0.004085224121809006 2023-01-23 18:44:55.008250: step: 2526/1541, loss: 0.008728012442588806 2023-01-23 18:44:55.959143: step: 2528/1541, loss: 0.015323418192565441 2023-01-23 18:44:56.848770: step: 2530/1541, loss: 0.14800913631916046 2023-01-23 18:44:57.740445: step: 2532/1541, loss: 0.010997431352734566 2023-01-23 18:44:58.605545: step: 2534/1541, loss: 0.001093863626010716 2023-01-23 18:44:59.482285: step: 2536/1541, loss: 0.09049130231142044 2023-01-23 18:45:00.379981: step: 2538/1541, loss: 0.0011324136285111308 2023-01-23 18:45:01.256116: step: 2540/1541, loss: 0.0007794578559696674 2023-01-23 18:45:02.160660: step: 2542/1541, loss: 0.009840180166065693 2023-01-23 18:45:02.991145: step: 2544/1541, loss: 0.0023970408365130424 2023-01-23 18:45:03.885534: step: 2546/1541, loss: 0.012410886585712433 2023-01-23 18:45:04.745209: step: 2548/1541, loss: 0.0005972948856651783 2023-01-23 18:45:05.629935: step: 2550/1541, loss: 0.0030352144967764616 2023-01-23 18:45:06.583024: step: 2552/1541, loss: 0.0019382500322535634 2023-01-23 18:45:07.482096: step: 2554/1541, loss: 0.0005362785886973143 2023-01-23 18:45:08.374571: step: 2556/1541, loss: 9.148120443569496e-05 2023-01-23 18:45:09.262614: step: 2558/1541, loss: 0.05310780555009842 2023-01-23 18:45:10.150275: step: 2560/1541, loss: 0.05501827597618103 2023-01-23 18:45:10.999286: step: 2562/1541, loss: 0.11031679809093475 2023-01-23 18:45:11.869764: step: 2564/1541, loss: 0.0012279573129490018 2023-01-23 18:45:12.823165: step: 2566/1541, loss: 0.00051295873709023 2023-01-23 18:45:13.790213: step: 2568/1541, loss: 0.00014064909191802144 2023-01-23 18:45:14.683018: step: 2570/1541, loss: 0.00045362376840785146 2023-01-23 18:45:15.577601: step: 2572/1541, loss: 5.3381110774353147e-05 2023-01-23 18:45:16.479396: step: 2574/1541, loss: 0.0416024774312973 2023-01-23 18:45:17.362137: step: 2576/1541, loss: 0.013106529600918293 2023-01-23 18:45:18.256228: step: 2578/1541, loss: 0.00035762556944973767 2023-01-23 18:45:19.122588: step: 2580/1541, loss: 0.011722679249942303 2023-01-23 18:45:20.098853: step: 2582/1541, loss: 0.09010802209377289 2023-01-23 18:45:21.001863: step: 2584/1541, loss: 0.0006287544965744019 2023-01-23 18:45:21.902152: step: 2586/1541, loss: 0.03215113654732704 2023-01-23 18:45:22.775615: step: 2588/1541, loss: 0.00029719845042563975 2023-01-23 18:45:23.656871: step: 2590/1541, loss: 0.027633998543024063 2023-01-23 18:45:24.570772: step: 2592/1541, loss: 0.027241995558142662 2023-01-23 18:45:25.493236: step: 2594/1541, loss: 0.00038702503661625087 2023-01-23 18:45:26.406471: step: 2596/1541, loss: 4.534599793259986e-05 2023-01-23 18:45:27.291021: step: 2598/1541, loss: 0.0007097414345480502 2023-01-23 18:45:28.152535: step: 2600/1541, loss: 0.0054937005043029785 2023-01-23 18:45:29.087182: step: 2602/1541, loss: 0.0005098019610159099 2023-01-23 18:45:29.966670: step: 2604/1541, loss: 0.0002035850629908964 2023-01-23 18:45:30.837480: step: 2606/1541, loss: 0.0795074999332428 2023-01-23 18:45:31.743108: step: 2608/1541, loss: 0.036724500358104706 2023-01-23 18:45:32.681851: step: 2610/1541, loss: 0.001963951624929905 2023-01-23 18:45:33.548102: step: 2612/1541, loss: 7.671346247661859e-05 2023-01-23 18:45:34.478766: step: 2614/1541, loss: 0.3960722088813782 2023-01-23 18:45:35.379440: step: 2616/1541, loss: 0.008636055514216423 2023-01-23 18:45:36.299957: step: 2618/1541, loss: 0.0771595910191536 2023-01-23 18:45:37.181009: step: 2620/1541, loss: 0.0004398995661176741 2023-01-23 18:45:38.093848: step: 2622/1541, loss: 0.13798914849758148 2023-01-23 18:45:38.975711: step: 2624/1541, loss: 0.015579065307974815 2023-01-23 18:45:39.827314: step: 2626/1541, loss: 0.0683143213391304 2023-01-23 18:45:40.704638: step: 2628/1541, loss: 0.012515440583229065 2023-01-23 18:45:41.551503: step: 2630/1541, loss: 0.0003059477312490344 2023-01-23 18:45:42.431814: step: 2632/1541, loss: 0.0018623464275151491 2023-01-23 18:45:43.340407: step: 2634/1541, loss: 0.007303132209926844 2023-01-23 18:45:44.217979: step: 2636/1541, loss: 0.1399877369403839 2023-01-23 18:45:45.132862: step: 2638/1541, loss: 0.0001064361713360995 2023-01-23 18:45:46.038934: step: 2640/1541, loss: 0.007412930019199848 2023-01-23 18:45:46.981810: step: 2642/1541, loss: 0.04903985932469368 2023-01-23 18:45:47.887955: step: 2644/1541, loss: 0.08165153115987778 2023-01-23 18:45:48.794497: step: 2646/1541, loss: 0.5534743666648865 2023-01-23 18:45:49.684302: step: 2648/1541, loss: 0.11540576815605164 2023-01-23 18:45:50.546211: step: 2650/1541, loss: 2.3620564937591553 2023-01-23 18:45:51.423298: step: 2652/1541, loss: 0.10825894773006439 2023-01-23 18:45:52.340800: step: 2654/1541, loss: 0.1040622740983963 2023-01-23 18:45:53.296500: step: 2656/1541, loss: 0.18661685287952423 2023-01-23 18:45:54.208015: step: 2658/1541, loss: 0.023895470425486565 2023-01-23 18:45:55.072743: step: 2660/1541, loss: 0.16866211593151093 2023-01-23 18:45:56.015015: step: 2662/1541, loss: 0.0001091568119591102 2023-01-23 18:45:56.971449: step: 2664/1541, loss: 0.022684214636683464 2023-01-23 18:45:57.852813: step: 2666/1541, loss: 0.02835877053439617 2023-01-23 18:45:58.722325: step: 2668/1541, loss: 0.0032414935994893312 2023-01-23 18:45:59.608374: step: 2670/1541, loss: 0.0003554006980266422 2023-01-23 18:46:00.476849: step: 2672/1541, loss: 0.22258001565933228 2023-01-23 18:46:01.352202: step: 2674/1541, loss: 0.0007537662168033421 2023-01-23 18:46:02.232173: step: 2676/1541, loss: 0.08419134467840195 2023-01-23 18:46:03.106416: step: 2678/1541, loss: 0.0005041506374254823 2023-01-23 18:46:04.007991: step: 2680/1541, loss: 0.10101281106472015 2023-01-23 18:46:04.893705: step: 2682/1541, loss: 0.005532308015972376 2023-01-23 18:46:05.850791: step: 2684/1541, loss: 0.015792537480592728 2023-01-23 18:46:06.801263: step: 2686/1541, loss: 0.13336454331874847 2023-01-23 18:46:07.673419: step: 2688/1541, loss: 0.026323705911636353 2023-01-23 18:46:08.539519: step: 2690/1541, loss: 0.04147502779960632 2023-01-23 18:46:09.410247: step: 2692/1541, loss: 0.01232953928411007 2023-01-23 18:46:10.237183: step: 2694/1541, loss: 0.00032980149262584746 2023-01-23 18:46:11.122457: step: 2696/1541, loss: 0.001208562171086669 2023-01-23 18:46:12.030422: step: 2698/1541, loss: 0.0033045834861695766 2023-01-23 18:46:12.944035: step: 2700/1541, loss: 0.9786041975021362 2023-01-23 18:46:13.835125: step: 2702/1541, loss: 0.00020012928871437907 2023-01-23 18:46:14.793169: step: 2704/1541, loss: 0.003499214770272374 2023-01-23 18:46:15.691087: step: 2706/1541, loss: 0.005048117134720087 2023-01-23 18:46:16.597123: step: 2708/1541, loss: 0.0017009183065965772 2023-01-23 18:46:17.482073: step: 2710/1541, loss: 0.0001970250013982877 2023-01-23 18:46:18.426339: step: 2712/1541, loss: 0.0006300851819105446 2023-01-23 18:46:19.321628: step: 2714/1541, loss: 0.06360874325037003 2023-01-23 18:46:20.266355: step: 2716/1541, loss: 0.0007716134423390031 2023-01-23 18:46:21.134208: step: 2718/1541, loss: 0.01458597183227539 2023-01-23 18:46:22.032161: step: 2720/1541, loss: 0.0031736523378640413 2023-01-23 18:46:22.900105: step: 2722/1541, loss: 0.001827104832045734 2023-01-23 18:46:23.808228: step: 2724/1541, loss: 0.06876889616250992 2023-01-23 18:46:24.673083: step: 2726/1541, loss: 0.003207452828064561 2023-01-23 18:46:25.602415: step: 2728/1541, loss: 0.018808843567967415 2023-01-23 18:46:26.575328: step: 2730/1541, loss: 0.0004250715719535947 2023-01-23 18:46:27.442110: step: 2732/1541, loss: 0.002437750343233347 2023-01-23 18:46:28.303645: step: 2734/1541, loss: 0.01293137762695551 2023-01-23 18:46:29.196231: step: 2736/1541, loss: 0.008630642667412758 2023-01-23 18:46:30.112469: step: 2738/1541, loss: 0.0014430193696171045 2023-01-23 18:46:31.047543: step: 2740/1541, loss: 0.00042453716741874814 2023-01-23 18:46:31.936969: step: 2742/1541, loss: 0.0013060381170362234 2023-01-23 18:46:32.861411: step: 2744/1541, loss: 0.06357290595769882 2023-01-23 18:46:33.794968: step: 2746/1541, loss: 0.006375435274094343 2023-01-23 18:46:34.679278: step: 2748/1541, loss: 0.006558528635650873 2023-01-23 18:46:35.549484: step: 2750/1541, loss: 0.0002828687720466405 2023-01-23 18:46:36.520227: step: 2752/1541, loss: 0.0004246959579177201 2023-01-23 18:46:37.455313: step: 2754/1541, loss: 0.06448114663362503 2023-01-23 18:46:38.347864: step: 2756/1541, loss: 0.009121283888816833 2023-01-23 18:46:39.331101: step: 2758/1541, loss: 5.732601130148396e-05 2023-01-23 18:46:40.221569: step: 2760/1541, loss: 0.0001318034337600693 2023-01-23 18:46:41.140249: step: 2762/1541, loss: 0.1646546721458435 2023-01-23 18:46:42.013162: step: 2764/1541, loss: 0.00033863727003335953 2023-01-23 18:46:42.879828: step: 2766/1541, loss: 0.0022070128470659256 2023-01-23 18:46:43.800615: step: 2768/1541, loss: 0.002594785066321492 2023-01-23 18:46:44.682408: step: 2770/1541, loss: 0.15852825343608856 2023-01-23 18:46:45.576080: step: 2772/1541, loss: 0.003376528387889266 2023-01-23 18:46:46.493805: step: 2774/1541, loss: 0.01524427905678749 2023-01-23 18:46:47.363486: step: 2776/1541, loss: 0.016597965732216835 2023-01-23 18:46:48.283286: step: 2778/1541, loss: 0.0868055447936058 2023-01-23 18:46:49.131899: step: 2780/1541, loss: 0.06699592620134354 2023-01-23 18:46:49.991203: step: 2782/1541, loss: 9.308289736509323e-05 2023-01-23 18:46:50.865301: step: 2784/1541, loss: 0.05790579319000244 2023-01-23 18:46:51.776573: step: 2786/1541, loss: 0.0011901423567906022 2023-01-23 18:46:52.731171: step: 2788/1541, loss: 0.0835874155163765 2023-01-23 18:46:53.627103: step: 2790/1541, loss: 0.004327512811869383 2023-01-23 18:46:54.493139: step: 2792/1541, loss: 0.05238733068108559 2023-01-23 18:46:55.414312: step: 2794/1541, loss: 0.0003569778928067535 2023-01-23 18:46:56.278589: step: 2796/1541, loss: 0.051326651126146317 2023-01-23 18:46:57.176280: step: 2798/1541, loss: 0.025861332193017006 2023-01-23 18:46:58.120996: step: 2800/1541, loss: 0.0002703338977880776 2023-01-23 18:46:59.023754: step: 2802/1541, loss: 0.00020050549937877804 2023-01-23 18:46:59.963145: step: 2804/1541, loss: 0.0005054573994129896 2023-01-23 18:47:00.882618: step: 2806/1541, loss: 0.0030530730728060007 2023-01-23 18:47:01.759382: step: 2808/1541, loss: 0.04852290451526642 2023-01-23 18:47:02.630837: step: 2810/1541, loss: 0.014503386802971363 2023-01-23 18:47:03.512726: step: 2812/1541, loss: 0.007107025943696499 2023-01-23 18:47:04.430342: step: 2814/1541, loss: 0.044005729258060455 2023-01-23 18:47:05.316780: step: 2816/1541, loss: 0.00024314410984516144 2023-01-23 18:47:06.214737: step: 2818/1541, loss: 0.0032125147990882397 2023-01-23 18:47:07.123640: step: 2820/1541, loss: 0.008756080642342567 2023-01-23 18:47:08.053132: step: 2822/1541, loss: 0.07650432735681534 2023-01-23 18:47:09.049807: step: 2824/1541, loss: 0.14133897423744202 2023-01-23 18:47:09.886304: step: 2826/1541, loss: 0.001666462980210781 2023-01-23 18:47:10.806400: step: 2828/1541, loss: 0.21182818710803986 2023-01-23 18:47:11.725701: step: 2830/1541, loss: 0.020454756915569305 2023-01-23 18:47:12.622875: step: 2832/1541, loss: 0.18836148083209991 2023-01-23 18:47:13.492126: step: 2834/1541, loss: 0.0008316341554746032 2023-01-23 18:47:14.388699: step: 2836/1541, loss: 0.07402729988098145 2023-01-23 18:47:15.279070: step: 2838/1541, loss: 0.0003516337019391358 2023-01-23 18:47:16.255229: step: 2840/1541, loss: 0.0075052049942314625 2023-01-23 18:47:17.169592: step: 2842/1541, loss: 0.3413507342338562 2023-01-23 18:47:18.102662: step: 2844/1541, loss: 0.0007980908849276602 2023-01-23 18:47:19.023800: step: 2846/1541, loss: 0.011123431846499443 2023-01-23 18:47:19.893401: step: 2848/1541, loss: 4.48264618171379e-05 2023-01-23 18:47:20.800246: step: 2850/1541, loss: 0.03970661759376526 2023-01-23 18:47:21.725595: step: 2852/1541, loss: 0.07751744240522385 2023-01-23 18:47:22.678917: step: 2854/1541, loss: 0.06289806216955185 2023-01-23 18:47:23.560955: step: 2856/1541, loss: 0.009440446272492409 2023-01-23 18:47:24.456746: step: 2858/1541, loss: 0.32557764649391174 2023-01-23 18:47:25.393997: step: 2860/1541, loss: 0.01978342793881893 2023-01-23 18:47:26.276983: step: 2862/1541, loss: 0.03672075271606445 2023-01-23 18:47:27.139813: step: 2864/1541, loss: 0.00016517285257577896 2023-01-23 18:47:28.004126: step: 2866/1541, loss: 0.38590025901794434 2023-01-23 18:47:28.834086: step: 2868/1541, loss: 0.0064093852415680885 2023-01-23 18:47:29.730518: step: 2870/1541, loss: 0.006018784828484058 2023-01-23 18:47:30.634297: step: 2872/1541, loss: 0.0005690797697752714 2023-01-23 18:47:31.522131: step: 2874/1541, loss: 0.001767846755683422 2023-01-23 18:47:32.401041: step: 2876/1541, loss: 0.004437893163412809 2023-01-23 18:47:33.325680: step: 2878/1541, loss: 0.51776123046875 2023-01-23 18:47:34.203191: step: 2880/1541, loss: 0.06326649338006973 2023-01-23 18:47:35.109756: step: 2882/1541, loss: 0.06727585941553116 2023-01-23 18:47:36.107892: step: 2884/1541, loss: 0.24848853051662445 2023-01-23 18:47:37.021773: step: 2886/1541, loss: 0.009245540015399456 2023-01-23 18:47:37.866724: step: 2888/1541, loss: 0.003966739401221275 2023-01-23 18:47:38.765467: step: 2890/1541, loss: 0.03761664777994156 2023-01-23 18:47:39.641984: step: 2892/1541, loss: 0.0010035816812887788 2023-01-23 18:47:40.525682: step: 2894/1541, loss: 0.003593317698687315 2023-01-23 18:47:41.392364: step: 2896/1541, loss: 0.11323904991149902 2023-01-23 18:47:42.370702: step: 2898/1541, loss: 0.0018801117548719049 2023-01-23 18:47:43.243611: step: 2900/1541, loss: 0.1873859465122223 2023-01-23 18:47:44.123424: step: 2902/1541, loss: 0.060458485037088394 2023-01-23 18:47:44.997822: step: 2904/1541, loss: 0.13279442489147186 2023-01-23 18:47:45.925711: step: 2906/1541, loss: 0.06257655471563339 2023-01-23 18:47:46.810468: step: 2908/1541, loss: 0.000545745249837637 2023-01-23 18:47:47.658820: step: 2910/1541, loss: 0.01797696016728878 2023-01-23 18:47:48.598943: step: 2912/1541, loss: 0.002962243976071477 2023-01-23 18:47:49.481811: step: 2914/1541, loss: 0.09922024607658386 2023-01-23 18:47:50.394422: step: 2916/1541, loss: 0.0147313941270113 2023-01-23 18:47:51.300389: step: 2918/1541, loss: 0.0008649885421618819 2023-01-23 18:47:52.187618: step: 2920/1541, loss: 0.02928191050887108 2023-01-23 18:47:53.092762: step: 2922/1541, loss: 0.0005275685689412057 2023-01-23 18:47:53.977306: step: 2924/1541, loss: 0.06944654881954193 2023-01-23 18:47:54.890272: step: 2926/1541, loss: 0.012097851373255253 2023-01-23 18:47:55.800445: step: 2928/1541, loss: 0.14528842270374298 2023-01-23 18:47:56.659060: step: 2930/1541, loss: 0.7632072567939758 2023-01-23 18:47:57.540239: step: 2932/1541, loss: 0.00413904944434762 2023-01-23 18:47:58.429174: step: 2934/1541, loss: 0.001068162266165018 2023-01-23 18:47:59.295315: step: 2936/1541, loss: 0.020106447860598564 2023-01-23 18:48:00.176757: step: 2938/1541, loss: 8.866078860592097e-05 2023-01-23 18:48:01.053233: step: 2940/1541, loss: 0.00021676940377801657 2023-01-23 18:48:02.037576: step: 2942/1541, loss: 0.02917640469968319 2023-01-23 18:48:02.953620: step: 2944/1541, loss: 0.15838177502155304 2023-01-23 18:48:03.819334: step: 2946/1541, loss: 0.015755226835608482 2023-01-23 18:48:04.758199: step: 2948/1541, loss: 0.012351973913609982 2023-01-23 18:48:05.662136: step: 2950/1541, loss: 0.006404614541679621 2023-01-23 18:48:06.527657: step: 2952/1541, loss: 1.3148431777954102 2023-01-23 18:48:07.437360: step: 2954/1541, loss: 0.0036018239334225655 2023-01-23 18:48:08.359739: step: 2956/1541, loss: 0.005789400078356266 2023-01-23 18:48:09.222744: step: 2958/1541, loss: 0.5001032948493958 2023-01-23 18:48:10.131300: step: 2960/1541, loss: 0.9889851808547974 2023-01-23 18:48:11.053817: step: 2962/1541, loss: 0.08535674959421158 2023-01-23 18:48:11.934400: step: 2964/1541, loss: 6.580613990081474e-05 2023-01-23 18:48:12.822473: step: 2966/1541, loss: 0.06985270977020264 2023-01-23 18:48:13.738258: step: 2968/1541, loss: 0.036196544766426086 2023-01-23 18:48:14.663269: step: 2970/1541, loss: 0.0010713513474911451 2023-01-23 18:48:15.568916: step: 2972/1541, loss: 0.0001639525726204738 2023-01-23 18:48:16.491356: step: 2974/1541, loss: 1.2029184103012085 2023-01-23 18:48:17.379474: step: 2976/1541, loss: 0.0023778195027261972 2023-01-23 18:48:18.277081: step: 2978/1541, loss: 0.0005149706266820431 2023-01-23 18:48:19.151882: step: 2980/1541, loss: 0.04661066085100174 2023-01-23 18:48:20.047371: step: 2982/1541, loss: 0.15552355349063873 2023-01-23 18:48:20.895555: step: 2984/1541, loss: 0.0026587825268507004 2023-01-23 18:48:21.821116: step: 2986/1541, loss: 0.0005935265799053013 2023-01-23 18:48:22.681555: step: 2988/1541, loss: 0.0005962229915894568 2023-01-23 18:48:23.556289: step: 2990/1541, loss: 0.009319383651018143 2023-01-23 18:48:24.411859: step: 2992/1541, loss: 9.385309385834262e-05 2023-01-23 18:48:25.294164: step: 2994/1541, loss: 0.0021019235718995333 2023-01-23 18:48:26.190223: step: 2996/1541, loss: 0.12561453878879547 2023-01-23 18:48:27.169437: step: 2998/1541, loss: 0.015017683617770672 2023-01-23 18:48:28.049049: step: 3000/1541, loss: 0.00012986203364562243 2023-01-23 18:48:28.911001: step: 3002/1541, loss: 0.0011095762019976974 2023-01-23 18:48:29.785344: step: 3004/1541, loss: 0.027644094079732895 2023-01-23 18:48:30.695423: step: 3006/1541, loss: 0.08452960103750229 2023-01-23 18:48:31.553880: step: 3008/1541, loss: 0.015932995826005936 2023-01-23 18:48:32.431316: step: 3010/1541, loss: 0.0034579038619995117 2023-01-23 18:48:33.346698: step: 3012/1541, loss: 0.08208220452070236 2023-01-23 18:48:34.195792: step: 3014/1541, loss: 0.018798135221004486 2023-01-23 18:48:35.070766: step: 3016/1541, loss: 0.002724897349253297 2023-01-23 18:48:35.973067: step: 3018/1541, loss: 0.012997993268072605 2023-01-23 18:48:36.891571: step: 3020/1541, loss: 0.00895716529339552 2023-01-23 18:48:37.824553: step: 3022/1541, loss: 0.14131282269954681 2023-01-23 18:48:38.728339: step: 3024/1541, loss: 0.0007189692696556449 2023-01-23 18:48:39.616332: step: 3026/1541, loss: 0.0002628892834763974 2023-01-23 18:48:40.539138: step: 3028/1541, loss: 0.000591571384575218 2023-01-23 18:48:41.423268: step: 3030/1541, loss: 0.03685275837779045 2023-01-23 18:48:42.352916: step: 3032/1541, loss: 0.00048603530740365386 2023-01-23 18:48:43.254857: step: 3034/1541, loss: 0.013291112147271633 2023-01-23 18:48:44.138350: step: 3036/1541, loss: 0.05706046149134636 2023-01-23 18:48:45.105934: step: 3038/1541, loss: 0.0007941719959490001 2023-01-23 18:48:46.077485: step: 3040/1541, loss: 0.05952935665845871 2023-01-23 18:48:47.011070: step: 3042/1541, loss: 0.29194706678390503 2023-01-23 18:48:47.956011: step: 3044/1541, loss: 0.031098591163754463 2023-01-23 18:48:48.820810: step: 3046/1541, loss: 0.08250036835670471 2023-01-23 18:48:49.696670: step: 3048/1541, loss: 0.0002394342445768416 2023-01-23 18:48:50.566409: step: 3050/1541, loss: 0.003695129882544279 2023-01-23 18:48:51.430280: step: 3052/1541, loss: 0.08553216606378555 2023-01-23 18:48:52.400143: step: 3054/1541, loss: 0.1356109231710434 2023-01-23 18:48:53.305623: step: 3056/1541, loss: 0.0008052756893448532 2023-01-23 18:48:54.218091: step: 3058/1541, loss: 0.0049927146174013615 2023-01-23 18:48:55.130393: step: 3060/1541, loss: 0.003712486708536744 2023-01-23 18:48:56.032860: step: 3062/1541, loss: 0.0024234720040112734 2023-01-23 18:48:56.942200: step: 3064/1541, loss: 0.006859096698462963 2023-01-23 18:48:57.805773: step: 3066/1541, loss: 0.00025342695880681276 2023-01-23 18:48:58.706182: step: 3068/1541, loss: 0.008405294269323349 2023-01-23 18:48:59.610119: step: 3070/1541, loss: 0.12695260345935822 2023-01-23 18:49:00.497536: step: 3072/1541, loss: 0.0005857809446752071 2023-01-23 18:49:01.417872: step: 3074/1541, loss: 0.09544482827186584 2023-01-23 18:49:02.283620: step: 3076/1541, loss: 0.0020904175471514463 2023-01-23 18:49:03.180001: step: 3078/1541, loss: 0.002852879697456956 2023-01-23 18:49:04.074584: step: 3080/1541, loss: 0.01873381994664669 2023-01-23 18:49:04.955036: step: 3082/1541, loss: 0.0888841450214386 2023-01-23 18:49:05.525800: step: 3084/1541, loss: 2.6121655537281185e-05 ================================================== Loss: 0.075 -------------------- Dev Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Dev Korean: {'event': {'p': 0.5676470588235294, 'r': 0.7709720372836218, 'f1': 0.6538678712591757}, 'argument': {'p': 0.6799844555377508, 'r': 0.6614349872603391, 'f1': 0.670581468161319}, 'combined': 0.43847167709249435, 'epoch': 28} Test Korean: {'event': {'p': 0.6101048617731173, 'r': 0.7669263031755542, 'f1': 0.6795858773559862}, 'argument': {'p': 0.7167363642351611, 'r': 0.7044135284991881, 'f1': 0.7105215205176734}, 'combined': 0.4828603909013124, 'epoch': 28} Dev Russian: {'event': {'p': 0.6528436018957346, 'r': 0.7336884154460719, 'f1': 0.6909090909090909}, 'argument': {'p': 0.6710021009827256, 'r': 0.6598777255331251, 'f1': 0.665393420807338}, 'combined': 0.45972636346688805, 'epoch': 28} Test Russian: {'event': {'p': 0.6514913657770801, 'r': 0.7459556620730977, 'f1': 0.6955307262569833}, 'argument': {'p': 0.7113260130880612, 'r': 0.6962728442314631, 'f1': 0.7037189377260537}, 'combined': 0.4894581438373949, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6119194591416814, 'r': 0.4236365486365487, 'f1': 0.5006613756613757}, 'combined': 0.29172395117277006, 'epoch': 28} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.561250654107797, 'r': 0.6395646988670244, 'f1': 0.5978539576365662}, 'combined': 0.2903862079949036, 'epoch': 28} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6636893013672245, 'r': 0.6465372447633335, 'f1': 0.6550010050584172}, 'combined': 0.4319825939628635, 'epoch': 14} Test for Korean: {'event': {'p': 0.6143946615824595, 'r': 0.7723187537447573, 'f1': 0.6843642155561453}, 'argument': {'p': 0.7065465973991794, 'r': 0.7144881116656963, 'f1': 0.7104951637835268}, 'combined': 0.48623746541914836, 'epoch': 14} Korean: {'event': {'p': 0.59375, 'r': 0.6031746031746031, 'f1': 0.5984251968503936}, 'argument': {'p': 0.6660879629629629, 'r': 0.42068713450292394, 'f1': 0.5156810035842294}, 'combined': 0.30859650608190103, 'epoch': 14} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 29 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 18:54:32.859098: step: 2/1541, loss: 0.03430541977286339 2023-01-23 18:54:33.776809: step: 4/1541, loss: 0.0002497680834494531 2023-01-23 18:54:34.667071: step: 6/1541, loss: 0.09760662168264389 2023-01-23 18:54:35.605472: step: 8/1541, loss: 0.0002610604860819876 2023-01-23 18:54:36.499773: step: 10/1541, loss: 0.0023885397240519524 2023-01-23 18:54:37.367606: step: 12/1541, loss: 5.919231261941604e-05 2023-01-23 18:54:38.256550: step: 14/1541, loss: 0.007572878152132034 2023-01-23 18:54:39.121144: step: 16/1541, loss: 0.001633936888538301 2023-01-23 18:54:40.015846: step: 18/1541, loss: 0.00020074719213880599 2023-01-23 18:54:40.918142: step: 20/1541, loss: 0.01344518642872572 2023-01-23 18:54:41.817979: step: 22/1541, loss: 0.004065827000886202 2023-01-23 18:54:42.686964: step: 24/1541, loss: 0.00034559948835521936 2023-01-23 18:54:43.657252: step: 26/1541, loss: 0.026239639148116112 2023-01-23 18:54:44.622506: step: 28/1541, loss: 8.303866343339905e-05 2023-01-23 18:54:45.554305: step: 30/1541, loss: 0.007678572088479996 2023-01-23 18:54:46.439339: step: 32/1541, loss: 0.00040870357770472765 2023-01-23 18:54:47.370173: step: 34/1541, loss: 0.04444056749343872 2023-01-23 18:54:48.254473: step: 36/1541, loss: 0.011112436652183533 2023-01-23 18:54:49.106517: step: 38/1541, loss: 0.00017029207083396614 2023-01-23 18:54:50.023224: step: 40/1541, loss: 0.00011322160571580753 2023-01-23 18:54:50.867221: step: 42/1541, loss: 0.0013405662029981613 2023-01-23 18:54:51.767434: step: 44/1541, loss: 0.005584117025136948 2023-01-23 18:54:52.628290: step: 46/1541, loss: 0.00284490454941988 2023-01-23 18:54:53.547216: step: 48/1541, loss: 0.007243636529892683 2023-01-23 18:54:54.408995: step: 50/1541, loss: 0.14808228611946106 2023-01-23 18:54:55.313116: step: 52/1541, loss: 0.5089493989944458 2023-01-23 18:54:56.218854: step: 54/1541, loss: 0.033143844455480576 2023-01-23 18:54:57.072530: step: 56/1541, loss: 0.005512006115168333 2023-01-23 18:54:57.979080: step: 58/1541, loss: 0.0005652106483466923 2023-01-23 18:54:58.928645: step: 60/1541, loss: 0.599121630191803 2023-01-23 18:54:59.779749: step: 62/1541, loss: 0.00023073615739122033 2023-01-23 18:55:00.658014: step: 64/1541, loss: 0.00034845879417844117 2023-01-23 18:55:01.536251: step: 66/1541, loss: 0.07956936955451965 2023-01-23 18:55:02.452705: step: 68/1541, loss: 0.02618575468659401 2023-01-23 18:55:03.337135: step: 70/1541, loss: 0.0007329515647143126 2023-01-23 18:55:04.221315: step: 72/1541, loss: 7.272881339304149e-05 2023-01-23 18:55:05.128427: step: 74/1541, loss: 0.0001280156138818711 2023-01-23 18:55:05.990890: step: 76/1541, loss: 0.03272828087210655 2023-01-23 18:55:06.894226: step: 78/1541, loss: 0.000697153271175921 2023-01-23 18:55:07.805000: step: 80/1541, loss: 0.0013139225775375962 2023-01-23 18:55:08.787813: step: 82/1541, loss: 0.0008275241125375032 2023-01-23 18:55:09.645397: step: 84/1541, loss: 0.0016756481491029263 2023-01-23 18:55:10.542922: step: 86/1541, loss: 0.00012894475366920233 2023-01-23 18:55:11.427679: step: 88/1541, loss: 0.0005806207191199064 2023-01-23 18:55:12.335978: step: 90/1541, loss: 0.20353682339191437 2023-01-23 18:55:13.314027: step: 92/1541, loss: 0.36949706077575684 2023-01-23 18:55:14.191682: step: 94/1541, loss: 0.057884056121110916 2023-01-23 18:55:15.107359: step: 96/1541, loss: 1.821218729019165 2023-01-23 18:55:16.012246: step: 98/1541, loss: 0.08767200261354446 2023-01-23 18:55:16.899234: step: 100/1541, loss: 0.0010382828768342733 2023-01-23 18:55:17.769325: step: 102/1541, loss: -9.914238944475073e-06 2023-01-23 18:55:18.690903: step: 104/1541, loss: 0.0003483613836579025 2023-01-23 18:55:19.559239: step: 106/1541, loss: 0.004373227246105671 2023-01-23 18:55:20.425739: step: 108/1541, loss: 0.003352327970787883 2023-01-23 18:55:21.372895: step: 110/1541, loss: 0.00013917595788370818 2023-01-23 18:55:22.291570: step: 112/1541, loss: 0.6010262966156006 2023-01-23 18:55:23.223548: step: 114/1541, loss: 0.03308961167931557 2023-01-23 18:55:24.153675: step: 116/1541, loss: 0.009427985176444054 2023-01-23 18:55:25.042093: step: 118/1541, loss: 0.13981041312217712 2023-01-23 18:55:25.951557: step: 120/1541, loss: 0.0002309402625542134 2023-01-23 18:55:26.817216: step: 122/1541, loss: 0.1035512313246727 2023-01-23 18:55:27.706264: step: 124/1541, loss: 0.00038200250128284097 2023-01-23 18:55:28.587220: step: 126/1541, loss: 0.0018866650061681867 2023-01-23 18:55:29.477936: step: 128/1541, loss: 0.0028922418132424355 2023-01-23 18:55:30.372314: step: 130/1541, loss: 0.008345631882548332 2023-01-23 18:55:31.246762: step: 132/1541, loss: 0.12525801360607147 2023-01-23 18:55:32.156477: step: 134/1541, loss: 0.005535878706723452 2023-01-23 18:55:33.046675: step: 136/1541, loss: 0.007039661519229412 2023-01-23 18:55:33.926573: step: 138/1541, loss: 0.0005722962087020278 2023-01-23 18:55:34.822119: step: 140/1541, loss: 0.014578628353774548 2023-01-23 18:55:35.753253: step: 142/1541, loss: 2.921854019165039 2023-01-23 18:55:36.665400: step: 144/1541, loss: 0.0018997283186763525 2023-01-23 18:55:37.546550: step: 146/1541, loss: 0.19178909063339233 2023-01-23 18:55:38.409253: step: 148/1541, loss: 0.07581336051225662 2023-01-23 18:55:39.288797: step: 150/1541, loss: 0.027187293395400047 2023-01-23 18:55:40.215661: step: 152/1541, loss: 0.005573634058237076 2023-01-23 18:55:41.159610: step: 154/1541, loss: 0.08407159894704819 2023-01-23 18:55:42.038994: step: 156/1541, loss: 0.0002724120859056711 2023-01-23 18:55:42.893451: step: 158/1541, loss: 0.0024164491333067417 2023-01-23 18:55:43.844545: step: 160/1541, loss: -1.5380985132651404e-05 2023-01-23 18:55:44.754745: step: 162/1541, loss: 0.00013686937745660543 2023-01-23 18:55:45.652904: step: 164/1541, loss: 0.02501458115875721 2023-01-23 18:55:46.575559: step: 166/1541, loss: 0.0008567710756324232 2023-01-23 18:55:47.467530: step: 168/1541, loss: 0.019506029784679413 2023-01-23 18:55:48.377511: step: 170/1541, loss: 0.06858809292316437 2023-01-23 18:55:49.289057: step: 172/1541, loss: 0.005453822202980518 2023-01-23 18:55:50.128434: step: 174/1541, loss: 0.0660586804151535 2023-01-23 18:55:51.064427: step: 176/1541, loss: 5.090230843052268e-05 2023-01-23 18:55:52.010467: step: 178/1541, loss: 0.00903302151709795 2023-01-23 18:55:52.941058: step: 180/1541, loss: 0.00016205347492359579 2023-01-23 18:55:53.812450: step: 182/1541, loss: 0.010184597223997116 2023-01-23 18:55:54.742851: step: 184/1541, loss: 0.09215205907821655 2023-01-23 18:55:55.641340: step: 186/1541, loss: 0.0006729663582518697 2023-01-23 18:55:56.524683: step: 188/1541, loss: 0.00014591212675441056 2023-01-23 18:55:57.425902: step: 190/1541, loss: 0.07856816798448563 2023-01-23 18:55:58.381632: step: 192/1541, loss: 0.013730880804359913 2023-01-23 18:55:59.256485: step: 194/1541, loss: 0.0020864380057901144 2023-01-23 18:56:00.273577: step: 196/1541, loss: 0.00036680797347798944 2023-01-23 18:56:01.141231: step: 198/1541, loss: 0.1175796315073967 2023-01-23 18:56:02.049545: step: 200/1541, loss: 0.42544275522232056 2023-01-23 18:56:02.961886: step: 202/1541, loss: 0.006349688861519098 2023-01-23 18:56:03.860303: step: 204/1541, loss: 0.008478387258946896 2023-01-23 18:56:04.812609: step: 206/1541, loss: 0.00032054228358902037 2023-01-23 18:56:05.667223: step: 208/1541, loss: 0.0003436529659666121 2023-01-23 18:56:06.564966: step: 210/1541, loss: 0.007475683931261301 2023-01-23 18:56:07.551642: step: 212/1541, loss: 0.5823183655738831 2023-01-23 18:56:08.451749: step: 214/1541, loss: 0.12738186120986938 2023-01-23 18:56:09.386186: step: 216/1541, loss: 4.626917507266626e-05 2023-01-23 18:56:10.357972: step: 218/1541, loss: -1.157124188466696e-05 2023-01-23 18:56:11.284188: step: 220/1541, loss: 0.0009338072268292308 2023-01-23 18:56:12.247202: step: 222/1541, loss: 0.004578731954097748 2023-01-23 18:56:13.163762: step: 224/1541, loss: 1.1455416824901477e-05 2023-01-23 18:56:14.039596: step: 226/1541, loss: 0.3712612986564636 2023-01-23 18:56:14.975175: step: 228/1541, loss: 0.005644103512167931 2023-01-23 18:56:15.866580: step: 230/1541, loss: 0.0029082847759127617 2023-01-23 18:56:16.724551: step: 232/1541, loss: 2.071291055472102e-05 2023-01-23 18:56:17.688182: step: 234/1541, loss: 0.17444171011447906 2023-01-23 18:56:18.596100: step: 236/1541, loss: 0.0005610614898614585 2023-01-23 18:56:19.442880: step: 238/1541, loss: 0.0009805120062083006 2023-01-23 18:56:20.333430: step: 240/1541, loss: 0.0018991093384101987 2023-01-23 18:56:21.220460: step: 242/1541, loss: 9.044651233125478e-05 2023-01-23 18:56:22.127027: step: 244/1541, loss: 0.021579263731837273 2023-01-23 18:56:23.039500: step: 246/1541, loss: 0.0014492286136373878 2023-01-23 18:56:23.939062: step: 248/1541, loss: 0.2807866036891937 2023-01-23 18:56:24.836310: step: 250/1541, loss: 4.947911293129437e-05 2023-01-23 18:56:25.662857: step: 252/1541, loss: 7.113975880201906e-05 2023-01-23 18:56:26.554292: step: 254/1541, loss: 0.0002489733451511711 2023-01-23 18:56:27.441887: step: 256/1541, loss: 0.06745066493749619 2023-01-23 18:56:28.430265: step: 258/1541, loss: 0.026166357100009918 2023-01-23 18:56:29.334452: step: 260/1541, loss: 0.0006368551985360682 2023-01-23 18:56:30.280000: step: 262/1541, loss: 0.008980436250567436 2023-01-23 18:56:31.208496: step: 264/1541, loss: 2.7676492209138814e-06 2023-01-23 18:56:32.095572: step: 266/1541, loss: 0.3610060214996338 2023-01-23 18:56:32.947247: step: 268/1541, loss: 0.0002256531734019518 2023-01-23 18:56:33.852349: step: 270/1541, loss: 0.016915570944547653 2023-01-23 18:56:34.815968: step: 272/1541, loss: 0.0003842556325253099 2023-01-23 18:56:35.656051: step: 274/1541, loss: 0.0020343950018286705 2023-01-23 18:56:36.500271: step: 276/1541, loss: 4.687246837420389e-05 2023-01-23 18:56:37.360149: step: 278/1541, loss: 3.8802612834842876e-05 2023-01-23 18:56:38.203030: step: 280/1541, loss: 0.00011546904715942219 2023-01-23 18:56:39.054029: step: 282/1541, loss: 0.0002577518462203443 2023-01-23 18:56:39.929787: step: 284/1541, loss: 0.0001769612281350419 2023-01-23 18:56:40.805559: step: 286/1541, loss: 0.00011225114576518536 2023-01-23 18:56:41.676354: step: 288/1541, loss: 0.009783076122403145 2023-01-23 18:56:42.598693: step: 290/1541, loss: 0.000981436693109572 2023-01-23 18:56:43.480108: step: 292/1541, loss: 0.08913194388151169 2023-01-23 18:56:44.362954: step: 294/1541, loss: 8.985100430436432e-05 2023-01-23 18:56:45.233194: step: 296/1541, loss: 0.0006440355791710317 2023-01-23 18:56:46.124147: step: 298/1541, loss: 0.0003459212020970881 2023-01-23 18:56:47.042917: step: 300/1541, loss: 0.01920977607369423 2023-01-23 18:56:47.915100: step: 302/1541, loss: 2.218600273132324 2023-01-23 18:56:48.777783: step: 304/1541, loss: 0.0005411154706962407 2023-01-23 18:56:49.686436: step: 306/1541, loss: -3.692227619467303e-05 2023-01-23 18:56:50.568062: step: 308/1541, loss: 0.00018735775665845722 2023-01-23 18:56:51.457762: step: 310/1541, loss: 0.0567275770008564 2023-01-23 18:56:52.329919: step: 312/1541, loss: 0.0015650374116376042 2023-01-23 18:56:53.233579: step: 314/1541, loss: 1.8610109691508114e-05 2023-01-23 18:56:54.079124: step: 316/1541, loss: 0.0002380623627686873 2023-01-23 18:56:55.055165: step: 318/1541, loss: 0.002815135521814227 2023-01-23 18:56:56.017836: step: 320/1541, loss: 0.01677607372403145 2023-01-23 18:56:56.917966: step: 322/1541, loss: 0.002198661444708705 2023-01-23 18:56:57.791176: step: 324/1541, loss: 0.0001083549068425782 2023-01-23 18:56:58.669766: step: 326/1541, loss: 0.37153348326683044 2023-01-23 18:56:59.612674: step: 328/1541, loss: 0.003737854538485408 2023-01-23 18:57:00.497154: step: 330/1541, loss: 0.19280600547790527 2023-01-23 18:57:01.359382: step: 332/1541, loss: 0.00015699495270382613 2023-01-23 18:57:02.235773: step: 334/1541, loss: 0.0011077297385782003 2023-01-23 18:57:03.119132: step: 336/1541, loss: 0.006180705968290567 2023-01-23 18:57:04.007303: step: 338/1541, loss: 0.0008147311746142805 2023-01-23 18:57:04.886386: step: 340/1541, loss: 0.09851937741041183 2023-01-23 18:57:05.767335: step: 342/1541, loss: 0.0010580419329926372 2023-01-23 18:57:06.622014: step: 344/1541, loss: 0.0107237808406353 2023-01-23 18:57:07.535553: step: 346/1541, loss: 0.0008284790674224496 2023-01-23 18:57:08.417159: step: 348/1541, loss: 0.00022832768445368856 2023-01-23 18:57:09.309247: step: 350/1541, loss: 0.1048545390367508 2023-01-23 18:57:10.221777: step: 352/1541, loss: 0.8050105571746826 2023-01-23 18:57:11.134263: step: 354/1541, loss: 0.0017663771286606789 2023-01-23 18:57:11.999168: step: 356/1541, loss: 0.08152199536561966 2023-01-23 18:57:12.880672: step: 358/1541, loss: 0.4593560993671417 2023-01-23 18:57:13.773500: step: 360/1541, loss: 0.007635404821485281 2023-01-23 18:57:14.617294: step: 362/1541, loss: 0.000970542139839381 2023-01-23 18:57:15.492816: step: 364/1541, loss: 0.0007230225601233542 2023-01-23 18:57:16.321151: step: 366/1541, loss: 0.05733858048915863 2023-01-23 18:57:17.183654: step: 368/1541, loss: 0.004593852441757917 2023-01-23 18:57:18.030036: step: 370/1541, loss: 0.02390325628221035 2023-01-23 18:57:18.950456: step: 372/1541, loss: 0.23988482356071472 2023-01-23 18:57:19.828597: step: 374/1541, loss: 0.035379763692617416 2023-01-23 18:57:20.698158: step: 376/1541, loss: 0.003567315870895982 2023-01-23 18:57:21.602128: step: 378/1541, loss: 0.01745901256799698 2023-01-23 18:57:22.480569: step: 380/1541, loss: 0.024688508361577988 2023-01-23 18:57:23.355550: step: 382/1541, loss: 0.0009395063971169293 2023-01-23 18:57:24.301579: step: 384/1541, loss: 0.41861218214035034 2023-01-23 18:57:25.249204: step: 386/1541, loss: 0.001551494700834155 2023-01-23 18:57:26.162206: step: 388/1541, loss: 0.008951143361628056 2023-01-23 18:57:27.081395: step: 390/1541, loss: 0.015019112266600132 2023-01-23 18:57:27.994112: step: 392/1541, loss: 0.06162261217832565 2023-01-23 18:57:28.914111: step: 394/1541, loss: 0.2325950264930725 2023-01-23 18:57:29.857999: step: 396/1541, loss: 0.30540135502815247 2023-01-23 18:57:30.731378: step: 398/1541, loss: 0.05003951117396355 2023-01-23 18:57:31.629317: step: 400/1541, loss: 0.006656430661678314 2023-01-23 18:57:32.574850: step: 402/1541, loss: 0.0018333983607590199 2023-01-23 18:57:33.464703: step: 404/1541, loss: 0.30600208044052124 2023-01-23 18:57:34.466188: step: 406/1541, loss: 0.08881815522909164 2023-01-23 18:57:35.323404: step: 408/1541, loss: 0.0004467998805921525 2023-01-23 18:57:36.233224: step: 410/1541, loss: 0.0006690867012366652 2023-01-23 18:57:37.116165: step: 412/1541, loss: 0.6016623973846436 2023-01-23 18:57:38.002928: step: 414/1541, loss: 0.10777319222688675 2023-01-23 18:57:38.961501: step: 416/1541, loss: 0.00031512725399807096 2023-01-23 18:57:39.861651: step: 418/1541, loss: 0.0036179167218506336 2023-01-23 18:57:40.787636: step: 420/1541, loss: 0.002054712036624551 2023-01-23 18:57:41.678620: step: 422/1541, loss: 0.006889587268233299 2023-01-23 18:57:42.503376: step: 424/1541, loss: 0.0002286244125571102 2023-01-23 18:57:43.362461: step: 426/1541, loss: 0.0052908118814229965 2023-01-23 18:57:44.247984: step: 428/1541, loss: 0.0270712748169899 2023-01-23 18:57:45.099392: step: 430/1541, loss: 0.0014419830404222012 2023-01-23 18:57:45.968736: step: 432/1541, loss: 0.026976319029927254 2023-01-23 18:57:46.872772: step: 434/1541, loss: 0.06475614756345749 2023-01-23 18:57:47.741904: step: 436/1541, loss: 5.2898823923897e-05 2023-01-23 18:57:48.600276: step: 438/1541, loss: 0.0009692495805211365 2023-01-23 18:57:49.585063: step: 440/1541, loss: 0.01255822740495205 2023-01-23 18:57:50.506511: step: 442/1541, loss: 0.05716044828295708 2023-01-23 18:57:51.408266: step: 444/1541, loss: -3.7210618756944314e-05 2023-01-23 18:57:52.281662: step: 446/1541, loss: 0.08466432243585587 2023-01-23 18:57:53.164164: step: 448/1541, loss: 0.05477479100227356 2023-01-23 18:57:54.026872: step: 450/1541, loss: 0.022454233840107918 2023-01-23 18:57:54.944187: step: 452/1541, loss: 0.0013740649446845055 2023-01-23 18:57:55.807575: step: 454/1541, loss: 3.770262992475182e-06 2023-01-23 18:57:56.644192: step: 456/1541, loss: 0.002941169310361147 2023-01-23 18:57:57.597616: step: 458/1541, loss: 0.0002947404282167554 2023-01-23 18:57:58.461903: step: 460/1541, loss: 0.00016601757670287043 2023-01-23 18:57:59.372305: step: 462/1541, loss: 0.00016220391262322664 2023-01-23 18:58:00.227118: step: 464/1541, loss: 0.07509937882423401 2023-01-23 18:58:01.089080: step: 466/1541, loss: 0.014126627705991268 2023-01-23 18:58:02.007507: step: 468/1541, loss: 0.12705160677433014 2023-01-23 18:58:02.845992: step: 470/1541, loss: 0.001156275044195354 2023-01-23 18:58:03.764784: step: 472/1541, loss: 0.14438267052173615 2023-01-23 18:58:04.727846: step: 474/1541, loss: 0.0003227002453058958 2023-01-23 18:58:05.638856: step: 476/1541, loss: 0.19267292320728302 2023-01-23 18:58:06.552322: step: 478/1541, loss: 0.008886006660759449 2023-01-23 18:58:07.384838: step: 480/1541, loss: 0.002096208045259118 2023-01-23 18:58:08.250116: step: 482/1541, loss: 0.012163723818957806 2023-01-23 18:58:09.172534: step: 484/1541, loss: 0.01915130764245987 2023-01-23 18:58:10.065259: step: 486/1541, loss: 8.24436210677959e-05 2023-01-23 18:58:10.959770: step: 488/1541, loss: 0.0004073774325661361 2023-01-23 18:58:11.840113: step: 490/1541, loss: 0.816019594669342 2023-01-23 18:58:12.690381: step: 492/1541, loss: 0.00012990905088372529 2023-01-23 18:58:13.607321: step: 494/1541, loss: 0.022501317784190178 2023-01-23 18:58:14.501982: step: 496/1541, loss: 0.03180040791630745 2023-01-23 18:58:15.363186: step: 498/1541, loss: 0.344331294298172 2023-01-23 18:58:16.263088: step: 500/1541, loss: 0.01381499320268631 2023-01-23 18:58:17.171388: step: 502/1541, loss: 0.133025661110878 2023-01-23 18:58:18.096792: step: 504/1541, loss: 0.010393341071903706 2023-01-23 18:58:19.053011: step: 506/1541, loss: 0.00026222103042528033 2023-01-23 18:58:19.969375: step: 508/1541, loss: 9.819823389989324e-06 2023-01-23 18:58:20.849123: step: 510/1541, loss: 0.018479105085134506 2023-01-23 18:58:21.731193: step: 512/1541, loss: 0.08663828670978546 2023-01-23 18:58:22.682718: step: 514/1541, loss: 0.00032379821641370654 2023-01-23 18:58:23.589306: step: 516/1541, loss: 0.0005623911274597049 2023-01-23 18:58:24.483890: step: 518/1541, loss: 4.2339124775025994e-05 2023-01-23 18:58:25.381193: step: 520/1541, loss: 0.0011232312535867095 2023-01-23 18:58:26.293039: step: 522/1541, loss: 0.0006369317998178303 2023-01-23 18:58:27.147638: step: 524/1541, loss: 0.12075009942054749 2023-01-23 18:58:28.030142: step: 526/1541, loss: 0.13524992763996124 2023-01-23 18:58:28.948911: step: 528/1541, loss: 0.7443218231201172 2023-01-23 18:58:29.908657: step: 530/1541, loss: 0.019418003037571907 2023-01-23 18:58:30.776093: step: 532/1541, loss: 0.057424597442150116 2023-01-23 18:58:31.656346: step: 534/1541, loss: 0.08427924662828445 2023-01-23 18:58:32.600782: step: 536/1541, loss: 0.0019728343468159437 2023-01-23 18:58:33.467438: step: 538/1541, loss: 0.3247078061103821 2023-01-23 18:58:34.363333: step: 540/1541, loss: 0.00025557083426974714 2023-01-23 18:58:35.259693: step: 542/1541, loss: 0.03003927506506443 2023-01-23 18:58:36.130176: step: 544/1541, loss: 0.00011306758096907288 2023-01-23 18:58:37.020526: step: 546/1541, loss: 0.0008139132405631244 2023-01-23 18:58:37.921792: step: 548/1541, loss: 8.353397424798459e-05 2023-01-23 18:58:38.875091: step: 550/1541, loss: 0.00986032746732235 2023-01-23 18:58:39.802042: step: 552/1541, loss: 0.024652812629938126 2023-01-23 18:58:40.724919: step: 554/1541, loss: 0.058995649218559265 2023-01-23 18:58:41.607555: step: 556/1541, loss: 0.1761719137430191 2023-01-23 18:58:42.468373: step: 558/1541, loss: 0.0037690012250095606 2023-01-23 18:58:43.375238: step: 560/1541, loss: 0.013804574497044086 2023-01-23 18:58:44.275033: step: 562/1541, loss: 0.013752298429608345 2023-01-23 18:58:45.203871: step: 564/1541, loss: 0.017656249925494194 2023-01-23 18:58:46.068452: step: 566/1541, loss: 0.0014892804902046919 2023-01-23 18:58:46.928009: step: 568/1541, loss: 0.00026615889510139823 2023-01-23 18:58:47.872919: step: 570/1541, loss: 0.005807006265968084 2023-01-23 18:58:48.781811: step: 572/1541, loss: 0.00023204028548207134 2023-01-23 18:58:49.670717: step: 574/1541, loss: 0.07230090349912643 2023-01-23 18:58:50.558233: step: 576/1541, loss: 0.06534271687269211 2023-01-23 18:58:51.522040: step: 578/1541, loss: 0.0027875732630491257 2023-01-23 18:58:52.436510: step: 580/1541, loss: 0.09318862855434418 2023-01-23 18:58:53.367692: step: 582/1541, loss: 0.0005182891618460417 2023-01-23 18:58:54.320948: step: 584/1541, loss: 0.00011576511315070093 2023-01-23 18:58:55.189805: step: 586/1541, loss: 8.239857561420649e-05 2023-01-23 18:58:56.041436: step: 588/1541, loss: 0.04800815135240555 2023-01-23 18:58:56.939562: step: 590/1541, loss: 0.03875810652971268 2023-01-23 18:58:57.853927: step: 592/1541, loss: 0.0005919131217524409 2023-01-23 18:58:58.734313: step: 594/1541, loss: 0.0011408522259443998 2023-01-23 18:58:59.651110: step: 596/1541, loss: 0.0004317587590776384 2023-01-23 18:59:00.533174: step: 598/1541, loss: 0.0005585463950410485 2023-01-23 18:59:01.451884: step: 600/1541, loss: 0.0030526292975991964 2023-01-23 18:59:02.378596: step: 602/1541, loss: 6.183353889355203e-06 2023-01-23 18:59:03.308242: step: 604/1541, loss: 0.0040661939419806 2023-01-23 18:59:04.201135: step: 606/1541, loss: 3.872752859024331e-05 2023-01-23 18:59:05.100000: step: 608/1541, loss: 0.00018567290680948645 2023-01-23 18:59:05.981029: step: 610/1541, loss: 0.0005095868837088346 2023-01-23 18:59:06.850473: step: 612/1541, loss: 0.004210461396723986 2023-01-23 18:59:07.741342: step: 614/1541, loss: 0.006784325465559959 2023-01-23 18:59:08.626210: step: 616/1541, loss: 0.06303336471319199 2023-01-23 18:59:09.574006: step: 618/1541, loss: 0.00020229758229106665 2023-01-23 18:59:10.547885: step: 620/1541, loss: 0.0014441122766584158 2023-01-23 18:59:11.440012: step: 622/1541, loss: 0.00814199447631836 2023-01-23 18:59:12.346672: step: 624/1541, loss: 0.0001094947729143314 2023-01-23 18:59:13.216144: step: 626/1541, loss: 0.001222840161062777 2023-01-23 18:59:14.096368: step: 628/1541, loss: 0.13129353523254395 2023-01-23 18:59:14.974232: step: 630/1541, loss: 0.0003319161478430033 2023-01-23 18:59:15.879616: step: 632/1541, loss: 0.00026487873401492834 2023-01-23 18:59:16.811465: step: 634/1541, loss: 4.958866338711232e-05 2023-01-23 18:59:17.696184: step: 636/1541, loss: 0.0015685880789533257 2023-01-23 18:59:18.622669: step: 638/1541, loss: 0.00033562275348231196 2023-01-23 18:59:19.468868: step: 640/1541, loss: 0.003861065488308668 2023-01-23 18:59:20.345113: step: 642/1541, loss: 0.12240895628929138 2023-01-23 18:59:21.245887: step: 644/1541, loss: 0.3637479245662689 2023-01-23 18:59:22.237139: step: 646/1541, loss: 0.0019489862024784088 2023-01-23 18:59:23.092893: step: 648/1541, loss: 6.872911762911826e-05 2023-01-23 18:59:23.975877: step: 650/1541, loss: 0.00047596171498298645 2023-01-23 18:59:24.884418: step: 652/1541, loss: 7.655532681383193e-05 2023-01-23 18:59:25.792723: step: 654/1541, loss: 5.0497437769081444e-05 2023-01-23 18:59:26.674089: step: 656/1541, loss: 0.26284632086753845 2023-01-23 18:59:27.575410: step: 658/1541, loss: 0.00025611737510189414 2023-01-23 18:59:28.462472: step: 660/1541, loss: 0.004659299273043871 2023-01-23 18:59:29.429883: step: 662/1541, loss: 0.2782260775566101 2023-01-23 18:59:30.351455: step: 664/1541, loss: 0.049527011811733246 2023-01-23 18:59:31.236353: step: 666/1541, loss: 0.0011038149241358042 2023-01-23 18:59:32.146318: step: 668/1541, loss: 0.20321470499038696 2023-01-23 18:59:33.077692: step: 670/1541, loss: 0.005940483883023262 2023-01-23 18:59:34.022281: step: 672/1541, loss: 0.0001227632019435987 2023-01-23 18:59:34.879537: step: 674/1541, loss: 0.000841516419313848 2023-01-23 18:59:35.790548: step: 676/1541, loss: 0.00012994706048630178 2023-01-23 18:59:36.701337: step: 678/1541, loss: 0.08840079605579376 2023-01-23 18:59:37.589571: step: 680/1541, loss: 0.00013486020907294005 2023-01-23 18:59:38.451032: step: 682/1541, loss: 0.00018332088075112551 2023-01-23 18:59:39.393265: step: 684/1541, loss: 0.15523764491081238 2023-01-23 18:59:40.304362: step: 686/1541, loss: 0.2689770460128784 2023-01-23 18:59:41.188041: step: 688/1541, loss: 0.00018513765826355666 2023-01-23 18:59:42.054156: step: 690/1541, loss: 0.0006829577032476664 2023-01-23 18:59:42.911292: step: 692/1541, loss: 0.18192262947559357 2023-01-23 18:59:43.760229: step: 694/1541, loss: 0.0014568695332854986 2023-01-23 18:59:44.700709: step: 696/1541, loss: 0.08436349779367447 2023-01-23 18:59:45.602391: step: 698/1541, loss: 0.0002926735323853791 2023-01-23 18:59:46.445312: step: 700/1541, loss: 0.005444007460027933 2023-01-23 18:59:47.433397: step: 702/1541, loss: 0.006659643724560738 2023-01-23 18:59:48.296093: step: 704/1541, loss: 0.004683103412389755 2023-01-23 18:59:49.138043: step: 706/1541, loss: 6.490867963293567e-05 2023-01-23 18:59:50.070911: step: 708/1541, loss: 0.22273927927017212 2023-01-23 18:59:50.996924: step: 710/1541, loss: 0.12832729518413544 2023-01-23 18:59:51.859512: step: 712/1541, loss: 0.00040922066546045244 2023-01-23 18:59:52.752182: step: 714/1541, loss: 0.024075979366898537 2023-01-23 18:59:53.604225: step: 716/1541, loss: 0.000706262537278235 2023-01-23 18:59:54.522208: step: 718/1541, loss: 0.0007175203645601869 2023-01-23 18:59:55.428358: step: 720/1541, loss: 0.0005043039564043283 2023-01-23 18:59:56.339164: step: 722/1541, loss: 0.0005921887932345271 2023-01-23 18:59:57.208450: step: 724/1541, loss: 0.03982851281762123 2023-01-23 18:59:58.062523: step: 726/1541, loss: 0.007772454060614109 2023-01-23 18:59:58.985390: step: 728/1541, loss: 0.010419442318379879 2023-01-23 18:59:59.874163: step: 730/1541, loss: 0.003556463634595275 2023-01-23 19:00:00.796082: step: 732/1541, loss: 0.00171733892057091 2023-01-23 19:00:01.676206: step: 734/1541, loss: 0.0009394639637321234 2023-01-23 19:00:02.600182: step: 736/1541, loss: 0.015829814597964287 2023-01-23 19:00:03.517358: step: 738/1541, loss: 0.06976888328790665 2023-01-23 19:00:04.402692: step: 740/1541, loss: 0.011948302388191223 2023-01-23 19:00:05.292793: step: 742/1541, loss: 0.016271987929940224 2023-01-23 19:00:06.168533: step: 744/1541, loss: 0.2001945525407791 2023-01-23 19:00:07.047157: step: 746/1541, loss: 0.0016082340152934194 2023-01-23 19:00:08.089646: step: 748/1541, loss: 0.00452328659594059 2023-01-23 19:00:08.951272: step: 750/1541, loss: 0.03566092252731323 2023-01-23 19:00:09.822054: step: 752/1541, loss: 0.0043708765879273415 2023-01-23 19:00:10.695872: step: 754/1541, loss: 0.5713350176811218 2023-01-23 19:00:11.537267: step: 756/1541, loss: 0.08071401715278625 2023-01-23 19:00:12.424674: step: 758/1541, loss: 0.03856879100203514 2023-01-23 19:00:13.339441: step: 760/1541, loss: 0.01870899088680744 2023-01-23 19:00:14.224769: step: 762/1541, loss: 0.03450974076986313 2023-01-23 19:00:15.112099: step: 764/1541, loss: 2.7572970793698914e-05 2023-01-23 19:00:15.954626: step: 766/1541, loss: 0.00017442026000935584 2023-01-23 19:00:16.852151: step: 768/1541, loss: 0.0053970711305737495 2023-01-23 19:00:17.779950: step: 770/1541, loss: 0.01628420315682888 2023-01-23 19:00:18.646058: step: 772/1541, loss: 0.0020339470356702805 2023-01-23 19:00:19.638498: step: 774/1541, loss: 0.14049281179904938 2023-01-23 19:00:20.523396: step: 776/1541, loss: 0.00020480276725720614 2023-01-23 19:00:21.440828: step: 778/1541, loss: 0.13503816723823547 2023-01-23 19:00:22.325211: step: 780/1541, loss: 0.000106185267213732 2023-01-23 19:00:23.223734: step: 782/1541, loss: 0.05401233956217766 2023-01-23 19:00:24.105252: step: 784/1541, loss: 0.02301451936364174 2023-01-23 19:00:25.103226: step: 786/1541, loss: 0.05458977073431015 2023-01-23 19:00:26.022581: step: 788/1541, loss: 0.06756984442472458 2023-01-23 19:00:26.983706: step: 790/1541, loss: 0.08308837562799454 2023-01-23 19:00:27.858784: step: 792/1541, loss: 0.03338282182812691 2023-01-23 19:00:28.937867: step: 794/1541, loss: 0.0031225262209773064 2023-01-23 19:00:29.852541: step: 796/1541, loss: 0.004202007316052914 2023-01-23 19:00:30.774369: step: 798/1541, loss: 0.982593297958374 2023-01-23 19:00:31.698442: step: 800/1541, loss: 0.005165192764252424 2023-01-23 19:00:32.671354: step: 802/1541, loss: 0.00032445063698105514 2023-01-23 19:00:33.586655: step: 804/1541, loss: 0.001565684680826962 2023-01-23 19:00:34.506966: step: 806/1541, loss: 0.07493606954813004 2023-01-23 19:00:35.411488: step: 808/1541, loss: 0.02939583919942379 2023-01-23 19:00:36.303813: step: 810/1541, loss: 0.0003304774290882051 2023-01-23 19:00:37.159288: step: 812/1541, loss: 0.0004254037339705974 2023-01-23 19:00:38.037971: step: 814/1541, loss: 0.0034398268908262253 2023-01-23 19:00:38.892879: step: 816/1541, loss: 0.0050661093555390835 2023-01-23 19:00:39.772068: step: 818/1541, loss: 0.18615840375423431 2023-01-23 19:00:40.686767: step: 820/1541, loss: 0.0004518833593465388 2023-01-23 19:00:41.555186: step: 822/1541, loss: 0.021386021748185158 2023-01-23 19:00:42.451822: step: 824/1541, loss: 0.09959539026021957 2023-01-23 19:00:43.356344: step: 826/1541, loss: 0.05077895149588585 2023-01-23 19:00:44.241813: step: 828/1541, loss: 0.24738000333309174 2023-01-23 19:00:45.151194: step: 830/1541, loss: 0.004480266477912664 2023-01-23 19:00:46.030760: step: 832/1541, loss: 0.00011475039354991168 2023-01-23 19:00:46.924622: step: 834/1541, loss: 0.17294616997241974 2023-01-23 19:00:47.783983: step: 836/1541, loss: 0.0001482337829656899 2023-01-23 19:00:48.668657: step: 838/1541, loss: 0.07461008429527283 2023-01-23 19:00:49.574564: step: 840/1541, loss: 0.17546455562114716 2023-01-23 19:00:50.393775: step: 842/1541, loss: 0.0007690083002671599 2023-01-23 19:00:51.256471: step: 844/1541, loss: 0.0043588802218437195 2023-01-23 19:00:52.117386: step: 846/1541, loss: 0.09118427336215973 2023-01-23 19:00:53.081430: step: 848/1541, loss: 0.034630950540304184 2023-01-23 19:00:53.946430: step: 850/1541, loss: 0.0011745549272745848 2023-01-23 19:00:54.817952: step: 852/1541, loss: 0.00014519886462949216 2023-01-23 19:00:55.712581: step: 854/1541, loss: 0.0028990309219807386 2023-01-23 19:00:56.583038: step: 856/1541, loss: 0.01914982683956623 2023-01-23 19:00:57.510719: step: 858/1541, loss: 0.0009092632099054754 2023-01-23 19:00:58.407694: step: 860/1541, loss: 0.4044910669326782 2023-01-23 19:00:59.299974: step: 862/1541, loss: 0.000494071573484689 2023-01-23 19:01:00.223815: step: 864/1541, loss: 0.0063010649755597115 2023-01-23 19:01:01.127041: step: 866/1541, loss: 0.01375279575586319 2023-01-23 19:01:02.050465: step: 868/1541, loss: 0.013659491203725338 2023-01-23 19:01:02.964909: step: 870/1541, loss: 0.0022560071665793657 2023-01-23 19:01:03.874641: step: 872/1541, loss: 0.05168786272406578 2023-01-23 19:01:04.731224: step: 874/1541, loss: 0.00215220358222723 2023-01-23 19:01:05.654241: step: 876/1541, loss: 0.011977797374129295 2023-01-23 19:01:06.513772: step: 878/1541, loss: 0.002212432213127613 2023-01-23 19:01:07.383443: step: 880/1541, loss: 0.011227604933083057 2023-01-23 19:01:08.254981: step: 882/1541, loss: 0.015075383707880974 2023-01-23 19:01:09.166813: step: 884/1541, loss: 0.003668441204354167 2023-01-23 19:01:10.016903: step: 886/1541, loss: 0.0016771538648754358 2023-01-23 19:01:10.881734: step: 888/1541, loss: 0.004329943563789129 2023-01-23 19:01:11.778253: step: 890/1541, loss: 0.02668949030339718 2023-01-23 19:01:12.729726: step: 892/1541, loss: 0.2334444522857666 2023-01-23 19:01:13.605482: step: 894/1541, loss: 0.00016646177391521633 2023-01-23 19:01:14.559759: step: 896/1541, loss: 0.005816618446260691 2023-01-23 19:01:15.502525: step: 898/1541, loss: 0.2015581578016281 2023-01-23 19:01:16.363145: step: 900/1541, loss: 0.0012313914485275745 2023-01-23 19:01:17.250021: step: 902/1541, loss: 0.1043182760477066 2023-01-23 19:01:18.169187: step: 904/1541, loss: 0.0656481459736824 2023-01-23 19:01:19.086439: step: 906/1541, loss: 0.0014996007084846497 2023-01-23 19:01:19.991275: step: 908/1541, loss: 0.03382836654782295 2023-01-23 19:01:20.877596: step: 910/1541, loss: 0.000558358384296298 2023-01-23 19:01:21.769356: step: 912/1541, loss: 0.01600506901741028 2023-01-23 19:01:22.696143: step: 914/1541, loss: 0.04128512367606163 2023-01-23 19:01:23.582004: step: 916/1541, loss: 5.3622574341716245e-05 2023-01-23 19:01:24.482834: step: 918/1541, loss: 0.007046093232929707 2023-01-23 19:01:25.392179: step: 920/1541, loss: 0.22145995497703552 2023-01-23 19:01:26.261692: step: 922/1541, loss: 0.00041562432306818664 2023-01-23 19:01:27.120325: step: 924/1541, loss: 0.0026834034360945225 2023-01-23 19:01:28.019751: step: 926/1541, loss: -1.561582030262798e-05 2023-01-23 19:01:28.937466: step: 928/1541, loss: 0.04467925801873207 2023-01-23 19:01:29.794898: step: 930/1541, loss: 0.3722824156284332 2023-01-23 19:01:30.650365: step: 932/1541, loss: 0.000139542855322361 2023-01-23 19:01:31.527041: step: 934/1541, loss: 6.0350357671268284e-05 2023-01-23 19:01:32.398801: step: 936/1541, loss: 0.013468594290316105 2023-01-23 19:01:33.314934: step: 938/1541, loss: 0.12829266488552094 2023-01-23 19:01:34.183967: step: 940/1541, loss: 0.00030043430160731077 2023-01-23 19:01:35.057793: step: 942/1541, loss: 0.1610116809606552 2023-01-23 19:01:35.972486: step: 944/1541, loss: 0.0013847147347405553 2023-01-23 19:01:36.858473: step: 946/1541, loss: 0.0006635079625993967 2023-01-23 19:01:37.788346: step: 948/1541, loss: 3.018975257873535e-05 2023-01-23 19:01:38.735962: step: 950/1541, loss: 3.2074156479211524e-05 2023-01-23 19:01:39.615734: step: 952/1541, loss: 0.042262349277734756 2023-01-23 19:01:40.469383: step: 954/1541, loss: 4.245722448104061e-05 2023-01-23 19:01:41.379369: step: 956/1541, loss: 0.0009602587088011205 2023-01-23 19:01:42.278402: step: 958/1541, loss: 0.053195152431726456 2023-01-23 19:01:43.163669: step: 960/1541, loss: 0.002910839393734932 2023-01-23 19:01:44.070952: step: 962/1541, loss: 0.27193108201026917 2023-01-23 19:01:44.963633: step: 964/1541, loss: 0.00034912649425677955 2023-01-23 19:01:45.894079: step: 966/1541, loss: 0.2026388794183731 2023-01-23 19:01:46.758720: step: 968/1541, loss: -8.137551049003378e-06 2023-01-23 19:01:47.657994: step: 970/1541, loss: 0.037335578352212906 2023-01-23 19:01:48.610024: step: 972/1541, loss: 0.0005264265928417444 2023-01-23 19:01:49.544277: step: 974/1541, loss: 0.0006182945799082518 2023-01-23 19:01:50.437315: step: 976/1541, loss: 0.0006281929672695696 2023-01-23 19:01:51.393673: step: 978/1541, loss: 0.0001956982014235109 2023-01-23 19:01:52.249718: step: 980/1541, loss: 0.007517526391893625 2023-01-23 19:01:53.141462: step: 982/1541, loss: 8.257279841927812e-05 2023-01-23 19:01:54.017997: step: 984/1541, loss: 0.0003002184384968132 2023-01-23 19:01:54.923675: step: 986/1541, loss: 0.015980681404471397 2023-01-23 19:01:55.831959: step: 988/1541, loss: 9.528188820695505e-05 2023-01-23 19:01:56.701864: step: 990/1541, loss: 0.07036075741052628 2023-01-23 19:01:57.593035: step: 992/1541, loss: 0.08147157728672028 2023-01-23 19:01:58.619866: step: 994/1541, loss: 4.988599357602652e-06 2023-01-23 19:01:59.500477: step: 996/1541, loss: 0.002349917544052005 2023-01-23 19:02:00.404585: step: 998/1541, loss: 0.027869416400790215 2023-01-23 19:02:01.301549: step: 1000/1541, loss: 0.03568970784544945 2023-01-23 19:02:02.228619: step: 1002/1541, loss: 0.004319201223552227 2023-01-23 19:02:03.129381: step: 1004/1541, loss: 0.14661279320716858 2023-01-23 19:02:04.019068: step: 1006/1541, loss: 3.3754851756384596e-05 2023-01-23 19:02:04.912931: step: 1008/1541, loss: 0.0011754547012969851 2023-01-23 19:02:05.781545: step: 1010/1541, loss: 0.004872155375778675 2023-01-23 19:02:06.665745: step: 1012/1541, loss: 0.00013658026000484824 2023-01-23 19:02:07.532817: step: 1014/1541, loss: 0.0002447803854010999 2023-01-23 19:02:08.430586: step: 1016/1541, loss: 0.00022263462597038597 2023-01-23 19:02:09.298444: step: 1018/1541, loss: 0.002868895186111331 2023-01-23 19:02:10.168657: step: 1020/1541, loss: 0.007241044659167528 2023-01-23 19:02:11.035065: step: 1022/1541, loss: 0.0006022776942700148 2023-01-23 19:02:11.930716: step: 1024/1541, loss: 0.18720677495002747 2023-01-23 19:02:12.823308: step: 1026/1541, loss: 0.002597893588244915 2023-01-23 19:02:13.700907: step: 1028/1541, loss: 0.015073015354573727 2023-01-23 19:02:14.612713: step: 1030/1541, loss: 0.015474695712327957 2023-01-23 19:02:15.557084: step: 1032/1541, loss: 0.0748949721455574 2023-01-23 19:02:16.437194: step: 1034/1541, loss: 0.003735582809895277 2023-01-23 19:02:17.327227: step: 1036/1541, loss: 7.162956899264827e-05 2023-01-23 19:02:18.244110: step: 1038/1541, loss: 0.054029688239097595 2023-01-23 19:02:19.100877: step: 1040/1541, loss: 0.10969747602939606 2023-01-23 19:02:19.997404: step: 1042/1541, loss: 0.007238627877086401 2023-01-23 19:02:20.930083: step: 1044/1541, loss: 0.00025747119798325 2023-01-23 19:02:21.827068: step: 1046/1541, loss: 0.2470264732837677 2023-01-23 19:02:22.707014: step: 1048/1541, loss: 0.07554453611373901 2023-01-23 19:02:23.633481: step: 1050/1541, loss: 0.00018244236707687378 2023-01-23 19:02:24.514552: step: 1052/1541, loss: 8.707233791938052e-05 2023-01-23 19:02:25.439916: step: 1054/1541, loss: 0.008283544331789017 2023-01-23 19:02:26.339918: step: 1056/1541, loss: 6.381916318787262e-05 2023-01-23 19:02:27.232107: step: 1058/1541, loss: -5.039208872403833e-08 2023-01-23 19:02:28.182859: step: 1060/1541, loss: 0.0005518211983144283 2023-01-23 19:02:29.181691: step: 1062/1541, loss: 0.0003538011806085706 2023-01-23 19:02:30.046830: step: 1064/1541, loss: 0.0006040595471858978 2023-01-23 19:02:30.926838: step: 1066/1541, loss: 0.010263575240969658 2023-01-23 19:02:31.939337: step: 1068/1541, loss: 0.0005246825749054551 2023-01-23 19:02:32.839767: step: 1070/1541, loss: 0.004902364686131477 2023-01-23 19:02:33.764585: step: 1072/1541, loss: 0.0830020010471344 2023-01-23 19:02:34.618952: step: 1074/1541, loss: 5.914923895034008e-05 2023-01-23 19:02:35.502454: step: 1076/1541, loss: 0.012536735273897648 2023-01-23 19:02:36.367936: step: 1078/1541, loss: 0.09419193863868713 2023-01-23 19:02:37.275231: step: 1080/1541, loss: 0.001138511230237782 2023-01-23 19:02:38.152325: step: 1082/1541, loss: 0.015631187707185745 2023-01-23 19:02:39.128498: step: 1084/1541, loss: 0.0003192605508957058 2023-01-23 19:02:40.029491: step: 1086/1541, loss: 0.3632625341415405 2023-01-23 19:02:40.919294: step: 1088/1541, loss: 0.012963298708200455 2023-01-23 19:02:41.842125: step: 1090/1541, loss: 0.029988233000040054 2023-01-23 19:02:42.719287: step: 1092/1541, loss: 0.00037390520446933806 2023-01-23 19:02:43.620806: step: 1094/1541, loss: 0.08880792558193207 2023-01-23 19:02:44.477170: step: 1096/1541, loss: 0.0048916214145720005 2023-01-23 19:02:45.345701: step: 1098/1541, loss: 0.00014972354983910918 2023-01-23 19:02:46.286930: step: 1100/1541, loss: 0.06849831342697144 2023-01-23 19:02:47.242775: step: 1102/1541, loss: 0.001139287487603724 2023-01-23 19:02:48.093227: step: 1104/1541, loss: 0.013165787793695927 2023-01-23 19:02:48.981656: step: 1106/1541, loss: 6.373206269927323e-05 2023-01-23 19:02:49.875466: step: 1108/1541, loss: 0.05147423967719078 2023-01-23 19:02:50.779042: step: 1110/1541, loss: 0.03792494162917137 2023-01-23 19:02:51.700589: step: 1112/1541, loss: 0.03231818228960037 2023-01-23 19:02:52.612250: step: 1114/1541, loss: 4.8901772970566526e-05 2023-01-23 19:02:53.531096: step: 1116/1541, loss: 0.016169151291251183 2023-01-23 19:02:54.405719: step: 1118/1541, loss: 0.1159491240978241 2023-01-23 19:02:55.282303: step: 1120/1541, loss: 0.4440317749977112 2023-01-23 19:02:56.186403: step: 1122/1541, loss: 0.03771618753671646 2023-01-23 19:02:57.086498: step: 1124/1541, loss: 0.00037287213490344584 2023-01-23 19:02:58.120713: step: 1126/1541, loss: 0.0002981251454912126 2023-01-23 19:02:59.011833: step: 1128/1541, loss: 0.24301694333553314 2023-01-23 19:02:59.933984: step: 1130/1541, loss: 0.009382719174027443 2023-01-23 19:03:00.809063: step: 1132/1541, loss: 4.349875234765932e-05 2023-01-23 19:03:01.671084: step: 1134/1541, loss: 0.0007169304881244898 2023-01-23 19:03:02.567597: step: 1136/1541, loss: 0.00011965887824771926 2023-01-23 19:03:03.553337: step: 1138/1541, loss: 0.0069291009567677975 2023-01-23 19:03:04.483536: step: 1140/1541, loss: 0.009681749157607555 2023-01-23 19:03:05.490687: step: 1142/1541, loss: 0.00741939153522253 2023-01-23 19:03:06.368344: step: 1144/1541, loss: 0.0001386975054629147 2023-01-23 19:03:07.299883: step: 1146/1541, loss: 0.003763117827475071 2023-01-23 19:03:08.185103: step: 1148/1541, loss: 0.050852593034505844 2023-01-23 19:03:09.084721: step: 1150/1541, loss: 0.29235658049583435 2023-01-23 19:03:09.952921: step: 1152/1541, loss: 0.002201038645580411 2023-01-23 19:03:10.871748: step: 1154/1541, loss: 0.0001792106922948733 2023-01-23 19:03:11.782110: step: 1156/1541, loss: 3.3541913580847904e-05 2023-01-23 19:03:12.640254: step: 1158/1541, loss: 0.11341913044452667 2023-01-23 19:03:13.544924: step: 1160/1541, loss: 0.00034192364546470344 2023-01-23 19:03:14.429212: step: 1162/1541, loss: 0.1474488377571106 2023-01-23 19:03:15.343452: step: 1164/1541, loss: 0.0011273763375356793 2023-01-23 19:03:16.285335: step: 1166/1541, loss: 0.0011164933675900102 2023-01-23 19:03:17.172606: step: 1168/1541, loss: 0.056758929044008255 2023-01-23 19:03:18.032576: step: 1170/1541, loss: 0.07694199681282043 2023-01-23 19:03:18.939767: step: 1172/1541, loss: 0.0004774473491124809 2023-01-23 19:03:19.874330: step: 1174/1541, loss: 0.04459192603826523 2023-01-23 19:03:20.810501: step: 1176/1541, loss: 6.182508514029905e-05 2023-01-23 19:03:21.738059: step: 1178/1541, loss: 0.004089312627911568 2023-01-23 19:03:22.678178: step: 1180/1541, loss: 0.05485604703426361 2023-01-23 19:03:23.564811: step: 1182/1541, loss: 0.06180302053689957 2023-01-23 19:03:24.493597: step: 1184/1541, loss: 0.022224929183721542 2023-01-23 19:03:25.393435: step: 1186/1541, loss: 0.001739123254083097 2023-01-23 19:03:26.284609: step: 1188/1541, loss: 0.0097165796905756 2023-01-23 19:03:27.189685: step: 1190/1541, loss: 0.02746947854757309 2023-01-23 19:03:28.098773: step: 1192/1541, loss: 0.16964682936668396 2023-01-23 19:03:29.024593: step: 1194/1541, loss: 0.0001855258597061038 2023-01-23 19:03:29.911161: step: 1196/1541, loss: 0.33576667308807373 2023-01-23 19:03:30.819898: step: 1198/1541, loss: 0.5231835246086121 2023-01-23 19:03:31.687271: step: 1200/1541, loss: 0.8581125140190125 2023-01-23 19:03:32.607064: step: 1202/1541, loss: 0.016949016600847244 2023-01-23 19:03:33.529077: step: 1204/1541, loss: 6.532300903927535e-05 2023-01-23 19:03:34.406911: step: 1206/1541, loss: 0.008437822572886944 2023-01-23 19:03:35.335311: step: 1208/1541, loss: 0.00047016359167173505 2023-01-23 19:03:36.227754: step: 1210/1541, loss: 4.471499414648861e-05 2023-01-23 19:03:37.106261: step: 1212/1541, loss: 0.005641972180455923 2023-01-23 19:03:38.004998: step: 1214/1541, loss: 0.1417974978685379 2023-01-23 19:03:38.906492: step: 1216/1541, loss: 0.0003984146751463413 2023-01-23 19:03:39.794478: step: 1218/1541, loss: 0.1737990528345108 2023-01-23 19:03:40.669122: step: 1220/1541, loss: 0.00022540890495292842 2023-01-23 19:03:41.546054: step: 1222/1541, loss: 0.03741609677672386 2023-01-23 19:03:42.446470: step: 1224/1541, loss: 0.005535002797842026 2023-01-23 19:03:43.345451: step: 1226/1541, loss: 0.0001972754835151136 2023-01-23 19:03:44.271258: step: 1228/1541, loss: 0.38110262155532837 2023-01-23 19:03:45.197389: step: 1230/1541, loss: 0.004380285274237394 2023-01-23 19:03:46.121242: step: 1232/1541, loss: 0.00600343756377697 2023-01-23 19:03:47.015648: step: 1234/1541, loss: 0.02046847529709339 2023-01-23 19:03:47.920156: step: 1236/1541, loss: 0.0002248916425742209 2023-01-23 19:03:48.811473: step: 1238/1541, loss: 0.21033816039562225 2023-01-23 19:03:49.714189: step: 1240/1541, loss: 0.0016967353876680136 2023-01-23 19:03:50.596774: step: 1242/1541, loss: 0.0001758315193001181 2023-01-23 19:03:51.476083: step: 1244/1541, loss: 0.0010143371764570475 2023-01-23 19:03:52.364863: step: 1246/1541, loss: 0.00023027769930195063 2023-01-23 19:03:53.262266: step: 1248/1541, loss: 0.0001420070620952174 2023-01-23 19:03:54.200345: step: 1250/1541, loss: 0.0009370298939757049 2023-01-23 19:03:55.077851: step: 1252/1541, loss: 0.20507289469242096 2023-01-23 19:03:56.003208: step: 1254/1541, loss: 0.022782450541853905 2023-01-23 19:03:57.031476: step: 1256/1541, loss: 0.2753523290157318 2023-01-23 19:03:57.920356: step: 1258/1541, loss: 6.57698474242352e-05 2023-01-23 19:03:58.861800: step: 1260/1541, loss: 0.001590827596373856 2023-01-23 19:03:59.764522: step: 1262/1541, loss: 0.00028180904337204993 2023-01-23 19:04:00.641202: step: 1264/1541, loss: 0.05127452686429024 2023-01-23 19:04:01.534009: step: 1266/1541, loss: 0.012263874523341656 2023-01-23 19:04:02.401103: step: 1268/1541, loss: 0.00042331317672505975 2023-01-23 19:04:03.275341: step: 1270/1541, loss: 0.0003318696399219334 2023-01-23 19:04:04.225337: step: 1272/1541, loss: 0.8190096020698547 2023-01-23 19:04:05.096223: step: 1274/1541, loss: 9.759065869729966e-05 2023-01-23 19:04:05.954197: step: 1276/1541, loss: 0.001041524694301188 2023-01-23 19:04:06.858952: step: 1278/1541, loss: 0.0004607029550243169 2023-01-23 19:04:07.763482: step: 1280/1541, loss: 0.011571274138987064 2023-01-23 19:04:08.664161: step: 1282/1541, loss: 0.0021171749103814363 2023-01-23 19:04:09.563798: step: 1284/1541, loss: 0.27400124073028564 2023-01-23 19:04:10.421514: step: 1286/1541, loss: 0.0005062585696578026 2023-01-23 19:04:11.276699: step: 1288/1541, loss: 0.007470917422324419 2023-01-23 19:04:12.196454: step: 1290/1541, loss: 0.00016498652985319495 2023-01-23 19:04:13.082723: step: 1292/1541, loss: 9.163061622530222e-05 2023-01-23 19:04:14.013003: step: 1294/1541, loss: 0.0007668337784707546 2023-01-23 19:04:14.893166: step: 1296/1541, loss: 0.00027796582435257733 2023-01-23 19:04:15.785675: step: 1298/1541, loss: 0.07534223049879074 2023-01-23 19:04:16.671037: step: 1300/1541, loss: 0.2801534831523895 2023-01-23 19:04:17.512627: step: 1302/1541, loss: 0.027742929756641388 2023-01-23 19:04:18.466468: step: 1304/1541, loss: 0.0005498199607245624 2023-01-23 19:04:19.374095: step: 1306/1541, loss: 0.09125224500894547 2023-01-23 19:04:20.221382: step: 1308/1541, loss: 0.02897615171968937 2023-01-23 19:04:21.211787: step: 1310/1541, loss: 0.004144096747040749 2023-01-23 19:04:22.087870: step: 1312/1541, loss: -1.6618147128610872e-05 2023-01-23 19:04:23.022159: step: 1314/1541, loss: 0.012572798877954483 2023-01-23 19:04:23.918328: step: 1316/1541, loss: 9.460021101403981e-05 2023-01-23 19:04:24.831396: step: 1318/1541, loss: 0.1257251352071762 2023-01-23 19:04:25.729945: step: 1320/1541, loss: 8.27267358545214e-05 2023-01-23 19:04:26.629769: step: 1322/1541, loss: 1.8343323972658254e-05 2023-01-23 19:04:27.518496: step: 1324/1541, loss: 0.0006494858535006642 2023-01-23 19:04:28.455802: step: 1326/1541, loss: 0.09868501871824265 2023-01-23 19:04:29.344885: step: 1328/1541, loss: 0.02118748426437378 2023-01-23 19:04:30.260301: step: 1330/1541, loss: 0.019349783658981323 2023-01-23 19:04:31.209643: step: 1332/1541, loss: 0.1399627923965454 2023-01-23 19:04:32.131454: step: 1334/1541, loss: 0.010176356881856918 2023-01-23 19:04:33.015167: step: 1336/1541, loss: 0.06015119329094887 2023-01-23 19:04:33.948501: step: 1338/1541, loss: 0.10550861805677414 2023-01-23 19:04:34.877920: step: 1340/1541, loss: 0.001009476836770773 2023-01-23 19:04:35.808619: step: 1342/1541, loss: 0.0010294199455529451 2023-01-23 19:04:36.706856: step: 1344/1541, loss: 0.007412455044686794 2023-01-23 19:04:37.644571: step: 1346/1541, loss: 0.015174276195466518 2023-01-23 19:04:38.541919: step: 1348/1541, loss: 0.0005064575234428048 2023-01-23 19:04:39.455669: step: 1350/1541, loss: 3.773427670239471e-05 2023-01-23 19:04:40.402987: step: 1352/1541, loss: 0.010368399322032928 2023-01-23 19:04:41.300235: step: 1354/1541, loss: 0.21008913218975067 2023-01-23 19:04:42.219404: step: 1356/1541, loss: 0.0011060755932703614 2023-01-23 19:04:43.099996: step: 1358/1541, loss: 0.0017068900633603334 2023-01-23 19:04:44.003475: step: 1360/1541, loss: 0.04717755317687988 2023-01-23 19:04:44.899470: step: 1362/1541, loss: 0.00311936903744936 2023-01-23 19:04:45.774234: step: 1364/1541, loss: 0.6918103098869324 2023-01-23 19:04:46.710585: step: 1366/1541, loss: 0.0003918817965313792 2023-01-23 19:04:47.609745: step: 1368/1541, loss: 0.14861567318439484 2023-01-23 19:04:48.512397: step: 1370/1541, loss: 0.00010169869347009808 2023-01-23 19:04:49.445738: step: 1372/1541, loss: 0.09087216109037399 2023-01-23 19:04:50.366477: step: 1374/1541, loss: 0.012894297018647194 2023-01-23 19:04:51.245598: step: 1376/1541, loss: 0.0014015014749020338 2023-01-23 19:04:52.182889: step: 1378/1541, loss: 0.0012955251149833202 2023-01-23 19:04:53.133814: step: 1380/1541, loss: 0.00624624639749527 2023-01-23 19:04:54.036118: step: 1382/1541, loss: 0.19386547803878784 2023-01-23 19:04:54.896584: step: 1384/1541, loss: 0.00014352428843267262 2023-01-23 19:04:55.864942: step: 1386/1541, loss: 9.792942728381604e-05 2023-01-23 19:04:56.792010: step: 1388/1541, loss: 6.050313822925091e-05 2023-01-23 19:04:57.693161: step: 1390/1541, loss: 0.004794289357960224 2023-01-23 19:04:58.603778: step: 1392/1541, loss: 0.012054556049406528 2023-01-23 19:04:59.498683: step: 1394/1541, loss: 0.01400203537195921 2023-01-23 19:05:00.442741: step: 1396/1541, loss: 4.496341716730967e-05 2023-01-23 19:05:01.399663: step: 1398/1541, loss: 0.02749750390648842 2023-01-23 19:05:02.299408: step: 1400/1541, loss: 0.05720960348844528 2023-01-23 19:05:03.192065: step: 1402/1541, loss: 0.007954617962241173 2023-01-23 19:05:04.106250: step: 1404/1541, loss: 0.35974451899528503 2023-01-23 19:05:05.026753: step: 1406/1541, loss: 3.997718158643693e-05 2023-01-23 19:05:05.889660: step: 1408/1541, loss: 0.004718072712421417 2023-01-23 19:05:06.775455: step: 1410/1541, loss: 0.22018884122371674 2023-01-23 19:05:07.709142: step: 1412/1541, loss: 0.015578532591462135 2023-01-23 19:05:08.598359: step: 1414/1541, loss: 0.13257597386837006 2023-01-23 19:05:09.507232: step: 1416/1541, loss: 0.0005715698935091496 2023-01-23 19:05:10.386261: step: 1418/1541, loss: 0.0025720305275171995 2023-01-23 19:05:11.263661: step: 1420/1541, loss: 0.003946686629205942 2023-01-23 19:05:12.161035: step: 1422/1541, loss: 0.012040294706821442 2023-01-23 19:05:13.077414: step: 1424/1541, loss: 0.04195189103484154 2023-01-23 19:05:13.967041: step: 1426/1541, loss: 0.03930435702204704 2023-01-23 19:05:14.907150: step: 1428/1541, loss: 0.001027073711156845 2023-01-23 19:05:15.812754: step: 1430/1541, loss: 0.0058553749695420265 2023-01-23 19:05:16.711636: step: 1432/1541, loss: 0.0022185691632330418 2023-01-23 19:05:17.586959: step: 1434/1541, loss: 0.0004357897269073874 2023-01-23 19:05:18.460423: step: 1436/1541, loss: 0.0003930528473574668 2023-01-23 19:05:19.338777: step: 1438/1541, loss: 0.0009322241530753672 2023-01-23 19:05:20.282886: step: 1440/1541, loss: 0.030416861176490784 2023-01-23 19:05:21.165908: step: 1442/1541, loss: 0.002586725866422057 2023-01-23 19:05:22.068574: step: 1444/1541, loss: 0.0007912660366855562 2023-01-23 19:05:22.968583: step: 1446/1541, loss: 0.22712506353855133 2023-01-23 19:05:23.856355: step: 1448/1541, loss: 0.0511968731880188 2023-01-23 19:05:24.779021: step: 1450/1541, loss: 0.24646979570388794 2023-01-23 19:05:25.735725: step: 1452/1541, loss: 0.0008586367475800216 2023-01-23 19:05:26.636781: step: 1454/1541, loss: 0.043207064270973206 2023-01-23 19:05:27.586452: step: 1456/1541, loss: 0.008344506844878197 2023-01-23 19:05:28.456541: step: 1458/1541, loss: 0.4309713840484619 2023-01-23 19:05:29.313001: step: 1460/1541, loss: 0.0009729436715133488 2023-01-23 19:05:30.222894: step: 1462/1541, loss: 0.0763111561536789 2023-01-23 19:05:31.072350: step: 1464/1541, loss: 0.2303963005542755 2023-01-23 19:05:31.988200: step: 1466/1541, loss: 0.023485437035560608 2023-01-23 19:05:32.891830: step: 1468/1541, loss: 0.004787149839103222 2023-01-23 19:05:33.776734: step: 1470/1541, loss: 0.0028071505948901176 2023-01-23 19:05:34.654343: step: 1472/1541, loss: 0.024306274950504303 2023-01-23 19:05:35.535191: step: 1474/1541, loss: 0.0009587162639945745 2023-01-23 19:05:36.494209: step: 1476/1541, loss: 0.11635639518499374 2023-01-23 19:05:37.373918: step: 1478/1541, loss: 0.007253024727106094 2023-01-23 19:05:38.269893: step: 1480/1541, loss: 0.02324015274643898 2023-01-23 19:05:39.188201: step: 1482/1541, loss: 0.0011778862681239843 2023-01-23 19:05:40.065714: step: 1484/1541, loss: 0.001038008602336049 2023-01-23 19:05:41.021294: step: 1486/1541, loss: 0.0011777773033827543 2023-01-23 19:05:41.919781: step: 1488/1541, loss: 0.0016690526390448213 2023-01-23 19:05:42.803443: step: 1490/1541, loss: 0.1347445398569107 2023-01-23 19:05:43.683841: step: 1492/1541, loss: 0.15221571922302246 2023-01-23 19:05:44.545527: step: 1494/1541, loss: 0.0026102291885763407 2023-01-23 19:05:45.455348: step: 1496/1541, loss: 0.09023944288492203 2023-01-23 19:05:46.362476: step: 1498/1541, loss: 0.012732391245663166 2023-01-23 19:05:47.219826: step: 1500/1541, loss: 0.0010715412208810449 2023-01-23 19:05:48.260033: step: 1502/1541, loss: 3.153407096862793 2023-01-23 19:05:49.212481: step: 1504/1541, loss: 0.01952918991446495 2023-01-23 19:05:50.160837: step: 1506/1541, loss: 0.35026633739471436 2023-01-23 19:05:51.063010: step: 1508/1541, loss: 0.14788611233234406 2023-01-23 19:05:51.971224: step: 1510/1541, loss: 0.006441375706344843 2023-01-23 19:05:52.869954: step: 1512/1541, loss: 0.03776387497782707 2023-01-23 19:05:53.749512: step: 1514/1541, loss: 0.0002903889981098473 2023-01-23 19:05:54.689355: step: 1516/1541, loss: 0.0030821352265775204 2023-01-23 19:05:55.546040: step: 1518/1541, loss: 0.0016652510967105627 2023-01-23 19:05:56.403578: step: 1520/1541, loss: 0.013011347502470016 2023-01-23 19:05:57.328618: step: 1522/1541, loss: 0.02858724258840084 2023-01-23 19:05:58.235912: step: 1524/1541, loss: 0.0020438889041543007 2023-01-23 19:05:59.172281: step: 1526/1541, loss: 0.0015779510140419006 2023-01-23 19:06:00.061267: step: 1528/1541, loss: 0.0025145686231553555 2023-01-23 19:06:01.019891: step: 1530/1541, loss: 0.0026123635470867157 2023-01-23 19:06:01.952343: step: 1532/1541, loss: 0.001458314131014049 2023-01-23 19:06:02.836900: step: 1534/1541, loss: 0.011088418774306774 2023-01-23 19:06:03.700695: step: 1536/1541, loss: 0.10892214626073837 2023-01-23 19:06:04.552744: step: 1538/1541, loss: 0.32878896594047546 2023-01-23 19:06:05.392641: step: 1540/1541, loss: 0.01621874049305916 2023-01-23 19:06:06.300227: step: 1542/1541, loss: 0.004677947144955397 2023-01-23 19:06:07.179245: step: 1544/1541, loss: 0.11044365912675858 2023-01-23 19:06:08.098788: step: 1546/1541, loss: 0.0026883212849497795 2023-01-23 19:06:09.021336: step: 1548/1541, loss: 0.045176710933446884 2023-01-23 19:06:09.890247: step: 1550/1541, loss: 0.0046112146228551865 2023-01-23 19:06:10.748478: step: 1552/1541, loss: 0.0002989911590702832 2023-01-23 19:06:11.648043: step: 1554/1541, loss: 0.0929848849773407 2023-01-23 19:06:12.591800: step: 1556/1541, loss: 0.0011050424072891474 2023-01-23 19:06:13.532313: step: 1558/1541, loss: 0.022787166759371758 2023-01-23 19:06:14.513992: step: 1560/1541, loss: 0.0007425983785651624 2023-01-23 19:06:15.463185: step: 1562/1541, loss: 0.008005950599908829 2023-01-23 19:06:16.394800: step: 1564/1541, loss: 0.053781669586896896 2023-01-23 19:06:17.301757: step: 1566/1541, loss: 0.003175705438479781 2023-01-23 19:06:18.272791: step: 1568/1541, loss: 0.0014694246929138899 2023-01-23 19:06:19.158804: step: 1570/1541, loss: 0.01082096341997385 2023-01-23 19:06:20.085825: step: 1572/1541, loss: 0.00280424440279603 2023-01-23 19:06:21.017254: step: 1574/1541, loss: 0.00010410163667984307 2023-01-23 19:06:21.959647: step: 1576/1541, loss: 0.0025669331662356853 2023-01-23 19:06:22.869523: step: 1578/1541, loss: 0.0004926088149659336 2023-01-23 19:06:23.792292: step: 1580/1541, loss: 0.00024924310855567455 2023-01-23 19:06:24.689963: step: 1582/1541, loss: 0.0005143845337443054 2023-01-23 19:06:25.568316: step: 1584/1541, loss: 0.00039490184281021357 2023-01-23 19:06:26.419068: step: 1586/1541, loss: 0.4413178563117981 2023-01-23 19:06:27.300148: step: 1588/1541, loss: 0.020988404750823975 2023-01-23 19:06:28.163750: step: 1590/1541, loss: 0.12454589456319809 2023-01-23 19:06:29.106136: step: 1592/1541, loss: 0.037043247371912 2023-01-23 19:06:29.967784: step: 1594/1541, loss: 0.0008703142520971596 2023-01-23 19:06:30.827172: step: 1596/1541, loss: 3.579006079235114e-05 2023-01-23 19:06:31.774494: step: 1598/1541, loss: 0.001132387900725007 2023-01-23 19:06:32.738664: step: 1600/1541, loss: 0.002496364526450634 2023-01-23 19:06:33.628125: step: 1602/1541, loss: 0.00020471926836762577 2023-01-23 19:06:34.528388: step: 1604/1541, loss: 0.004741146229207516 2023-01-23 19:06:35.431517: step: 1606/1541, loss: 0.0032561812549829483 2023-01-23 19:06:36.318968: step: 1608/1541, loss: 0.025468239560723305 2023-01-23 19:06:37.222085: step: 1610/1541, loss: 4.2549610952846706e-05 2023-01-23 19:06:38.125440: step: 1612/1541, loss: 1.7682423276710324e-05 2023-01-23 19:06:39.010241: step: 1614/1541, loss: 0.0008962037973105907 2023-01-23 19:06:39.863287: step: 1616/1541, loss: 0.0022182699758559465 2023-01-23 19:06:40.714466: step: 1618/1541, loss: 6.437131378334016e-05 2023-01-23 19:06:41.575013: step: 1620/1541, loss: 2.0646948541980237e-05 2023-01-23 19:06:42.469592: step: 1622/1541, loss: 0.017123598605394363 2023-01-23 19:06:43.401219: step: 1624/1541, loss: 0.002464784774929285 2023-01-23 19:06:44.353123: step: 1626/1541, loss: 0.0005737977335229516 2023-01-23 19:06:45.194997: step: 1628/1541, loss: 0.003014585003256798 2023-01-23 19:06:46.070750: step: 1630/1541, loss: 0.021613050252199173 2023-01-23 19:06:46.975233: step: 1632/1541, loss: 0.06972196698188782 2023-01-23 19:06:47.873520: step: 1634/1541, loss: 0.0004616064834408462 2023-01-23 19:06:48.778887: step: 1636/1541, loss: 0.01630263403058052 2023-01-23 19:06:49.736936: step: 1638/1541, loss: 0.0005351700820028782 2023-01-23 19:06:50.641534: step: 1640/1541, loss: 0.11944413930177689 2023-01-23 19:06:51.518822: step: 1642/1541, loss: 0.0014713669661432505 2023-01-23 19:06:52.378498: step: 1644/1541, loss: 0.0019947055261582136 2023-01-23 19:06:53.284478: step: 1646/1541, loss: 0.00011264260683674365 2023-01-23 19:06:54.155117: step: 1648/1541, loss: 0.024892285466194153 2023-01-23 19:06:55.001092: step: 1650/1541, loss: 0.00100371812004596 2023-01-23 19:06:55.933584: step: 1652/1541, loss: 0.2852706015110016 2023-01-23 19:06:56.852919: step: 1654/1541, loss: 0.0021490631625056267 2023-01-23 19:06:57.744830: step: 1656/1541, loss: 0.0002982516889460385 2023-01-23 19:06:58.625310: step: 1658/1541, loss: 0.0007468568510375917 2023-01-23 19:06:59.546480: step: 1660/1541, loss: 0.004599161446094513 2023-01-23 19:07:00.502802: step: 1662/1541, loss: 0.00018249292043037713 2023-01-23 19:07:01.394204: step: 1664/1541, loss: 0.0008548217010684311 2023-01-23 19:07:02.288825: step: 1666/1541, loss: 0.0001727019262034446 2023-01-23 19:07:03.164417: step: 1668/1541, loss: 0.060235701501369476 2023-01-23 19:07:04.074817: step: 1670/1541, loss: 4.625180008588359e-05 2023-01-23 19:07:04.982456: step: 1672/1541, loss: 0.0002645898493938148 2023-01-23 19:07:05.886749: step: 1674/1541, loss: 0.0012291934108361602 2023-01-23 19:07:06.763486: step: 1676/1541, loss: 0.02927200309932232 2023-01-23 19:07:07.632342: step: 1678/1541, loss: 7.398710295092314e-05 2023-01-23 19:07:08.493298: step: 1680/1541, loss: 2.950742782559246e-06 2023-01-23 19:07:09.382223: step: 1682/1541, loss: 0.13150738179683685 2023-01-23 19:07:10.276780: step: 1684/1541, loss: 0.006306726485490799 2023-01-23 19:07:11.159128: step: 1686/1541, loss: 0.004766205325722694 2023-01-23 19:07:12.046435: step: 1688/1541, loss: 4.424456710694358e-05 2023-01-23 19:07:12.968130: step: 1690/1541, loss: 0.0004654046497307718 2023-01-23 19:07:13.845273: step: 1692/1541, loss: 0.0011935289949178696 2023-01-23 19:07:14.778781: step: 1694/1541, loss: 0.0774320587515831 2023-01-23 19:07:15.647457: step: 1696/1541, loss: 0.0009357549715787172 2023-01-23 19:07:16.575087: step: 1698/1541, loss: 0.002468332415446639 2023-01-23 19:07:17.477602: step: 1700/1541, loss: 0.5328259468078613 2023-01-23 19:07:18.363026: step: 1702/1541, loss: 0.003010737942531705 2023-01-23 19:07:19.222039: step: 1704/1541, loss: 0.0008535108063369989 2023-01-23 19:07:20.164805: step: 1706/1541, loss: 0.0016717531252652407 2023-01-23 19:07:21.054363: step: 1708/1541, loss: 0.1387290060520172 2023-01-23 19:07:21.931025: step: 1710/1541, loss: 0.05727413296699524 2023-01-23 19:07:22.842981: step: 1712/1541, loss: 0.0009785466827452183 2023-01-23 19:07:23.768563: step: 1714/1541, loss: 5.5635850003454834e-05 2023-01-23 19:07:24.766151: step: 1716/1541, loss: 0.004187557380646467 2023-01-23 19:07:25.643098: step: 1718/1541, loss: 0.0897512212395668 2023-01-23 19:07:26.575940: step: 1720/1541, loss: 0.0007367263897322118 2023-01-23 19:07:27.480409: step: 1722/1541, loss: 0.6331450939178467 2023-01-23 19:07:28.349817: step: 1724/1541, loss: 0.0008989358902908862 2023-01-23 19:07:29.281403: step: 1726/1541, loss: 0.0056005483493208885 2023-01-23 19:07:30.164450: step: 1728/1541, loss: 0.0002475571818649769 2023-01-23 19:07:31.027285: step: 1730/1541, loss: 2.8986120014451444e-05 2023-01-23 19:07:31.948113: step: 1732/1541, loss: 0.00015253701712936163 2023-01-23 19:07:32.811984: step: 1734/1541, loss: 6.594409933313727e-05 2023-01-23 19:07:33.687812: step: 1736/1541, loss: 1.6224976207013242e-05 2023-01-23 19:07:34.579143: step: 1738/1541, loss: 0.00034714810317382216 2023-01-23 19:07:35.505439: step: 1740/1541, loss: 0.029479995369911194 2023-01-23 19:07:36.418473: step: 1742/1541, loss: 0.29956379532814026 2023-01-23 19:07:37.299982: step: 1744/1541, loss: 0.10096843540668488 2023-01-23 19:07:38.195459: step: 1746/1541, loss: 0.008441848680377007 2023-01-23 19:07:39.133369: step: 1748/1541, loss: 0.0012674637837335467 2023-01-23 19:07:40.119286: step: 1750/1541, loss: 0.00017410599684808403 2023-01-23 19:07:40.995417: step: 1752/1541, loss: 0.10163871198892593 2023-01-23 19:07:41.893869: step: 1754/1541, loss: 0.0003235786862205714 2023-01-23 19:07:42.772008: step: 1756/1541, loss: 9.651656728237867e-05 2023-01-23 19:07:43.708010: step: 1758/1541, loss: 0.0006856971886008978 2023-01-23 19:07:44.690733: step: 1760/1541, loss: 0.016582414507865906 2023-01-23 19:07:45.606322: step: 1762/1541, loss: 2.8087906684959307e-05 2023-01-23 19:07:46.484793: step: 1764/1541, loss: 0.05739012733101845 2023-01-23 19:07:47.431241: step: 1766/1541, loss: 0.25447604060173035 2023-01-23 19:07:48.313437: step: 1768/1541, loss: 0.08944927901029587 2023-01-23 19:07:49.249535: step: 1770/1541, loss: 0.007114575710147619 2023-01-23 19:07:50.161973: step: 1772/1541, loss: 0.0010937575716525316 2023-01-23 19:07:51.037678: step: 1774/1541, loss: 0.2770388126373291 2023-01-23 19:07:51.941524: step: 1776/1541, loss: 0.15853826701641083 2023-01-23 19:07:52.851867: step: 1778/1541, loss: 0.013276352547109127 2023-01-23 19:07:53.689158: step: 1780/1541, loss: 0.08156000822782516 2023-01-23 19:07:54.549684: step: 1782/1541, loss: 0.009233609773218632 2023-01-23 19:07:55.421509: step: 1784/1541, loss: 0.13848401606082916 2023-01-23 19:07:56.307551: step: 1786/1541, loss: 0.08349043130874634 2023-01-23 19:07:57.194083: step: 1788/1541, loss: 0.12585659325122833 2023-01-23 19:07:58.091187: step: 1790/1541, loss: 0.0013074440648779273 2023-01-23 19:07:59.011954: step: 1792/1541, loss: 0.005794572178274393 2023-01-23 19:07:59.855047: step: 1794/1541, loss: 0.0016090997960418463 2023-01-23 19:08:00.720889: step: 1796/1541, loss: 0.0110838133841753 2023-01-23 19:08:01.610365: step: 1798/1541, loss: 0.0009030263754539192 2023-01-23 19:08:02.463391: step: 1800/1541, loss: 0.026045341044664383 2023-01-23 19:08:03.357230: step: 1802/1541, loss: 0.0019849769305437803 2023-01-23 19:08:04.232046: step: 1804/1541, loss: 0.003174922661855817 2023-01-23 19:08:05.181031: step: 1806/1541, loss: 0.0004467506369110197 2023-01-23 19:08:06.156429: step: 1808/1541, loss: 0.0020481275860220194 2023-01-23 19:08:07.081416: step: 1810/1541, loss: 0.013245146721601486 2023-01-23 19:08:07.998123: step: 1812/1541, loss: 0.00036272313445806503 2023-01-23 19:08:08.918547: step: 1814/1541, loss: 0.022291040048003197 2023-01-23 19:08:09.882500: step: 1816/1541, loss: 0.562724232673645 2023-01-23 19:08:10.790285: step: 1818/1541, loss: 0.003472699550911784 2023-01-23 19:08:11.689931: step: 1820/1541, loss: 0.0006245008553378284 2023-01-23 19:08:12.597660: step: 1822/1541, loss: 0.00020737586601171643 2023-01-23 19:08:13.543569: step: 1824/1541, loss: 0.10696475207805634 2023-01-23 19:08:14.398592: step: 1826/1541, loss: 0.0006363311549648643 2023-01-23 19:08:15.281841: step: 1828/1541, loss: 0.1075821965932846 2023-01-23 19:08:16.156510: step: 1830/1541, loss: 0.0004335540870670229 2023-01-23 19:08:17.034761: step: 1832/1541, loss: 0.0351397879421711 2023-01-23 19:08:17.932500: step: 1834/1541, loss: 0.0014955197693780065 2023-01-23 19:08:18.818641: step: 1836/1541, loss: 0.01444387435913086 2023-01-23 19:08:19.724719: step: 1838/1541, loss: 0.0017484166892245412 2023-01-23 19:08:20.579546: step: 1840/1541, loss: 0.13749639689922333 2023-01-23 19:08:21.465513: step: 1842/1541, loss: 0.017087846994400024 2023-01-23 19:08:22.352383: step: 1844/1541, loss: 0.0010747074848040938 2023-01-23 19:08:23.286661: step: 1846/1541, loss: 0.00016261794371530414 2023-01-23 19:08:24.154152: step: 1848/1541, loss: 0.06218577176332474 2023-01-23 19:08:25.042248: step: 1850/1541, loss: 0.022317031398415565 2023-01-23 19:08:25.943033: step: 1852/1541, loss: 0.13491559028625488 2023-01-23 19:08:26.816346: step: 1854/1541, loss: 0.01380821131169796 2023-01-23 19:08:27.658299: step: 1856/1541, loss: 0.0002663242048583925 2023-01-23 19:08:28.567821: step: 1858/1541, loss: 0.17877866327762604 2023-01-23 19:08:29.461337: step: 1860/1541, loss: 0.010997530072927475 2023-01-23 19:08:30.389580: step: 1862/1541, loss: 0.2591208517551422 2023-01-23 19:08:31.340933: step: 1864/1541, loss: 2.1533585822908208e-05 2023-01-23 19:08:32.271300: step: 1866/1541, loss: 0.01185980997979641 2023-01-23 19:08:33.140328: step: 1868/1541, loss: 0.006602066569030285 2023-01-23 19:08:34.005081: step: 1870/1541, loss: 0.0009943771874532104 2023-01-23 19:08:34.979071: step: 1872/1541, loss: 0.003238918026909232 2023-01-23 19:08:35.883085: step: 1874/1541, loss: 0.0001038554692058824 2023-01-23 19:08:36.828350: step: 1876/1541, loss: 0.00010462623322382569 2023-01-23 19:08:37.736342: step: 1878/1541, loss: 0.00014810657012276351 2023-01-23 19:08:38.599799: step: 1880/1541, loss: 0.08733803033828735 2023-01-23 19:08:39.471803: step: 1882/1541, loss: 0.0062319994904100895 2023-01-23 19:08:40.378134: step: 1884/1541, loss: 0.002555957529693842 2023-01-23 19:08:41.258359: step: 1886/1541, loss: 0.00459017651155591 2023-01-23 19:08:42.135811: step: 1888/1541, loss: 0.5376939177513123 2023-01-23 19:08:43.034215: step: 1890/1541, loss: 0.0027153976261615753 2023-01-23 19:08:43.974418: step: 1892/1541, loss: 4.971552698407322e-05 2023-01-23 19:08:44.859952: step: 1894/1541, loss: 0.00025554309831932187 2023-01-23 19:08:45.765409: step: 1896/1541, loss: 0.00017568189650774002 2023-01-23 19:08:46.667152: step: 1898/1541, loss: 0.030361784622073174 2023-01-23 19:08:47.555167: step: 1900/1541, loss: 0.001660091569647193 2023-01-23 19:08:48.485095: step: 1902/1541, loss: 0.0013880879851058125 2023-01-23 19:08:49.384348: step: 1904/1541, loss: 2.7528776627150364e-05 2023-01-23 19:08:50.308581: step: 1906/1541, loss: 0.2136687934398651 2023-01-23 19:08:51.199444: step: 1908/1541, loss: 0.08609475195407867 2023-01-23 19:08:52.058355: step: 1910/1541, loss: 0.3026983141899109 2023-01-23 19:08:52.921776: step: 1912/1541, loss: 0.002640125108882785 2023-01-23 19:08:53.834767: step: 1914/1541, loss: 0.00904797948896885 2023-01-23 19:08:54.702021: step: 1916/1541, loss: 0.0005914412322454154 2023-01-23 19:08:55.582292: step: 1918/1541, loss: 0.0023897343780845404 2023-01-23 19:08:56.455730: step: 1920/1541, loss: 0.06744451820850372 2023-01-23 19:08:57.392118: step: 1922/1541, loss: 0.1277303844690323 2023-01-23 19:08:58.385216: step: 1924/1541, loss: 1.2540863752365112 2023-01-23 19:08:59.272906: step: 1926/1541, loss: 0.0019286375027149916 2023-01-23 19:09:00.181423: step: 1928/1541, loss: 0.0007439687033183873 2023-01-23 19:09:01.038840: step: 1930/1541, loss: 0.005461074411869049 2023-01-23 19:09:01.948248: step: 1932/1541, loss: 0.02881855145096779 2023-01-23 19:09:02.852075: step: 1934/1541, loss: 0.0015423961449414492 2023-01-23 19:09:03.718768: step: 1936/1541, loss: 0.00854963157325983 2023-01-23 19:09:04.620954: step: 1938/1541, loss: 0.004257619380950928 2023-01-23 19:09:05.574514: step: 1940/1541, loss: 0.0002534484083298594 2023-01-23 19:09:06.516263: step: 1942/1541, loss: 0.0016108573181554675 2023-01-23 19:09:07.392577: step: 1944/1541, loss: 0.16544906795024872 2023-01-23 19:09:08.273649: step: 1946/1541, loss: 0.0010070448042824864 2023-01-23 19:09:09.132929: step: 1948/1541, loss: 0.0011284251231700182 2023-01-23 19:09:09.981521: step: 1950/1541, loss: 0.0005947837489657104 2023-01-23 19:09:10.856718: step: 1952/1541, loss: 0.10762914270162582 2023-01-23 19:09:11.745585: step: 1954/1541, loss: 0.0778413861989975 2023-01-23 19:09:12.645214: step: 1956/1541, loss: 0.00370365590788424 2023-01-23 19:09:13.529157: step: 1958/1541, loss: 0.004650984425097704 2023-01-23 19:09:14.468235: step: 1960/1541, loss: 0.13103362917900085 2023-01-23 19:09:15.346106: step: 1962/1541, loss: 0.002387415384873748 2023-01-23 19:09:16.262495: step: 1964/1541, loss: 0.0012739088851958513 2023-01-23 19:09:17.150180: step: 1966/1541, loss: 0.03660201653838158 2023-01-23 19:09:18.028967: step: 1968/1541, loss: 0.0013503599911928177 2023-01-23 19:09:18.914013: step: 1970/1541, loss: 0.004979690536856651 2023-01-23 19:09:19.791177: step: 1972/1541, loss: 0.0006073609692975879 2023-01-23 19:09:20.681292: step: 1974/1541, loss: 0.07156278192996979 2023-01-23 19:09:21.541528: step: 1976/1541, loss: 0.002298657549545169 2023-01-23 19:09:22.403682: step: 1978/1541, loss: 0.0016051535494625568 2023-01-23 19:09:23.293863: step: 1980/1541, loss: 0.10008499771356583 2023-01-23 19:09:24.192446: step: 1982/1541, loss: 0.014762374572455883 2023-01-23 19:09:25.080100: step: 1984/1541, loss: 0.004819625057280064 2023-01-23 19:09:25.991059: step: 1986/1541, loss: 0.2229979932308197 2023-01-23 19:09:26.981543: step: 1988/1541, loss: 0.1707465946674347 2023-01-23 19:09:27.875885: step: 1990/1541, loss: 0.041930049657821655 2023-01-23 19:09:28.789531: step: 1992/1541, loss: 0.2235928624868393 2023-01-23 19:09:29.691102: step: 1994/1541, loss: 0.02179146558046341 2023-01-23 19:09:30.572537: step: 1996/1541, loss: 0.000122302895761095 2023-01-23 19:09:31.468143: step: 1998/1541, loss: 0.09801207482814789 2023-01-23 19:09:32.363405: step: 2000/1541, loss: 0.0001315425761276856 2023-01-23 19:09:33.259582: step: 2002/1541, loss: 0.49499398469924927 2023-01-23 19:09:34.139795: step: 2004/1541, loss: 0.0071784392930567265 2023-01-23 19:09:35.075994: step: 2006/1541, loss: 0.13051070272922516 2023-01-23 19:09:35.948776: step: 2008/1541, loss: 0.14471301436424255 2023-01-23 19:09:36.836598: step: 2010/1541, loss: 0.10789690166711807 2023-01-23 19:09:37.703138: step: 2012/1541, loss: 0.0001053496016538702 2023-01-23 19:09:38.625500: step: 2014/1541, loss: 0.006011891178786755 2023-01-23 19:09:39.517433: step: 2016/1541, loss: 0.0007992051541805267 2023-01-23 19:09:40.407687: step: 2018/1541, loss: 0.0020180530846118927 2023-01-23 19:09:41.303578: step: 2020/1541, loss: 0.0007797240396030247 2023-01-23 19:09:42.194837: step: 2022/1541, loss: 0.0004078725178260356 2023-01-23 19:09:43.081758: step: 2024/1541, loss: 0.006661277264356613 2023-01-23 19:09:44.027713: step: 2026/1541, loss: 0.02218330278992653 2023-01-23 19:09:44.920257: step: 2028/1541, loss: 0.0006785230943933129 2023-01-23 19:09:45.826467: step: 2030/1541, loss: 0.01419571228325367 2023-01-23 19:09:46.722399: step: 2032/1541, loss: 0.03298937529325485 2023-01-23 19:09:47.600487: step: 2034/1541, loss: 0.0001274099777219817 2023-01-23 19:09:48.524143: step: 2036/1541, loss: 0.00887481402605772 2023-01-23 19:09:49.403244: step: 2038/1541, loss: 0.0002603020111564547 2023-01-23 19:09:50.273849: step: 2040/1541, loss: 0.07374908030033112 2023-01-23 19:09:51.170886: step: 2042/1541, loss: 0.05384300276637077 2023-01-23 19:09:52.080309: step: 2044/1541, loss: 0.0010987842688336968 2023-01-23 19:09:52.975735: step: 2046/1541, loss: 0.0009125038632191718 2023-01-23 19:09:53.864545: step: 2048/1541, loss: 0.24067191779613495 2023-01-23 19:09:54.774417: step: 2050/1541, loss: 0.1657118946313858 2023-01-23 19:09:55.698638: step: 2052/1541, loss: 0.00012467341730371118 2023-01-23 19:09:56.579597: step: 2054/1541, loss: 0.0036521940492093563 2023-01-23 19:09:57.538295: step: 2056/1541, loss: 0.005186883267015219 2023-01-23 19:09:58.408373: step: 2058/1541, loss: 0.004174421075731516 2023-01-23 19:09:59.311427: step: 2060/1541, loss: 0.35147625207901 2023-01-23 19:10:00.200211: step: 2062/1541, loss: 8.886377327144146e-05 2023-01-23 19:10:01.059338: step: 2064/1541, loss: 0.003726782975718379 2023-01-23 19:10:01.963325: step: 2066/1541, loss: 0.13330739736557007 2023-01-23 19:10:02.875212: step: 2068/1541, loss: 0.3903401792049408 2023-01-23 19:10:03.789763: step: 2070/1541, loss: 0.0007638748502358794 2023-01-23 19:10:04.657485: step: 2072/1541, loss: 0.4015355110168457 2023-01-23 19:10:05.549975: step: 2074/1541, loss: 0.01942872814834118 2023-01-23 19:10:06.449295: step: 2076/1541, loss: 0.006141613703221083 2023-01-23 19:10:07.370255: step: 2078/1541, loss: 0.1879260390996933 2023-01-23 19:10:08.241215: step: 2080/1541, loss: 0.000983453937806189 2023-01-23 19:10:09.115953: step: 2082/1541, loss: 0.10650278627872467 2023-01-23 19:10:09.989644: step: 2084/1541, loss: 0.025850309059023857 2023-01-23 19:10:10.869546: step: 2086/1541, loss: 0.0005718439933843911 2023-01-23 19:10:11.763482: step: 2088/1541, loss: 0.23785650730133057 2023-01-23 19:10:12.682812: step: 2090/1541, loss: 0.0013584236148744822 2023-01-23 19:10:13.568934: step: 2092/1541, loss: 0.0001626790181035176 2023-01-23 19:10:14.424910: step: 2094/1541, loss: 0.0006919180159457028 2023-01-23 19:10:15.321017: step: 2096/1541, loss: 8.734890434425324e-05 2023-01-23 19:10:16.267576: step: 2098/1541, loss: 0.004727391991764307 2023-01-23 19:10:17.227938: step: 2100/1541, loss: 0.003874905174598098 2023-01-23 19:10:18.089571: step: 2102/1541, loss: 0.004555603489279747 2023-01-23 19:10:18.976269: step: 2104/1541, loss: 0.0004914343007840216 2023-01-23 19:10:19.893389: step: 2106/1541, loss: 0.008997132070362568 2023-01-23 19:10:20.823542: step: 2108/1541, loss: 0.03420455753803253 2023-01-23 19:10:21.704008: step: 2110/1541, loss: 0.1775713562965393 2023-01-23 19:10:22.574065: step: 2112/1541, loss: 8.566216274630278e-05 2023-01-23 19:10:23.504751: step: 2114/1541, loss: 0.0002386751293670386 2023-01-23 19:10:24.391055: step: 2116/1541, loss: 0.0006354121724143624 2023-01-23 19:10:25.270007: step: 2118/1541, loss: 7.438593456754461e-05 2023-01-23 19:10:26.162015: step: 2120/1541, loss: 0.0350695438683033 2023-01-23 19:10:27.026325: step: 2122/1541, loss: 0.008704813197255135 2023-01-23 19:10:27.891956: step: 2124/1541, loss: 0.18292413651943207 2023-01-23 19:10:28.765068: step: 2126/1541, loss: 2.0444342681003036e-06 2023-01-23 19:10:29.735600: step: 2128/1541, loss: 0.22701753675937653 2023-01-23 19:10:30.699787: step: 2130/1541, loss: 0.0022175502963364124 2023-01-23 19:10:31.604370: step: 2132/1541, loss: 0.0001247942418558523 2023-01-23 19:10:32.518137: step: 2134/1541, loss: 0.00016158525249920785 2023-01-23 19:10:33.426421: step: 2136/1541, loss: 0.08812937140464783 2023-01-23 19:10:34.334816: step: 2138/1541, loss: 0.15188005566596985 2023-01-23 19:10:35.203824: step: 2140/1541, loss: 3.553542046574876e-05 2023-01-23 19:10:36.085761: step: 2142/1541, loss: 0.00012117921869503334 2023-01-23 19:10:36.983058: step: 2144/1541, loss: 0.22219255566596985 2023-01-23 19:10:37.894118: step: 2146/1541, loss: 0.04222751036286354 2023-01-23 19:10:38.764563: step: 2148/1541, loss: 0.0006764030549675226 2023-01-23 19:10:39.668519: step: 2150/1541, loss: 0.016769876703619957 2023-01-23 19:10:40.571321: step: 2152/1541, loss: 0.00027009446057491004 2023-01-23 19:10:41.431192: step: 2154/1541, loss: 0.045905470848083496 2023-01-23 19:10:42.329866: step: 2156/1541, loss: 0.37380605936050415 2023-01-23 19:10:43.242763: step: 2158/1541, loss: 0.3504713177680969 2023-01-23 19:10:44.135125: step: 2160/1541, loss: 0.6690375208854675 2023-01-23 19:10:45.061935: step: 2162/1541, loss: 0.012936299666762352 2023-01-23 19:10:45.937268: step: 2164/1541, loss: 0.1121324822306633 2023-01-23 19:10:46.868809: step: 2166/1541, loss: 0.05487419664859772 2023-01-23 19:10:47.783310: step: 2168/1541, loss: 0.10547357052564621 2023-01-23 19:10:48.716698: step: 2170/1541, loss: 0.5430811047554016 2023-01-23 19:10:49.600778: step: 2172/1541, loss: 0.01050642691552639 2023-01-23 19:10:50.508934: step: 2174/1541, loss: 0.00968796294182539 2023-01-23 19:10:51.399629: step: 2176/1541, loss: 0.00017619563732296228 2023-01-23 19:10:52.298026: step: 2178/1541, loss: 0.2154679298400879 2023-01-23 19:10:53.169845: step: 2180/1541, loss: 0.012520242482423782 2023-01-23 19:10:54.082493: step: 2182/1541, loss: 0.0013913526199758053 2023-01-23 19:10:54.964337: step: 2184/1541, loss: 0.003181586042046547 2023-01-23 19:10:55.846360: step: 2186/1541, loss: 0.16941821575164795 2023-01-23 19:10:56.835325: step: 2188/1541, loss: 0.008153421804308891 2023-01-23 19:10:57.726744: step: 2190/1541, loss: 0.006759459152817726 2023-01-23 19:10:58.687598: step: 2192/1541, loss: 0.04144582152366638 2023-01-23 19:10:59.589692: step: 2194/1541, loss: 0.0029061362147331238 2023-01-23 19:11:00.548486: step: 2196/1541, loss: 0.012294607236981392 2023-01-23 19:11:01.400166: step: 2198/1541, loss: 3.500023376545869e-05 2023-01-23 19:11:02.277943: step: 2200/1541, loss: 0.38660672307014465 2023-01-23 19:11:03.215032: step: 2202/1541, loss: 0.03275272622704506 2023-01-23 19:11:04.085393: step: 2204/1541, loss: 0.00024146516807377338 2023-01-23 19:11:04.987422: step: 2206/1541, loss: 1.0328764915466309 2023-01-23 19:11:05.905624: step: 2208/1541, loss: 0.18927964568138123 2023-01-23 19:11:06.784210: step: 2210/1541, loss: 0.29447102546691895 2023-01-23 19:11:07.753288: step: 2212/1541, loss: 0.1106187105178833 2023-01-23 19:11:08.651037: step: 2214/1541, loss: 0.0014419483486562967 2023-01-23 19:11:09.536282: step: 2216/1541, loss: 0.005046421196311712 2023-01-23 19:11:10.478643: step: 2218/1541, loss: 0.0359308123588562 2023-01-23 19:11:11.408671: step: 2220/1541, loss: 0.004074673168361187 2023-01-23 19:11:12.339358: step: 2222/1541, loss: 0.0017109641339629889 2023-01-23 19:11:13.222044: step: 2224/1541, loss: 0.4591328203678131 2023-01-23 19:11:14.127420: step: 2226/1541, loss: 0.22774019837379456 2023-01-23 19:11:14.972363: step: 2228/1541, loss: 0.0005724327638745308 2023-01-23 19:11:15.884002: step: 2230/1541, loss: 0.7092633247375488 2023-01-23 19:11:16.802795: step: 2232/1541, loss: 0.07966399192810059 2023-01-23 19:11:17.676082: step: 2234/1541, loss: 0.01106156874448061 2023-01-23 19:11:18.574644: step: 2236/1541, loss: 0.002034718869253993 2023-01-23 19:11:19.530473: step: 2238/1541, loss: 0.0010459180921316147 2023-01-23 19:11:20.435983: step: 2240/1541, loss: 0.18826347589492798 2023-01-23 19:11:21.288584: step: 2242/1541, loss: 0.019176142290234566 2023-01-23 19:11:22.134910: step: 2244/1541, loss: 0.0027479021809995174 2023-01-23 19:11:23.016006: step: 2246/1541, loss: 0.10093491524457932 2023-01-23 19:11:23.925666: step: 2248/1541, loss: 0.07218790054321289 2023-01-23 19:11:24.874846: step: 2250/1541, loss: 0.016384610906243324 2023-01-23 19:11:25.736457: step: 2252/1541, loss: 0.005880022421479225 2023-01-23 19:11:26.662012: step: 2254/1541, loss: 0.01265286561101675 2023-01-23 19:11:27.606150: step: 2256/1541, loss: 0.0020499820820987225 2023-01-23 19:11:28.537153: step: 2258/1541, loss: 0.006755335256457329 2023-01-23 19:11:29.478914: step: 2260/1541, loss: 0.41990599036216736 2023-01-23 19:11:30.343772: step: 2262/1541, loss: 8.730431000003591e-05 2023-01-23 19:11:31.220514: step: 2264/1541, loss: 0.01153340470045805 2023-01-23 19:11:32.161079: step: 2266/1541, loss: 0.1424867957830429 2023-01-23 19:11:33.021994: step: 2268/1541, loss: 0.14236600697040558 2023-01-23 19:11:33.918513: step: 2270/1541, loss: 0.00020842309459112585 2023-01-23 19:11:34.809446: step: 2272/1541, loss: 0.4863755404949188 2023-01-23 19:11:35.715182: step: 2274/1541, loss: 0.012280279770493507 2023-01-23 19:11:36.629186: step: 2276/1541, loss: 0.0043357620015740395 2023-01-23 19:11:37.528161: step: 2278/1541, loss: 0.0007679163245484233 2023-01-23 19:11:38.421159: step: 2280/1541, loss: 0.0002588460920378566 2023-01-23 19:11:39.321495: step: 2282/1541, loss: 0.023469462990760803 2023-01-23 19:11:40.246257: step: 2284/1541, loss: 0.004590969067066908 2023-01-23 19:11:41.139519: step: 2286/1541, loss: 0.1144055724143982 2023-01-23 19:11:42.059647: step: 2288/1541, loss: 0.006514562293887138 2023-01-23 19:11:42.969892: step: 2290/1541, loss: 0.0793546810746193 2023-01-23 19:11:43.858877: step: 2292/1541, loss: 0.01660500466823578 2023-01-23 19:11:44.730566: step: 2294/1541, loss: 0.00037935227737762034 2023-01-23 19:11:45.575458: step: 2296/1541, loss: 0.0010099587962031364 2023-01-23 19:11:46.511225: step: 2298/1541, loss: 0.0002968019980471581 2023-01-23 19:11:47.385097: step: 2300/1541, loss: 8.04563969722949e-05 2023-01-23 19:11:48.295211: step: 2302/1541, loss: 0.00010406671208329499 2023-01-23 19:11:49.174619: step: 2304/1541, loss: 0.04600522294640541 2023-01-23 19:11:50.106093: step: 2306/1541, loss: 0.07668125629425049 2023-01-23 19:11:50.990605: step: 2308/1541, loss: 0.0032958018127828836 2023-01-23 19:11:51.898386: step: 2310/1541, loss: 0.00034556229365989566 2023-01-23 19:11:52.806927: step: 2312/1541, loss: 0.35591816902160645 2023-01-23 19:11:53.707446: step: 2314/1541, loss: 0.251150906085968 2023-01-23 19:11:54.652492: step: 2316/1541, loss: 0.881886899471283 2023-01-23 19:11:55.636823: step: 2318/1541, loss: 0.29527732729911804 2023-01-23 19:11:56.513574: step: 2320/1541, loss: 0.0018380254041403532 2023-01-23 19:11:57.378723: step: 2322/1541, loss: 0.00036336941411718726 2023-01-23 19:11:58.253874: step: 2324/1541, loss: 0.0231272391974926 2023-01-23 19:11:59.145947: step: 2326/1541, loss: 0.0313018299639225 2023-01-23 19:12:00.065054: step: 2328/1541, loss: 0.010974327102303505 2023-01-23 19:12:00.927586: step: 2330/1541, loss: 0.24550139904022217 2023-01-23 19:12:01.850648: step: 2332/1541, loss: 0.004225886892527342 2023-01-23 19:12:02.759901: step: 2334/1541, loss: 0.0020051756873726845 2023-01-23 19:12:03.673137: step: 2336/1541, loss: 0.1432708501815796 2023-01-23 19:12:04.557062: step: 2338/1541, loss: 1.8264057871419936e-05 2023-01-23 19:12:05.504080: step: 2340/1541, loss: 0.13416297733783722 2023-01-23 19:12:06.408191: step: 2342/1541, loss: 0.055596236139535904 2023-01-23 19:12:07.283139: step: 2344/1541, loss: 0.004482504911720753 2023-01-23 19:12:08.176462: step: 2346/1541, loss: 0.0376172736287117 2023-01-23 19:12:09.107857: step: 2348/1541, loss: 0.009636964648962021 2023-01-23 19:12:10.002626: step: 2350/1541, loss: 0.029913775622844696 2023-01-23 19:12:10.933321: step: 2352/1541, loss: 0.0011647427454590797 2023-01-23 19:12:11.815738: step: 2354/1541, loss: 0.0007335279369726777 2023-01-23 19:12:12.693802: step: 2356/1541, loss: 0.0007017824100330472 2023-01-23 19:12:13.561323: step: 2358/1541, loss: 0.001795999240130186 2023-01-23 19:12:14.457703: step: 2360/1541, loss: 0.055653832852840424 2023-01-23 19:12:15.330847: step: 2362/1541, loss: 5.899165626033209e-05 2023-01-23 19:12:16.243891: step: 2364/1541, loss: 0.012123982422053814 2023-01-23 19:12:17.231129: step: 2366/1541, loss: 0.00035356602165848017 2023-01-23 19:12:18.152017: step: 2368/1541, loss: 0.05782677233219147 2023-01-23 19:12:19.125945: step: 2370/1541, loss: 0.040247730910778046 2023-01-23 19:12:20.024536: step: 2372/1541, loss: 0.029754430055618286 2023-01-23 19:12:20.886144: step: 2374/1541, loss: 0.0015450575156137347 2023-01-23 19:12:21.803718: step: 2376/1541, loss: 0.11622966080904007 2023-01-23 19:12:22.758992: step: 2378/1541, loss: 0.523967981338501 2023-01-23 19:12:23.658377: step: 2380/1541, loss: 0.000395788811147213 2023-01-23 19:12:24.548790: step: 2382/1541, loss: 0.05294426530599594 2023-01-23 19:12:25.437006: step: 2384/1541, loss: 0.0011320251505821943 2023-01-23 19:12:26.345383: step: 2386/1541, loss: 0.0007858938770368695 2023-01-23 19:12:27.269017: step: 2388/1541, loss: 0.2675539553165436 2023-01-23 19:12:28.288412: step: 2390/1541, loss: 0.0005688412929885089 2023-01-23 19:12:29.178091: step: 2392/1541, loss: 0.0006392563227564096 2023-01-23 19:12:30.032665: step: 2394/1541, loss: 0.0001221293641719967 2023-01-23 19:12:30.929947: step: 2396/1541, loss: 0.004989506211131811 2023-01-23 19:12:31.883503: step: 2398/1541, loss: 0.00024183292407542467 2023-01-23 19:12:32.765164: step: 2400/1541, loss: 0.06294256448745728 2023-01-23 19:12:33.655520: step: 2402/1541, loss: 0.0008615129627287388 2023-01-23 19:12:34.566103: step: 2404/1541, loss: 0.021852442994713783 2023-01-23 19:12:35.465034: step: 2406/1541, loss: 4.10326974815689e-05 2023-01-23 19:12:36.358422: step: 2408/1541, loss: 0.00042121042497456074 2023-01-23 19:12:37.309464: step: 2410/1541, loss: 0.00016366419731639326 2023-01-23 19:12:38.250168: step: 2412/1541, loss: 0.27188974618911743 2023-01-23 19:12:39.124896: step: 2414/1541, loss: 0.016609974205493927 2023-01-23 19:12:40.046745: step: 2416/1541, loss: 0.499693363904953 2023-01-23 19:12:40.901250: step: 2418/1541, loss: 0.0020542754791677 2023-01-23 19:12:41.770308: step: 2420/1541, loss: 0.0884384885430336 2023-01-23 19:12:42.645270: step: 2422/1541, loss: 0.0006999325705692172 2023-01-23 19:12:43.502815: step: 2424/1541, loss: 0.00010182873666053638 2023-01-23 19:12:44.382007: step: 2426/1541, loss: 0.00015059576253406703 2023-01-23 19:12:45.260256: step: 2428/1541, loss: 0.011717393063008785 2023-01-23 19:12:46.127544: step: 2430/1541, loss: 0.0013244581641629338 2023-01-23 19:12:47.001715: step: 2432/1541, loss: 0.0004090913862455636 2023-01-23 19:12:47.915470: step: 2434/1541, loss: 0.00038342937477864325 2023-01-23 19:12:48.773402: step: 2436/1541, loss: 0.00218523433431983 2023-01-23 19:12:49.646101: step: 2438/1541, loss: 0.0002611567615531385 2023-01-23 19:12:50.569407: step: 2440/1541, loss: 0.0056921313516795635 2023-01-23 19:12:51.485850: step: 2442/1541, loss: 2.490782026143279e-05 2023-01-23 19:12:52.403015: step: 2444/1541, loss: 0.35125699639320374 2023-01-23 19:12:53.301100: step: 2446/1541, loss: 0.022426433861255646 2023-01-23 19:12:54.246158: step: 2448/1541, loss: 0.26968592405319214 2023-01-23 19:12:55.164167: step: 2450/1541, loss: 0.11577856540679932 2023-01-23 19:12:56.063983: step: 2452/1541, loss: 0.08951006829738617 2023-01-23 19:12:56.941141: step: 2454/1541, loss: 0.01760815642774105 2023-01-23 19:12:57.868407: step: 2456/1541, loss: 0.07062698155641556 2023-01-23 19:12:58.752453: step: 2458/1541, loss: 0.0024232598952949047 2023-01-23 19:12:59.684547: step: 2460/1541, loss: 0.00010670772462617606 2023-01-23 19:13:00.584578: step: 2462/1541, loss: 0.10511425137519836 2023-01-23 19:13:01.463442: step: 2464/1541, loss: 0.11663831025362015 2023-01-23 19:13:02.350951: step: 2466/1541, loss: 0.001433780649676919 2023-01-23 19:13:03.253691: step: 2468/1541, loss: 0.0013324671890586615 2023-01-23 19:13:04.217524: step: 2470/1541, loss: 0.08400089293718338 2023-01-23 19:13:05.079350: step: 2472/1541, loss: 0.7222298383712769 2023-01-23 19:13:05.925099: step: 2474/1541, loss: 0.019900182262063026 2023-01-23 19:13:06.879543: step: 2476/1541, loss: 0.7091194987297058 2023-01-23 19:13:07.794700: step: 2478/1541, loss: 0.0025844089686870575 2023-01-23 19:13:08.691087: step: 2480/1541, loss: 0.02552756480872631 2023-01-23 19:13:09.580857: step: 2482/1541, loss: 0.007651653606444597 2023-01-23 19:13:10.448972: step: 2484/1541, loss: 0.04635840281844139 2023-01-23 19:13:11.343203: step: 2486/1541, loss: 0.0010735993273556232 2023-01-23 19:13:12.223708: step: 2488/1541, loss: 0.0037633765023201704 2023-01-23 19:13:13.146691: step: 2490/1541, loss: 0.0002146955521311611 2023-01-23 19:13:14.025343: step: 2492/1541, loss: 0.00104005157481879 2023-01-23 19:13:15.018801: step: 2494/1541, loss: 0.05442185327410698 2023-01-23 19:13:15.970994: step: 2496/1541, loss: 0.2837851047515869 2023-01-23 19:13:16.849158: step: 2498/1541, loss: 0.026440395042300224 2023-01-23 19:13:17.738929: step: 2500/1541, loss: 0.04527684673666954 2023-01-23 19:13:18.637499: step: 2502/1541, loss: 0.0003352368948981166 2023-01-23 19:13:19.512977: step: 2504/1541, loss: 0.03273975849151611 2023-01-23 19:13:20.447794: step: 2506/1541, loss: 0.0014465436106547713 2023-01-23 19:13:21.320780: step: 2508/1541, loss: 0.009625301696360111 2023-01-23 19:13:22.245291: step: 2510/1541, loss: 0.00031344976741820574 2023-01-23 19:13:23.275848: step: 2512/1541, loss: 0.011722099035978317 2023-01-23 19:13:24.162333: step: 2514/1541, loss: 0.09809228777885437 2023-01-23 19:13:25.067769: step: 2516/1541, loss: 0.07406605035066605 2023-01-23 19:13:25.945442: step: 2518/1541, loss: 0.04539290443062782 2023-01-23 19:13:26.846348: step: 2520/1541, loss: 0.0813581570982933 2023-01-23 19:13:27.778684: step: 2522/1541, loss: 0.012888411991298199 2023-01-23 19:13:28.711880: step: 2524/1541, loss: 0.0397169254720211 2023-01-23 19:13:29.632484: step: 2526/1541, loss: 0.00015032374358270317 2023-01-23 19:13:30.588299: step: 2528/1541, loss: 0.0022198467049747705 2023-01-23 19:13:31.460961: step: 2530/1541, loss: 2.8324870072538033e-05 2023-01-23 19:13:32.362017: step: 2532/1541, loss: 0.00042307560215704143 2023-01-23 19:13:33.219912: step: 2534/1541, loss: 0.0028872766997665167 2023-01-23 19:13:34.059526: step: 2536/1541, loss: 0.3006042540073395 2023-01-23 19:13:34.961963: step: 2538/1541, loss: 0.005800372920930386 2023-01-23 19:13:35.859806: step: 2540/1541, loss: 0.20467858016490936 2023-01-23 19:13:36.746906: step: 2542/1541, loss: 0.000373381917597726 2023-01-23 19:13:37.613476: step: 2544/1541, loss: 0.01504158228635788 2023-01-23 19:13:38.484687: step: 2546/1541, loss: 0.00016318599227815866 2023-01-23 19:13:39.378423: step: 2548/1541, loss: 0.0001383874041493982 2023-01-23 19:13:40.285643: step: 2550/1541, loss: 0.04037104547023773 2023-01-23 19:13:41.176130: step: 2552/1541, loss: 0.0002471599727869034 2023-01-23 19:13:42.049316: step: 2554/1541, loss: 0.008715265430510044 2023-01-23 19:13:42.943012: step: 2556/1541, loss: 0.018454870209097862 2023-01-23 19:13:43.838841: step: 2558/1541, loss: 0.37830325961112976 2023-01-23 19:13:44.710305: step: 2560/1541, loss: 0.1730293333530426 2023-01-23 19:13:45.603302: step: 2562/1541, loss: 0.0003109350800514221 2023-01-23 19:13:46.496427: step: 2564/1541, loss: 0.00028467009542509913 2023-01-23 19:13:47.404575: step: 2566/1541, loss: 0.015645824372768402 2023-01-23 19:13:48.321266: step: 2568/1541, loss: 0.04624829441308975 2023-01-23 19:13:49.214159: step: 2570/1541, loss: 0.09679853916168213 2023-01-23 19:13:50.116858: step: 2572/1541, loss: 0.009155020117759705 2023-01-23 19:13:50.991103: step: 2574/1541, loss: 0.06214912608265877 2023-01-23 19:13:51.888963: step: 2576/1541, loss: 0.0008760288474150002 2023-01-23 19:13:52.779241: step: 2578/1541, loss: 0.1458345651626587 2023-01-23 19:13:53.661141: step: 2580/1541, loss: 0.0002998587442561984 2023-01-23 19:13:54.550869: step: 2582/1541, loss: 0.02987804263830185 2023-01-23 19:13:55.474447: step: 2584/1541, loss: 0.06776571273803711 2023-01-23 19:13:56.396961: step: 2586/1541, loss: 0.021583247929811478 2023-01-23 19:13:57.291236: step: 2588/1541, loss: 0.023271769285202026 2023-01-23 19:13:58.190050: step: 2590/1541, loss: 0.0025735273957252502 2023-01-23 19:13:59.065095: step: 2592/1541, loss: 0.0015215830644592643 2023-01-23 19:13:59.956150: step: 2594/1541, loss: 0.0026569024194031954 2023-01-23 19:14:00.832505: step: 2596/1541, loss: 0.0002262130583403632 2023-01-23 19:14:01.823179: step: 2598/1541, loss: 0.05407854542136192 2023-01-23 19:14:02.690027: step: 2600/1541, loss: 0.14256753027439117 2023-01-23 19:14:03.568393: step: 2602/1541, loss: 0.07535062730312347 2023-01-23 19:14:04.475519: step: 2604/1541, loss: 0.002582205692306161 2023-01-23 19:14:05.387114: step: 2606/1541, loss: 0.022913189604878426 2023-01-23 19:14:06.280419: step: 2608/1541, loss: 0.16974292695522308 2023-01-23 19:14:07.140690: step: 2610/1541, loss: 0.006545140407979488 2023-01-23 19:14:08.066426: step: 2612/1541, loss: 0.0009313920163549483 2023-01-23 19:14:09.013597: step: 2614/1541, loss: 0.0009010875364765525 2023-01-23 19:14:09.895138: step: 2616/1541, loss: 0.029247447848320007 2023-01-23 19:14:10.814855: step: 2618/1541, loss: 0.02242494747042656 2023-01-23 19:14:11.799125: step: 2620/1541, loss: 0.00039076668326742947 2023-01-23 19:14:12.661821: step: 2622/1541, loss: 0.11894520372152328 2023-01-23 19:14:13.563461: step: 2624/1541, loss: 0.0005927286110818386 2023-01-23 19:14:14.432534: step: 2626/1541, loss: 0.0006060004816390574 2023-01-23 19:14:15.295515: step: 2628/1541, loss: 0.10596288740634918 2023-01-23 19:14:16.214033: step: 2630/1541, loss: 0.037326667457818985 2023-01-23 19:14:17.080606: step: 2632/1541, loss: 0.07610056549310684 2023-01-23 19:14:17.986284: step: 2634/1541, loss: 0.06642400473356247 2023-01-23 19:14:18.916053: step: 2636/1541, loss: 0.00025877621374092996 2023-01-23 19:14:19.785076: step: 2638/1541, loss: 0.00027097942074760795 2023-01-23 19:14:20.684931: step: 2640/1541, loss: 0.16581076383590698 2023-01-23 19:14:21.596364: step: 2642/1541, loss: 0.007158711086958647 2023-01-23 19:14:22.473788: step: 2644/1541, loss: 3.616480898926966e-05 2023-01-23 19:14:23.403258: step: 2646/1541, loss: 2.1613523131236434e-05 2023-01-23 19:14:24.304997: step: 2648/1541, loss: 0.14643165469169617 2023-01-23 19:14:25.181456: step: 2650/1541, loss: 0.18324443697929382 2023-01-23 19:14:26.104530: step: 2652/1541, loss: 0.0012123704655095935 2023-01-23 19:14:26.958906: step: 2654/1541, loss: 0.0011356602190062404 2023-01-23 19:14:27.848528: step: 2656/1541, loss: 0.00043945974903181195 2023-01-23 19:14:28.705166: step: 2658/1541, loss: 0.00017383141675963998 2023-01-23 19:14:29.679171: step: 2660/1541, loss: 0.20444932579994202 2023-01-23 19:14:30.569417: step: 2662/1541, loss: 1.1908403394045308e-05 2023-01-23 19:14:31.517035: step: 2664/1541, loss: 0.003744517918676138 2023-01-23 19:14:32.376645: step: 2666/1541, loss: 0.07670686393976212 2023-01-23 19:14:33.259343: step: 2668/1541, loss: 0.05917711183428764 2023-01-23 19:14:34.125554: step: 2670/1541, loss: 0.0002094489464070648 2023-01-23 19:14:34.998957: step: 2672/1541, loss: 0.08430355787277222 2023-01-23 19:14:35.895144: step: 2674/1541, loss: 0.0030191668774932623 2023-01-23 19:14:36.847633: step: 2676/1541, loss: 0.006105050444602966 2023-01-23 19:14:37.736394: step: 2678/1541, loss: 0.08187217265367508 2023-01-23 19:14:38.615034: step: 2680/1541, loss: 0.10411862283945084 2023-01-23 19:14:39.512817: step: 2682/1541, loss: 8.078296377789229e-05 2023-01-23 19:14:40.425145: step: 2684/1541, loss: 0.25079721212387085 2023-01-23 19:14:41.305207: step: 2686/1541, loss: 0.02266809716820717 2023-01-23 19:14:42.215781: step: 2688/1541, loss: 0.01073444727808237 2023-01-23 19:14:43.080440: step: 2690/1541, loss: 0.040169961750507355 2023-01-23 19:14:43.979924: step: 2692/1541, loss: 0.0009607939864508808 2023-01-23 19:14:44.842887: step: 2694/1541, loss: 0.0010674456134438515 2023-01-23 19:14:45.748179: step: 2696/1541, loss: 0.03630829229950905 2023-01-23 19:14:46.674880: step: 2698/1541, loss: 0.011733686551451683 2023-01-23 19:14:47.582834: step: 2700/1541, loss: 0.04238433390855789 2023-01-23 19:14:48.423399: step: 2702/1541, loss: 0.0005455065402202308 2023-01-23 19:14:49.281303: step: 2704/1541, loss: 0.0057714818976819515 2023-01-23 19:14:50.115279: step: 2706/1541, loss: 0.0006836959510110319 2023-01-23 19:14:51.004156: step: 2708/1541, loss: 0.1390453428030014 2023-01-23 19:14:51.969894: step: 2710/1541, loss: 0.1636328399181366 2023-01-23 19:14:52.863388: step: 2712/1541, loss: 0.04507429525256157 2023-01-23 19:14:53.799988: step: 2714/1541, loss: 0.01248463336378336 2023-01-23 19:14:54.669737: step: 2716/1541, loss: 0.0006792958010919392 2023-01-23 19:14:55.548357: step: 2718/1541, loss: 0.0007380973547697067 2023-01-23 19:14:56.462187: step: 2720/1541, loss: 0.003668026067316532 2023-01-23 19:14:57.373558: step: 2722/1541, loss: 0.002380253281444311 2023-01-23 19:14:58.269937: step: 2724/1541, loss: 0.03436939790844917 2023-01-23 19:14:59.141005: step: 2726/1541, loss: 3.0529892683262005e-05 2023-01-23 19:15:00.007516: step: 2728/1541, loss: 0.021148810163140297 2023-01-23 19:15:00.915287: step: 2730/1541, loss: 0.04207135736942291 2023-01-23 19:15:01.848801: step: 2732/1541, loss: 0.0002863744448404759 2023-01-23 19:15:02.764241: step: 2734/1541, loss: 0.04476859048008919 2023-01-23 19:15:03.678224: step: 2736/1541, loss: 0.029977522790431976 2023-01-23 19:15:04.588990: step: 2738/1541, loss: 0.008698347955942154 2023-01-23 19:15:05.486099: step: 2740/1541, loss: 0.006534046493470669 2023-01-23 19:15:06.370669: step: 2742/1541, loss: 0.0008324815426021814 2023-01-23 19:15:07.266177: step: 2744/1541, loss: 0.0012110361130908132 2023-01-23 19:15:08.138660: step: 2746/1541, loss: 0.01823766715824604 2023-01-23 19:15:09.052401: step: 2748/1541, loss: 0.0007440869230777025 2023-01-23 19:15:09.930149: step: 2750/1541, loss: 0.0002108181652147323 2023-01-23 19:15:10.915941: step: 2752/1541, loss: 0.028070101514458656 2023-01-23 19:15:11.775131: step: 2754/1541, loss: 0.0012587697710841894 2023-01-23 19:15:12.658837: step: 2756/1541, loss: 0.4745275676250458 2023-01-23 19:15:13.606921: step: 2758/1541, loss: 0.0050104521214962006 2023-01-23 19:15:14.525545: step: 2760/1541, loss: 0.001162068685516715 2023-01-23 19:15:15.387622: step: 2762/1541, loss: 0.0009538923623040318 2023-01-23 19:15:16.280757: step: 2764/1541, loss: 0.0017415593611076474 2023-01-23 19:15:17.221000: step: 2766/1541, loss: 0.33154749870300293 2023-01-23 19:15:18.150803: step: 2768/1541, loss: 0.0008892021141946316 2023-01-23 19:15:19.008483: step: 2770/1541, loss: 0.061937496066093445 2023-01-23 19:15:19.932118: step: 2772/1541, loss: 0.09748966246843338 2023-01-23 19:15:20.815315: step: 2774/1541, loss: 0.00944150798022747 2023-01-23 19:15:21.710360: step: 2776/1541, loss: 0.06353820115327835 2023-01-23 19:15:22.610385: step: 2778/1541, loss: 0.0014508216409012675 2023-01-23 19:15:23.494511: step: 2780/1541, loss: 6.53552997391671e-05 2023-01-23 19:15:24.392074: step: 2782/1541, loss: 0.00029037348576821387 2023-01-23 19:15:25.305342: step: 2784/1541, loss: 0.00013781491725239903 2023-01-23 19:15:26.184177: step: 2786/1541, loss: 0.25454181432724 2023-01-23 19:15:27.059650: step: 2788/1541, loss: 0.016751887276768684 2023-01-23 19:15:28.023409: step: 2790/1541, loss: 7.888893014751375e-05 2023-01-23 19:15:28.894588: step: 2792/1541, loss: 0.0012363082496449351 2023-01-23 19:15:29.818502: step: 2794/1541, loss: 0.0017343206563964486 2023-01-23 19:15:30.731900: step: 2796/1541, loss: 0.042040999978780746 2023-01-23 19:15:31.612074: step: 2798/1541, loss: 0.0024288238491863012 2023-01-23 19:15:32.529304: step: 2800/1541, loss: 0.2689690887928009 2023-01-23 19:15:33.430332: step: 2802/1541, loss: 0.10161226987838745 2023-01-23 19:15:34.328250: step: 2804/1541, loss: 3.601637945394032e-05 2023-01-23 19:15:35.189909: step: 2806/1541, loss: -2.20281035581138e-05 2023-01-23 19:15:36.083752: step: 2808/1541, loss: 0.0016291242791339755 2023-01-23 19:15:36.966582: step: 2810/1541, loss: 0.2509758770465851 2023-01-23 19:15:37.849534: step: 2812/1541, loss: 0.0038055931217968464 2023-01-23 19:15:38.777120: step: 2814/1541, loss: 0.06635815650224686 2023-01-23 19:15:39.595746: step: 2816/1541, loss: 0.031077668070793152 2023-01-23 19:15:40.504556: step: 2818/1541, loss: 0.039175864309072495 2023-01-23 19:15:41.425124: step: 2820/1541, loss: 0.049862075597047806 2023-01-23 19:15:42.340518: step: 2822/1541, loss: 0.0003107521333731711 2023-01-23 19:15:43.406510: step: 2824/1541, loss: 0.004558094777166843 2023-01-23 19:15:44.310905: step: 2826/1541, loss: 0.0003691230667755008 2023-01-23 19:15:45.173826: step: 2828/1541, loss: 0.001252168556675315 2023-01-23 19:15:46.071070: step: 2830/1541, loss: 0.01998029462993145 2023-01-23 19:15:46.968905: step: 2832/1541, loss: 0.017175227403640747 2023-01-23 19:15:47.874202: step: 2834/1541, loss: 0.13113419711589813 2023-01-23 19:15:48.762805: step: 2836/1541, loss: 0.19497285783290863 2023-01-23 19:15:49.648443: step: 2838/1541, loss: 0.001217524753883481 2023-01-23 19:15:50.567813: step: 2840/1541, loss: 0.0002802888920996338 2023-01-23 19:15:51.564111: step: 2842/1541, loss: 0.1245177760720253 2023-01-23 19:15:52.492653: step: 2844/1541, loss: 0.00015812648052815348 2023-01-23 19:15:53.401548: step: 2846/1541, loss: 0.030338283628225327 2023-01-23 19:15:54.283314: step: 2848/1541, loss: 0.0036032311618328094 2023-01-23 19:15:55.137055: step: 2850/1541, loss: 4.8476282245246693e-05 2023-01-23 19:15:56.019668: step: 2852/1541, loss: 0.001980955945327878 2023-01-23 19:15:57.032785: step: 2854/1541, loss: 0.004544090945273638 2023-01-23 19:15:57.914504: step: 2856/1541, loss: 7.854994328226894e-05 2023-01-23 19:15:58.817434: step: 2858/1541, loss: 0.000289932475425303 2023-01-23 19:15:59.768989: step: 2860/1541, loss: 0.0010406597284600139 2023-01-23 19:16:00.628479: step: 2862/1541, loss: 0.011888200417160988 2023-01-23 19:16:01.615249: step: 2864/1541, loss: 0.0001746994093991816 2023-01-23 19:16:02.526038: step: 2866/1541, loss: 0.00196938868612051 2023-01-23 19:16:03.427789: step: 2868/1541, loss: 0.00019875381258316338 2023-01-23 19:16:04.296052: step: 2870/1541, loss: 0.00020970289187971503 2023-01-23 19:16:05.210621: step: 2872/1541, loss: 0.006764139514416456 2023-01-23 19:16:06.103012: step: 2874/1541, loss: 0.7480329275131226 2023-01-23 19:16:06.993909: step: 2876/1541, loss: 0.04751351475715637 2023-01-23 19:16:07.875229: step: 2878/1541, loss: 0.004060591571033001 2023-01-23 19:16:08.760616: step: 2880/1541, loss: 1.0895981788635254 2023-01-23 19:16:09.668952: step: 2882/1541, loss: 0.3335067927837372 2023-01-23 19:16:10.613850: step: 2884/1541, loss: 0.0006897274288348854 2023-01-23 19:16:11.494200: step: 2886/1541, loss: 0.00013144113472662866 2023-01-23 19:16:12.441885: step: 2888/1541, loss: 0.4365535378456116 2023-01-23 19:16:13.323599: step: 2890/1541, loss: 0.003362407675012946 2023-01-23 19:16:14.226339: step: 2892/1541, loss: 0.0003171642601955682 2023-01-23 19:16:15.141199: step: 2894/1541, loss: 0.15373988449573517 2023-01-23 19:16:16.043595: step: 2896/1541, loss: 0.0021269330754876137 2023-01-23 19:16:16.951483: step: 2898/1541, loss: 0.001756128971464932 2023-01-23 19:16:17.835632: step: 2900/1541, loss: 0.23488685488700867 2023-01-23 19:16:18.715435: step: 2902/1541, loss: 0.004102866631001234 2023-01-23 19:16:19.580411: step: 2904/1541, loss: 0.0009082413744181395 2023-01-23 19:16:20.482964: step: 2906/1541, loss: 0.004100542049854994 2023-01-23 19:16:21.444058: step: 2908/1541, loss: 0.008979905396699905 2023-01-23 19:16:22.403266: step: 2910/1541, loss: 0.04474852606654167 2023-01-23 19:16:23.329951: step: 2912/1541, loss: 0.0006775177316740155 2023-01-23 19:16:24.229767: step: 2914/1541, loss: 0.009056279435753822 2023-01-23 19:16:25.090281: step: 2916/1541, loss: 0.5839191675186157 2023-01-23 19:16:25.990511: step: 2918/1541, loss: 0.06592472642660141 2023-01-23 19:16:26.927288: step: 2920/1541, loss: 0.0005492103518918157 2023-01-23 19:16:27.850487: step: 2922/1541, loss: 0.02856653369963169 2023-01-23 19:16:28.740499: step: 2924/1541, loss: 0.023980762809515 2023-01-23 19:16:29.635291: step: 2926/1541, loss: 0.21893049776554108 2023-01-23 19:16:30.673474: step: 2928/1541, loss: 0.05452817305922508 2023-01-23 19:16:31.541086: step: 2930/1541, loss: 0.0006304702255874872 2023-01-23 19:16:32.455655: step: 2932/1541, loss: 0.010184705257415771 2023-01-23 19:16:33.348385: step: 2934/1541, loss: 0.0008099536062218249 2023-01-23 19:16:34.228577: step: 2936/1541, loss: 0.0003433512756600976 2023-01-23 19:16:35.103311: step: 2938/1541, loss: 0.2125566303730011 2023-01-23 19:16:35.997011: step: 2940/1541, loss: 0.3271543085575104 2023-01-23 19:16:36.910064: step: 2942/1541, loss: 0.15968884527683258 2023-01-23 19:16:37.748978: step: 2944/1541, loss: 0.010984441265463829 2023-01-23 19:16:38.619566: step: 2946/1541, loss: 0.0005554666859097779 2023-01-23 19:16:39.495750: step: 2948/1541, loss: 0.0032138449605554342 2023-01-23 19:16:40.379766: step: 2950/1541, loss: 0.00018621505296323448 2023-01-23 19:16:41.270176: step: 2952/1541, loss: 0.21366263926029205 2023-01-23 19:16:42.172209: step: 2954/1541, loss: 0.020940274000167847 2023-01-23 19:16:43.073294: step: 2956/1541, loss: 0.1859321892261505 2023-01-23 19:16:43.969270: step: 2958/1541, loss: 0.006292345467954874 2023-01-23 19:16:44.846829: step: 2960/1541, loss: 0.004255941137671471 2023-01-23 19:16:45.772228: step: 2962/1541, loss: 0.008115425705909729 2023-01-23 19:16:46.660027: step: 2964/1541, loss: 0.00010471393761690706 2023-01-23 19:16:47.548803: step: 2966/1541, loss: 0.0016014809953048825 2023-01-23 19:16:48.444264: step: 2968/1541, loss: 0.06146223098039627 2023-01-23 19:16:49.350075: step: 2970/1541, loss: 0.09314342588186264 2023-01-23 19:16:50.275408: step: 2972/1541, loss: 0.008541432209312916 2023-01-23 19:16:51.205655: step: 2974/1541, loss: 0.000175679728272371 2023-01-23 19:16:52.066652: step: 2976/1541, loss: 0.010459236800670624 2023-01-23 19:16:52.955544: step: 2978/1541, loss: 0.23297779262065887 2023-01-23 19:16:53.839037: step: 2980/1541, loss: 0.15163534879684448 2023-01-23 19:16:54.809979: step: 2982/1541, loss: 0.07106296718120575 2023-01-23 19:16:55.720480: step: 2984/1541, loss: 0.17424112558364868 2023-01-23 19:16:56.616429: step: 2986/1541, loss: 0.19798544049263 2023-01-23 19:16:57.499463: step: 2988/1541, loss: 0.017734678462147713 2023-01-23 19:16:58.434113: step: 2990/1541, loss: 0.012004237622022629 2023-01-23 19:16:59.348947: step: 2992/1541, loss: 0.21869109570980072 2023-01-23 19:17:00.236417: step: 2994/1541, loss: 0.5099053382873535 2023-01-23 19:17:01.199846: step: 2996/1541, loss: 0.19037771224975586 2023-01-23 19:17:02.131956: step: 2998/1541, loss: 0.0016996682388707995 2023-01-23 19:17:03.021281: step: 3000/1541, loss: 1.2458504438400269 2023-01-23 19:17:03.882537: step: 3002/1541, loss: 0.0001516097690910101 2023-01-23 19:17:04.769976: step: 3004/1541, loss: 0.011531208641827106 2023-01-23 19:17:05.661828: step: 3006/1541, loss: 0.0005537240649573505 2023-01-23 19:17:06.561337: step: 3008/1541, loss: 0.23779916763305664 2023-01-23 19:17:07.447346: step: 3010/1541, loss: 0.00058191322023049 2023-01-23 19:17:08.321612: step: 3012/1541, loss: 0.15909065306186676 2023-01-23 19:17:09.227274: step: 3014/1541, loss: 0.00071867456426844 2023-01-23 19:17:10.134282: step: 3016/1541, loss: 0.0212581604719162 2023-01-23 19:17:11.031203: step: 3018/1541, loss: 0.07973947376012802 2023-01-23 19:17:11.950002: step: 3020/1541, loss: 0.483587384223938 2023-01-23 19:17:12.859743: step: 3022/1541, loss: 0.017055034637451172 2023-01-23 19:17:13.752901: step: 3024/1541, loss: 0.016956672072410583 2023-01-23 19:17:14.648610: step: 3026/1541, loss: 0.04150335490703583 2023-01-23 19:17:15.535098: step: 3028/1541, loss: 0.00465871999040246 2023-01-23 19:17:16.434057: step: 3030/1541, loss: 0.0022703344002366066 2023-01-23 19:17:17.373696: step: 3032/1541, loss: 0.018571577966213226 2023-01-23 19:17:18.297501: step: 3034/1541, loss: 0.00039244061917997897 2023-01-23 19:17:19.200781: step: 3036/1541, loss: 1.9956061840057373 2023-01-23 19:17:20.127673: step: 3038/1541, loss: 0.09644407033920288 2023-01-23 19:17:21.002966: step: 3040/1541, loss: 0.005328746512532234 2023-01-23 19:17:21.947914: step: 3042/1541, loss: 0.005395527463406324 2023-01-23 19:17:22.892842: step: 3044/1541, loss: 0.0004195222572889179 2023-01-23 19:17:23.788922: step: 3046/1541, loss: 0.0045770686119794846 2023-01-23 19:17:24.639698: step: 3048/1541, loss: 0.0011383409146219492 2023-01-23 19:17:25.531005: step: 3050/1541, loss: 0.008945167995989323 2023-01-23 19:17:26.440014: step: 3052/1541, loss: 0.07481503486633301 2023-01-23 19:17:27.425391: step: 3054/1541, loss: 0.02149173431098461 2023-01-23 19:17:28.300598: step: 3056/1541, loss: 0.41362807154655457 2023-01-23 19:17:29.184815: step: 3058/1541, loss: 0.05264386534690857 2023-01-23 19:17:30.076323: step: 3060/1541, loss: 0.00042669044341892004 2023-01-23 19:17:30.939002: step: 3062/1541, loss: 0.0024302450474351645 2023-01-23 19:17:31.823305: step: 3064/1541, loss: 0.015658415853977203 2023-01-23 19:17:32.732525: step: 3066/1541, loss: 0.005352812819182873 2023-01-23 19:17:33.620586: step: 3068/1541, loss: 0.0007878539036028087 2023-01-23 19:17:34.547115: step: 3070/1541, loss: 0.029281392693519592 2023-01-23 19:17:35.398560: step: 3072/1541, loss: 0.0010246715974062681 2023-01-23 19:17:36.371433: step: 3074/1541, loss: 0.17260752618312836 2023-01-23 19:17:37.245741: step: 3076/1541, loss: 0.002550003817304969 2023-01-23 19:17:38.135129: step: 3078/1541, loss: 0.0005082319839857519 2023-01-23 19:17:39.045575: step: 3080/1541, loss: 0.0002513639337848872 2023-01-23 19:17:39.947503: step: 3082/1541, loss: 0.023656900972127914 2023-01-23 19:17:40.595547: step: 3084/1541, loss: 5.870097447768785e-05 ================================================== Loss: 0.067 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6785674678881605, 'r': 0.6511506004987399, 'f1': 0.6645763860760334}, 'combined': 0.43942393285883824, 'epoch': 29} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7160994091858569, 'r': 0.6967876904250139, 'f1': 0.7063115709369184}, 'combined': 0.4600618073366767, 'epoch': 29} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6750913025256584, 'r': 0.6593323941082673, 'f1': 0.6671187957148792}, 'combined': 0.46175495327224236, 'epoch': 29} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.7161345911435275, 'r': 0.702263978546056, 'f1': 0.7091314640298458}, 'combined': 0.4916380708720887, 'epoch': 29} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6429075235109718, 'r': 0.6599758648431215, 'f1': 0.6513298928146091}, 'combined': 0.48988915015115897, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5761148272017838, 'r': 0.6163088849135361, 'f1': 0.5955344281186978}, 'combined': 0.28925957937193897, 'epoch': 29} New best korean model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 30 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 19:23:11.393136: step: 2/1541, loss: 0.00038541521644219756 2023-01-23 19:23:12.262962: step: 4/1541, loss: 0.04341582953929901 2023-01-23 19:23:13.150114: step: 6/1541, loss: 0.24681887030601501 2023-01-23 19:23:14.053268: step: 8/1541, loss: 0.0003212057345081121 2023-01-23 19:23:14.957915: step: 10/1541, loss: 0.06660483032464981 2023-01-23 19:23:15.883357: step: 12/1541, loss: 0.2528073787689209 2023-01-23 19:23:16.822994: step: 14/1541, loss: 0.0011931355111300945 2023-01-23 19:23:17.773501: step: 16/1541, loss: 0.00029494406771846116 2023-01-23 19:23:18.673918: step: 18/1541, loss: 0.0038490132428705692 2023-01-23 19:23:19.572790: step: 20/1541, loss: 0.002455643843859434 2023-01-23 19:23:20.487581: step: 22/1541, loss: 0.3647923767566681 2023-01-23 19:23:21.427767: step: 24/1541, loss: 0.3259512484073639 2023-01-23 19:23:22.296868: step: 26/1541, loss: 0.05657140538096428 2023-01-23 19:23:23.166164: step: 28/1541, loss: 0.003517427248880267 2023-01-23 19:23:24.062105: step: 30/1541, loss: 0.03797813877463341 2023-01-23 19:23:24.974417: step: 32/1541, loss: 0.0005670725950039923 2023-01-23 19:23:25.877241: step: 34/1541, loss: 0.001816058182157576 2023-01-23 19:23:26.782205: step: 36/1541, loss: 0.0004408370004966855 2023-01-23 19:23:27.688037: step: 38/1541, loss: 0.0012591604609042406 2023-01-23 19:23:28.606831: step: 40/1541, loss: 0.0005184180336073041 2023-01-23 19:23:29.548561: step: 42/1541, loss: 0.00040584761882200837 2023-01-23 19:23:30.449202: step: 44/1541, loss: 0.04696871340274811 2023-01-23 19:23:31.358517: step: 46/1541, loss: 0.00032166976598091424 2023-01-23 19:23:32.340298: step: 48/1541, loss: 0.4148523807525635 2023-01-23 19:23:33.198531: step: 50/1541, loss: 0.001506004948168993 2023-01-23 19:23:34.068996: step: 52/1541, loss: 1.219585657119751 2023-01-23 19:23:34.942960: step: 54/1541, loss: 0.00027393834898248315 2023-01-23 19:23:35.839075: step: 56/1541, loss: 0.08314666152000427 2023-01-23 19:23:36.729999: step: 58/1541, loss: 0.026497088372707367 2023-01-23 19:23:37.614654: step: 60/1541, loss: 0.0002011082979151979 2023-01-23 19:23:38.507141: step: 62/1541, loss: 0.001482466235756874 2023-01-23 19:23:39.405588: step: 64/1541, loss: 0.16202788054943085 2023-01-23 19:23:40.308845: step: 66/1541, loss: 0.015191365964710712 2023-01-23 19:23:41.310923: step: 68/1541, loss: 0.16546370089054108 2023-01-23 19:23:42.241772: step: 70/1541, loss: 0.00012650422286242247 2023-01-23 19:23:43.149034: step: 72/1541, loss: 0.08646831661462784 2023-01-23 19:23:44.006285: step: 74/1541, loss: 0.0011420836672186852 2023-01-23 19:23:44.895269: step: 76/1541, loss: 0.10320570319890976 2023-01-23 19:23:45.797558: step: 78/1541, loss: 3.0833874916424975e-05 2023-01-23 19:23:46.741193: step: 80/1541, loss: 0.12179717421531677 2023-01-23 19:23:47.660496: step: 82/1541, loss: 0.00044799179886467755 2023-01-23 19:23:48.545603: step: 84/1541, loss: 0.0015018227277323604 2023-01-23 19:23:49.469553: step: 86/1541, loss: 3.0403431082959287e-05 2023-01-23 19:23:50.388002: step: 88/1541, loss: 0.00037637760397046804 2023-01-23 19:23:51.308904: step: 90/1541, loss: 0.0016028615646064281 2023-01-23 19:23:52.206823: step: 92/1541, loss: 0.0016840416938066483 2023-01-23 19:23:53.069264: step: 94/1541, loss: 0.03334381431341171 2023-01-23 19:23:53.936397: step: 96/1541, loss: 0.21029208600521088 2023-01-23 19:23:54.831837: step: 98/1541, loss: 0.031430505216121674 2023-01-23 19:23:55.718158: step: 100/1541, loss: 3.476140773273073e-05 2023-01-23 19:23:56.599487: step: 102/1541, loss: 0.0001056039473041892 2023-01-23 19:23:57.515980: step: 104/1541, loss: 0.001203229301609099 2023-01-23 19:23:58.398329: step: 106/1541, loss: 0.08375049382448196 2023-01-23 19:23:59.273170: step: 108/1541, loss: 0.1278296858072281 2023-01-23 19:24:00.179721: step: 110/1541, loss: -1.538121978228446e-05 2023-01-23 19:24:01.108775: step: 112/1541, loss: 0.020915884524583817 2023-01-23 19:24:02.008175: step: 114/1541, loss: 0.06476886570453644 2023-01-23 19:24:02.848356: step: 116/1541, loss: 0.0006823487346991897 2023-01-23 19:24:03.703411: step: 118/1541, loss: 3.514634590828791e-05 2023-01-23 19:24:04.631276: step: 120/1541, loss: 0.001277028233744204 2023-01-23 19:24:05.514829: step: 122/1541, loss: 0.15501205623149872 2023-01-23 19:24:06.497553: step: 124/1541, loss: 0.1369229257106781 2023-01-23 19:24:07.341125: step: 126/1541, loss: 0.0021212813444435596 2023-01-23 19:24:08.276127: step: 128/1541, loss: 0.09104583412408829 2023-01-23 19:24:09.141905: step: 130/1541, loss: 0.0010160037782043219 2023-01-23 19:24:10.108360: step: 132/1541, loss: 0.03818478062748909 2023-01-23 19:24:10.974097: step: 134/1541, loss: 0.00035762679181061685 2023-01-23 19:24:11.850944: step: 136/1541, loss: 0.0033046184107661247 2023-01-23 19:24:12.763585: step: 138/1541, loss: 0.0026874379254877567 2023-01-23 19:24:13.646580: step: 140/1541, loss: 0.001971643650904298 2023-01-23 19:24:14.601021: step: 142/1541, loss: 0.16064311563968658 2023-01-23 19:24:15.475737: step: 144/1541, loss: 0.14274825155735016 2023-01-23 19:24:16.371829: step: 146/1541, loss: 1.0774432666948996e-05 2023-01-23 19:24:17.282986: step: 148/1541, loss: 0.11526800692081451 2023-01-23 19:24:18.213187: step: 150/1541, loss: 7.660868868697435e-05 2023-01-23 19:24:19.131467: step: 152/1541, loss: 0.3720688223838806 2023-01-23 19:24:20.106298: step: 154/1541, loss: 0.0004255011444911361 2023-01-23 19:24:21.011371: step: 156/1541, loss: 0.10044754296541214 2023-01-23 19:24:21.986051: step: 158/1541, loss: 0.0020758453756570816 2023-01-23 19:24:22.900504: step: 160/1541, loss: 0.005169909913092852 2023-01-23 19:24:23.864024: step: 162/1541, loss: 0.0005839696968905628 2023-01-23 19:24:24.768986: step: 164/1541, loss: 0.03024001605808735 2023-01-23 19:24:25.646178: step: 166/1541, loss: 0.0005109526100568473 2023-01-23 19:24:26.619693: step: 168/1541, loss: 1.1889311075210571 2023-01-23 19:24:27.518590: step: 170/1541, loss: 0.0008212760440073907 2023-01-23 19:24:28.380312: step: 172/1541, loss: 0.001487507950514555 2023-01-23 19:24:29.234482: step: 174/1541, loss: 0.0020027344580739737 2023-01-23 19:24:30.135029: step: 176/1541, loss: 0.002557943807914853 2023-01-23 19:24:31.028827: step: 178/1541, loss: 0.03461230918765068 2023-01-23 19:24:31.947778: step: 180/1541, loss: 0.0016069014091044664 2023-01-23 19:24:32.814275: step: 182/1541, loss: 0.03983587399125099 2023-01-23 19:24:33.680874: step: 184/1541, loss: 0.06510518491268158 2023-01-23 19:24:34.576910: step: 186/1541, loss: 0.05337444692850113 2023-01-23 19:24:35.495535: step: 188/1541, loss: 0.00013792916433885694 2023-01-23 19:24:36.384620: step: 190/1541, loss: 0.027823366224765778 2023-01-23 19:24:37.294604: step: 192/1541, loss: 0.005326287820935249 2023-01-23 19:24:38.194021: step: 194/1541, loss: 0.14040495455265045 2023-01-23 19:24:39.051154: step: 196/1541, loss: 0.05454308167099953 2023-01-23 19:24:39.917543: step: 198/1541, loss: 0.0005233807605691254 2023-01-23 19:24:40.764493: step: 200/1541, loss: 0.01849653199315071 2023-01-23 19:24:41.739761: step: 202/1541, loss: 0.0031952750869095325 2023-01-23 19:24:42.642418: step: 204/1541, loss: 0.002554644364863634 2023-01-23 19:24:43.523507: step: 206/1541, loss: 0.06293267011642456 2023-01-23 19:24:44.392405: step: 208/1541, loss: 0.0038221338763833046 2023-01-23 19:24:45.289256: step: 210/1541, loss: 0.009239614009857178 2023-01-23 19:24:46.249171: step: 212/1541, loss: 0.0013134137261658907 2023-01-23 19:24:47.146548: step: 214/1541, loss: 0.004850102122873068 2023-01-23 19:24:48.157766: step: 216/1541, loss: 0.014773013070225716 2023-01-23 19:24:49.097966: step: 218/1541, loss: 0.004406817257404327 2023-01-23 19:24:49.975218: step: 220/1541, loss: 0.0004039460909552872 2023-01-23 19:24:50.841722: step: 222/1541, loss: 0.00028179757646285 2023-01-23 19:24:51.704039: step: 224/1541, loss: 0.000239940665778704 2023-01-23 19:24:52.583058: step: 226/1541, loss: 0.00036719985655508935 2023-01-23 19:24:53.515924: step: 228/1541, loss: 3.61037346010562e-05 2023-01-23 19:24:54.396877: step: 230/1541, loss: 4.323988105170429e-05 2023-01-23 19:24:55.301042: step: 232/1541, loss: 0.017957651987671852 2023-01-23 19:24:56.146807: step: 234/1541, loss: 0.001487251021899283 2023-01-23 19:24:57.037940: step: 236/1541, loss: 0.0559205487370491 2023-01-23 19:24:57.915825: step: 238/1541, loss: 0.018494706600904465 2023-01-23 19:24:58.786808: step: 240/1541, loss: 0.2890370190143585 2023-01-23 19:24:59.707185: step: 242/1541, loss: 0.02541477605700493 2023-01-23 19:25:00.591564: step: 244/1541, loss: -3.1524585210718215e-05 2023-01-23 19:25:01.439776: step: 246/1541, loss: 0.0005501921405084431 2023-01-23 19:25:02.409896: step: 248/1541, loss: 0.3403070867061615 2023-01-23 19:25:03.282340: step: 250/1541, loss: 1.6280919226119295e-06 2023-01-23 19:25:04.193583: step: 252/1541, loss: 0.00020543165737763047 2023-01-23 19:25:05.195474: step: 254/1541, loss: 6.961688632145524e-05 2023-01-23 19:25:06.075089: step: 256/1541, loss: 0.11265359073877335 2023-01-23 19:25:06.970796: step: 258/1541, loss: 8.058291132329032e-05 2023-01-23 19:25:07.837760: step: 260/1541, loss: 0.34956008195877075 2023-01-23 19:25:08.754958: step: 262/1541, loss: 0.005939038470387459 2023-01-23 19:25:09.639494: step: 264/1541, loss: 0.0020375207532197237 2023-01-23 19:25:10.602039: step: 266/1541, loss: 0.48005831241607666 2023-01-23 19:25:11.497259: step: 268/1541, loss: 0.49322113394737244 2023-01-23 19:25:12.429986: step: 270/1541, loss: 0.0655529573559761 2023-01-23 19:25:13.337656: step: 272/1541, loss: 0.0937994047999382 2023-01-23 19:25:14.238007: step: 274/1541, loss: 0.00018060456204693764 2023-01-23 19:25:15.147004: step: 276/1541, loss: 0.00018778421508613974 2023-01-23 19:25:16.010682: step: 278/1541, loss: 0.3451794385910034 2023-01-23 19:25:16.931479: step: 280/1541, loss: 0.001428141025826335 2023-01-23 19:25:17.828708: step: 282/1541, loss: 5.546250395127572e-05 2023-01-23 19:25:18.692593: step: 284/1541, loss: 0.0006280058878473938 2023-01-23 19:25:19.592164: step: 286/1541, loss: 0.004601827822625637 2023-01-23 19:25:20.494283: step: 288/1541, loss: 0.00048122761654667556 2023-01-23 19:25:21.369304: step: 290/1541, loss: 0.00012414240336511284 2023-01-23 19:25:22.240677: step: 292/1541, loss: 0.017275774851441383 2023-01-23 19:25:23.131057: step: 294/1541, loss: 0.00016038387548178434 2023-01-23 19:25:24.022917: step: 296/1541, loss: 0.36696621775627136 2023-01-23 19:25:24.985154: step: 298/1541, loss: 0.3225466012954712 2023-01-23 19:25:25.888645: step: 300/1541, loss: 0.15301118791103363 2023-01-23 19:25:26.762345: step: 302/1541, loss: 0.015280510298907757 2023-01-23 19:25:27.678237: step: 304/1541, loss: 0.12749090790748596 2023-01-23 19:25:28.572161: step: 306/1541, loss: 0.013338622637093067 2023-01-23 19:25:29.449796: step: 308/1541, loss: 0.1802721470594406 2023-01-23 19:25:30.330403: step: 310/1541, loss: 0.12574706971645355 2023-01-23 19:25:31.219910: step: 312/1541, loss: 0.0032888122368603945 2023-01-23 19:25:32.083454: step: 314/1541, loss: 0.03682565316557884 2023-01-23 19:25:33.022711: step: 316/1541, loss: 0.1092943325638771 2023-01-23 19:25:33.940761: step: 318/1541, loss: 0.13186223804950714 2023-01-23 19:25:34.832671: step: 320/1541, loss: 0.056640028953552246 2023-01-23 19:25:35.775375: step: 322/1541, loss: 0.09985089302062988 2023-01-23 19:25:36.727351: step: 324/1541, loss: 0.00026932722539640963 2023-01-23 19:25:37.582973: step: 326/1541, loss: 0.0005613721441477537 2023-01-23 19:25:38.496982: step: 328/1541, loss: 0.007872761227190495 2023-01-23 19:25:39.439050: step: 330/1541, loss: 0.0016033044084906578 2023-01-23 19:25:40.334216: step: 332/1541, loss: 0.010982553474605083 2023-01-23 19:25:41.220170: step: 334/1541, loss: 0.5989691615104675 2023-01-23 19:25:42.116778: step: 336/1541, loss: 0.011298082768917084 2023-01-23 19:25:43.060034: step: 338/1541, loss: 2.8895807190565392e-05 2023-01-23 19:25:43.980140: step: 340/1541, loss: 0.047598108649253845 2023-01-23 19:25:44.811535: step: 342/1541, loss: 0.002665772335603833 2023-01-23 19:25:45.687721: step: 344/1541, loss: 0.0006284031551331282 2023-01-23 19:25:46.556534: step: 346/1541, loss: 0.005391888786107302 2023-01-23 19:25:47.450647: step: 348/1541, loss: 0.0020071014296263456 2023-01-23 19:25:48.349414: step: 350/1541, loss: 0.07440845668315887 2023-01-23 19:25:49.242093: step: 352/1541, loss: 0.00046194548485800624 2023-01-23 19:25:50.115540: step: 354/1541, loss: 0.1754174530506134 2023-01-23 19:25:50.969489: step: 356/1541, loss: 0.0010790432570502162 2023-01-23 19:25:51.847952: step: 358/1541, loss: 0.018704283982515335 2023-01-23 19:25:52.849875: step: 360/1541, loss: 0.001146146678365767 2023-01-23 19:25:53.714210: step: 362/1541, loss: 0.006246503442525864 2023-01-23 19:25:54.589941: step: 364/1541, loss: 0.004065703134983778 2023-01-23 19:25:55.547808: step: 366/1541, loss: 0.000571881071664393 2023-01-23 19:25:56.417200: step: 368/1541, loss: 0.002022018190473318 2023-01-23 19:25:57.316446: step: 370/1541, loss: 0.01950700394809246 2023-01-23 19:25:58.183036: step: 372/1541, loss: 0.0001830681721912697 2023-01-23 19:25:59.079019: step: 374/1541, loss: 0.007667983882129192 2023-01-23 19:26:00.100791: step: 376/1541, loss: 0.0014582719886675477 2023-01-23 19:26:00.989378: step: 378/1541, loss: 0.010565542615950108 2023-01-23 19:26:01.892509: step: 380/1541, loss: 0.0008486206061206758 2023-01-23 19:26:02.757604: step: 382/1541, loss: 0.0013419648166745901 2023-01-23 19:26:03.616798: step: 384/1541, loss: 0.010564355179667473 2023-01-23 19:26:04.590249: step: 386/1541, loss: 0.0011759281624108553 2023-01-23 19:26:05.513808: step: 388/1541, loss: 0.17793317139148712 2023-01-23 19:26:06.410121: step: 390/1541, loss: 0.003584659192711115 2023-01-23 19:26:07.245811: step: 392/1541, loss: 0.0002177067071897909 2023-01-23 19:26:08.120342: step: 394/1541, loss: 0.00026962775154970586 2023-01-23 19:26:09.044628: step: 396/1541, loss: 0.0013227639719843864 2023-01-23 19:26:09.943522: step: 398/1541, loss: 0.007393260952085257 2023-01-23 19:26:10.847351: step: 400/1541, loss: 0.007134603802114725 2023-01-23 19:26:11.778329: step: 402/1541, loss: 0.06513292342424393 2023-01-23 19:26:12.686768: step: 404/1541, loss: 0.0003546794760040939 2023-01-23 19:26:13.631038: step: 406/1541, loss: 0.006888632662594318 2023-01-23 19:26:14.554760: step: 408/1541, loss: 0.0007300589932128787 2023-01-23 19:26:15.423901: step: 410/1541, loss: 0.08341812342405319 2023-01-23 19:26:16.303630: step: 412/1541, loss: 0.006426959298551083 2023-01-23 19:26:17.180111: step: 414/1541, loss: 0.10667189955711365 2023-01-23 19:26:18.043888: step: 416/1541, loss: 0.0007103733369149268 2023-01-23 19:26:18.931667: step: 418/1541, loss: 0.01730635203421116 2023-01-23 19:26:19.812080: step: 420/1541, loss: 0.0010047814575955272 2023-01-23 19:26:20.693816: step: 422/1541, loss: 0.005647421348839998 2023-01-23 19:26:21.564830: step: 424/1541, loss: 0.5074038505554199 2023-01-23 19:26:22.512739: step: 426/1541, loss: 0.07859756052494049 2023-01-23 19:26:23.393194: step: 428/1541, loss: 0.05672073736786842 2023-01-23 19:26:24.265510: step: 430/1541, loss: 0.06648657470941544 2023-01-23 19:26:25.166608: step: 432/1541, loss: 0.004197355825453997 2023-01-23 19:26:26.116002: step: 434/1541, loss: 0.1628609448671341 2023-01-23 19:26:26.971176: step: 436/1541, loss: 0.0004122865793760866 2023-01-23 19:26:27.841919: step: 438/1541, loss: 0.00026476700440980494 2023-01-23 19:26:28.734681: step: 440/1541, loss: 0.0007458805339410901 2023-01-23 19:26:29.649401: step: 442/1541, loss: 0.001512509654276073 2023-01-23 19:26:30.544137: step: 444/1541, loss: 0.0004250294005032629 2023-01-23 19:26:31.408408: step: 446/1541, loss: 0.00041988666635006666 2023-01-23 19:26:32.312273: step: 448/1541, loss: 0.4812350869178772 2023-01-23 19:26:33.156791: step: 450/1541, loss: 0.0009324729908257723 2023-01-23 19:26:34.034366: step: 452/1541, loss: 0.004816718865185976 2023-01-23 19:26:34.898467: step: 454/1541, loss: 0.004116961266845465 2023-01-23 19:26:35.821139: step: 456/1541, loss: 0.061735834926366806 2023-01-23 19:26:36.681031: step: 458/1541, loss: 0.11589184403419495 2023-01-23 19:26:37.572157: step: 460/1541, loss: 0.3498118221759796 2023-01-23 19:26:38.442121: step: 462/1541, loss: 0.0008112787618301809 2023-01-23 19:26:39.356695: step: 464/1541, loss: 0.07938243448734283 2023-01-23 19:26:40.257738: step: 466/1541, loss: 0.00022541025828104466 2023-01-23 19:26:41.118610: step: 468/1541, loss: 0.00109301065094769 2023-01-23 19:26:41.993955: step: 470/1541, loss: 0.05112815648317337 2023-01-23 19:26:42.888879: step: 472/1541, loss: 0.0005441700923256576 2023-01-23 19:26:43.756632: step: 474/1541, loss: 0.0008441523532383144 2023-01-23 19:26:44.639096: step: 476/1541, loss: 0.5549581050872803 2023-01-23 19:26:45.612613: step: 478/1541, loss: 0.0063877105712890625 2023-01-23 19:26:46.519911: step: 480/1541, loss: 0.0010206613223999739 2023-01-23 19:26:47.400216: step: 482/1541, loss: 0.004621011205017567 2023-01-23 19:26:48.325372: step: 484/1541, loss: 0.0003737136139534414 2023-01-23 19:26:49.201241: step: 486/1541, loss: 0.0003170908021274954 2023-01-23 19:26:50.095922: step: 488/1541, loss: 6.139575270935893e-05 2023-01-23 19:26:50.974573: step: 490/1541, loss: 0.029475659132003784 2023-01-23 19:26:51.845261: step: 492/1541, loss: 0.0015330408932641149 2023-01-23 19:26:52.763520: step: 494/1541, loss: 0.00023257164866663516 2023-01-23 19:26:53.640059: step: 496/1541, loss: 0.000655831303447485 2023-01-23 19:26:54.508982: step: 498/1541, loss: 0.16348788142204285 2023-01-23 19:26:55.377195: step: 500/1541, loss: 0.0012335748178884387 2023-01-23 19:26:56.230176: step: 502/1541, loss: 0.032656215131282806 2023-01-23 19:26:57.135671: step: 504/1541, loss: 0.13364623486995697 2023-01-23 19:26:58.022128: step: 506/1541, loss: 0.29178643226623535 2023-01-23 19:26:58.902601: step: 508/1541, loss: 0.0068770949728786945 2023-01-23 19:26:59.786877: step: 510/1541, loss: 0.0006336075603030622 2023-01-23 19:27:00.653496: step: 512/1541, loss: 0.011934278532862663 2023-01-23 19:27:01.587376: step: 514/1541, loss: 0.2430134117603302 2023-01-23 19:27:02.433165: step: 516/1541, loss: 0.014552174136042595 2023-01-23 19:27:03.319032: step: 518/1541, loss: 0.10041619837284088 2023-01-23 19:27:04.186150: step: 520/1541, loss: 7.015527080511674e-05 2023-01-23 19:27:05.079100: step: 522/1541, loss: 0.037385933101177216 2023-01-23 19:27:05.999890: step: 524/1541, loss: 0.12425775080919266 2023-01-23 19:27:06.923941: step: 526/1541, loss: 0.0011025697458535433 2023-01-23 19:27:07.821200: step: 528/1541, loss: 0.0007472276338376105 2023-01-23 19:27:08.759622: step: 530/1541, loss: 0.0001933082821778953 2023-01-23 19:27:09.644048: step: 532/1541, loss: 7.662943971809e-05 2023-01-23 19:27:10.501642: step: 534/1541, loss: 0.0015900360886007547 2023-01-23 19:27:11.383452: step: 536/1541, loss: 0.011928202584385872 2023-01-23 19:27:12.287622: step: 538/1541, loss: 0.11841004341840744 2023-01-23 19:27:13.180080: step: 540/1541, loss: 0.010989471338689327 2023-01-23 19:27:14.089140: step: 542/1541, loss: 0.020040487870573997 2023-01-23 19:27:15.006847: step: 544/1541, loss: 0.023668624460697174 2023-01-23 19:27:15.848524: step: 546/1541, loss: 0.000808346550911665 2023-01-23 19:27:16.717327: step: 548/1541, loss: 0.49547505378723145 2023-01-23 19:27:17.608386: step: 550/1541, loss: 0.03282307833433151 2023-01-23 19:27:18.529787: step: 552/1541, loss: 0.020932385697960854 2023-01-23 19:27:19.405089: step: 554/1541, loss: 0.002219389658421278 2023-01-23 19:27:20.288562: step: 556/1541, loss: 0.0073494198732078075 2023-01-23 19:27:21.201301: step: 558/1541, loss: 7.00495729688555e-05 2023-01-23 19:27:22.122611: step: 560/1541, loss: 0.034878913313150406 2023-01-23 19:27:23.028263: step: 562/1541, loss: 0.000263517489656806 2023-01-23 19:27:23.916978: step: 564/1541, loss: 0.0006916356505826116 2023-01-23 19:27:24.827362: step: 566/1541, loss: 0.002245893469080329 2023-01-23 19:27:25.702959: step: 568/1541, loss: 0.010470351204276085 2023-01-23 19:27:26.589925: step: 570/1541, loss: 0.0003824048035312444 2023-01-23 19:27:27.541856: step: 572/1541, loss: 0.0006680625374428928 2023-01-23 19:27:28.439476: step: 574/1541, loss: 0.0008537366520613432 2023-01-23 19:27:29.341712: step: 576/1541, loss: 0.5724825859069824 2023-01-23 19:27:30.260996: step: 578/1541, loss: 0.014306274242699146 2023-01-23 19:27:31.160061: step: 580/1541, loss: 0.27133700251579285 2023-01-23 19:27:32.055903: step: 582/1541, loss: 0.3424444794654846 2023-01-23 19:27:32.987465: step: 584/1541, loss: 0.27277863025665283 2023-01-23 19:27:33.858082: step: 586/1541, loss: 0.00172917649615556 2023-01-23 19:27:34.708703: step: 588/1541, loss: 0.0008587312186136842 2023-01-23 19:27:35.595604: step: 590/1541, loss: 0.17327219247817993 2023-01-23 19:27:36.484171: step: 592/1541, loss: 0.001684945891611278 2023-01-23 19:27:37.419621: step: 594/1541, loss: 0.031951405107975006 2023-01-23 19:27:38.306301: step: 596/1541, loss: 0.03665515407919884 2023-01-23 19:27:39.227905: step: 598/1541, loss: 0.0004029753035865724 2023-01-23 19:27:40.123518: step: 600/1541, loss: -2.19731118704658e-05 2023-01-23 19:27:40.944529: step: 602/1541, loss: 0.0006368036265484989 2023-01-23 19:27:41.845606: step: 604/1541, loss: 0.001498479745350778 2023-01-23 19:27:42.702185: step: 606/1541, loss: 0.001158747705630958 2023-01-23 19:27:43.629599: step: 608/1541, loss: 0.00023709192464593798 2023-01-23 19:27:44.497944: step: 610/1541, loss: 0.001041296636685729 2023-01-23 19:27:45.424568: step: 612/1541, loss: 0.035841088742017746 2023-01-23 19:27:46.358479: step: 614/1541, loss: 0.005236076191067696 2023-01-23 19:27:47.295945: step: 616/1541, loss: 0.00014039562665857375 2023-01-23 19:27:48.163541: step: 618/1541, loss: 0.0013492016587406397 2023-01-23 19:27:49.019254: step: 620/1541, loss: 0.0007432579877786338 2023-01-23 19:27:49.922242: step: 622/1541, loss: 0.009283008053898811 2023-01-23 19:27:50.815943: step: 624/1541, loss: 0.1452566236257553 2023-01-23 19:27:51.684664: step: 626/1541, loss: 0.004529625177383423 2023-01-23 19:27:52.553135: step: 628/1541, loss: 0.0019453666172921658 2023-01-23 19:27:53.427482: step: 630/1541, loss: 0.13602085411548615 2023-01-23 19:27:54.300403: step: 632/1541, loss: 0.053420159965753555 2023-01-23 19:27:55.199188: step: 634/1541, loss: 0.03235480561852455 2023-01-23 19:27:56.066755: step: 636/1541, loss: 0.06450559943914413 2023-01-23 19:27:56.936775: step: 638/1541, loss: 0.004641187842935324 2023-01-23 19:27:57.803707: step: 640/1541, loss: 0.0033395891077816486 2023-01-23 19:27:58.701183: step: 642/1541, loss: 0.022273922339081764 2023-01-23 19:27:59.639322: step: 644/1541, loss: 0.04926111549139023 2023-01-23 19:28:00.576474: step: 646/1541, loss: 0.0006976211443543434 2023-01-23 19:28:01.586601: step: 648/1541, loss: 0.007528889924287796 2023-01-23 19:28:02.476208: step: 650/1541, loss: 0.0019514558371156454 2023-01-23 19:28:03.397215: step: 652/1541, loss: 0.3538547158241272 2023-01-23 19:28:04.293161: step: 654/1541, loss: 0.15582585334777832 2023-01-23 19:28:05.176332: step: 656/1541, loss: 0.12918058037757874 2023-01-23 19:28:06.106864: step: 658/1541, loss: 0.001943927607499063 2023-01-23 19:28:06.967376: step: 660/1541, loss: 0.09177008271217346 2023-01-23 19:28:07.839961: step: 662/1541, loss: 0.0005418489454314113 2023-01-23 19:28:08.761274: step: 664/1541, loss: 0.00039016566006466746 2023-01-23 19:28:09.637305: step: 666/1541, loss: 0.00047790308599360287 2023-01-23 19:28:10.543008: step: 668/1541, loss: 0.1838318407535553 2023-01-23 19:28:11.498244: step: 670/1541, loss: 0.007844536565244198 2023-01-23 19:28:12.437329: step: 672/1541, loss: 0.14724427461624146 2023-01-23 19:28:13.307643: step: 674/1541, loss: 0.0002906442969106138 2023-01-23 19:28:14.181776: step: 676/1541, loss: 9.771824261406437e-05 2023-01-23 19:28:15.063880: step: 678/1541, loss: 0.0025954325683414936 2023-01-23 19:28:15.953889: step: 680/1541, loss: 0.1321965754032135 2023-01-23 19:28:16.979013: step: 682/1541, loss: 0.01863110437989235 2023-01-23 19:28:17.872675: step: 684/1541, loss: 0.01757136732339859 2023-01-23 19:28:18.797648: step: 686/1541, loss: 0.0003576671879272908 2023-01-23 19:28:19.753191: step: 688/1541, loss: 0.0005480002146214247 2023-01-23 19:28:20.631579: step: 690/1541, loss: 0.0017233899561688304 2023-01-23 19:28:21.500338: step: 692/1541, loss: 0.03560537472367287 2023-01-23 19:28:22.399835: step: 694/1541, loss: 0.06812966614961624 2023-01-23 19:28:23.231590: step: 696/1541, loss: 0.0016296063549816608 2023-01-23 19:28:24.119739: step: 698/1541, loss: 0.0013425308279693127 2023-01-23 19:28:24.990931: step: 700/1541, loss: 0.0002436064532957971 2023-01-23 19:28:25.867169: step: 702/1541, loss: 0.003243909915909171 2023-01-23 19:28:26.805928: step: 704/1541, loss: 0.0013185227289795876 2023-01-23 19:28:27.691965: step: 706/1541, loss: 0.009405548684298992 2023-01-23 19:28:28.571094: step: 708/1541, loss: 0.16396021842956543 2023-01-23 19:28:29.437016: step: 710/1541, loss: 0.0030276039615273476 2023-01-23 19:28:30.301855: step: 712/1541, loss: 0.00015081166930031031 2023-01-23 19:28:31.191886: step: 714/1541, loss: 0.004391700495034456 2023-01-23 19:28:32.153551: step: 716/1541, loss: 0.014825254678726196 2023-01-23 19:28:33.131182: step: 718/1541, loss: 0.06649062037467957 2023-01-23 19:28:34.094748: step: 720/1541, loss: 0.0351804718375206 2023-01-23 19:28:34.977892: step: 722/1541, loss: 0.0003297653456684202 2023-01-23 19:28:35.942685: step: 724/1541, loss: 0.048211049288511276 2023-01-23 19:28:36.874458: step: 726/1541, loss: 3.148592440993525e-05 2023-01-23 19:28:37.798317: step: 728/1541, loss: 5.903436249354854e-05 2023-01-23 19:28:38.722288: step: 730/1541, loss: 2.196391869802028e-05 2023-01-23 19:28:39.638297: step: 732/1541, loss: 0.008095704950392246 2023-01-23 19:28:40.688646: step: 734/1541, loss: 0.0004083871899638325 2023-01-23 19:28:41.655462: step: 736/1541, loss: 0.0027098411228507757 2023-01-23 19:28:42.581889: step: 738/1541, loss: 0.06458275765180588 2023-01-23 19:28:43.525505: step: 740/1541, loss: 0.003908369690179825 2023-01-23 19:28:44.371568: step: 742/1541, loss: 0.004169669467955828 2023-01-23 19:28:45.230524: step: 744/1541, loss: 0.02480456233024597 2023-01-23 19:28:46.185654: step: 746/1541, loss: 0.0017152053769677877 2023-01-23 19:28:47.168570: step: 748/1541, loss: 0.00013386463979259133 2023-01-23 19:28:48.036620: step: 750/1541, loss: 0.00020106100419070572 2023-01-23 19:28:48.894619: step: 752/1541, loss: 7.04850535839796e-05 2023-01-23 19:28:49.766127: step: 754/1541, loss: 0.02427457645535469 2023-01-23 19:28:50.776058: step: 756/1541, loss: 0.016656167805194855 2023-01-23 19:28:51.690944: step: 758/1541, loss: 0.00028902781195938587 2023-01-23 19:28:52.612288: step: 760/1541, loss: 7.04306221450679e-05 2023-01-23 19:28:53.529326: step: 762/1541, loss: 0.008226866833865643 2023-01-23 19:28:54.419277: step: 764/1541, loss: -1.3235508049547207e-05 2023-01-23 19:28:55.348273: step: 766/1541, loss: 0.03783026337623596 2023-01-23 19:28:56.224084: step: 768/1541, loss: 0.08343475311994553 2023-01-23 19:28:57.109543: step: 770/1541, loss: 0.0011497624218463898 2023-01-23 19:28:58.004479: step: 772/1541, loss: 0.003105418523773551 2023-01-23 19:28:58.897135: step: 774/1541, loss: 4.373887713882141e-05 2023-01-23 19:28:59.744525: step: 776/1541, loss: 0.00030005883309058845 2023-01-23 19:29:00.610662: step: 778/1541, loss: 6.876850420667324e-06 2023-01-23 19:29:01.525046: step: 780/1541, loss: -2.31548874580767e-05 2023-01-23 19:29:02.460990: step: 782/1541, loss: 0.16364280879497528 2023-01-23 19:29:03.357025: step: 784/1541, loss: 0.00019172648899257183 2023-01-23 19:29:04.267656: step: 786/1541, loss: 0.009988393634557724 2023-01-23 19:29:05.233535: step: 788/1541, loss: 0.013044923543930054 2023-01-23 19:29:06.161179: step: 790/1541, loss: 0.0005269358516670763 2023-01-23 19:29:07.112024: step: 792/1541, loss: 0.008161191828548908 2023-01-23 19:29:08.009710: step: 794/1541, loss: 0.021887296810746193 2023-01-23 19:29:08.873862: step: 796/1541, loss: 0.00056259281700477 2023-01-23 19:29:09.769567: step: 798/1541, loss: 0.05440925434231758 2023-01-23 19:29:10.632893: step: 800/1541, loss: 0.052479084581136703 2023-01-23 19:29:11.472314: step: 802/1541, loss: 0.005967844743281603 2023-01-23 19:29:12.311880: step: 804/1541, loss: 0.00024361509713344276 2023-01-23 19:29:13.292430: step: 806/1541, loss: 0.012504949234426022 2023-01-23 19:29:14.205151: step: 808/1541, loss: 0.20348234474658966 2023-01-23 19:29:15.104039: step: 810/1541, loss: 0.0001607864978723228 2023-01-23 19:29:16.004630: step: 812/1541, loss: 0.02287112921476364 2023-01-23 19:29:16.904817: step: 814/1541, loss: 0.00017314800061285496 2023-01-23 19:29:17.804254: step: 816/1541, loss: 0.035804908722639084 2023-01-23 19:29:18.741580: step: 818/1541, loss: 0.03094545006752014 2023-01-23 19:29:19.668340: step: 820/1541, loss: 0.43969306349754333 2023-01-23 19:29:20.541829: step: 822/1541, loss: 0.00012831634376198053 2023-01-23 19:29:21.431375: step: 824/1541, loss: 0.0002448412706144154 2023-01-23 19:29:22.337706: step: 826/1541, loss: 5.7006200222531334e-05 2023-01-23 19:29:23.197453: step: 828/1541, loss: 0.004879501182585955 2023-01-23 19:29:24.135526: step: 830/1541, loss: 0.0001711374643491581 2023-01-23 19:29:25.013527: step: 832/1541, loss: 0.0031995470635592937 2023-01-23 19:29:25.887363: step: 834/1541, loss: 0.0005197346908971667 2023-01-23 19:29:26.823247: step: 836/1541, loss: 0.006433949805796146 2023-01-23 19:29:27.753251: step: 838/1541, loss: 0.051392536610364914 2023-01-23 19:29:28.616241: step: 840/1541, loss: 3.717856088769622e-05 2023-01-23 19:29:29.578348: step: 842/1541, loss: 0.45689722895622253 2023-01-23 19:29:30.502785: step: 844/1541, loss: 0.004993211943656206 2023-01-23 19:29:31.368740: step: 846/1541, loss: 0.015245924703776836 2023-01-23 19:29:32.236671: step: 848/1541, loss: 5.661993782268837e-05 2023-01-23 19:29:33.126627: step: 850/1541, loss: 0.005999152548611164 2023-01-23 19:29:33.986006: step: 852/1541, loss: 0.32306283712387085 2023-01-23 19:29:34.850064: step: 854/1541, loss: 0.17169997096061707 2023-01-23 19:29:35.731459: step: 856/1541, loss: 0.00692429905757308 2023-01-23 19:29:36.586629: step: 858/1541, loss: 9.173155376629438e-06 2023-01-23 19:29:37.446000: step: 860/1541, loss: 5.8453530073165894e-05 2023-01-23 19:29:38.314802: step: 862/1541, loss: 0.0004025443340651691 2023-01-23 19:29:39.218714: step: 864/1541, loss: 0.017086993902921677 2023-01-23 19:29:40.073562: step: 866/1541, loss: 8.843787509249523e-05 2023-01-23 19:29:40.955614: step: 868/1541, loss: 0.0008764983504079282 2023-01-23 19:29:41.856838: step: 870/1541, loss: 0.000891638221219182 2023-01-23 19:29:42.771032: step: 872/1541, loss: 0.00343385711312294 2023-01-23 19:29:43.662733: step: 874/1541, loss: 6.543134804815054e-05 2023-01-23 19:29:44.599304: step: 876/1541, loss: 0.13894861936569214 2023-01-23 19:29:45.486552: step: 878/1541, loss: 0.0015006695175543427 2023-01-23 19:29:46.364325: step: 880/1541, loss: 0.008166637271642685 2023-01-23 19:29:47.269666: step: 882/1541, loss: 0.004715633578598499 2023-01-23 19:29:48.159784: step: 884/1541, loss: 0.002583767520263791 2023-01-23 19:29:49.063786: step: 886/1541, loss: 0.12013586610555649 2023-01-23 19:29:49.979218: step: 888/1541, loss: 0.019049763679504395 2023-01-23 19:29:50.859714: step: 890/1541, loss: 0.003244160208851099 2023-01-23 19:29:51.746133: step: 892/1541, loss: 0.6033757328987122 2023-01-23 19:29:52.624752: step: 894/1541, loss: 0.628921627998352 2023-01-23 19:29:53.540508: step: 896/1541, loss: 0.16226869821548462 2023-01-23 19:29:54.465770: step: 898/1541, loss: 0.4773520231246948 2023-01-23 19:29:55.357957: step: 900/1541, loss: 0.000640150741674006 2023-01-23 19:29:56.239002: step: 902/1541, loss: 0.03821320831775665 2023-01-23 19:29:57.104550: step: 904/1541, loss: 0.0013807707000523806 2023-01-23 19:29:57.959699: step: 906/1541, loss: 0.0011581453727558255 2023-01-23 19:29:58.811871: step: 908/1541, loss: 0.0001146555005107075 2023-01-23 19:29:59.662475: step: 910/1541, loss: 0.0024669289123266935 2023-01-23 19:30:00.559045: step: 912/1541, loss: 0.05260273069143295 2023-01-23 19:30:01.453027: step: 914/1541, loss: 0.000637256947811693 2023-01-23 19:30:02.388917: step: 916/1541, loss: 0.00016341939044650644 2023-01-23 19:30:03.258282: step: 918/1541, loss: 0.016705038025975227 2023-01-23 19:30:04.167067: step: 920/1541, loss: 0.00029992772033438087 2023-01-23 19:30:05.163754: step: 922/1541, loss: 0.1838279515504837 2023-01-23 19:30:06.047418: step: 924/1541, loss: 0.000512852449901402 2023-01-23 19:30:06.909234: step: 926/1541, loss: 0.00016040605260059237 2023-01-23 19:30:07.824655: step: 928/1541, loss: 0.01792685128748417 2023-01-23 19:30:08.695288: step: 930/1541, loss: 0.000125848688185215 2023-01-23 19:30:09.624666: step: 932/1541, loss: 0.01565997675061226 2023-01-23 19:30:10.541380: step: 934/1541, loss: 0.0009641572250984609 2023-01-23 19:30:11.493789: step: 936/1541, loss: 0.0001423341454938054 2023-01-23 19:30:12.376413: step: 938/1541, loss: 0.017708314582705498 2023-01-23 19:30:13.277536: step: 940/1541, loss: 0.019705062732100487 2023-01-23 19:30:14.162184: step: 942/1541, loss: 0.24480904638767242 2023-01-23 19:30:15.122296: step: 944/1541, loss: 0.09425787627696991 2023-01-23 19:30:16.019413: step: 946/1541, loss: 0.0075170258060097694 2023-01-23 19:30:16.918763: step: 948/1541, loss: 0.09391709417104721 2023-01-23 19:30:17.770304: step: 950/1541, loss: 0.00015119176532607526 2023-01-23 19:30:18.684243: step: 952/1541, loss: 0.009611472487449646 2023-01-23 19:30:19.571520: step: 954/1541, loss: 0.0002722803910728544 2023-01-23 19:30:20.447424: step: 956/1541, loss: 0.004874152597039938 2023-01-23 19:30:21.322467: step: 958/1541, loss: 0.08395083248615265 2023-01-23 19:30:22.208890: step: 960/1541, loss: 0.007093718741089106 2023-01-23 19:30:23.062074: step: 962/1541, loss: 0.00015254327445290983 2023-01-23 19:30:23.969214: step: 964/1541, loss: 0.44949519634246826 2023-01-23 19:30:24.851270: step: 966/1541, loss: 0.02724725939333439 2023-01-23 19:30:25.755757: step: 968/1541, loss: 0.007768324110656977 2023-01-23 19:30:26.644792: step: 970/1541, loss: 0.00010242427379125729 2023-01-23 19:30:27.510342: step: 972/1541, loss: 0.007307328283786774 2023-01-23 19:30:28.417333: step: 974/1541, loss: 0.0017697757575660944 2023-01-23 19:30:29.303627: step: 976/1541, loss: 0.0058044851757586 2023-01-23 19:30:30.188650: step: 978/1541, loss: 0.22225190699100494 2023-01-23 19:30:31.172677: step: 980/1541, loss: 0.08936434239149094 2023-01-23 19:30:32.086503: step: 982/1541, loss: 0.11971232295036316 2023-01-23 19:30:32.951111: step: 984/1541, loss: 0.0004975741030648351 2023-01-23 19:30:33.831829: step: 986/1541, loss: 0.18402370810508728 2023-01-23 19:30:34.719852: step: 988/1541, loss: 0.04041365161538124 2023-01-23 19:30:35.665502: step: 990/1541, loss: 0.0008481921395286918 2023-01-23 19:30:36.582161: step: 992/1541, loss: 0.001340103684924543 2023-01-23 19:30:37.430126: step: 994/1541, loss: 5.548047192860395e-05 2023-01-23 19:30:38.368657: step: 996/1541, loss: 0.022457558661699295 2023-01-23 19:30:39.222833: step: 998/1541, loss: 0.000697269628290087 2023-01-23 19:30:40.102787: step: 1000/1541, loss: 0.00023539550602436066 2023-01-23 19:30:40.989470: step: 1002/1541, loss: 0.0035154642537236214 2023-01-23 19:30:41.855261: step: 1004/1541, loss: 0.006089816801249981 2023-01-23 19:30:42.763946: step: 1006/1541, loss: 0.013383904471993446 2023-01-23 19:30:43.679770: step: 1008/1541, loss: 0.013628971762955189 2023-01-23 19:30:44.583904: step: 1010/1541, loss: 0.0013331698719412088 2023-01-23 19:30:45.482971: step: 1012/1541, loss: 0.021131008863449097 2023-01-23 19:30:46.399221: step: 1014/1541, loss: 0.17610907554626465 2023-01-23 19:30:47.305431: step: 1016/1541, loss: 0.0011578609701246023 2023-01-23 19:30:48.196389: step: 1018/1541, loss: 0.00030707119731232524 2023-01-23 19:30:49.105712: step: 1020/1541, loss: 0.0021345550194382668 2023-01-23 19:30:49.981374: step: 1022/1541, loss: 0.0004960161750204861 2023-01-23 19:30:50.912448: step: 1024/1541, loss: 0.00011513743083924055 2023-01-23 19:30:51.819103: step: 1026/1541, loss: 0.0023565092124044895 2023-01-23 19:30:52.726597: step: 1028/1541, loss: 0.03531489148736 2023-01-23 19:30:53.642179: step: 1030/1541, loss: 0.3855813145637512 2023-01-23 19:30:54.567990: step: 1032/1541, loss: 0.0029302206821739674 2023-01-23 19:30:55.450808: step: 1034/1541, loss: 0.08034300059080124 2023-01-23 19:30:56.309177: step: 1036/1541, loss: 0.002323680091649294 2023-01-23 19:30:57.317902: step: 1038/1541, loss: 0.16539233922958374 2023-01-23 19:30:58.182726: step: 1040/1541, loss: 0.008229624480009079 2023-01-23 19:30:59.063086: step: 1042/1541, loss: 0.12001198530197144 2023-01-23 19:30:59.971965: step: 1044/1541, loss: 0.5637159943580627 2023-01-23 19:31:00.854673: step: 1046/1541, loss: 0.04682346433401108 2023-01-23 19:31:01.784741: step: 1048/1541, loss: 0.00012915604747831821 2023-01-23 19:31:02.712276: step: 1050/1541, loss: 0.001034131390042603 2023-01-23 19:31:03.665936: step: 1052/1541, loss: 5.539299672818743e-05 2023-01-23 19:31:04.631474: step: 1054/1541, loss: 0.00011294543219264597 2023-01-23 19:31:05.541520: step: 1056/1541, loss: 0.03711366280913353 2023-01-23 19:31:06.467417: step: 1058/1541, loss: 0.0028402141761034727 2023-01-23 19:31:07.390869: step: 1060/1541, loss: 0.1126108393073082 2023-01-23 19:31:08.283852: step: 1062/1541, loss: 0.00040133611764758825 2023-01-23 19:31:09.170542: step: 1064/1541, loss: 0.003448771545663476 2023-01-23 19:31:10.012589: step: 1066/1541, loss: 0.0018984503112733364 2023-01-23 19:31:10.919008: step: 1068/1541, loss: 0.010551645420491695 2023-01-23 19:31:11.763785: step: 1070/1541, loss: 0.0010762757156044245 2023-01-23 19:31:12.700555: step: 1072/1541, loss: 0.08435997366905212 2023-01-23 19:31:13.604696: step: 1074/1541, loss: 0.00031264303834177554 2023-01-23 19:31:14.524639: step: 1076/1541, loss: 0.0023746134247630835 2023-01-23 19:31:15.366630: step: 1078/1541, loss: 0.00018657516920939088 2023-01-23 19:31:16.259976: step: 1080/1541, loss: 0.0006531771505251527 2023-01-23 19:31:17.138963: step: 1082/1541, loss: 0.00018672370060812682 2023-01-23 19:31:18.000366: step: 1084/1541, loss: 0.00018061733862850815 2023-01-23 19:31:18.849126: step: 1086/1541, loss: 7.40414543543011e-05 2023-01-23 19:31:19.770179: step: 1088/1541, loss: 0.002073802286759019 2023-01-23 19:31:20.692322: step: 1090/1541, loss: 0.009271051734685898 2023-01-23 19:31:21.610645: step: 1092/1541, loss: 0.01763242669403553 2023-01-23 19:31:22.562015: step: 1094/1541, loss: 0.009491140022873878 2023-01-23 19:31:23.530536: step: 1096/1541, loss: 0.0005346795078366995 2023-01-23 19:31:24.446385: step: 1098/1541, loss: 0.00827011652290821 2023-01-23 19:31:25.323820: step: 1100/1541, loss: 0.211802676320076 2023-01-23 19:31:26.218612: step: 1102/1541, loss: 2.9510396416299045e-05 2023-01-23 19:31:27.053334: step: 1104/1541, loss: 0.00033150595845654607 2023-01-23 19:31:28.023568: step: 1106/1541, loss: 0.061365317553281784 2023-01-23 19:31:28.981185: step: 1108/1541, loss: 0.006062671076506376 2023-01-23 19:31:29.839895: step: 1110/1541, loss: 0.0004081530205439776 2023-01-23 19:31:30.692650: step: 1112/1541, loss: 0.3406819999217987 2023-01-23 19:31:31.571547: step: 1114/1541, loss: 0.0028703962452709675 2023-01-23 19:31:32.467699: step: 1116/1541, loss: 0.0001764914923114702 2023-01-23 19:31:33.328976: step: 1118/1541, loss: 0.008104484528303146 2023-01-23 19:31:34.218922: step: 1120/1541, loss: 0.0031089724507182837 2023-01-23 19:31:35.145688: step: 1122/1541, loss: 0.00026043702382594347 2023-01-23 19:31:36.039581: step: 1124/1541, loss: 0.01213852223008871 2023-01-23 19:31:36.925050: step: 1126/1541, loss: 0.0008819971117191017 2023-01-23 19:31:37.839904: step: 1128/1541, loss: 0.0006998040480539203 2023-01-23 19:31:38.781943: step: 1130/1541, loss: 0.08774050325155258 2023-01-23 19:31:39.681163: step: 1132/1541, loss: 0.19385145604610443 2023-01-23 19:31:40.565438: step: 1134/1541, loss: 0.8225146532058716 2023-01-23 19:31:41.560051: step: 1136/1541, loss: 0.037754300981760025 2023-01-23 19:31:42.445780: step: 1138/1541, loss: 0.014988407492637634 2023-01-23 19:31:43.374205: step: 1140/1541, loss: 0.0002921978011727333 2023-01-23 19:31:44.261444: step: 1142/1541, loss: 0.0006224720273166895 2023-01-23 19:31:45.122767: step: 1144/1541, loss: 0.002344567561522126 2023-01-23 19:31:46.105867: step: 1146/1541, loss: 0.11131437122821808 2023-01-23 19:31:46.984561: step: 1148/1541, loss: 0.0027384981513023376 2023-01-23 19:31:47.914714: step: 1150/1541, loss: 0.0024760703090578318 2023-01-23 19:31:48.807329: step: 1152/1541, loss: 0.018828898668289185 2023-01-23 19:31:49.712621: step: 1154/1541, loss: 0.014613858424127102 2023-01-23 19:31:50.594801: step: 1156/1541, loss: 0.27476823329925537 2023-01-23 19:31:51.516776: step: 1158/1541, loss: 0.011175653897225857 2023-01-23 19:31:52.408659: step: 1160/1541, loss: 0.3359881341457367 2023-01-23 19:31:53.298320: step: 1162/1541, loss: 0.00044892251025885344 2023-01-23 19:31:54.328278: step: 1164/1541, loss: 0.0046952650882303715 2023-01-23 19:31:55.219348: step: 1166/1541, loss: 0.0004236675740685314 2023-01-23 19:31:56.083353: step: 1168/1541, loss: 0.001492191688157618 2023-01-23 19:31:56.960480: step: 1170/1541, loss: 0.00578120956197381 2023-01-23 19:31:57.834664: step: 1172/1541, loss: 0.09803566336631775 2023-01-23 19:31:58.714311: step: 1174/1541, loss: 0.44139260053634644 2023-01-23 19:31:59.600294: step: 1176/1541, loss: 0.0007870644330978394 2023-01-23 19:32:00.473670: step: 1178/1541, loss: 0.0005625002086162567 2023-01-23 19:32:01.448785: step: 1180/1541, loss: 0.003099046880379319 2023-01-23 19:32:02.417022: step: 1182/1541, loss: 0.18285426497459412 2023-01-23 19:32:03.335269: step: 1184/1541, loss: 0.00010018013563239947 2023-01-23 19:32:04.206910: step: 1186/1541, loss: 0.00011928086314583197 2023-01-23 19:32:05.114796: step: 1188/1541, loss: 0.0005151189398020506 2023-01-23 19:32:06.021589: step: 1190/1541, loss: 0.00013366490020416677 2023-01-23 19:32:06.955818: step: 1192/1541, loss: 0.023097047582268715 2023-01-23 19:32:07.884998: step: 1194/1541, loss: 0.02138654515147209 2023-01-23 19:32:08.799056: step: 1196/1541, loss: 0.25514739751815796 2023-01-23 19:32:09.764398: step: 1198/1541, loss: 0.17220300436019897 2023-01-23 19:32:10.677833: step: 1200/1541, loss: 0.008893485181033611 2023-01-23 19:32:11.595872: step: 1202/1541, loss: 0.003885545302182436 2023-01-23 19:32:12.508526: step: 1204/1541, loss: 0.07870180159807205 2023-01-23 19:32:13.431623: step: 1206/1541, loss: 0.00030097621493041515 2023-01-23 19:32:14.309947: step: 1208/1541, loss: 0.09603412449359894 2023-01-23 19:32:15.195403: step: 1210/1541, loss: 0.0032669296488165855 2023-01-23 19:32:16.107575: step: 1212/1541, loss: 0.00023728140513412654 2023-01-23 19:32:17.037553: step: 1214/1541, loss: 0.09066935628652573 2023-01-23 19:32:17.960198: step: 1216/1541, loss: 0.04672108590602875 2023-01-23 19:32:18.879517: step: 1218/1541, loss: 0.019007466733455658 2023-01-23 19:32:19.770538: step: 1220/1541, loss: 0.019706495106220245 2023-01-23 19:32:20.650268: step: 1222/1541, loss: 0.060733385384082794 2023-01-23 19:32:21.530255: step: 1224/1541, loss: 0.0009769059251993895 2023-01-23 19:32:22.523731: step: 1226/1541, loss: 0.07905218005180359 2023-01-23 19:32:23.463861: step: 1228/1541, loss: 0.001003420096822083 2023-01-23 19:32:24.358687: step: 1230/1541, loss: 9.053555550053716e-05 2023-01-23 19:32:25.305504: step: 1232/1541, loss: 0.06348244845867157 2023-01-23 19:32:26.179800: step: 1234/1541, loss: 0.003238206496462226 2023-01-23 19:32:27.096150: step: 1236/1541, loss: 0.13679328560829163 2023-01-23 19:32:27.925606: step: 1238/1541, loss: 0.0001379019522573799 2023-01-23 19:32:28.849764: step: 1240/1541, loss: 0.14913804829120636 2023-01-23 19:32:29.792748: step: 1242/1541, loss: 0.00021415649098344147 2023-01-23 19:32:30.703460: step: 1244/1541, loss: 0.004689592868089676 2023-01-23 19:32:31.573388: step: 1246/1541, loss: 2.5150236979243346e-05 2023-01-23 19:32:32.462166: step: 1248/1541, loss: 0.23124724626541138 2023-01-23 19:32:33.375367: step: 1250/1541, loss: 0.0009743962436914444 2023-01-23 19:32:34.253266: step: 1252/1541, loss: 0.000286128866719082 2023-01-23 19:32:35.173054: step: 1254/1541, loss: 0.017527969554066658 2023-01-23 19:32:36.092320: step: 1256/1541, loss: 0.0034991756547242403 2023-01-23 19:32:37.002672: step: 1258/1541, loss: 0.003906577825546265 2023-01-23 19:32:37.904575: step: 1260/1541, loss: 0.00027637724997475743 2023-01-23 19:32:38.840427: step: 1262/1541, loss: 9.758372470969334e-05 2023-01-23 19:32:39.767876: step: 1264/1541, loss: 0.0018804841674864292 2023-01-23 19:32:40.671879: step: 1266/1541, loss: 0.016998829320073128 2023-01-23 19:32:41.568158: step: 1268/1541, loss: 0.05007085204124451 2023-01-23 19:32:42.489917: step: 1270/1541, loss: 0.018261313438415527 2023-01-23 19:32:43.361880: step: 1272/1541, loss: 0.0016496884636580944 2023-01-23 19:32:44.225247: step: 1274/1541, loss: 0.00021785199351143092 2023-01-23 19:32:45.078673: step: 1276/1541, loss: 0.003995390143245459 2023-01-23 19:32:45.974985: step: 1278/1541, loss: 0.00046311839832924306 2023-01-23 19:32:46.855351: step: 1280/1541, loss: 0.0061610909178853035 2023-01-23 19:32:47.734156: step: 1282/1541, loss: 0.0008613745449110866 2023-01-23 19:32:48.662849: step: 1284/1541, loss: 0.02560526318848133 2023-01-23 19:32:49.543551: step: 1286/1541, loss: 0.08365069329738617 2023-01-23 19:32:50.421633: step: 1288/1541, loss: 0.10404693335294724 2023-01-23 19:32:51.343798: step: 1290/1541, loss: 0.07360729575157166 2023-01-23 19:32:52.200316: step: 1292/1541, loss: 9.37479198910296e-05 2023-01-23 19:32:53.058162: step: 1294/1541, loss: 0.00013981226948089898 2023-01-23 19:32:53.977285: step: 1296/1541, loss: 0.26809436082839966 2023-01-23 19:32:54.825910: step: 1298/1541, loss: 0.0002995503891725093 2023-01-23 19:32:55.709224: step: 1300/1541, loss: 0.0005029637250117958 2023-01-23 19:32:56.634349: step: 1302/1541, loss: 3.690419907798059e-05 2023-01-23 19:32:57.554523: step: 1304/1541, loss: 0.032646894454956055 2023-01-23 19:32:58.583839: step: 1306/1541, loss: 0.20814387500286102 2023-01-23 19:32:59.525327: step: 1308/1541, loss: 0.11671449989080429 2023-01-23 19:33:00.396445: step: 1310/1541, loss: 0.00629847589880228 2023-01-23 19:33:01.298312: step: 1312/1541, loss: 0.00023313131532631814 2023-01-23 19:33:02.282278: step: 1314/1541, loss: 0.08708832412958145 2023-01-23 19:33:03.195035: step: 1316/1541, loss: 0.13611207902431488 2023-01-23 19:33:04.108962: step: 1318/1541, loss: 7.637314411113039e-05 2023-01-23 19:33:04.991940: step: 1320/1541, loss: 0.3955792486667633 2023-01-23 19:33:05.962275: step: 1322/1541, loss: 0.011554191820323467 2023-01-23 19:33:06.873837: step: 1324/1541, loss: 0.0005845529376529157 2023-01-23 19:33:07.758558: step: 1326/1541, loss: 0.1138959750533104 2023-01-23 19:33:08.742153: step: 1328/1541, loss: 0.0007909265113994479 2023-01-23 19:33:09.625438: step: 1330/1541, loss: 5.002352554583922e-05 2023-01-23 19:33:10.571561: step: 1332/1541, loss: 0.0007668096804991364 2023-01-23 19:33:11.461933: step: 1334/1541, loss: 0.005555364303290844 2023-01-23 19:33:12.383318: step: 1336/1541, loss: 0.048405345529317856 2023-01-23 19:33:13.290390: step: 1338/1541, loss: 0.0018100111046805978 2023-01-23 19:33:14.180754: step: 1340/1541, loss: 0.18476688861846924 2023-01-23 19:33:15.096273: step: 1342/1541, loss: 0.12026525288820267 2023-01-23 19:33:15.987755: step: 1344/1541, loss: 0.1889609396457672 2023-01-23 19:33:16.955570: step: 1346/1541, loss: 0.01973024569451809 2023-01-23 19:33:17.844381: step: 1348/1541, loss: 0.0430428609251976 2023-01-23 19:33:18.740960: step: 1350/1541, loss: 0.07587207853794098 2023-01-23 19:33:19.625291: step: 1352/1541, loss: 0.24027009308338165 2023-01-23 19:33:20.528872: step: 1354/1541, loss: 0.0006667295820079744 2023-01-23 19:33:21.418005: step: 1356/1541, loss: 0.7436737418174744 2023-01-23 19:33:22.315009: step: 1358/1541, loss: 0.00018531858222559094 2023-01-23 19:33:23.270886: step: 1360/1541, loss: 0.004033934324979782 2023-01-23 19:33:24.197774: step: 1362/1541, loss: 0.0025201220996677876 2023-01-23 19:33:25.090321: step: 1364/1541, loss: 0.00011986183380940929 2023-01-23 19:33:25.966082: step: 1366/1541, loss: 0.03162260353565216 2023-01-23 19:33:26.855809: step: 1368/1541, loss: 0.003942593466490507 2023-01-23 19:33:27.742144: step: 1370/1541, loss: 0.025700824335217476 2023-01-23 19:33:28.611066: step: 1372/1541, loss: 0.007562682498246431 2023-01-23 19:33:29.501496: step: 1374/1541, loss: 0.04858323186635971 2023-01-23 19:33:30.420379: step: 1376/1541, loss: 0.006066053174436092 2023-01-23 19:33:31.294619: step: 1378/1541, loss: 0.07889886945486069 2023-01-23 19:33:32.212350: step: 1380/1541, loss: 0.0014267901424318552 2023-01-23 19:33:33.084920: step: 1382/1541, loss: 0.018457286059856415 2023-01-23 19:33:33.939948: step: 1384/1541, loss: 0.001795439631678164 2023-01-23 19:33:34.882791: step: 1386/1541, loss: 0.011448616161942482 2023-01-23 19:33:35.746095: step: 1388/1541, loss: 0.006462187971919775 2023-01-23 19:33:36.676240: step: 1390/1541, loss: 0.12324553728103638 2023-01-23 19:33:37.559647: step: 1392/1541, loss: 0.02282160520553589 2023-01-23 19:33:38.433417: step: 1394/1541, loss: 0.0020124404691159725 2023-01-23 19:33:39.367030: step: 1396/1541, loss: 0.04271768033504486 2023-01-23 19:33:40.286606: step: 1398/1541, loss: 0.31850793957710266 2023-01-23 19:33:41.171708: step: 1400/1541, loss: 0.000402073870645836 2023-01-23 19:33:42.032966: step: 1402/1541, loss: 0.20552991330623627 2023-01-23 19:33:42.910158: step: 1404/1541, loss: 0.004138223826885223 2023-01-23 19:33:43.836782: step: 1406/1541, loss: 0.006347555201500654 2023-01-23 19:33:44.722035: step: 1408/1541, loss: 0.0044988105073571205 2023-01-23 19:33:45.641281: step: 1410/1541, loss: 0.00041617773240432143 2023-01-23 19:33:46.556778: step: 1412/1541, loss: 0.015239167958498001 2023-01-23 19:33:47.443152: step: 1414/1541, loss: 0.002676410833373666 2023-01-23 19:33:48.338099: step: 1416/1541, loss: 0.00038099527591839433 2023-01-23 19:33:49.237946: step: 1418/1541, loss: 0.007833126001060009 2023-01-23 19:33:50.115134: step: 1420/1541, loss: 0.00017707730876281857 2023-01-23 19:33:51.014351: step: 1422/1541, loss: 0.014296198263764381 2023-01-23 19:33:51.900414: step: 1424/1541, loss: 0.0003177147009409964 2023-01-23 19:33:52.776803: step: 1426/1541, loss: 0.11100509762763977 2023-01-23 19:33:53.722291: step: 1428/1541, loss: 0.05982479825615883 2023-01-23 19:33:54.605998: step: 1430/1541, loss: 0.019176039844751358 2023-01-23 19:33:55.534619: step: 1432/1541, loss: 0.007897756062448025 2023-01-23 19:33:56.491164: step: 1434/1541, loss: 0.0004699050623457879 2023-01-23 19:33:57.407241: step: 1436/1541, loss: 0.021988771855831146 2023-01-23 19:33:58.283928: step: 1438/1541, loss: 0.0044639939442276955 2023-01-23 19:33:59.158775: step: 1440/1541, loss: 0.008003191091120243 2023-01-23 19:34:00.046725: step: 1442/1541, loss: 0.3034881353378296 2023-01-23 19:34:00.946910: step: 1444/1541, loss: 0.019930928945541382 2023-01-23 19:34:01.902134: step: 1446/1541, loss: 0.20588819682598114 2023-01-23 19:34:02.777804: step: 1448/1541, loss: 0.43416810035705566 2023-01-23 19:34:03.717153: step: 1450/1541, loss: 0.006313880439847708 2023-01-23 19:34:04.641021: step: 1452/1541, loss: 0.0017959552351385355 2023-01-23 19:34:05.560806: step: 1454/1541, loss: 0.0004727238556370139 2023-01-23 19:34:06.460768: step: 1456/1541, loss: 0.00030718400375917554 2023-01-23 19:34:07.367151: step: 1458/1541, loss: 0.004478264134377241 2023-01-23 19:34:08.272852: step: 1460/1541, loss: 0.008784702979028225 2023-01-23 19:34:09.139843: step: 1462/1541, loss: 0.140488862991333 2023-01-23 19:34:10.043431: step: 1464/1541, loss: 0.010387244634330273 2023-01-23 19:34:10.959652: step: 1466/1541, loss: 0.0004713377566076815 2023-01-23 19:34:11.822667: step: 1468/1541, loss: 0.11311785876750946 2023-01-23 19:34:12.694157: step: 1470/1541, loss: 0.07953192293643951 2023-01-23 19:34:13.572702: step: 1472/1541, loss: 0.010353957302868366 2023-01-23 19:34:14.461424: step: 1474/1541, loss: 0.05159437283873558 2023-01-23 19:34:15.413586: step: 1476/1541, loss: 0.0017449244624003768 2023-01-23 19:34:16.286762: step: 1478/1541, loss: 0.0019457241287454963 2023-01-23 19:34:17.173235: step: 1480/1541, loss: 2.219991445541382 2023-01-23 19:34:18.076274: step: 1482/1541, loss: 0.0005385245895013213 2023-01-23 19:34:18.933405: step: 1484/1541, loss: 0.0018761875107884407 2023-01-23 19:34:19.805061: step: 1486/1541, loss: 0.002904691733419895 2023-01-23 19:34:20.631405: step: 1488/1541, loss: 0.0011080580297857523 2023-01-23 19:34:21.591212: step: 1490/1541, loss: 0.15883095562458038 2023-01-23 19:34:22.488539: step: 1492/1541, loss: 0.09567072242498398 2023-01-23 19:34:23.405879: step: 1494/1541, loss: 0.0008037827210500836 2023-01-23 19:34:24.295159: step: 1496/1541, loss: 0.0048533412627875805 2023-01-23 19:34:25.209921: step: 1498/1541, loss: 0.0010113655589520931 2023-01-23 19:34:26.101913: step: 1500/1541, loss: 0.01489954348653555 2023-01-23 19:34:27.042838: step: 1502/1541, loss: 0.02150113508105278 2023-01-23 19:34:27.954089: step: 1504/1541, loss: 0.003013217356055975 2023-01-23 19:34:28.826052: step: 1506/1541, loss: 0.001815954688936472 2023-01-23 19:34:29.698763: step: 1508/1541, loss: 0.0003062652249354869 2023-01-23 19:34:30.599479: step: 1510/1541, loss: 0.000441129261162132 2023-01-23 19:34:31.505228: step: 1512/1541, loss: 0.00029862308292649686 2023-01-23 19:34:32.420592: step: 1514/1541, loss: 8.11105128377676e-05 2023-01-23 19:34:33.333710: step: 1516/1541, loss: 0.1651371270418167 2023-01-23 19:34:34.244889: step: 1518/1541, loss: 0.4837077558040619 2023-01-23 19:34:35.150821: step: 1520/1541, loss: 0.003563877195119858 2023-01-23 19:34:36.093514: step: 1522/1541, loss: 0.006272983271628618 2023-01-23 19:34:36.932514: step: 1524/1541, loss: 0.00038340891478583217 2023-01-23 19:34:37.825851: step: 1526/1541, loss: 0.0108925960958004 2023-01-23 19:34:38.698440: step: 1528/1541, loss: 0.06638489663600922 2023-01-23 19:34:39.597061: step: 1530/1541, loss: 0.17003503441810608 2023-01-23 19:34:40.517035: step: 1532/1541, loss: 0.0012413341319188476 2023-01-23 19:34:41.455248: step: 1534/1541, loss: 0.00016947687254287302 2023-01-23 19:34:42.361689: step: 1536/1541, loss: 0.1287035495042801 2023-01-23 19:34:43.244160: step: 1538/1541, loss: 0.0013404989149421453 2023-01-23 19:34:44.137047: step: 1540/1541, loss: 0.013521643355488777 2023-01-23 19:34:45.044659: step: 1542/1541, loss: 0.005721050314605236 2023-01-23 19:34:45.946438: step: 1544/1541, loss: 0.042868517339229584 2023-01-23 19:34:46.862544: step: 1546/1541, loss: 0.0021631084382534027 2023-01-23 19:34:47.841518: step: 1548/1541, loss: 0.08638039231300354 2023-01-23 19:34:48.764823: step: 1550/1541, loss: 0.005667935125529766 2023-01-23 19:34:49.653967: step: 1552/1541, loss: 0.0010697015095502138 2023-01-23 19:34:50.519759: step: 1554/1541, loss: 0.017823664471507072 2023-01-23 19:34:51.387075: step: 1556/1541, loss: 0.011648287996649742 2023-01-23 19:34:52.238306: step: 1558/1541, loss: 0.0003727045841515064 2023-01-23 19:34:53.094937: step: 1560/1541, loss: 0.0014414973556995392 2023-01-23 19:34:54.041017: step: 1562/1541, loss: 0.0018082733731716871 2023-01-23 19:34:54.979781: step: 1564/1541, loss: 0.013339833356440067 2023-01-23 19:34:55.871589: step: 1566/1541, loss: 0.7372056841850281 2023-01-23 19:34:56.783270: step: 1568/1541, loss: 0.01856830343604088 2023-01-23 19:34:57.723737: step: 1570/1541, loss: 0.0001488158741267398 2023-01-23 19:34:58.605144: step: 1572/1541, loss: 0.002595486119389534 2023-01-23 19:34:59.513295: step: 1574/1541, loss: 0.0005970253841951489 2023-01-23 19:35:00.461321: step: 1576/1541, loss: 0.0003489162481855601 2023-01-23 19:35:01.389273: step: 1578/1541, loss: 0.0005704618524760008 2023-01-23 19:35:02.377393: step: 1580/1541, loss: 0.0017154577653855085 2023-01-23 19:35:03.283273: step: 1582/1541, loss: 0.052374303340911865 2023-01-23 19:35:04.189255: step: 1584/1541, loss: 0.043962664902210236 2023-01-23 19:35:05.079510: step: 1586/1541, loss: 0.002043994842097163 2023-01-23 19:35:05.990673: step: 1588/1541, loss: 0.0010926120448857546 2023-01-23 19:35:06.868591: step: 1590/1541, loss: 0.00026892125606536865 2023-01-23 19:35:07.740672: step: 1592/1541, loss: 0.20673424005508423 2023-01-23 19:35:08.637258: step: 1594/1541, loss: 0.0029736580327153206 2023-01-23 19:35:09.521378: step: 1596/1541, loss: 0.06742477416992188 2023-01-23 19:35:10.400216: step: 1598/1541, loss: 0.00027840171242132783 2023-01-23 19:35:11.385045: step: 1600/1541, loss: 0.0003027996572200209 2023-01-23 19:35:12.286328: step: 1602/1541, loss: 7.70072074374184e-05 2023-01-23 19:35:13.127167: step: 1604/1541, loss: 3.131776611553505e-05 2023-01-23 19:35:13.982385: step: 1606/1541, loss: 9.774132922757417e-05 2023-01-23 19:35:14.868701: step: 1608/1541, loss: 0.00144388887565583 2023-01-23 19:35:15.726115: step: 1610/1541, loss: 0.0009453344391658902 2023-01-23 19:35:16.620981: step: 1612/1541, loss: 0.542276918888092 2023-01-23 19:35:17.495133: step: 1614/1541, loss: 0.049233753234148026 2023-01-23 19:35:18.406966: step: 1616/1541, loss: 0.0010762711754068732 2023-01-23 19:35:19.269300: step: 1618/1541, loss: 0.001399463159032166 2023-01-23 19:35:20.200326: step: 1620/1541, loss: 0.06561420857906342 2023-01-23 19:35:21.115171: step: 1622/1541, loss: 0.19474589824676514 2023-01-23 19:35:22.007374: step: 1624/1541, loss: 0.27401798963546753 2023-01-23 19:35:22.886425: step: 1626/1541, loss: 0.05046159774065018 2023-01-23 19:35:23.754893: step: 1628/1541, loss: 0.002245624316856265 2023-01-23 19:35:24.615345: step: 1630/1541, loss: 0.004938756115734577 2023-01-23 19:35:25.513275: step: 1632/1541, loss: 0.00010187643056269735 2023-01-23 19:35:26.428012: step: 1634/1541, loss: 0.0022564043756574392 2023-01-23 19:35:27.293261: step: 1636/1541, loss: 0.03562038391828537 2023-01-23 19:35:28.157473: step: 1638/1541, loss: 0.00319181801751256 2023-01-23 19:35:29.084204: step: 1640/1541, loss: 0.00033887059544213116 2023-01-23 19:35:30.013144: step: 1642/1541, loss: 0.0006348302122205496 2023-01-23 19:35:30.948874: step: 1644/1541, loss: 0.005426627583801746 2023-01-23 19:35:31.892539: step: 1646/1541, loss: 0.0002999742282554507 2023-01-23 19:35:32.791643: step: 1648/1541, loss: 1.6859788956935517e-05 2023-01-23 19:35:33.720017: step: 1650/1541, loss: 0.00019998203788418323 2023-01-23 19:35:34.593568: step: 1652/1541, loss: 0.0011579348938539624 2023-01-23 19:35:35.477886: step: 1654/1541, loss: 0.001051741768606007 2023-01-23 19:35:36.368586: step: 1656/1541, loss: 0.0014261356554925442 2023-01-23 19:35:37.255716: step: 1658/1541, loss: 0.017679918557405472 2023-01-23 19:35:38.161515: step: 1660/1541, loss: 0.001166842645034194 2023-01-23 19:35:39.053151: step: 1662/1541, loss: 0.0031759561970829964 2023-01-23 19:35:39.983525: step: 1664/1541, loss: 0.000489158439449966 2023-01-23 19:35:40.862349: step: 1666/1541, loss: 0.001492423820309341 2023-01-23 19:35:41.803055: step: 1668/1541, loss: 0.001692136051133275 2023-01-23 19:35:42.694381: step: 1670/1541, loss: 6.228103302419186e-05 2023-01-23 19:35:43.579586: step: 1672/1541, loss: 0.04422502964735031 2023-01-23 19:35:44.497589: step: 1674/1541, loss: 0.27372631430625916 2023-01-23 19:35:45.488381: step: 1676/1541, loss: 0.00023586268071085215 2023-01-23 19:35:46.358861: step: 1678/1541, loss: 0.000892651965841651 2023-01-23 19:35:47.236182: step: 1680/1541, loss: 0.001019969000481069 2023-01-23 19:35:48.115709: step: 1682/1541, loss: 0.00023947987938299775 2023-01-23 19:35:49.002502: step: 1684/1541, loss: 0.00040535887819714844 2023-01-23 19:35:49.825628: step: 1686/1541, loss: 0.004146584775298834 2023-01-23 19:35:50.735233: step: 1688/1541, loss: 0.008465040475130081 2023-01-23 19:35:51.583865: step: 1690/1541, loss: 0.002446940867230296 2023-01-23 19:35:52.458329: step: 1692/1541, loss: 0.00010599707457004115 2023-01-23 19:35:53.327388: step: 1694/1541, loss: 0.06674487143754959 2023-01-23 19:35:54.213388: step: 1696/1541, loss: 0.0003003712045028806 2023-01-23 19:35:55.114181: step: 1698/1541, loss: 0.0317075289785862 2023-01-23 19:35:55.999720: step: 1700/1541, loss: 1.1551873683929443 2023-01-23 19:35:56.976270: step: 1702/1541, loss: 8.935837831813842e-05 2023-01-23 19:35:57.839292: step: 1704/1541, loss: 0.005736171267926693 2023-01-23 19:35:58.749708: step: 1706/1541, loss: 0.021891318261623383 2023-01-23 19:35:59.627370: step: 1708/1541, loss: 0.0004395728465169668 2023-01-23 19:36:00.625766: step: 1710/1541, loss: 1.0706671476364136 2023-01-23 19:36:01.481961: step: 1712/1541, loss: 4.8681329644750804e-05 2023-01-23 19:36:02.361362: step: 1714/1541, loss: 0.00107289245352149 2023-01-23 19:36:03.231708: step: 1716/1541, loss: 0.0001350548554910347 2023-01-23 19:36:04.115524: step: 1718/1541, loss: 0.029285043478012085 2023-01-23 19:36:04.995445: step: 1720/1541, loss: 7.797591388225555e-05 2023-01-23 19:36:05.903955: step: 1722/1541, loss: 0.1310400366783142 2023-01-23 19:36:06.824986: step: 1724/1541, loss: 0.031169245019555092 2023-01-23 19:36:07.714027: step: 1726/1541, loss: 0.2165357619524002 2023-01-23 19:36:08.584861: step: 1728/1541, loss: 0.014413748867809772 2023-01-23 19:36:09.461492: step: 1730/1541, loss: 0.00012183935177745298 2023-01-23 19:36:10.357574: step: 1732/1541, loss: 0.03267902508378029 2023-01-23 19:36:11.310846: step: 1734/1541, loss: 0.0013902649516239762 2023-01-23 19:36:12.222787: step: 1736/1541, loss: 0.0029403010848909616 2023-01-23 19:36:13.119606: step: 1738/1541, loss: 0.0026663970202207565 2023-01-23 19:36:14.028512: step: 1740/1541, loss: 0.0003310385509394109 2023-01-23 19:36:14.911890: step: 1742/1541, loss: 0.0002731321728788316 2023-01-23 19:36:15.769698: step: 1744/1541, loss: 0.006991517264395952 2023-01-23 19:36:16.700752: step: 1746/1541, loss: 0.01707383058965206 2023-01-23 19:36:17.684703: step: 1748/1541, loss: 0.016166292130947113 2023-01-23 19:36:18.571641: step: 1750/1541, loss: 0.012498113326728344 2023-01-23 19:36:19.458426: step: 1752/1541, loss: 0.0016700312262400985 2023-01-23 19:36:20.324841: step: 1754/1541, loss: 0.028430044651031494 2023-01-23 19:36:21.162320: step: 1756/1541, loss: 0.00017605451284907758 2023-01-23 19:36:22.098398: step: 1758/1541, loss: 0.001168915769085288 2023-01-23 19:36:22.986583: step: 1760/1541, loss: 0.07905610650777817 2023-01-23 19:36:23.858914: step: 1762/1541, loss: 0.00039340241346508265 2023-01-23 19:36:24.756335: step: 1764/1541, loss: 0.0012690394651144743 2023-01-23 19:36:25.645078: step: 1766/1541, loss: 0.14387419819831848 2023-01-23 19:36:26.570664: step: 1768/1541, loss: 0.020762694999575615 2023-01-23 19:36:27.438801: step: 1770/1541, loss: 0.0010869153775274754 2023-01-23 19:36:28.309453: step: 1772/1541, loss: 0.0009434508974663913 2023-01-23 19:36:29.228082: step: 1774/1541, loss: 0.0008598488057032228 2023-01-23 19:36:30.112754: step: 1776/1541, loss: 0.02055230177938938 2023-01-23 19:36:31.105423: step: 1778/1541, loss: 0.00022248324239626527 2023-01-23 19:36:32.050936: step: 1780/1541, loss: 4.3493178964126855e-05 2023-01-23 19:36:32.929335: step: 1782/1541, loss: 0.0011945851147174835 2023-01-23 19:36:33.806053: step: 1784/1541, loss: 1.4122422933578491 2023-01-23 19:36:34.711686: step: 1786/1541, loss: 0.001292603905312717 2023-01-23 19:36:35.598488: step: 1788/1541, loss: 7.242213177960366e-05 2023-01-23 19:36:36.519902: step: 1790/1541, loss: 0.0033442245330661535 2023-01-23 19:36:37.539765: step: 1792/1541, loss: 0.00015456175606232136 2023-01-23 19:36:38.399271: step: 1794/1541, loss: 0.0007308347267098725 2023-01-23 19:36:39.264117: step: 1796/1541, loss: 0.00053423922508955 2023-01-23 19:36:40.199872: step: 1798/1541, loss: 1.468074917793274 2023-01-23 19:36:41.121721: step: 1800/1541, loss: 0.000254006969043985 2023-01-23 19:36:42.007573: step: 1802/1541, loss: 0.0025990314316004515 2023-01-23 19:36:42.897554: step: 1804/1541, loss: 0.0002545449824538082 2023-01-23 19:36:43.800171: step: 1806/1541, loss: 5.622435492114164e-05 2023-01-23 19:36:44.723192: step: 1808/1541, loss: 0.24705512821674347 2023-01-23 19:36:45.584834: step: 1810/1541, loss: 0.03927737846970558 2023-01-23 19:36:46.502772: step: 1812/1541, loss: 0.00010212750930804759 2023-01-23 19:36:47.372523: step: 1814/1541, loss: 0.00026522911502979696 2023-01-23 19:36:48.258949: step: 1816/1541, loss: 0.591282844543457 2023-01-23 19:36:49.165892: step: 1818/1541, loss: 0.24056775867938995 2023-01-23 19:36:50.098702: step: 1820/1541, loss: 0.056793056428432465 2023-01-23 19:36:51.035884: step: 1822/1541, loss: 0.003445268142968416 2023-01-23 19:36:51.923882: step: 1824/1541, loss: 0.017099514603614807 2023-01-23 19:36:52.837956: step: 1826/1541, loss: 0.03485588729381561 2023-01-23 19:36:53.711851: step: 1828/1541, loss: 0.02632046677172184 2023-01-23 19:36:54.621547: step: 1830/1541, loss: 0.0032610807102173567 2023-01-23 19:36:55.497810: step: 1832/1541, loss: 0.12928763031959534 2023-01-23 19:36:56.372817: step: 1834/1541, loss: 0.18920259177684784 2023-01-23 19:36:57.230104: step: 1836/1541, loss: 0.0014664381742477417 2023-01-23 19:36:58.123459: step: 1838/1541, loss: 0.0009339434327557683 2023-01-23 19:36:58.997837: step: 1840/1541, loss: 0.06997893005609512 2023-01-23 19:36:59.878158: step: 1842/1541, loss: 0.02643461711704731 2023-01-23 19:37:00.739226: step: 1844/1541, loss: 0.1369856595993042 2023-01-23 19:37:01.651246: step: 1846/1541, loss: 0.009882133454084396 2023-01-23 19:37:02.616151: step: 1848/1541, loss: 0.10709737986326218 2023-01-23 19:37:03.497460: step: 1850/1541, loss: 0.3126679062843323 2023-01-23 19:37:04.394546: step: 1852/1541, loss: 0.0053770411759614944 2023-01-23 19:37:05.259074: step: 1854/1541, loss: 0.22694025933742523 2023-01-23 19:37:06.180861: step: 1856/1541, loss: 0.015870915725827217 2023-01-23 19:37:07.076183: step: 1858/1541, loss: 5.723436697735451e-05 2023-01-23 19:37:07.927805: step: 1860/1541, loss: 0.004685687832534313 2023-01-23 19:37:08.776928: step: 1862/1541, loss: 0.01234241109341383 2023-01-23 19:37:09.670054: step: 1864/1541, loss: 9.64236241998151e-05 2023-01-23 19:37:10.616109: step: 1866/1541, loss: 0.00017507703159935772 2023-01-23 19:37:11.509568: step: 1868/1541, loss: 0.043235812336206436 2023-01-23 19:37:12.383271: step: 1870/1541, loss: 0.04642597213387489 2023-01-23 19:37:13.264038: step: 1872/1541, loss: 0.0024859700351953506 2023-01-23 19:37:14.153658: step: 1874/1541, loss: 0.024416543543338776 2023-01-23 19:37:15.062283: step: 1876/1541, loss: 1.752276330080349e-05 2023-01-23 19:37:16.000419: step: 1878/1541, loss: 0.040659476071596146 2023-01-23 19:37:16.881693: step: 1880/1541, loss: 0.005303018726408482 2023-01-23 19:37:17.797960: step: 1882/1541, loss: 0.0015352295013144612 2023-01-23 19:37:18.681454: step: 1884/1541, loss: 0.00011740947229554877 2023-01-23 19:37:19.585561: step: 1886/1541, loss: 0.0002265995426569134 2023-01-23 19:37:20.470471: step: 1888/1541, loss: 0.00035864097299054265 2023-01-23 19:37:21.355436: step: 1890/1541, loss: 6.696504715364426e-05 2023-01-23 19:37:22.257701: step: 1892/1541, loss: 0.0001698934065643698 2023-01-23 19:37:23.117025: step: 1894/1541, loss: 4.9151665734825656e-05 2023-01-23 19:37:24.025322: step: 1896/1541, loss: 0.0004819756140932441 2023-01-23 19:37:24.882425: step: 1898/1541, loss: 0.0006382321589626372 2023-01-23 19:37:25.758532: step: 1900/1541, loss: -8.446490937785711e-06 2023-01-23 19:37:26.600454: step: 1902/1541, loss: 0.0002764714590739459 2023-01-23 19:37:27.523723: step: 1904/1541, loss: 0.11557739973068237 2023-01-23 19:37:28.397274: step: 1906/1541, loss: 0.004555385559797287 2023-01-23 19:37:29.325377: step: 1908/1541, loss: 0.0001412715791957453 2023-01-23 19:37:30.218411: step: 1910/1541, loss: 0.00114439707249403 2023-01-23 19:37:31.085229: step: 1912/1541, loss: 0.00043113273568451405 2023-01-23 19:37:31.976275: step: 1914/1541, loss: 0.6587910652160645 2023-01-23 19:37:32.863067: step: 1916/1541, loss: 0.5358087420463562 2023-01-23 19:37:33.737927: step: 1918/1541, loss: 0.0052065253257751465 2023-01-23 19:37:34.591315: step: 1920/1541, loss: 0.0006185645470395684 2023-01-23 19:37:35.449067: step: 1922/1541, loss: 0.0029679196886718273 2023-01-23 19:37:36.369504: step: 1924/1541, loss: 0.26273593306541443 2023-01-23 19:37:37.261917: step: 1926/1541, loss: 0.03802986815571785 2023-01-23 19:37:38.164335: step: 1928/1541, loss: 0.010638143867254257 2023-01-23 19:37:39.039306: step: 1930/1541, loss: 0.12260273098945618 2023-01-23 19:37:39.907725: step: 1932/1541, loss: 0.003357481909915805 2023-01-23 19:37:40.784619: step: 1934/1541, loss: 6.681421655230224e-05 2023-01-23 19:37:41.712916: step: 1936/1541, loss: 0.000255794235272333 2023-01-23 19:37:42.604653: step: 1938/1541, loss: 0.0059972223825752735 2023-01-23 19:37:43.474341: step: 1940/1541, loss: 0.00017668507643975317 2023-01-23 19:37:44.436362: step: 1942/1541, loss: 0.0007474934100173414 2023-01-23 19:37:45.322315: step: 1944/1541, loss: 0.0004375749849714339 2023-01-23 19:37:46.227406: step: 1946/1541, loss: 0.0033090468496084213 2023-01-23 19:37:47.122903: step: 1948/1541, loss: 0.00052937789587304 2023-01-23 19:37:48.010501: step: 1950/1541, loss: 0.04808099567890167 2023-01-23 19:37:48.912838: step: 1952/1541, loss: 6.108092202339321e-05 2023-01-23 19:37:49.811566: step: 1954/1541, loss: 0.00020222031162120402 2023-01-23 19:37:50.699378: step: 1956/1541, loss: 0.15825681388378143 2023-01-23 19:37:51.574130: step: 1958/1541, loss: 0.05518701672554016 2023-01-23 19:37:52.444406: step: 1960/1541, loss: 0.029140686616301537 2023-01-23 19:37:53.330219: step: 1962/1541, loss: 0.10149417817592621 2023-01-23 19:37:54.244681: step: 1964/1541, loss: 0.001836207346059382 2023-01-23 19:37:55.170569: step: 1966/1541, loss: 0.0012523784535005689 2023-01-23 19:37:56.057914: step: 1968/1541, loss: 0.027284665033221245 2023-01-23 19:37:57.020143: step: 1970/1541, loss: 0.03000395931303501 2023-01-23 19:37:57.898574: step: 1972/1541, loss: 0.045175664126873016 2023-01-23 19:37:58.792750: step: 1974/1541, loss: 0.00015594527940265834 2023-01-23 19:37:59.745572: step: 1976/1541, loss: 0.16443786025047302 2023-01-23 19:38:00.618712: step: 1978/1541, loss: 5.5091833928599954e-05 2023-01-23 19:38:01.514967: step: 1980/1541, loss: 0.0004635256773326546 2023-01-23 19:38:02.398602: step: 1982/1541, loss: 0.0025627214927226305 2023-01-23 19:38:03.276931: step: 1984/1541, loss: 0.38220101594924927 2023-01-23 19:38:04.184946: step: 1986/1541, loss: 0.16886167228221893 2023-01-23 19:38:05.096354: step: 1988/1541, loss: 0.22916707396507263 2023-01-23 19:38:05.964380: step: 1990/1541, loss: 0.0010517379269003868 2023-01-23 19:38:06.828227: step: 1992/1541, loss: 0.0002449653984513134 2023-01-23 19:38:07.785653: step: 1994/1541, loss: 0.10999580472707748 2023-01-23 19:38:08.671459: step: 1996/1541, loss: 0.07282693684101105 2023-01-23 19:38:09.496466: step: 1998/1541, loss: 4.2051815398735926e-05 2023-01-23 19:38:10.360981: step: 2000/1541, loss: 0.0004952703602612019 2023-01-23 19:38:11.268326: step: 2002/1541, loss: 0.008344143629074097 2023-01-23 19:38:12.174821: step: 2004/1541, loss: 0.002541928319260478 2023-01-23 19:38:13.106609: step: 2006/1541, loss: 0.00024569989182054996 2023-01-23 19:38:13.977173: step: 2008/1541, loss: 0.00013596523785963655 2023-01-23 19:38:14.903996: step: 2010/1541, loss: 0.2424147129058838 2023-01-23 19:38:15.767134: step: 2012/1541, loss: 0.001728618866764009 2023-01-23 19:38:16.692361: step: 2014/1541, loss: 0.00018023565644398332 2023-01-23 19:38:17.577739: step: 2016/1541, loss: 0.0001714690588414669 2023-01-23 19:38:18.486122: step: 2018/1541, loss: 0.019365355372428894 2023-01-23 19:38:19.362199: step: 2020/1541, loss: 0.0005192872486077249 2023-01-23 19:38:20.225486: step: 2022/1541, loss: 0.0012824086006730795 2023-01-23 19:38:21.127287: step: 2024/1541, loss: 0.014171065762639046 2023-01-23 19:38:21.972606: step: 2026/1541, loss: 0.0004142946854699403 2023-01-23 19:38:22.886114: step: 2028/1541, loss: 0.27928584814071655 2023-01-23 19:38:23.813941: step: 2030/1541, loss: 0.014251954853534698 2023-01-23 19:38:24.679438: step: 2032/1541, loss: 0.0002604915644042194 2023-01-23 19:38:25.592047: step: 2034/1541, loss: 0.008158047683537006 2023-01-23 19:38:26.485748: step: 2036/1541, loss: 0.04578536003828049 2023-01-23 19:38:27.362980: step: 2038/1541, loss: 0.0005063217831775546 2023-01-23 19:38:28.246381: step: 2040/1541, loss: 0.00012966606300324202 2023-01-23 19:38:29.107524: step: 2042/1541, loss: 5.7382843806408346e-05 2023-01-23 19:38:29.992112: step: 2044/1541, loss: 0.00484971608966589 2023-01-23 19:38:30.866804: step: 2046/1541, loss: 0.05349759757518768 2023-01-23 19:38:31.728249: step: 2048/1541, loss: 0.2669268846511841 2023-01-23 19:38:32.687070: step: 2050/1541, loss: 0.0024159064050763845 2023-01-23 19:38:33.556001: step: 2052/1541, loss: 0.00019505745149217546 2023-01-23 19:38:34.433905: step: 2054/1541, loss: 0.0005404376424849033 2023-01-23 19:38:35.307357: step: 2056/1541, loss: 0.0014202614547684789 2023-01-23 19:38:36.215061: step: 2058/1541, loss: 0.00020185996254440397 2023-01-23 19:38:37.099095: step: 2060/1541, loss: 0.00021988079242873937 2023-01-23 19:38:37.992846: step: 2062/1541, loss: 0.10787598788738251 2023-01-23 19:38:38.906902: step: 2064/1541, loss: 0.00016349999350495636 2023-01-23 19:38:39.804621: step: 2066/1541, loss: 0.006142713129520416 2023-01-23 19:38:40.664558: step: 2068/1541, loss: 0.00032011602888815105 2023-01-23 19:38:41.568989: step: 2070/1541, loss: 0.11711421608924866 2023-01-23 19:38:42.569968: step: 2072/1541, loss: 0.0014237440191209316 2023-01-23 19:38:43.579885: step: 2074/1541, loss: 0.0012076674029231071 2023-01-23 19:38:44.547990: step: 2076/1541, loss: 0.11657847464084625 2023-01-23 19:38:45.447041: step: 2078/1541, loss: 0.00964452512562275 2023-01-23 19:38:46.310241: step: 2080/1541, loss: 0.3605060577392578 2023-01-23 19:38:47.273081: step: 2082/1541, loss: 0.33070993423461914 2023-01-23 19:38:48.166497: step: 2084/1541, loss: 8.250639075413346e-05 2023-01-23 19:38:49.114776: step: 2086/1541, loss: 0.0056533291935920715 2023-01-23 19:38:50.046770: step: 2088/1541, loss: 0.0008502023410983384 2023-01-23 19:38:50.927735: step: 2090/1541, loss: 0.004465044941753149 2023-01-23 19:38:51.825944: step: 2092/1541, loss: 8.218983566621318e-06 2023-01-23 19:38:52.703337: step: 2094/1541, loss: 0.020003193989396095 2023-01-23 19:38:53.645629: step: 2096/1541, loss: 0.001097525586374104 2023-01-23 19:38:54.598605: step: 2098/1541, loss: 7.541956438217312e-05 2023-01-23 19:38:55.521064: step: 2100/1541, loss: 0.0020482465624809265 2023-01-23 19:38:56.372376: step: 2102/1541, loss: 0.00022222283587325364 2023-01-23 19:38:57.251603: step: 2104/1541, loss: 0.13211044669151306 2023-01-23 19:38:58.165532: step: 2106/1541, loss: 0.4270341992378235 2023-01-23 19:38:59.035986: step: 2108/1541, loss: 9.919396688928828e-06 2023-01-23 19:38:59.917442: step: 2110/1541, loss: 0.002155188238248229 2023-01-23 19:39:00.786652: step: 2112/1541, loss: 0.004869662690907717 2023-01-23 19:39:01.693275: step: 2114/1541, loss: 0.2857859432697296 2023-01-23 19:39:02.632592: step: 2116/1541, loss: 0.003004391910508275 2023-01-23 19:39:03.490859: step: 2118/1541, loss: 0.17673614621162415 2023-01-23 19:39:04.358719: step: 2120/1541, loss: 0.29816505312919617 2023-01-23 19:39:05.344848: step: 2122/1541, loss: 0.0019862945191562176 2023-01-23 19:39:06.271716: step: 2124/1541, loss: 0.0036630600225180387 2023-01-23 19:39:07.212512: step: 2126/1541, loss: 0.03111778385937214 2023-01-23 19:39:08.113304: step: 2128/1541, loss: 0.0001167066438938491 2023-01-23 19:39:08.992563: step: 2130/1541, loss: 0.0031302757561206818 2023-01-23 19:39:09.847086: step: 2132/1541, loss: 0.06118592619895935 2023-01-23 19:39:10.769328: step: 2134/1541, loss: 0.0184163860976696 2023-01-23 19:39:11.634242: step: 2136/1541, loss: 0.04668022319674492 2023-01-23 19:39:12.551353: step: 2138/1541, loss: 0.03782735764980316 2023-01-23 19:39:13.460639: step: 2140/1541, loss: 0.0035382569767534733 2023-01-23 19:39:14.342888: step: 2142/1541, loss: 0.00022866783547215164 2023-01-23 19:39:15.253783: step: 2144/1541, loss: 0.00020876229973509908 2023-01-23 19:39:16.129576: step: 2146/1541, loss: 0.00016508399858139455 2023-01-23 19:39:17.049901: step: 2148/1541, loss: 0.0023679002188146114 2023-01-23 19:39:17.960170: step: 2150/1541, loss: 0.32764196395874023 2023-01-23 19:39:18.911035: step: 2152/1541, loss: 0.0016124423127621412 2023-01-23 19:39:19.772679: step: 2154/1541, loss: 0.022007418796420097 2023-01-23 19:39:20.663100: step: 2156/1541, loss: 0.01877695508301258 2023-01-23 19:39:21.583466: step: 2158/1541, loss: 0.004261089488863945 2023-01-23 19:39:22.500856: step: 2160/1541, loss: 0.0003997108433395624 2023-01-23 19:39:23.379202: step: 2162/1541, loss: 0.00029951459146104753 2023-01-23 19:39:24.286857: step: 2164/1541, loss: 0.0031453100964426994 2023-01-23 19:39:25.231329: step: 2166/1541, loss: 0.0044930074363946915 2023-01-23 19:39:26.111494: step: 2168/1541, loss: 0.0012130002724006772 2023-01-23 19:39:26.992473: step: 2170/1541, loss: 9.255349141312763e-06 2023-01-23 19:39:27.872303: step: 2172/1541, loss: 0.13824136555194855 2023-01-23 19:39:28.764048: step: 2174/1541, loss: 0.33651024103164673 2023-01-23 19:39:29.704487: step: 2176/1541, loss: 0.023091671988368034 2023-01-23 19:39:30.563326: step: 2178/1541, loss: 0.10613702237606049 2023-01-23 19:39:31.470760: step: 2180/1541, loss: 0.009000753983855247 2023-01-23 19:39:32.492938: step: 2182/1541, loss: 0.00024963729083538055 2023-01-23 19:39:33.408847: step: 2184/1541, loss: 0.0001395609142491594 2023-01-23 19:39:34.297517: step: 2186/1541, loss: 0.00013862001651432365 2023-01-23 19:39:35.241849: step: 2188/1541, loss: 0.011904518119990826 2023-01-23 19:39:36.134419: step: 2190/1541, loss: 0.20026756823062897 2023-01-23 19:39:37.072210: step: 2192/1541, loss: 0.00010748968634288758 2023-01-23 19:39:37.957992: step: 2194/1541, loss: 0.005297766998410225 2023-01-23 19:39:38.837585: step: 2196/1541, loss: 0.008190744556486607 2023-01-23 19:39:39.746639: step: 2198/1541, loss: 0.0004317389684729278 2023-01-23 19:39:40.657353: step: 2200/1541, loss: 0.09670157730579376 2023-01-23 19:39:41.548454: step: 2202/1541, loss: 0.00016786405467428267 2023-01-23 19:39:42.485693: step: 2204/1541, loss: 0.15408247709274292 2023-01-23 19:39:43.353235: step: 2206/1541, loss: 0.33851537108421326 2023-01-23 19:39:44.281648: step: 2208/1541, loss: 0.0002806400298140943 2023-01-23 19:39:45.168160: step: 2210/1541, loss: 0.034003257751464844 2023-01-23 19:39:46.063772: step: 2212/1541, loss: 0.0012948599178344011 2023-01-23 19:39:46.960638: step: 2214/1541, loss: 0.09294541925191879 2023-01-23 19:39:47.879022: step: 2216/1541, loss: 0.040322721004486084 2023-01-23 19:39:48.798142: step: 2218/1541, loss: 0.25184983015060425 2023-01-23 19:39:49.704568: step: 2220/1541, loss: 0.12468982487916946 2023-01-23 19:39:50.558399: step: 2222/1541, loss: 0.005011339671909809 2023-01-23 19:39:51.437262: step: 2224/1541, loss: 0.011609156616032124 2023-01-23 19:39:52.366562: step: 2226/1541, loss: 0.06470268219709396 2023-01-23 19:39:53.272660: step: 2228/1541, loss: 0.0003037518763449043 2023-01-23 19:39:54.158171: step: 2230/1541, loss: 7.280511636054143e-05 2023-01-23 19:39:55.038114: step: 2232/1541, loss: 0.24678024649620056 2023-01-23 19:39:55.922852: step: 2234/1541, loss: 5.886155486223288e-05 2023-01-23 19:39:56.822669: step: 2236/1541, loss: 0.018174806609749794 2023-01-23 19:39:57.680014: step: 2238/1541, loss: 5.114823215990327e-05 2023-01-23 19:39:58.606746: step: 2240/1541, loss: 0.06842624396085739 2023-01-23 19:39:59.539141: step: 2242/1541, loss: 0.0008140943828038871 2023-01-23 19:40:00.380939: step: 2244/1541, loss: 0.5437707304954529 2023-01-23 19:40:01.305470: step: 2246/1541, loss: 0.0014144900487735868 2023-01-23 19:40:02.209374: step: 2248/1541, loss: 0.00010630816541379318 2023-01-23 19:40:03.181005: step: 2250/1541, loss: 0.0007053684093989432 2023-01-23 19:40:04.088617: step: 2252/1541, loss: 0.01747826300561428 2023-01-23 19:40:04.982227: step: 2254/1541, loss: 0.012140033766627312 2023-01-23 19:40:05.889211: step: 2256/1541, loss: 0.0002627855574246496 2023-01-23 19:40:06.747360: step: 2258/1541, loss: 0.004709724802523851 2023-01-23 19:40:07.585875: step: 2260/1541, loss: 0.01162385381758213 2023-01-23 19:40:08.518879: step: 2262/1541, loss: 0.07127156108617783 2023-01-23 19:40:09.411651: step: 2264/1541, loss: 0.00015655657625757158 2023-01-23 19:40:10.349089: step: 2266/1541, loss: 0.019845981150865555 2023-01-23 19:40:11.232069: step: 2268/1541, loss: 0.0008409044239670038 2023-01-23 19:40:12.115734: step: 2270/1541, loss: 0.08272430300712585 2023-01-23 19:40:12.968875: step: 2272/1541, loss: 0.004632021300494671 2023-01-23 19:40:13.859670: step: 2274/1541, loss: 0.018394578248262405 2023-01-23 19:40:14.737218: step: 2276/1541, loss: 0.0005677968147210777 2023-01-23 19:40:15.641075: step: 2278/1541, loss: 0.0007913943263702095 2023-01-23 19:40:16.591910: step: 2280/1541, loss: 0.00014675954298581928 2023-01-23 19:40:17.535378: step: 2282/1541, loss: 0.005112248007208109 2023-01-23 19:40:18.414961: step: 2284/1541, loss: 0.00020356607274152339 2023-01-23 19:40:19.328762: step: 2286/1541, loss: 0.0002638173755258322 2023-01-23 19:40:20.223832: step: 2288/1541, loss: 0.001153445104137063 2023-01-23 19:40:21.109256: step: 2290/1541, loss: 0.00015492957027163357 2023-01-23 19:40:21.970174: step: 2292/1541, loss: 0.0005677985027432442 2023-01-23 19:40:22.841722: step: 2294/1541, loss: 0.052564337849617004 2023-01-23 19:40:23.766365: step: 2296/1541, loss: 0.07225310057401657 2023-01-23 19:40:24.651693: step: 2298/1541, loss: 0.0666363313794136 2023-01-23 19:40:25.518892: step: 2300/1541, loss: 0.08272698521614075 2023-01-23 19:40:26.422445: step: 2302/1541, loss: 0.05363866314291954 2023-01-23 19:40:27.286645: step: 2304/1541, loss: 0.03337268903851509 2023-01-23 19:40:28.148568: step: 2306/1541, loss: 0.2412102222442627 2023-01-23 19:40:29.062458: step: 2308/1541, loss: 0.00197805417701602 2023-01-23 19:40:30.002749: step: 2310/1541, loss: 0.011402657255530357 2023-01-23 19:40:30.864230: step: 2312/1541, loss: 0.09091762453317642 2023-01-23 19:40:31.735317: step: 2314/1541, loss: 0.062076181173324585 2023-01-23 19:40:32.624543: step: 2316/1541, loss: 1.5331723261624575e-05 2023-01-23 19:40:33.551112: step: 2318/1541, loss: 0.005133432801812887 2023-01-23 19:40:34.432984: step: 2320/1541, loss: 0.0020563669968396425 2023-01-23 19:40:35.310507: step: 2322/1541, loss: 0.018283436074852943 2023-01-23 19:40:36.240963: step: 2324/1541, loss: 0.13808472454547882 2023-01-23 19:40:37.146114: step: 2326/1541, loss: 0.07779127359390259 2023-01-23 19:40:38.015861: step: 2328/1541, loss: 0.0001787648070603609 2023-01-23 19:40:38.935877: step: 2330/1541, loss: 0.24404515326023102 2023-01-23 19:40:39.845573: step: 2332/1541, loss: 0.00022437503503169864 2023-01-23 19:40:40.711360: step: 2334/1541, loss: 0.8532313108444214 2023-01-23 19:40:41.617062: step: 2336/1541, loss: 0.00044373335549607873 2023-01-23 19:40:42.494958: step: 2338/1541, loss: 0.011476288549602032 2023-01-23 19:40:43.379467: step: 2340/1541, loss: 0.1457480937242508 2023-01-23 19:40:44.342835: step: 2342/1541, loss: 0.00429916987195611 2023-01-23 19:40:45.235450: step: 2344/1541, loss: 0.036072731018066406 2023-01-23 19:40:46.181873: step: 2346/1541, loss: 0.0010557882487773895 2023-01-23 19:40:47.063743: step: 2348/1541, loss: 0.002253169659525156 2023-01-23 19:40:48.029375: step: 2350/1541, loss: 0.07879827916622162 2023-01-23 19:40:48.936714: step: 2352/1541, loss: 0.013873676769435406 2023-01-23 19:40:49.835757: step: 2354/1541, loss: 8.208146755350754e-05 2023-01-23 19:40:50.709653: step: 2356/1541, loss: 0.04304254800081253 2023-01-23 19:40:51.620362: step: 2358/1541, loss: 0.14765611290931702 2023-01-23 19:40:52.559184: step: 2360/1541, loss: 0.05967286229133606 2023-01-23 19:40:53.449234: step: 2362/1541, loss: 0.0020991116762161255 2023-01-23 19:40:54.354074: step: 2364/1541, loss: 0.025145266205072403 2023-01-23 19:40:55.218423: step: 2366/1541, loss: 0.0002282049390487373 2023-01-23 19:40:56.141737: step: 2368/1541, loss: 0.007407668046653271 2023-01-23 19:40:57.056594: step: 2370/1541, loss: 0.22555981576442719 2023-01-23 19:40:57.965030: step: 2372/1541, loss: 0.0001700358698144555 2023-01-23 19:40:58.872696: step: 2374/1541, loss: 0.10441219806671143 2023-01-23 19:40:59.803990: step: 2376/1541, loss: 0.01321872603148222 2023-01-23 19:41:00.681905: step: 2378/1541, loss: 0.0005080804112367332 2023-01-23 19:41:01.548247: step: 2380/1541, loss: 0.0015809491742402315 2023-01-23 19:41:02.413778: step: 2382/1541, loss: 0.0035230531357228756 2023-01-23 19:41:03.310312: step: 2384/1541, loss: 0.0012753787450492382 2023-01-23 19:41:04.180894: step: 2386/1541, loss: 0.004752176348119974 2023-01-23 19:41:05.077558: step: 2388/1541, loss: 0.00010916283645201474 2023-01-23 19:41:05.955670: step: 2390/1541, loss: 0.0042657083831727505 2023-01-23 19:41:06.911297: step: 2392/1541, loss: 0.048770464956760406 2023-01-23 19:41:07.809697: step: 2394/1541, loss: 0.07971592992544174 2023-01-23 19:41:08.712592: step: 2396/1541, loss: 0.0011237141443416476 2023-01-23 19:41:09.629004: step: 2398/1541, loss: 0.001765624270774424 2023-01-23 19:41:10.587254: step: 2400/1541, loss: 0.03472456336021423 2023-01-23 19:41:11.487338: step: 2402/1541, loss: 0.004030253738164902 2023-01-23 19:41:12.364542: step: 2404/1541, loss: 0.00918328296393156 2023-01-23 19:41:13.259843: step: 2406/1541, loss: 0.05906734988093376 2023-01-23 19:41:14.202317: step: 2408/1541, loss: 0.04798445478081703 2023-01-23 19:41:15.067883: step: 2410/1541, loss: 0.001857694354839623 2023-01-23 19:41:15.973339: step: 2412/1541, loss: 7.491960423067212e-05 2023-01-23 19:41:16.842822: step: 2414/1541, loss: 0.0001345458731520921 2023-01-23 19:41:17.765072: step: 2416/1541, loss: 0.29167309403419495 2023-01-23 19:41:18.673257: step: 2418/1541, loss: 0.002319943392649293 2023-01-23 19:41:19.576344: step: 2420/1541, loss: 0.0019341462757438421 2023-01-23 19:41:20.523928: step: 2422/1541, loss: 0.005009128712117672 2023-01-23 19:41:21.428808: step: 2424/1541, loss: 0.0004616461810655892 2023-01-23 19:41:22.351280: step: 2426/1541, loss: 0.0005289742257446051 2023-01-23 19:41:23.308569: step: 2428/1541, loss: 0.10643112659454346 2023-01-23 19:41:24.171054: step: 2430/1541, loss: 0.503416121006012 2023-01-23 19:41:25.059464: step: 2432/1541, loss: 0.18835902214050293 2023-01-23 19:41:25.934219: step: 2434/1541, loss: 0.0010598654625937343 2023-01-23 19:41:26.848771: step: 2436/1541, loss: 0.00720311189070344 2023-01-23 19:41:27.697435: step: 2438/1541, loss: 0.14445653557777405 2023-01-23 19:41:28.627137: step: 2440/1541, loss: 0.001310808933340013 2023-01-23 19:41:29.571805: step: 2442/1541, loss: 0.0007995758787728846 2023-01-23 19:41:30.464477: step: 2444/1541, loss: 0.0059379399754107 2023-01-23 19:41:31.297834: step: 2446/1541, loss: 5.635741035803221e-05 2023-01-23 19:41:32.267667: step: 2448/1541, loss: 0.07601162791252136 2023-01-23 19:41:33.154124: step: 2450/1541, loss: 0.00040591711876913905 2023-01-23 19:41:34.106416: step: 2452/1541, loss: 0.001965549774467945 2023-01-23 19:41:34.971023: step: 2454/1541, loss: 0.0011698317248374224 2023-01-23 19:41:35.912954: step: 2456/1541, loss: 0.013170727528631687 2023-01-23 19:41:36.803916: step: 2458/1541, loss: 0.0042629120871424675 2023-01-23 19:41:37.709487: step: 2460/1541, loss: 0.012625334784388542 2023-01-23 19:41:38.609082: step: 2462/1541, loss: 0.055733051151037216 2023-01-23 19:41:39.448311: step: 2464/1541, loss: 0.001955613726750016 2023-01-23 19:41:40.365103: step: 2466/1541, loss: 0.0008550088969059289 2023-01-23 19:41:41.218915: step: 2468/1541, loss: 0.001043491531163454 2023-01-23 19:41:42.098143: step: 2470/1541, loss: 0.007775064557790756 2023-01-23 19:41:43.035058: step: 2472/1541, loss: 0.012267797254025936 2023-01-23 19:41:43.904549: step: 2474/1541, loss: 0.0021347966976463795 2023-01-23 19:41:44.763543: step: 2476/1541, loss: 0.0025586821138858795 2023-01-23 19:41:45.658070: step: 2478/1541, loss: 0.005296459887176752 2023-01-23 19:41:46.570211: step: 2480/1541, loss: 0.0653989166021347 2023-01-23 19:41:47.462289: step: 2482/1541, loss: 0.0008613356039859354 2023-01-23 19:41:48.402128: step: 2484/1541, loss: 0.001163451699540019 2023-01-23 19:41:49.289352: step: 2486/1541, loss: 0.0005271266563795507 2023-01-23 19:41:50.189299: step: 2488/1541, loss: 0.0006562903290614486 2023-01-23 19:41:51.067474: step: 2490/1541, loss: 0.0010953645687550306 2023-01-23 19:41:52.040385: step: 2492/1541, loss: 0.021554429084062576 2023-01-23 19:41:52.891981: step: 2494/1541, loss: 0.0003044290642719716 2023-01-23 19:41:53.782265: step: 2496/1541, loss: 0.13001231849193573 2023-01-23 19:41:54.673490: step: 2498/1541, loss: 0.001712979283183813 2023-01-23 19:41:55.644199: step: 2500/1541, loss: 0.15939991176128387 2023-01-23 19:41:56.558719: step: 2502/1541, loss: 0.0037198523059487343 2023-01-23 19:41:57.474578: step: 2504/1541, loss: 0.09466369450092316 2023-01-23 19:41:58.379994: step: 2506/1541, loss: 4.6302400733111426e-05 2023-01-23 19:41:59.317746: step: 2508/1541, loss: 0.01388125866651535 2023-01-23 19:42:00.205731: step: 2510/1541, loss: 0.002714853733778 2023-01-23 19:42:01.068480: step: 2512/1541, loss: 0.001435613608919084 2023-01-23 19:42:01.947347: step: 2514/1541, loss: 0.006837575696408749 2023-01-23 19:42:02.826488: step: 2516/1541, loss: 0.05781414359807968 2023-01-23 19:42:03.778876: step: 2518/1541, loss: 0.005838498007506132 2023-01-23 19:42:04.664370: step: 2520/1541, loss: 0.00174560712184757 2023-01-23 19:42:05.546363: step: 2522/1541, loss: 0.0016242474084720016 2023-01-23 19:42:06.439008: step: 2524/1541, loss: 0.0008505556033924222 2023-01-23 19:42:07.332017: step: 2526/1541, loss: 0.0006140039185993373 2023-01-23 19:42:08.193076: step: 2528/1541, loss: 0.004405696410685778 2023-01-23 19:42:09.080441: step: 2530/1541, loss: 5.072205749456771e-05 2023-01-23 19:42:09.994382: step: 2532/1541, loss: 0.00039819974335841835 2023-01-23 19:42:10.874797: step: 2534/1541, loss: 0.009159969165921211 2023-01-23 19:42:11.770290: step: 2536/1541, loss: 0.1136830672621727 2023-01-23 19:42:12.677383: step: 2538/1541, loss: 0.0007220904808491468 2023-01-23 19:42:13.548923: step: 2540/1541, loss: 0.006101843435317278 2023-01-23 19:42:14.411200: step: 2542/1541, loss: 5.315196631272556e-06 2023-01-23 19:42:15.341044: step: 2544/1541, loss: 0.00011274960706941783 2023-01-23 19:42:16.251484: step: 2546/1541, loss: 0.007483682129532099 2023-01-23 19:42:17.140024: step: 2548/1541, loss: 0.20904114842414856 2023-01-23 19:42:18.064322: step: 2550/1541, loss: 0.24506662786006927 2023-01-23 19:42:18.995901: step: 2552/1541, loss: 0.0006770399049855769 2023-01-23 19:42:20.014888: step: 2554/1541, loss: 0.3177778124809265 2023-01-23 19:42:20.966860: step: 2556/1541, loss: 0.05403151363134384 2023-01-23 19:42:21.857816: step: 2558/1541, loss: 0.00023781695927027613 2023-01-23 19:42:22.744295: step: 2560/1541, loss: 5.152223093318753e-05 2023-01-23 19:42:23.615312: step: 2562/1541, loss: 0.07486424595117569 2023-01-23 19:42:24.522333: step: 2564/1541, loss: 0.007617342751473188 2023-01-23 19:42:25.407644: step: 2566/1541, loss: 0.048586033284664154 2023-01-23 19:42:26.247417: step: 2568/1541, loss: 0.016582908108830452 2023-01-23 19:42:27.149757: step: 2570/1541, loss: 0.04583177715539932 2023-01-23 19:42:28.114368: step: 2572/1541, loss: 0.07438325881958008 2023-01-23 19:42:29.009589: step: 2574/1541, loss: 0.000521042849868536 2023-01-23 19:42:29.934131: step: 2576/1541, loss: 0.01110475230962038 2023-01-23 19:42:30.893455: step: 2578/1541, loss: 0.03473125770688057 2023-01-23 19:42:31.805191: step: 2580/1541, loss: 0.24804535508155823 2023-01-23 19:42:32.670793: step: 2582/1541, loss: 0.0315818265080452 2023-01-23 19:42:33.595938: step: 2584/1541, loss: 0.0596892312169075 2023-01-23 19:42:34.480947: step: 2586/1541, loss: 0.4357808232307434 2023-01-23 19:42:35.378721: step: 2588/1541, loss: 0.0003520411264616996 2023-01-23 19:42:36.260036: step: 2590/1541, loss: 0.0004944922984577715 2023-01-23 19:42:37.165515: step: 2592/1541, loss: 0.046670615673065186 2023-01-23 19:42:38.067054: step: 2594/1541, loss: 0.05450175330042839 2023-01-23 19:42:38.951812: step: 2596/1541, loss: 0.012725697830319405 2023-01-23 19:42:39.846908: step: 2598/1541, loss: 0.00026230444200336933 2023-01-23 19:42:40.816110: step: 2600/1541, loss: 0.00010936992475762963 2023-01-23 19:42:41.738797: step: 2602/1541, loss: 0.00010534711327636614 2023-01-23 19:42:42.612101: step: 2604/1541, loss: 0.0003897424030583352 2023-01-23 19:42:43.513710: step: 2606/1541, loss: 0.0007188613526523113 2023-01-23 19:42:44.365647: step: 2608/1541, loss: 0.00785757414996624 2023-01-23 19:42:45.281721: step: 2610/1541, loss: 0.002459232695400715 2023-01-23 19:42:46.149106: step: 2612/1541, loss: 3.149299664073624e-05 2023-01-23 19:42:47.033547: step: 2614/1541, loss: 0.14378845691680908 2023-01-23 19:42:47.905061: step: 2616/1541, loss: 0.00817098654806614 2023-01-23 19:42:48.777342: step: 2618/1541, loss: 0.003003292717039585 2023-01-23 19:42:49.712984: step: 2620/1541, loss: 0.0020898866932839155 2023-01-23 19:42:50.613072: step: 2622/1541, loss: 0.4260185658931732 2023-01-23 19:42:51.546928: step: 2624/1541, loss: 0.00012589048128575087 2023-01-23 19:42:52.419151: step: 2626/1541, loss: 0.000696286209858954 2023-01-23 19:42:53.304597: step: 2628/1541, loss: 0.0801912322640419 2023-01-23 19:42:54.204113: step: 2630/1541, loss: 0.2478860765695572 2023-01-23 19:42:55.150214: step: 2632/1541, loss: 0.07998157292604446 2023-01-23 19:42:56.075992: step: 2634/1541, loss: 0.0004668831534218043 2023-01-23 19:42:56.973689: step: 2636/1541, loss: 0.00029604072915390134 2023-01-23 19:42:57.899049: step: 2638/1541, loss: 0.011526022106409073 2023-01-23 19:42:58.846218: step: 2640/1541, loss: 0.3895290195941925 2023-01-23 19:42:59.734535: step: 2642/1541, loss: 0.0006687741260975599 2023-01-23 19:43:00.604715: step: 2644/1541, loss: 0.0012158987810835242 2023-01-23 19:43:01.485946: step: 2646/1541, loss: 0.0004014430451206863 2023-01-23 19:43:02.358662: step: 2648/1541, loss: 0.0034220765810459852 2023-01-23 19:43:03.231440: step: 2650/1541, loss: 6.70611480018124e-05 2023-01-23 19:43:04.119201: step: 2652/1541, loss: 0.04353050887584686 2023-01-23 19:43:05.017039: step: 2654/1541, loss: 0.04817232862114906 2023-01-23 19:43:05.925404: step: 2656/1541, loss: 0.3131755590438843 2023-01-23 19:43:06.867756: step: 2658/1541, loss: 0.0003425095637794584 2023-01-23 19:43:07.768315: step: 2660/1541, loss: 0.0006641981890425086 2023-01-23 19:43:08.644687: step: 2662/1541, loss: 0.0053944154642522335 2023-01-23 19:43:09.524863: step: 2664/1541, loss: 0.04063589498400688 2023-01-23 19:43:10.446050: step: 2666/1541, loss: 0.015409178100526333 2023-01-23 19:43:11.364345: step: 2668/1541, loss: 0.14726762473583221 2023-01-23 19:43:12.234716: step: 2670/1541, loss: 0.001311384025029838 2023-01-23 19:43:13.107705: step: 2672/1541, loss: 0.001131783239543438 2023-01-23 19:43:14.082125: step: 2674/1541, loss: 0.017990972846746445 2023-01-23 19:43:14.943040: step: 2676/1541, loss: 0.011891940608620644 2023-01-23 19:43:15.806601: step: 2678/1541, loss: 0.33112525939941406 2023-01-23 19:43:16.751508: step: 2680/1541, loss: 0.2564079761505127 2023-01-23 19:43:17.620521: step: 2682/1541, loss: 0.0025901368353515863 2023-01-23 19:43:18.505305: step: 2684/1541, loss: 0.002340981736779213 2023-01-23 19:43:19.420296: step: 2686/1541, loss: 0.008267903700470924 2023-01-23 19:43:20.270870: step: 2688/1541, loss: 0.00044549370068125427 2023-01-23 19:43:21.124039: step: 2690/1541, loss: 0.0018982269102707505 2023-01-23 19:43:22.015367: step: 2692/1541, loss: 0.011140665970742702 2023-01-23 19:43:22.983239: step: 2694/1541, loss: 0.7443578243255615 2023-01-23 19:43:23.880062: step: 2696/1541, loss: 0.011854689568281174 2023-01-23 19:43:24.774401: step: 2698/1541, loss: 0.00463098892942071 2023-01-23 19:43:25.687919: step: 2700/1541, loss: 0.06521961092948914 2023-01-23 19:43:26.555048: step: 2702/1541, loss: 0.054517343640327454 2023-01-23 19:43:27.463300: step: 2704/1541, loss: 0.03071785159409046 2023-01-23 19:43:28.321075: step: 2706/1541, loss: 0.025041045621037483 2023-01-23 19:43:29.206167: step: 2708/1541, loss: 0.004140967503190041 2023-01-23 19:43:30.088687: step: 2710/1541, loss: 0.008142841048538685 2023-01-23 19:43:30.944747: step: 2712/1541, loss: 0.0005279884207993746 2023-01-23 19:43:31.870798: step: 2714/1541, loss: 0.02742721326649189 2023-01-23 19:43:32.714779: step: 2716/1541, loss: 0.02538585104048252 2023-01-23 19:43:33.608699: step: 2718/1541, loss: 0.00045150122605264187 2023-01-23 19:43:34.572916: step: 2720/1541, loss: 0.000317309342790395 2023-01-23 19:43:35.486460: step: 2722/1541, loss: 0.45306047797203064 2023-01-23 19:43:36.461614: step: 2724/1541, loss: 0.0005506046582013369 2023-01-23 19:43:37.411607: step: 2726/1541, loss: 0.042366549372673035 2023-01-23 19:43:38.325453: step: 2728/1541, loss: 0.033891916275024414 2023-01-23 19:43:39.230008: step: 2730/1541, loss: 0.00021805847063660622 2023-01-23 19:43:40.135893: step: 2732/1541, loss: 0.07529390603303909 2023-01-23 19:43:41.057807: step: 2734/1541, loss: 0.11112610995769501 2023-01-23 19:43:41.981336: step: 2736/1541, loss: 9.167387906927615e-05 2023-01-23 19:43:42.895875: step: 2738/1541, loss: 0.24224308133125305 2023-01-23 19:43:43.811174: step: 2740/1541, loss: 0.02507983334362507 2023-01-23 19:43:44.716977: step: 2742/1541, loss: 0.4828532934188843 2023-01-23 19:43:45.609528: step: 2744/1541, loss: 0.0001079109642887488 2023-01-23 19:43:46.474814: step: 2746/1541, loss: 0.0054639833979308605 2023-01-23 19:43:47.341925: step: 2748/1541, loss: 0.01668177917599678 2023-01-23 19:43:48.235939: step: 2750/1541, loss: 0.0001076329208444804 2023-01-23 19:43:49.163032: step: 2752/1541, loss: 0.8713012337684631 2023-01-23 19:43:50.078409: step: 2754/1541, loss: 0.0006091546965762973 2023-01-23 19:43:50.953088: step: 2756/1541, loss: 0.02158040925860405 2023-01-23 19:43:51.821662: step: 2758/1541, loss: 2.941223382949829 2023-01-23 19:43:52.719476: step: 2760/1541, loss: 0.17472617328166962 2023-01-23 19:43:53.601782: step: 2762/1541, loss: 0.001483248546719551 2023-01-23 19:43:54.505689: step: 2764/1541, loss: 0.0018707902636379004 2023-01-23 19:43:55.485236: step: 2766/1541, loss: 0.7955536842346191 2023-01-23 19:43:56.382612: step: 2768/1541, loss: 0.0021197039168328047 2023-01-23 19:43:57.381579: step: 2770/1541, loss: 7.393669511657208e-05 2023-01-23 19:43:58.270802: step: 2772/1541, loss: 0.054002806544303894 2023-01-23 19:43:59.139354: step: 2774/1541, loss: 0.010388128459453583 2023-01-23 19:44:00.008589: step: 2776/1541, loss: 0.00028417777502909303 2023-01-23 19:44:00.901558: step: 2778/1541, loss: 0.008015468716621399 2023-01-23 19:44:01.802492: step: 2780/1541, loss: 0.057362839579582214 2023-01-23 19:44:02.697276: step: 2782/1541, loss: 0.14611539244651794 2023-01-23 19:44:03.596842: step: 2784/1541, loss: 0.0013367041246965528 2023-01-23 19:44:04.507538: step: 2786/1541, loss: 0.0002377538476139307 2023-01-23 19:44:05.408127: step: 2788/1541, loss: 0.0015251020668074489 2023-01-23 19:44:06.332858: step: 2790/1541, loss: 0.02440246008336544 2023-01-23 19:44:07.202308: step: 2792/1541, loss: 0.010547097772359848 2023-01-23 19:44:08.109823: step: 2794/1541, loss: 0.011339616030454636 2023-01-23 19:44:09.051794: step: 2796/1541, loss: 0.05616648867726326 2023-01-23 19:44:09.949359: step: 2798/1541, loss: 0.042342375963926315 2023-01-23 19:44:10.787556: step: 2800/1541, loss: 0.08504730463027954 2023-01-23 19:44:11.689387: step: 2802/1541, loss: 0.0011369910789653659 2023-01-23 19:44:12.591194: step: 2804/1541, loss: 0.004599344916641712 2023-01-23 19:44:13.464146: step: 2806/1541, loss: 0.00038089422741904855 2023-01-23 19:44:14.358856: step: 2808/1541, loss: 0.20831547677516937 2023-01-23 19:44:15.270222: step: 2810/1541, loss: 0.001117982785217464 2023-01-23 19:44:16.220327: step: 2812/1541, loss: 0.2523496747016907 2023-01-23 19:44:17.082791: step: 2814/1541, loss: 0.09381263703107834 2023-01-23 19:44:18.044517: step: 2816/1541, loss: 0.008207304403185844 2023-01-23 19:44:18.935466: step: 2818/1541, loss: 0.0019370695808902383 2023-01-23 19:44:19.845319: step: 2820/1541, loss: 0.04577997699379921 2023-01-23 19:44:20.709390: step: 2822/1541, loss: 0.0007577361539006233 2023-01-23 19:44:21.587491: step: 2824/1541, loss: 0.026839720085263252 2023-01-23 19:44:22.483324: step: 2826/1541, loss: 0.005251922179013491 2023-01-23 19:44:23.347586: step: 2828/1541, loss: 0.0005571527872234583 2023-01-23 19:44:24.206402: step: 2830/1541, loss: 0.08206959813833237 2023-01-23 19:44:25.097664: step: 2832/1541, loss: 0.0012764723505824804 2023-01-23 19:44:25.983406: step: 2834/1541, loss: 0.0024554820265620947 2023-01-23 19:44:26.861637: step: 2836/1541, loss: 0.008319002576172352 2023-01-23 19:44:27.766720: step: 2838/1541, loss: 0.005777946673333645 2023-01-23 19:44:28.616594: step: 2840/1541, loss: 0.004466793034225702 2023-01-23 19:44:29.460705: step: 2842/1541, loss: 0.0015416651731356978 2023-01-23 19:44:30.321077: step: 2844/1541, loss: 0.0025720836129039526 2023-01-23 19:44:31.230178: step: 2846/1541, loss: 0.11562220752239227 2023-01-23 19:44:32.127486: step: 2848/1541, loss: 0.0013475625310093164 2023-01-23 19:44:33.025337: step: 2850/1541, loss: 0.008094675838947296 2023-01-23 19:44:33.945736: step: 2852/1541, loss: 0.21451586484909058 2023-01-23 19:44:34.805119: step: 2854/1541, loss: 0.0035397468600422144 2023-01-23 19:44:35.701014: step: 2856/1541, loss: 0.015546000562608242 2023-01-23 19:44:36.669595: step: 2858/1541, loss: 0.07487167418003082 2023-01-23 19:44:37.552260: step: 2860/1541, loss: 0.01138377096503973 2023-01-23 19:44:38.476674: step: 2862/1541, loss: 0.049893517047166824 2023-01-23 19:44:39.355545: step: 2864/1541, loss: 0.011711657047271729 2023-01-23 19:44:40.247079: step: 2866/1541, loss: 0.013509027659893036 2023-01-23 19:44:41.140557: step: 2868/1541, loss: 0.052541665732860565 2023-01-23 19:44:42.118190: step: 2870/1541, loss: 0.4196416139602661 2023-01-23 19:44:42.984612: step: 2872/1541, loss: 0.12796971201896667 2023-01-23 19:44:43.943268: step: 2874/1541, loss: 0.028307050466537476 2023-01-23 19:44:44.840418: step: 2876/1541, loss: 0.040725596249103546 2023-01-23 19:44:45.725940: step: 2878/1541, loss: 0.17579182982444763 2023-01-23 19:44:46.598381: step: 2880/1541, loss: 0.0027606417424976826 2023-01-23 19:44:47.475195: step: 2882/1541, loss: 0.05826295539736748 2023-01-23 19:44:48.420186: step: 2884/1541, loss: 0.010823510587215424 2023-01-23 19:44:49.282813: step: 2886/1541, loss: 0.04790689796209335 2023-01-23 19:44:50.179084: step: 2888/1541, loss: 0.41675928235054016 2023-01-23 19:44:51.044492: step: 2890/1541, loss: 0.015592093579471111 2023-01-23 19:44:52.040265: step: 2892/1541, loss: 0.07565279304981232 2023-01-23 19:44:52.967591: step: 2894/1541, loss: 0.0016365915071219206 2023-01-23 19:44:53.884941: step: 2896/1541, loss: 0.6036463379859924 2023-01-23 19:44:54.827124: step: 2898/1541, loss: 0.2993064522743225 2023-01-23 19:44:55.737812: step: 2900/1541, loss: 0.035516321659088135 2023-01-23 19:44:56.608519: step: 2902/1541, loss: 0.0476401224732399 2023-01-23 19:44:57.494160: step: 2904/1541, loss: 0.0552261546254158 2023-01-23 19:44:58.370764: step: 2906/1541, loss: 0.0022684703581035137 2023-01-23 19:44:59.325608: step: 2908/1541, loss: 0.0583154559135437 2023-01-23 19:45:00.255552: step: 2910/1541, loss: 0.0016995846526697278 2023-01-23 19:45:01.142903: step: 2912/1541, loss: 0.12188724428415298 2023-01-23 19:45:02.041205: step: 2914/1541, loss: 0.0014260284369811416 2023-01-23 19:45:02.914918: step: 2916/1541, loss: 0.0020960073452442884 2023-01-23 19:45:03.823333: step: 2918/1541, loss: 0.1674845665693283 2023-01-23 19:45:04.687362: step: 2920/1541, loss: 0.0028555335011333227 2023-01-23 19:45:05.583630: step: 2922/1541, loss: 0.0002111557696480304 2023-01-23 19:45:06.459291: step: 2924/1541, loss: 0.005839626770466566 2023-01-23 19:45:07.355248: step: 2926/1541, loss: 0.49564146995544434 2023-01-23 19:45:08.235110: step: 2928/1541, loss: 0.36967936158180237 2023-01-23 19:45:09.143867: step: 2930/1541, loss: 0.001861653057858348 2023-01-23 19:45:10.107952: step: 2932/1541, loss: 0.006165409926325083 2023-01-23 19:45:11.024572: step: 2934/1541, loss: 0.0002662753686308861 2023-01-23 19:45:11.966510: step: 2936/1541, loss: 0.7761531472206116 2023-01-23 19:45:12.845696: step: 2938/1541, loss: 0.0033309643622487783 2023-01-23 19:45:13.791160: step: 2940/1541, loss: 0.0037167032714933157 2023-01-23 19:45:14.648884: step: 2942/1541, loss: -1.7223550457856618e-05 2023-01-23 19:45:15.555427: step: 2944/1541, loss: 0.039249271154403687 2023-01-23 19:45:16.424795: step: 2946/1541, loss: 0.0014412227319553494 2023-01-23 19:45:17.332607: step: 2948/1541, loss: 0.005032448098063469 2023-01-23 19:45:18.230788: step: 2950/1541, loss: 0.00440558698028326 2023-01-23 19:45:19.151312: step: 2952/1541, loss: 0.0012964531779289246 2023-01-23 19:45:20.048017: step: 2954/1541, loss: 0.39032429456710815 2023-01-23 19:45:20.935292: step: 2956/1541, loss: 0.002946840599179268 2023-01-23 19:45:21.836114: step: 2958/1541, loss: 0.09172113984823227 2023-01-23 19:45:22.732314: step: 2960/1541, loss: 0.0007366271456703544 2023-01-23 19:45:23.663095: step: 2962/1541, loss: 0.009594087488949299 2023-01-23 19:45:24.590635: step: 2964/1541, loss: 0.0012110774405300617 2023-01-23 19:45:25.488481: step: 2966/1541, loss: 0.04883279651403427 2023-01-23 19:45:26.389960: step: 2968/1541, loss: 0.6977794766426086 2023-01-23 19:45:27.273938: step: 2970/1541, loss: 0.012319293804466724 2023-01-23 19:45:28.246641: step: 2972/1541, loss: 0.0002344996464671567 2023-01-23 19:45:29.152624: step: 2974/1541, loss: 0.00013335535186342895 2023-01-23 19:45:30.066820: step: 2976/1541, loss: 0.2851448357105255 2023-01-23 19:45:31.026308: step: 2978/1541, loss: 0.0023328422103077173 2023-01-23 19:45:31.938898: step: 2980/1541, loss: 7.874528091633692e-05 2023-01-23 19:45:32.836647: step: 2982/1541, loss: 0.05143680423498154 2023-01-23 19:45:33.730468: step: 2984/1541, loss: 0.00267293443903327 2023-01-23 19:45:34.627265: step: 2986/1541, loss: 0.15726229548454285 2023-01-23 19:45:35.502537: step: 2988/1541, loss: 1.767694354057312 2023-01-23 19:45:36.486734: step: 2990/1541, loss: 0.4608880281448364 2023-01-23 19:45:37.387421: step: 2992/1541, loss: 0.003000875236466527 2023-01-23 19:45:38.255767: step: 2994/1541, loss: 0.01153561007231474 2023-01-23 19:45:39.150494: step: 2996/1541, loss: 0.00038272293750196695 2023-01-23 19:45:40.013552: step: 2998/1541, loss: 0.0003819481935352087 2023-01-23 19:45:40.886371: step: 3000/1541, loss: 0.01623118296265602 2023-01-23 19:45:41.780692: step: 3002/1541, loss: 0.0444735623896122 2023-01-23 19:45:42.687890: step: 3004/1541, loss: 0.1422315388917923 2023-01-23 19:45:43.630326: step: 3006/1541, loss: 0.15095402300357819 2023-01-23 19:45:44.516021: step: 3008/1541, loss: 0.006730640772730112 2023-01-23 19:45:45.434157: step: 3010/1541, loss: 0.05790168419480324 2023-01-23 19:45:46.327330: step: 3012/1541, loss: 0.007545131258666515 2023-01-23 19:45:47.255617: step: 3014/1541, loss: 0.001534163486212492 2023-01-23 19:45:48.111733: step: 3016/1541, loss: 0.2829339802265167 2023-01-23 19:45:49.004184: step: 3018/1541, loss: 0.0017211362719535828 2023-01-23 19:45:49.856896: step: 3020/1541, loss: 0.027205081656575203 2023-01-23 19:45:50.744488: step: 3022/1541, loss: 0.279494971036911 2023-01-23 19:45:51.642921: step: 3024/1541, loss: 0.11389851570129395 2023-01-23 19:45:52.487922: step: 3026/1541, loss: 0.0008138457778841257 2023-01-23 19:45:53.434347: step: 3028/1541, loss: 0.0016145245172083378 2023-01-23 19:45:54.331448: step: 3030/1541, loss: 0.020896758884191513 2023-01-23 19:45:55.225498: step: 3032/1541, loss: 0.0978320762515068 2023-01-23 19:45:56.139926: step: 3034/1541, loss: 0.0008156134281307459 2023-01-23 19:45:57.073598: step: 3036/1541, loss: 0.01228778250515461 2023-01-23 19:45:57.956066: step: 3038/1541, loss: 0.001151782926172018 2023-01-23 19:45:58.847225: step: 3040/1541, loss: 0.015605120919644833 2023-01-23 19:45:59.724793: step: 3042/1541, loss: 0.0005267411470413208 2023-01-23 19:46:00.691531: step: 3044/1541, loss: 0.007615739479660988 2023-01-23 19:46:01.607065: step: 3046/1541, loss: 0.2350577712059021 2023-01-23 19:46:02.549368: step: 3048/1541, loss: 0.13422776758670807 2023-01-23 19:46:03.478368: step: 3050/1541, loss: 0.0013171187601983547 2023-01-23 19:46:04.336950: step: 3052/1541, loss: 0.00010082943481393158 2023-01-23 19:46:05.278889: step: 3054/1541, loss: 0.0008984804153442383 2023-01-23 19:46:06.150449: step: 3056/1541, loss: 0.007847219705581665 2023-01-23 19:46:07.060212: step: 3058/1541, loss: 0.05435424670577049 2023-01-23 19:46:07.949070: step: 3060/1541, loss: 0.0002461251860950142 2023-01-23 19:46:08.846500: step: 3062/1541, loss: 0.00034699676325544715 2023-01-23 19:46:09.800862: step: 3064/1541, loss: 0.001343929790891707 2023-01-23 19:46:10.699245: step: 3066/1541, loss: 0.0009739311644807458 2023-01-23 19:46:11.601442: step: 3068/1541, loss: 0.05488096922636032 2023-01-23 19:46:12.503701: step: 3070/1541, loss: 0.3539043962955475 2023-01-23 19:46:13.402150: step: 3072/1541, loss: 0.0004889121046289802 2023-01-23 19:46:14.318067: step: 3074/1541, loss: 5.6310673244297504e-05 2023-01-23 19:46:15.228633: step: 3076/1541, loss: 0.025127124041318893 2023-01-23 19:46:16.125178: step: 3078/1541, loss: 0.29117250442504883 2023-01-23 19:46:16.988286: step: 3080/1541, loss: 0.0003486349596641958 2023-01-23 19:46:17.925674: step: 3082/1541, loss: 0.0026847580447793007 2023-01-23 19:46:18.536957: step: 3084/1541, loss: 0.006110209506005049 ================================================== Loss: 0.064 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6792395718258888, 'r': 0.6544728845676911, 'f1': 0.6666262727836854}, 'combined': 0.440779336538331, 'epoch': 30} Test Chinese: {'event': {'p': 0.6072555205047319, 'r': 0.6920311563810665, 'f1': 0.6468776253150378}, 'argument': {'p': 0.7308811240268683, 'r': 0.695442848600119, 'f1': 0.7127217387297314}, 'combined': 0.46104374585989344, 'epoch': 30} Dev Korean: {'event': {'p': 0.5735294117647058, 'r': 0.7789613848202397, 'f1': 0.6606437041219649}, 'argument': {'p': 0.6632830910458017, 'r': 0.6632830910458017, 'f1': 0.6632830910458017}, 'combined': 0.43819379814996495, 'epoch': 30} Test Korean: {'event': {'p': 0.6105815061963775, 'r': 0.7675254643499101, 'f1': 0.6801168038226706}, 'argument': {'p': 0.7241965327770653, 'r': 0.7089395833505823, 'f1': 0.7164868464806451}, 'combined': 0.4872947440094008, 'epoch': 30} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.6721237710474757, 'r': 0.6620014250979654, 'f1': 0.6670241976707877}, 'combined': 0.46168947600536026, 'epoch': 30} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.722776026511748, 'r': 0.7025331072773077, 'f1': 0.7125108171038972}, 'combined': 0.49517511535041786, 'epoch': 30} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.662472179421332, 'r': 0.6917851077143113, 'f1': 0.6768114040841314}, 'combined': 0.5090547312769536, 'epoch': 30} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6009700176366843, 'r': 0.3512162440733869, 'f1': 0.4433385376008327}, 'combined': 0.25832324238158755, 'epoch': 30} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5812767094017094, 'r': 0.6488670244484197, 'f1': 0.613214990138067}, 'combined': 0.297847280924204, 'epoch': 30} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 31 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 19:51:40.333565: step: 2/1541, loss: 0.001238915603607893 2023-01-23 19:51:41.251122: step: 4/1541, loss: 0.07711662352085114 2023-01-23 19:51:42.138540: step: 6/1541, loss: 0.010057744570076466 2023-01-23 19:51:43.080324: step: 8/1541, loss: 0.07948507368564606 2023-01-23 19:51:43.982300: step: 10/1541, loss: 0.6112045049667358 2023-01-23 19:51:44.911707: step: 12/1541, loss: -6.35177229924011e-07 2023-01-23 19:51:45.806885: step: 14/1541, loss: 0.0036503183655440807 2023-01-23 19:51:46.725755: step: 16/1541, loss: 0.00038125060382299125 2023-01-23 19:51:47.637035: step: 18/1541, loss: 0.036002203822135925 2023-01-23 19:51:48.573736: step: 20/1541, loss: 0.0213986374437809 2023-01-23 19:51:49.441901: step: 22/1541, loss: 0.001246610190719366 2023-01-23 19:51:50.299086: step: 24/1541, loss: 0.08159514516592026 2023-01-23 19:51:51.199518: step: 26/1541, loss: 0.019952313974499702 2023-01-23 19:51:52.168596: step: 28/1541, loss: 0.03987856209278107 2023-01-23 19:51:53.064199: step: 30/1541, loss: 0.0015379042597487569 2023-01-23 19:51:53.964600: step: 32/1541, loss: 0.0011799747589975595 2023-01-23 19:51:54.854302: step: 34/1541, loss: 0.006951871328055859 2023-01-23 19:51:55.751144: step: 36/1541, loss: 0.00016875665460247546 2023-01-23 19:51:56.614256: step: 38/1541, loss: 0.003253714647144079 2023-01-23 19:51:57.494311: step: 40/1541, loss: 0.0034870041999965906 2023-01-23 19:51:58.391803: step: 42/1541, loss: 0.0005891985492780805 2023-01-23 19:51:59.300973: step: 44/1541, loss: 0.3339942693710327 2023-01-23 19:52:00.196615: step: 46/1541, loss: 0.016712479293346405 2023-01-23 19:52:01.152633: step: 48/1541, loss: 0.25771987438201904 2023-01-23 19:52:02.048444: step: 50/1541, loss: 0.03057350218296051 2023-01-23 19:52:02.980139: step: 52/1541, loss: 0.00020278454758226871 2023-01-23 19:52:03.856615: step: 54/1541, loss: 0.16400004923343658 2023-01-23 19:52:04.725563: step: 56/1541, loss: 0.0012649681884795427 2023-01-23 19:52:05.612207: step: 58/1541, loss: 0.005412232130765915 2023-01-23 19:52:06.538032: step: 60/1541, loss: 0.0007524950196966529 2023-01-23 19:52:07.403332: step: 62/1541, loss: 0.000528972246684134 2023-01-23 19:52:08.275104: step: 64/1541, loss: 0.009732184000313282 2023-01-23 19:52:09.174020: step: 66/1541, loss: 0.0010952389566227794 2023-01-23 19:52:10.084781: step: 68/1541, loss: 0.00823968555778265 2023-01-23 19:52:11.007749: step: 70/1541, loss: 0.14407825469970703 2023-01-23 19:52:11.911191: step: 72/1541, loss: 0.47242218255996704 2023-01-23 19:52:12.886819: step: 74/1541, loss: 0.0038741326425224543 2023-01-23 19:52:13.909586: step: 76/1541, loss: 0.0032383205834776163 2023-01-23 19:52:14.791349: step: 78/1541, loss: 0.00011949706095037982 2023-01-23 19:52:15.711447: step: 80/1541, loss: 0.00019955378957092762 2023-01-23 19:52:16.648696: step: 82/1541, loss: 0.0891556516289711 2023-01-23 19:52:17.586007: step: 84/1541, loss: 0.05717528238892555 2023-01-23 19:52:18.505814: step: 86/1541, loss: 1.5260564088821411 2023-01-23 19:52:19.417828: step: 88/1541, loss: 0.0027293767780065536 2023-01-23 19:52:20.297330: step: 90/1541, loss: 0.0002406170970061794 2023-01-23 19:52:21.169251: step: 92/1541, loss: 0.0011888225562870502 2023-01-23 19:52:22.087333: step: 94/1541, loss: 0.5375372767448425 2023-01-23 19:52:22.991401: step: 96/1541, loss: 0.0009513114346191287 2023-01-23 19:52:23.959505: step: 98/1541, loss: 0.00020033025066368282 2023-01-23 19:52:24.850694: step: 100/1541, loss: 0.00022056353918742388 2023-01-23 19:52:25.746553: step: 102/1541, loss: 4.443118814378977e-05 2023-01-23 19:52:26.718301: step: 104/1541, loss: 0.0616399385035038 2023-01-23 19:52:27.621805: step: 106/1541, loss: 0.0539407879114151 2023-01-23 19:52:28.490508: step: 108/1541, loss: 0.0012600335758179426 2023-01-23 19:52:29.417617: step: 110/1541, loss: 0.3728964328765869 2023-01-23 19:52:30.301299: step: 112/1541, loss: 0.0009884568862617016 2023-01-23 19:52:31.240470: step: 114/1541, loss: 2.4575693607330322 2023-01-23 19:52:32.154674: step: 116/1541, loss: 0.00023814589076209813 2023-01-23 19:52:33.053796: step: 118/1541, loss: 0.002446345519274473 2023-01-23 19:52:33.929344: step: 120/1541, loss: 0.00036719226045534015 2023-01-23 19:52:34.831958: step: 122/1541, loss: 0.00015659596829209477 2023-01-23 19:52:35.750311: step: 124/1541, loss: 0.00014348218974191695 2023-01-23 19:52:36.665679: step: 126/1541, loss: 0.000854981190059334 2023-01-23 19:52:37.575737: step: 128/1541, loss: 0.006963150575757027 2023-01-23 19:52:38.430564: step: 130/1541, loss: 0.0007458667969331145 2023-01-23 19:52:39.420447: step: 132/1541, loss: 0.001968397293239832 2023-01-23 19:52:40.349354: step: 134/1541, loss: 0.00013406190555542707 2023-01-23 19:52:41.231846: step: 136/1541, loss: 0.006107444874942303 2023-01-23 19:52:42.151466: step: 138/1541, loss: 0.02124881185591221 2023-01-23 19:52:43.024584: step: 140/1541, loss: 0.031118357554078102 2023-01-23 19:52:43.919017: step: 142/1541, loss: 0.041394248604774475 2023-01-23 19:52:44.869193: step: 144/1541, loss: 0.044672891497612 2023-01-23 19:52:45.783511: step: 146/1541, loss: 0.001114162616431713 2023-01-23 19:52:46.693382: step: 148/1541, loss: 0.00022214165073819458 2023-01-23 19:52:47.578247: step: 150/1541, loss: 0.0009594914154149592 2023-01-23 19:52:48.478299: step: 152/1541, loss: 0.028531372547149658 2023-01-23 19:52:49.357582: step: 154/1541, loss: 0.0011378745548427105 2023-01-23 19:52:50.262705: step: 156/1541, loss: -2.5316336177638732e-05 2023-01-23 19:52:51.142366: step: 158/1541, loss: 6.110786489443853e-05 2023-01-23 19:52:52.100310: step: 160/1541, loss: 0.03571510687470436 2023-01-23 19:52:52.973010: step: 162/1541, loss: 9.286045678891242e-05 2023-01-23 19:52:53.871448: step: 164/1541, loss: 0.2798074185848236 2023-01-23 19:52:54.811051: step: 166/1541, loss: 0.027576886117458344 2023-01-23 19:52:55.733125: step: 168/1541, loss: 0.0015105269849300385 2023-01-23 19:52:56.610958: step: 170/1541, loss: 0.00019411738321650773 2023-01-23 19:52:57.516182: step: 172/1541, loss: 8.874892955645919e-05 2023-01-23 19:52:58.387020: step: 174/1541, loss: 0.11539123207330704 2023-01-23 19:52:59.270409: step: 176/1541, loss: 0.025276601314544678 2023-01-23 19:53:00.194285: step: 178/1541, loss: 0.00028734843363054097 2023-01-23 19:53:01.150450: step: 180/1541, loss: 0.10702866315841675 2023-01-23 19:53:02.044249: step: 182/1541, loss: 0.0008522017160430551 2023-01-23 19:53:02.937903: step: 184/1541, loss: 0.010971171781420708 2023-01-23 19:53:03.884523: step: 186/1541, loss: 0.004337850026786327 2023-01-23 19:53:04.770571: step: 188/1541, loss: 0.0817614421248436 2023-01-23 19:53:05.755828: step: 190/1541, loss: 0.090997613966465 2023-01-23 19:53:06.653510: step: 192/1541, loss: 0.0005610503721982241 2023-01-23 19:53:07.558261: step: 194/1541, loss: 0.0004867705865763128 2023-01-23 19:53:08.438817: step: 196/1541, loss: 0.06251796334981918 2023-01-23 19:53:09.295797: step: 198/1541, loss: 0.005370832979679108 2023-01-23 19:53:10.190081: step: 200/1541, loss: 9.098628652282059e-05 2023-01-23 19:53:11.092462: step: 202/1541, loss: 0.003268780652433634 2023-01-23 19:53:11.989226: step: 204/1541, loss: 0.018536243587732315 2023-01-23 19:53:12.845269: step: 206/1541, loss: 0.025813771411776543 2023-01-23 19:53:13.687791: step: 208/1541, loss: 0.006883857771754265 2023-01-23 19:53:14.588237: step: 210/1541, loss: 0.008690916933119297 2023-01-23 19:53:15.452659: step: 212/1541, loss: 0.0032718705479055643 2023-01-23 19:53:16.353868: step: 214/1541, loss: 0.00043977657333016396 2023-01-23 19:53:17.251134: step: 216/1541, loss: 2.7514637622516602e-05 2023-01-23 19:53:18.169121: step: 218/1541, loss: 0.0004955792683176696 2023-01-23 19:53:19.152286: step: 220/1541, loss: 0.0006814050721004605 2023-01-23 19:53:20.041968: step: 222/1541, loss: 0.001078763627447188 2023-01-23 19:53:20.954916: step: 224/1541, loss: 0.06736496835947037 2023-01-23 19:53:21.809352: step: 226/1541, loss: 0.00022426752548199147 2023-01-23 19:53:22.690090: step: 228/1541, loss: 0.001511680195108056 2023-01-23 19:53:23.609603: step: 230/1541, loss: 0.001466374029405415 2023-01-23 19:53:24.479969: step: 232/1541, loss: 0.035208817571401596 2023-01-23 19:53:25.361494: step: 234/1541, loss: 0.0038602855056524277 2023-01-23 19:53:26.234444: step: 236/1541, loss: 3.4480003705539275e-06 2023-01-23 19:53:27.076619: step: 238/1541, loss: 0.0035599153488874435 2023-01-23 19:53:28.016084: step: 240/1541, loss: 6.989179382799193e-05 2023-01-23 19:53:28.906516: step: 242/1541, loss: 0.0013846068177372217 2023-01-23 19:53:29.801158: step: 244/1541, loss: 0.0004039801424369216 2023-01-23 19:53:30.664444: step: 246/1541, loss: 0.1277277022600174 2023-01-23 19:53:31.508898: step: 248/1541, loss: 0.0005727537209168077 2023-01-23 19:53:32.383909: step: 250/1541, loss: 0.0009404401644133031 2023-01-23 19:53:33.250068: step: 252/1541, loss: 0.017422480508685112 2023-01-23 19:53:34.150882: step: 254/1541, loss: 0.05706951022148132 2023-01-23 19:53:35.026916: step: 256/1541, loss: 0.0014799266355112195 2023-01-23 19:53:35.973854: step: 258/1541, loss: 8.124102350848261e-06 2023-01-23 19:53:36.879939: step: 260/1541, loss: 0.07580848783254623 2023-01-23 19:53:37.759483: step: 262/1541, loss: 0.02055254951119423 2023-01-23 19:53:38.643638: step: 264/1541, loss: 0.04874804988503456 2023-01-23 19:53:39.518250: step: 266/1541, loss: 0.00010785990889417008 2023-01-23 19:53:40.396305: step: 268/1541, loss: 0.010011068545281887 2023-01-23 19:53:41.314919: step: 270/1541, loss: 8.701995830051601e-05 2023-01-23 19:53:42.213887: step: 272/1541, loss: 0.00012389947369229048 2023-01-23 19:53:43.084436: step: 274/1541, loss: 0.028277350589632988 2023-01-23 19:53:43.949640: step: 276/1541, loss: 0.12954853475093842 2023-01-23 19:53:44.835250: step: 278/1541, loss: 0.0003461726300884038 2023-01-23 19:53:45.733978: step: 280/1541, loss: 3.8059530197642744e-05 2023-01-23 19:53:46.606014: step: 282/1541, loss: 0.011838440783321857 2023-01-23 19:53:47.471887: step: 284/1541, loss: 0.000436901580542326 2023-01-23 19:53:48.348625: step: 286/1541, loss: 0.22128920257091522 2023-01-23 19:53:49.223050: step: 288/1541, loss: 3.682942406157963e-05 2023-01-23 19:53:50.182683: step: 290/1541, loss: 0.021856019273400307 2023-01-23 19:53:51.005714: step: 292/1541, loss: 0.011931746266782284 2023-01-23 19:53:51.942132: step: 294/1541, loss: 4.612707138061523 2023-01-23 19:53:52.864163: step: 296/1541, loss: 0.019828934222459793 2023-01-23 19:53:53.744193: step: 298/1541, loss: 0.2514141798019409 2023-01-23 19:53:54.649776: step: 300/1541, loss: 0.030234752222895622 2023-01-23 19:53:55.569488: step: 302/1541, loss: 0.00018029460625257343 2023-01-23 19:53:56.410859: step: 304/1541, loss: 0.3462374806404114 2023-01-23 19:53:57.335992: step: 306/1541, loss: 0.05259915068745613 2023-01-23 19:53:58.255412: step: 308/1541, loss: 0.00025120051577687263 2023-01-23 19:53:59.164505: step: 310/1541, loss: 0.0009345478611066937 2023-01-23 19:54:00.074653: step: 312/1541, loss: 0.031226178631186485 2023-01-23 19:54:00.907906: step: 314/1541, loss: 0.028341546654701233 2023-01-23 19:54:01.873706: step: 316/1541, loss: 4.022546636406332e-05 2023-01-23 19:54:02.801681: step: 318/1541, loss: 0.0012591335689648986 2023-01-23 19:54:03.741039: step: 320/1541, loss: 0.16674168407917023 2023-01-23 19:54:04.637997: step: 322/1541, loss: 0.4328012764453888 2023-01-23 19:54:05.565373: step: 324/1541, loss: 0.005613358225673437 2023-01-23 19:54:06.462764: step: 326/1541, loss: 0.015110270120203495 2023-01-23 19:54:07.352266: step: 328/1541, loss: 4.141724275541492e-05 2023-01-23 19:54:08.243161: step: 330/1541, loss: 0.000795786501839757 2023-01-23 19:54:09.103583: step: 332/1541, loss: 0.02239866368472576 2023-01-23 19:54:09.942567: step: 334/1541, loss: 0.010596094653010368 2023-01-23 19:54:10.883082: step: 336/1541, loss: 0.0002161335141863674 2023-01-23 19:54:11.792278: step: 338/1541, loss: 0.009418027475476265 2023-01-23 19:54:12.753539: step: 340/1541, loss: 0.00018780752725433558 2023-01-23 19:54:13.631471: step: 342/1541, loss: 0.006820732727646828 2023-01-23 19:54:14.524047: step: 344/1541, loss: 0.0006551193655468524 2023-01-23 19:54:15.392225: step: 346/1541, loss: 0.00015070235531311482 2023-01-23 19:54:16.325130: step: 348/1541, loss: 0.1016061007976532 2023-01-23 19:54:17.188622: step: 350/1541, loss: 0.07705061137676239 2023-01-23 19:54:18.070337: step: 352/1541, loss: 0.0015636268071830273 2023-01-23 19:54:18.942858: step: 354/1541, loss: 7.885004015406594e-05 2023-01-23 19:54:19.792827: step: 356/1541, loss: 0.007775785867124796 2023-01-23 19:54:20.723207: step: 358/1541, loss: 0.0008302276837639511 2023-01-23 19:54:21.620357: step: 360/1541, loss: 0.00019993557361885905 2023-01-23 19:54:22.537969: step: 362/1541, loss: 0.00011214578262297437 2023-01-23 19:54:23.445144: step: 364/1541, loss: 0.00294479844160378 2023-01-23 19:54:24.365029: step: 366/1541, loss: 1.672909638728015e-05 2023-01-23 19:54:25.267203: step: 368/1541, loss: 0.0001487556437496096 2023-01-23 19:54:26.146283: step: 370/1541, loss: 0.004007507581263781 2023-01-23 19:54:27.026267: step: 372/1541, loss: 0.05455286428332329 2023-01-23 19:54:27.886789: step: 374/1541, loss: 0.011526277288794518 2023-01-23 19:54:28.738774: step: 376/1541, loss: 0.0011187947820872068 2023-01-23 19:54:29.627077: step: 378/1541, loss: 0.029469015076756477 2023-01-23 19:54:30.541440: step: 380/1541, loss: 0.0005441135144792497 2023-01-23 19:54:31.409340: step: 382/1541, loss: 0.14045245945453644 2023-01-23 19:54:32.299840: step: 384/1541, loss: 0.0007549230940639973 2023-01-23 19:54:33.141322: step: 386/1541, loss: 0.0006776501541025937 2023-01-23 19:54:34.075606: step: 388/1541, loss: 0.0036756901536136866 2023-01-23 19:54:34.977176: step: 390/1541, loss: 0.0027692841831594706 2023-01-23 19:54:35.863695: step: 392/1541, loss: 0.0005199926672503352 2023-01-23 19:54:36.750044: step: 394/1541, loss: 5.651983155985363e-05 2023-01-23 19:54:37.619052: step: 396/1541, loss: 0.6598929166793823 2023-01-23 19:54:38.511743: step: 398/1541, loss: 0.0009634990128688514 2023-01-23 19:54:39.343287: step: 400/1541, loss: 0.00019529805285856128 2023-01-23 19:54:40.245930: step: 402/1541, loss: 6.74001348670572e-05 2023-01-23 19:54:41.143798: step: 404/1541, loss: 0.00034711998887360096 2023-01-23 19:54:42.038533: step: 406/1541, loss: 0.003066041972488165 2023-01-23 19:54:42.960620: step: 408/1541, loss: 0.004024328198283911 2023-01-23 19:54:43.872229: step: 410/1541, loss: 0.0003762244596146047 2023-01-23 19:54:44.727335: step: 412/1541, loss: 1.2816475646104664e-05 2023-01-23 19:54:45.647634: step: 414/1541, loss: 0.004502903204411268 2023-01-23 19:54:46.523154: step: 416/1541, loss: 0.007429934106767178 2023-01-23 19:54:47.418548: step: 418/1541, loss: 0.03249379247426987 2023-01-23 19:54:48.295248: step: 420/1541, loss: 1.153846642409917e-05 2023-01-23 19:54:49.161924: step: 422/1541, loss: 0.2640909254550934 2023-01-23 19:54:50.056051: step: 424/1541, loss: -3.249606743338518e-05 2023-01-23 19:54:50.973900: step: 426/1541, loss: 0.00026020046789199114 2023-01-23 19:54:51.846623: step: 428/1541, loss: 0.00010499614290893078 2023-01-23 19:54:52.752579: step: 430/1541, loss: 0.1392693817615509 2023-01-23 19:54:53.716206: step: 432/1541, loss: 0.2489045113325119 2023-01-23 19:54:54.597012: step: 434/1541, loss: 0.15300244092941284 2023-01-23 19:54:55.474045: step: 436/1541, loss: 7.778097642585635e-05 2023-01-23 19:54:56.423979: step: 438/1541, loss: 0.02210254780948162 2023-01-23 19:54:57.288171: step: 440/1541, loss: 0.0886964201927185 2023-01-23 19:54:58.287956: step: 442/1541, loss: 0.0011417821515351534 2023-01-23 19:54:59.207501: step: 444/1541, loss: 0.000719908275641501 2023-01-23 19:55:00.040912: step: 446/1541, loss: 0.12475493550300598 2023-01-23 19:55:00.942662: step: 448/1541, loss: 0.0010481377830728889 2023-01-23 19:55:01.878368: step: 450/1541, loss: 0.7663816213607788 2023-01-23 19:55:02.819211: step: 452/1541, loss: 0.0014263158664107323 2023-01-23 19:55:03.822015: step: 454/1541, loss: 0.0011799359926953912 2023-01-23 19:55:04.696537: step: 456/1541, loss: 0.004470211919397116 2023-01-23 19:55:05.581689: step: 458/1541, loss: 0.09505914896726608 2023-01-23 19:55:06.481514: step: 460/1541, loss: 0.5163630247116089 2023-01-23 19:55:07.357723: step: 462/1541, loss: 0.011087723076343536 2023-01-23 19:55:08.205916: step: 464/1541, loss: 0.0060644024051725864 2023-01-23 19:55:09.056473: step: 466/1541, loss: 0.0027626892551779747 2023-01-23 19:55:10.039227: step: 468/1541, loss: 0.03296844661235809 2023-01-23 19:55:10.886179: step: 470/1541, loss: 0.11112022399902344 2023-01-23 19:55:11.763835: step: 472/1541, loss: 0.001777965808287263 2023-01-23 19:55:12.699808: step: 474/1541, loss: 0.002185878111049533 2023-01-23 19:55:13.622046: step: 476/1541, loss: 0.03794363513588905 2023-01-23 19:55:14.501107: step: 478/1541, loss: 0.0013957973569631577 2023-01-23 19:55:15.537842: step: 480/1541, loss: 0.0021980488672852516 2023-01-23 19:55:16.409795: step: 482/1541, loss: 0.006536322180181742 2023-01-23 19:55:17.324678: step: 484/1541, loss: 0.0002477189409546554 2023-01-23 19:55:18.221482: step: 486/1541, loss: 0.08334305882453918 2023-01-23 19:55:19.151825: step: 488/1541, loss: 5.0488441047491506e-05 2023-01-23 19:55:19.999823: step: 490/1541, loss: 0.0018750526942312717 2023-01-23 19:55:20.916301: step: 492/1541, loss: 0.05496201664209366 2023-01-23 19:55:21.881391: step: 494/1541, loss: 0.13721628487110138 2023-01-23 19:55:22.761482: step: 496/1541, loss: 0.000343867955962196 2023-01-23 19:55:23.648791: step: 498/1541, loss: 0.0800001472234726 2023-01-23 19:55:24.577650: step: 500/1541, loss: 0.15405194461345673 2023-01-23 19:55:25.514574: step: 502/1541, loss: 0.001378071028739214 2023-01-23 19:55:26.410217: step: 504/1541, loss: 0.15831618010997772 2023-01-23 19:55:27.280763: step: 506/1541, loss: 4.147634535911493e-05 2023-01-23 19:55:28.158117: step: 508/1541, loss: 0.06702002882957458 2023-01-23 19:55:29.055787: step: 510/1541, loss: 0.020859146490693092 2023-01-23 19:55:29.981572: step: 512/1541, loss: 0.0013950291322544217 2023-01-23 19:55:30.886680: step: 514/1541, loss: 0.07887575775384903 2023-01-23 19:55:31.861236: step: 516/1541, loss: 0.07520148903131485 2023-01-23 19:55:32.762218: step: 518/1541, loss: 0.2639191448688507 2023-01-23 19:55:33.640997: step: 520/1541, loss: 0.0028059734031558037 2023-01-23 19:55:34.531014: step: 522/1541, loss: 0.10547379404306412 2023-01-23 19:55:35.428001: step: 524/1541, loss: 0.003717361483722925 2023-01-23 19:55:36.307225: step: 526/1541, loss: 0.04792820289731026 2023-01-23 19:55:37.222034: step: 528/1541, loss: 0.014153610914945602 2023-01-23 19:55:38.091911: step: 530/1541, loss: 0.03801492974162102 2023-01-23 19:55:39.016193: step: 532/1541, loss: 0.1419743150472641 2023-01-23 19:55:39.907142: step: 534/1541, loss: -1.4208153515937738e-05 2023-01-23 19:55:40.775844: step: 536/1541, loss: 0.018551062792539597 2023-01-23 19:55:41.700209: step: 538/1541, loss: 0.0002646847569849342 2023-01-23 19:55:42.635526: step: 540/1541, loss: 0.010287226177752018 2023-01-23 19:55:43.548223: step: 542/1541, loss: 2.5876583094941452e-05 2023-01-23 19:55:44.446156: step: 544/1541, loss: -1.6975784092210233e-05 2023-01-23 19:55:45.375743: step: 546/1541, loss: 0.0035606157034635544 2023-01-23 19:55:46.318710: step: 548/1541, loss: 0.01878461241722107 2023-01-23 19:55:47.196624: step: 550/1541, loss: 0.06732051819562912 2023-01-23 19:55:48.038616: step: 552/1541, loss: 0.0012640099739655852 2023-01-23 19:55:48.947746: step: 554/1541, loss: 0.3764043152332306 2023-01-23 19:55:49.866886: step: 556/1541, loss: 0.08008529245853424 2023-01-23 19:55:50.767237: step: 558/1541, loss: 0.011593005619943142 2023-01-23 19:55:51.642089: step: 560/1541, loss: 0.1657639890909195 2023-01-23 19:55:52.503184: step: 562/1541, loss: 0.0020825557876378298 2023-01-23 19:55:53.370976: step: 564/1541, loss: 0.0005524629377759993 2023-01-23 19:55:54.251347: step: 566/1541, loss: 0.0009179721237160265 2023-01-23 19:55:55.200672: step: 568/1541, loss: 0.0007994403713382781 2023-01-23 19:55:56.084162: step: 570/1541, loss: 0.011699081398546696 2023-01-23 19:55:56.973327: step: 572/1541, loss: 0.00873550958931446 2023-01-23 19:55:57.877876: step: 574/1541, loss: 0.003497425001114607 2023-01-23 19:55:58.767545: step: 576/1541, loss: 0.0016702680150046945 2023-01-23 19:55:59.631836: step: 578/1541, loss: 0.0001772963732946664 2023-01-23 19:56:00.532140: step: 580/1541, loss: 0.005036309827119112 2023-01-23 19:56:01.429823: step: 582/1541, loss: 0.17521153390407562 2023-01-23 19:56:02.332305: step: 584/1541, loss: 0.0008299991022795439 2023-01-23 19:56:03.209812: step: 586/1541, loss: 0.0002739136398304254 2023-01-23 19:56:04.192395: step: 588/1541, loss: 0.06057954579591751 2023-01-23 19:56:05.111819: step: 590/1541, loss: 0.0008052678313106298 2023-01-23 19:56:05.977957: step: 592/1541, loss: 0.004132740665227175 2023-01-23 19:56:06.901765: step: 594/1541, loss: 0.0008254069834947586 2023-01-23 19:56:07.803668: step: 596/1541, loss: 6.545330688823014e-05 2023-01-23 19:56:08.683008: step: 598/1541, loss: 0.10606876760721207 2023-01-23 19:56:09.578999: step: 600/1541, loss: 0.06599655002355576 2023-01-23 19:56:10.463433: step: 602/1541, loss: 0.022464146837592125 2023-01-23 19:56:11.408941: step: 604/1541, loss: 0.023266639560461044 2023-01-23 19:56:12.356453: step: 606/1541, loss: 0.012916267849504948 2023-01-23 19:56:13.316844: step: 608/1541, loss: 0.001244581420905888 2023-01-23 19:56:14.200514: step: 610/1541, loss: 7.495787576772273e-05 2023-01-23 19:56:15.072077: step: 612/1541, loss: 0.00316198100335896 2023-01-23 19:56:15.988582: step: 614/1541, loss: 0.0038524819537997246 2023-01-23 19:56:16.864413: step: 616/1541, loss: 0.00360656064003706 2023-01-23 19:56:17.775439: step: 618/1541, loss: 0.0025610607117414474 2023-01-23 19:56:18.618527: step: 620/1541, loss: 0.000726833357475698 2023-01-23 19:56:19.554578: step: 622/1541, loss: 0.10314011573791504 2023-01-23 19:56:20.434889: step: 624/1541, loss: 0.0010545639088377357 2023-01-23 19:56:21.325388: step: 626/1541, loss: 0.0001851528650149703 2023-01-23 19:56:22.208883: step: 628/1541, loss: 0.000629593210760504 2023-01-23 19:56:23.068199: step: 630/1541, loss: 0.11745457351207733 2023-01-23 19:56:24.000914: step: 632/1541, loss: 0.009002527222037315 2023-01-23 19:56:24.875157: step: 634/1541, loss: 0.0011835935292765498 2023-01-23 19:56:25.759685: step: 636/1541, loss: 0.008525951765477657 2023-01-23 19:56:26.619206: step: 638/1541, loss: 0.00442040478810668 2023-01-23 19:56:27.497178: step: 640/1541, loss: 0.054522402584552765 2023-01-23 19:56:28.347282: step: 642/1541, loss: 0.13172085583209991 2023-01-23 19:56:29.248860: step: 644/1541, loss: 0.3499695658683777 2023-01-23 19:56:30.163100: step: 646/1541, loss: 0.7681554555892944 2023-01-23 19:56:31.079889: step: 648/1541, loss: 0.06345761567354202 2023-01-23 19:56:31.949424: step: 650/1541, loss: 0.055903784930706024 2023-01-23 19:56:32.866407: step: 652/1541, loss: 0.0029519684612751007 2023-01-23 19:56:33.756801: step: 654/1541, loss: 0.06976109743118286 2023-01-23 19:56:34.639483: step: 656/1541, loss: 0.01946842297911644 2023-01-23 19:56:35.539319: step: 658/1541, loss: 0.0003661238879431039 2023-01-23 19:56:36.418645: step: 660/1541, loss: 0.0008367137052118778 2023-01-23 19:56:37.322700: step: 662/1541, loss: 0.06630824506282806 2023-01-23 19:56:38.193563: step: 664/1541, loss: 0.0037861058954149485 2023-01-23 19:56:39.036924: step: 666/1541, loss: 0.004137822892516851 2023-01-23 19:56:39.880890: step: 668/1541, loss: 0.07358146458864212 2023-01-23 19:56:40.755790: step: 670/1541, loss: 0.0015137320151552558 2023-01-23 19:56:41.646180: step: 672/1541, loss: 0.08482114225625992 2023-01-23 19:56:42.614142: step: 674/1541, loss: 0.0003854153328575194 2023-01-23 19:56:43.494003: step: 676/1541, loss: 0.09796621650457382 2023-01-23 19:56:44.362844: step: 678/1541, loss: 0.005779509898275137 2023-01-23 19:56:45.281696: step: 680/1541, loss: 0.001079936628229916 2023-01-23 19:56:46.290471: step: 682/1541, loss: 0.25055885314941406 2023-01-23 19:56:47.178510: step: 684/1541, loss: 0.0007968880818225443 2023-01-23 19:56:48.069749: step: 686/1541, loss: 0.1941101998090744 2023-01-23 19:56:48.965498: step: 688/1541, loss: 0.0008348210831172764 2023-01-23 19:56:49.874055: step: 690/1541, loss: 0.00035601065610535443 2023-01-23 19:56:50.735219: step: 692/1541, loss: 0.17330364882946014 2023-01-23 19:56:51.626567: step: 694/1541, loss: 4.701626300811768 2023-01-23 19:56:52.563897: step: 696/1541, loss: 0.14418798685073853 2023-01-23 19:56:53.414485: step: 698/1541, loss: 0.000749138300307095 2023-01-23 19:56:54.320002: step: 700/1541, loss: 0.03685843572020531 2023-01-23 19:56:55.304961: step: 702/1541, loss: 0.0002489882754161954 2023-01-23 19:56:56.168392: step: 704/1541, loss: 0.00044497204362414777 2023-01-23 19:56:57.128369: step: 706/1541, loss: 0.0010658307000994682 2023-01-23 19:56:57.987532: step: 708/1541, loss: 0.002606150461360812 2023-01-23 19:56:58.869077: step: 710/1541, loss: 1.3104056119918823 2023-01-23 19:56:59.766917: step: 712/1541, loss: 0.0006781378178857267 2023-01-23 19:57:00.654064: step: 714/1541, loss: 0.01690634898841381 2023-01-23 19:57:01.563453: step: 716/1541, loss: 0.00021287276467774063 2023-01-23 19:57:02.425055: step: 718/1541, loss: 0.09754498302936554 2023-01-23 19:57:03.289850: step: 720/1541, loss: 0.13277842104434967 2023-01-23 19:57:04.208216: step: 722/1541, loss: 0.000913132622372359 2023-01-23 19:57:05.096587: step: 724/1541, loss: 0.00025272872881032526 2023-01-23 19:57:06.072927: step: 726/1541, loss: 0.10891498625278473 2023-01-23 19:57:06.975454: step: 728/1541, loss: 0.0002171690430259332 2023-01-23 19:57:07.870101: step: 730/1541, loss: 0.143055260181427 2023-01-23 19:57:08.762630: step: 732/1541, loss: 0.2960297167301178 2023-01-23 19:57:09.657136: step: 734/1541, loss: 0.39921632409095764 2023-01-23 19:57:10.551211: step: 736/1541, loss: 0.000578188628423959 2023-01-23 19:57:11.467519: step: 738/1541, loss: 0.007076953537762165 2023-01-23 19:57:12.309645: step: 740/1541, loss: 0.0005386968259699643 2023-01-23 19:57:13.183685: step: 742/1541, loss: 0.0006332477205432951 2023-01-23 19:57:14.083792: step: 744/1541, loss: 0.0001685098686721176 2023-01-23 19:57:14.979723: step: 746/1541, loss: 0.0016919930931180716 2023-01-23 19:57:15.967070: step: 748/1541, loss: 3.128743264824152e-05 2023-01-23 19:57:16.871912: step: 750/1541, loss: 0.02297019213438034 2023-01-23 19:57:17.753967: step: 752/1541, loss: 1.693904232524801e-05 2023-01-23 19:57:18.601314: step: 754/1541, loss: 0.0004608367453329265 2023-01-23 19:57:19.510135: step: 756/1541, loss: 0.00017386628314852715 2023-01-23 19:57:20.402081: step: 758/1541, loss: 0.00882378313690424 2023-01-23 19:57:21.277665: step: 760/1541, loss: 0.5556791424751282 2023-01-23 19:57:22.209020: step: 762/1541, loss: 0.0038365311920642853 2023-01-23 19:57:23.103595: step: 764/1541, loss: 0.0004973644390702248 2023-01-23 19:57:24.042999: step: 766/1541, loss: 0.004055412486195564 2023-01-23 19:57:24.933342: step: 768/1541, loss: 0.4894099235534668 2023-01-23 19:57:25.845668: step: 770/1541, loss: 0.010777512565255165 2023-01-23 19:57:26.701870: step: 772/1541, loss: 0.0007201260305009782 2023-01-23 19:57:27.609977: step: 774/1541, loss: 0.028128739446401596 2023-01-23 19:57:28.469649: step: 776/1541, loss: 0.0001093935061362572 2023-01-23 19:57:29.348928: step: 778/1541, loss: 0.03078591637313366 2023-01-23 19:57:30.245188: step: 780/1541, loss: 0.063689686357975 2023-01-23 19:57:31.087723: step: 782/1541, loss: 0.0003644872922450304 2023-01-23 19:57:31.985796: step: 784/1541, loss: 3.712385660037398e-05 2023-01-23 19:57:32.898534: step: 786/1541, loss: 7.590485438413452e-06 2023-01-23 19:57:33.825532: step: 788/1541, loss: 0.00037006568163633347 2023-01-23 19:57:34.746154: step: 790/1541, loss: 0.35667556524276733 2023-01-23 19:57:35.676867: step: 792/1541, loss: 0.0008099857368506491 2023-01-23 19:57:36.539295: step: 794/1541, loss: 6.913914694450796e-05 2023-01-23 19:57:37.450367: step: 796/1541, loss: 0.00027607654919847846 2023-01-23 19:57:38.348012: step: 798/1541, loss: 0.16673527657985687 2023-01-23 19:57:39.224349: step: 800/1541, loss: 0.006826607044786215 2023-01-23 19:57:40.085857: step: 802/1541, loss: 0.04093402996659279 2023-01-23 19:57:40.952356: step: 804/1541, loss: 0.00029243106837384403 2023-01-23 19:57:41.807234: step: 806/1541, loss: 0.00019768287893384695 2023-01-23 19:57:42.780263: step: 808/1541, loss: 0.07462867349386215 2023-01-23 19:57:43.768686: step: 810/1541, loss: 0.0027853164356201887 2023-01-23 19:57:44.732115: step: 812/1541, loss: 0.02013576403260231 2023-01-23 19:57:45.588288: step: 814/1541, loss: 2.1611729607684538e-05 2023-01-23 19:57:46.475123: step: 816/1541, loss: 0.04002024605870247 2023-01-23 19:57:47.372369: step: 818/1541, loss: 0.00015799755055923015 2023-01-23 19:57:48.253906: step: 820/1541, loss: 0.08876875787973404 2023-01-23 19:57:49.165790: step: 822/1541, loss: 0.08154312521219254 2023-01-23 19:57:50.106749: step: 824/1541, loss: 0.00193198979832232 2023-01-23 19:57:51.033981: step: 826/1541, loss: 0.0008697830489836633 2023-01-23 19:57:51.929168: step: 828/1541, loss: 0.000500878959428519 2023-01-23 19:57:52.833204: step: 830/1541, loss: 0.0021492354571819305 2023-01-23 19:57:53.728750: step: 832/1541, loss: 0.0024169159587472677 2023-01-23 19:57:54.597285: step: 834/1541, loss: 0.2275310605764389 2023-01-23 19:57:55.499094: step: 836/1541, loss: 0.00011846348934341222 2023-01-23 19:57:56.383870: step: 838/1541, loss: 0.0031932450365275145 2023-01-23 19:57:57.280185: step: 840/1541, loss: 0.016988001763820648 2023-01-23 19:57:58.214566: step: 842/1541, loss: 0.017990535125136375 2023-01-23 19:57:59.082182: step: 844/1541, loss: 0.0036965664476156235 2023-01-23 19:57:59.981040: step: 846/1541, loss: 0.00012147649249527603 2023-01-23 19:58:00.832928: step: 848/1541, loss: 0.003910656087100506 2023-01-23 19:58:01.779113: step: 850/1541, loss: 0.0013783088652417064 2023-01-23 19:58:02.747219: step: 852/1541, loss: 0.016850320622324944 2023-01-23 19:58:03.705976: step: 854/1541, loss: 0.0007883690996095538 2023-01-23 19:58:04.582415: step: 856/1541, loss: 0.0014914445346221328 2023-01-23 19:58:05.451672: step: 858/1541, loss: 0.001871175947599113 2023-01-23 19:58:06.351889: step: 860/1541, loss: -1.0877223758143373e-05 2023-01-23 19:58:07.252789: step: 862/1541, loss: 0.2277849018573761 2023-01-23 19:58:08.228933: step: 864/1541, loss: 0.071397565305233 2023-01-23 19:58:09.108861: step: 866/1541, loss: 0.0016029292019084096 2023-01-23 19:58:10.019215: step: 868/1541, loss: 0.005873962305486202 2023-01-23 19:58:10.919770: step: 870/1541, loss: 0.00018573940906208009 2023-01-23 19:58:11.770239: step: 872/1541, loss: 0.0004110414010938257 2023-01-23 19:58:12.681114: step: 874/1541, loss: 0.006699929945170879 2023-01-23 19:58:13.673230: step: 876/1541, loss: 0.00019116017210762948 2023-01-23 19:58:14.601208: step: 878/1541, loss: 0.5799528956413269 2023-01-23 19:58:15.501226: step: 880/1541, loss: 5.3268431656761095e-06 2023-01-23 19:58:16.385504: step: 882/1541, loss: 0.2526939809322357 2023-01-23 19:58:17.311531: step: 884/1541, loss: 1.8242297301185317e-05 2023-01-23 19:58:18.183235: step: 886/1541, loss: 0.0016014124266803265 2023-01-23 19:58:19.068144: step: 888/1541, loss: 0.03074544481933117 2023-01-23 19:58:19.966230: step: 890/1541, loss: 0.10062696039676666 2023-01-23 19:58:20.856818: step: 892/1541, loss: 0.28966569900512695 2023-01-23 19:58:21.691103: step: 894/1541, loss: 0.0006897225975990295 2023-01-23 19:58:22.539876: step: 896/1541, loss: 0.15590260922908783 2023-01-23 19:58:23.402123: step: 898/1541, loss: 0.003143665613606572 2023-01-23 19:58:24.259076: step: 900/1541, loss: 5.511194467544556e-05 2023-01-23 19:58:25.144941: step: 902/1541, loss: 0.0028679054230451584 2023-01-23 19:58:26.023860: step: 904/1541, loss: 0.0023431668523699045 2023-01-23 19:58:26.906659: step: 906/1541, loss: 0.2339869886636734 2023-01-23 19:58:27.778938: step: 908/1541, loss: 1.527809500694275 2023-01-23 19:58:28.643489: step: 910/1541, loss: 0.0017977298703044653 2023-01-23 19:58:29.552133: step: 912/1541, loss: 0.005816061981022358 2023-01-23 19:58:30.453519: step: 914/1541, loss: 0.11722835898399353 2023-01-23 19:58:31.340134: step: 916/1541, loss: 0.0005530784837901592 2023-01-23 19:58:32.301000: step: 918/1541, loss: 0.1339164823293686 2023-01-23 19:58:33.189956: step: 920/1541, loss: 0.00023624609457328916 2023-01-23 19:58:34.063350: step: 922/1541, loss: 0.001171771204099059 2023-01-23 19:58:34.955960: step: 924/1541, loss: 0.0007383649353869259 2023-01-23 19:58:35.929394: step: 926/1541, loss: 0.0005351410945877433 2023-01-23 19:58:36.820640: step: 928/1541, loss: 0.00034650342422537506 2023-01-23 19:58:37.719066: step: 930/1541, loss: 0.0004313074750825763 2023-01-23 19:58:38.597574: step: 932/1541, loss: 0.1352768838405609 2023-01-23 19:58:39.484358: step: 934/1541, loss: 0.007877221331000328 2023-01-23 19:58:40.365603: step: 936/1541, loss: 0.00450742244720459 2023-01-23 19:58:41.301402: step: 938/1541, loss: 0.00274839554913342 2023-01-23 19:58:42.197302: step: 940/1541, loss: 0.009175991639494896 2023-01-23 19:58:43.184578: step: 942/1541, loss: 0.0006107556982897222 2023-01-23 19:58:44.119269: step: 944/1541, loss: 0.024500291794538498 2023-01-23 19:58:45.128338: step: 946/1541, loss: 0.14371490478515625 2023-01-23 19:58:46.023483: step: 948/1541, loss: 4.159879972576164e-05 2023-01-23 19:58:46.897597: step: 950/1541, loss: 0.0009933511028066278 2023-01-23 19:58:47.764138: step: 952/1541, loss: 0.0034961840137839317 2023-01-23 19:58:48.656834: step: 954/1541, loss: 0.0164937824010849 2023-01-23 19:58:49.580985: step: 956/1541, loss: 0.030940325930714607 2023-01-23 19:58:50.431396: step: 958/1541, loss: 0.07754611223936081 2023-01-23 19:58:51.294491: step: 960/1541, loss: 0.03015454299747944 2023-01-23 19:58:52.171277: step: 962/1541, loss: 0.016632499173283577 2023-01-23 19:58:53.113265: step: 964/1541, loss: 0.0003963506023865193 2023-01-23 19:58:54.002101: step: 966/1541, loss: 0.07921195775270462 2023-01-23 19:58:54.982668: step: 968/1541, loss: 0.01643635332584381 2023-01-23 19:58:55.885692: step: 970/1541, loss: 0.001385314972139895 2023-01-23 19:58:56.779492: step: 972/1541, loss: 0.004333584103733301 2023-01-23 19:58:57.672322: step: 974/1541, loss: 0.0014355932362377644 2023-01-23 19:58:58.607593: step: 976/1541, loss: 0.0008366358815692365 2023-01-23 19:58:59.477790: step: 978/1541, loss: 0.09410247951745987 2023-01-23 19:59:00.339360: step: 980/1541, loss: 0.2574513554573059 2023-01-23 19:59:01.201833: step: 982/1541, loss: 0.0010885426308959723 2023-01-23 19:59:02.097616: step: 984/1541, loss: 0.011166204698383808 2023-01-23 19:59:02.996520: step: 986/1541, loss: 0.3160668611526489 2023-01-23 19:59:03.901988: step: 988/1541, loss: 0.035986799746751785 2023-01-23 19:59:04.824613: step: 990/1541, loss: 0.02986771985888481 2023-01-23 19:59:05.696782: step: 992/1541, loss: 0.0004590658354572952 2023-01-23 19:59:06.595492: step: 994/1541, loss: 0.15561918914318085 2023-01-23 19:59:07.492496: step: 996/1541, loss: 0.3099207580089569 2023-01-23 19:59:08.373912: step: 998/1541, loss: 0.5223041772842407 2023-01-23 19:59:09.286968: step: 1000/1541, loss: 0.002481292001903057 2023-01-23 19:59:10.173464: step: 1002/1541, loss: 0.00633441936224699 2023-01-23 19:59:11.120571: step: 1004/1541, loss: 0.00027932884404435754 2023-01-23 19:59:12.039481: step: 1006/1541, loss: 0.0002431452157907188 2023-01-23 19:59:12.929792: step: 1008/1541, loss: 0.0009051894885487854 2023-01-23 19:59:13.833266: step: 1010/1541, loss: 0.00043317602830938995 2023-01-23 19:59:14.739092: step: 1012/1541, loss: 0.00017852177552413195 2023-01-23 19:59:15.606870: step: 1014/1541, loss: 0.002234897343441844 2023-01-23 19:59:16.496744: step: 1016/1541, loss: 5.887547740712762e-05 2023-01-23 19:59:17.403631: step: 1018/1541, loss: 0.00370397069491446 2023-01-23 19:59:18.271167: step: 1020/1541, loss: 0.0004552606260403991 2023-01-23 19:59:19.185255: step: 1022/1541, loss: 0.13179390132427216 2023-01-23 19:59:20.093398: step: 1024/1541, loss: 0.0008750552078709006 2023-01-23 19:59:20.966799: step: 1026/1541, loss: 0.2885045111179352 2023-01-23 19:59:21.861922: step: 1028/1541, loss: 0.06119997426867485 2023-01-23 19:59:22.725981: step: 1030/1541, loss: 0.006839648820459843 2023-01-23 19:59:23.644186: step: 1032/1541, loss: 0.17862898111343384 2023-01-23 19:59:24.554320: step: 1034/1541, loss: 0.002981899306178093 2023-01-23 19:59:25.448694: step: 1036/1541, loss: 0.002437350805848837 2023-01-23 19:59:26.373766: step: 1038/1541, loss: 0.00017016188940033317 2023-01-23 19:59:27.266393: step: 1040/1541, loss: 0.006312074139714241 2023-01-23 19:59:28.169166: step: 1042/1541, loss: 6.071434836485423e-05 2023-01-23 19:59:29.051977: step: 1044/1541, loss: 0.0009288007277064025 2023-01-23 19:59:29.950778: step: 1046/1541, loss: 0.00026726536452770233 2023-01-23 19:59:30.837500: step: 1048/1541, loss: 0.0019922677893191576 2023-01-23 19:59:31.749205: step: 1050/1541, loss: 0.013258561491966248 2023-01-23 19:59:32.625578: step: 1052/1541, loss: 5.528277688426897e-05 2023-01-23 19:59:33.523685: step: 1054/1541, loss: 0.0019732879009097815 2023-01-23 19:59:34.409045: step: 1056/1541, loss: 0.00026052308385260403 2023-01-23 19:59:35.285543: step: 1058/1541, loss: 0.0835162028670311 2023-01-23 19:59:36.194242: step: 1060/1541, loss: 0.0007896032766439021 2023-01-23 19:59:37.072887: step: 1062/1541, loss: 0.00014152127550914884 2023-01-23 19:59:37.953862: step: 1064/1541, loss: 0.03620356321334839 2023-01-23 19:59:38.832777: step: 1066/1541, loss: 0.0034912959672510624 2023-01-23 19:59:39.747870: step: 1068/1541, loss: 0.0014868222642689943 2023-01-23 19:59:40.610818: step: 1070/1541, loss: 0.00770982401445508 2023-01-23 19:59:41.527520: step: 1072/1541, loss: 0.006158025935292244 2023-01-23 19:59:42.418081: step: 1074/1541, loss: 0.19401755928993225 2023-01-23 19:59:43.348240: step: 1076/1541, loss: 0.004698640666902065 2023-01-23 19:59:44.246757: step: 1078/1541, loss: 0.00748712383210659 2023-01-23 19:59:45.129637: step: 1080/1541, loss: 4.628148235497065e-05 2023-01-23 19:59:45.963744: step: 1082/1541, loss: 6.752909393981099e-05 2023-01-23 19:59:46.889555: step: 1084/1541, loss: 0.005145952105522156 2023-01-23 19:59:47.848047: step: 1086/1541, loss: 0.016175933182239532 2023-01-23 19:59:48.716916: step: 1088/1541, loss: 0.000840745575260371 2023-01-23 19:59:49.590823: step: 1090/1541, loss: 0.024083716794848442 2023-01-23 19:59:50.529084: step: 1092/1541, loss: 0.08029612898826599 2023-01-23 19:59:51.404952: step: 1094/1541, loss: 0.000628767884336412 2023-01-23 19:59:52.275248: step: 1096/1541, loss: 0.0014262552140280604 2023-01-23 19:59:53.111923: step: 1098/1541, loss: 0.023800330236554146 2023-01-23 19:59:53.994783: step: 1100/1541, loss: 0.0020309449173510075 2023-01-23 19:59:54.861122: step: 1102/1541, loss: 0.0018595076398923993 2023-01-23 19:59:55.767999: step: 1104/1541, loss: 0.06870588660240173 2023-01-23 19:59:56.657287: step: 1106/1541, loss: 0.012903759256005287 2023-01-23 19:59:57.542865: step: 1108/1541, loss: 0.9703381061553955 2023-01-23 19:59:58.430841: step: 1110/1541, loss: 0.0008480456890538335 2023-01-23 19:59:59.325487: step: 1112/1541, loss: 0.001844768994487822 2023-01-23 20:00:00.225715: step: 1114/1541, loss: 0.0001312973618041724 2023-01-23 20:00:01.123196: step: 1116/1541, loss: 3.0237339160521515e-05 2023-01-23 20:00:02.084384: step: 1118/1541, loss: 0.0001037101392284967 2023-01-23 20:00:02.980463: step: 1120/1541, loss: 0.00022470099793281406 2023-01-23 20:00:03.849864: step: 1122/1541, loss: 0.0010323491878807545 2023-01-23 20:00:04.740870: step: 1124/1541, loss: 0.06386001408100128 2023-01-23 20:00:05.594574: step: 1126/1541, loss: 0.00014671444660052657 2023-01-23 20:00:06.545740: step: 1128/1541, loss: 0.0004223905270919204 2023-01-23 20:00:07.420338: step: 1130/1541, loss: 0.019328966736793518 2023-01-23 20:00:08.320945: step: 1132/1541, loss: 0.09052020311355591 2023-01-23 20:00:09.231314: step: 1134/1541, loss: 0.00035304404445923865 2023-01-23 20:00:10.114847: step: 1136/1541, loss: 0.00011644407641142607 2023-01-23 20:00:11.043357: step: 1138/1541, loss: 7.3846778832376e-05 2023-01-23 20:00:11.940083: step: 1140/1541, loss: 0.0009283435647375882 2023-01-23 20:00:12.857283: step: 1142/1541, loss: 0.673041045665741 2023-01-23 20:00:13.733232: step: 1144/1541, loss: 0.12472909688949585 2023-01-23 20:00:14.658175: step: 1146/1541, loss: 0.025891322642564774 2023-01-23 20:00:15.549142: step: 1148/1541, loss: 0.0019040359184145927 2023-01-23 20:00:16.510138: step: 1150/1541, loss: 0.07840500771999359 2023-01-23 20:00:17.439851: step: 1152/1541, loss: 0.00020418174972292036 2023-01-23 20:00:18.382696: step: 1154/1541, loss: 0.0019574821926653385 2023-01-23 20:00:19.291389: step: 1156/1541, loss: 0.0027260161004960537 2023-01-23 20:00:20.255243: step: 1158/1541, loss: 0.0009340682299807668 2023-01-23 20:00:21.193883: step: 1160/1541, loss: 0.00015159175381995738 2023-01-23 20:00:22.086563: step: 1162/1541, loss: 0.0018156940350309014 2023-01-23 20:00:22.996084: step: 1164/1541, loss: 0.0013286551693454385 2023-01-23 20:00:23.959053: step: 1166/1541, loss: 0.4319537281990051 2023-01-23 20:00:24.880796: step: 1168/1541, loss: 0.0019173084292560816 2023-01-23 20:00:25.806047: step: 1170/1541, loss: 0.008656635880470276 2023-01-23 20:00:26.665571: step: 1172/1541, loss: 0.0005407995195128024 2023-01-23 20:00:27.518814: step: 1174/1541, loss: 0.016120461747050285 2023-01-23 20:00:28.474105: step: 1176/1541, loss: 0.06472426652908325 2023-01-23 20:00:29.428488: step: 1178/1541, loss: 0.0002786841068882495 2023-01-23 20:00:30.303330: step: 1180/1541, loss: 0.001959319459274411 2023-01-23 20:00:31.161122: step: 1182/1541, loss: 0.1411847025156021 2023-01-23 20:00:32.028897: step: 1184/1541, loss: 0.02642570249736309 2023-01-23 20:00:32.939904: step: 1186/1541, loss: 0.0007068678387440741 2023-01-23 20:00:33.846380: step: 1188/1541, loss: 0.035994090139865875 2023-01-23 20:00:34.750686: step: 1190/1541, loss: 0.009144000709056854 2023-01-23 20:00:35.676344: step: 1192/1541, loss: 0.37907636165618896 2023-01-23 20:00:36.557381: step: 1194/1541, loss: 0.00014284100325312465 2023-01-23 20:00:37.425717: step: 1196/1541, loss: 0.0001894235610961914 2023-01-23 20:00:38.357138: step: 1198/1541, loss: 0.004626648034900427 2023-01-23 20:00:39.210828: step: 1200/1541, loss: 0.00530595937743783 2023-01-23 20:00:40.073148: step: 1202/1541, loss: 0.00019959181372541934 2023-01-23 20:00:40.931109: step: 1204/1541, loss: 0.007564004510641098 2023-01-23 20:00:41.804597: step: 1206/1541, loss: 0.1046348437666893 2023-01-23 20:00:42.714863: step: 1208/1541, loss: 0.001962208654731512 2023-01-23 20:00:43.595859: step: 1210/1541, loss: 0.010535377077758312 2023-01-23 20:00:44.467142: step: 1212/1541, loss: 0.12899582087993622 2023-01-23 20:00:45.350530: step: 1214/1541, loss: 0.001805448904633522 2023-01-23 20:00:46.256093: step: 1216/1541, loss: 0.001293090172111988 2023-01-23 20:00:47.240913: step: 1218/1541, loss: 0.0008053393103182316 2023-01-23 20:00:48.090362: step: 1220/1541, loss: 0.00041918217903003097 2023-01-23 20:00:48.983295: step: 1222/1541, loss: 0.028300171718001366 2023-01-23 20:00:49.862186: step: 1224/1541, loss: 9.650979336583987e-05 2023-01-23 20:00:50.755450: step: 1226/1541, loss: 0.06639038026332855 2023-01-23 20:00:51.632524: step: 1228/1541, loss: 0.2048145979642868 2023-01-23 20:00:52.496384: step: 1230/1541, loss: 0.0072282347828149796 2023-01-23 20:00:53.377303: step: 1232/1541, loss: 0.0007149633602239192 2023-01-23 20:00:54.265963: step: 1234/1541, loss: 0.03374669328331947 2023-01-23 20:00:55.236772: step: 1236/1541, loss: 0.0015452756779268384 2023-01-23 20:00:56.134325: step: 1238/1541, loss: 0.13658659160137177 2023-01-23 20:00:57.009998: step: 1240/1541, loss: 0.07922050356864929 2023-01-23 20:00:57.942255: step: 1242/1541, loss: 0.0014981661224737763 2023-01-23 20:00:58.843194: step: 1244/1541, loss: 0.1454458087682724 2023-01-23 20:00:59.695429: step: 1246/1541, loss: 0.004053423181176186 2023-01-23 20:01:00.567182: step: 1248/1541, loss: 0.0002524059673305601 2023-01-23 20:01:01.479826: step: 1250/1541, loss: 0.002094478579238057 2023-01-23 20:01:02.391744: step: 1252/1541, loss: 0.00215061753988266 2023-01-23 20:01:03.314195: step: 1254/1541, loss: 0.10459718853235245 2023-01-23 20:01:04.206670: step: 1256/1541, loss: 0.12051813304424286 2023-01-23 20:01:05.161438: step: 1258/1541, loss: 0.13588210940361023 2023-01-23 20:01:06.045218: step: 1260/1541, loss: 0.019489066675305367 2023-01-23 20:01:06.914926: step: 1262/1541, loss: 0.9828168153762817 2023-01-23 20:01:07.824853: step: 1264/1541, loss: 0.03118269145488739 2023-01-23 20:01:08.753669: step: 1266/1541, loss: 0.25407078862190247 2023-01-23 20:01:09.716792: step: 1268/1541, loss: 0.0005622763419523835 2023-01-23 20:01:10.709455: step: 1270/1541, loss: 0.00014635382103733718 2023-01-23 20:01:11.678289: step: 1272/1541, loss: 0.00014381890650838614 2023-01-23 20:01:12.549275: step: 1274/1541, loss: 0.013342531397938728 2023-01-23 20:01:13.427013: step: 1276/1541, loss: 0.03553009033203125 2023-01-23 20:01:14.358269: step: 1278/1541, loss: 0.00026945260469801724 2023-01-23 20:01:15.331642: step: 1280/1541, loss: 0.3680209219455719 2023-01-23 20:01:16.239924: step: 1282/1541, loss: 0.07703421264886856 2023-01-23 20:01:17.113867: step: 1284/1541, loss: 0.14025302231311798 2023-01-23 20:01:17.993783: step: 1286/1541, loss: 0.015483864583075047 2023-01-23 20:01:18.887000: step: 1288/1541, loss: 0.0016668150201439857 2023-01-23 20:01:19.775907: step: 1290/1541, loss: 0.0018987971125170588 2023-01-23 20:01:20.681082: step: 1292/1541, loss: 0.0128376679494977 2023-01-23 20:01:21.576027: step: 1294/1541, loss: 0.02698882296681404 2023-01-23 20:01:22.504496: step: 1296/1541, loss: 0.01466536894440651 2023-01-23 20:01:23.456018: step: 1298/1541, loss: 0.00019035131845157593 2023-01-23 20:01:24.315307: step: 1300/1541, loss: 0.0001804089843062684 2023-01-23 20:01:25.218109: step: 1302/1541, loss: 0.0013814154081046581 2023-01-23 20:01:26.160694: step: 1304/1541, loss: 7.908149564173073e-05 2023-01-23 20:01:27.102628: step: 1306/1541, loss: 0.0010984994005411863 2023-01-23 20:01:27.997953: step: 1308/1541, loss: 0.0004105370317120105 2023-01-23 20:01:28.922077: step: 1310/1541, loss: 0.022461585700511932 2023-01-23 20:01:29.790362: step: 1312/1541, loss: 7.274943345692009e-05 2023-01-23 20:01:30.651208: step: 1314/1541, loss: 0.0003213214222341776 2023-01-23 20:01:31.516964: step: 1316/1541, loss: 0.0023943879641592503 2023-01-23 20:01:32.421138: step: 1318/1541, loss: 0.03287962079048157 2023-01-23 20:01:33.327363: step: 1320/1541, loss: 0.0021240650676190853 2023-01-23 20:01:34.271556: step: 1322/1541, loss: 0.001413289806805551 2023-01-23 20:01:35.175989: step: 1324/1541, loss: 0.011658300645649433 2023-01-23 20:01:36.029190: step: 1326/1541, loss: 0.013232581317424774 2023-01-23 20:01:36.934478: step: 1328/1541, loss: 6.818555993959308e-05 2023-01-23 20:01:37.876471: step: 1330/1541, loss: 5.13233280798886e-05 2023-01-23 20:01:38.790510: step: 1332/1541, loss: 0.10114584863185883 2023-01-23 20:01:39.692240: step: 1334/1541, loss: 0.001817749347537756 2023-01-23 20:01:40.599593: step: 1336/1541, loss: 0.0005763734807260334 2023-01-23 20:01:41.491735: step: 1338/1541, loss: 1.7115923166275024 2023-01-23 20:01:42.388779: step: 1340/1541, loss: 7.898244803072885e-05 2023-01-23 20:01:43.255647: step: 1342/1541, loss: 0.00015754616470076144 2023-01-23 20:01:44.128198: step: 1344/1541, loss: 0.040880344808101654 2023-01-23 20:01:45.031817: step: 1346/1541, loss: 0.0008701436454430223 2023-01-23 20:01:45.915220: step: 1348/1541, loss: 0.0011042207479476929 2023-01-23 20:01:46.769644: step: 1350/1541, loss: 0.00035180614213459194 2023-01-23 20:01:47.663211: step: 1352/1541, loss: 0.0001012274224194698 2023-01-23 20:01:48.523198: step: 1354/1541, loss: 1.3803261026623659e-05 2023-01-23 20:01:49.448176: step: 1356/1541, loss: 0.012304592877626419 2023-01-23 20:01:50.372969: step: 1358/1541, loss: 0.05354879051446915 2023-01-23 20:01:51.272358: step: 1360/1541, loss: 7.988724246388301e-05 2023-01-23 20:01:52.209750: step: 1362/1541, loss: 0.06686458736658096 2023-01-23 20:01:53.094757: step: 1364/1541, loss: 0.07625488936901093 2023-01-23 20:01:54.014896: step: 1366/1541, loss: 0.0014521413249894977 2023-01-23 20:01:54.895259: step: 1368/1541, loss: -2.672783193702344e-05 2023-01-23 20:01:55.849779: step: 1370/1541, loss: 0.04719983786344528 2023-01-23 20:01:56.768470: step: 1372/1541, loss: 8.806557161733508e-05 2023-01-23 20:01:57.671703: step: 1374/1541, loss: 0.01400737464427948 2023-01-23 20:01:58.549017: step: 1376/1541, loss: 7.81118797021918e-06 2023-01-23 20:01:59.455982: step: 1378/1541, loss: 0.0001238998956978321 2023-01-23 20:02:00.304134: step: 1380/1541, loss: 1.3138653230271302e-05 2023-01-23 20:02:01.246893: step: 1382/1541, loss: 0.07124336808919907 2023-01-23 20:02:02.209689: step: 1384/1541, loss: 0.0945734903216362 2023-01-23 20:02:03.099734: step: 1386/1541, loss: 0.00012956018326804042 2023-01-23 20:02:03.997873: step: 1388/1541, loss: 3.086403012275696e-05 2023-01-23 20:02:04.914420: step: 1390/1541, loss: -2.5760064090718515e-05 2023-01-23 20:02:05.848319: step: 1392/1541, loss: 0.006724216975271702 2023-01-23 20:02:06.721975: step: 1394/1541, loss: 0.0802624374628067 2023-01-23 20:02:07.615467: step: 1396/1541, loss: 0.11259549111127853 2023-01-23 20:02:08.492443: step: 1398/1541, loss: 0.00010890864359680563 2023-01-23 20:02:09.381211: step: 1400/1541, loss: 0.0010208514286205173 2023-01-23 20:02:10.282410: step: 1402/1541, loss: 0.5504513382911682 2023-01-23 20:02:11.162672: step: 1404/1541, loss: 0.0001051120416377671 2023-01-23 20:02:12.077552: step: 1406/1541, loss: 0.2338494062423706 2023-01-23 20:02:13.016181: step: 1408/1541, loss: 0.0001351617684122175 2023-01-23 20:02:13.901766: step: 1410/1541, loss: 0.011955926194787025 2023-01-23 20:02:14.829237: step: 1412/1541, loss: 0.0003328946477267891 2023-01-23 20:02:15.685090: step: 1414/1541, loss: 7.904579979367554e-05 2023-01-23 20:02:16.609451: step: 1416/1541, loss: 0.014353455044329166 2023-01-23 20:02:17.528597: step: 1418/1541, loss: 0.0003713947662618011 2023-01-23 20:02:18.399280: step: 1420/1541, loss: 0.0002467639569658786 2023-01-23 20:02:19.309308: step: 1422/1541, loss: 0.01682843267917633 2023-01-23 20:02:20.276008: step: 1424/1541, loss: 0.0040341829881072044 2023-01-23 20:02:21.179662: step: 1426/1541, loss: 0.023455273360013962 2023-01-23 20:02:22.089192: step: 1428/1541, loss: 0.015011253766715527 2023-01-23 20:02:23.005279: step: 1430/1541, loss: 0.00862300954759121 2023-01-23 20:02:23.881177: step: 1432/1541, loss: 0.018403397873044014 2023-01-23 20:02:24.756994: step: 1434/1541, loss: 0.19178491830825806 2023-01-23 20:02:25.666971: step: 1436/1541, loss: 0.0488445870578289 2023-01-23 20:02:26.650383: step: 1438/1541, loss: 0.008306741714477539 2023-01-23 20:02:27.532178: step: 1440/1541, loss: 0.03318487107753754 2023-01-23 20:02:28.406527: step: 1442/1541, loss: 0.1446533501148224 2023-01-23 20:02:29.325881: step: 1444/1541, loss: 0.016799263656139374 2023-01-23 20:02:30.196110: step: 1446/1541, loss: 0.04240843653678894 2023-01-23 20:02:31.113022: step: 1448/1541, loss: 0.0008973989170044661 2023-01-23 20:02:32.011599: step: 1450/1541, loss: 0.0006250993465073407 2023-01-23 20:02:32.913022: step: 1452/1541, loss: 0.12743251025676727 2023-01-23 20:02:33.891836: step: 1454/1541, loss: 0.008485479280352592 2023-01-23 20:02:34.824819: step: 1456/1541, loss: 6.994475552346557e-05 2023-01-23 20:02:35.735266: step: 1458/1541, loss: 0.22826774418354034 2023-01-23 20:02:36.661857: step: 1460/1541, loss: 0.000642425031401217 2023-01-23 20:02:37.532097: step: 1462/1541, loss: 0.19182616472244263 2023-01-23 20:02:38.394410: step: 1464/1541, loss: 0.007856630720198154 2023-01-23 20:02:39.263604: step: 1466/1541, loss: 0.00019637374498415738 2023-01-23 20:02:40.166071: step: 1468/1541, loss: 0.007365709636360407 2023-01-23 20:02:41.034464: step: 1470/1541, loss: 0.00011789458949351683 2023-01-23 20:02:41.993620: step: 1472/1541, loss: 0.0008823085809126496 2023-01-23 20:02:42.878645: step: 1474/1541, loss: 0.0021435143426060677 2023-01-23 20:02:43.779631: step: 1476/1541, loss: 0.17485259473323822 2023-01-23 20:02:44.690349: step: 1478/1541, loss: 0.08952105045318604 2023-01-23 20:02:45.615945: step: 1480/1541, loss: 0.0012769214808940887 2023-01-23 20:02:46.517247: step: 1482/1541, loss: 0.0052517615258693695 2023-01-23 20:02:47.435177: step: 1484/1541, loss: 0.13342006504535675 2023-01-23 20:02:48.439782: step: 1486/1541, loss: 0.0012216659961268306 2023-01-23 20:02:49.324897: step: 1488/1541, loss: 0.0036937780678272247 2023-01-23 20:02:50.218790: step: 1490/1541, loss: 0.05182451754808426 2023-01-23 20:02:51.063606: step: 1492/1541, loss: 0.0005062664858996868 2023-01-23 20:02:51.953850: step: 1494/1541, loss: 0.4429317116737366 2023-01-23 20:02:52.822671: step: 1496/1541, loss: 0.004379982128739357 2023-01-23 20:02:53.726122: step: 1498/1541, loss: 0.03905002027750015 2023-01-23 20:02:54.697145: step: 1500/1541, loss: 0.00030043290462344885 2023-01-23 20:02:55.647851: step: 1502/1541, loss: 0.017638741061091423 2023-01-23 20:02:56.557598: step: 1504/1541, loss: 0.0075405375100672245 2023-01-23 20:02:57.452901: step: 1506/1541, loss: 0.03745165467262268 2023-01-23 20:02:58.358337: step: 1508/1541, loss: 0.0003978745953645557 2023-01-23 20:02:59.261946: step: 1510/1541, loss: 0.003167622722685337 2023-01-23 20:03:00.128710: step: 1512/1541, loss: 0.040593914687633514 2023-01-23 20:03:01.046998: step: 1514/1541, loss: 0.002154436893761158 2023-01-23 20:03:01.949899: step: 1516/1541, loss: 1.694237232208252 2023-01-23 20:03:02.820565: step: 1518/1541, loss: 0.002052388619631529 2023-01-23 20:03:03.716399: step: 1520/1541, loss: 0.05242736637592316 2023-01-23 20:03:04.591315: step: 1522/1541, loss: 0.020761361345648766 2023-01-23 20:03:05.487557: step: 1524/1541, loss: 0.03302888199687004 2023-01-23 20:03:06.369326: step: 1526/1541, loss: 0.0037004933692514896 2023-01-23 20:03:07.260769: step: 1528/1541, loss: 0.13522708415985107 2023-01-23 20:03:08.150038: step: 1530/1541, loss: 0.000566842092666775 2023-01-23 20:03:09.034302: step: 1532/1541, loss: 0.5028311610221863 2023-01-23 20:03:09.936812: step: 1534/1541, loss: 7.924921374069527e-05 2023-01-23 20:03:10.867926: step: 1536/1541, loss: 0.00048637576401233673 2023-01-23 20:03:11.727824: step: 1538/1541, loss: 0.23068448901176453 2023-01-23 20:03:12.603082: step: 1540/1541, loss: 0.03558395802974701 2023-01-23 20:03:13.465448: step: 1542/1541, loss: 0.00310795521363616 2023-01-23 20:03:14.303103: step: 1544/1541, loss: 0.0005549495690502226 2023-01-23 20:03:15.176579: step: 1546/1541, loss: 0.2847992181777954 2023-01-23 20:03:16.057750: step: 1548/1541, loss: 0.00027845497243106365 2023-01-23 20:03:17.024224: step: 1550/1541, loss: 8.618037827545777e-05 2023-01-23 20:03:17.912280: step: 1552/1541, loss: 0.0005657074507325888 2023-01-23 20:03:18.817783: step: 1554/1541, loss: 0.0002119416749337688 2023-01-23 20:03:19.697659: step: 1556/1541, loss: 0.15958261489868164 2023-01-23 20:03:20.613354: step: 1558/1541, loss: 0.00030833500204607844 2023-01-23 20:03:21.488201: step: 1560/1541, loss: 0.016818730160593987 2023-01-23 20:03:22.385148: step: 1562/1541, loss: 0.0004289042844902724 2023-01-23 20:03:23.277651: step: 1564/1541, loss: 0.07429281622171402 2023-01-23 20:03:24.139581: step: 1566/1541, loss: 0.00019929705013055354 2023-01-23 20:03:25.058806: step: 1568/1541, loss: 0.00028660637326538563 2023-01-23 20:03:25.934422: step: 1570/1541, loss: 0.0435340590775013 2023-01-23 20:03:26.806110: step: 1572/1541, loss: 0.031084604561328888 2023-01-23 20:03:27.667171: step: 1574/1541, loss: 0.00028782489243894815 2023-01-23 20:03:28.575564: step: 1576/1541, loss: 0.00916782021522522 2023-01-23 20:03:29.407954: step: 1578/1541, loss: 0.0002745308156590909 2023-01-23 20:03:30.317200: step: 1580/1541, loss: 0.013607306405901909 2023-01-23 20:03:31.173973: step: 1582/1541, loss: 0.3170168101787567 2023-01-23 20:03:32.068314: step: 1584/1541, loss: 0.0013586271088570356 2023-01-23 20:03:33.003359: step: 1586/1541, loss: 0.00037341887946240604 2023-01-23 20:03:33.895676: step: 1588/1541, loss: 4.740131043945439e-05 2023-01-23 20:03:34.877124: step: 1590/1541, loss: 0.0028538673650473356 2023-01-23 20:03:35.749993: step: 1592/1541, loss: 0.006721526384353638 2023-01-23 20:03:36.616394: step: 1594/1541, loss: 0.00034363020677119493 2023-01-23 20:03:37.533721: step: 1596/1541, loss: 0.001342296483926475 2023-01-23 20:03:38.469001: step: 1598/1541, loss: 0.04950116574764252 2023-01-23 20:03:39.337510: step: 1600/1541, loss: 0.001802512677386403 2023-01-23 20:03:40.217010: step: 1602/1541, loss: 0.0005340416682884097 2023-01-23 20:03:41.128225: step: 1604/1541, loss: 0.005707127042114735 2023-01-23 20:03:42.080236: step: 1606/1541, loss: 0.009819273836910725 2023-01-23 20:03:42.970285: step: 1608/1541, loss: 0.01867619715631008 2023-01-23 20:03:43.850688: step: 1610/1541, loss: 0.011797968298196793 2023-01-23 20:03:44.733774: step: 1612/1541, loss: 0.010658862069249153 2023-01-23 20:03:45.642694: step: 1614/1541, loss: 0.002645005937665701 2023-01-23 20:03:46.470387: step: 1616/1541, loss: 0.0012657238403335214 2023-01-23 20:03:47.402116: step: 1618/1541, loss: 0.0003253124305047095 2023-01-23 20:03:48.268173: step: 1620/1541, loss: 0.0003834376693703234 2023-01-23 20:03:49.195792: step: 1622/1541, loss: 0.0003074509440921247 2023-01-23 20:03:50.058114: step: 1624/1541, loss: 0.1392667442560196 2023-01-23 20:03:50.931237: step: 1626/1541, loss: 0.030936088413000107 2023-01-23 20:03:51.840587: step: 1628/1541, loss: 0.12505975365638733 2023-01-23 20:03:52.741466: step: 1630/1541, loss: 0.14302505552768707 2023-01-23 20:03:53.673799: step: 1632/1541, loss: 0.13905006647109985 2023-01-23 20:03:54.545702: step: 1634/1541, loss: 0.00021078817371744663 2023-01-23 20:03:55.404274: step: 1636/1541, loss: 0.02057303674519062 2023-01-23 20:03:56.366890: step: 1638/1541, loss: 0.03353145718574524 2023-01-23 20:03:57.232861: step: 1640/1541, loss: 0.0007325045298784971 2023-01-23 20:03:58.167038: step: 1642/1541, loss: -2.6617653929861262e-05 2023-01-23 20:03:59.079023: step: 1644/1541, loss: 0.00015588989481329918 2023-01-23 20:03:59.959279: step: 1646/1541, loss: 0.001069746445864439 2023-01-23 20:04:00.838980: step: 1648/1541, loss: 0.011656239628791809 2023-01-23 20:04:01.781219: step: 1650/1541, loss: -7.727350748609751e-06 2023-01-23 20:04:02.655681: step: 1652/1541, loss: 0.010921292006969452 2023-01-23 20:04:03.551016: step: 1654/1541, loss: 6.761754320905311e-06 2023-01-23 20:04:04.425523: step: 1656/1541, loss: 0.001535880728624761 2023-01-23 20:04:05.272780: step: 1658/1541, loss: 0.03896917402744293 2023-01-23 20:04:06.135299: step: 1660/1541, loss: 0.00020753886201418936 2023-01-23 20:04:06.973890: step: 1662/1541, loss: -6.439931894419715e-07 2023-01-23 20:04:07.873481: step: 1664/1541, loss: 0.3616971969604492 2023-01-23 20:04:08.757672: step: 1666/1541, loss: 3.383839430171065e-05 2023-01-23 20:04:09.676123: step: 1668/1541, loss: 0.009549479931592941 2023-01-23 20:04:10.597040: step: 1670/1541, loss: 0.03810696303844452 2023-01-23 20:04:11.523347: step: 1672/1541, loss: 0.023793315514922142 2023-01-23 20:04:12.409944: step: 1674/1541, loss: 0.0006896818522363901 2023-01-23 20:04:13.289442: step: 1676/1541, loss: 0.000212829138035886 2023-01-23 20:04:14.178428: step: 1678/1541, loss: 0.1652003526687622 2023-01-23 20:04:15.088784: step: 1680/1541, loss: 0.00023879928630776703 2023-01-23 20:04:15.955660: step: 1682/1541, loss: 0.00018564678612165153 2023-01-23 20:04:16.818045: step: 1684/1541, loss: 0.1710907220840454 2023-01-23 20:04:17.744373: step: 1686/1541, loss: -9.579330253473017e-06 2023-01-23 20:04:18.609229: step: 1688/1541, loss: 0.0005080952541902661 2023-01-23 20:04:19.531868: step: 1690/1541, loss: 0.003778722370043397 2023-01-23 20:04:20.425897: step: 1692/1541, loss: 0.000783939438406378 2023-01-23 20:04:21.295499: step: 1694/1541, loss: 0.004933774936944246 2023-01-23 20:04:22.182098: step: 1696/1541, loss: 0.0016851251712068915 2023-01-23 20:04:23.035990: step: 1698/1541, loss: 0.00022062385687604547 2023-01-23 20:04:23.918805: step: 1700/1541, loss: 0.00336281955242157 2023-01-23 20:04:24.817542: step: 1702/1541, loss: -8.601696208643261e-06 2023-01-23 20:04:25.756345: step: 1704/1541, loss: 0.16295896470546722 2023-01-23 20:04:26.639058: step: 1706/1541, loss: 0.2797928750514984 2023-01-23 20:04:27.533662: step: 1708/1541, loss: 0.000128651357954368 2023-01-23 20:04:28.395490: step: 1710/1541, loss: -6.07154106546659e-06 2023-01-23 20:04:29.256144: step: 1712/1541, loss: 0.00022864065249450505 2023-01-23 20:04:30.126013: step: 1714/1541, loss: 0.09488805383443832 2023-01-23 20:04:31.022840: step: 1716/1541, loss: 0.003338838694617152 2023-01-23 20:04:31.900686: step: 1718/1541, loss: 0.0008755470626056194 2023-01-23 20:04:32.761388: step: 1720/1541, loss: 0.0011001923121511936 2023-01-23 20:04:33.662422: step: 1722/1541, loss: 0.011938749812543392 2023-01-23 20:04:34.519637: step: 1724/1541, loss: 0.000380056444555521 2023-01-23 20:04:35.386162: step: 1726/1541, loss: 7.56331646698527e-05 2023-01-23 20:04:36.267317: step: 1728/1541, loss: 0.0001554811024107039 2023-01-23 20:04:37.131873: step: 1730/1541, loss: 0.0017712113913148642 2023-01-23 20:04:38.002072: step: 1732/1541, loss: 0.017856750637292862 2023-01-23 20:04:38.942617: step: 1734/1541, loss: 0.014549423940479755 2023-01-23 20:04:39.823828: step: 1736/1541, loss: 0.030006498098373413 2023-01-23 20:04:40.661520: step: 1738/1541, loss: 0.0002411048481008038 2023-01-23 20:04:41.550603: step: 1740/1541, loss: 0.004487945698201656 2023-01-23 20:04:42.461937: step: 1742/1541, loss: 5.717635576729663e-05 2023-01-23 20:04:43.349959: step: 1744/1541, loss: 3.34748619934544e-05 2023-01-23 20:04:44.267508: step: 1746/1541, loss: 0.0002911749470513314 2023-01-23 20:04:45.175647: step: 1748/1541, loss: 0.001238383585587144 2023-01-23 20:04:46.074129: step: 1750/1541, loss: 9.948881779564545e-05 2023-01-23 20:04:46.979664: step: 1752/1541, loss: -1.7069200112018734e-05 2023-01-23 20:04:47.939021: step: 1754/1541, loss: 0.058828819543123245 2023-01-23 20:04:48.824064: step: 1756/1541, loss: 0.0002906025620177388 2023-01-23 20:04:49.747302: step: 1758/1541, loss: 0.15229807794094086 2023-01-23 20:04:50.656160: step: 1760/1541, loss: 0.0017815213650465012 2023-01-23 20:04:51.570778: step: 1762/1541, loss: 0.0005679734167642891 2023-01-23 20:04:52.486147: step: 1764/1541, loss: 0.0015532486140727997 2023-01-23 20:04:53.395530: step: 1766/1541, loss: 0.00019383427570573986 2023-01-23 20:04:54.312015: step: 1768/1541, loss: 0.08392231911420822 2023-01-23 20:04:55.191837: step: 1770/1541, loss: 0.2862301766872406 2023-01-23 20:04:56.088670: step: 1772/1541, loss: 0.06055387109518051 2023-01-23 20:04:56.999192: step: 1774/1541, loss: 0.12580251693725586 2023-01-23 20:04:57.870632: step: 1776/1541, loss: 6.925046182004735e-05 2023-01-23 20:04:58.765839: step: 1778/1541, loss: 0.0015707403654232621 2023-01-23 20:04:59.659313: step: 1780/1541, loss: 0.1429755538702011 2023-01-23 20:05:00.539661: step: 1782/1541, loss: 9.315461647929624e-05 2023-01-23 20:05:01.419913: step: 1784/1541, loss: 3.450065923971124e-05 2023-01-23 20:05:02.297759: step: 1786/1541, loss: 0.0005072295316495001 2023-01-23 20:05:03.229374: step: 1788/1541, loss: 1.3062729835510254 2023-01-23 20:05:04.102003: step: 1790/1541, loss: -6.064773060643347e-06 2023-01-23 20:05:04.969418: step: 1792/1541, loss: 0.0012802457204088569 2023-01-23 20:05:05.915020: step: 1794/1541, loss: 0.29813095927238464 2023-01-23 20:05:06.848877: step: 1796/1541, loss: 0.00140476378146559 2023-01-23 20:05:07.704322: step: 1798/1541, loss: 5.433783371699974e-05 2023-01-23 20:05:08.573015: step: 1800/1541, loss: 0.0008323884103447199 2023-01-23 20:05:09.443539: step: 1802/1541, loss: 8.982802683021873e-05 2023-01-23 20:05:10.329652: step: 1804/1541, loss: -2.0931072867824696e-05 2023-01-23 20:05:11.236002: step: 1806/1541, loss: 2.2449832613347098e-05 2023-01-23 20:05:12.161877: step: 1808/1541, loss: 0.001996910898014903 2023-01-23 20:05:13.131742: step: 1810/1541, loss: 0.0012909292709082365 2023-01-23 20:05:13.991744: step: 1812/1541, loss: 0.00011561246355995536 2023-01-23 20:05:14.917900: step: 1814/1541, loss: 0.022648002952337265 2023-01-23 20:05:15.799089: step: 1816/1541, loss: 0.19569525122642517 2023-01-23 20:05:16.815295: step: 1818/1541, loss: 0.029284141957759857 2023-01-23 20:05:17.747072: step: 1820/1541, loss: 0.09875226765871048 2023-01-23 20:05:18.724209: step: 1822/1541, loss: 0.22662121057510376 2023-01-23 20:05:19.584580: step: 1824/1541, loss: 0.003018822753801942 2023-01-23 20:05:20.489143: step: 1826/1541, loss: 0.499362051486969 2023-01-23 20:05:21.386267: step: 1828/1541, loss: 0.1477733850479126 2023-01-23 20:05:22.305571: step: 1830/1541, loss: 3.497501165838912e-05 2023-01-23 20:05:23.194274: step: 1832/1541, loss: 0.0019035369623452425 2023-01-23 20:05:24.096571: step: 1834/1541, loss: 0.06411891430616379 2023-01-23 20:05:25.074662: step: 1836/1541, loss: 0.0061241514049470425 2023-01-23 20:05:25.972692: step: 1838/1541, loss: 5.280311233946122e-05 2023-01-23 20:05:26.883520: step: 1840/1541, loss: 7.554553303634748e-05 2023-01-23 20:05:27.790297: step: 1842/1541, loss: 0.0005085743614472449 2023-01-23 20:05:28.707165: step: 1844/1541, loss: 0.04245306923985481 2023-01-23 20:05:29.576769: step: 1846/1541, loss: 0.006472240202128887 2023-01-23 20:05:30.453109: step: 1848/1541, loss: 0.21317392587661743 2023-01-23 20:05:31.337312: step: 1850/1541, loss: 0.04551315680146217 2023-01-23 20:05:32.260456: step: 1852/1541, loss: 0.001474865130148828 2023-01-23 20:05:33.158958: step: 1854/1541, loss: 0.1549016386270523 2023-01-23 20:05:34.052254: step: 1856/1541, loss: 0.04823394864797592 2023-01-23 20:05:34.936198: step: 1858/1541, loss: 0.0001455515593988821 2023-01-23 20:05:35.836844: step: 1860/1541, loss: 0.0046546803787350655 2023-01-23 20:05:36.747268: step: 1862/1541, loss: 0.0011404952965676785 2023-01-23 20:05:37.659635: step: 1864/1541, loss: 0.0034333723597228527 2023-01-23 20:05:38.556981: step: 1866/1541, loss: 0.004423885140568018 2023-01-23 20:05:39.465449: step: 1868/1541, loss: 0.1931753307580948 2023-01-23 20:05:40.360509: step: 1870/1541, loss: 0.010126972571015358 2023-01-23 20:05:41.275882: step: 1872/1541, loss: 0.01471518725156784 2023-01-23 20:05:42.140416: step: 1874/1541, loss: 3.609709892771207e-05 2023-01-23 20:05:43.032109: step: 1876/1541, loss: 0.0030261368956416845 2023-01-23 20:05:43.901722: step: 1878/1541, loss: 0.0016587238060310483 2023-01-23 20:05:44.779314: step: 1880/1541, loss: 0.00029417945188470185 2023-01-23 20:05:45.684626: step: 1882/1541, loss: 0.003511643735691905 2023-01-23 20:05:46.541474: step: 1884/1541, loss: 0.011584769934415817 2023-01-23 20:05:47.389292: step: 1886/1541, loss: 0.011927069164812565 2023-01-23 20:05:48.295107: step: 1888/1541, loss: 3.509744783514179e-05 2023-01-23 20:05:49.201873: step: 1890/1541, loss: 1.0841971635818481 2023-01-23 20:05:50.114435: step: 1892/1541, loss: 0.09006096422672272 2023-01-23 20:05:51.011201: step: 1894/1541, loss: 0.042794119566679 2023-01-23 20:05:51.905979: step: 1896/1541, loss: 0.0024100400041788816 2023-01-23 20:05:52.869234: step: 1898/1541, loss: 0.00184748531319201 2023-01-23 20:05:53.753871: step: 1900/1541, loss: 0.002380385994911194 2023-01-23 20:05:54.658608: step: 1902/1541, loss: 0.19410254061222076 2023-01-23 20:05:55.551663: step: 1904/1541, loss: 0.0005114302039146423 2023-01-23 20:05:56.446379: step: 1906/1541, loss: 0.001232311362400651 2023-01-23 20:05:57.275332: step: 1908/1541, loss: 0.00018993401317857206 2023-01-23 20:05:58.198734: step: 1910/1541, loss: 0.4301918148994446 2023-01-23 20:05:59.148059: step: 1912/1541, loss: 0.13701921701431274 2023-01-23 20:06:00.063613: step: 1914/1541, loss: 0.0031191816087812185 2023-01-23 20:06:00.966361: step: 1916/1541, loss: 0.0008027133299037814 2023-01-23 20:06:01.892693: step: 1918/1541, loss: 0.005992885213345289 2023-01-23 20:06:02.755947: step: 1920/1541, loss: 0.0012116028228774667 2023-01-23 20:06:03.638532: step: 1922/1541, loss: 0.0013516341568902135 2023-01-23 20:06:04.529585: step: 1924/1541, loss: 0.009511050768196583 2023-01-23 20:06:05.401156: step: 1926/1541, loss: 0.05726928636431694 2023-01-23 20:06:06.325224: step: 1928/1541, loss: 0.0018402280984446406 2023-01-23 20:06:07.181450: step: 1930/1541, loss: 0.06068737804889679 2023-01-23 20:06:08.048783: step: 1932/1541, loss: 0.00012860656715929508 2023-01-23 20:06:08.924301: step: 1934/1541, loss: 0.0554007813334465 2023-01-23 20:06:09.805785: step: 1936/1541, loss: 0.0007343717734329402 2023-01-23 20:06:10.674159: step: 1938/1541, loss: 0.00020771696290466934 2023-01-23 20:06:11.590865: step: 1940/1541, loss: 0.004086408298462629 2023-01-23 20:06:12.536942: step: 1942/1541, loss: 0.003899166826158762 2023-01-23 20:06:13.406926: step: 1944/1541, loss: 0.025817787274718285 2023-01-23 20:06:14.353991: step: 1946/1541, loss: 0.8324107527732849 2023-01-23 20:06:15.196389: step: 1948/1541, loss: 0.00010484499216545373 2023-01-23 20:06:16.069533: step: 1950/1541, loss: 0.005227556452155113 2023-01-23 20:06:16.994271: step: 1952/1541, loss: 0.0010474054142832756 2023-01-23 20:06:17.911921: step: 1954/1541, loss: 0.00024434435181319714 2023-01-23 20:06:18.817198: step: 1956/1541, loss: 0.0003312088083475828 2023-01-23 20:06:19.660610: step: 1958/1541, loss: 0.01345750130712986 2023-01-23 20:06:20.559171: step: 1960/1541, loss: 0.2399091124534607 2023-01-23 20:06:21.437174: step: 1962/1541, loss: 0.02558124251663685 2023-01-23 20:06:22.314876: step: 1964/1541, loss: 0.00022350465587805957 2023-01-23 20:06:23.194203: step: 1966/1541, loss: 0.00016925409727264196 2023-01-23 20:06:24.087961: step: 1968/1541, loss: 0.005789607763290405 2023-01-23 20:06:24.963303: step: 1970/1541, loss: 0.0001362700277240947 2023-01-23 20:06:25.837004: step: 1972/1541, loss: 0.0035781743936240673 2023-01-23 20:06:26.742090: step: 1974/1541, loss: 0.011145733296871185 2023-01-23 20:06:27.607864: step: 1976/1541, loss: 0.0007644937140867114 2023-01-23 20:06:28.487694: step: 1978/1541, loss: 0.00357167050242424 2023-01-23 20:06:29.367684: step: 1980/1541, loss: 0.07469545304775238 2023-01-23 20:06:30.239865: step: 1982/1541, loss: 0.010530912317335606 2023-01-23 20:06:31.127888: step: 1984/1541, loss: 0.0008289588149636984 2023-01-23 20:06:32.026823: step: 1986/1541, loss: 1.5297026038751937e-05 2023-01-23 20:06:32.974764: step: 1988/1541, loss: 0.3924303650856018 2023-01-23 20:06:33.882225: step: 1990/1541, loss: 0.01945655234158039 2023-01-23 20:06:34.742549: step: 1992/1541, loss: 0.00012522510951384902 2023-01-23 20:06:35.601746: step: 1994/1541, loss: 0.0003801639541052282 2023-01-23 20:06:36.475219: step: 1996/1541, loss: 0.0006152316927909851 2023-01-23 20:06:37.364182: step: 1998/1541, loss: 2.444742676743772e-05 2023-01-23 20:06:38.242991: step: 2000/1541, loss: 0.10394255816936493 2023-01-23 20:06:39.124606: step: 2002/1541, loss: 0.0016066794050857425 2023-01-23 20:06:40.006758: step: 2004/1541, loss: 6.044009205652401e-05 2023-01-23 20:06:40.890982: step: 2006/1541, loss: 0.00026342031196691096 2023-01-23 20:06:41.815523: step: 2008/1541, loss: -2.8236539947101846e-05 2023-01-23 20:06:42.732016: step: 2010/1541, loss: 0.007006209343671799 2023-01-23 20:06:43.576088: step: 2012/1541, loss: 4.4861226342618465e-05 2023-01-23 20:06:44.426178: step: 2014/1541, loss: 6.065854904591106e-05 2023-01-23 20:06:45.321803: step: 2016/1541, loss: 5.661195609718561e-05 2023-01-23 20:06:46.246004: step: 2018/1541, loss: 3.397216278244741e-05 2023-01-23 20:06:47.145159: step: 2020/1541, loss: 0.0035180822014808655 2023-01-23 20:06:48.047119: step: 2022/1541, loss: 0.0002705036895349622 2023-01-23 20:06:48.944663: step: 2024/1541, loss: 0.0005873183836229146 2023-01-23 20:06:49.835962: step: 2026/1541, loss: 0.0009784505236893892 2023-01-23 20:06:50.694583: step: 2028/1541, loss: 0.023873718455433846 2023-01-23 20:06:51.672907: step: 2030/1541, loss: 0.00036717255716212094 2023-01-23 20:06:52.556575: step: 2032/1541, loss: 0.04493589699268341 2023-01-23 20:06:53.435666: step: 2034/1541, loss: -2.989019594679121e-05 2023-01-23 20:06:54.376515: step: 2036/1541, loss: 0.22270147502422333 2023-01-23 20:06:55.304474: step: 2038/1541, loss: 0.0001330645172856748 2023-01-23 20:06:56.210152: step: 2040/1541, loss: 0.0032388821709901094 2023-01-23 20:06:57.082126: step: 2042/1541, loss: 0.62986820936203 2023-01-23 20:06:58.024436: step: 2044/1541, loss: 4.448222171049565e-05 2023-01-23 20:06:58.902190: step: 2046/1541, loss: 0.06250633299350739 2023-01-23 20:06:59.796524: step: 2048/1541, loss: 0.14573897421360016 2023-01-23 20:07:00.687217: step: 2050/1541, loss: 0.04922148212790489 2023-01-23 20:07:01.530349: step: 2052/1541, loss: 0.28862470388412476 2023-01-23 20:07:02.532795: step: 2054/1541, loss: 0.0002790137950796634 2023-01-23 20:07:03.420748: step: 2056/1541, loss: 0.010110808536410332 2023-01-23 20:07:04.287911: step: 2058/1541, loss: 0.0010796734131872654 2023-01-23 20:07:05.220010: step: 2060/1541, loss: 0.08964843302965164 2023-01-23 20:07:06.113057: step: 2062/1541, loss: 8.771172724664211e-05 2023-01-23 20:07:07.009441: step: 2064/1541, loss: 9.250095172319561e-05 2023-01-23 20:07:07.901118: step: 2066/1541, loss: 0.01540794875472784 2023-01-23 20:07:08.769798: step: 2068/1541, loss: 0.00012427949695847929 2023-01-23 20:07:09.667197: step: 2070/1541, loss: 0.15887998044490814 2023-01-23 20:07:10.589767: step: 2072/1541, loss: 0.09642999619245529 2023-01-23 20:07:11.488145: step: 2074/1541, loss: 0.05415074899792671 2023-01-23 20:07:12.405291: step: 2076/1541, loss: 0.1166563481092453 2023-01-23 20:07:13.345361: step: 2078/1541, loss: 0.052642107009887695 2023-01-23 20:07:14.255963: step: 2080/1541, loss: 0.13271954655647278 2023-01-23 20:07:15.154521: step: 2082/1541, loss: 0.6474720239639282 2023-01-23 20:07:16.051126: step: 2084/1541, loss: 0.10202149301767349 2023-01-23 20:07:16.962630: step: 2086/1541, loss: 0.00853061769157648 2023-01-23 20:07:17.864369: step: 2088/1541, loss: 0.0001696922117844224 2023-01-23 20:07:18.719540: step: 2090/1541, loss: 0.0011537248501554132 2023-01-23 20:07:19.605634: step: 2092/1541, loss: 0.07856898009777069 2023-01-23 20:07:20.478922: step: 2094/1541, loss: 0.0005314630689099431 2023-01-23 20:07:21.341682: step: 2096/1541, loss: 0.0028980665374547243 2023-01-23 20:07:22.242858: step: 2098/1541, loss: 0.010283689945936203 2023-01-23 20:07:23.119063: step: 2100/1541, loss: 0.00040815467946231365 2023-01-23 20:07:24.074437: step: 2102/1541, loss: 0.07094042003154755 2023-01-23 20:07:24.983196: step: 2104/1541, loss: 0.002449783030897379 2023-01-23 20:07:25.839580: step: 2106/1541, loss: 0.0002196411951445043 2023-01-23 20:07:26.771869: step: 2108/1541, loss: 0.03361988812685013 2023-01-23 20:07:27.675329: step: 2110/1541, loss: 0.03721889853477478 2023-01-23 20:07:28.559922: step: 2112/1541, loss: 0.031501077115535736 2023-01-23 20:07:29.458148: step: 2114/1541, loss: 0.06370873749256134 2023-01-23 20:07:30.343822: step: 2116/1541, loss: 0.01090075820684433 2023-01-23 20:07:31.262221: step: 2118/1541, loss: 0.00606408528983593 2023-01-23 20:07:32.176366: step: 2120/1541, loss: 0.006530304905027151 2023-01-23 20:07:33.028176: step: 2122/1541, loss: 0.0068243504501879215 2023-01-23 20:07:33.871675: step: 2124/1541, loss: 0.001487530767917633 2023-01-23 20:07:34.773101: step: 2126/1541, loss: 5.151798632141436e-07 2023-01-23 20:07:35.742393: step: 2128/1541, loss: 0.06016694754362106 2023-01-23 20:07:36.665339: step: 2130/1541, loss: 0.0002345020475331694 2023-01-23 20:07:37.583398: step: 2132/1541, loss: 0.06817098706960678 2023-01-23 20:07:38.464794: step: 2134/1541, loss: 0.0010155595373362303 2023-01-23 20:07:39.382689: step: 2136/1541, loss: 0.050294551998376846 2023-01-23 20:07:40.256401: step: 2138/1541, loss: 0.0030138781294226646 2023-01-23 20:07:41.154492: step: 2140/1541, loss: 0.00033167534274980426 2023-01-23 20:07:42.056520: step: 2142/1541, loss: 0.04086105152964592 2023-01-23 20:07:42.975856: step: 2144/1541, loss: 0.015366636216640472 2023-01-23 20:07:43.855944: step: 2146/1541, loss: 0.018816279247403145 2023-01-23 20:07:44.747245: step: 2148/1541, loss: 5.763522858615033e-05 2023-01-23 20:07:45.641108: step: 2150/1541, loss: 0.13972604274749756 2023-01-23 20:07:46.558616: step: 2152/1541, loss: 0.02843300811946392 2023-01-23 20:07:47.482531: step: 2154/1541, loss: 0.009601774625480175 2023-01-23 20:07:48.405406: step: 2156/1541, loss: 0.0649074912071228 2023-01-23 20:07:49.325000: step: 2158/1541, loss: 0.1322522610425949 2023-01-23 20:07:50.241542: step: 2160/1541, loss: 0.025145133957266808 2023-01-23 20:07:51.193514: step: 2162/1541, loss: 0.2972726821899414 2023-01-23 20:07:52.070832: step: 2164/1541, loss: 0.003201565472409129 2023-01-23 20:07:52.948425: step: 2166/1541, loss: 0.004857107996940613 2023-01-23 20:07:53.854265: step: 2168/1541, loss: 0.2770887613296509 2023-01-23 20:07:54.718718: step: 2170/1541, loss: 0.21463392674922943 2023-01-23 20:07:55.595636: step: 2172/1541, loss: 0.02248520962893963 2023-01-23 20:07:56.556858: step: 2174/1541, loss: 0.018354609608650208 2023-01-23 20:07:57.453150: step: 2176/1541, loss: 1.9109973907470703 2023-01-23 20:07:58.348244: step: 2178/1541, loss: 3.6954820643586572e-06 2023-01-23 20:07:59.355098: step: 2180/1541, loss: 5.540251095226267e-06 2023-01-23 20:08:00.263485: step: 2182/1541, loss: 0.00012409350892994553 2023-01-23 20:08:01.162093: step: 2184/1541, loss: 0.061501991003751755 2023-01-23 20:08:02.134246: step: 2186/1541, loss: 0.0003304584533907473 2023-01-23 20:08:03.036509: step: 2188/1541, loss: 0.09502432495355606 2023-01-23 20:08:03.928826: step: 2190/1541, loss: 0.23134943842887878 2023-01-23 20:08:04.790848: step: 2192/1541, loss: 0.03224105015397072 2023-01-23 20:08:05.701358: step: 2194/1541, loss: 0.00010859069880098104 2023-01-23 20:08:06.616399: step: 2196/1541, loss: 0.0002846546412911266 2023-01-23 20:08:07.482309: step: 2198/1541, loss: 9.524685447104275e-05 2023-01-23 20:08:08.372050: step: 2200/1541, loss: 0.15495246648788452 2023-01-23 20:08:09.275370: step: 2202/1541, loss: 0.20384946465492249 2023-01-23 20:08:10.163324: step: 2204/1541, loss: 0.0016018657479435205 2023-01-23 20:08:11.045031: step: 2206/1541, loss: 0.3338239789009094 2023-01-23 20:08:11.994707: step: 2208/1541, loss: 0.3671554923057556 2023-01-23 20:08:12.898544: step: 2210/1541, loss: 0.007630787789821625 2023-01-23 20:08:13.824123: step: 2212/1541, loss: 0.025161469355225563 2023-01-23 20:08:14.711528: step: 2214/1541, loss: 0.00029315362917259336 2023-01-23 20:08:15.623420: step: 2216/1541, loss: 0.012839209288358688 2023-01-23 20:08:16.500042: step: 2218/1541, loss: 0.025392567738890648 2023-01-23 20:08:17.421696: step: 2220/1541, loss: 0.018189702183008194 2023-01-23 20:08:18.285199: step: 2222/1541, loss: 2.7115009288536385e-05 2023-01-23 20:08:19.196193: step: 2224/1541, loss: 0.0001538748329039663 2023-01-23 20:08:20.078120: step: 2226/1541, loss: 0.0028929118998348713 2023-01-23 20:08:20.948875: step: 2228/1541, loss: 0.0023074958007782698 2023-01-23 20:08:21.815390: step: 2230/1541, loss: 0.001075970008969307 2023-01-23 20:08:22.728627: step: 2232/1541, loss: 0.11185841262340546 2023-01-23 20:08:23.618868: step: 2234/1541, loss: 0.09031414240598679 2023-01-23 20:08:24.527662: step: 2236/1541, loss: -3.541470141499303e-05 2023-01-23 20:08:25.385330: step: 2238/1541, loss: 0.013713210821151733 2023-01-23 20:08:26.290585: step: 2240/1541, loss: 0.000130259373690933 2023-01-23 20:08:27.198982: step: 2242/1541, loss: 0.00014145375462248921 2023-01-23 20:08:28.103056: step: 2244/1541, loss: 0.00013361641322262585 2023-01-23 20:08:29.047660: step: 2246/1541, loss: 0.0038303823675960302 2023-01-23 20:08:30.049916: step: 2248/1541, loss: 0.009256389923393726 2023-01-23 20:08:30.911555: step: 2250/1541, loss: 0.000563708774279803 2023-01-23 20:08:31.803745: step: 2252/1541, loss: 0.047585442662239075 2023-01-23 20:08:32.707772: step: 2254/1541, loss: 0.2970777750015259 2023-01-23 20:08:33.576871: step: 2256/1541, loss: 0.09489607065916061 2023-01-23 20:08:34.491922: step: 2258/1541, loss: 0.015705738216638565 2023-01-23 20:08:35.431397: step: 2260/1541, loss: 0.0001450321142328903 2023-01-23 20:08:36.329693: step: 2262/1541, loss: 0.000731182808522135 2023-01-23 20:08:37.387204: step: 2264/1541, loss: 0.00909408275038004 2023-01-23 20:08:38.345823: step: 2266/1541, loss: 0.0013838387094438076 2023-01-23 20:08:39.274942: step: 2268/1541, loss: 0.0031560217030346394 2023-01-23 20:08:40.202037: step: 2270/1541, loss: 0.06185700744390488 2023-01-23 20:08:41.064076: step: 2272/1541, loss: 0.023916268721222878 2023-01-23 20:08:41.921386: step: 2274/1541, loss: 0.01931118778884411 2023-01-23 20:08:42.838405: step: 2276/1541, loss: 0.17385341227054596 2023-01-23 20:08:43.715531: step: 2278/1541, loss: 0.011527741327881813 2023-01-23 20:08:44.621493: step: 2280/1541, loss: 0.0005034268251620233 2023-01-23 20:08:45.484103: step: 2282/1541, loss: 0.0020104951690882444 2023-01-23 20:08:46.383862: step: 2284/1541, loss: 0.0011679645394906402 2023-01-23 20:08:47.416808: step: 2286/1541, loss: 0.016626404598355293 2023-01-23 20:08:48.266384: step: 2288/1541, loss: 0.0011301107006147504 2023-01-23 20:08:49.244592: step: 2290/1541, loss: 0.004798569716513157 2023-01-23 20:08:50.116339: step: 2292/1541, loss: 0.009653029032051563 2023-01-23 20:08:50.979235: step: 2294/1541, loss: 0.005722849164158106 2023-01-23 20:08:51.820254: step: 2296/1541, loss: 0.0064823501743376255 2023-01-23 20:08:52.713322: step: 2298/1541, loss: 0.024199925363063812 2023-01-23 20:08:53.649105: step: 2300/1541, loss: 0.00010919585474766791 2023-01-23 20:08:54.540572: step: 2302/1541, loss: 0.15015633404254913 2023-01-23 20:08:55.445056: step: 2304/1541, loss: 0.0015580363105982542 2023-01-23 20:08:56.292987: step: 2306/1541, loss: 0.18712805211544037 2023-01-23 20:08:57.168457: step: 2308/1541, loss: 4.433419235283509e-05 2023-01-23 20:08:58.035999: step: 2310/1541, loss: 0.30292391777038574 2023-01-23 20:08:58.910569: step: 2312/1541, loss: 0.01231371983885765 2023-01-23 20:08:59.792956: step: 2314/1541, loss: 0.5790797472000122 2023-01-23 20:09:00.650467: step: 2316/1541, loss: 0.002752478700131178 2023-01-23 20:09:01.508342: step: 2318/1541, loss: 0.0003845060709863901 2023-01-23 20:09:02.417653: step: 2320/1541, loss: 6.0732818383257836e-05 2023-01-23 20:09:03.339322: step: 2322/1541, loss: 0.10318417102098465 2023-01-23 20:09:04.272546: step: 2324/1541, loss: 0.00018673115118872374 2023-01-23 20:09:05.173115: step: 2326/1541, loss: 0.00022432062542065978 2023-01-23 20:09:06.058547: step: 2328/1541, loss: 0.008971993811428547 2023-01-23 20:09:06.991147: step: 2330/1541, loss: 0.1723683625459671 2023-01-23 20:09:07.858324: step: 2332/1541, loss: 0.00018158341117668897 2023-01-23 20:09:08.725083: step: 2334/1541, loss: 0.007307000923901796 2023-01-23 20:09:09.658445: step: 2336/1541, loss: 0.011564862914383411 2023-01-23 20:09:10.569203: step: 2338/1541, loss: 0.0004734384419862181 2023-01-23 20:09:11.512683: step: 2340/1541, loss: 0.00048045176663435996 2023-01-23 20:09:12.400119: step: 2342/1541, loss: 0.008575187996029854 2023-01-23 20:09:13.312605: step: 2344/1541, loss: 0.0413195863366127 2023-01-23 20:09:14.270445: step: 2346/1541, loss: 0.1417083442211151 2023-01-23 20:09:15.221428: step: 2348/1541, loss: 0.3275150954723358 2023-01-23 20:09:16.118868: step: 2350/1541, loss: 0.0023805187083780766 2023-01-23 20:09:16.999878: step: 2352/1541, loss: 0.03264910355210304 2023-01-23 20:09:17.893094: step: 2354/1541, loss: 0.019180037081241608 2023-01-23 20:09:18.765573: step: 2356/1541, loss: 0.08818043023347855 2023-01-23 20:09:19.658821: step: 2358/1541, loss: 0.001332852873019874 2023-01-23 20:09:20.561944: step: 2360/1541, loss: 0.23623643815517426 2023-01-23 20:09:21.487088: step: 2362/1541, loss: 0.020962141454219818 2023-01-23 20:09:22.481107: step: 2364/1541, loss: 0.049283694475889206 2023-01-23 20:09:23.432838: step: 2366/1541, loss: 0.00020106867305003107 2023-01-23 20:09:24.425076: step: 2368/1541, loss: 0.15260183811187744 2023-01-23 20:09:25.296003: step: 2370/1541, loss: 0.01845197007060051 2023-01-23 20:09:26.181772: step: 2372/1541, loss: 0.016680995002388954 2023-01-23 20:09:27.108189: step: 2374/1541, loss: 0.005712907295674086 2023-01-23 20:09:28.069622: step: 2376/1541, loss: 0.5352446436882019 2023-01-23 20:09:28.976900: step: 2378/1541, loss: 0.002447013510391116 2023-01-23 20:09:29.906248: step: 2380/1541, loss: 0.004806159064173698 2023-01-23 20:09:30.851287: step: 2382/1541, loss: 0.025371719151735306 2023-01-23 20:09:31.734389: step: 2384/1541, loss: 0.01423491071909666 2023-01-23 20:09:32.672679: step: 2386/1541, loss: 0.19723299145698547 2023-01-23 20:09:33.641886: step: 2388/1541, loss: 0.10947418212890625 2023-01-23 20:09:34.592742: step: 2390/1541, loss: 0.0016291356878355145 2023-01-23 20:09:35.511101: step: 2392/1541, loss: 0.04981265962123871 2023-01-23 20:09:36.409328: step: 2394/1541, loss: 0.003580392338335514 2023-01-23 20:09:37.364864: step: 2396/1541, loss: 0.005289362743496895 2023-01-23 20:09:38.268215: step: 2398/1541, loss: 0.0006320218089967966 2023-01-23 20:09:39.142868: step: 2400/1541, loss: 0.0029015582986176014 2023-01-23 20:09:40.006553: step: 2402/1541, loss: 0.0761084258556366 2023-01-23 20:09:40.876854: step: 2404/1541, loss: 0.008336476050317287 2023-01-23 20:09:41.790978: step: 2406/1541, loss: 0.7911558747291565 2023-01-23 20:09:42.778659: step: 2408/1541, loss: 0.007912946864962578 2023-01-23 20:09:43.645210: step: 2410/1541, loss: 0.0637885332107544 2023-01-23 20:09:44.616313: step: 2412/1541, loss: 0.0006974138086661696 2023-01-23 20:09:45.539634: step: 2414/1541, loss: 0.018373984843492508 2023-01-23 20:09:46.420988: step: 2416/1541, loss: 0.0065536354668438435 2023-01-23 20:09:47.312194: step: 2418/1541, loss: 0.0003611061256378889 2023-01-23 20:09:48.220842: step: 2420/1541, loss: 0.0027408155146986246 2023-01-23 20:09:49.138775: step: 2422/1541, loss: 0.0007016644813120365 2023-01-23 20:09:50.095883: step: 2424/1541, loss: 0.04564021900296211 2023-01-23 20:09:51.015222: step: 2426/1541, loss: 0.0048529920168221 2023-01-23 20:09:51.930003: step: 2428/1541, loss: 0.004105532541871071 2023-01-23 20:09:52.847027: step: 2430/1541, loss: 0.011664851568639278 2023-01-23 20:09:53.726377: step: 2432/1541, loss: 7.9744728282094e-05 2023-01-23 20:09:54.604245: step: 2434/1541, loss: 0.0005490723415277898 2023-01-23 20:09:55.505736: step: 2436/1541, loss: 0.011576232500374317 2023-01-23 20:09:56.407633: step: 2438/1541, loss: 0.0007741290610283613 2023-01-23 20:09:57.273184: step: 2440/1541, loss: 0.0016818201402202249 2023-01-23 20:09:58.120357: step: 2442/1541, loss: 0.0009761427063494921 2023-01-23 20:09:59.009473: step: 2444/1541, loss: 0.00010551593004493043 2023-01-23 20:09:59.937458: step: 2446/1541, loss: 0.004371508955955505 2023-01-23 20:10:00.792386: step: 2448/1541, loss: 0.050043974071741104 2023-01-23 20:10:01.706085: step: 2450/1541, loss: 0.0005623122560791671 2023-01-23 20:10:02.698140: step: 2452/1541, loss: 0.015591112896800041 2023-01-23 20:10:03.566218: step: 2454/1541, loss: 0.030586671084165573 2023-01-23 20:10:04.481671: step: 2456/1541, loss: 0.002868247451260686 2023-01-23 20:10:05.408548: step: 2458/1541, loss: 0.006839386653155088 2023-01-23 20:10:06.308237: step: 2460/1541, loss: 0.0007442277856171131 2023-01-23 20:10:07.198705: step: 2462/1541, loss: 0.39793267846107483 2023-01-23 20:10:08.098996: step: 2464/1541, loss: 0.00227122800424695 2023-01-23 20:10:08.955633: step: 2466/1541, loss: 0.00048652419354766607 2023-01-23 20:10:09.859527: step: 2468/1541, loss: 0.00014005495177116245 2023-01-23 20:10:10.734526: step: 2470/1541, loss: 0.0009313629707321525 2023-01-23 20:10:11.612867: step: 2472/1541, loss: 0.01000697910785675 2023-01-23 20:10:12.521914: step: 2474/1541, loss: 0.0014089687028899789 2023-01-23 20:10:13.425374: step: 2476/1541, loss: 0.003135352861136198 2023-01-23 20:10:14.294671: step: 2478/1541, loss: 0.0849263072013855 2023-01-23 20:10:15.238659: step: 2480/1541, loss: 0.0007315604016184807 2023-01-23 20:10:16.109484: step: 2482/1541, loss: 0.0007322581368498504 2023-01-23 20:10:17.006368: step: 2484/1541, loss: 5.73253637412563e-05 2023-01-23 20:10:17.887112: step: 2486/1541, loss: -2.130598295480013e-05 2023-01-23 20:10:18.741462: step: 2488/1541, loss: 0.00037847599014639854 2023-01-23 20:10:19.645982: step: 2490/1541, loss: 0.0005919431569054723 2023-01-23 20:10:20.538961: step: 2492/1541, loss: 0.003683879505842924 2023-01-23 20:10:21.434734: step: 2494/1541, loss: 0.07788904011249542 2023-01-23 20:10:22.331895: step: 2496/1541, loss: 0.00013905168452765793 2023-01-23 20:10:23.256752: step: 2498/1541, loss: 0.020984331145882607 2023-01-23 20:10:24.149931: step: 2500/1541, loss: 0.00017459149239584804 2023-01-23 20:10:25.042123: step: 2502/1541, loss: 0.009162267670035362 2023-01-23 20:10:25.908656: step: 2504/1541, loss: 0.04566648229956627 2023-01-23 20:10:26.797907: step: 2506/1541, loss: 0.2218938171863556 2023-01-23 20:10:27.731152: step: 2508/1541, loss: 0.0033212658017873764 2023-01-23 20:10:28.615433: step: 2510/1541, loss: 0.11781833320856094 2023-01-23 20:10:29.599434: step: 2512/1541, loss: 0.0002032980410149321 2023-01-23 20:10:30.491872: step: 2514/1541, loss: 0.012908735312521458 2023-01-23 20:10:31.365472: step: 2516/1541, loss: 0.003251298563554883 2023-01-23 20:10:32.282203: step: 2518/1541, loss: 0.006817059591412544 2023-01-23 20:10:33.184501: step: 2520/1541, loss: 0.02256987988948822 2023-01-23 20:10:34.005645: step: 2522/1541, loss: 0.0007516462937928736 2023-01-23 20:10:34.903222: step: 2524/1541, loss: 0.02671956457197666 2023-01-23 20:10:35.833063: step: 2526/1541, loss: 8.407505811192095e-05 2023-01-23 20:10:36.724823: step: 2528/1541, loss: 0.38067713379859924 2023-01-23 20:10:37.672616: step: 2530/1541, loss: 4.0044284105533734e-05 2023-01-23 20:10:38.618733: step: 2532/1541, loss: 0.0048242188058793545 2023-01-23 20:10:39.529653: step: 2534/1541, loss: 0.00040626339614391327 2023-01-23 20:10:40.455079: step: 2536/1541, loss: 0.8441546559333801 2023-01-23 20:10:41.352079: step: 2538/1541, loss: 0.0031220950186252594 2023-01-23 20:10:42.242073: step: 2540/1541, loss: 0.020289799198508263 2023-01-23 20:10:43.144100: step: 2542/1541, loss: 0.26045212149620056 2023-01-23 20:10:44.074035: step: 2544/1541, loss: 0.0007291844231076539 2023-01-23 20:10:44.975912: step: 2546/1541, loss: 0.003662725677713752 2023-01-23 20:10:45.892060: step: 2548/1541, loss: 0.574233889579773 2023-01-23 20:10:46.774978: step: 2550/1541, loss: 0.01791350357234478 2023-01-23 20:10:47.625185: step: 2552/1541, loss: 5.349403727450408e-05 2023-01-23 20:10:48.565732: step: 2554/1541, loss: 0.00085863983258605 2023-01-23 20:10:49.428983: step: 2556/1541, loss: 1.0169780254364014 2023-01-23 20:10:50.301143: step: 2558/1541, loss: 0.0011864821426570415 2023-01-23 20:10:51.155322: step: 2560/1541, loss: 0.0001373815321130678 2023-01-23 20:10:52.036307: step: 2562/1541, loss: 0.0033367937430739403 2023-01-23 20:10:52.912006: step: 2564/1541, loss: 0.0046937051229178905 2023-01-23 20:10:53.770166: step: 2566/1541, loss: 0.007885044440627098 2023-01-23 20:10:54.687506: step: 2568/1541, loss: 0.00754478620365262 2023-01-23 20:10:55.549403: step: 2570/1541, loss: 0.0001324547192780301 2023-01-23 20:10:56.474623: step: 2572/1541, loss: 0.010557024739682674 2023-01-23 20:10:57.397080: step: 2574/1541, loss: 0.003840178716927767 2023-01-23 20:10:58.278325: step: 2576/1541, loss: 5.7005847338587046e-05 2023-01-23 20:10:59.191186: step: 2578/1541, loss: 0.0035661030560731888 2023-01-23 20:11:00.176334: step: 2580/1541, loss: 4.2253337596775964e-05 2023-01-23 20:11:01.067821: step: 2582/1541, loss: 0.0013916692696511745 2023-01-23 20:11:01.961329: step: 2584/1541, loss: 0.0025875000283122063 2023-01-23 20:11:02.815008: step: 2586/1541, loss: 0.0051795789040625095 2023-01-23 20:11:03.694144: step: 2588/1541, loss: 0.00019387448264751583 2023-01-23 20:11:04.600167: step: 2590/1541, loss: -3.974169885623269e-06 2023-01-23 20:11:05.488206: step: 2592/1541, loss: 0.0028251761104911566 2023-01-23 20:11:06.370118: step: 2594/1541, loss: 4.079617792740464e-05 2023-01-23 20:11:07.237405: step: 2596/1541, loss: 0.10499913990497589 2023-01-23 20:11:08.105315: step: 2598/1541, loss: 0.0011319696204736829 2023-01-23 20:11:09.004793: step: 2600/1541, loss: 0.006038092076778412 2023-01-23 20:11:09.862301: step: 2602/1541, loss: 3.19236314680893e-05 2023-01-23 20:11:10.751432: step: 2604/1541, loss: 0.0004707840271294117 2023-01-23 20:11:11.639018: step: 2606/1541, loss: 0.11171968281269073 2023-01-23 20:11:12.536446: step: 2608/1541, loss: 0.011639698408544064 2023-01-23 20:11:13.441906: step: 2610/1541, loss: 6.829958874732256e-05 2023-01-23 20:11:14.359548: step: 2612/1541, loss: 0.013750478625297546 2023-01-23 20:11:15.224524: step: 2614/1541, loss: 0.06486613303422928 2023-01-23 20:11:16.128161: step: 2616/1541, loss: 0.05153707414865494 2023-01-23 20:11:17.003840: step: 2618/1541, loss: 0.00039440530235879123 2023-01-23 20:11:17.883073: step: 2620/1541, loss: 0.0020364467054605484 2023-01-23 20:11:18.805536: step: 2622/1541, loss: 0.00739673338830471 2023-01-23 20:11:19.740409: step: 2624/1541, loss: 0.0024135736748576164 2023-01-23 20:11:20.702397: step: 2626/1541, loss: 0.0004935486358590424 2023-01-23 20:11:21.638557: step: 2628/1541, loss: 0.10379484295845032 2023-01-23 20:11:22.526475: step: 2630/1541, loss: 0.003779325168579817 2023-01-23 20:11:23.441824: step: 2632/1541, loss: 0.13151304423809052 2023-01-23 20:11:24.414920: step: 2634/1541, loss: 0.014904113486409187 2023-01-23 20:11:25.340984: step: 2636/1541, loss: 0.02181772142648697 2023-01-23 20:11:26.212601: step: 2638/1541, loss: 0.0010695521486923099 2023-01-23 20:11:27.121432: step: 2640/1541, loss: 0.0009547041263431311 2023-01-23 20:11:27.963270: step: 2642/1541, loss: 0.011938750743865967 2023-01-23 20:11:28.920706: step: 2644/1541, loss: 0.12556755542755127 2023-01-23 20:11:29.821897: step: 2646/1541, loss: 0.08893915265798569 2023-01-23 20:11:30.693123: step: 2648/1541, loss: 0.0001620937546249479 2023-01-23 20:11:31.548608: step: 2650/1541, loss: 0.001424242276698351 2023-01-23 20:11:32.404491: step: 2652/1541, loss: 0.05714583769440651 2023-01-23 20:11:33.296957: step: 2654/1541, loss: 1.7377096810378134e-05 2023-01-23 20:11:34.155111: step: 2656/1541, loss: 0.22063694894313812 2023-01-23 20:11:35.033266: step: 2658/1541, loss: 7.934695895528421e-05 2023-01-23 20:11:35.918849: step: 2660/1541, loss: 0.00038071078597567976 2023-01-23 20:11:36.807493: step: 2662/1541, loss: 4.8719844926381484e-05 2023-01-23 20:11:37.671600: step: 2664/1541, loss: 0.0370454341173172 2023-01-23 20:11:38.533548: step: 2666/1541, loss: 0.00046362480497919023 2023-01-23 20:11:39.398369: step: 2668/1541, loss: 0.007111223880201578 2023-01-23 20:11:40.234796: step: 2670/1541, loss: 0.21163290739059448 2023-01-23 20:11:41.135204: step: 2672/1541, loss: 0.21361123025417328 2023-01-23 20:11:42.048017: step: 2674/1541, loss: 0.0037391777150332928 2023-01-23 20:11:42.962841: step: 2676/1541, loss: 0.004674505442380905 2023-01-23 20:11:43.848355: step: 2678/1541, loss: 0.0008004853734746575 2023-01-23 20:11:44.741712: step: 2680/1541, loss: 0.0014753632713109255 2023-01-23 20:11:45.650810: step: 2682/1541, loss: 0.0008200713200494647 2023-01-23 20:11:46.529100: step: 2684/1541, loss: 0.001445608213543892 2023-01-23 20:11:47.479806: step: 2686/1541, loss: 0.08045701682567596 2023-01-23 20:11:48.338938: step: 2688/1541, loss: 0.0018113708356395364 2023-01-23 20:11:49.231415: step: 2690/1541, loss: 0.00020316203881520778 2023-01-23 20:11:50.100268: step: 2692/1541, loss: 0.00013132170715834945 2023-01-23 20:11:50.953677: step: 2694/1541, loss: 0.010522847063839436 2023-01-23 20:11:51.840942: step: 2696/1541, loss: 0.0005243256455287337 2023-01-23 20:11:52.728464: step: 2698/1541, loss: 2.2234704374568537e-05 2023-01-23 20:11:53.637200: step: 2700/1541, loss: 2.765031240414828e-05 2023-01-23 20:11:54.536341: step: 2702/1541, loss: 0.0005414055194705725 2023-01-23 20:11:55.417570: step: 2704/1541, loss: 0.008891659788787365 2023-01-23 20:11:56.325425: step: 2706/1541, loss: 2.49508648266783e-05 2023-01-23 20:11:57.228605: step: 2708/1541, loss: 0.00024663418298587203 2023-01-23 20:11:58.117417: step: 2710/1541, loss: 0.0003344565338920802 2023-01-23 20:11:59.009528: step: 2712/1541, loss: 0.00039329336141236126 2023-01-23 20:11:59.904555: step: 2714/1541, loss: 0.00010221442789770663 2023-01-23 20:12:00.766655: step: 2716/1541, loss: 3.247612767154351e-05 2023-01-23 20:12:01.682008: step: 2718/1541, loss: 0.0002692263515200466 2023-01-23 20:12:02.608588: step: 2720/1541, loss: 0.004668904934078455 2023-01-23 20:12:03.485170: step: 2722/1541, loss: 0.001707356539554894 2023-01-23 20:12:04.366387: step: 2724/1541, loss: 0.03136691078543663 2023-01-23 20:12:05.230107: step: 2726/1541, loss: 0.001362113980576396 2023-01-23 20:12:06.098943: step: 2728/1541, loss: 0.059889089316129684 2023-01-23 20:12:07.008465: step: 2730/1541, loss: 0.000976672861725092 2023-01-23 20:12:07.890106: step: 2732/1541, loss: 0.004775463603436947 2023-01-23 20:12:08.828217: step: 2734/1541, loss: 0.2402850091457367 2023-01-23 20:12:09.696580: step: 2736/1541, loss: 0.0004898374900221825 2023-01-23 20:12:10.581380: step: 2738/1541, loss: 0.0005835401825606823 2023-01-23 20:12:11.427542: step: 2740/1541, loss: 0.0010561122326180339 2023-01-23 20:12:12.300275: step: 2742/1541, loss: 0.22244958579540253 2023-01-23 20:12:13.189936: step: 2744/1541, loss: 0.0002177975547965616 2023-01-23 20:12:14.076938: step: 2746/1541, loss: 0.6300485730171204 2023-01-23 20:12:14.971611: step: 2748/1541, loss: 0.030246270820498466 2023-01-23 20:12:15.863584: step: 2750/1541, loss: 0.010597849264740944 2023-01-23 20:12:16.776937: step: 2752/1541, loss: 0.00024084278265945613 2023-01-23 20:12:17.669862: step: 2754/1541, loss: 6.341146945487708e-05 2023-01-23 20:12:18.519599: step: 2756/1541, loss: 0.007889439351856709 2023-01-23 20:12:19.426683: step: 2758/1541, loss: 0.014948228374123573 2023-01-23 20:12:20.330464: step: 2760/1541, loss: 0.006496722344309092 2023-01-23 20:12:21.201330: step: 2762/1541, loss: 0.0026687798090279102 2023-01-23 20:12:22.077971: step: 2764/1541, loss: 0.030072810128331184 2023-01-23 20:12:22.996615: step: 2766/1541, loss: 0.035603806376457214 2023-01-23 20:12:23.875717: step: 2768/1541, loss: 0.0004939929349347949 2023-01-23 20:12:24.768359: step: 2770/1541, loss: 6.360199040500447e-05 2023-01-23 20:12:25.646997: step: 2772/1541, loss: 2.9573366191470996e-05 2023-01-23 20:12:26.544062: step: 2774/1541, loss: 0.12642434239387512 2023-01-23 20:12:27.473290: step: 2776/1541, loss: 4.570396413328126e-05 2023-01-23 20:12:28.365163: step: 2778/1541, loss: 0.0003715443308465183 2023-01-23 20:12:29.258460: step: 2780/1541, loss: 0.2236720770597458 2023-01-23 20:12:30.155217: step: 2782/1541, loss: 0.231027752161026 2023-01-23 20:12:31.098928: step: 2784/1541, loss: 0.17646439373493195 2023-01-23 20:12:32.002022: step: 2786/1541, loss: 0.04715577885508537 2023-01-23 20:12:32.883275: step: 2788/1541, loss: 0.47168901562690735 2023-01-23 20:12:33.732879: step: 2790/1541, loss: 0.00010080239007947966 2023-01-23 20:12:34.584472: step: 2792/1541, loss: 0.03363121673464775 2023-01-23 20:12:35.494012: step: 2794/1541, loss: 0.03951621800661087 2023-01-23 20:12:36.371588: step: 2796/1541, loss: 0.01896614022552967 2023-01-23 20:12:37.307332: step: 2798/1541, loss: 0.0008995337411761284 2023-01-23 20:12:38.190606: step: 2800/1541, loss: 0.0009411562932655215 2023-01-23 20:12:39.082183: step: 2802/1541, loss: 0.004955223295837641 2023-01-23 20:12:39.973431: step: 2804/1541, loss: 0.06387781351804733 2023-01-23 20:12:40.833277: step: 2806/1541, loss: 0.0011862525716423988 2023-01-23 20:12:41.722195: step: 2808/1541, loss: 0.07299686223268509 2023-01-23 20:12:42.614269: step: 2810/1541, loss: 0.00039741338696330786 2023-01-23 20:12:43.545609: step: 2812/1541, loss: 0.3008910119533539 2023-01-23 20:12:44.453686: step: 2814/1541, loss: 0.030977707356214523 2023-01-23 20:12:45.313875: step: 2816/1541, loss: 0.07392959296703339 2023-01-23 20:12:46.247913: step: 2818/1541, loss: 0.001465491601265967 2023-01-23 20:12:47.141156: step: 2820/1541, loss: 0.00016551838780287653 2023-01-23 20:12:48.025584: step: 2822/1541, loss: 0.11439280956983566 2023-01-23 20:12:48.899765: step: 2824/1541, loss: 0.000916414603125304 2023-01-23 20:12:49.810964: step: 2826/1541, loss: 0.02506275661289692 2023-01-23 20:12:50.755230: step: 2828/1541, loss: 0.4991838037967682 2023-01-23 20:12:51.687981: step: 2830/1541, loss: 0.02469056285917759 2023-01-23 20:12:52.574398: step: 2832/1541, loss: 0.006960256490856409 2023-01-23 20:12:53.517234: step: 2834/1541, loss: 0.0026193817611783743 2023-01-23 20:12:54.399285: step: 2836/1541, loss: 0.002434896305203438 2023-01-23 20:12:55.287201: step: 2838/1541, loss: 0.003831598674878478 2023-01-23 20:12:56.185083: step: 2840/1541, loss: 0.0030081807635724545 2023-01-23 20:12:57.077126: step: 2842/1541, loss: 0.0003794811200350523 2023-01-23 20:12:57.989277: step: 2844/1541, loss: 0.16477569937705994 2023-01-23 20:12:58.913049: step: 2846/1541, loss: 0.00271418085321784 2023-01-23 20:12:59.795724: step: 2848/1541, loss: 8.66765039972961e-05 2023-01-23 20:13:00.630300: step: 2850/1541, loss: 0.0011198431020602584 2023-01-23 20:13:01.485696: step: 2852/1541, loss: 0.0009458669228479266 2023-01-23 20:13:02.416362: step: 2854/1541, loss: 0.0057974583469331264 2023-01-23 20:13:03.306660: step: 2856/1541, loss: 0.0030153528787195683 2023-01-23 20:13:04.211468: step: 2858/1541, loss: 0.0005295865121297538 2023-01-23 20:13:05.091755: step: 2860/1541, loss: 0.006076700519770384 2023-01-23 20:13:06.000274: step: 2862/1541, loss: 0.20676375925540924 2023-01-23 20:13:06.878100: step: 2864/1541, loss: 0.34764236211776733 2023-01-23 20:13:07.783183: step: 2866/1541, loss: 0.028596391901373863 2023-01-23 20:13:08.718149: step: 2868/1541, loss: 0.0007899164920672774 2023-01-23 20:13:09.594402: step: 2870/1541, loss: 4.092938615940511e-05 2023-01-23 20:13:10.483265: step: 2872/1541, loss: 0.0037257138174027205 2023-01-23 20:13:11.415428: step: 2874/1541, loss: 0.9199328422546387 2023-01-23 20:13:12.293154: step: 2876/1541, loss: 0.0002275816659675911 2023-01-23 20:13:13.258744: step: 2878/1541, loss: 0.02541792020201683 2023-01-23 20:13:14.228358: step: 2880/1541, loss: 0.037275172770023346 2023-01-23 20:13:15.147252: step: 2882/1541, loss: 0.07740180939435959 2023-01-23 20:13:16.038901: step: 2884/1541, loss: 0.28824907541275024 2023-01-23 20:13:16.905802: step: 2886/1541, loss: 0.0005371100851334631 2023-01-23 20:13:17.866443: step: 2888/1541, loss: 0.021871820092201233 2023-01-23 20:13:18.732158: step: 2890/1541, loss: 0.31491559743881226 2023-01-23 20:13:19.602720: step: 2892/1541, loss: 0.0006499089067801833 2023-01-23 20:13:20.482077: step: 2894/1541, loss: 0.0013710586354136467 2023-01-23 20:13:21.427569: step: 2896/1541, loss: 0.0015158116584643722 2023-01-23 20:13:22.309207: step: 2898/1541, loss: 0.0016173231415450573 2023-01-23 20:13:23.187802: step: 2900/1541, loss: 0.005965544376522303 2023-01-23 20:13:24.089032: step: 2902/1541, loss: 0.3141747713088989 2023-01-23 20:13:25.104773: step: 2904/1541, loss: 0.00013832715922035277 2023-01-23 20:13:26.012961: step: 2906/1541, loss: 0.06440665572881699 2023-01-23 20:13:26.884365: step: 2908/1541, loss: 0.00045421786489896476 2023-01-23 20:13:27.737015: step: 2910/1541, loss: 0.00018062899471260607 2023-01-23 20:13:28.683247: step: 2912/1541, loss: 0.04193718731403351 2023-01-23 20:13:29.567958: step: 2914/1541, loss: 0.14075711369514465 2023-01-23 20:13:30.421078: step: 2916/1541, loss: 0.03662041947245598 2023-01-23 20:13:31.313286: step: 2918/1541, loss: 0.0006731445901095867 2023-01-23 20:13:32.211890: step: 2920/1541, loss: 0.000353671086486429 2023-01-23 20:13:33.112986: step: 2922/1541, loss: 0.0030126236379146576 2023-01-23 20:13:34.113788: step: 2924/1541, loss: 0.019661854952573776 2023-01-23 20:13:34.987712: step: 2926/1541, loss: 0.00781825091689825 2023-01-23 20:13:35.854450: step: 2928/1541, loss: 0.0029405015520751476 2023-01-23 20:13:36.736573: step: 2930/1541, loss: 0.000849194242618978 2023-01-23 20:13:37.730590: step: 2932/1541, loss: 0.005340857431292534 2023-01-23 20:13:38.654178: step: 2934/1541, loss: 0.0009716870845295489 2023-01-23 20:13:39.534921: step: 2936/1541, loss: 0.0022853657137602568 2023-01-23 20:13:40.485584: step: 2938/1541, loss: 0.11257592588663101 2023-01-23 20:13:41.369030: step: 2940/1541, loss: 0.10058094561100006 2023-01-23 20:13:42.247079: step: 2942/1541, loss: 0.014469006098806858 2023-01-23 20:13:43.194669: step: 2944/1541, loss: 0.0003264163970015943 2023-01-23 20:13:44.117491: step: 2946/1541, loss: 0.011816898360848427 2023-01-23 20:13:45.054964: step: 2948/1541, loss: 0.04899107664823532 2023-01-23 20:13:45.947192: step: 2950/1541, loss: 0.00021518586436286569 2023-01-23 20:13:46.835448: step: 2952/1541, loss: 0.08138445019721985 2023-01-23 20:13:47.707974: step: 2954/1541, loss: 0.0008579050190746784 2023-01-23 20:13:48.570904: step: 2956/1541, loss: 0.003776203840970993 2023-01-23 20:13:49.499051: step: 2958/1541, loss: 0.00019413052359595895 2023-01-23 20:13:50.356242: step: 2960/1541, loss: 0.0004892595461569726 2023-01-23 20:13:51.222356: step: 2962/1541, loss: 0.010825357399880886 2023-01-23 20:13:52.134475: step: 2964/1541, loss: 0.09736035019159317 2023-01-23 20:13:53.038187: step: 2966/1541, loss: 0.002526924479752779 2023-01-23 20:13:53.939945: step: 2968/1541, loss: 0.002970131579786539 2023-01-23 20:13:54.815213: step: 2970/1541, loss: 0.0003117485321126878 2023-01-23 20:13:55.726370: step: 2972/1541, loss: 0.0013134093023836613 2023-01-23 20:13:56.622171: step: 2974/1541, loss: 0.00040347379399463534 2023-01-23 20:13:57.528425: step: 2976/1541, loss: 0.004207713529467583 2023-01-23 20:13:58.391274: step: 2978/1541, loss: 0.007573672570288181 2023-01-23 20:13:59.262982: step: 2980/1541, loss: 0.0001041141731548123 2023-01-23 20:14:00.143373: step: 2982/1541, loss: 0.00031921712798066437 2023-01-23 20:14:01.061368: step: 2984/1541, loss: 0.005401455331593752 2023-01-23 20:14:01.962262: step: 2986/1541, loss: 0.1718316376209259 2023-01-23 20:14:02.867463: step: 2988/1541, loss: 0.0009210953139699996 2023-01-23 20:14:03.783314: step: 2990/1541, loss: 0.06844060868024826 2023-01-23 20:14:04.686000: step: 2992/1541, loss: 0.037628911435604095 2023-01-23 20:14:05.554590: step: 2994/1541, loss: 0.003122338093817234 2023-01-23 20:14:06.467636: step: 2996/1541, loss: 9.070707892533392e-05 2023-01-23 20:14:07.354538: step: 2998/1541, loss: 0.0011162132723256946 2023-01-23 20:14:08.228076: step: 3000/1541, loss: 0.005808340851217508 2023-01-23 20:14:09.131497: step: 3002/1541, loss: 0.02885391376912594 2023-01-23 20:14:10.047860: step: 3004/1541, loss: 0.002283615991473198 2023-01-23 20:14:10.947120: step: 3006/1541, loss: 0.10140477120876312 2023-01-23 20:14:11.836059: step: 3008/1541, loss: 0.00027573166880756617 2023-01-23 20:14:12.725310: step: 3010/1541, loss: 0.00023820689239073545 2023-01-23 20:14:13.611983: step: 3012/1541, loss: 0.001212224131450057 2023-01-23 20:14:14.521341: step: 3014/1541, loss: 0.1692071557044983 2023-01-23 20:14:15.423935: step: 3016/1541, loss: 0.10370637476444244 2023-01-23 20:14:16.333046: step: 3018/1541, loss: 0.14911924302577972 2023-01-23 20:14:17.205129: step: 3020/1541, loss: 0.00015230546705424786 2023-01-23 20:14:18.075807: step: 3022/1541, loss: 0.0036408344749361277 2023-01-23 20:14:18.967936: step: 3024/1541, loss: 0.0016973880119621754 2023-01-23 20:14:19.903736: step: 3026/1541, loss: 0.013313649222254753 2023-01-23 20:14:20.858725: step: 3028/1541, loss: 0.2698816955089569 2023-01-23 20:14:21.759008: step: 3030/1541, loss: 1.0476466741238255e-05 2023-01-23 20:14:22.617129: step: 3032/1541, loss: 0.0755164846777916 2023-01-23 20:14:23.491847: step: 3034/1541, loss: -2.1299030777299777e-05 2023-01-23 20:14:24.399086: step: 3036/1541, loss: 0.012014677748084068 2023-01-23 20:14:25.305416: step: 3038/1541, loss: 0.018606500700116158 2023-01-23 20:14:26.262409: step: 3040/1541, loss: 0.00020067888544872403 2023-01-23 20:14:27.156121: step: 3042/1541, loss: 6.968521120143123e-06 2023-01-23 20:14:28.010521: step: 3044/1541, loss: 0.0005730970297008753 2023-01-23 20:14:28.890996: step: 3046/1541, loss: 0.004149948246777058 2023-01-23 20:14:29.784796: step: 3048/1541, loss: 0.4821852147579193 2023-01-23 20:14:30.651899: step: 3050/1541, loss: -1.3696965652343351e-05 2023-01-23 20:14:31.506649: step: 3052/1541, loss: 4.7398229071404785e-05 2023-01-23 20:14:32.374620: step: 3054/1541, loss: 0.0006770469481125474 2023-01-23 20:14:33.295039: step: 3056/1541, loss: 0.00028530677082017064 2023-01-23 20:14:34.181745: step: 3058/1541, loss: 0.0006530247628688812 2023-01-23 20:14:35.064970: step: 3060/1541, loss: 0.25938326120376587 2023-01-23 20:14:35.982745: step: 3062/1541, loss: 0.230860635638237 2023-01-23 20:14:36.865345: step: 3064/1541, loss: 0.3084745705127716 2023-01-23 20:14:37.727895: step: 3066/1541, loss: 0.0004817779699806124 2023-01-23 20:14:38.598614: step: 3068/1541, loss: 0.5968751311302185 2023-01-23 20:14:39.496507: step: 3070/1541, loss: 6.299078086158261e-05 2023-01-23 20:14:40.396982: step: 3072/1541, loss: 0.06956083327531815 2023-01-23 20:14:41.297258: step: 3074/1541, loss: 0.05328879877924919 2023-01-23 20:14:42.165059: step: 3076/1541, loss: 0.00011631956294877455 2023-01-23 20:14:43.085125: step: 3078/1541, loss: 0.019345231354236603 2023-01-23 20:14:43.946193: step: 3080/1541, loss: 0.0026929136365652084 2023-01-23 20:14:44.841455: step: 3082/1541, loss: 0.0029203947633504868 2023-01-23 20:14:45.557495: step: 3084/1541, loss: 3.368854595464654e-05 ================================================== Loss: 0.065 -------------------- Dev Chinese: {'event': {'p': 0.6284348864994026, 'r': 0.7003994673768309, 'f1': 0.6624685138539043}, 'argument': {'p': 0.6619326401514553, 'r': 0.6567814133798486, 'f1': 0.6593469657758637}, 'combined': 0.4367966045316175, 'epoch': 31} Test Chinese: {'event': {'p': 0.6056782334384858, 'r': 0.6902336728579987, 'f1': 0.6451974236908429}, 'argument': {'p': 0.7146322356778463, 'r': 0.7062952938106603, 'f1': 0.7104393072690287}, 'combined': 0.45837361073868443, 'epoch': 31} Dev Korean: {'event': {'p': 0.5686274509803921, 'r': 0.7723035952063915, 'f1': 0.6549971767363071}, 'argument': {'p': 0.6473164976644228, 'r': 0.6626403458874687, 'f1': 0.6548887927452189}, 'combined': 0.4289503103243669, 'epoch': 31} Test Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.7099116610679151, 'r': 0.7181201701131994, 'f1': 0.7139923238122525}, 'combined': 0.48294463527306064, 'epoch': 31} Dev Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6602125580861498, 'r': 0.6656811862028272, 'f1': 0.6629355945087024}, 'combined': 0.45719696173013963, 'epoch': 31} Test Russian: {'event': {'p': 0.6467817896389325, 'r': 0.7405632115038946, 'f1': 0.6905027932960894}, 'argument': {'p': 0.7085942728862843, 'r': 0.7139277781660736, 'f1': 0.7112510269892324}, 'combined': 0.49112082087077724, 'epoch': 31} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6657224418093983, 'r': 0.6775051398945203, 'f1': 0.671562112351586}, 'combined': 0.49362685181398624, 'epoch': 31} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6135819107517221, 'r': 0.41692104192104196, 'f1': 0.4964861262571187}, 'combined': 0.28929112868525025, 'epoch': 31} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5755840455840456, 'r': 0.6023553965414431, 'f1': 0.5886655011655012}, 'combined': 0.28592324342324343, 'epoch': 31} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 32 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 20:20:09.207162: step: 2/1541, loss: 0.27946367859840393 2023-01-23 20:20:10.102644: step: 4/1541, loss: 0.0004815998545382172 2023-01-23 20:20:11.059076: step: 6/1541, loss: 0.0016038863686844707 2023-01-23 20:20:11.911994: step: 8/1541, loss: 3.907667996827513e-05 2023-01-23 20:20:12.785185: step: 10/1541, loss: 0.005859008524566889 2023-01-23 20:20:13.665880: step: 12/1541, loss: 8.379944483749568e-05 2023-01-23 20:20:14.551855: step: 14/1541, loss: 0.00010252413630951196 2023-01-23 20:20:15.456751: step: 16/1541, loss: 0.0010895368177443743 2023-01-23 20:20:16.394246: step: 18/1541, loss: 0.003624377539381385 2023-01-23 20:20:17.245483: step: 20/1541, loss: 0.00011861746315844357 2023-01-23 20:20:18.127471: step: 22/1541, loss: 0.0008193108369596303 2023-01-23 20:20:18.992211: step: 24/1541, loss: 0.0005605347687378526 2023-01-23 20:20:19.932632: step: 26/1541, loss: 0.0010549419093877077 2023-01-23 20:20:20.833739: step: 28/1541, loss: 0.0005732519784942269 2023-01-23 20:20:21.862377: step: 30/1541, loss: 0.003281723940744996 2023-01-23 20:20:22.733532: step: 32/1541, loss: 3.739449311979115e-05 2023-01-23 20:20:23.608156: step: 34/1541, loss: 0.00010492301953490824 2023-01-23 20:20:24.523246: step: 36/1541, loss: 0.0007360527524724603 2023-01-23 20:20:25.377917: step: 38/1541, loss: 0.0005710842087864876 2023-01-23 20:20:26.269292: step: 40/1541, loss: 0.028295151889324188 2023-01-23 20:20:27.142684: step: 42/1541, loss: 0.1234268769621849 2023-01-23 20:20:28.050540: step: 44/1541, loss: 0.1701793521642685 2023-01-23 20:20:28.946366: step: 46/1541, loss: 0.0005574565730057657 2023-01-23 20:20:29.799239: step: 48/1541, loss: 0.0003904185723513365 2023-01-23 20:20:30.708501: step: 50/1541, loss: 6.163310899864882e-05 2023-01-23 20:20:31.665558: step: 52/1541, loss: 0.04670026898384094 2023-01-23 20:20:32.564133: step: 54/1541, loss: 0.05358346551656723 2023-01-23 20:20:33.396650: step: 56/1541, loss: 0.0008525141747668386 2023-01-23 20:20:34.273528: step: 58/1541, loss: 0.004178058356046677 2023-01-23 20:20:35.148779: step: 60/1541, loss: 0.00043459507287479937 2023-01-23 20:20:36.006266: step: 62/1541, loss: 9.764184505911544e-05 2023-01-23 20:20:36.878604: step: 64/1541, loss: 0.0016413652338087559 2023-01-23 20:20:37.768172: step: 66/1541, loss: 0.00019229899044148624 2023-01-23 20:20:38.624877: step: 68/1541, loss: 0.00753899198025465 2023-01-23 20:20:39.557722: step: 70/1541, loss: 0.010267181321978569 2023-01-23 20:20:40.447324: step: 72/1541, loss: 6.9218008320603985e-06 2023-01-23 20:20:41.300799: step: 74/1541, loss: 0.010905453003942966 2023-01-23 20:20:42.216616: step: 76/1541, loss: 0.0354011133313179 2023-01-23 20:20:43.126667: step: 78/1541, loss: 9.640601638238877e-05 2023-01-23 20:20:43.993623: step: 80/1541, loss: 0.013375961221754551 2023-01-23 20:20:44.881147: step: 82/1541, loss: 0.5639797449111938 2023-01-23 20:20:45.718967: step: 84/1541, loss: 0.0001294043322559446 2023-01-23 20:20:46.634923: step: 86/1541, loss: 0.08050260692834854 2023-01-23 20:20:47.587513: step: 88/1541, loss: 2.40360951423645 2023-01-23 20:20:48.488870: step: 90/1541, loss: 0.026585089042782784 2023-01-23 20:20:49.372617: step: 92/1541, loss: 0.143968865275383 2023-01-23 20:20:50.261770: step: 94/1541, loss: 6.388826295733452e-05 2023-01-23 20:20:51.125795: step: 96/1541, loss: 0.004397165961563587 2023-01-23 20:20:52.033783: step: 98/1541, loss: 8.786726539256051e-05 2023-01-23 20:20:53.017968: step: 100/1541, loss: 0.0007242488791234791 2023-01-23 20:20:53.891813: step: 102/1541, loss: 0.0001693796511972323 2023-01-23 20:20:54.787000: step: 104/1541, loss: 6.150852277642116e-05 2023-01-23 20:20:55.640147: step: 106/1541, loss: 0.018735814839601517 2023-01-23 20:20:56.503531: step: 108/1541, loss: 0.0003482711035758257 2023-01-23 20:20:57.443450: step: 110/1541, loss: 0.0012045309413224459 2023-01-23 20:20:58.349791: step: 112/1541, loss: 0.0005120832938700914 2023-01-23 20:20:59.231778: step: 114/1541, loss: -4.352275936980732e-05 2023-01-23 20:21:00.097654: step: 116/1541, loss: 0.3332425355911255 2023-01-23 20:21:00.936725: step: 118/1541, loss: 0.04683909937739372 2023-01-23 20:21:01.816889: step: 120/1541, loss: 0.0004389646928757429 2023-01-23 20:21:02.762210: step: 122/1541, loss: 0.024111725389957428 2023-01-23 20:21:03.661376: step: 124/1541, loss: 0.0002913949720095843 2023-01-23 20:21:04.579263: step: 126/1541, loss: 2.0628172933356836e-05 2023-01-23 20:21:05.459289: step: 128/1541, loss: 6.904750625835732e-05 2023-01-23 20:21:06.321040: step: 130/1541, loss: 0.1331428736448288 2023-01-23 20:21:07.315412: step: 132/1541, loss: 0.04319412261247635 2023-01-23 20:21:08.189206: step: 134/1541, loss: 2.752532736849389e-06 2023-01-23 20:21:09.111000: step: 136/1541, loss: 0.21968959271907806 2023-01-23 20:21:09.961674: step: 138/1541, loss: 9.762524314282928e-06 2023-01-23 20:21:10.834082: step: 140/1541, loss: 0.0001473897573305294 2023-01-23 20:21:11.693642: step: 142/1541, loss: 0.0005138754495419562 2023-01-23 20:21:12.550849: step: 144/1541, loss: -7.275252755789552e-06 2023-01-23 20:21:13.450645: step: 146/1541, loss: 0.002409194130450487 2023-01-23 20:21:14.361638: step: 148/1541, loss: 0.008200019598007202 2023-01-23 20:21:15.249552: step: 150/1541, loss: 0.0005927648162469268 2023-01-23 20:21:16.107831: step: 152/1541, loss: 0.0009701648959890008 2023-01-23 20:21:16.966653: step: 154/1541, loss: 0.00529930554330349 2023-01-23 20:21:17.839884: step: 156/1541, loss: 0.00011096167145296931 2023-01-23 20:21:18.699876: step: 158/1541, loss: 0.005645491182804108 2023-01-23 20:21:19.599681: step: 160/1541, loss: 0.003610054962337017 2023-01-23 20:21:20.520480: step: 162/1541, loss: 5.6178938393713906e-05 2023-01-23 20:21:21.386659: step: 164/1541, loss: 0.00031427154317498207 2023-01-23 20:21:22.246415: step: 166/1541, loss: 0.0026862765662372112 2023-01-23 20:21:23.117116: step: 168/1541, loss: 0.23629765212535858 2023-01-23 20:21:24.008771: step: 170/1541, loss: -3.34333126374986e-05 2023-01-23 20:21:24.897092: step: 172/1541, loss: 0.06655076891183853 2023-01-23 20:21:25.762195: step: 174/1541, loss: -1.1395940418879036e-05 2023-01-23 20:21:26.664205: step: 176/1541, loss: 0.02533421292901039 2023-01-23 20:21:27.553399: step: 178/1541, loss: 0.1203896552324295 2023-01-23 20:21:28.438290: step: 180/1541, loss: 0.0760180726647377 2023-01-23 20:21:29.387940: step: 182/1541, loss: 0.001496936776675284 2023-01-23 20:21:30.290724: step: 184/1541, loss: 8.431022433796898e-05 2023-01-23 20:21:31.176878: step: 186/1541, loss: 2.2170692318468355e-05 2023-01-23 20:21:32.017997: step: 188/1541, loss: 0.005232313182204962 2023-01-23 20:21:32.934939: step: 190/1541, loss: 0.001823666738346219 2023-01-23 20:21:33.830898: step: 192/1541, loss: 0.000619884638581425 2023-01-23 20:21:34.691147: step: 194/1541, loss: 0.0016635081265121698 2023-01-23 20:21:35.556695: step: 196/1541, loss: 0.01342188660055399 2023-01-23 20:21:36.447856: step: 198/1541, loss: 0.0014049881137907505 2023-01-23 20:21:37.359481: step: 200/1541, loss: 0.07890751957893372 2023-01-23 20:21:38.245225: step: 202/1541, loss: 0.0010242502903565764 2023-01-23 20:21:39.111622: step: 204/1541, loss: 0.00833591166883707 2023-01-23 20:21:39.999936: step: 206/1541, loss: 6.203869997989386e-05 2023-01-23 20:21:40.857652: step: 208/1541, loss: 0.15032050013542175 2023-01-23 20:21:41.751669: step: 210/1541, loss: 0.04618315026164055 2023-01-23 20:21:42.687174: step: 212/1541, loss: 0.00021024384477641433 2023-01-23 20:21:43.608563: step: 214/1541, loss: 0.00035435205791145563 2023-01-23 20:21:44.558060: step: 216/1541, loss: 8.979659469332546e-05 2023-01-23 20:21:45.468619: step: 218/1541, loss: 0.0007337208371609449 2023-01-23 20:21:46.402683: step: 220/1541, loss: 0.04914514347910881 2023-01-23 20:21:47.252956: step: 222/1541, loss: 0.00024429819313809276 2023-01-23 20:21:48.138719: step: 224/1541, loss: 0.0005822451785206795 2023-01-23 20:21:49.002063: step: 226/1541, loss: 0.000337190373102203 2023-01-23 20:21:49.890632: step: 228/1541, loss: 0.0002921796403825283 2023-01-23 20:21:50.786556: step: 230/1541, loss: 0.09081102907657623 2023-01-23 20:21:51.709710: step: 232/1541, loss: -2.2075655579101294e-06 2023-01-23 20:21:52.580742: step: 234/1541, loss: 0.0009684249525889754 2023-01-23 20:21:53.459982: step: 236/1541, loss: 0.03368476778268814 2023-01-23 20:21:54.425956: step: 238/1541, loss: 0.010007808916270733 2023-01-23 20:21:55.362572: step: 240/1541, loss: 9.560326725477353e-05 2023-01-23 20:21:56.240306: step: 242/1541, loss: 9.40849567996338e-05 2023-01-23 20:21:57.099986: step: 244/1541, loss: 3.931820538127795e-05 2023-01-23 20:21:57.996151: step: 246/1541, loss: 0.00014441536040976644 2023-01-23 20:21:58.897797: step: 248/1541, loss: 8.380063809454441e-05 2023-01-23 20:21:59.837758: step: 250/1541, loss: 0.09495389461517334 2023-01-23 20:22:00.710949: step: 252/1541, loss: 0.006124449893832207 2023-01-23 20:22:01.572886: step: 254/1541, loss: 0.0005096419481560588 2023-01-23 20:22:02.535910: step: 256/1541, loss: 0.00029551584157161415 2023-01-23 20:22:03.403931: step: 258/1541, loss: 0.0002778275520540774 2023-01-23 20:22:04.376872: step: 260/1541, loss: 0.0003679537621792406 2023-01-23 20:22:05.261072: step: 262/1541, loss: 0.00014117732644081116 2023-01-23 20:22:06.086515: step: 264/1541, loss: 2.8124872187618166e-05 2023-01-23 20:22:06.992951: step: 266/1541, loss: 3.968759119743481e-05 2023-01-23 20:22:07.906351: step: 268/1541, loss: 0.00017594303062651306 2023-01-23 20:22:08.818183: step: 270/1541, loss: 7.703348819632083e-05 2023-01-23 20:22:09.676124: step: 272/1541, loss: 0.3960779905319214 2023-01-23 20:22:10.644657: step: 274/1541, loss: 0.008234893903136253 2023-01-23 20:22:11.566883: step: 276/1541, loss: 0.6067779064178467 2023-01-23 20:22:12.514792: step: 278/1541, loss: 0.2721737325191498 2023-01-23 20:22:13.374970: step: 280/1541, loss: 0.0013686887687072158 2023-01-23 20:22:14.257350: step: 282/1541, loss: 0.020494239404797554 2023-01-23 20:22:15.105299: step: 284/1541, loss: 0.0017724415520206094 2023-01-23 20:22:15.971983: step: 286/1541, loss: 0.0027789922896772623 2023-01-23 20:22:16.907468: step: 288/1541, loss: 0.08651421964168549 2023-01-23 20:22:17.774290: step: 290/1541, loss: 0.3169184923171997 2023-01-23 20:22:18.697410: step: 292/1541, loss: 0.002244934206828475 2023-01-23 20:22:19.672016: step: 294/1541, loss: 0.0013914336450397968 2023-01-23 20:22:20.528189: step: 296/1541, loss: 0.07202493399381638 2023-01-23 20:22:21.435794: step: 298/1541, loss: 0.00017853896133601665 2023-01-23 20:22:22.423420: step: 300/1541, loss: 0.09988865256309509 2023-01-23 20:22:23.301397: step: 302/1541, loss: 0.099082812666893 2023-01-23 20:22:24.182080: step: 304/1541, loss: 3.4386121114948764e-05 2023-01-23 20:22:25.072320: step: 306/1541, loss: 0.0036768359132111073 2023-01-23 20:22:25.960474: step: 308/1541, loss: 0.0628865584731102 2023-01-23 20:22:26.869343: step: 310/1541, loss: 0.013429675251245499 2023-01-23 20:22:27.738467: step: 312/1541, loss: 0.0013526990078389645 2023-01-23 20:22:28.669005: step: 314/1541, loss: 0.04618849977850914 2023-01-23 20:22:29.549002: step: 316/1541, loss: 7.543552783317864e-05 2023-01-23 20:22:30.429427: step: 318/1541, loss: 5.027199586038478e-05 2023-01-23 20:22:31.321875: step: 320/1541, loss: 9.10171729628928e-05 2023-01-23 20:22:32.227751: step: 322/1541, loss: 0.03952264413237572 2023-01-23 20:22:33.106080: step: 324/1541, loss: 0.10247410833835602 2023-01-23 20:22:33.962668: step: 326/1541, loss: 0.012998152524232864 2023-01-23 20:22:34.860274: step: 328/1541, loss: 0.011641589924693108 2023-01-23 20:22:35.847759: step: 330/1541, loss: 0.007629279047250748 2023-01-23 20:22:36.766468: step: 332/1541, loss: 5.09030869579874e-05 2023-01-23 20:22:37.687513: step: 334/1541, loss: 0.034175872802734375 2023-01-23 20:22:38.549517: step: 336/1541, loss: 0.0004904231172986329 2023-01-23 20:22:39.467317: step: 338/1541, loss: 0.03671657666563988 2023-01-23 20:22:40.311726: step: 340/1541, loss: 0.025322478264570236 2023-01-23 20:22:41.167191: step: 342/1541, loss: 0.03935772925615311 2023-01-23 20:22:42.034422: step: 344/1541, loss: 0.0006858569104224443 2023-01-23 20:22:42.944535: step: 346/1541, loss: 0.04368803650140762 2023-01-23 20:22:43.822842: step: 348/1541, loss: 0.0012239179341122508 2023-01-23 20:22:44.728084: step: 350/1541, loss: 0.0002259121829411015 2023-01-23 20:22:45.607427: step: 352/1541, loss: 0.0003109806275460869 2023-01-23 20:22:46.499945: step: 354/1541, loss: 0.5276966691017151 2023-01-23 20:22:47.386423: step: 356/1541, loss: 0.13705746829509735 2023-01-23 20:22:48.281527: step: 358/1541, loss: 0.08865664154291153 2023-01-23 20:22:49.235603: step: 360/1541, loss: 0.07191098481416702 2023-01-23 20:22:50.146508: step: 362/1541, loss: 0.0908002182841301 2023-01-23 20:22:51.012380: step: 364/1541, loss: 3.471672607702203e-05 2023-01-23 20:22:51.922556: step: 366/1541, loss: 0.0008100606501102448 2023-01-23 20:22:52.824242: step: 368/1541, loss: 0.10961402952671051 2023-01-23 20:22:53.747692: step: 370/1541, loss: 0.0008659141603857279 2023-01-23 20:22:54.653610: step: 372/1541, loss: 0.008871341124176979 2023-01-23 20:22:55.513417: step: 374/1541, loss: 0.00037144756061024964 2023-01-23 20:22:56.387934: step: 376/1541, loss: 0.07558833807706833 2023-01-23 20:22:57.245102: step: 378/1541, loss: 4.6460329031106085e-05 2023-01-23 20:22:58.156963: step: 380/1541, loss: 0.01810554414987564 2023-01-23 20:22:59.054360: step: 382/1541, loss: 0.03599146381020546 2023-01-23 20:22:59.916114: step: 384/1541, loss: 6.090512761147693e-05 2023-01-23 20:23:00.841146: step: 386/1541, loss: 0.004225147422403097 2023-01-23 20:23:01.681373: step: 388/1541, loss: 3.223656676709652e-05 2023-01-23 20:23:02.579907: step: 390/1541, loss: 0.00028569845017045736 2023-01-23 20:23:03.504332: step: 392/1541, loss: 9.245399996871129e-05 2023-01-23 20:23:04.351429: step: 394/1541, loss: 0.002350186463445425 2023-01-23 20:23:05.217080: step: 396/1541, loss: 0.0008301758789457381 2023-01-23 20:23:06.136151: step: 398/1541, loss: 0.007193912286311388 2023-01-23 20:23:07.073238: step: 400/1541, loss: 1.3851125004293863e-05 2023-01-23 20:23:07.963402: step: 402/1541, loss: 5.318913463270292e-05 2023-01-23 20:23:08.855713: step: 404/1541, loss: -6.648010185017483e-06 2023-01-23 20:23:09.776585: step: 406/1541, loss: 0.013720018789172173 2023-01-23 20:23:10.726074: step: 408/1541, loss: 0.0009523819899186492 2023-01-23 20:23:11.591327: step: 410/1541, loss: 0.00014446317800320685 2023-01-23 20:23:12.523587: step: 412/1541, loss: 0.0021937196142971516 2023-01-23 20:23:13.381881: step: 414/1541, loss: 0.38523393869400024 2023-01-23 20:23:14.237090: step: 416/1541, loss: 0.0051973541267216206 2023-01-23 20:23:15.120861: step: 418/1541, loss: 8.030731987673789e-05 2023-01-23 20:23:15.971468: step: 420/1541, loss: 0.008770554326474667 2023-01-23 20:23:16.910610: step: 422/1541, loss: 0.04168718308210373 2023-01-23 20:23:17.766499: step: 424/1541, loss: -2.348039924982004e-05 2023-01-23 20:23:18.648083: step: 426/1541, loss: 0.00032463346724398434 2023-01-23 20:23:19.542268: step: 428/1541, loss: 0.09077369421720505 2023-01-23 20:23:20.456852: step: 430/1541, loss: 0.0002820205991156399 2023-01-23 20:23:21.379475: step: 432/1541, loss: 0.050510358065366745 2023-01-23 20:23:22.263182: step: 434/1541, loss: 2.3338592654909007e-05 2023-01-23 20:23:23.164191: step: 436/1541, loss: 6.007518095429987e-05 2023-01-23 20:23:24.082653: step: 438/1541, loss: 0.013500627130270004 2023-01-23 20:23:24.953672: step: 440/1541, loss: 2.4827719244058244e-05 2023-01-23 20:23:25.819353: step: 442/1541, loss: 9.92189143289579e-06 2023-01-23 20:23:26.703655: step: 444/1541, loss: 0.009890049695968628 2023-01-23 20:23:27.563402: step: 446/1541, loss: 0.015539453364908695 2023-01-23 20:23:28.449969: step: 448/1541, loss: 1.3576440323959105e-05 2023-01-23 20:23:29.329823: step: 450/1541, loss: 0.029600298032164574 2023-01-23 20:23:30.214262: step: 452/1541, loss: 1.6794354451121762e-05 2023-01-23 20:23:31.147791: step: 454/1541, loss: 0.014165927655994892 2023-01-23 20:23:32.061191: step: 456/1541, loss: 0.6018339395523071 2023-01-23 20:23:32.960292: step: 458/1541, loss: 2.59196476690704e-05 2023-01-23 20:23:33.830338: step: 460/1541, loss: 7.955242472235113e-05 2023-01-23 20:23:34.791054: step: 462/1541, loss: 0.0007925406098365784 2023-01-23 20:23:35.717759: step: 464/1541, loss: 0.025212086737155914 2023-01-23 20:23:36.611033: step: 466/1541, loss: 0.637891948223114 2023-01-23 20:23:37.539123: step: 468/1541, loss: 2.2172671378939413e-06 2023-01-23 20:23:38.386477: step: 470/1541, loss: 4.5927907194709405e-05 2023-01-23 20:23:39.272857: step: 472/1541, loss: 0.03216332942247391 2023-01-23 20:23:40.170382: step: 474/1541, loss: 0.06953612715005875 2023-01-23 20:23:41.055772: step: 476/1541, loss: 0.005542275961488485 2023-01-23 20:23:41.976620: step: 478/1541, loss: 0.03635973483324051 2023-01-23 20:23:42.861767: step: 480/1541, loss: 0.5081778168678284 2023-01-23 20:23:43.749548: step: 482/1541, loss: 0.0004378400044515729 2023-01-23 20:23:44.628201: step: 484/1541, loss: -2.9161583370296285e-06 2023-01-23 20:23:45.504262: step: 486/1541, loss: 0.014999305829405785 2023-01-23 20:23:46.379655: step: 488/1541, loss: 0.00019840376626234502 2023-01-23 20:23:47.375526: step: 490/1541, loss: -5.5096574214985594e-05 2023-01-23 20:23:48.335827: step: 492/1541, loss: 0.03259136900305748 2023-01-23 20:23:49.184504: step: 494/1541, loss: 0.019523898139595985 2023-01-23 20:23:50.131927: step: 496/1541, loss: 0.09740910679101944 2023-01-23 20:23:51.065302: step: 498/1541, loss: 0.005983480252325535 2023-01-23 20:23:51.915683: step: 500/1541, loss: 0.0003405947354622185 2023-01-23 20:23:52.800185: step: 502/1541, loss: 0.00041466139373369515 2023-01-23 20:23:53.680755: step: 504/1541, loss: 0.0003028393548447639 2023-01-23 20:23:54.581145: step: 506/1541, loss: 0.16769716143608093 2023-01-23 20:23:55.481097: step: 508/1541, loss: 0.019885821267962456 2023-01-23 20:23:56.466989: step: 510/1541, loss: 0.00018679641652852297 2023-01-23 20:23:57.334367: step: 512/1541, loss: 1.4379112720489502 2023-01-23 20:23:58.237148: step: 514/1541, loss: 0.00016705249436199665 2023-01-23 20:23:59.139038: step: 516/1541, loss: 0.0030710259452462196 2023-01-23 20:24:00.044689: step: 518/1541, loss: 0.01365535706281662 2023-01-23 20:24:00.917028: step: 520/1541, loss: 0.33654657006263733 2023-01-23 20:24:01.875969: step: 522/1541, loss: 0.07200229167938232 2023-01-23 20:24:02.772022: step: 524/1541, loss: 0.00032844935776665807 2023-01-23 20:24:03.684142: step: 526/1541, loss: 0.001638267538510263 2023-01-23 20:24:04.599455: step: 528/1541, loss: 0.07178512960672379 2023-01-23 20:24:05.472675: step: 530/1541, loss: 1.3672205209732056 2023-01-23 20:24:06.342943: step: 532/1541, loss: 0.011917879804968834 2023-01-23 20:24:07.251517: step: 534/1541, loss: 0.0001168886010418646 2023-01-23 20:24:08.154041: step: 536/1541, loss: 0.00404878705739975 2023-01-23 20:24:09.067303: step: 538/1541, loss: -4.405439176480286e-05 2023-01-23 20:24:09.937869: step: 540/1541, loss: 2.581655644462444e-05 2023-01-23 20:24:10.847338: step: 542/1541, loss: 0.05301285907626152 2023-01-23 20:24:11.708826: step: 544/1541, loss: 0.0003939337329939008 2023-01-23 20:24:12.636853: step: 546/1541, loss: 0.020082570612430573 2023-01-23 20:24:13.520516: step: 548/1541, loss: 0.0006053732940927148 2023-01-23 20:24:14.431254: step: 550/1541, loss: 0.07757345587015152 2023-01-23 20:24:15.324312: step: 552/1541, loss: 0.2551177740097046 2023-01-23 20:24:16.246188: step: 554/1541, loss: 0.0019986059051007032 2023-01-23 20:24:17.134868: step: 556/1541, loss: 0.14400237798690796 2023-01-23 20:24:17.998290: step: 558/1541, loss: 0.0023634699173271656 2023-01-23 20:24:18.912110: step: 560/1541, loss: 0.00033725594403222203 2023-01-23 20:24:19.798841: step: 562/1541, loss: 0.01828111708164215 2023-01-23 20:24:20.720899: step: 564/1541, loss: 0.17902666330337524 2023-01-23 20:24:21.636284: step: 566/1541, loss: 0.007102818228304386 2023-01-23 20:24:22.533649: step: 568/1541, loss: 0.016512274742126465 2023-01-23 20:24:23.480674: step: 570/1541, loss: 3.83310834877193e-05 2023-01-23 20:24:24.361954: step: 572/1541, loss: 4.190011532045901e-05 2023-01-23 20:24:25.260234: step: 574/1541, loss: 0.09744785726070404 2023-01-23 20:24:26.137031: step: 576/1541, loss: 0.0024513325188308954 2023-01-23 20:24:27.013620: step: 578/1541, loss: 0.0024076623376458883 2023-01-23 20:24:27.927789: step: 580/1541, loss: 0.1462857574224472 2023-01-23 20:24:28.841753: step: 582/1541, loss: 0.01905454695224762 2023-01-23 20:24:29.751466: step: 584/1541, loss: 0.0008444017730653286 2023-01-23 20:24:30.667597: step: 586/1541, loss: 0.12726622819900513 2023-01-23 20:24:31.573176: step: 588/1541, loss: 0.003783585038036108 2023-01-23 20:24:32.451603: step: 590/1541, loss: 0.021238524466753006 2023-01-23 20:24:33.358449: step: 592/1541, loss: 0.0642058327794075 2023-01-23 20:24:34.233995: step: 594/1541, loss: 0.0528760701417923 2023-01-23 20:24:35.147518: step: 596/1541, loss: 0.0007801475003361702 2023-01-23 20:24:36.052328: step: 598/1541, loss: 0.0005305394297465682 2023-01-23 20:24:36.984646: step: 600/1541, loss: 0.0005755179445259273 2023-01-23 20:24:37.932123: step: 602/1541, loss: 0.11975265294313431 2023-01-23 20:24:38.810966: step: 604/1541, loss: 0.006330449134111404 2023-01-23 20:24:39.750557: step: 606/1541, loss: 0.03302617743611336 2023-01-23 20:24:40.630140: step: 608/1541, loss: 0.005978397559374571 2023-01-23 20:24:41.507196: step: 610/1541, loss: 0.0001608994643902406 2023-01-23 20:24:42.412009: step: 612/1541, loss: 0.010345853865146637 2023-01-23 20:24:43.332586: step: 614/1541, loss: 0.0005393121391534805 2023-01-23 20:24:44.235254: step: 616/1541, loss: 0.11733781546354294 2023-01-23 20:24:45.125353: step: 618/1541, loss: 0.01093974057585001 2023-01-23 20:24:46.027754: step: 620/1541, loss: 0.0017163361189886928 2023-01-23 20:24:46.952379: step: 622/1541, loss: 0.00933746062219143 2023-01-23 20:24:47.848407: step: 624/1541, loss: 0.00011256038851570338 2023-01-23 20:24:48.718007: step: 626/1541, loss: 0.1406393051147461 2023-01-23 20:24:49.594909: step: 628/1541, loss: 0.000123173143947497 2023-01-23 20:24:50.564510: step: 630/1541, loss: 0.028215091675519943 2023-01-23 20:24:51.453634: step: 632/1541, loss: 0.08661540597677231 2023-01-23 20:24:52.351272: step: 634/1541, loss: 0.00010634463978931308 2023-01-23 20:24:53.265505: step: 636/1541, loss: 0.0001377393346047029 2023-01-23 20:24:54.160421: step: 638/1541, loss: 0.003215080127120018 2023-01-23 20:24:55.086350: step: 640/1541, loss: -2.836164412656217e-06 2023-01-23 20:24:55.994955: step: 642/1541, loss: 0.00021640982595272362 2023-01-23 20:24:56.890420: step: 644/1541, loss: -2.723244506341871e-06 2023-01-23 20:24:57.791416: step: 646/1541, loss: 0.00023333024000748992 2023-01-23 20:24:58.707632: step: 648/1541, loss: 0.00025950322742573917 2023-01-23 20:24:59.641071: step: 650/1541, loss: 9.120732283918187e-05 2023-01-23 20:25:00.491338: step: 652/1541, loss: 4.9592057621339336e-05 2023-01-23 20:25:01.391475: step: 654/1541, loss: 0.0015541138127446175 2023-01-23 20:25:02.331215: step: 656/1541, loss: 0.2861427664756775 2023-01-23 20:25:03.290380: step: 658/1541, loss: 0.040049608796834946 2023-01-23 20:25:04.133677: step: 660/1541, loss: 0.0007745518814772367 2023-01-23 20:25:05.028466: step: 662/1541, loss: 0.0014320644550025463 2023-01-23 20:25:05.930175: step: 664/1541, loss: 0.26430243253707886 2023-01-23 20:25:06.777557: step: 666/1541, loss: 0.004110902547836304 2023-01-23 20:25:07.689957: step: 668/1541, loss: 0.02117481641471386 2023-01-23 20:25:08.553000: step: 670/1541, loss: 0.07788321375846863 2023-01-23 20:25:09.425082: step: 672/1541, loss: 3.070075763389468e-05 2023-01-23 20:25:10.277623: step: 674/1541, loss: 0.00800804328173399 2023-01-23 20:25:11.224691: step: 676/1541, loss: 0.013614615425467491 2023-01-23 20:25:12.094898: step: 678/1541, loss: 0.00018696107144933194 2023-01-23 20:25:13.037856: step: 680/1541, loss: 0.0021569717209786177 2023-01-23 20:25:13.955149: step: 682/1541, loss: 0.058451879769563675 2023-01-23 20:25:14.870297: step: 684/1541, loss: 0.0032046812120825052 2023-01-23 20:25:15.747707: step: 686/1541, loss: 0.01979820430278778 2023-01-23 20:25:16.648129: step: 688/1541, loss: 0.0710785910487175 2023-01-23 20:25:17.577867: step: 690/1541, loss: 0.15823866426944733 2023-01-23 20:25:18.461026: step: 692/1541, loss: 9.301007230533287e-05 2023-01-23 20:25:19.345742: step: 694/1541, loss: 0.011394968256354332 2023-01-23 20:25:20.185564: step: 696/1541, loss: 0.016347963362932205 2023-01-23 20:25:21.092528: step: 698/1541, loss: 1.5740992239443585e-05 2023-01-23 20:25:21.959615: step: 700/1541, loss: 0.12957386672496796 2023-01-23 20:25:22.861894: step: 702/1541, loss: 0.0018338959198445082 2023-01-23 20:25:23.759057: step: 704/1541, loss: 0.0003212353913113475 2023-01-23 20:25:24.639523: step: 706/1541, loss: 0.00013522090739570558 2023-01-23 20:25:25.505690: step: 708/1541, loss: 6.815492815803736e-05 2023-01-23 20:25:26.402457: step: 710/1541, loss: 0.00022573006572201848 2023-01-23 20:25:27.377894: step: 712/1541, loss: 0.005502437241375446 2023-01-23 20:25:28.271652: step: 714/1541, loss: 0.18289250135421753 2023-01-23 20:25:29.137388: step: 716/1541, loss: 0.09200199693441391 2023-01-23 20:25:30.157806: step: 718/1541, loss: 0.011296767741441727 2023-01-23 20:25:31.067360: step: 720/1541, loss: 0.23597578704357147 2023-01-23 20:25:31.953295: step: 722/1541, loss: 0.0013062889920547605 2023-01-23 20:25:32.886839: step: 724/1541, loss: 0.00359870376996696 2023-01-23 20:25:33.803164: step: 726/1541, loss: 0.008343297056853771 2023-01-23 20:25:34.747102: step: 728/1541, loss: 0.02333882451057434 2023-01-23 20:25:35.630759: step: 730/1541, loss: 0.00990677997469902 2023-01-23 20:25:36.500339: step: 732/1541, loss: 0.03849126398563385 2023-01-23 20:25:37.407180: step: 734/1541, loss: 0.0033616425935178995 2023-01-23 20:25:38.344765: step: 736/1541, loss: 0.00027000566478818655 2023-01-23 20:25:39.230817: step: 738/1541, loss: 0.00025884920614771545 2023-01-23 20:25:40.142549: step: 740/1541, loss: 0.0010163509286940098 2023-01-23 20:25:41.014018: step: 742/1541, loss: 0.002041976433247328 2023-01-23 20:25:41.856570: step: 744/1541, loss: 3.97903895645868e-05 2023-01-23 20:25:42.808863: step: 746/1541, loss: 0.0001399953616783023 2023-01-23 20:25:43.728050: step: 748/1541, loss: 0.025299062952399254 2023-01-23 20:25:44.646237: step: 750/1541, loss: 0.0008889115415513515 2023-01-23 20:25:45.509288: step: 752/1541, loss: 0.00012496068666223437 2023-01-23 20:25:46.379306: step: 754/1541, loss: 0.27047625184059143 2023-01-23 20:25:47.271216: step: 756/1541, loss: 0.0002371974114794284 2023-01-23 20:25:48.193933: step: 758/1541, loss: 0.0003838955017272383 2023-01-23 20:25:49.075054: step: 760/1541, loss: 0.009045782499015331 2023-01-23 20:25:49.945562: step: 762/1541, loss: 0.055736761540174484 2023-01-23 20:25:50.813769: step: 764/1541, loss: 0.001378806191496551 2023-01-23 20:25:51.779482: step: 766/1541, loss: 0.009182076901197433 2023-01-23 20:25:52.672184: step: 768/1541, loss: 0.00015739782247692347 2023-01-23 20:25:53.557807: step: 770/1541, loss: 0.005610853433609009 2023-01-23 20:25:54.416328: step: 772/1541, loss: 0.00014642711903434247 2023-01-23 20:25:55.309122: step: 774/1541, loss: 0.00751736294478178 2023-01-23 20:25:56.210324: step: 776/1541, loss: 0.0034590265713632107 2023-01-23 20:25:57.195168: step: 778/1541, loss: 0.014596215449273586 2023-01-23 20:25:58.087293: step: 780/1541, loss: 1.520719051361084 2023-01-23 20:25:59.035928: step: 782/1541, loss: 0.017331967130303383 2023-01-23 20:25:59.917939: step: 784/1541, loss: 0.00020321916963439435 2023-01-23 20:26:00.778335: step: 786/1541, loss: 0.005982531700283289 2023-01-23 20:26:01.655703: step: 788/1541, loss: 0.004799028858542442 2023-01-23 20:26:02.561312: step: 790/1541, loss: 0.00024272545124404132 2023-01-23 20:26:03.425791: step: 792/1541, loss: 9.319041419075802e-05 2023-01-23 20:26:04.336692: step: 794/1541, loss: 0.0001531353045720607 2023-01-23 20:26:05.271081: step: 796/1541, loss: 0.5812057852745056 2023-01-23 20:26:06.163457: step: 798/1541, loss: 0.002838718704879284 2023-01-23 20:26:07.096295: step: 800/1541, loss: 0.0081239128485322 2023-01-23 20:26:07.982109: step: 802/1541, loss: 0.09549415856599808 2023-01-23 20:26:08.884827: step: 804/1541, loss: 0.02738887071609497 2023-01-23 20:26:09.755168: step: 806/1541, loss: 0.10692130029201508 2023-01-23 20:26:10.696039: step: 808/1541, loss: 0.0011432584142312407 2023-01-23 20:26:11.615856: step: 810/1541, loss: 0.1521000862121582 2023-01-23 20:26:12.520642: step: 812/1541, loss: 0.00033961431472562253 2023-01-23 20:26:13.398790: step: 814/1541, loss: 0.00016298482660204172 2023-01-23 20:26:14.280556: step: 816/1541, loss: 0.0028484463691711426 2023-01-23 20:26:15.159565: step: 818/1541, loss: 0.0002932685019914061 2023-01-23 20:26:16.033053: step: 820/1541, loss: 0.0014356145402416587 2023-01-23 20:26:16.900349: step: 822/1541, loss: 0.0021740298252552748 2023-01-23 20:26:17.823490: step: 824/1541, loss: 0.00015934917610138655 2023-01-23 20:26:18.826530: step: 826/1541, loss: 0.0011444676201790571 2023-01-23 20:26:19.683695: step: 828/1541, loss: 0.0012822800781577826 2023-01-23 20:26:20.637477: step: 830/1541, loss: 0.006306820549070835 2023-01-23 20:26:21.531348: step: 832/1541, loss: 0.0009668181301094592 2023-01-23 20:26:22.539769: step: 834/1541, loss: 0.18944872915744781 2023-01-23 20:26:23.441502: step: 836/1541, loss: 0.032904841005802155 2023-01-23 20:26:24.378758: step: 838/1541, loss: 0.00013381648750510067 2023-01-23 20:26:25.274310: step: 840/1541, loss: 9.548159141559154e-05 2023-01-23 20:26:26.142950: step: 842/1541, loss: 0.0011201818706467748 2023-01-23 20:26:27.051282: step: 844/1541, loss: 0.17690756916999817 2023-01-23 20:26:27.925083: step: 846/1541, loss: 0.00046082527842372656 2023-01-23 20:26:28.816749: step: 848/1541, loss: 0.012899006716907024 2023-01-23 20:26:29.742046: step: 850/1541, loss: 1.3653031601279508e-05 2023-01-23 20:26:30.606001: step: 852/1541, loss: 6.267127173487097e-05 2023-01-23 20:26:31.510992: step: 854/1541, loss: 0.07664132118225098 2023-01-23 20:26:32.475483: step: 856/1541, loss: 0.0073067848570644855 2023-01-23 20:26:33.437241: step: 858/1541, loss: 0.02924329973757267 2023-01-23 20:26:34.391020: step: 860/1541, loss: 0.00019205936405342072 2023-01-23 20:26:35.352397: step: 862/1541, loss: 0.010173587128520012 2023-01-23 20:26:36.256979: step: 864/1541, loss: 0.002092681359499693 2023-01-23 20:26:37.119347: step: 866/1541, loss: 0.0003760014078579843 2023-01-23 20:26:38.049697: step: 868/1541, loss: 0.032878078520298004 2023-01-23 20:26:38.921328: step: 870/1541, loss: 0.050253089517354965 2023-01-23 20:26:39.811176: step: 872/1541, loss: 0.15302826464176178 2023-01-23 20:26:40.685951: step: 874/1541, loss: 0.007362186908721924 2023-01-23 20:26:41.593970: step: 876/1541, loss: 0.017882628366351128 2023-01-23 20:26:42.471836: step: 878/1541, loss: 0.00024306047998834401 2023-01-23 20:26:43.322429: step: 880/1541, loss: 0.04486801102757454 2023-01-23 20:26:44.171465: step: 882/1541, loss: 0.002995179034769535 2023-01-23 20:26:45.065234: step: 884/1541, loss: 0.00030975911067798734 2023-01-23 20:26:45.977654: step: 886/1541, loss: 0.007317284122109413 2023-01-23 20:26:46.854805: step: 888/1541, loss: 0.00011854639160446823 2023-01-23 20:26:47.753308: step: 890/1541, loss: 0.0004220552509650588 2023-01-23 20:26:48.662075: step: 892/1541, loss: 0.00020767129899468273 2023-01-23 20:26:49.562611: step: 894/1541, loss: 0.011620876379311085 2023-01-23 20:26:50.451784: step: 896/1541, loss: 0.00023844910901971161 2023-01-23 20:26:51.316040: step: 898/1541, loss: 0.002691864036023617 2023-01-23 20:26:52.254255: step: 900/1541, loss: -3.317925438750535e-05 2023-01-23 20:26:53.162174: step: 902/1541, loss: 0.26611530780792236 2023-01-23 20:26:54.044101: step: 904/1541, loss: 0.0004488707345444709 2023-01-23 20:26:54.921926: step: 906/1541, loss: 0.003251161891967058 2023-01-23 20:26:55.829007: step: 908/1541, loss: 0.0018401378765702248 2023-01-23 20:26:56.717356: step: 910/1541, loss: 5.3220683184918016e-05 2023-01-23 20:26:57.699458: step: 912/1541, loss: 0.24174410104751587 2023-01-23 20:26:58.618165: step: 914/1541, loss: 0.0002457673544995487 2023-01-23 20:26:59.501765: step: 916/1541, loss: 0.00038013875018805265 2023-01-23 20:27:00.446479: step: 918/1541, loss: 0.0025903084315359592 2023-01-23 20:27:01.370777: step: 920/1541, loss: 0.01007423922419548 2023-01-23 20:27:02.290735: step: 922/1541, loss: 0.043332088738679886 2023-01-23 20:27:03.178393: step: 924/1541, loss: 0.0001802755577955395 2023-01-23 20:27:04.088067: step: 926/1541, loss: 0.00035447085974738 2023-01-23 20:27:04.969332: step: 928/1541, loss: 0.010701649822294712 2023-01-23 20:27:05.854834: step: 930/1541, loss: 0.4761281907558441 2023-01-23 20:27:06.789406: step: 932/1541, loss: 0.001049558981321752 2023-01-23 20:27:07.647641: step: 934/1541, loss: 0.001124034752137959 2023-01-23 20:27:08.551373: step: 936/1541, loss: 0.0024587935768067837 2023-01-23 20:27:09.438305: step: 938/1541, loss: 0.3920917510986328 2023-01-23 20:27:10.341368: step: 940/1541, loss: 1.632274866104126 2023-01-23 20:27:11.274847: step: 942/1541, loss: 0.0011486727744340897 2023-01-23 20:27:12.147365: step: 944/1541, loss: 0.009779070504009724 2023-01-23 20:27:13.007455: step: 946/1541, loss: 0.001171721494756639 2023-01-23 20:27:13.913163: step: 948/1541, loss: 0.027715658769011497 2023-01-23 20:27:14.841778: step: 950/1541, loss: 0.00014669611118733883 2023-01-23 20:27:15.719616: step: 952/1541, loss: 0.055216770619153976 2023-01-23 20:27:16.632161: step: 954/1541, loss: 0.004834720864892006 2023-01-23 20:27:17.536753: step: 956/1541, loss: 0.0011133650550618768 2023-01-23 20:27:18.405507: step: 958/1541, loss: 0.596472978591919 2023-01-23 20:27:19.300008: step: 960/1541, loss: 0.0035290499217808247 2023-01-23 20:27:20.213589: step: 962/1541, loss: 0.02381066419184208 2023-01-23 20:27:21.192210: step: 964/1541, loss: 0.01886889711022377 2023-01-23 20:27:22.069563: step: 966/1541, loss: 0.16716183722019196 2023-01-23 20:27:22.958545: step: 968/1541, loss: 0.026388375088572502 2023-01-23 20:27:23.840032: step: 970/1541, loss: 0.05686789005994797 2023-01-23 20:27:24.743914: step: 972/1541, loss: 0.004420807119458914 2023-01-23 20:27:25.596788: step: 974/1541, loss: 0.0015813750214874744 2023-01-23 20:27:26.530361: step: 976/1541, loss: 0.000958816846832633 2023-01-23 20:27:27.456859: step: 978/1541, loss: 3.442222805460915e-05 2023-01-23 20:27:28.383614: step: 980/1541, loss: 0.06203751266002655 2023-01-23 20:27:29.291585: step: 982/1541, loss: 0.15660201013088226 2023-01-23 20:27:30.171015: step: 984/1541, loss: 0.0008477020310238004 2023-01-23 20:27:31.030757: step: 986/1541, loss: 0.002210839418694377 2023-01-23 20:27:31.947061: step: 988/1541, loss: 0.23832033574581146 2023-01-23 20:27:32.827921: step: 990/1541, loss: 0.007815130054950714 2023-01-23 20:27:33.703005: step: 992/1541, loss: 0.0013858230086043477 2023-01-23 20:27:34.630476: step: 994/1541, loss: 0.001936599612236023 2023-01-23 20:27:35.530166: step: 996/1541, loss: 0.13315603137016296 2023-01-23 20:27:36.405637: step: 998/1541, loss: 0.012695259414613247 2023-01-23 20:27:37.372398: step: 1000/1541, loss: 0.00018351190374232829 2023-01-23 20:27:38.241614: step: 1002/1541, loss: 1.9161758245900273e-05 2023-01-23 20:27:39.126350: step: 1004/1541, loss: 0.0007038315525278449 2023-01-23 20:27:39.992007: step: 1006/1541, loss: 4.6005938202142715e-05 2023-01-23 20:27:40.873128: step: 1008/1541, loss: 0.00022809427173342556 2023-01-23 20:27:41.762870: step: 1010/1541, loss: 0.10535331070423126 2023-01-23 20:27:42.647652: step: 1012/1541, loss: 0.000533412559889257 2023-01-23 20:27:43.534149: step: 1014/1541, loss: 8.037149382289499e-05 2023-01-23 20:27:44.407442: step: 1016/1541, loss: 4.701377292803954e-06 2023-01-23 20:27:45.295213: step: 1018/1541, loss: 0.36091578006744385 2023-01-23 20:27:46.177974: step: 1020/1541, loss: 0.00027866181335411966 2023-01-23 20:27:47.021106: step: 1022/1541, loss: 0.0001344605698250234 2023-01-23 20:27:47.920483: step: 1024/1541, loss: 0.00022340306895785034 2023-01-23 20:27:48.802493: step: 1026/1541, loss: 0.00021700913202948868 2023-01-23 20:27:49.725266: step: 1028/1541, loss: 0.007125677540898323 2023-01-23 20:27:50.620598: step: 1030/1541, loss: 0.021087128669023514 2023-01-23 20:27:51.532611: step: 1032/1541, loss: 0.04965204373002052 2023-01-23 20:27:52.402159: step: 1034/1541, loss: 0.0008305566152557731 2023-01-23 20:27:53.256604: step: 1036/1541, loss: 0.00015264797548297793 2023-01-23 20:27:54.157079: step: 1038/1541, loss: 0.1280178725719452 2023-01-23 20:27:55.059635: step: 1040/1541, loss: -1.1044086932088248e-05 2023-01-23 20:27:55.977786: step: 1042/1541, loss: 0.0034780835267156363 2023-01-23 20:27:56.830519: step: 1044/1541, loss: 0.0034757531248033047 2023-01-23 20:27:57.716689: step: 1046/1541, loss: 0.027435192838311195 2023-01-23 20:27:58.631649: step: 1048/1541, loss: 0.0004178447707090527 2023-01-23 20:27:59.570343: step: 1050/1541, loss: 0.13658061623573303 2023-01-23 20:28:00.480448: step: 1052/1541, loss: -2.4054057575995103e-05 2023-01-23 20:28:01.388336: step: 1054/1541, loss: 0.008773989975452423 2023-01-23 20:28:02.306038: step: 1056/1541, loss: 0.0008752946741878986 2023-01-23 20:28:03.204054: step: 1058/1541, loss: 0.01348045002669096 2023-01-23 20:28:04.124432: step: 1060/1541, loss: 2.862532710423693e-05 2023-01-23 20:28:05.042278: step: 1062/1541, loss: 0.00925470795482397 2023-01-23 20:28:05.926037: step: 1064/1541, loss: 5.1943450671387836e-05 2023-01-23 20:28:06.826575: step: 1066/1541, loss: 0.0007633328787051141 2023-01-23 20:28:07.725203: step: 1068/1541, loss: 0.02158956229686737 2023-01-23 20:28:08.614699: step: 1070/1541, loss: -3.095830834354274e-06 2023-01-23 20:28:09.585958: step: 1072/1541, loss: 0.0006708358414471149 2023-01-23 20:28:10.564686: step: 1074/1541, loss: 6.604442751267925e-05 2023-01-23 20:28:11.465015: step: 1076/1541, loss: 0.0005242041661404073 2023-01-23 20:28:12.368162: step: 1078/1541, loss: 0.08983185142278671 2023-01-23 20:28:13.292985: step: 1080/1541, loss: 0.11479561030864716 2023-01-23 20:28:14.188392: step: 1082/1541, loss: 0.3795938491821289 2023-01-23 20:28:15.088428: step: 1084/1541, loss: 0.0010686920722946525 2023-01-23 20:28:15.989146: step: 1086/1541, loss: 0.004278452135622501 2023-01-23 20:28:16.925364: step: 1088/1541, loss: 0.00030463121947832406 2023-01-23 20:28:17.877656: step: 1090/1541, loss: 0.00010362563625676557 2023-01-23 20:28:18.828608: step: 1092/1541, loss: 0.04431629553437233 2023-01-23 20:28:19.674111: step: 1094/1541, loss: 0.12428224831819534 2023-01-23 20:28:20.528690: step: 1096/1541, loss: 0.0016743041342124343 2023-01-23 20:28:21.449021: step: 1098/1541, loss: 0.0002840412489604205 2023-01-23 20:28:22.309969: step: 1100/1541, loss: 3.898105205735192e-05 2023-01-23 20:28:23.162361: step: 1102/1541, loss: 0.0017336279852315784 2023-01-23 20:28:24.041921: step: 1104/1541, loss: 0.21729964017868042 2023-01-23 20:28:24.983610: step: 1106/1541, loss: 0.0023970366455614567 2023-01-23 20:28:25.868843: step: 1108/1541, loss: 0.15572647750377655 2023-01-23 20:28:26.832642: step: 1110/1541, loss: 0.000821294728666544 2023-01-23 20:28:27.751119: step: 1112/1541, loss: 0.002141367644071579 2023-01-23 20:28:28.602036: step: 1114/1541, loss: 0.013688678853213787 2023-01-23 20:28:29.474936: step: 1116/1541, loss: 0.0005498692626133561 2023-01-23 20:28:30.364687: step: 1118/1541, loss: 0.0023144353181123734 2023-01-23 20:28:31.231995: step: 1120/1541, loss: 0.0006291532190516591 2023-01-23 20:28:32.199244: step: 1122/1541, loss: 0.031680211424827576 2023-01-23 20:28:33.088508: step: 1124/1541, loss: 0.001180316903628409 2023-01-23 20:28:33.993308: step: 1126/1541, loss: 0.0004544087569229305 2023-01-23 20:28:34.960635: step: 1128/1541, loss: 1.1867723515024409e-05 2023-01-23 20:28:35.828734: step: 1130/1541, loss: 0.15829268097877502 2023-01-23 20:28:36.671470: step: 1132/1541, loss: 0.005381581839174032 2023-01-23 20:28:37.592770: step: 1134/1541, loss: 0.0006586898234672844 2023-01-23 20:28:38.462482: step: 1136/1541, loss: 0.655377209186554 2023-01-23 20:28:39.345962: step: 1138/1541, loss: 2.0208196929161204e-06 2023-01-23 20:28:40.202753: step: 1140/1541, loss: 0.0003358394023962319 2023-01-23 20:28:41.112208: step: 1142/1541, loss: 8.58804487506859e-05 2023-01-23 20:28:42.073016: step: 1144/1541, loss: -1.6374138795072213e-05 2023-01-23 20:28:42.965136: step: 1146/1541, loss: 0.07509604096412659 2023-01-23 20:28:43.882994: step: 1148/1541, loss: 0.2235143929719925 2023-01-23 20:28:44.810239: step: 1150/1541, loss: 0.005919259507209063 2023-01-23 20:28:45.707760: step: 1152/1541, loss: 0.26347634196281433 2023-01-23 20:28:46.576695: step: 1154/1541, loss: 0.21831756830215454 2023-01-23 20:28:47.495620: step: 1156/1541, loss: 0.0010935425525531173 2023-01-23 20:28:48.376922: step: 1158/1541, loss: 0.004022629931569099 2023-01-23 20:28:49.290336: step: 1160/1541, loss: 0.0023574901279062033 2023-01-23 20:28:50.208918: step: 1162/1541, loss: 0.011404970660805702 2023-01-23 20:28:51.103397: step: 1164/1541, loss: 0.0012866068864241242 2023-01-23 20:28:51.979396: step: 1166/1541, loss: 0.14939232170581818 2023-01-23 20:28:52.882905: step: 1168/1541, loss: 0.019635258242487907 2023-01-23 20:28:53.785655: step: 1170/1541, loss: -2.2882968551130034e-05 2023-01-23 20:28:54.669648: step: 1172/1541, loss: 0.00029037543572485447 2023-01-23 20:28:55.564084: step: 1174/1541, loss: 0.0045828926376998425 2023-01-23 20:28:56.542146: step: 1176/1541, loss: 0.00818694569170475 2023-01-23 20:28:57.506513: step: 1178/1541, loss: 0.02146092988550663 2023-01-23 20:28:58.418787: step: 1180/1541, loss: 7.154270861065015e-05 2023-01-23 20:28:59.319141: step: 1182/1541, loss: 0.002333403332158923 2023-01-23 20:29:00.212616: step: 1184/1541, loss: 0.008510718122124672 2023-01-23 20:29:01.144336: step: 1186/1541, loss: 0.0004833227430935949 2023-01-23 20:29:02.028880: step: 1188/1541, loss: 0.02634487859904766 2023-01-23 20:29:02.861779: step: 1190/1541, loss: 0.0007775568519718945 2023-01-23 20:29:03.757482: step: 1192/1541, loss: 0.0011407551355659962 2023-01-23 20:29:04.645378: step: 1194/1541, loss: 0.8882042765617371 2023-01-23 20:29:05.517124: step: 1196/1541, loss: 0.10282571613788605 2023-01-23 20:29:06.489274: step: 1198/1541, loss: 0.0009183805668726563 2023-01-23 20:29:07.418079: step: 1200/1541, loss: 0.010714491829276085 2023-01-23 20:29:08.363520: step: 1202/1541, loss: 0.0062899719923734665 2023-01-23 20:29:09.225471: step: 1204/1541, loss: 0.4629112780094147 2023-01-23 20:29:10.098223: step: 1206/1541, loss: 0.00019984447862952948 2023-01-23 20:29:11.027600: step: 1208/1541, loss: 0.09484866261482239 2023-01-23 20:29:11.913838: step: 1210/1541, loss: 0.009485688991844654 2023-01-23 20:29:12.811310: step: 1212/1541, loss: 0.001804903382435441 2023-01-23 20:29:13.734571: step: 1214/1541, loss: 0.7257403135299683 2023-01-23 20:29:14.665325: step: 1216/1541, loss: 0.07145953178405762 2023-01-23 20:29:15.527407: step: 1218/1541, loss: 0.0020592333748936653 2023-01-23 20:29:16.389153: step: 1220/1541, loss: 0.00021467191982083023 2023-01-23 20:29:17.283883: step: 1222/1541, loss: 0.2192053347826004 2023-01-23 20:29:18.144973: step: 1224/1541, loss: 0.00789803359657526 2023-01-23 20:29:19.014773: step: 1226/1541, loss: 4.122670361539349e-06 2023-01-23 20:29:19.929568: step: 1228/1541, loss: 0.0003771409683395177 2023-01-23 20:29:20.896942: step: 1230/1541, loss: 0.023044254630804062 2023-01-23 20:29:21.787462: step: 1232/1541, loss: 0.0001373140257783234 2023-01-23 20:29:22.671255: step: 1234/1541, loss: 0.1632205992937088 2023-01-23 20:29:23.655676: step: 1236/1541, loss: 0.00032984468271024525 2023-01-23 20:29:24.584645: step: 1238/1541, loss: 0.00020675337873399258 2023-01-23 20:29:25.462565: step: 1240/1541, loss: 0.00245766737498343 2023-01-23 20:29:26.368778: step: 1242/1541, loss: 0.0012542289914563298 2023-01-23 20:29:27.268689: step: 1244/1541, loss: 0.24265117943286896 2023-01-23 20:29:28.244546: step: 1246/1541, loss: 0.002067611087113619 2023-01-23 20:29:29.139389: step: 1248/1541, loss: 0.183873251080513 2023-01-23 20:29:30.026884: step: 1250/1541, loss: 9.569898247718811e-05 2023-01-23 20:29:30.876337: step: 1252/1541, loss: 0.005145070143043995 2023-01-23 20:29:31.772505: step: 1254/1541, loss: 0.0008030590834096074 2023-01-23 20:29:32.768672: step: 1256/1541, loss: 0.0016243810532614589 2023-01-23 20:29:33.654380: step: 1258/1541, loss: 0.11231909692287445 2023-01-23 20:29:34.590053: step: 1260/1541, loss: 0.03357161581516266 2023-01-23 20:29:35.557717: step: 1262/1541, loss: 0.09958113729953766 2023-01-23 20:29:36.435285: step: 1264/1541, loss: 0.00027304625837132335 2023-01-23 20:29:37.316993: step: 1266/1541, loss: 0.012822090648114681 2023-01-23 20:29:38.222521: step: 1268/1541, loss: 0.006421986967325211 2023-01-23 20:29:39.141354: step: 1270/1541, loss: 0.003411621553823352 2023-01-23 20:29:40.088567: step: 1272/1541, loss: 0.11282313615083694 2023-01-23 20:29:40.981756: step: 1274/1541, loss: 0.026336969807744026 2023-01-23 20:29:41.870751: step: 1276/1541, loss: 0.04532705992460251 2023-01-23 20:29:42.878695: step: 1278/1541, loss: 0.0024060355499386787 2023-01-23 20:29:43.754726: step: 1280/1541, loss: 0.0002284282527398318 2023-01-23 20:29:44.644586: step: 1282/1541, loss: 0.16054223477840424 2023-01-23 20:29:45.586621: step: 1284/1541, loss: 0.004919599276036024 2023-01-23 20:29:46.503811: step: 1286/1541, loss: 0.000354752060957253 2023-01-23 20:29:47.395877: step: 1288/1541, loss: 0.0003937861474696547 2023-01-23 20:29:48.371493: step: 1290/1541, loss: 0.0012024124152958393 2023-01-23 20:29:49.299300: step: 1292/1541, loss: 0.0005252439295873046 2023-01-23 20:29:50.176249: step: 1294/1541, loss: 0.13961029052734375 2023-01-23 20:29:51.074506: step: 1296/1541, loss: 0.6105654239654541 2023-01-23 20:29:51.978799: step: 1298/1541, loss: 0.007216195110231638 2023-01-23 20:29:52.821785: step: 1300/1541, loss: 0.0007420220063067973 2023-01-23 20:29:53.706080: step: 1302/1541, loss: 0.049577876925468445 2023-01-23 20:29:54.642467: step: 1304/1541, loss: 0.003182291053235531 2023-01-23 20:29:55.624634: step: 1306/1541, loss: 0.23016108572483063 2023-01-23 20:29:56.498501: step: 1308/1541, loss: 0.049587640911340714 2023-01-23 20:29:57.378450: step: 1310/1541, loss: 0.0030829552561044693 2023-01-23 20:29:58.253979: step: 1312/1541, loss: 0.0012501697055995464 2023-01-23 20:29:59.234540: step: 1314/1541, loss: 0.015552465803921223 2023-01-23 20:30:00.220247: step: 1316/1541, loss: 0.021767016500234604 2023-01-23 20:30:01.111587: step: 1318/1541, loss: 0.11586932092905045 2023-01-23 20:30:02.008575: step: 1320/1541, loss: 0.03094916045665741 2023-01-23 20:30:02.940223: step: 1322/1541, loss: 0.015098600648343563 2023-01-23 20:30:03.826216: step: 1324/1541, loss: 0.0063714864663779736 2023-01-23 20:30:04.757552: step: 1326/1541, loss: 0.0024302725214511156 2023-01-23 20:30:05.621893: step: 1328/1541, loss: 0.0018695970065891743 2023-01-23 20:30:06.529493: step: 1330/1541, loss: 0.0037644035182893276 2023-01-23 20:30:07.444709: step: 1332/1541, loss: 0.012405658140778542 2023-01-23 20:30:08.339530: step: 1334/1541, loss: 0.00258970120921731 2023-01-23 20:30:09.222262: step: 1336/1541, loss: 0.0003312679473310709 2023-01-23 20:30:10.048880: step: 1338/1541, loss: 0.026411738246679306 2023-01-23 20:30:10.977108: step: 1340/1541, loss: 0.08551593869924545 2023-01-23 20:30:11.869137: step: 1342/1541, loss: 0.008649986237287521 2023-01-23 20:30:12.780351: step: 1344/1541, loss: 0.3682907223701477 2023-01-23 20:30:13.646753: step: 1346/1541, loss: 0.007555790711194277 2023-01-23 20:30:14.528099: step: 1348/1541, loss: 0.0017315183067694306 2023-01-23 20:30:15.464873: step: 1350/1541, loss: 0.0007720692083239555 2023-01-23 20:30:16.441401: step: 1352/1541, loss: 0.0012329984456300735 2023-01-23 20:30:17.327315: step: 1354/1541, loss: 0.0005023017292842269 2023-01-23 20:30:18.271443: step: 1356/1541, loss: 0.0011267387308180332 2023-01-23 20:30:19.149010: step: 1358/1541, loss: 0.3404217064380646 2023-01-23 20:30:20.036784: step: 1360/1541, loss: 0.010392152704298496 2023-01-23 20:30:20.948130: step: 1362/1541, loss: 0.17716850340366364 2023-01-23 20:30:21.822107: step: 1364/1541, loss: 0.24746327102184296 2023-01-23 20:30:22.795545: step: 1366/1541, loss: 0.000931429909542203 2023-01-23 20:30:23.669391: step: 1368/1541, loss: 0.009582188911736012 2023-01-23 20:30:24.512269: step: 1370/1541, loss: 1.683563232421875 2023-01-23 20:30:25.401651: step: 1372/1541, loss: 0.00937632191926241 2023-01-23 20:30:26.303706: step: 1374/1541, loss: 0.14497072994709015 2023-01-23 20:30:27.198513: step: 1376/1541, loss: 0.15808023512363434 2023-01-23 20:30:28.080417: step: 1378/1541, loss: 0.06888014823198318 2023-01-23 20:30:28.939151: step: 1380/1541, loss: 0.010813998058438301 2023-01-23 20:30:29.831834: step: 1382/1541, loss: 9.148152457782999e-05 2023-01-23 20:30:30.769831: step: 1384/1541, loss: 0.000756696448661387 2023-01-23 20:30:31.656754: step: 1386/1541, loss: 0.002656793687492609 2023-01-23 20:30:32.613261: step: 1388/1541, loss: 0.001622985117137432 2023-01-23 20:30:33.547673: step: 1390/1541, loss: 0.0006165202939882874 2023-01-23 20:30:34.479365: step: 1392/1541, loss: 0.3945455551147461 2023-01-23 20:30:35.405512: step: 1394/1541, loss: 0.0011561298742890358 2023-01-23 20:30:36.252740: step: 1396/1541, loss: 0.00014225892664398998 2023-01-23 20:30:37.110331: step: 1398/1541, loss: 0.0016395747661590576 2023-01-23 20:30:38.031539: step: 1400/1541, loss: 0.0008687402587383986 2023-01-23 20:30:38.953458: step: 1402/1541, loss: 0.13200941681861877 2023-01-23 20:30:39.876992: step: 1404/1541, loss: 0.14576490223407745 2023-01-23 20:30:40.783795: step: 1406/1541, loss: 0.2484939992427826 2023-01-23 20:30:41.679956: step: 1408/1541, loss: 0.0006056373240426183 2023-01-23 20:30:42.596082: step: 1410/1541, loss: 0.00011176000407431275 2023-01-23 20:30:43.517994: step: 1412/1541, loss: 0.18028806149959564 2023-01-23 20:30:44.404558: step: 1414/1541, loss: 9.110481914831325e-05 2023-01-23 20:30:45.287447: step: 1416/1541, loss: 0.0778321847319603 2023-01-23 20:30:46.188266: step: 1418/1541, loss: 0.12191271036863327 2023-01-23 20:30:47.131399: step: 1420/1541, loss: 0.003720622044056654 2023-01-23 20:30:48.016104: step: 1422/1541, loss: 0.0001456235331716016 2023-01-23 20:30:48.915093: step: 1424/1541, loss: 0.05070500448346138 2023-01-23 20:30:49.782848: step: 1426/1541, loss: 0.0002658904704730958 2023-01-23 20:30:50.638246: step: 1428/1541, loss: 0.0002517219400033355 2023-01-23 20:30:51.537626: step: 1430/1541, loss: 0.32436951994895935 2023-01-23 20:30:52.459840: step: 1432/1541, loss: 0.045972879976034164 2023-01-23 20:30:53.340451: step: 1434/1541, loss: 0.003998016007244587 2023-01-23 20:30:54.274249: step: 1436/1541, loss: 0.0034549448173493147 2023-01-23 20:30:55.237473: step: 1438/1541, loss: 6.929765368113294e-05 2023-01-23 20:30:56.158492: step: 1440/1541, loss: 0.001768546411767602 2023-01-23 20:30:57.052349: step: 1442/1541, loss: 0.45232856273651123 2023-01-23 20:30:57.972280: step: 1444/1541, loss: 0.032290730625391006 2023-01-23 20:30:58.952840: step: 1446/1541, loss: 0.00016850177780725062 2023-01-23 20:30:59.841954: step: 1448/1541, loss: 0.028213277459144592 2023-01-23 20:31:00.744249: step: 1450/1541, loss: 0.3117856979370117 2023-01-23 20:31:01.709656: step: 1452/1541, loss: 0.0028601207304745913 2023-01-23 20:31:02.684254: step: 1454/1541, loss: 0.1553610861301422 2023-01-23 20:31:03.574351: step: 1456/1541, loss: 0.3852808475494385 2023-01-23 20:31:04.464393: step: 1458/1541, loss: 0.0007385039934888482 2023-01-23 20:31:05.328570: step: 1460/1541, loss: 0.0035567055456340313 2023-01-23 20:31:06.279416: step: 1462/1541, loss: 0.02796165831387043 2023-01-23 20:31:07.149192: step: 1464/1541, loss: 0.0016969945281744003 2023-01-23 20:31:08.015523: step: 1466/1541, loss: 0.0006564957438968122 2023-01-23 20:31:08.870656: step: 1468/1541, loss: 0.002751338528469205 2023-01-23 20:31:09.739695: step: 1470/1541, loss: 0.0006794023793190718 2023-01-23 20:31:10.634883: step: 1472/1541, loss: 0.0025783348828554153 2023-01-23 20:31:11.538337: step: 1474/1541, loss: 0.1488138884305954 2023-01-23 20:31:12.405726: step: 1476/1541, loss: 0.3948831260204315 2023-01-23 20:31:13.395570: step: 1478/1541, loss: 0.007810015231370926 2023-01-23 20:31:14.256341: step: 1480/1541, loss: 3.692110112751834e-05 2023-01-23 20:31:15.158665: step: 1482/1541, loss: 0.0011376463808119297 2023-01-23 20:31:16.076874: step: 1484/1541, loss: 0.019511429592967033 2023-01-23 20:31:17.024863: step: 1486/1541, loss: 0.0023971786722540855 2023-01-23 20:31:17.967861: step: 1488/1541, loss: 0.00047631142660975456 2023-01-23 20:31:18.876592: step: 1490/1541, loss: 0.026113459840416908 2023-01-23 20:31:19.769295: step: 1492/1541, loss: 0.06457991153001785 2023-01-23 20:31:20.670793: step: 1494/1541, loss: 0.07185173779726028 2023-01-23 20:31:21.563349: step: 1496/1541, loss: 0.0011221268214285374 2023-01-23 20:31:22.446241: step: 1498/1541, loss: 0.0061850473284721375 2023-01-23 20:31:23.328020: step: 1500/1541, loss: 0.040516216307878494 2023-01-23 20:31:24.198139: step: 1502/1541, loss: 0.0010779007570818067 2023-01-23 20:31:25.079052: step: 1504/1541, loss: 0.09050111472606659 2023-01-23 20:31:25.936037: step: 1506/1541, loss: 8.930977492127568e-05 2023-01-23 20:31:26.817958: step: 1508/1541, loss: 0.004136964678764343 2023-01-23 20:31:27.698633: step: 1510/1541, loss: 0.0015944677870720625 2023-01-23 20:31:28.598368: step: 1512/1541, loss: 0.057790111750364304 2023-01-23 20:31:29.466556: step: 1514/1541, loss: 0.005435428582131863 2023-01-23 20:31:30.349710: step: 1516/1541, loss: 0.000216678308788687 2023-01-23 20:31:31.258793: step: 1518/1541, loss: 0.020356930792331696 2023-01-23 20:31:32.156692: step: 1520/1541, loss: 0.006932556163519621 2023-01-23 20:31:33.066121: step: 1522/1541, loss: 0.5337927937507629 2023-01-23 20:31:33.946340: step: 1524/1541, loss: 8.512308704666793e-05 2023-01-23 20:31:34.861893: step: 1526/1541, loss: 0.2315802276134491 2023-01-23 20:31:35.775460: step: 1528/1541, loss: 0.00047470146091654897 2023-01-23 20:31:36.675737: step: 1530/1541, loss: 0.001688829273916781 2023-01-23 20:31:37.581175: step: 1532/1541, loss: 0.0003766710578929633 2023-01-23 20:31:38.469213: step: 1534/1541, loss: 0.045672763139009476 2023-01-23 20:31:39.387647: step: 1536/1541, loss: 0.00035894548636861145 2023-01-23 20:31:40.271597: step: 1538/1541, loss: 0.007587975822389126 2023-01-23 20:31:41.143922: step: 1540/1541, loss: 0.001102966838516295 2023-01-23 20:31:42.052398: step: 1542/1541, loss: 0.8748167157173157 2023-01-23 20:31:42.946620: step: 1544/1541, loss: 0.002757895737886429 2023-01-23 20:31:43.838356: step: 1546/1541, loss: 0.0010830272221937776 2023-01-23 20:31:44.719959: step: 1548/1541, loss: 0.00017568276962265372 2023-01-23 20:31:45.579455: step: 1550/1541, loss: 9.52371337916702e-05 2023-01-23 20:31:46.509335: step: 1552/1541, loss: 0.0014388811541721225 2023-01-23 20:31:47.462061: step: 1554/1541, loss: 0.0006298029329627752 2023-01-23 20:31:48.354249: step: 1556/1541, loss: 0.002054489217698574 2023-01-23 20:31:49.272589: step: 1558/1541, loss: -3.356757588335313e-05 2023-01-23 20:31:50.116808: step: 1560/1541, loss: 0.00010837092850124463 2023-01-23 20:31:50.966390: step: 1562/1541, loss: 0.00024007540196180344 2023-01-23 20:31:51.836620: step: 1564/1541, loss: 0.00023153347137849778 2023-01-23 20:31:52.746952: step: 1566/1541, loss: 8.9483437477611e-05 2023-01-23 20:31:53.671218: step: 1568/1541, loss: 0.08059622347354889 2023-01-23 20:31:54.551653: step: 1570/1541, loss: 0.0029987625312060118 2023-01-23 20:31:55.452013: step: 1572/1541, loss: 0.00012701604282483459 2023-01-23 20:31:56.383509: step: 1574/1541, loss: 0.14716961979866028 2023-01-23 20:31:57.266085: step: 1576/1541, loss: 0.008640352636575699 2023-01-23 20:31:58.186549: step: 1578/1541, loss: 1.967298703675624e-05 2023-01-23 20:31:59.080408: step: 1580/1541, loss: 8.101973071461543e-05 2023-01-23 20:32:00.009474: step: 1582/1541, loss: 0.0003529535315465182 2023-01-23 20:32:00.930857: step: 1584/1541, loss: -7.2941297730722e-06 2023-01-23 20:32:01.870698: step: 1586/1541, loss: 0.03650623559951782 2023-01-23 20:32:02.813708: step: 1588/1541, loss: 0.00015625923697371036 2023-01-23 20:32:03.814289: step: 1590/1541, loss: 0.005050370469689369 2023-01-23 20:32:04.742693: step: 1592/1541, loss: 1.3301594663062133e-05 2023-01-23 20:32:05.649687: step: 1594/1541, loss: 0.003161761909723282 2023-01-23 20:32:06.551114: step: 1596/1541, loss: 0.0017998849507421255 2023-01-23 20:32:07.506453: step: 1598/1541, loss: 0.00022385446936823428 2023-01-23 20:32:08.406739: step: 1600/1541, loss: 0.029953381046652794 2023-01-23 20:32:09.326171: step: 1602/1541, loss: 8.046964649111032e-05 2023-01-23 20:32:10.190041: step: 1604/1541, loss: 0.001130463439039886 2023-01-23 20:32:11.063733: step: 1606/1541, loss: 0.18729443848133087 2023-01-23 20:32:11.922906: step: 1608/1541, loss: 0.0001652335049584508 2023-01-23 20:32:12.785926: step: 1610/1541, loss: 6.856116669951007e-07 2023-01-23 20:32:13.647250: step: 1612/1541, loss: 0.0002797315828502178 2023-01-23 20:32:14.561662: step: 1614/1541, loss: 0.00044552164035849273 2023-01-23 20:32:15.447294: step: 1616/1541, loss: 0.16096380352973938 2023-01-23 20:32:16.325502: step: 1618/1541, loss: 0.009777815081179142 2023-01-23 20:32:17.234033: step: 1620/1541, loss: 0.00016525287355761975 2023-01-23 20:32:18.082359: step: 1622/1541, loss: 3.249491419410333e-05 2023-01-23 20:32:19.016377: step: 1624/1541, loss: 0.057373419404029846 2023-01-23 20:32:19.935057: step: 1626/1541, loss: 4.711411747848615e-05 2023-01-23 20:32:20.813741: step: 1628/1541, loss: 0.1904723048210144 2023-01-23 20:32:21.694849: step: 1630/1541, loss: 0.004491833504289389 2023-01-23 20:32:22.578600: step: 1632/1541, loss: 0.08229263871908188 2023-01-23 20:32:23.466059: step: 1634/1541, loss: 0.0033268616534769535 2023-01-23 20:32:24.318393: step: 1636/1541, loss: 0.0022193340118974447 2023-01-23 20:32:25.243172: step: 1638/1541, loss: 0.0020302035845816135 2023-01-23 20:32:26.151061: step: 1640/1541, loss: 0.0007128375000320375 2023-01-23 20:32:27.038134: step: 1642/1541, loss: 0.04487147927284241 2023-01-23 20:32:27.939622: step: 1644/1541, loss: 0.017971865832805634 2023-01-23 20:32:28.812711: step: 1646/1541, loss: 0.0009505946654826403 2023-01-23 20:32:29.704667: step: 1648/1541, loss: 0.005272937007248402 2023-01-23 20:32:30.578197: step: 1650/1541, loss: 3.414874299778603e-05 2023-01-23 20:32:31.520274: step: 1652/1541, loss: 0.011118771508336067 2023-01-23 20:32:32.438317: step: 1654/1541, loss: 0.05803602188825607 2023-01-23 20:32:33.359371: step: 1656/1541, loss: 0.1825578510761261 2023-01-23 20:32:34.237914: step: 1658/1541, loss: 0.002173356944695115 2023-01-23 20:32:35.139374: step: 1660/1541, loss: 0.0006135970470495522 2023-01-23 20:32:36.067919: step: 1662/1541, loss: 0.003591272048652172 2023-01-23 20:32:36.932112: step: 1664/1541, loss: 0.003183837281540036 2023-01-23 20:32:37.788776: step: 1666/1541, loss: 6.8533453941345215 2023-01-23 20:32:38.651693: step: 1668/1541, loss: 0.012211622670292854 2023-01-23 20:32:39.530109: step: 1670/1541, loss: 0.012406655587255955 2023-01-23 20:32:40.404713: step: 1672/1541, loss: 0.00042347965063527226 2023-01-23 20:32:41.311909: step: 1674/1541, loss: 0.0005719056935049593 2023-01-23 20:32:42.178821: step: 1676/1541, loss: 0.0328974574804306 2023-01-23 20:32:43.085654: step: 1678/1541, loss: 0.01230250671505928 2023-01-23 20:32:43.985453: step: 1680/1541, loss: 0.46570858359336853 2023-01-23 20:32:44.904207: step: 1682/1541, loss: 0.018855692818760872 2023-01-23 20:32:45.804355: step: 1684/1541, loss: 0.0009347524028271437 2023-01-23 20:32:46.686933: step: 1686/1541, loss: 0.047096654772758484 2023-01-23 20:32:47.578582: step: 1688/1541, loss: 0.0016161322128027678 2023-01-23 20:32:48.456546: step: 1690/1541, loss: 0.00018267333507537842 2023-01-23 20:32:49.299309: step: 1692/1541, loss: 5.6261924328282475e-05 2023-01-23 20:32:50.163678: step: 1694/1541, loss: 0.00014946261944714934 2023-01-23 20:32:51.070854: step: 1696/1541, loss: 0.0010202592238783836 2023-01-23 20:32:51.992259: step: 1698/1541, loss: 0.026662103831768036 2023-01-23 20:32:52.858088: step: 1700/1541, loss: 0.010970844887197018 2023-01-23 20:32:53.743835: step: 1702/1541, loss: 0.00013201333058532327 2023-01-23 20:32:54.653339: step: 1704/1541, loss: -2.670408684934955e-05 2023-01-23 20:32:55.619052: step: 1706/1541, loss: 0.020399965345859528 2023-01-23 20:32:56.521467: step: 1708/1541, loss: 0.0003532195114530623 2023-01-23 20:32:57.395546: step: 1710/1541, loss: 0.00014208229549694806 2023-01-23 20:32:58.317182: step: 1712/1541, loss: 0.011699787341058254 2023-01-23 20:32:59.214484: step: 1714/1541, loss: 0.013149290345609188 2023-01-23 20:33:00.105685: step: 1716/1541, loss: 0.02435847744345665 2023-01-23 20:33:00.982353: step: 1718/1541, loss: 0.0004359629238024354 2023-01-23 20:33:01.949422: step: 1720/1541, loss: 0.45202675461769104 2023-01-23 20:33:02.820479: step: 1722/1541, loss: 0.0002886352885980159 2023-01-23 20:33:03.713523: step: 1724/1541, loss: 0.003538097022101283 2023-01-23 20:33:04.623218: step: 1726/1541, loss: 0.026178620755672455 2023-01-23 20:33:05.502166: step: 1728/1541, loss: 0.1857801228761673 2023-01-23 20:33:06.392325: step: 1730/1541, loss: 0.5751001834869385 2023-01-23 20:33:07.350685: step: 1732/1541, loss: 0.00015449835336767137 2023-01-23 20:33:08.238564: step: 1734/1541, loss: -1.65464971360052e-05 2023-01-23 20:33:09.120264: step: 1736/1541, loss: 0.000794659077655524 2023-01-23 20:33:10.020995: step: 1738/1541, loss: 2.5323006411781535e-05 2023-01-23 20:33:10.905422: step: 1740/1541, loss: 0.00043743313290178776 2023-01-23 20:33:11.790073: step: 1742/1541, loss: 0.00036688512773253024 2023-01-23 20:33:12.681166: step: 1744/1541, loss: 0.046444300562143326 2023-01-23 20:33:13.553475: step: 1746/1541, loss: 0.43704459071159363 2023-01-23 20:33:14.474292: step: 1748/1541, loss: 0.0002860378008335829 2023-01-23 20:33:15.348677: step: 1750/1541, loss: 0.0013885843800380826 2023-01-23 20:33:16.260260: step: 1752/1541, loss: 0.0011850765440613031 2023-01-23 20:33:17.142777: step: 1754/1541, loss: 0.19221064448356628 2023-01-23 20:33:18.033604: step: 1756/1541, loss: 0.08004844188690186 2023-01-23 20:33:18.875087: step: 1758/1541, loss: 0.00017695028509479016 2023-01-23 20:33:19.751331: step: 1760/1541, loss: 0.004926144611090422 2023-01-23 20:33:20.721167: step: 1762/1541, loss: 0.00039014281355775893 2023-01-23 20:33:21.600435: step: 1764/1541, loss: 0.00019390371744520962 2023-01-23 20:33:22.522890: step: 1766/1541, loss: 0.027042442932724953 2023-01-23 20:33:23.397172: step: 1768/1541, loss: 0.28129327297210693 2023-01-23 20:33:24.238227: step: 1770/1541, loss: 0.00016138031787704676 2023-01-23 20:33:25.133340: step: 1772/1541, loss: 2.1034869860159233e-05 2023-01-23 20:33:26.042328: step: 1774/1541, loss: 0.0003997732128482312 2023-01-23 20:33:26.940270: step: 1776/1541, loss: 0.0011700670002028346 2023-01-23 20:33:27.815810: step: 1778/1541, loss: 0.00016282216529361904 2023-01-23 20:33:28.723567: step: 1780/1541, loss: 5.2557796152541414e-05 2023-01-23 20:33:29.594666: step: 1782/1541, loss: 0.06488345563411713 2023-01-23 20:33:30.530215: step: 1784/1541, loss: 8.545279706595466e-05 2023-01-23 20:33:31.425439: step: 1786/1541, loss: 0.04457515850663185 2023-01-23 20:33:32.413306: step: 1788/1541, loss: 0.37080368399620056 2023-01-23 20:33:33.285350: step: 1790/1541, loss: 0.0008532918291166425 2023-01-23 20:33:34.300472: step: 1792/1541, loss: 0.037678103893995285 2023-01-23 20:33:35.228937: step: 1794/1541, loss: 0.0005315780290402472 2023-01-23 20:33:36.082786: step: 1796/1541, loss: 0.011720619164407253 2023-01-23 20:33:36.986407: step: 1798/1541, loss: 1.4859307157166768e-05 2023-01-23 20:33:37.983680: step: 1800/1541, loss: 0.18261630833148956 2023-01-23 20:33:38.886029: step: 1802/1541, loss: -8.585781324654818e-05 2023-01-23 20:33:39.766285: step: 1804/1541, loss: 2.4672412109794095e-05 2023-01-23 20:33:40.660947: step: 1806/1541, loss: 0.006224098615348339 2023-01-23 20:33:41.581389: step: 1808/1541, loss: 0.00037727842573076487 2023-01-23 20:33:42.592110: step: 1810/1541, loss: 0.00022667087614536285 2023-01-23 20:33:43.497369: step: 1812/1541, loss: 0.11732850968837738 2023-01-23 20:33:44.382242: step: 1814/1541, loss: 0.0024862971622496843 2023-01-23 20:33:45.251880: step: 1816/1541, loss: 0.022130504250526428 2023-01-23 20:33:46.135152: step: 1818/1541, loss: -4.292456651455723e-05 2023-01-23 20:33:46.993998: step: 1820/1541, loss: 6.934187695151195e-07 2023-01-23 20:33:47.874526: step: 1822/1541, loss: 0.0027839550748467445 2023-01-23 20:33:48.769285: step: 1824/1541, loss: 9.276165656046942e-05 2023-01-23 20:33:49.687619: step: 1826/1541, loss: 0.01649361103773117 2023-01-23 20:33:50.578135: step: 1828/1541, loss: 0.6799137592315674 2023-01-23 20:33:51.456704: step: 1830/1541, loss: 0.059080369770526886 2023-01-23 20:33:52.323824: step: 1832/1541, loss: 0.24308156967163086 2023-01-23 20:33:53.212369: step: 1834/1541, loss: 0.0015145880170166492 2023-01-23 20:33:54.113016: step: 1836/1541, loss: 0.08578111231327057 2023-01-23 20:33:55.040945: step: 1838/1541, loss: 0.009261930361390114 2023-01-23 20:33:55.939768: step: 1840/1541, loss: -1.698835512797814e-05 2023-01-23 20:33:56.769499: step: 1842/1541, loss: 4.081104998476803e-05 2023-01-23 20:33:57.647990: step: 1844/1541, loss: 0.06588970869779587 2023-01-23 20:33:58.517782: step: 1846/1541, loss: 0.0022632181644439697 2023-01-23 20:33:59.446734: step: 1848/1541, loss: -1.315996178163914e-05 2023-01-23 20:34:00.325920: step: 1850/1541, loss: 0.00019206108117941767 2023-01-23 20:34:01.260807: step: 1852/1541, loss: 0.00010459981422172859 2023-01-23 20:34:02.208954: step: 1854/1541, loss: 8.30938370199874e-05 2023-01-23 20:34:03.096031: step: 1856/1541, loss: 0.02953367307782173 2023-01-23 20:34:03.974639: step: 1858/1541, loss: 0.030666982755064964 2023-01-23 20:34:04.862343: step: 1860/1541, loss: 0.008787202648818493 2023-01-23 20:34:05.721999: step: 1862/1541, loss: 8.775426249485463e-05 2023-01-23 20:34:06.605992: step: 1864/1541, loss: 0.012591746635735035 2023-01-23 20:34:07.474271: step: 1866/1541, loss: 0.0033830651082098484 2023-01-23 20:34:08.363747: step: 1868/1541, loss: 0.0006149904220364988 2023-01-23 20:34:09.238088: step: 1870/1541, loss: 2.815420157276094e-05 2023-01-23 20:34:10.105811: step: 1872/1541, loss: 0.0026895070914179087 2023-01-23 20:34:11.094222: step: 1874/1541, loss: 0.02624945156276226 2023-01-23 20:34:12.053250: step: 1876/1541, loss: 0.0008588299388065934 2023-01-23 20:34:12.902519: step: 1878/1541, loss: 0.018146423622965813 2023-01-23 20:34:13.848208: step: 1880/1541, loss: 0.0006698824581690133 2023-01-23 20:34:14.759153: step: 1882/1541, loss: 0.003423302434384823 2023-01-23 20:34:15.647491: step: 1884/1541, loss: 0.16912856698036194 2023-01-23 20:34:16.550651: step: 1886/1541, loss: 0.006807517725974321 2023-01-23 20:34:17.476351: step: 1888/1541, loss: 0.020481403917074203 2023-01-23 20:34:18.379025: step: 1890/1541, loss: 0.00032320411992259324 2023-01-23 20:34:19.250709: step: 1892/1541, loss: 0.005024343263357878 2023-01-23 20:34:20.151191: step: 1894/1541, loss: 0.013257242739200592 2023-01-23 20:34:21.072823: step: 1896/1541, loss: 0.00026337214512750506 2023-01-23 20:34:21.935706: step: 1898/1541, loss: 0.010427944362163544 2023-01-23 20:34:22.804880: step: 1900/1541, loss: 0.0065169730223715305 2023-01-23 20:34:23.677403: step: 1902/1541, loss: 0.1320117563009262 2023-01-23 20:34:24.558003: step: 1904/1541, loss: 0.020101375877857208 2023-01-23 20:34:25.461568: step: 1906/1541, loss: 0.3276626169681549 2023-01-23 20:34:26.387275: step: 1908/1541, loss: 0.00040186167461797595 2023-01-23 20:34:27.310719: step: 1910/1541, loss: -9.94629408523906e-06 2023-01-23 20:34:28.210383: step: 1912/1541, loss: 0.0017523816786706448 2023-01-23 20:34:29.095772: step: 1914/1541, loss: 0.06996075063943863 2023-01-23 20:34:30.031345: step: 1916/1541, loss: 0.00014951074263080955 2023-01-23 20:34:31.014092: step: 1918/1541, loss: 0.00015485806216020137 2023-01-23 20:34:31.928101: step: 1920/1541, loss: 0.0017732130363583565 2023-01-23 20:34:32.879328: step: 1922/1541, loss: 0.3661891222000122 2023-01-23 20:34:33.809396: step: 1924/1541, loss: 0.766818642616272 2023-01-23 20:34:34.816940: step: 1926/1541, loss: 0.00036922909202985466 2023-01-23 20:34:35.719218: step: 1928/1541, loss: 0.0005578272975981236 2023-01-23 20:34:36.587474: step: 1930/1541, loss: 0.00012874501408077776 2023-01-23 20:34:37.472678: step: 1932/1541, loss: 0.0005386636476032436 2023-01-23 20:34:38.341875: step: 1934/1541, loss: 0.0047758519649505615 2023-01-23 20:34:39.259040: step: 1936/1541, loss: 0.0006279017543420196 2023-01-23 20:34:40.214402: step: 1938/1541, loss: 0.0006796172237955034 2023-01-23 20:34:41.122213: step: 1940/1541, loss: 0.0020214987453073263 2023-01-23 20:34:42.007817: step: 1942/1541, loss: 0.001477795303799212 2023-01-23 20:34:42.871296: step: 1944/1541, loss: 6.583239155588672e-05 2023-01-23 20:34:43.790815: step: 1946/1541, loss: 0.08510057628154755 2023-01-23 20:34:44.670333: step: 1948/1541, loss: 0.0013215318322181702 2023-01-23 20:34:45.568956: step: 1950/1541, loss: 0.0008213178953155875 2023-01-23 20:34:46.522282: step: 1952/1541, loss: 0.0010012610582634807 2023-01-23 20:34:47.440229: step: 1954/1541, loss: 0.0005025164573453367 2023-01-23 20:34:48.331887: step: 1956/1541, loss: 0.00043889755033887923 2023-01-23 20:34:49.195333: step: 1958/1541, loss: 0.002910896437242627 2023-01-23 20:34:50.088136: step: 1960/1541, loss: 0.002120331861078739 2023-01-23 20:34:50.992611: step: 1962/1541, loss: 0.00016693832003511488 2023-01-23 20:34:51.881640: step: 1964/1541, loss: 0.000158899201778695 2023-01-23 20:34:52.811362: step: 1966/1541, loss: 0.0020281767938286066 2023-01-23 20:34:53.673186: step: 1968/1541, loss: 1.9194585547666065e-05 2023-01-23 20:34:54.525778: step: 1970/1541, loss: 0.0012854051310569048 2023-01-23 20:34:55.442637: step: 1972/1541, loss: 0.0005161215667612851 2023-01-23 20:34:56.337685: step: 1974/1541, loss: 0.026288634166121483 2023-01-23 20:34:57.236757: step: 1976/1541, loss: 4.2031526390928775e-05 2023-01-23 20:34:58.128074: step: 1978/1541, loss: 0.000804591691121459 2023-01-23 20:34:59.003401: step: 1980/1541, loss: 0.0069060781970620155 2023-01-23 20:34:59.946509: step: 1982/1541, loss: 0.003570426953956485 2023-01-23 20:35:00.844813: step: 1984/1541, loss: 0.23913682997226715 2023-01-23 20:35:01.763745: step: 1986/1541, loss: 0.012987266294658184 2023-01-23 20:35:02.648312: step: 1988/1541, loss: 0.002669594483450055 2023-01-23 20:35:03.535539: step: 1990/1541, loss: 0.12884238362312317 2023-01-23 20:35:04.439825: step: 1992/1541, loss: 0.07331059128046036 2023-01-23 20:35:05.364322: step: 1994/1541, loss: 9.468187636230141e-05 2023-01-23 20:35:06.269112: step: 1996/1541, loss: -1.673211772867944e-05 2023-01-23 20:35:07.150693: step: 1998/1541, loss: 7.154913328122348e-05 2023-01-23 20:35:08.024749: step: 2000/1541, loss: 0.0002429228916298598 2023-01-23 20:35:08.967165: step: 2002/1541, loss: 0.00031399395084008574 2023-01-23 20:35:09.841713: step: 2004/1541, loss: 4.653321229852736e-05 2023-01-23 20:35:10.716310: step: 2006/1541, loss: 0.0003727758303284645 2023-01-23 20:35:11.727806: step: 2008/1541, loss: 5.824751133332029e-05 2023-01-23 20:35:12.629954: step: 2010/1541, loss: 0.010762332007288933 2023-01-23 20:35:13.541387: step: 2012/1541, loss: 0.13917464017868042 2023-01-23 20:35:14.442766: step: 2014/1541, loss: 0.00016653173952363431 2023-01-23 20:35:15.308496: step: 2016/1541, loss: 0.0006692010210826993 2023-01-23 20:35:16.199884: step: 2018/1541, loss: 0.0006151358829811215 2023-01-23 20:35:17.103213: step: 2020/1541, loss: 0.00760282576084137 2023-01-23 20:35:17.987896: step: 2022/1541, loss: 0.00015240708307828754 2023-01-23 20:35:18.854773: step: 2024/1541, loss: 0.0008700324688106775 2023-01-23 20:35:19.749862: step: 2026/1541, loss: 0.00040856475243344903 2023-01-23 20:35:20.644419: step: 2028/1541, loss: 0.018528813496232033 2023-01-23 20:35:21.521352: step: 2030/1541, loss: 0.00022822761093266308 2023-01-23 20:35:22.405366: step: 2032/1541, loss: 0.00024439446860924363 2023-01-23 20:35:23.292024: step: 2034/1541, loss: 0.002358405152335763 2023-01-23 20:35:24.185105: step: 2036/1541, loss: 0.0009503602632321417 2023-01-23 20:35:25.107004: step: 2038/1541, loss: 0.04552142322063446 2023-01-23 20:35:25.989052: step: 2040/1541, loss: 0.0001599669922143221 2023-01-23 20:35:26.903300: step: 2042/1541, loss: 0.00015514435654040426 2023-01-23 20:35:27.795947: step: 2044/1541, loss: 0.07254724204540253 2023-01-23 20:35:28.788421: step: 2046/1541, loss: 4.186674777884036e-05 2023-01-23 20:35:29.763916: step: 2048/1541, loss: 0.03528952598571777 2023-01-23 20:35:30.639852: step: 2050/1541, loss: 0.006290025543421507 2023-01-23 20:35:31.543545: step: 2052/1541, loss: 0.04096014425158501 2023-01-23 20:35:32.407610: step: 2054/1541, loss: 1.572476321598515e-05 2023-01-23 20:35:33.298666: step: 2056/1541, loss: 0.0006821042043156922 2023-01-23 20:35:34.144850: step: 2058/1541, loss: 0.028168462216854095 2023-01-23 20:35:35.028614: step: 2060/1541, loss: 0.028033073991537094 2023-01-23 20:35:35.921865: step: 2062/1541, loss: 0.00024751678574830294 2023-01-23 20:35:36.809595: step: 2064/1541, loss: 0.0326320044696331 2023-01-23 20:35:37.658337: step: 2066/1541, loss: 0.052146442234516144 2023-01-23 20:35:38.533500: step: 2068/1541, loss: 0.05699435994029045 2023-01-23 20:35:39.442516: step: 2070/1541, loss: 0.002534252591431141 2023-01-23 20:35:40.355729: step: 2072/1541, loss: 0.0033919820562005043 2023-01-23 20:35:41.258730: step: 2074/1541, loss: 1.3543102741241455 2023-01-23 20:35:42.128691: step: 2076/1541, loss: 0.0006251645972952247 2023-01-23 20:35:42.997210: step: 2078/1541, loss: 0.030689416453242302 2023-01-23 20:35:43.950547: step: 2080/1541, loss: 0.0003340833354741335 2023-01-23 20:35:44.882276: step: 2082/1541, loss: 6.189977284520864e-05 2023-01-23 20:35:45.770044: step: 2084/1541, loss: 0.0001442716602468863 2023-01-23 20:35:46.655393: step: 2086/1541, loss: 0.06563056260347366 2023-01-23 20:35:47.532180: step: 2088/1541, loss: 0.0020980965346097946 2023-01-23 20:35:48.432273: step: 2090/1541, loss: 0.000927710032556206 2023-01-23 20:35:49.404235: step: 2092/1541, loss: 0.04328975826501846 2023-01-23 20:35:50.353100: step: 2094/1541, loss: 0.0003626657708082348 2023-01-23 20:35:51.290976: step: 2096/1541, loss: 0.03446965292096138 2023-01-23 20:35:52.186532: step: 2098/1541, loss: 0.00012607945245690644 2023-01-23 20:35:53.110180: step: 2100/1541, loss: 0.002883823122829199 2023-01-23 20:35:53.978687: step: 2102/1541, loss: 0.03260721638798714 2023-01-23 20:35:54.880046: step: 2104/1541, loss: 0.002009748015552759 2023-01-23 20:35:55.754312: step: 2106/1541, loss: 0.0006248726276680827 2023-01-23 20:35:56.611193: step: 2108/1541, loss: 0.3074079751968384 2023-01-23 20:35:57.463749: step: 2110/1541, loss: 0.00016023388889152557 2023-01-23 20:35:58.393304: step: 2112/1541, loss: 0.00010224703146377578 2023-01-23 20:35:59.263805: step: 2114/1541, loss: 0.011198819614946842 2023-01-23 20:36:00.133748: step: 2116/1541, loss: 0.0985424742102623 2023-01-23 20:36:01.024486: step: 2118/1541, loss: 0.0807553231716156 2023-01-23 20:36:01.928462: step: 2120/1541, loss: 0.002009832300245762 2023-01-23 20:36:02.808378: step: 2122/1541, loss: 0.03545719012618065 2023-01-23 20:36:03.665926: step: 2124/1541, loss: 0.009027186781167984 2023-01-23 20:36:04.589239: step: 2126/1541, loss: 0.23746052384376526 2023-01-23 20:36:05.504470: step: 2128/1541, loss: 0.00030135546694509685 2023-01-23 20:36:06.384742: step: 2130/1541, loss: 0.0003226588887628168 2023-01-23 20:36:07.233938: step: 2132/1541, loss: 2.2597834686166607e-05 2023-01-23 20:36:08.111649: step: 2134/1541, loss: 0.000366580905392766 2023-01-23 20:36:09.072122: step: 2136/1541, loss: 4.308580537326634e-05 2023-01-23 20:36:09.971716: step: 2138/1541, loss: 0.0007834862917661667 2023-01-23 20:36:10.835696: step: 2140/1541, loss: 0.00019571598386391997 2023-01-23 20:36:11.718786: step: 2142/1541, loss: 0.0029430475551635027 2023-01-23 20:36:12.582854: step: 2144/1541, loss: 0.0019676529336720705 2023-01-23 20:36:13.499641: step: 2146/1541, loss: 0.0005205277702771127 2023-01-23 20:36:14.414200: step: 2148/1541, loss: 0.0001484657113905996 2023-01-23 20:36:15.331849: step: 2150/1541, loss: 0.0013257772661745548 2023-01-23 20:36:16.238974: step: 2152/1541, loss: 0.0713030993938446 2023-01-23 20:36:17.150498: step: 2154/1541, loss: 0.030374588444828987 2023-01-23 20:36:18.060310: step: 2156/1541, loss: 0.0001012788707157597 2023-01-23 20:36:18.937009: step: 2158/1541, loss: 0.04876530542969704 2023-01-23 20:36:19.829684: step: 2160/1541, loss: 0.0021014816593378782 2023-01-23 20:36:20.683742: step: 2162/1541, loss: 0.0002480547409504652 2023-01-23 20:36:21.634433: step: 2164/1541, loss: 0.11844879388809204 2023-01-23 20:36:22.484265: step: 2166/1541, loss: 0.004729751031845808 2023-01-23 20:36:23.338085: step: 2168/1541, loss: 0.00015461981820408255 2023-01-23 20:36:24.257232: step: 2170/1541, loss: 0.03706074878573418 2023-01-23 20:36:25.127918: step: 2172/1541, loss: 0.10955441743135452 2023-01-23 20:36:25.994654: step: 2174/1541, loss: 0.00023567893367726356 2023-01-23 20:36:26.858127: step: 2176/1541, loss: 0.0004503987147472799 2023-01-23 20:36:27.764862: step: 2178/1541, loss: 0.07779793441295624 2023-01-23 20:36:28.666796: step: 2180/1541, loss: 0.03143680840730667 2023-01-23 20:36:29.537923: step: 2182/1541, loss: 0.017702247947454453 2023-01-23 20:36:30.464664: step: 2184/1541, loss: 0.1874057799577713 2023-01-23 20:36:31.392919: step: 2186/1541, loss: 0.00015365985746029764 2023-01-23 20:36:32.316705: step: 2188/1541, loss: 8.353272278327495e-06 2023-01-23 20:36:33.259916: step: 2190/1541, loss: 0.11140952259302139 2023-01-23 20:36:34.193001: step: 2192/1541, loss: 0.0001813586859498173 2023-01-23 20:36:35.096361: step: 2194/1541, loss: 6.306816067080945e-05 2023-01-23 20:36:35.991322: step: 2196/1541, loss: 0.00029964427812956274 2023-01-23 20:36:36.890626: step: 2198/1541, loss: 0.00029943554545752704 2023-01-23 20:36:37.765135: step: 2200/1541, loss: 0.0005355863249860704 2023-01-23 20:36:38.684783: step: 2202/1541, loss: 0.03825792670249939 2023-01-23 20:36:39.561506: step: 2204/1541, loss: 0.0008218280854634941 2023-01-23 20:36:40.468171: step: 2206/1541, loss: 0.0007201900007203221 2023-01-23 20:36:41.344141: step: 2208/1541, loss: 0.00024436612147837877 2023-01-23 20:36:42.243855: step: 2210/1541, loss: 0.00013989463332109153 2023-01-23 20:36:43.166750: step: 2212/1541, loss: 0.005885795224457979 2023-01-23 20:36:44.059862: step: 2214/1541, loss: 2.7736261472455226e-05 2023-01-23 20:36:44.937431: step: 2216/1541, loss: 0.00021729801665060222 2023-01-23 20:36:45.798186: step: 2218/1541, loss: 0.008851325139403343 2023-01-23 20:36:46.658324: step: 2220/1541, loss: 0.0005218583974055946 2023-01-23 20:36:47.582077: step: 2222/1541, loss: 0.0012528632069006562 2023-01-23 20:36:48.432151: step: 2224/1541, loss: 8.720355981495231e-05 2023-01-23 20:36:49.324370: step: 2226/1541, loss: 0.003122063586488366 2023-01-23 20:36:50.290136: step: 2228/1541, loss: 0.001711125485599041 2023-01-23 20:36:51.219456: step: 2230/1541, loss: 3.8892652810318395e-05 2023-01-23 20:36:52.096414: step: 2232/1541, loss: 0.0009655940812081099 2023-01-23 20:36:53.001600: step: 2234/1541, loss: 0.008924859575927258 2023-01-23 20:36:53.880356: step: 2236/1541, loss: 0.011832588352262974 2023-01-23 20:36:54.836975: step: 2238/1541, loss: 0.0075200749561190605 2023-01-23 20:36:55.752076: step: 2240/1541, loss: 0.004929418675601482 2023-01-23 20:36:56.618009: step: 2242/1541, loss: 0.011322950944304466 2023-01-23 20:36:57.528701: step: 2244/1541, loss: 0.00668665487319231 2023-01-23 20:36:58.407357: step: 2246/1541, loss: 0.0027480979915708303 2023-01-23 20:36:59.354065: step: 2248/1541, loss: 5.804254215036053e-06 2023-01-23 20:37:00.240406: step: 2250/1541, loss: 0.0037862665485590696 2023-01-23 20:37:01.226405: step: 2252/1541, loss: 7.12080072844401e-05 2023-01-23 20:37:02.147069: step: 2254/1541, loss: 0.00047934299800544977 2023-01-23 20:37:03.025747: step: 2256/1541, loss: 0.0012164011131972075 2023-01-23 20:37:03.875962: step: 2258/1541, loss: 0.0006625604000873864 2023-01-23 20:37:04.792175: step: 2260/1541, loss: 0.0028122244402766228 2023-01-23 20:37:05.674153: step: 2262/1541, loss: 0.00014019469381310046 2023-01-23 20:37:06.557833: step: 2264/1541, loss: 0.00018756717327050865 2023-01-23 20:37:07.460133: step: 2266/1541, loss: 0.004107109736651182 2023-01-23 20:37:08.391038: step: 2268/1541, loss: 0.11046569049358368 2023-01-23 20:37:09.311820: step: 2270/1541, loss: 0.023220714181661606 2023-01-23 20:37:10.232126: step: 2272/1541, loss: 0.00019393513503018767 2023-01-23 20:37:11.101432: step: 2274/1541, loss: 0.0004260822315700352 2023-01-23 20:37:11.996837: step: 2276/1541, loss: 0.008237740024924278 2023-01-23 20:37:12.845765: step: 2278/1541, loss: 0.0005306213861331344 2023-01-23 20:37:13.740137: step: 2280/1541, loss: 6.513347761938348e-05 2023-01-23 20:37:14.662577: step: 2282/1541, loss: 0.002013204153627157 2023-01-23 20:37:15.556723: step: 2284/1541, loss: 0.000725599704310298 2023-01-23 20:37:16.480289: step: 2286/1541, loss: 0.00048646292998455465 2023-01-23 20:37:17.313427: step: 2288/1541, loss: 0.20269322395324707 2023-01-23 20:37:18.211073: step: 2290/1541, loss: 0.016701877117156982 2023-01-23 20:37:19.148969: step: 2292/1541, loss: 0.010647818446159363 2023-01-23 20:37:20.092075: step: 2294/1541, loss: -1.6075049643404782e-05 2023-01-23 20:37:20.996054: step: 2296/1541, loss: 0.004409324377775192 2023-01-23 20:37:21.892850: step: 2298/1541, loss: 0.0004909735289402306 2023-01-23 20:37:22.795452: step: 2300/1541, loss: 0.450376957654953 2023-01-23 20:37:23.676848: step: 2302/1541, loss: 0.00021540746092796326 2023-01-23 20:37:24.543905: step: 2304/1541, loss: 0.0014480833197012544 2023-01-23 20:37:25.516676: step: 2306/1541, loss: 0.00170215277466923 2023-01-23 20:37:26.403093: step: 2308/1541, loss: 0.0031495820730924606 2023-01-23 20:37:27.337234: step: 2310/1541, loss: 0.01883126236498356 2023-01-23 20:37:28.227382: step: 2312/1541, loss: 0.0008027036674320698 2023-01-23 20:37:29.129187: step: 2314/1541, loss: 0.004527904558926821 2023-01-23 20:37:30.039982: step: 2316/1541, loss: 0.01625274494290352 2023-01-23 20:37:30.966971: step: 2318/1541, loss: 0.001147158327512443 2023-01-23 20:37:31.892356: step: 2320/1541, loss: 0.0004959393409080803 2023-01-23 20:37:32.740536: step: 2322/1541, loss: 0.0025131436996161938 2023-01-23 20:37:33.642547: step: 2324/1541, loss: 0.0918908417224884 2023-01-23 20:37:34.590099: step: 2326/1541, loss: -4.060649735038169e-05 2023-01-23 20:37:35.467893: step: 2328/1541, loss: 0.04185405746102333 2023-01-23 20:37:36.335537: step: 2330/1541, loss: 1.9649294699775055e-05 2023-01-23 20:37:37.253700: step: 2332/1541, loss: 0.0012536438880488276 2023-01-23 20:37:38.165108: step: 2334/1541, loss: 0.0001219113401020877 2023-01-23 20:37:39.076792: step: 2336/1541, loss: 0.1981416642665863 2023-01-23 20:37:40.003051: step: 2338/1541, loss: 0.001167358597740531 2023-01-23 20:37:40.926703: step: 2340/1541, loss: 0.0024204771034419537 2023-01-23 20:37:41.830984: step: 2342/1541, loss: 0.001159260282292962 2023-01-23 20:37:42.724527: step: 2344/1541, loss: 0.001970732817426324 2023-01-23 20:37:43.640989: step: 2346/1541, loss: 0.11085130274295807 2023-01-23 20:37:44.536413: step: 2348/1541, loss: 0.0001140421227319166 2023-01-23 20:37:45.450812: step: 2350/1541, loss: 0.027403879910707474 2023-01-23 20:37:46.363920: step: 2352/1541, loss: 0.00031276093795895576 2023-01-23 20:37:47.203320: step: 2354/1541, loss: 0.0007567327120341361 2023-01-23 20:37:48.125757: step: 2356/1541, loss: 0.06353452801704407 2023-01-23 20:37:49.031353: step: 2358/1541, loss: 0.0028519872575998306 2023-01-23 20:37:49.930433: step: 2360/1541, loss: 3.3584168704692274e-05 2023-01-23 20:37:50.840008: step: 2362/1541, loss: 0.030354391783475876 2023-01-23 20:37:51.735812: step: 2364/1541, loss: 0.017979275435209274 2023-01-23 20:37:52.635748: step: 2366/1541, loss: 0.0005807082634419203 2023-01-23 20:37:53.596457: step: 2368/1541, loss: 0.005346335005015135 2023-01-23 20:37:54.508807: step: 2370/1541, loss: 0.011163129471242428 2023-01-23 20:37:55.402252: step: 2372/1541, loss: 0.0008815345936454833 2023-01-23 20:37:56.302401: step: 2374/1541, loss: 0.0001276983239222318 2023-01-23 20:37:57.159161: step: 2376/1541, loss: 9.905009937938303e-05 2023-01-23 20:37:58.026371: step: 2378/1541, loss: 0.011336062103509903 2023-01-23 20:37:58.942786: step: 2380/1541, loss: 0.0002000837557716295 2023-01-23 20:37:59.802956: step: 2382/1541, loss: 0.00021652858413290232 2023-01-23 20:38:00.688886: step: 2384/1541, loss: 0.0020453387405723333 2023-01-23 20:38:01.560905: step: 2386/1541, loss: 0.011666389182209969 2023-01-23 20:38:02.433186: step: 2388/1541, loss: 0.00018289769650436938 2023-01-23 20:38:03.359509: step: 2390/1541, loss: 0.0006752196932211518 2023-01-23 20:38:04.285786: step: 2392/1541, loss: 0.0003724698908627033 2023-01-23 20:38:05.180764: step: 2394/1541, loss: 0.24400249123573303 2023-01-23 20:38:06.055125: step: 2396/1541, loss: 4.328694558353163e-05 2023-01-23 20:38:06.922214: step: 2398/1541, loss: 0.0024306883569806814 2023-01-23 20:38:07.802035: step: 2400/1541, loss: 4.251375867170282e-05 2023-01-23 20:38:08.676883: step: 2402/1541, loss: 0.31037917733192444 2023-01-23 20:38:09.607443: step: 2404/1541, loss: 0.36429643630981445 2023-01-23 20:38:10.500465: step: 2406/1541, loss: 7.470903074136004e-05 2023-01-23 20:38:11.379773: step: 2408/1541, loss: 0.08926025778055191 2023-01-23 20:38:12.302453: step: 2410/1541, loss: 0.08654992282390594 2023-01-23 20:38:13.199537: step: 2412/1541, loss: 0.002442097757011652 2023-01-23 20:38:14.128255: step: 2414/1541, loss: 1.4600455870095175e-05 2023-01-23 20:38:15.017650: step: 2416/1541, loss: 0.2039758712053299 2023-01-23 20:38:15.894154: step: 2418/1541, loss: 0.0006663671229034662 2023-01-23 20:38:16.824774: step: 2420/1541, loss: 0.06550055742263794 2023-01-23 20:38:17.673206: step: 2422/1541, loss: 0.020269816741347313 2023-01-23 20:38:18.543452: step: 2424/1541, loss: 0.0526873804628849 2023-01-23 20:38:19.476269: step: 2426/1541, loss: 0.0035282208118587732 2023-01-23 20:38:20.344414: step: 2428/1541, loss: 5.017378498450853e-05 2023-01-23 20:38:21.223700: step: 2430/1541, loss: 0.0002791336737573147 2023-01-23 20:38:22.101715: step: 2432/1541, loss: 0.000221921072807163 2023-01-23 20:38:22.973393: step: 2434/1541, loss: 0.009336402639746666 2023-01-23 20:38:23.925456: step: 2436/1541, loss: 0.0017550329212099314 2023-01-23 20:38:24.809154: step: 2438/1541, loss: 0.002061261096969247 2023-01-23 20:38:25.772111: step: 2440/1541, loss: 0.007334115915000439 2023-01-23 20:38:26.653867: step: 2442/1541, loss: 3.241038939449936e-05 2023-01-23 20:38:27.524610: step: 2444/1541, loss: 0.00028177749481983483 2023-01-23 20:38:28.429673: step: 2446/1541, loss: 0.0006393957301042974 2023-01-23 20:38:29.391400: step: 2448/1541, loss: 0.0006457935669459403 2023-01-23 20:38:30.328525: step: 2450/1541, loss: 0.12970532476902008 2023-01-23 20:38:31.193584: step: 2452/1541, loss: 0.0015222812071442604 2023-01-23 20:38:32.123787: step: 2454/1541, loss: 0.06896024197340012 2023-01-23 20:38:33.007318: step: 2456/1541, loss: 9.339449024992064e-05 2023-01-23 20:38:33.869003: step: 2458/1541, loss: 0.00010682166612241417 2023-01-23 20:38:34.747724: step: 2460/1541, loss: 0.00023803366639185697 2023-01-23 20:38:35.652312: step: 2462/1541, loss: 0.0004236119566485286 2023-01-23 20:38:36.506668: step: 2464/1541, loss: 0.41189929842948914 2023-01-23 20:38:37.363371: step: 2466/1541, loss: 0.005789887625724077 2023-01-23 20:38:38.258003: step: 2468/1541, loss: 0.1030089408159256 2023-01-23 20:38:39.118329: step: 2470/1541, loss: 0.027040593326091766 2023-01-23 20:38:40.031883: step: 2472/1541, loss: 0.00042163877515122294 2023-01-23 20:38:40.937681: step: 2474/1541, loss: 0.030598221346735954 2023-01-23 20:38:41.840767: step: 2476/1541, loss: 0.00014009405276738107 2023-01-23 20:38:42.707806: step: 2478/1541, loss: 0.004358184523880482 2023-01-23 20:38:43.616455: step: 2480/1541, loss: 0.00638876436278224 2023-01-23 20:38:44.534816: step: 2482/1541, loss: 0.00014530884800478816 2023-01-23 20:38:45.418006: step: 2484/1541, loss: 0.027467932552099228 2023-01-23 20:38:46.325274: step: 2486/1541, loss: 0.0015364985447376966 2023-01-23 20:38:47.211469: step: 2488/1541, loss: 0.45452892780303955 2023-01-23 20:38:48.123326: step: 2490/1541, loss: 0.001046196324750781 2023-01-23 20:38:48.988003: step: 2492/1541, loss: 0.03440776839852333 2023-01-23 20:38:49.890406: step: 2494/1541, loss: 0.002030384261161089 2023-01-23 20:38:50.793670: step: 2496/1541, loss: 0.0007911153370514512 2023-01-23 20:38:51.662787: step: 2498/1541, loss: 0.00029232422821223736 2023-01-23 20:38:52.579329: step: 2500/1541, loss: 0.20160998404026031 2023-01-23 20:38:53.516166: step: 2502/1541, loss: 0.17893537878990173 2023-01-23 20:38:54.425768: step: 2504/1541, loss: 0.00021574409038294107 2023-01-23 20:38:55.324757: step: 2506/1541, loss: 0.0005720561021007597 2023-01-23 20:38:56.211278: step: 2508/1541, loss: 0.011298468336462975 2023-01-23 20:38:57.115812: step: 2510/1541, loss: -5.637908998323837e-06 2023-01-23 20:38:57.956380: step: 2512/1541, loss: 0.04298822209239006 2023-01-23 20:38:58.826019: step: 2514/1541, loss: 4.44930701632984e-05 2023-01-23 20:38:59.793347: step: 2516/1541, loss: 0.0035995154175907373 2023-01-23 20:39:00.704083: step: 2518/1541, loss: 0.0915520116686821 2023-01-23 20:39:01.563487: step: 2520/1541, loss: 0.00032320566242560744 2023-01-23 20:39:02.438937: step: 2522/1541, loss: 0.0004999506054446101 2023-01-23 20:39:03.295005: step: 2524/1541, loss: 0.1026003435254097 2023-01-23 20:39:04.255765: step: 2526/1541, loss: 2.4872615337371826 2023-01-23 20:39:05.136218: step: 2528/1541, loss: 0.0028842268511652946 2023-01-23 20:39:06.044941: step: 2530/1541, loss: 0.0046663531102240086 2023-01-23 20:39:06.912028: step: 2532/1541, loss: 7.69407779444009e-05 2023-01-23 20:39:07.861842: step: 2534/1541, loss: 4.090605580131523e-05 2023-01-23 20:39:08.772271: step: 2536/1541, loss: 0.000396006740629673 2023-01-23 20:39:09.646903: step: 2538/1541, loss: 0.005593983922153711 2023-01-23 20:39:10.537318: step: 2540/1541, loss: 0.0004189168394077569 2023-01-23 20:39:11.416191: step: 2542/1541, loss: 0.00019032393174711615 2023-01-23 20:39:12.317235: step: 2544/1541, loss: 0.055920567363500595 2023-01-23 20:39:13.236051: step: 2546/1541, loss: 0.016844993457198143 2023-01-23 20:39:14.160906: step: 2548/1541, loss: 0.0003134851576760411 2023-01-23 20:39:15.054782: step: 2550/1541, loss: 0.0006566254887729883 2023-01-23 20:39:16.037028: step: 2552/1541, loss: 0.00022064143558964133 2023-01-23 20:39:16.994328: step: 2554/1541, loss: 0.00017966605082619935 2023-01-23 20:39:17.891094: step: 2556/1541, loss: 0.01089048944413662 2023-01-23 20:39:18.806257: step: 2558/1541, loss: 0.33807849884033203 2023-01-23 20:39:19.768857: step: 2560/1541, loss: 0.25034648180007935 2023-01-23 20:39:20.747599: step: 2562/1541, loss: 0.0007084034150466323 2023-01-23 20:39:21.678961: step: 2564/1541, loss: 4.65063531009946e-05 2023-01-23 20:39:22.545670: step: 2566/1541, loss: 0.07907659560441971 2023-01-23 20:39:23.450333: step: 2568/1541, loss: 0.00767540792003274 2023-01-23 20:39:24.367539: step: 2570/1541, loss: 0.000468096521217376 2023-01-23 20:39:25.244164: step: 2572/1541, loss: 0.00020762006170116365 2023-01-23 20:39:26.102181: step: 2574/1541, loss: 0.0007867555250413716 2023-01-23 20:39:27.009572: step: 2576/1541, loss: 0.001164012704975903 2023-01-23 20:39:27.877912: step: 2578/1541, loss: 0.05729788541793823 2023-01-23 20:39:28.784604: step: 2580/1541, loss: 0.08324015885591507 2023-01-23 20:39:29.678031: step: 2582/1541, loss: 0.020489443093538284 2023-01-23 20:39:30.604039: step: 2584/1541, loss: 0.01455403957515955 2023-01-23 20:39:31.489234: step: 2586/1541, loss: 0.047889430075883865 2023-01-23 20:39:32.342881: step: 2588/1541, loss: 0.0006950535462237895 2023-01-23 20:39:33.221444: step: 2590/1541, loss: 0.01039110217243433 2023-01-23 20:39:34.136725: step: 2592/1541, loss: 0.0003940190945286304 2023-01-23 20:39:35.036974: step: 2594/1541, loss: 0.002655266784131527 2023-01-23 20:39:35.930100: step: 2596/1541, loss: 0.17866744101047516 2023-01-23 20:39:36.805083: step: 2598/1541, loss: 0.0009834340307861567 2023-01-23 20:39:37.724127: step: 2600/1541, loss: 0.02942698262631893 2023-01-23 20:39:38.620129: step: 2602/1541, loss: 0.0009490478551015258 2023-01-23 20:39:39.558208: step: 2604/1541, loss: 0.00048675748985260725 2023-01-23 20:39:40.499917: step: 2606/1541, loss: 0.028774293139576912 2023-01-23 20:39:41.392540: step: 2608/1541, loss: 0.035119350999593735 2023-01-23 20:39:42.278004: step: 2610/1541, loss: 7.246219320222735e-05 2023-01-23 20:39:43.150798: step: 2612/1541, loss: 0.2415577918291092 2023-01-23 20:39:44.003475: step: 2614/1541, loss: 0.004830038640648127 2023-01-23 20:39:44.922574: step: 2616/1541, loss: 0.0006849608034826815 2023-01-23 20:39:45.826862: step: 2618/1541, loss: 0.26586005091667175 2023-01-23 20:39:46.723454: step: 2620/1541, loss: 0.21157696843147278 2023-01-23 20:39:47.712704: step: 2622/1541, loss: 0.010862153023481369 2023-01-23 20:39:48.653034: step: 2624/1541, loss: 0.00016589998267591 2023-01-23 20:39:49.542622: step: 2626/1541, loss: 0.00015233896556310356 2023-01-23 20:39:50.411293: step: 2628/1541, loss: 0.021019132807850838 2023-01-23 20:39:51.315234: step: 2630/1541, loss: 0.031264692544937134 2023-01-23 20:39:52.216193: step: 2632/1541, loss: 0.028504323214292526 2023-01-23 20:39:53.092574: step: 2634/1541, loss: 0.0032260981388390064 2023-01-23 20:39:54.050256: step: 2636/1541, loss: 0.008749932050704956 2023-01-23 20:39:54.901374: step: 2638/1541, loss: 0.00010283544543199241 2023-01-23 20:39:55.760057: step: 2640/1541, loss: 0.022813033312559128 2023-01-23 20:39:56.682139: step: 2642/1541, loss: 0.00042498900438658893 2023-01-23 20:39:57.577899: step: 2644/1541, loss: 0.018973641097545624 2023-01-23 20:39:58.558180: step: 2646/1541, loss: 0.20040147006511688 2023-01-23 20:39:59.475129: step: 2648/1541, loss: 0.05625272914767265 2023-01-23 20:40:00.391274: step: 2650/1541, loss: 0.06496436148881912 2023-01-23 20:40:01.293944: step: 2652/1541, loss: 0.18144921958446503 2023-01-23 20:40:02.292525: step: 2654/1541, loss: 0.0799909383058548 2023-01-23 20:40:03.181837: step: 2656/1541, loss: 0.0001967136631719768 2023-01-23 20:40:04.089059: step: 2658/1541, loss: 0.0012257930357009172 2023-01-23 20:40:04.917265: step: 2660/1541, loss: -1.098780739994254e-05 2023-01-23 20:40:05.807836: step: 2662/1541, loss: 0.007059829775243998 2023-01-23 20:40:06.671880: step: 2664/1541, loss: 0.20343056321144104 2023-01-23 20:40:07.575836: step: 2666/1541, loss: 0.06334967166185379 2023-01-23 20:40:08.489616: step: 2668/1541, loss: 0.000731915351934731 2023-01-23 20:40:09.419518: step: 2670/1541, loss: 0.007411099039018154 2023-01-23 20:40:10.308776: step: 2672/1541, loss: 0.00012227791012264788 2023-01-23 20:40:11.195544: step: 2674/1541, loss: 0.007441341411322355 2023-01-23 20:40:12.079343: step: 2676/1541, loss: 0.00013083970407024026 2023-01-23 20:40:12.978057: step: 2678/1541, loss: 0.09529239684343338 2023-01-23 20:40:13.887435: step: 2680/1541, loss: 0.0004525969852693379 2023-01-23 20:40:14.809127: step: 2682/1541, loss: 0.0893329530954361 2023-01-23 20:40:15.716706: step: 2684/1541, loss: 0.00019078249169979244 2023-01-23 20:40:16.649093: step: 2686/1541, loss: 0.008008647710084915 2023-01-23 20:40:17.522885: step: 2688/1541, loss: 0.07452679425477982 2023-01-23 20:40:18.394237: step: 2690/1541, loss: 4.4438209442887455e-05 2023-01-23 20:40:19.307439: step: 2692/1541, loss: 0.000385850464226678 2023-01-23 20:40:20.206856: step: 2694/1541, loss: 0.32389384508132935 2023-01-23 20:40:21.105002: step: 2696/1541, loss: 0.005217378959059715 2023-01-23 20:40:22.022573: step: 2698/1541, loss: 0.975280225276947 2023-01-23 20:40:22.918533: step: 2700/1541, loss: 0.00011964859731961042 2023-01-23 20:40:23.810070: step: 2702/1541, loss: 0.0005808992427773774 2023-01-23 20:40:24.665210: step: 2704/1541, loss: 0.0023945271968841553 2023-01-23 20:40:25.537160: step: 2706/1541, loss: 0.0499611422419548 2023-01-23 20:40:26.461520: step: 2708/1541, loss: 0.5467968583106995 2023-01-23 20:40:27.378819: step: 2710/1541, loss: 0.0007309893844649196 2023-01-23 20:40:28.294761: step: 2712/1541, loss: 0.021345864981412888 2023-01-23 20:40:29.208303: step: 2714/1541, loss: 0.04054350033402443 2023-01-23 20:40:30.103868: step: 2716/1541, loss: 0.21506932377815247 2023-01-23 20:40:31.025590: step: 2718/1541, loss: 0.1379457712173462 2023-01-23 20:40:31.939120: step: 2720/1541, loss: 0.001565308659337461 2023-01-23 20:40:32.846256: step: 2722/1541, loss: 0.04550608992576599 2023-01-23 20:40:33.761520: step: 2724/1541, loss: 0.0004755911650136113 2023-01-23 20:40:34.637450: step: 2726/1541, loss: 5.634612534777261e-05 2023-01-23 20:40:35.564949: step: 2728/1541, loss: 0.03597886860370636 2023-01-23 20:40:36.505001: step: 2730/1541, loss: 0.00029977993108332157 2023-01-23 20:40:37.394534: step: 2732/1541, loss: 0.00010494610614841804 2023-01-23 20:40:38.249119: step: 2734/1541, loss: 0.0001415965671185404 2023-01-23 20:40:39.163508: step: 2736/1541, loss: 0.31316885352134705 2023-01-23 20:40:40.080370: step: 2738/1541, loss: 0.00737959798425436 2023-01-23 20:40:40.959628: step: 2740/1541, loss: 5.458117084344849e-05 2023-01-23 20:40:41.862301: step: 2742/1541, loss: 0.21190766990184784 2023-01-23 20:40:42.791446: step: 2744/1541, loss: 0.17811669409275055 2023-01-23 20:40:43.717181: step: 2746/1541, loss: 0.5307702422142029 2023-01-23 20:40:44.616561: step: 2748/1541, loss: 0.00029688188806176186 2023-01-23 20:40:45.497238: step: 2750/1541, loss: 6.428199412766844e-05 2023-01-23 20:40:46.399510: step: 2752/1541, loss: 0.033766694366931915 2023-01-23 20:40:47.281016: step: 2754/1541, loss: 0.0478597953915596 2023-01-23 20:40:48.145077: step: 2756/1541, loss: 0.00024086951452773064 2023-01-23 20:40:49.065317: step: 2758/1541, loss: 0.0018322516698390245 2023-01-23 20:40:49.964608: step: 2760/1541, loss: 0.005280117969959974 2023-01-23 20:40:50.896271: step: 2762/1541, loss: 0.07184728235006332 2023-01-23 20:40:51.827736: step: 2764/1541, loss: 0.0003301110118627548 2023-01-23 20:40:52.728312: step: 2766/1541, loss: 0.014854435808956623 2023-01-23 20:40:53.631500: step: 2768/1541, loss: 0.0009651071741245687 2023-01-23 20:40:54.497067: step: 2770/1541, loss: 0.4417804479598999 2023-01-23 20:40:55.346599: step: 2772/1541, loss: 0.0001375968859065324 2023-01-23 20:40:56.231526: step: 2774/1541, loss: 0.028804130852222443 2023-01-23 20:40:57.117493: step: 2776/1541, loss: 0.00015886002802290022 2023-01-23 20:40:58.024295: step: 2778/1541, loss: 0.0001775336277205497 2023-01-23 20:40:58.902661: step: 2780/1541, loss: 0.004833010956645012 2023-01-23 20:40:59.826844: step: 2782/1541, loss: 0.001225874526426196 2023-01-23 20:41:00.739434: step: 2784/1541, loss: 0.0012873547384515405 2023-01-23 20:41:01.588415: step: 2786/1541, loss: 0.017811346799135208 2023-01-23 20:41:02.487804: step: 2788/1541, loss: 0.005428043659776449 2023-01-23 20:41:03.370629: step: 2790/1541, loss: 0.00481982110068202 2023-01-23 20:41:04.280188: step: 2792/1541, loss: 0.0011517630191519856 2023-01-23 20:41:05.185373: step: 2794/1541, loss: 0.19518209993839264 2023-01-23 20:41:06.061941: step: 2796/1541, loss: 0.14241386950016022 2023-01-23 20:41:06.957008: step: 2798/1541, loss: 0.11756020784378052 2023-01-23 20:41:07.841802: step: 2800/1541, loss: 0.266104519367218 2023-01-23 20:41:08.714169: step: 2802/1541, loss: 0.0007358071743510664 2023-01-23 20:41:09.611202: step: 2804/1541, loss: 0.007949532940983772 2023-01-23 20:41:10.490177: step: 2806/1541, loss: 0.0012698292266577482 2023-01-23 20:41:11.375099: step: 2808/1541, loss: 0.053886570036411285 2023-01-23 20:41:12.273913: step: 2810/1541, loss: 0.052740391343832016 2023-01-23 20:41:13.148282: step: 2812/1541, loss: 0.0010097827762365341 2023-01-23 20:41:14.055003: step: 2814/1541, loss: 0.047950729727745056 2023-01-23 20:41:15.072066: step: 2816/1541, loss: 0.0010395047720521688 2023-01-23 20:41:16.062560: step: 2818/1541, loss: 0.0018482391024008393 2023-01-23 20:41:16.944934: step: 2820/1541, loss: 0.0016008351230993867 2023-01-23 20:41:17.871516: step: 2822/1541, loss: 0.016609754413366318 2023-01-23 20:41:18.738152: step: 2824/1541, loss: 0.04260462149977684 2023-01-23 20:41:19.635681: step: 2826/1541, loss: 0.008545373566448689 2023-01-23 20:41:20.524734: step: 2828/1541, loss: 0.05515683814883232 2023-01-23 20:41:21.425650: step: 2830/1541, loss: 0.009757949970662594 2023-01-23 20:41:22.301930: step: 2832/1541, loss: 0.2796180248260498 2023-01-23 20:41:23.180626: step: 2834/1541, loss: 0.005059078335762024 2023-01-23 20:41:24.069093: step: 2836/1541, loss: 0.05082816630601883 2023-01-23 20:41:24.975466: step: 2838/1541, loss: 0.029254542663693428 2023-01-23 20:41:25.853963: step: 2840/1541, loss: 0.024554969742894173 2023-01-23 20:41:26.771361: step: 2842/1541, loss: 0.13170796632766724 2023-01-23 20:41:27.674355: step: 2844/1541, loss: 0.0031722192652523518 2023-01-23 20:41:28.529336: step: 2846/1541, loss: 0.0008330440614372492 2023-01-23 20:41:29.420218: step: 2848/1541, loss: 0.11309266835451126 2023-01-23 20:41:30.300580: step: 2850/1541, loss: 0.00019598827930167317 2023-01-23 20:41:31.237879: step: 2852/1541, loss: 0.0004428318643476814 2023-01-23 20:41:32.106293: step: 2854/1541, loss: 0.0015510404482483864 2023-01-23 20:41:32.981686: step: 2856/1541, loss: 0.04488251358270645 2023-01-23 20:41:33.832827: step: 2858/1541, loss: 0.021967876702547073 2023-01-23 20:41:34.727568: step: 2860/1541, loss: 0.00022642937256023288 2023-01-23 20:41:35.596814: step: 2862/1541, loss: 0.000317832687869668 2023-01-23 20:41:36.544291: step: 2864/1541, loss: 0.11119487136602402 2023-01-23 20:41:37.446947: step: 2866/1541, loss: 0.002401108853518963 2023-01-23 20:41:38.421686: step: 2868/1541, loss: 0.018271056935191154 2023-01-23 20:41:39.291279: step: 2870/1541, loss: 0.00019261885609012097 2023-01-23 20:41:40.180838: step: 2872/1541, loss: 0.20641987025737762 2023-01-23 20:41:41.066123: step: 2874/1541, loss: 0.1957511305809021 2023-01-23 20:41:42.044626: step: 2876/1541, loss: 0.0023995807860046625 2023-01-23 20:41:42.956605: step: 2878/1541, loss: 0.021614056080579758 2023-01-23 20:41:43.834493: step: 2880/1541, loss: 0.015887226909399033 2023-01-23 20:41:44.727462: step: 2882/1541, loss: 0.00011144835298182443 2023-01-23 20:41:45.636024: step: 2884/1541, loss: 0.00017124068108387291 2023-01-23 20:41:46.496970: step: 2886/1541, loss: 0.0030597217846661806 2023-01-23 20:41:47.404225: step: 2888/1541, loss: 0.0016437650192528963 2023-01-23 20:41:48.323257: step: 2890/1541, loss: 0.031779129058122635 2023-01-23 20:41:49.284423: step: 2892/1541, loss: 0.0005526530439965427 2023-01-23 20:41:50.200453: step: 2894/1541, loss: 0.0020535567309707403 2023-01-23 20:41:51.154032: step: 2896/1541, loss: 0.03465987741947174 2023-01-23 20:41:52.024740: step: 2898/1541, loss: 0.030075592920184135 2023-01-23 20:41:52.910002: step: 2900/1541, loss: 0.014701427891850471 2023-01-23 20:41:53.860273: step: 2902/1541, loss: 0.0010966045083478093 2023-01-23 20:41:54.743521: step: 2904/1541, loss: 0.0033048996701836586 2023-01-23 20:41:55.622774: step: 2906/1541, loss: 0.0007892552530393004 2023-01-23 20:41:56.573655: step: 2908/1541, loss: 0.22780823707580566 2023-01-23 20:41:57.518150: step: 2910/1541, loss: 0.3218807578086853 2023-01-23 20:41:58.397940: step: 2912/1541, loss: 0.0030728995334357023 2023-01-23 20:41:59.289071: step: 2914/1541, loss: 0.10676160454750061 2023-01-23 20:42:00.218842: step: 2916/1541, loss: 0.0014215023256838322 2023-01-23 20:42:01.120397: step: 2918/1541, loss: 1.23745596408844 2023-01-23 20:42:02.015398: step: 2920/1541, loss: 0.023255852982401848 2023-01-23 20:42:02.918651: step: 2922/1541, loss: 0.0978999063372612 2023-01-23 20:42:03.784458: step: 2924/1541, loss: 0.15669871866703033 2023-01-23 20:42:04.681145: step: 2926/1541, loss: 0.00037670062738470733 2023-01-23 20:42:05.578287: step: 2928/1541, loss: 0.003691186895594001 2023-01-23 20:42:06.454136: step: 2930/1541, loss: 0.010475493967533112 2023-01-23 20:42:07.340716: step: 2932/1541, loss: 0.10350453853607178 2023-01-23 20:42:08.225751: step: 2934/1541, loss: 0.003123528091236949 2023-01-23 20:42:09.167398: step: 2936/1541, loss: 0.003003431484103203 2023-01-23 20:42:10.080383: step: 2938/1541, loss: 0.013503093272447586 2023-01-23 20:42:10.946997: step: 2940/1541, loss: 0.005474340170621872 2023-01-23 20:42:11.848986: step: 2942/1541, loss: 0.0011927741579711437 2023-01-23 20:42:12.689234: step: 2944/1541, loss: 0.00020110432524234056 2023-01-23 20:42:13.553314: step: 2946/1541, loss: 0.005385172087699175 2023-01-23 20:42:14.416035: step: 2948/1541, loss: 0.0025337543338537216 2023-01-23 20:42:15.324281: step: 2950/1541, loss: 0.008152061142027378 2023-01-23 20:42:16.185697: step: 2952/1541, loss: 0.11657823622226715 2023-01-23 20:42:17.073026: step: 2954/1541, loss: 0.007169280666857958 2023-01-23 20:42:18.049202: step: 2956/1541, loss: 0.0012001341674476862 2023-01-23 20:42:18.954855: step: 2958/1541, loss: 0.1309671252965927 2023-01-23 20:42:19.964154: step: 2960/1541, loss: 0.001943891984410584 2023-01-23 20:42:20.856927: step: 2962/1541, loss: 0.08038666099309921 2023-01-23 20:42:21.749639: step: 2964/1541, loss: 0.0317736454308033 2023-01-23 20:42:22.689103: step: 2966/1541, loss: 0.003313407301902771 2023-01-23 20:42:23.613607: step: 2968/1541, loss: 0.0011530560441315174 2023-01-23 20:42:24.509152: step: 2970/1541, loss: 0.03628644719719887 2023-01-23 20:42:25.444658: step: 2972/1541, loss: 0.10039439797401428 2023-01-23 20:42:26.318466: step: 2974/1541, loss: 3.2599439236946637e-06 2023-01-23 20:42:27.210810: step: 2976/1541, loss: 0.005174407735466957 2023-01-23 20:42:28.208745: step: 2978/1541, loss: 0.2183784395456314 2023-01-23 20:42:29.091409: step: 2980/1541, loss: 0.0002854566555470228 2023-01-23 20:42:29.990525: step: 2982/1541, loss: 0.035945020616054535 2023-01-23 20:42:30.859302: step: 2984/1541, loss: 0.00040433340473100543 2023-01-23 20:42:31.876832: step: 2986/1541, loss: 0.0010421447223052382 2023-01-23 20:42:32.824576: step: 2988/1541, loss: 0.002969450317323208 2023-01-23 20:42:33.737785: step: 2990/1541, loss: 0.0014998825499787927 2023-01-23 20:42:34.697652: step: 2992/1541, loss: 0.006608820054680109 2023-01-23 20:42:35.619210: step: 2994/1541, loss: 0.001626590732485056 2023-01-23 20:42:36.457178: step: 2996/1541, loss: 0.0014368613483384252 2023-01-23 20:42:37.334183: step: 2998/1541, loss: 0.0010114802280440927 2023-01-23 20:42:38.222213: step: 3000/1541, loss: 0.10137902200222015 2023-01-23 20:42:39.123063: step: 3002/1541, loss: 0.00011599401477724314 2023-01-23 20:42:40.069754: step: 3004/1541, loss: 0.003457268001511693 2023-01-23 20:42:40.952077: step: 3006/1541, loss: 0.21448442339897156 2023-01-23 20:42:41.835373: step: 3008/1541, loss: 0.015778254717588425 2023-01-23 20:42:42.741722: step: 3010/1541, loss: 0.02261636033654213 2023-01-23 20:42:43.586912: step: 3012/1541, loss: 0.0020008618012070656 2023-01-23 20:42:44.530263: step: 3014/1541, loss: 0.017669418826699257 2023-01-23 20:42:45.457915: step: 3016/1541, loss: 0.44828420877456665 2023-01-23 20:42:46.333053: step: 3018/1541, loss: 0.011298341676592827 2023-01-23 20:42:47.246193: step: 3020/1541, loss: 0.00700178649276495 2023-01-23 20:42:48.125254: step: 3022/1541, loss: 0.006677985191345215 2023-01-23 20:42:49.024510: step: 3024/1541, loss: 0.07630922645330429 2023-01-23 20:42:49.921104: step: 3026/1541, loss: 0.0005559102864935994 2023-01-23 20:42:50.776377: step: 3028/1541, loss: 0.00010196337098022923 2023-01-23 20:42:51.659515: step: 3030/1541, loss: 0.00027359422529116273 2023-01-23 20:42:52.589143: step: 3032/1541, loss: 0.002141502220183611 2023-01-23 20:42:53.498311: step: 3034/1541, loss: 0.10922287404537201 2023-01-23 20:42:54.393749: step: 3036/1541, loss: 0.00013492183643393219 2023-01-23 20:42:55.279194: step: 3038/1541, loss: 0.00032135119545273483 2023-01-23 20:42:56.176833: step: 3040/1541, loss: 0.019128087908029556 2023-01-23 20:42:57.037868: step: 3042/1541, loss: 0.0665164589881897 2023-01-23 20:42:57.960700: step: 3044/1541, loss: 0.0028985950630158186 2023-01-23 20:42:58.890704: step: 3046/1541, loss: 0.13059712946414948 2023-01-23 20:42:59.754526: step: 3048/1541, loss: 0.001827652333304286 2023-01-23 20:43:00.631370: step: 3050/1541, loss: 0.00022813442046754062 2023-01-23 20:43:01.522821: step: 3052/1541, loss: 0.003339186077937484 2023-01-23 20:43:02.443383: step: 3054/1541, loss: -1.3121094525558874e-05 2023-01-23 20:43:03.447012: step: 3056/1541, loss: 0.0006162789068184793 2023-01-23 20:43:04.304239: step: 3058/1541, loss: 7.896783790783957e-05 2023-01-23 20:43:05.215400: step: 3060/1541, loss: 0.42215418815612793 2023-01-23 20:43:06.083836: step: 3062/1541, loss: 0.14667505025863647 2023-01-23 20:43:06.984762: step: 3064/1541, loss: 0.007808796130120754 2023-01-23 20:43:07.897995: step: 3066/1541, loss: 0.00013577990466728806 2023-01-23 20:43:08.756089: step: 3068/1541, loss: 0.0002484230208210647 2023-01-23 20:43:09.661951: step: 3070/1541, loss: 0.0001552991452626884 2023-01-23 20:43:10.559274: step: 3072/1541, loss: 0.0002520380076020956 2023-01-23 20:43:11.491629: step: 3074/1541, loss: 0.00027366005815565586 2023-01-23 20:43:12.371875: step: 3076/1541, loss: 0.019926315173506737 2023-01-23 20:43:13.301095: step: 3078/1541, loss: 0.03194064646959305 2023-01-23 20:43:14.167159: step: 3080/1541, loss: 0.018216485157608986 2023-01-23 20:43:15.025611: step: 3082/1541, loss: 0.0011802781373262405 2023-01-23 20:43:15.713581: step: 3084/1541, loss: 0.3598845899105072 ================================================== Loss: 0.058 -------------------- Dev Chinese: {'event': {'p': 0.6320191158900836, 'r': 0.7043941411451398, 'f1': 0.6662468513853904}, 'argument': {'p': 0.7064294519407052, 'r': 0.6290350830800685, 'f1': 0.6654896439983955}, 'combined': 0.44338037994351537, 'epoch': 32} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7537890914616138, 'r': 0.677915353590204, 'f1': 0.7138417432965972}, 'combined': 0.4649666465717965, 'epoch': 32} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.7001227469542347, 'r': 0.6333726929435443, 'f1': 0.6650770844232173}, 'combined': 0.4378768381916835, 'epoch': 32} Test Korean: {'event': {'p': 0.6105815061963775, 'r': 0.7675254643499101, 'f1': 0.6801168038226706}, 'argument': {'p': 0.7525362332215595, 'r': 0.6901771914237055, 'f1': 0.7200090260713234}, 'combined': 0.4896902375351024, 'epoch': 32} Dev Russian: {'event': {'p': 0.6540284360189573, 'r': 0.7350199733688415, 'f1': 0.6921630094043887}, 'argument': {'p': 0.7095563060360507, 'r': 0.6348661685585717, 'f1': 0.67013651125627}, 'combined': 0.46384370434289784, 'epoch': 32} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.7505422699780423, 'r': 0.6888280222850542, 'f1': 0.7183621202261284}, 'combined': 0.49924160757614733, 'epoch': 32} Chinese: {'event': {'p': 0.6666666666666666, 'r': 0.7777777777777778, 'f1': 0.717948717948718}, 'argument': {'p': 0.6883971291866028, 'r': 0.6944891391794046, 'f1': 0.6914297156587905}, 'combined': 0.49641107790887523, 'epoch': 32} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5905943937858831, 'r': 0.3701058201058201, 'f1': 0.455048139474369}, 'combined': 0.26514616000868746, 'epoch': 32} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5533618233618234, 'r': 0.5790995825879547, 'f1': 0.5659382284382285}, 'combined': 0.29105394605394613, 'epoch': 32} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 33 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 20:48:33.439034: step: 2/1541, loss: 0.0019863639026880264 2023-01-23 20:48:34.340247: step: 4/1541, loss: 0.12260644137859344 2023-01-23 20:48:35.216751: step: 6/1541, loss: 0.010598946362733841 2023-01-23 20:48:36.151382: step: 8/1541, loss: 0.001376852160319686 2023-01-23 20:48:37.059819: step: 10/1541, loss: 0.03806724026799202 2023-01-23 20:48:37.930049: step: 12/1541, loss: 0.00601884163916111 2023-01-23 20:48:38.845806: step: 14/1541, loss: 0.043683506548404694 2023-01-23 20:48:39.792578: step: 16/1541, loss: 3.6531804653350264e-05 2023-01-23 20:48:40.681894: step: 18/1541, loss: 0.011142605915665627 2023-01-23 20:48:41.559182: step: 20/1541, loss: 0.0006245546392165124 2023-01-23 20:48:42.422445: step: 22/1541, loss: 0.000733500171918422 2023-01-23 20:48:43.346330: step: 24/1541, loss: 0.0012352492194622755 2023-01-23 20:48:44.260199: step: 26/1541, loss: 0.001108256634324789 2023-01-23 20:48:45.183980: step: 28/1541, loss: 0.024884402751922607 2023-01-23 20:48:46.107981: step: 30/1541, loss: 0.7194418907165527 2023-01-23 20:48:47.000052: step: 32/1541, loss: -1.2689712093560956e-05 2023-01-23 20:48:47.872563: step: 34/1541, loss: 9.392245556227863e-05 2023-01-23 20:48:48.756243: step: 36/1541, loss: 0.00042230371036566794 2023-01-23 20:48:49.666082: step: 38/1541, loss: 0.0019162055104970932 2023-01-23 20:48:50.588511: step: 40/1541, loss: 0.02195885218679905 2023-01-23 20:48:51.477002: step: 42/1541, loss: 0.008397498168051243 2023-01-23 20:48:52.386060: step: 44/1541, loss: 0.0004533854662440717 2023-01-23 20:48:53.288730: step: 46/1541, loss: 0.02988284081220627 2023-01-23 20:48:54.175378: step: 48/1541, loss: 0.0005998621927574277 2023-01-23 20:48:55.102405: step: 50/1541, loss: 0.00027447345200926065 2023-01-23 20:48:55.948764: step: 52/1541, loss: 0.025014184415340424 2023-01-23 20:48:56.851381: step: 54/1541, loss: 0.0014869928127154708 2023-01-23 20:48:57.796525: step: 56/1541, loss: 0.012294110842049122 2023-01-23 20:48:58.718126: step: 58/1541, loss: 0.013963899575173855 2023-01-23 20:48:59.578972: step: 60/1541, loss: 0.004018956329673529 2023-01-23 20:49:00.475973: step: 62/1541, loss: 0.007261103484779596 2023-01-23 20:49:01.392238: step: 64/1541, loss: 0.0815432146191597 2023-01-23 20:49:02.321122: step: 66/1541, loss: 0.013428413309156895 2023-01-23 20:49:03.198276: step: 68/1541, loss: 0.0261865071952343 2023-01-23 20:49:04.091048: step: 70/1541, loss: 0.017564259469509125 2023-01-23 20:49:04.965534: step: 72/1541, loss: 0.023816179484128952 2023-01-23 20:49:05.893689: step: 74/1541, loss: 0.0002615654084365815 2023-01-23 20:49:06.764590: step: 76/1541, loss: 0.00030631094705313444 2023-01-23 20:49:07.631853: step: 78/1541, loss: 3.5099110391456634e-05 2023-01-23 20:49:08.496392: step: 80/1541, loss: 4.776864807354286e-05 2023-01-23 20:49:09.375579: step: 82/1541, loss: 0.00015476832049898803 2023-01-23 20:49:10.238831: step: 84/1541, loss: 0.00045983499148860574 2023-01-23 20:49:11.137763: step: 86/1541, loss: 0.09149181842803955 2023-01-23 20:49:12.042165: step: 88/1541, loss: 5.1129427447449416e-05 2023-01-23 20:49:12.895427: step: 90/1541, loss: 0.0003164028166793287 2023-01-23 20:49:13.823168: step: 92/1541, loss: 0.37967967987060547 2023-01-23 20:49:14.697749: step: 94/1541, loss: 0.07770407199859619 2023-01-23 20:49:15.555322: step: 96/1541, loss: 8.589544449932873e-05 2023-01-23 20:49:16.461556: step: 98/1541, loss: 0.006236601155251265 2023-01-23 20:49:17.372551: step: 100/1541, loss: -1.663213515712414e-05 2023-01-23 20:49:18.278420: step: 102/1541, loss: 1.782166145858355e-05 2023-01-23 20:49:19.151474: step: 104/1541, loss: 0.008644640445709229 2023-01-23 20:49:20.090771: step: 106/1541, loss: 5.5329841416096315e-05 2023-01-23 20:49:21.046382: step: 108/1541, loss: 0.00033421567059122026 2023-01-23 20:49:21.926747: step: 110/1541, loss: 0.00026001708465628326 2023-01-23 20:49:22.800266: step: 112/1541, loss: 0.050769269466400146 2023-01-23 20:49:23.666264: step: 114/1541, loss: 0.01551266573369503 2023-01-23 20:49:24.577040: step: 116/1541, loss: 0.0037007746286690235 2023-01-23 20:49:25.443129: step: 118/1541, loss: 0.004969281144440174 2023-01-23 20:49:26.363910: step: 120/1541, loss: 0.8608496189117432 2023-01-23 20:49:27.266351: step: 122/1541, loss: 0.0004830908146686852 2023-01-23 20:49:28.130935: step: 124/1541, loss: 0.12048612534999847 2023-01-23 20:49:29.016123: step: 126/1541, loss: -1.1894881026819348e-06 2023-01-23 20:49:29.929814: step: 128/1541, loss: 0.0007639182731509209 2023-01-23 20:49:30.867015: step: 130/1541, loss: 0.1199057474732399 2023-01-23 20:49:31.775735: step: 132/1541, loss: 0.0012889282079413533 2023-01-23 20:49:32.775780: step: 134/1541, loss: 3.9239028410520405e-05 2023-01-23 20:49:33.645118: step: 136/1541, loss: 0.08004717528820038 2023-01-23 20:49:34.544712: step: 138/1541, loss: 1.1032784641429316e-06 2023-01-23 20:49:35.466208: step: 140/1541, loss: 0.2676035463809967 2023-01-23 20:49:36.374734: step: 142/1541, loss: 0.014277375303208828 2023-01-23 20:49:37.252485: step: 144/1541, loss: 0.00023435737239196897 2023-01-23 20:49:38.161759: step: 146/1541, loss: 0.44996052980422974 2023-01-23 20:49:39.084865: step: 148/1541, loss: 0.00019986297411378473 2023-01-23 20:49:40.058844: step: 150/1541, loss: 0.059540051966905594 2023-01-23 20:49:41.014671: step: 152/1541, loss: 0.00880240648984909 2023-01-23 20:49:41.915631: step: 154/1541, loss: 0.0008933331700973213 2023-01-23 20:49:42.824625: step: 156/1541, loss: 0.18852072954177856 2023-01-23 20:49:43.693751: step: 158/1541, loss: 0.0057089370675385 2023-01-23 20:49:44.565286: step: 160/1541, loss: 0.00014513068890664726 2023-01-23 20:49:45.453416: step: 162/1541, loss: 0.015478681772947311 2023-01-23 20:49:46.325286: step: 164/1541, loss: 0.05311661213636398 2023-01-23 20:49:47.253287: step: 166/1541, loss: 0.008661356754601002 2023-01-23 20:49:48.119459: step: 168/1541, loss: 0.00024180224863812327 2023-01-23 20:49:48.983289: step: 170/1541, loss: 0.0005349138518795371 2023-01-23 20:49:49.868702: step: 172/1541, loss: 0.06468959152698517 2023-01-23 20:49:50.761337: step: 174/1541, loss: 0.34883472323417664 2023-01-23 20:49:51.666809: step: 176/1541, loss: 0.29703354835510254 2023-01-23 20:49:52.597983: step: 178/1541, loss: 0.0031183497048914433 2023-01-23 20:49:53.460029: step: 180/1541, loss: 0.00023768722894601524 2023-01-23 20:49:54.456596: step: 182/1541, loss: 0.09913773834705353 2023-01-23 20:49:55.351539: step: 184/1541, loss: 0.00023129492183215916 2023-01-23 20:49:56.236276: step: 186/1541, loss: 0.0007878186879679561 2023-01-23 20:49:57.139226: step: 188/1541, loss: 0.01418466679751873 2023-01-23 20:49:58.016454: step: 190/1541, loss: 0.004868834279477596 2023-01-23 20:49:58.937847: step: 192/1541, loss: 0.000743336568120867 2023-01-23 20:49:59.783710: step: 194/1541, loss: 6.155914888950065e-05 2023-01-23 20:50:00.706140: step: 196/1541, loss: 0.0018181302584707737 2023-01-23 20:50:01.576461: step: 198/1541, loss: 0.014841808006167412 2023-01-23 20:50:02.495651: step: 200/1541, loss: 0.001476916833780706 2023-01-23 20:50:03.370901: step: 202/1541, loss: 0.045342158526182175 2023-01-23 20:50:04.327147: step: 204/1541, loss: 0.14760640263557434 2023-01-23 20:50:05.300082: step: 206/1541, loss: 0.0176116731017828 2023-01-23 20:50:06.200419: step: 208/1541, loss: 0.022844575345516205 2023-01-23 20:50:07.077208: step: 210/1541, loss: 0.00034488216624595225 2023-01-23 20:50:08.037563: step: 212/1541, loss: 0.0019359007710590959 2023-01-23 20:50:08.916036: step: 214/1541, loss: 0.0004886342794634402 2023-01-23 20:50:09.823018: step: 216/1541, loss: 0.024270767346024513 2023-01-23 20:50:10.719017: step: 218/1541, loss: 0.04537827894091606 2023-01-23 20:50:11.641935: step: 220/1541, loss: 0.38713955879211426 2023-01-23 20:50:12.566601: step: 222/1541, loss: 0.0003273270558565855 2023-01-23 20:50:13.429426: step: 224/1541, loss: 0.0012174013536423445 2023-01-23 20:50:14.332892: step: 226/1541, loss: 0.3022518455982208 2023-01-23 20:50:15.234299: step: 228/1541, loss: 0.001930514583364129 2023-01-23 20:50:16.227732: step: 230/1541, loss: 0.0009455413674004376 2023-01-23 20:50:17.103721: step: 232/1541, loss: 0.04238840937614441 2023-01-23 20:50:18.016825: step: 234/1541, loss: 7.161288522183895e-05 2023-01-23 20:50:18.874201: step: 236/1541, loss: 0.08840087056159973 2023-01-23 20:50:19.795644: step: 238/1541, loss: 8.585785690229386e-05 2023-01-23 20:50:20.671285: step: 240/1541, loss: 0.00013052656140644103 2023-01-23 20:50:21.582756: step: 242/1541, loss: 0.0002575737307779491 2023-01-23 20:50:22.504204: step: 244/1541, loss: 0.02139807865023613 2023-01-23 20:50:23.414796: step: 246/1541, loss: 0.0918063372373581 2023-01-23 20:50:24.263897: step: 248/1541, loss: 0.08057812601327896 2023-01-23 20:50:25.155862: step: 250/1541, loss: 0.00017360092897433788 2023-01-23 20:50:26.045982: step: 252/1541, loss: 7.878414908191189e-05 2023-01-23 20:50:26.938264: step: 254/1541, loss: 0.04655442759394646 2023-01-23 20:50:27.798906: step: 256/1541, loss: 0.13779865205287933 2023-01-23 20:50:28.634992: step: 258/1541, loss: 0.0001635711087146774 2023-01-23 20:50:29.564074: step: 260/1541, loss: 0.01859242096543312 2023-01-23 20:50:30.430886: step: 262/1541, loss: 0.2161344736814499 2023-01-23 20:50:31.301873: step: 264/1541, loss: 0.000663465354591608 2023-01-23 20:50:32.241904: step: 266/1541, loss: 0.2947727143764496 2023-01-23 20:50:33.124564: step: 268/1541, loss: 0.520010769367218 2023-01-23 20:50:34.017483: step: 270/1541, loss: 0.01729070395231247 2023-01-23 20:50:34.904714: step: 272/1541, loss: 0.00030713557498529553 2023-01-23 20:50:35.779116: step: 274/1541, loss: 0.0010710930218920112 2023-01-23 20:50:36.654654: step: 276/1541, loss: 1.3763354218099266e-05 2023-01-23 20:50:37.542821: step: 278/1541, loss: 0.10185234248638153 2023-01-23 20:50:38.515395: step: 280/1541, loss: 0.01668473519384861 2023-01-23 20:50:39.431488: step: 282/1541, loss: 0.002096649492159486 2023-01-23 20:50:40.357964: step: 284/1541, loss: 0.03630489483475685 2023-01-23 20:50:41.296013: step: 286/1541, loss: 0.00011636613635346293 2023-01-23 20:50:42.172304: step: 288/1541, loss: 0.06961864978075027 2023-01-23 20:50:43.042679: step: 290/1541, loss: 0.03030470572412014 2023-01-23 20:50:43.919392: step: 292/1541, loss: 0.011395910754799843 2023-01-23 20:50:44.849859: step: 294/1541, loss: 0.0441909022629261 2023-01-23 20:50:45.780846: step: 296/1541, loss: 0.10834509879350662 2023-01-23 20:50:46.720573: step: 298/1541, loss: 0.000350505142705515 2023-01-23 20:50:47.618355: step: 300/1541, loss: 0.0006792660569772124 2023-01-23 20:50:48.510333: step: 302/1541, loss: 0.0016500996425747871 2023-01-23 20:50:49.374185: step: 304/1541, loss: 0.0014273462584242225 2023-01-23 20:50:50.223395: step: 306/1541, loss: 0.0008210509549826384 2023-01-23 20:50:51.108471: step: 308/1541, loss: 5.584455357166007e-05 2023-01-23 20:50:51.968140: step: 310/1541, loss: 0.00011580604768823832 2023-01-23 20:50:52.911107: step: 312/1541, loss: 0.0001753501419443637 2023-01-23 20:50:53.821101: step: 314/1541, loss: 0.0017226856434717774 2023-01-23 20:50:54.682520: step: 316/1541, loss: 0.0025149076245725155 2023-01-23 20:50:55.570594: step: 318/1541, loss: 0.0001221198617713526 2023-01-23 20:50:56.445860: step: 320/1541, loss: 0.0011121499119326472 2023-01-23 20:50:57.352156: step: 322/1541, loss: 0.00020415004109963775 2023-01-23 20:50:58.234227: step: 324/1541, loss: 0.1832299530506134 2023-01-23 20:50:59.128665: step: 326/1541, loss: 0.32487753033638 2023-01-23 20:51:00.005245: step: 328/1541, loss: 0.0004807819495908916 2023-01-23 20:51:00.877565: step: 330/1541, loss: 0.001817051088437438 2023-01-23 20:51:01.756864: step: 332/1541, loss: 0.08842656016349792 2023-01-23 20:51:02.627866: step: 334/1541, loss: 0.002027393551543355 2023-01-23 20:51:03.567042: step: 336/1541, loss: 0.020893152803182602 2023-01-23 20:51:04.424959: step: 338/1541, loss: 0.0048991101793944836 2023-01-23 20:51:05.278345: step: 340/1541, loss: 0.10025953501462936 2023-01-23 20:51:06.130751: step: 342/1541, loss: 0.0018211054848507047 2023-01-23 20:51:07.021101: step: 344/1541, loss: 0.004656096920371056 2023-01-23 20:51:07.917769: step: 346/1541, loss: 0.09913274645805359 2023-01-23 20:51:08.845515: step: 348/1541, loss: 0.018566381186246872 2023-01-23 20:51:09.731955: step: 350/1541, loss: 0.0002274678845424205 2023-01-23 20:51:10.653962: step: 352/1541, loss: 0.376068651676178 2023-01-23 20:51:11.518741: step: 354/1541, loss: 0.00020631086954381317 2023-01-23 20:51:12.441982: step: 356/1541, loss: 0.0012320841196924448 2023-01-23 20:51:13.340364: step: 358/1541, loss: 0.0008429645095020533 2023-01-23 20:51:14.228454: step: 360/1541, loss: 0.021316850557923317 2023-01-23 20:51:15.193180: step: 362/1541, loss: 0.0004676127282436937 2023-01-23 20:51:16.034468: step: 364/1541, loss: 4.753136454382911e-05 2023-01-23 20:51:16.921777: step: 366/1541, loss: 0.008531276136636734 2023-01-23 20:51:17.815287: step: 368/1541, loss: 0.050348296761512756 2023-01-23 20:51:18.694317: step: 370/1541, loss: 0.0005111757782287896 2023-01-23 20:51:19.591699: step: 372/1541, loss: 0.10877352207899094 2023-01-23 20:51:20.473611: step: 374/1541, loss: 0.00011177583655808121 2023-01-23 20:51:21.344537: step: 376/1541, loss: 5.218618389335461e-05 2023-01-23 20:51:22.240726: step: 378/1541, loss: 0.00042455631773918867 2023-01-23 20:51:23.122184: step: 380/1541, loss: 0.021218303591012955 2023-01-23 20:51:23.977387: step: 382/1541, loss: 0.0002671888214536011 2023-01-23 20:51:24.873168: step: 384/1541, loss: 2.787267294479534e-05 2023-01-23 20:51:25.750221: step: 386/1541, loss: 0.0011212892131879926 2023-01-23 20:51:26.635304: step: 388/1541, loss: 0.006479067262262106 2023-01-23 20:51:27.562058: step: 390/1541, loss: 0.18802981078624725 2023-01-23 20:51:28.381553: step: 392/1541, loss: 0.00017948384629562497 2023-01-23 20:51:29.301185: step: 394/1541, loss: 0.00018316558271180838 2023-01-23 20:51:30.190868: step: 396/1541, loss: 0.03597750514745712 2023-01-23 20:51:31.112478: step: 398/1541, loss: 0.013156572356820107 2023-01-23 20:51:31.974230: step: 400/1541, loss: 0.0007029626285657287 2023-01-23 20:51:32.856224: step: 402/1541, loss: 0.0031973321456462145 2023-01-23 20:51:33.784855: step: 404/1541, loss: 0.02826135791838169 2023-01-23 20:51:34.666412: step: 406/1541, loss: 0.0020575234666466713 2023-01-23 20:51:35.575923: step: 408/1541, loss: 0.00036741126677952707 2023-01-23 20:51:36.482552: step: 410/1541, loss: 0.00017309709801338613 2023-01-23 20:51:37.371077: step: 412/1541, loss: 0.218618705868721 2023-01-23 20:51:38.360959: step: 414/1541, loss: 0.0005876351497136056 2023-01-23 20:51:39.230906: step: 416/1541, loss: 0.00038491201121360064 2023-01-23 20:51:40.246223: step: 418/1541, loss: 0.0281513798981905 2023-01-23 20:51:41.121853: step: 420/1541, loss: 0.0003082386392634362 2023-01-23 20:51:41.984137: step: 422/1541, loss: 0.10766483098268509 2023-01-23 20:51:42.836902: step: 424/1541, loss: 0.0002239034656668082 2023-01-23 20:51:43.722430: step: 426/1541, loss: 0.00029446661937981844 2023-01-23 20:51:44.597040: step: 428/1541, loss: 0.0013983924873173237 2023-01-23 20:51:45.491576: step: 430/1541, loss: 0.0007649559993296862 2023-01-23 20:51:46.368632: step: 432/1541, loss: 3.265538907726295e-05 2023-01-23 20:51:47.243636: step: 434/1541, loss: 0.006644151173532009 2023-01-23 20:51:48.122095: step: 436/1541, loss: 0.003892486449331045 2023-01-23 20:51:49.018286: step: 438/1541, loss: 0.00027999718440696597 2023-01-23 20:51:49.900136: step: 440/1541, loss: 8.784983947407454e-05 2023-01-23 20:51:50.813255: step: 442/1541, loss: 0.0003299257077742368 2023-01-23 20:51:51.756276: step: 444/1541, loss: 0.024991177022457123 2023-01-23 20:51:52.650445: step: 446/1541, loss: 0.005010606721043587 2023-01-23 20:51:53.508102: step: 448/1541, loss: 0.0002644456399139017 2023-01-23 20:51:54.400419: step: 450/1541, loss: 0.00012682478700298816 2023-01-23 20:51:55.272987: step: 452/1541, loss: 0.005298309028148651 2023-01-23 20:51:56.167708: step: 454/1541, loss: 0.10801304876804352 2023-01-23 20:51:57.104434: step: 456/1541, loss: 0.00019401437020860612 2023-01-23 20:51:58.005408: step: 458/1541, loss: 0.0002971435897052288 2023-01-23 20:51:58.912857: step: 460/1541, loss: 0.042494915425777435 2023-01-23 20:51:59.781548: step: 462/1541, loss: 0.0007651486666873097 2023-01-23 20:52:00.727354: step: 464/1541, loss: 0.08564005047082901 2023-01-23 20:52:01.634482: step: 466/1541, loss: 0.734062910079956 2023-01-23 20:52:02.563259: step: 468/1541, loss: 5.266136577120051e-05 2023-01-23 20:52:03.439789: step: 470/1541, loss: 2.8956637834198773e-05 2023-01-23 20:52:04.349731: step: 472/1541, loss: 0.00020605904865078628 2023-01-23 20:52:05.255462: step: 474/1541, loss: 0.21433699131011963 2023-01-23 20:52:06.088273: step: 476/1541, loss: 0.0004365718050394207 2023-01-23 20:52:06.960239: step: 478/1541, loss: 0.013620329089462757 2023-01-23 20:52:07.853996: step: 480/1541, loss: 0.198518767952919 2023-01-23 20:52:08.744430: step: 482/1541, loss: 0.0005057731177657843 2023-01-23 20:52:09.669917: step: 484/1541, loss: 0.00010278389527229592 2023-01-23 20:52:10.564344: step: 486/1541, loss: 0.0009607518441043794 2023-01-23 20:52:11.442556: step: 488/1541, loss: 0.00020803326333407313 2023-01-23 20:52:12.441133: step: 490/1541, loss: 0.0001999730011448264 2023-01-23 20:52:13.359763: step: 492/1541, loss: 0.00013677289825864136 2023-01-23 20:52:14.306895: step: 494/1541, loss: -1.3112295164319221e-05 2023-01-23 20:52:15.180712: step: 496/1541, loss: 8.41667570057325e-05 2023-01-23 20:52:16.040560: step: 498/1541, loss: 0.0054290760308504105 2023-01-23 20:52:16.937486: step: 500/1541, loss: 0.0008041227702051401 2023-01-23 20:52:17.825317: step: 502/1541, loss: 0.0002121272700605914 2023-01-23 20:52:18.802444: step: 504/1541, loss: 0.09664112329483032 2023-01-23 20:52:19.647148: step: 506/1541, loss: 0.00026512189651839435 2023-01-23 20:52:20.585835: step: 508/1541, loss: 0.16892629861831665 2023-01-23 20:52:21.476606: step: 510/1541, loss: 0.0004547432763502002 2023-01-23 20:52:22.424917: step: 512/1541, loss: 0.00013522052904590964 2023-01-23 20:52:23.326011: step: 514/1541, loss: 0.0021704111713916063 2023-01-23 20:52:24.234413: step: 516/1541, loss: 0.00035126032889820635 2023-01-23 20:52:25.146625: step: 518/1541, loss: 0.002301161177456379 2023-01-23 20:52:26.007288: step: 520/1541, loss: 0.0002572291996330023 2023-01-23 20:52:26.896650: step: 522/1541, loss: 0.012210400775074959 2023-01-23 20:52:27.821811: step: 524/1541, loss: 0.00016865717770997435 2023-01-23 20:52:28.710488: step: 526/1541, loss: 0.00730912433937192 2023-01-23 20:52:29.587655: step: 528/1541, loss: 0.00041976114152930677 2023-01-23 20:52:30.559267: step: 530/1541, loss: 0.021224526688456535 2023-01-23 20:52:31.468324: step: 532/1541, loss: 0.007002943195402622 2023-01-23 20:52:32.331271: step: 534/1541, loss: 0.0027329560834914446 2023-01-23 20:52:33.208218: step: 536/1541, loss: 0.011013412848114967 2023-01-23 20:52:34.139025: step: 538/1541, loss: 0.0012261095689609647 2023-01-23 20:52:35.000117: step: 540/1541, loss: 0.0056032040156424046 2023-01-23 20:52:35.885538: step: 542/1541, loss: 0.02270134724676609 2023-01-23 20:52:36.769052: step: 544/1541, loss: 0.2848830223083496 2023-01-23 20:52:37.649970: step: 546/1541, loss: 6.565019430126995e-05 2023-01-23 20:52:38.615143: step: 548/1541, loss: 0.08959905803203583 2023-01-23 20:52:39.527852: step: 550/1541, loss: 0.00023102462000679225 2023-01-23 20:52:40.448219: step: 552/1541, loss: 0.0005634616245515645 2023-01-23 20:52:41.445860: step: 554/1541, loss: 7.904994708951563e-05 2023-01-23 20:52:42.320679: step: 556/1541, loss: 0.3054051101207733 2023-01-23 20:52:43.179872: step: 558/1541, loss: 0.005875128321349621 2023-01-23 20:52:44.084585: step: 560/1541, loss: 0.0006414131494238973 2023-01-23 20:52:45.012971: step: 562/1541, loss: 0.02022915706038475 2023-01-23 20:52:45.948143: step: 564/1541, loss: 0.05828925967216492 2023-01-23 20:52:46.807940: step: 566/1541, loss: 0.0001027386388159357 2023-01-23 20:52:47.735441: step: 568/1541, loss: 0.015096058137714863 2023-01-23 20:52:48.626853: step: 570/1541, loss: 9.828458132687956e-05 2023-01-23 20:52:49.545961: step: 572/1541, loss: 7.257292600115761e-05 2023-01-23 20:52:50.454609: step: 574/1541, loss: 0.632934033870697 2023-01-23 20:52:51.324138: step: 576/1541, loss: 0.000512563216034323 2023-01-23 20:52:52.224130: step: 578/1541, loss: 0.00010301768634235486 2023-01-23 20:52:53.086965: step: 580/1541, loss: 0.011961942538619041 2023-01-23 20:52:53.966912: step: 582/1541, loss: 0.0017141925636678934 2023-01-23 20:52:54.921810: step: 584/1541, loss: 0.0012350426986813545 2023-01-23 20:52:55.833736: step: 586/1541, loss: 0.006851653568446636 2023-01-23 20:52:56.755126: step: 588/1541, loss: 0.00048812656314112246 2023-01-23 20:52:57.708414: step: 590/1541, loss: 0.004688572604209185 2023-01-23 20:52:58.583266: step: 592/1541, loss: 0.007699321955442429 2023-01-23 20:52:59.557960: step: 594/1541, loss: 0.00011453926708782092 2023-01-23 20:53:00.469716: step: 596/1541, loss: 9.949156810762361e-05 2023-01-23 20:53:01.365081: step: 598/1541, loss: 0.3740856349468231 2023-01-23 20:53:02.254254: step: 600/1541, loss: 0.011970335617661476 2023-01-23 20:53:03.155038: step: 602/1541, loss: 0.01903104968369007 2023-01-23 20:53:04.074859: step: 604/1541, loss: 0.006706113461405039 2023-01-23 20:53:04.952178: step: 606/1541, loss: 0.0014552130596712232 2023-01-23 20:53:05.831954: step: 608/1541, loss: 0.005504705011844635 2023-01-23 20:53:06.830929: step: 610/1541, loss: 0.00023003372189123183 2023-01-23 20:53:07.676738: step: 612/1541, loss: 0.026254579424858093 2023-01-23 20:53:08.608005: step: 614/1541, loss: 0.17746150493621826 2023-01-23 20:53:09.492466: step: 616/1541, loss: 5.532577051781118e-05 2023-01-23 20:53:10.334561: step: 618/1541, loss: 0.0018887746846303344 2023-01-23 20:53:11.209943: step: 620/1541, loss: 0.26728197932243347 2023-01-23 20:53:12.068080: step: 622/1541, loss: 0.0010247916216030717 2023-01-23 20:53:12.981137: step: 624/1541, loss: 0.0006124391220510006 2023-01-23 20:53:13.854080: step: 626/1541, loss: 0.0001445706293452531 2023-01-23 20:53:14.728346: step: 628/1541, loss: 6.130225665401667e-05 2023-01-23 20:53:15.611359: step: 630/1541, loss: 3.189080598531291e-05 2023-01-23 20:53:16.469280: step: 632/1541, loss: 0.07498583197593689 2023-01-23 20:53:17.371759: step: 634/1541, loss: 0.01023029163479805 2023-01-23 20:53:18.249276: step: 636/1541, loss: 0.00018057823763228953 2023-01-23 20:53:19.164973: step: 638/1541, loss: 0.0007250287453643978 2023-01-23 20:53:20.063712: step: 640/1541, loss: 0.00020531089103315026 2023-01-23 20:53:21.024815: step: 642/1541, loss: 0.0279985461384058 2023-01-23 20:53:21.887194: step: 644/1541, loss: 0.01183097530156374 2023-01-23 20:53:22.791209: step: 646/1541, loss: 0.00044490653090178967 2023-01-23 20:53:23.688096: step: 648/1541, loss: 7.748678035568446e-05 2023-01-23 20:53:24.619493: step: 650/1541, loss: 0.0023261853493750095 2023-01-23 20:53:25.505995: step: 652/1541, loss: -1.0164556442759931e-05 2023-01-23 20:53:26.411837: step: 654/1541, loss: 0.4479518532752991 2023-01-23 20:53:27.322655: step: 656/1541, loss: 0.0022410748060792685 2023-01-23 20:53:28.201602: step: 658/1541, loss: 0.11555987596511841 2023-01-23 20:53:29.108748: step: 660/1541, loss: 0.11907108128070831 2023-01-23 20:53:30.014754: step: 662/1541, loss: 0.7446237802505493 2023-01-23 20:53:30.908280: step: 664/1541, loss: 0.08087024092674255 2023-01-23 20:53:31.787892: step: 666/1541, loss: 0.009449128992855549 2023-01-23 20:53:32.662517: step: 668/1541, loss: 0.0015485751209780574 2023-01-23 20:53:33.520981: step: 670/1541, loss: 0.14063455164432526 2023-01-23 20:53:34.469234: step: 672/1541, loss: 0.0003151187556795776 2023-01-23 20:53:35.338492: step: 674/1541, loss: 0.10578972846269608 2023-01-23 20:53:36.268559: step: 676/1541, loss: 0.0003957370645366609 2023-01-23 20:53:37.185965: step: 678/1541, loss: 0.0031498137395828962 2023-01-23 20:53:38.080243: step: 680/1541, loss: 0.00010549161379458383 2023-01-23 20:53:38.954520: step: 682/1541, loss: 0.0011607041815295815 2023-01-23 20:53:39.806860: step: 684/1541, loss: 5.820812293677591e-05 2023-01-23 20:53:40.690992: step: 686/1541, loss: 2.5994768293458037e-05 2023-01-23 20:53:41.559633: step: 688/1541, loss: 9.401548595633358e-06 2023-01-23 20:53:42.585810: step: 690/1541, loss: 0.008988441899418831 2023-01-23 20:53:43.458144: step: 692/1541, loss: 0.001620969269424677 2023-01-23 20:53:44.371627: step: 694/1541, loss: 0.007321220822632313 2023-01-23 20:53:45.266981: step: 696/1541, loss: 0.0009950852254405618 2023-01-23 20:53:46.159810: step: 698/1541, loss: 0.14058484137058258 2023-01-23 20:53:47.076391: step: 700/1541, loss: 0.028624268248677254 2023-01-23 20:53:47.983468: step: 702/1541, loss: 0.01186834555119276 2023-01-23 20:53:48.864421: step: 704/1541, loss: 0.0003027374332305044 2023-01-23 20:53:49.713612: step: 706/1541, loss: 0.0007698649424128234 2023-01-23 20:53:50.616368: step: 708/1541, loss: 0.035818543285131454 2023-01-23 20:53:51.532073: step: 710/1541, loss: 0.0001052686056937091 2023-01-23 20:53:52.440270: step: 712/1541, loss: 3.855327668134123e-05 2023-01-23 20:53:53.387090: step: 714/1541, loss: 0.011217448860406876 2023-01-23 20:53:54.239208: step: 716/1541, loss: 0.0004215584776829928 2023-01-23 20:53:55.167316: step: 718/1541, loss: 0.035308171063661575 2023-01-23 20:53:56.027021: step: 720/1541, loss: 0.508634626865387 2023-01-23 20:53:56.921121: step: 722/1541, loss: 0.0007614953792653978 2023-01-23 20:53:57.844224: step: 724/1541, loss: 0.011216173879802227 2023-01-23 20:53:58.726343: step: 726/1541, loss: 0.0003380347625352442 2023-01-23 20:53:59.626304: step: 728/1541, loss: 0.011974094435572624 2023-01-23 20:54:00.594372: step: 730/1541, loss: 0.2438141405582428 2023-01-23 20:54:01.486782: step: 732/1541, loss: 0.01227005384862423 2023-01-23 20:54:02.434297: step: 734/1541, loss: 0.00032593507785350084 2023-01-23 20:54:03.303297: step: 736/1541, loss: 0.0005175643600523472 2023-01-23 20:54:04.183966: step: 738/1541, loss: 0.00011838448699563742 2023-01-23 20:54:05.062803: step: 740/1541, loss: 0.0014769744593650103 2023-01-23 20:54:05.945148: step: 742/1541, loss: 0.003990615252405405 2023-01-23 20:54:06.847715: step: 744/1541, loss: 0.18106524646282196 2023-01-23 20:54:07.736475: step: 746/1541, loss: 0.00047690121573396027 2023-01-23 20:54:08.622823: step: 748/1541, loss: 0.0038053097669035196 2023-01-23 20:54:09.532372: step: 750/1541, loss: 0.00011024386913049966 2023-01-23 20:54:10.401734: step: 752/1541, loss: 0.026464885100722313 2023-01-23 20:54:11.315478: step: 754/1541, loss: 0.00011837937927339226 2023-01-23 20:54:12.189267: step: 756/1541, loss: 9.118799061980098e-05 2023-01-23 20:54:13.218910: step: 758/1541, loss: 7.728220953140408e-05 2023-01-23 20:54:14.114695: step: 760/1541, loss: 7.783890032442287e-05 2023-01-23 20:54:15.013981: step: 762/1541, loss: 0.00011911439651157707 2023-01-23 20:54:15.927315: step: 764/1541, loss: 0.0713546872138977 2023-01-23 20:54:16.838848: step: 766/1541, loss: 0.012784603051841259 2023-01-23 20:54:17.689804: step: 768/1541, loss: 0.012545359320938587 2023-01-23 20:54:18.558562: step: 770/1541, loss: 0.0023368638940155506 2023-01-23 20:54:19.437470: step: 772/1541, loss: 0.0011727729579433799 2023-01-23 20:54:20.331884: step: 774/1541, loss: 0.0072127776220440865 2023-01-23 20:54:21.272283: step: 776/1541, loss: 0.0001841124176280573 2023-01-23 20:54:22.219431: step: 778/1541, loss: 0.0008955256198532879 2023-01-23 20:54:23.073639: step: 780/1541, loss: 0.0013140912633389235 2023-01-23 20:54:23.955254: step: 782/1541, loss: 0.00792678352445364 2023-01-23 20:54:24.832844: step: 784/1541, loss: 0.07265780121088028 2023-01-23 20:54:25.729874: step: 786/1541, loss: 2.2514124793815427e-05 2023-01-23 20:54:26.621956: step: 788/1541, loss: 0.006298926658928394 2023-01-23 20:54:27.537081: step: 790/1541, loss: 0.1269044131040573 2023-01-23 20:54:28.464474: step: 792/1541, loss: 0.0004530496953520924 2023-01-23 20:54:29.349980: step: 794/1541, loss: 0.5435492992401123 2023-01-23 20:54:30.201033: step: 796/1541, loss: 1.9457256712485105e-05 2023-01-23 20:54:31.062450: step: 798/1541, loss: 0.0012690768344327807 2023-01-23 20:54:31.952008: step: 800/1541, loss: 0.021548567339777946 2023-01-23 20:54:32.835060: step: 802/1541, loss: 0.011786730028688908 2023-01-23 20:54:33.739603: step: 804/1541, loss: 0.001920399721711874 2023-01-23 20:54:34.666220: step: 806/1541, loss: 0.0012796127703040838 2023-01-23 20:54:35.536827: step: 808/1541, loss: 0.12107667326927185 2023-01-23 20:54:36.428430: step: 810/1541, loss: 3.893578832503408e-05 2023-01-23 20:54:37.351164: step: 812/1541, loss: -4.75577235192759e-06 2023-01-23 20:54:38.222580: step: 814/1541, loss: 0.001222920254804194 2023-01-23 20:54:39.146558: step: 816/1541, loss: 0.00027470538043417037 2023-01-23 20:54:40.003914: step: 818/1541, loss: 7.351438398472965e-05 2023-01-23 20:54:40.931460: step: 820/1541, loss: 0.0076974136754870415 2023-01-23 20:54:41.854768: step: 822/1541, loss: 2.1310806914698333e-05 2023-01-23 20:54:42.704646: step: 824/1541, loss: 0.0001928858255269006 2023-01-23 20:54:43.598071: step: 826/1541, loss: 0.027904709801077843 2023-01-23 20:54:44.563270: step: 828/1541, loss: 0.024185897782444954 2023-01-23 20:54:45.480998: step: 830/1541, loss: 0.0004940718063153327 2023-01-23 20:54:46.400025: step: 832/1541, loss: 0.019618792459368706 2023-01-23 20:54:47.250891: step: 834/1541, loss: 0.0008598534623160958 2023-01-23 20:54:48.124592: step: 836/1541, loss: 0.001611558604054153 2023-01-23 20:54:49.011433: step: 838/1541, loss: 0.13798809051513672 2023-01-23 20:54:49.886884: step: 840/1541, loss: 0.0005639190785586834 2023-01-23 20:54:50.782814: step: 842/1541, loss: 0.004700222983956337 2023-01-23 20:54:51.775270: step: 844/1541, loss: 0.028464075177907944 2023-01-23 20:54:52.698363: step: 846/1541, loss: 0.01577700302004814 2023-01-23 20:54:53.604749: step: 848/1541, loss: 0.002834104932844639 2023-01-23 20:54:54.487634: step: 850/1541, loss: 0.4085927903652191 2023-01-23 20:54:55.386488: step: 852/1541, loss: 0.00011268749949522316 2023-01-23 20:54:56.255106: step: 854/1541, loss: 0.0006547432858496904 2023-01-23 20:54:57.148764: step: 856/1541, loss: 0.00041200456325896084 2023-01-23 20:54:58.034186: step: 858/1541, loss: 0.03275186941027641 2023-01-23 20:54:58.926100: step: 860/1541, loss: 0.0031834798865020275 2023-01-23 20:54:59.820571: step: 862/1541, loss: 0.0020894554909318686 2023-01-23 20:55:00.730273: step: 864/1541, loss: 0.0002562272420618683 2023-01-23 20:55:01.605744: step: 866/1541, loss: 0.00017648458015173674 2023-01-23 20:55:02.549982: step: 868/1541, loss: 7.893738802522421e-05 2023-01-23 20:55:03.511861: step: 870/1541, loss: 0.00010671019117580727 2023-01-23 20:55:04.378888: step: 872/1541, loss: 0.0020546342711895704 2023-01-23 20:55:05.281308: step: 874/1541, loss: 0.22021682560443878 2023-01-23 20:55:06.151640: step: 876/1541, loss: 0.3191380500793457 2023-01-23 20:55:07.038632: step: 878/1541, loss: 0.018738945946097374 2023-01-23 20:55:07.965151: step: 880/1541, loss: 0.00021085632033646107 2023-01-23 20:55:08.871034: step: 882/1541, loss: 0.2681218981742859 2023-01-23 20:55:09.792611: step: 884/1541, loss: 0.042132820934057236 2023-01-23 20:55:10.659582: step: 886/1541, loss: 0.1767261028289795 2023-01-23 20:55:11.546020: step: 888/1541, loss: 0.0001665885210968554 2023-01-23 20:55:12.475701: step: 890/1541, loss: 0.008991234004497528 2023-01-23 20:55:13.379056: step: 892/1541, loss: 0.09542547166347504 2023-01-23 20:55:14.282398: step: 894/1541, loss: 0.0006552222184836864 2023-01-23 20:55:15.158382: step: 896/1541, loss: 0.41137316823005676 2023-01-23 20:55:16.110006: step: 898/1541, loss: 0.30118194222450256 2023-01-23 20:55:16.980311: step: 900/1541, loss: 2.2165897462400608e-05 2023-01-23 20:55:17.915723: step: 902/1541, loss: 0.0011061710538342595 2023-01-23 20:55:18.839272: step: 904/1541, loss: 0.2518455684185028 2023-01-23 20:55:19.727808: step: 906/1541, loss: 0.10871373116970062 2023-01-23 20:55:20.597301: step: 908/1541, loss: 0.0006191082648001611 2023-01-23 20:55:21.483511: step: 910/1541, loss: 0.00894853938370943 2023-01-23 20:55:22.369921: step: 912/1541, loss: 0.000447538128355518 2023-01-23 20:55:23.214439: step: 914/1541, loss: 0.00020652412786148489 2023-01-23 20:55:24.154599: step: 916/1541, loss: 0.10008338838815689 2023-01-23 20:55:25.120127: step: 918/1541, loss: 0.0004794076958205551 2023-01-23 20:55:26.015679: step: 920/1541, loss: 0.009933939203619957 2023-01-23 20:55:26.926577: step: 922/1541, loss: 0.004155793227255344 2023-01-23 20:55:27.791713: step: 924/1541, loss: 0.001283707795664668 2023-01-23 20:55:28.715447: step: 926/1541, loss: 0.37706097960472107 2023-01-23 20:55:29.576504: step: 928/1541, loss: 0.0003424258902668953 2023-01-23 20:55:30.455462: step: 930/1541, loss: 0.014915683306753635 2023-01-23 20:55:31.300630: step: 932/1541, loss: 0.0015872283838689327 2023-01-23 20:55:32.299251: step: 934/1541, loss: 0.007734021637588739 2023-01-23 20:55:33.214713: step: 936/1541, loss: 0.00949182640761137 2023-01-23 20:55:34.161599: step: 938/1541, loss: 0.0012992651900276542 2023-01-23 20:55:35.122072: step: 940/1541, loss: 0.09959187358617783 2023-01-23 20:55:36.015833: step: 942/1541, loss: 0.0006533190025947988 2023-01-23 20:55:36.924512: step: 944/1541, loss: 0.0003636754699982703 2023-01-23 20:55:37.793206: step: 946/1541, loss: 0.05278439819812775 2023-01-23 20:55:38.700926: step: 948/1541, loss: 0.1577397733926773 2023-01-23 20:55:39.580785: step: 950/1541, loss: 0.047566212713718414 2023-01-23 20:55:40.526749: step: 952/1541, loss: 0.051353272050619125 2023-01-23 20:55:41.476516: step: 954/1541, loss: 0.0008535129600204527 2023-01-23 20:55:42.341887: step: 956/1541, loss: 0.0003944610361941159 2023-01-23 20:55:43.245433: step: 958/1541, loss: 0.09885665029287338 2023-01-23 20:55:44.122624: step: 960/1541, loss: 0.009689549915492535 2023-01-23 20:55:45.063605: step: 962/1541, loss: 0.006806755904108286 2023-01-23 20:55:45.955025: step: 964/1541, loss: 0.005172341596335173 2023-01-23 20:55:46.844959: step: 966/1541, loss: 0.004630926996469498 2023-01-23 20:55:47.708842: step: 968/1541, loss: 0.4661586880683899 2023-01-23 20:55:48.581343: step: 970/1541, loss: 0.08059528470039368 2023-01-23 20:55:49.498653: step: 972/1541, loss: 0.07278312742710114 2023-01-23 20:55:50.374733: step: 974/1541, loss: 0.00016946325195021927 2023-01-23 20:55:51.301760: step: 976/1541, loss: 0.025291407480835915 2023-01-23 20:55:52.187689: step: 978/1541, loss: 0.00011567889305297285 2023-01-23 20:55:53.092225: step: 980/1541, loss: 0.003398940432816744 2023-01-23 20:55:54.027613: step: 982/1541, loss: 0.007976521737873554 2023-01-23 20:55:54.978568: step: 984/1541, loss: 0.3607330620288849 2023-01-23 20:55:55.822407: step: 986/1541, loss: 0.0005710571422241628 2023-01-23 20:55:56.728140: step: 988/1541, loss: 0.0003887048806063831 2023-01-23 20:55:57.649379: step: 990/1541, loss: 0.015759821981191635 2023-01-23 20:55:58.549501: step: 992/1541, loss: 0.0006096322322264314 2023-01-23 20:55:59.421684: step: 994/1541, loss: 2.636759563756641e-05 2023-01-23 20:56:00.311757: step: 996/1541, loss: 0.0004146481223870069 2023-01-23 20:56:01.184372: step: 998/1541, loss: 0.0027164153289049864 2023-01-23 20:56:02.172178: step: 1000/1541, loss: 0.08113157749176025 2023-01-23 20:56:03.088812: step: 1002/1541, loss: 0.06735852360725403 2023-01-23 20:56:03.938831: step: 1004/1541, loss: 0.0007675622473470867 2023-01-23 20:56:04.841400: step: 1006/1541, loss: 0.0007731748628430068 2023-01-23 20:56:05.730910: step: 1008/1541, loss: 0.06760529428720474 2023-01-23 20:56:06.672221: step: 1010/1541, loss: 0.14811624586582184 2023-01-23 20:56:07.576631: step: 1012/1541, loss: 0.0004761231248266995 2023-01-23 20:56:08.478494: step: 1014/1541, loss: 0.001708093099296093 2023-01-23 20:56:09.350617: step: 1016/1541, loss: 0.003159727668389678 2023-01-23 20:56:10.264745: step: 1018/1541, loss: 0.0473596528172493 2023-01-23 20:56:11.180908: step: 1020/1541, loss: 0.0007454408332705498 2023-01-23 20:56:12.064713: step: 1022/1541, loss: 0.2849040627479553 2023-01-23 20:56:13.014534: step: 1024/1541, loss: 0.0031750171910971403 2023-01-23 20:56:13.853961: step: 1026/1541, loss: 0.008649368770420551 2023-01-23 20:56:14.726475: step: 1028/1541, loss: 0.003159638261422515 2023-01-23 20:56:15.618467: step: 1030/1541, loss: 0.004041976295411587 2023-01-23 20:56:16.486392: step: 1032/1541, loss: 0.00045263467472977936 2023-01-23 20:56:17.359459: step: 1034/1541, loss: 0.00023635580146219581 2023-01-23 20:56:18.244938: step: 1036/1541, loss: 0.0002955969830509275 2023-01-23 20:56:19.152903: step: 1038/1541, loss: 0.034730248153209686 2023-01-23 20:56:20.071865: step: 1040/1541, loss: 6.775550446036505e-06 2023-01-23 20:56:21.014519: step: 1042/1541, loss: 0.19081979990005493 2023-01-23 20:56:21.890317: step: 1044/1541, loss: 0.002174011431634426 2023-01-23 20:56:22.846578: step: 1046/1541, loss: 0.03835834190249443 2023-01-23 20:56:23.739793: step: 1048/1541, loss: 0.00030382740078493953 2023-01-23 20:56:24.630053: step: 1050/1541, loss: 0.0015869280323386192 2023-01-23 20:56:25.545429: step: 1052/1541, loss: 0.022548263892531395 2023-01-23 20:56:26.453949: step: 1054/1541, loss: 0.0018355193315073848 2023-01-23 20:56:27.349570: step: 1056/1541, loss: 0.0037513659335672855 2023-01-23 20:56:28.199739: step: 1058/1541, loss: 7.516861660405993e-05 2023-01-23 20:56:29.128345: step: 1060/1541, loss: 0.0009039075812324882 2023-01-23 20:56:29.989340: step: 1062/1541, loss: 0.00010745100007625297 2023-01-23 20:56:30.978893: step: 1064/1541, loss: 0.017826108261942863 2023-01-23 20:56:31.937197: step: 1066/1541, loss: 0.0021576269064098597 2023-01-23 20:56:32.825063: step: 1068/1541, loss: 0.6395137310028076 2023-01-23 20:56:33.705244: step: 1070/1541, loss: 0.00011751345300581306 2023-01-23 20:56:34.579165: step: 1072/1541, loss: 0.010862614028155804 2023-01-23 20:56:35.455540: step: 1074/1541, loss: 0.010402331128716469 2023-01-23 20:56:36.372332: step: 1076/1541, loss: 0.0020473783370107412 2023-01-23 20:56:37.350079: step: 1078/1541, loss: 5.362587398849428e-05 2023-01-23 20:56:38.202308: step: 1080/1541, loss: 0.00025565983378328383 2023-01-23 20:56:39.155352: step: 1082/1541, loss: 0.0068541476503014565 2023-01-23 20:56:40.094607: step: 1084/1541, loss: 0.4702117145061493 2023-01-23 20:56:40.975305: step: 1086/1541, loss: 0.10896822065114975 2023-01-23 20:56:41.839420: step: 1088/1541, loss: 0.0010288532357662916 2023-01-23 20:56:42.704402: step: 1090/1541, loss: 5.224750202614814e-05 2023-01-23 20:56:43.656795: step: 1092/1541, loss: 0.013874351978302002 2023-01-23 20:56:44.626466: step: 1094/1541, loss: 8.086692832875997e-05 2023-01-23 20:56:45.504323: step: 1096/1541, loss: 0.014590194448828697 2023-01-23 20:56:46.387799: step: 1098/1541, loss: 8.823339885566384e-05 2023-01-23 20:56:47.300031: step: 1100/1541, loss: 0.00022929652186576277 2023-01-23 20:56:48.156297: step: 1102/1541, loss: 0.0002558017731644213 2023-01-23 20:56:49.051118: step: 1104/1541, loss: 6.262300303205848e-05 2023-01-23 20:56:49.985867: step: 1106/1541, loss: 0.00027859830879606307 2023-01-23 20:56:50.909308: step: 1108/1541, loss: 5.378977584769018e-05 2023-01-23 20:56:51.778196: step: 1110/1541, loss: -5.641002644551918e-06 2023-01-23 20:56:52.682030: step: 1112/1541, loss: 0.38315925002098083 2023-01-23 20:56:53.573025: step: 1114/1541, loss: 3.2112984627019614e-07 2023-01-23 20:56:54.488952: step: 1116/1541, loss: 0.034286465495824814 2023-01-23 20:56:55.422519: step: 1118/1541, loss: 0.007332270033657551 2023-01-23 20:56:56.363091: step: 1120/1541, loss: 0.0279623381793499 2023-01-23 20:56:57.259181: step: 1122/1541, loss: 0.050847675651311874 2023-01-23 20:56:58.132536: step: 1124/1541, loss: 0.0014286086661741138 2023-01-23 20:56:59.019139: step: 1126/1541, loss: 0.056352805346250534 2023-01-23 20:56:59.877439: step: 1128/1541, loss: 0.0008927494054660201 2023-01-23 20:57:00.724808: step: 1130/1541, loss: 0.8941194415092468 2023-01-23 20:57:01.581049: step: 1132/1541, loss: 0.021498149260878563 2023-01-23 20:57:02.565546: step: 1134/1541, loss: 0.00022524353698827326 2023-01-23 20:57:03.426984: step: 1136/1541, loss: 0.0003109401441179216 2023-01-23 20:57:04.291228: step: 1138/1541, loss: 0.0010389495873823762 2023-01-23 20:57:05.167098: step: 1140/1541, loss: 0.10457923263311386 2023-01-23 20:57:06.039082: step: 1142/1541, loss: 0.00036005230504088104 2023-01-23 20:57:06.911642: step: 1144/1541, loss: 0.05885860696434975 2023-01-23 20:57:07.798657: step: 1146/1541, loss: 0.004344116430729628 2023-01-23 20:57:08.671676: step: 1148/1541, loss: 0.00013537058839574456 2023-01-23 20:57:09.569233: step: 1150/1541, loss: 0.012174548581242561 2023-01-23 20:57:10.482102: step: 1152/1541, loss: 0.0009736731881275773 2023-01-23 20:57:11.351080: step: 1154/1541, loss: 0.0016868810635060072 2023-01-23 20:57:12.256288: step: 1156/1541, loss: 9.263306128559634e-05 2023-01-23 20:57:13.146292: step: 1158/1541, loss: 0.00015112564142327756 2023-01-23 20:57:14.040511: step: 1160/1541, loss: 0.001241320394910872 2023-01-23 20:57:14.960676: step: 1162/1541, loss: 0.08784817904233932 2023-01-23 20:57:15.833187: step: 1164/1541, loss: 0.013871275819838047 2023-01-23 20:57:16.735143: step: 1166/1541, loss: 0.0019330751383677125 2023-01-23 20:57:17.648619: step: 1168/1541, loss: 1.9691771268844604 2023-01-23 20:57:18.490299: step: 1170/1541, loss: 0.00016874611901585013 2023-01-23 20:57:19.374063: step: 1172/1541, loss: 4.1211398638552055e-05 2023-01-23 20:57:20.300867: step: 1174/1541, loss: 0.019198313355445862 2023-01-23 20:57:21.160715: step: 1176/1541, loss: 0.00013558095088228583 2023-01-23 20:57:22.128370: step: 1178/1541, loss: 0.0009180433116853237 2023-01-23 20:57:23.077687: step: 1180/1541, loss: 0.05307162553071976 2023-01-23 20:57:23.971274: step: 1182/1541, loss: 0.0022654400672763586 2023-01-23 20:57:24.897429: step: 1184/1541, loss: 7.722854934399948e-05 2023-01-23 20:57:25.747869: step: 1186/1541, loss: -1.449584942747606e-05 2023-01-23 20:57:26.611085: step: 1188/1541, loss: -3.953576378989965e-05 2023-01-23 20:57:27.489772: step: 1190/1541, loss: 0.014229650609195232 2023-01-23 20:57:28.437897: step: 1192/1541, loss: 0.00027190669788978994 2023-01-23 20:57:29.317813: step: 1194/1541, loss: 0.03637346625328064 2023-01-23 20:57:30.255546: step: 1196/1541, loss: 0.789290189743042 2023-01-23 20:57:31.155906: step: 1198/1541, loss: -1.2143207641202025e-05 2023-01-23 20:57:32.085414: step: 1200/1541, loss: 0.0026116662193089724 2023-01-23 20:57:32.963773: step: 1202/1541, loss: 0.00016207509906962514 2023-01-23 20:57:33.914910: step: 1204/1541, loss: 0.010295902378857136 2023-01-23 20:57:34.840343: step: 1206/1541, loss: 0.01820451021194458 2023-01-23 20:57:35.690363: step: 1208/1541, loss: 0.021117066964507103 2023-01-23 20:57:36.537148: step: 1210/1541, loss: 0.00013656765804626048 2023-01-23 20:57:37.407021: step: 1212/1541, loss: 0.0006054014666005969 2023-01-23 20:57:38.307061: step: 1214/1541, loss: 0.43101629614830017 2023-01-23 20:57:39.170375: step: 1216/1541, loss: 6.60656369291246e-05 2023-01-23 20:57:40.058244: step: 1218/1541, loss: 0.00011942646960960701 2023-01-23 20:57:40.969438: step: 1220/1541, loss: 0.0005142409936524928 2023-01-23 20:57:41.841252: step: 1222/1541, loss: 0.00021989588276483119 2023-01-23 20:57:42.726553: step: 1224/1541, loss: 0.04904676601290703 2023-01-23 20:57:43.718138: step: 1226/1541, loss: 0.010576926171779633 2023-01-23 20:57:44.600034: step: 1228/1541, loss: 0.03434457629919052 2023-01-23 20:57:45.506673: step: 1230/1541, loss: 0.00022833322873339057 2023-01-23 20:57:46.379599: step: 1232/1541, loss: 0.000277808983810246 2023-01-23 20:57:47.295887: step: 1234/1541, loss: 0.015759816393256187 2023-01-23 20:57:48.167207: step: 1236/1541, loss: 0.2773607671260834 2023-01-23 20:57:49.045289: step: 1238/1541, loss: 0.21249055862426758 2023-01-23 20:57:49.982157: step: 1240/1541, loss: 1.6620445251464844 2023-01-23 20:57:50.882689: step: 1242/1541, loss: 0.03598848357796669 2023-01-23 20:57:51.821544: step: 1244/1541, loss: 0.013760913163423538 2023-01-23 20:57:52.700893: step: 1246/1541, loss: 0.05725674331188202 2023-01-23 20:57:53.605343: step: 1248/1541, loss: 0.05074037238955498 2023-01-23 20:57:54.550426: step: 1250/1541, loss: 0.0009578064200468361 2023-01-23 20:57:55.450173: step: 1252/1541, loss: 0.24415114521980286 2023-01-23 20:57:56.318192: step: 1254/1541, loss: 0.00039958179695531726 2023-01-23 20:57:57.188503: step: 1256/1541, loss: 0.3687843084335327 2023-01-23 20:57:58.064680: step: 1258/1541, loss: 0.17509661614894867 2023-01-23 20:57:58.925962: step: 1260/1541, loss: 0.0033909885678440332 2023-01-23 20:57:59.798039: step: 1262/1541, loss: 0.02656189166009426 2023-01-23 20:58:00.753058: step: 1264/1541, loss: 0.06523459404706955 2023-01-23 20:58:01.687395: step: 1266/1541, loss: 0.0008716428419575095 2023-01-23 20:58:02.645976: step: 1268/1541, loss: 0.03138154745101929 2023-01-23 20:58:03.540764: step: 1270/1541, loss: 0.0002831299207173288 2023-01-23 20:58:04.435027: step: 1272/1541, loss: 0.004167343955487013 2023-01-23 20:58:05.301586: step: 1274/1541, loss: 0.07462435215711594 2023-01-23 20:58:06.205302: step: 1276/1541, loss: 0.3380657136440277 2023-01-23 20:58:07.074746: step: 1278/1541, loss: 0.027690598741173744 2023-01-23 20:58:07.991247: step: 1280/1541, loss: 0.0030008202884346247 2023-01-23 20:58:08.872564: step: 1282/1541, loss: 0.12326871603727341 2023-01-23 20:58:09.759971: step: 1284/1541, loss: 0.04738566279411316 2023-01-23 20:58:10.702556: step: 1286/1541, loss: 0.00484673585742712 2023-01-23 20:58:11.582918: step: 1288/1541, loss: 0.009500779211521149 2023-01-23 20:58:12.454433: step: 1290/1541, loss: 0.0022384067997336388 2023-01-23 20:58:13.358423: step: 1292/1541, loss: 0.0009201390203088522 2023-01-23 20:58:14.235846: step: 1294/1541, loss: 0.0006374649237841368 2023-01-23 20:58:15.122963: step: 1296/1541, loss: 0.0005849961889907718 2023-01-23 20:58:16.011715: step: 1298/1541, loss: 0.0010449510300531983 2023-01-23 20:58:16.922184: step: 1300/1541, loss: 0.03135167807340622 2023-01-23 20:58:17.804178: step: 1302/1541, loss: 0.0002787470875773579 2023-01-23 20:58:18.678525: step: 1304/1541, loss: 0.011025155894458294 2023-01-23 20:58:19.553779: step: 1306/1541, loss: 0.002503421390429139 2023-01-23 20:58:20.469639: step: 1308/1541, loss: 0.11030849069356918 2023-01-23 20:58:21.320919: step: 1310/1541, loss: 0.005195175763219595 2023-01-23 20:58:22.252909: step: 1312/1541, loss: 0.0005538333207368851 2023-01-23 20:58:23.153668: step: 1314/1541, loss: 0.007973814383149147 2023-01-23 20:58:24.061229: step: 1316/1541, loss: 0.3425777852535248 2023-01-23 20:58:24.945637: step: 1318/1541, loss: 0.00012455224350560457 2023-01-23 20:58:25.849735: step: 1320/1541, loss: 0.00026730215176939964 2023-01-23 20:58:26.808438: step: 1322/1541, loss: 0.2220114916563034 2023-01-23 20:58:27.667864: step: 1324/1541, loss: -1.6355779735022224e-06 2023-01-23 20:58:28.553938: step: 1326/1541, loss: 0.0008254452841356397 2023-01-23 20:58:29.458480: step: 1328/1541, loss: 0.00018235445895697922 2023-01-23 20:58:30.348094: step: 1330/1541, loss: 0.022214610129594803 2023-01-23 20:58:31.246229: step: 1332/1541, loss: 0.005795160308480263 2023-01-23 20:58:32.181492: step: 1334/1541, loss: 0.0007490816060453653 2023-01-23 20:58:33.049057: step: 1336/1541, loss: 0.00015963662008289248 2023-01-23 20:58:34.000668: step: 1338/1541, loss: 0.3776090443134308 2023-01-23 20:58:34.871999: step: 1340/1541, loss: 0.0007600969984196126 2023-01-23 20:58:35.752482: step: 1342/1541, loss: 0.17136122286319733 2023-01-23 20:58:36.669862: step: 1344/1541, loss: 0.030228719115257263 2023-01-23 20:58:37.569242: step: 1346/1541, loss: 2.491170380380936e-05 2023-01-23 20:58:38.436200: step: 1348/1541, loss: 6.64322215016e-05 2023-01-23 20:58:39.336719: step: 1350/1541, loss: 0.01009936910122633 2023-01-23 20:58:40.216417: step: 1352/1541, loss: 0.0406375490128994 2023-01-23 20:58:41.054364: step: 1354/1541, loss: 3.397438194951974e-05 2023-01-23 20:58:41.953441: step: 1356/1541, loss: 0.00015216582687571645 2023-01-23 20:58:42.850461: step: 1358/1541, loss: 0.0005924854194745421 2023-01-23 20:58:43.725460: step: 1360/1541, loss: 0.007034752052277327 2023-01-23 20:58:44.607204: step: 1362/1541, loss: 0.0006970782415010035 2023-01-23 20:58:45.499133: step: 1364/1541, loss: 0.0013900160556659102 2023-01-23 20:58:46.387548: step: 1366/1541, loss: 0.0011822794331237674 2023-01-23 20:58:47.352053: step: 1368/1541, loss: 0.0008783430675975978 2023-01-23 20:58:48.260545: step: 1370/1541, loss: 2.1281909942626953 2023-01-23 20:58:49.166224: step: 1372/1541, loss: 0.03530533239245415 2023-01-23 20:58:50.110482: step: 1374/1541, loss: 0.008496236056089401 2023-01-23 20:58:51.053155: step: 1376/1541, loss: 0.25713056325912476 2023-01-23 20:58:51.932359: step: 1378/1541, loss: 6.300351378740743e-05 2023-01-23 20:58:52.862554: step: 1380/1541, loss: 3.603348523029126e-05 2023-01-23 20:58:53.795359: step: 1382/1541, loss: 0.02406744286417961 2023-01-23 20:58:54.710912: step: 1384/1541, loss: 0.02420426346361637 2023-01-23 20:58:55.638590: step: 1386/1541, loss: 0.0003583460347726941 2023-01-23 20:58:56.620208: step: 1388/1541, loss: 0.007416603621095419 2023-01-23 20:58:57.493652: step: 1390/1541, loss: 0.07327577471733093 2023-01-23 20:58:58.379521: step: 1392/1541, loss: 9.104620403377339e-05 2023-01-23 20:58:59.313581: step: 1394/1541, loss: 0.01738172397017479 2023-01-23 20:59:00.187632: step: 1396/1541, loss: 0.0023863567039370537 2023-01-23 20:59:01.066294: step: 1398/1541, loss: 0.0002685706131160259 2023-01-23 20:59:01.941880: step: 1400/1541, loss: 0.029426584020256996 2023-01-23 20:59:02.939407: step: 1402/1541, loss: 0.0005389255820773542 2023-01-23 20:59:03.835015: step: 1404/1541, loss: 0.15240919589996338 2023-01-23 20:59:04.791821: step: 1406/1541, loss: 0.001842189347371459 2023-01-23 20:59:05.714255: step: 1408/1541, loss: 0.3611629009246826 2023-01-23 20:59:06.630189: step: 1410/1541, loss: 0.06635546684265137 2023-01-23 20:59:07.477284: step: 1412/1541, loss: 0.0008691084804013371 2023-01-23 20:59:08.395865: step: 1414/1541, loss: 0.0008877187501639128 2023-01-23 20:59:09.251998: step: 1416/1541, loss: 0.00034555094316601753 2023-01-23 20:59:10.117830: step: 1418/1541, loss: 0.004203790333122015 2023-01-23 20:59:11.004867: step: 1420/1541, loss: 0.017768017947673798 2023-01-23 20:59:11.876205: step: 1422/1541, loss: 0.0046630981378257275 2023-01-23 20:59:12.742102: step: 1424/1541, loss: 0.0029546753503382206 2023-01-23 20:59:13.618570: step: 1426/1541, loss: 0.18264704942703247 2023-01-23 20:59:14.486893: step: 1428/1541, loss: 0.24884626269340515 2023-01-23 20:59:15.369716: step: 1430/1541, loss: 0.16196057200431824 2023-01-23 20:59:16.282478: step: 1432/1541, loss: 0.008341723121702671 2023-01-23 20:59:17.145135: step: 1434/1541, loss: 0.00022777637059334666 2023-01-23 20:59:18.050155: step: 1436/1541, loss: 0.0036248178221285343 2023-01-23 20:59:18.986263: step: 1438/1541, loss: 0.05410528928041458 2023-01-23 20:59:19.979142: step: 1440/1541, loss: 0.0036272616125643253 2023-01-23 20:59:20.876078: step: 1442/1541, loss: 0.09093426167964935 2023-01-23 20:59:21.827398: step: 1444/1541, loss: 0.04204288125038147 2023-01-23 20:59:22.726016: step: 1446/1541, loss: 0.09390047192573547 2023-01-23 20:59:23.633788: step: 1448/1541, loss: 0.08611136674880981 2023-01-23 20:59:24.616322: step: 1450/1541, loss: 0.00037291404441930354 2023-01-23 20:59:25.491512: step: 1452/1541, loss: 0.010543463751673698 2023-01-23 20:59:26.447181: step: 1454/1541, loss: 0.029452983289957047 2023-01-23 20:59:27.330742: step: 1456/1541, loss: 0.0008391070878133178 2023-01-23 20:59:28.257312: step: 1458/1541, loss: 0.0015728070866316557 2023-01-23 20:59:29.169137: step: 1460/1541, loss: 0.21967868506908417 2023-01-23 20:59:30.135038: step: 1462/1541, loss: 0.026518555358052254 2023-01-23 20:59:31.044260: step: 1464/1541, loss: 0.00014645859482698143 2023-01-23 20:59:31.954362: step: 1466/1541, loss: 0.0014366770628839731 2023-01-23 20:59:32.866709: step: 1468/1541, loss: 0.0003633206360973418 2023-01-23 20:59:33.751636: step: 1470/1541, loss: 0.008034328930079937 2023-01-23 20:59:34.608809: step: 1472/1541, loss: 0.0005979895358905196 2023-01-23 20:59:35.500390: step: 1474/1541, loss: 0.00016214924107771367 2023-01-23 20:59:36.417457: step: 1476/1541, loss: 0.006165842991322279 2023-01-23 20:59:37.311788: step: 1478/1541, loss: 0.01684718020260334 2023-01-23 20:59:38.224421: step: 1480/1541, loss: 0.16334283351898193 2023-01-23 20:59:39.125661: step: 1482/1541, loss: 0.5141158699989319 2023-01-23 20:59:39.997415: step: 1484/1541, loss: 0.001057770336046815 2023-01-23 20:59:40.954590: step: 1486/1541, loss: 0.002183933975175023 2023-01-23 20:59:41.819128: step: 1488/1541, loss: 0.0006139687611721456 2023-01-23 20:59:42.702534: step: 1490/1541, loss: 0.06053713336586952 2023-01-23 20:59:43.621079: step: 1492/1541, loss: 0.050016582012176514 2023-01-23 20:59:44.542104: step: 1494/1541, loss: 0.02013646811246872 2023-01-23 20:59:45.464776: step: 1496/1541, loss: 0.05829484760761261 2023-01-23 20:59:46.349456: step: 1498/1541, loss: 0.06407885998487473 2023-01-23 20:59:47.251017: step: 1500/1541, loss: 0.03721193969249725 2023-01-23 20:59:48.144437: step: 1502/1541, loss: 6.976391887292266e-05 2023-01-23 20:59:49.040463: step: 1504/1541, loss: 0.05652607977390289 2023-01-23 20:59:49.934307: step: 1506/1541, loss: 0.01024524588137865 2023-01-23 20:59:50.884116: step: 1508/1541, loss: 0.00027978545404039323 2023-01-23 20:59:51.760517: step: 1510/1541, loss: 0.0004085524124093354 2023-01-23 20:59:52.672577: step: 1512/1541, loss: 0.10350088775157928 2023-01-23 20:59:53.582732: step: 1514/1541, loss: 0.00010925062088062987 2023-01-23 20:59:54.451143: step: 1516/1541, loss: 0.0620395727455616 2023-01-23 20:59:55.375182: step: 1518/1541, loss: 0.00023272457474377006 2023-01-23 20:59:56.230199: step: 1520/1541, loss: -2.544829658290837e-05 2023-01-23 20:59:57.134324: step: 1522/1541, loss: 0.0012405554298311472 2023-01-23 20:59:58.025605: step: 1524/1541, loss: 0.37409237027168274 2023-01-23 20:59:58.916328: step: 1526/1541, loss: 0.0004120832309126854 2023-01-23 20:59:59.857021: step: 1528/1541, loss: 9.641995711717755e-05 2023-01-23 21:00:00.703982: step: 1530/1541, loss: 0.0019134820904582739 2023-01-23 21:00:01.583925: step: 1532/1541, loss: 0.07625327259302139 2023-01-23 21:00:02.484704: step: 1534/1541, loss: 0.06027067452669144 2023-01-23 21:00:03.374325: step: 1536/1541, loss: 0.00027621391927823424 2023-01-23 21:00:04.311691: step: 1538/1541, loss: 0.005015907343477011 2023-01-23 21:00:05.182253: step: 1540/1541, loss: 0.000495420244988054 2023-01-23 21:00:06.158074: step: 1542/1541, loss: 0.00012090068776160479 2023-01-23 21:00:07.037308: step: 1544/1541, loss: 0.00337231089361012 2023-01-23 21:00:07.946156: step: 1546/1541, loss: 0.0006671040900982916 2023-01-23 21:00:08.856394: step: 1548/1541, loss: 0.005343620199710131 2023-01-23 21:00:09.705100: step: 1550/1541, loss: 6.228972051758319e-05 2023-01-23 21:00:10.551041: step: 1552/1541, loss: 0.0024357757065445185 2023-01-23 21:00:11.447676: step: 1554/1541, loss: 8.240489842137322e-05 2023-01-23 21:00:12.366059: step: 1556/1541, loss: 0.000221906098886393 2023-01-23 21:00:13.265166: step: 1558/1541, loss: 0.0002663128252606839 2023-01-23 21:00:14.252540: step: 1560/1541, loss: 0.00015022144361864775 2023-01-23 21:00:15.193131: step: 1562/1541, loss: 6.826105527579784e-05 2023-01-23 21:00:16.098590: step: 1564/1541, loss: 0.00011144523887196556 2023-01-23 21:00:17.022825: step: 1566/1541, loss: 0.015397683717310429 2023-01-23 21:00:17.989796: step: 1568/1541, loss: 0.0005579429562203586 2023-01-23 21:00:18.914002: step: 1570/1541, loss: 0.0016632410697638988 2023-01-23 21:00:19.848429: step: 1572/1541, loss: 0.08344486355781555 2023-01-23 21:00:20.707048: step: 1574/1541, loss: 3.9941594877745956e-05 2023-01-23 21:00:21.620054: step: 1576/1541, loss: 0.05916443467140198 2023-01-23 21:00:22.509986: step: 1578/1541, loss: 3.6897177778882906e-05 2023-01-23 21:00:23.398209: step: 1580/1541, loss: 6.295001367107034e-05 2023-01-23 21:00:24.408921: step: 1582/1541, loss: 0.0013286452740430832 2023-01-23 21:00:25.338899: step: 1584/1541, loss: 0.22237078845500946 2023-01-23 21:00:26.220124: step: 1586/1541, loss: 0.00038679499994032085 2023-01-23 21:00:27.138179: step: 1588/1541, loss: 0.0009638919145800173 2023-01-23 21:00:28.009927: step: 1590/1541, loss: 0.009265617467463017 2023-01-23 21:00:28.892370: step: 1592/1541, loss: 0.00013616232899948955 2023-01-23 21:00:29.769028: step: 1594/1541, loss: 0.001650877296924591 2023-01-23 21:00:30.644392: step: 1596/1541, loss: 0.04195941239595413 2023-01-23 21:00:31.536396: step: 1598/1541, loss: 0.00014557114627677947 2023-01-23 21:00:32.430873: step: 1600/1541, loss: 0.0003707711584866047 2023-01-23 21:00:33.356880: step: 1602/1541, loss: 0.0011799419298768044 2023-01-23 21:00:34.209992: step: 1604/1541, loss: -2.3797374524292536e-05 2023-01-23 21:00:35.089091: step: 1606/1541, loss: 0.10532215237617493 2023-01-23 21:00:35.947808: step: 1608/1541, loss: 0.0004256390093360096 2023-01-23 21:00:36.832117: step: 1610/1541, loss: 0.14836624264717102 2023-01-23 21:00:37.774705: step: 1612/1541, loss: 0.0031794612295925617 2023-01-23 21:00:38.654388: step: 1614/1541, loss: 0.0053642261773347855 2023-01-23 21:00:39.568621: step: 1616/1541, loss: 0.00020617691916413605 2023-01-23 21:00:40.441172: step: 1618/1541, loss: 0.0039032420609146357 2023-01-23 21:00:41.322381: step: 1620/1541, loss: 0.16189049184322357 2023-01-23 21:00:42.220435: step: 1622/1541, loss: 6.676714110653847e-05 2023-01-23 21:00:43.091821: step: 1624/1541, loss: 0.0007903036312200129 2023-01-23 21:00:43.962332: step: 1626/1541, loss: 4.897903272649273e-05 2023-01-23 21:00:44.891555: step: 1628/1541, loss: 0.00764113524928689 2023-01-23 21:00:45.767988: step: 1630/1541, loss: 0.002384575316682458 2023-01-23 21:00:46.635305: step: 1632/1541, loss: 0.00010682223364710808 2023-01-23 21:00:47.525972: step: 1634/1541, loss: 0.0021567977964878082 2023-01-23 21:00:48.505598: step: 1636/1541, loss: 8.335067832376808e-05 2023-01-23 21:00:49.414713: step: 1638/1541, loss: 0.0021554261911660433 2023-01-23 21:00:50.360750: step: 1640/1541, loss: 0.0004349230439402163 2023-01-23 21:00:51.265307: step: 1642/1541, loss: 0.0002034870267380029 2023-01-23 21:00:52.154201: step: 1644/1541, loss: 0.0012620599009096622 2023-01-23 21:00:53.051738: step: 1646/1541, loss: 0.009390571154654026 2023-01-23 21:00:53.919648: step: 1648/1541, loss: 0.007722593378275633 2023-01-23 21:00:54.806228: step: 1650/1541, loss: 0.1342860609292984 2023-01-23 21:00:55.823194: step: 1652/1541, loss: 0.04871678724884987 2023-01-23 21:00:56.699643: step: 1654/1541, loss: 0.017684679478406906 2023-01-23 21:00:57.607157: step: 1656/1541, loss: 0.04179704561829567 2023-01-23 21:00:58.502044: step: 1658/1541, loss: 0.12052088230848312 2023-01-23 21:00:59.388335: step: 1660/1541, loss: 3.1561961804982275e-05 2023-01-23 21:01:00.279920: step: 1662/1541, loss: 0.001505503780208528 2023-01-23 21:01:01.157494: step: 1664/1541, loss: 0.015471534803509712 2023-01-23 21:01:02.036447: step: 1666/1541, loss: 0.005615405272692442 2023-01-23 21:01:02.949436: step: 1668/1541, loss: 0.000389368535252288 2023-01-23 21:01:03.829310: step: 1670/1541, loss: 0.014777586795389652 2023-01-23 21:01:04.733990: step: 1672/1541, loss: 0.004386621993035078 2023-01-23 21:01:05.601543: step: 1674/1541, loss: 0.01037483848631382 2023-01-23 21:01:06.485080: step: 1676/1541, loss: 0.7433446049690247 2023-01-23 21:01:07.379517: step: 1678/1541, loss: 0.0478799045085907 2023-01-23 21:01:08.243703: step: 1680/1541, loss: 0.0003821362042799592 2023-01-23 21:01:09.165959: step: 1682/1541, loss: 0.38576561212539673 2023-01-23 21:01:10.061676: step: 1684/1541, loss: 0.03909101337194443 2023-01-23 21:01:10.959185: step: 1686/1541, loss: 0.1659930944442749 2023-01-23 21:01:11.888426: step: 1688/1541, loss: 0.00019485276425257325 2023-01-23 21:01:12.752825: step: 1690/1541, loss: 0.0007640371331945062 2023-01-23 21:01:13.646085: step: 1692/1541, loss: 0.19044356048107147 2023-01-23 21:01:14.534807: step: 1694/1541, loss: 0.057304661720991135 2023-01-23 21:01:15.455934: step: 1696/1541, loss: 0.0016943499213084579 2023-01-23 21:01:16.401804: step: 1698/1541, loss: 0.0024266173131763935 2023-01-23 21:01:17.348084: step: 1700/1541, loss: 0.07360801100730896 2023-01-23 21:01:18.256274: step: 1702/1541, loss: 0.006565725896507502 2023-01-23 21:01:19.158862: step: 1704/1541, loss: 0.00026290310779586434 2023-01-23 21:01:20.071147: step: 1706/1541, loss: 0.09546279162168503 2023-01-23 21:01:20.952632: step: 1708/1541, loss: 0.004100848454982042 2023-01-23 21:01:21.824251: step: 1710/1541, loss: 0.00044043955858796835 2023-01-23 21:01:22.712593: step: 1712/1541, loss: 0.011777319014072418 2023-01-23 21:01:23.591247: step: 1714/1541, loss: 0.11110689491033554 2023-01-23 21:01:24.478275: step: 1716/1541, loss: 0.002732980065047741 2023-01-23 21:01:25.399976: step: 1718/1541, loss: 0.001945417607203126 2023-01-23 21:01:26.268910: step: 1720/1541, loss: 0.001205409993417561 2023-01-23 21:01:27.186988: step: 1722/1541, loss: 0.000990667613223195 2023-01-23 21:01:28.060647: step: 1724/1541, loss: 0.0037176143378019333 2023-01-23 21:01:28.929897: step: 1726/1541, loss: 0.06904233992099762 2023-01-23 21:01:29.883176: step: 1728/1541, loss: 0.00045278790639713407 2023-01-23 21:01:30.812891: step: 1730/1541, loss: 0.15053319931030273 2023-01-23 21:01:31.675001: step: 1732/1541, loss: 0.252000629901886 2023-01-23 21:01:32.565466: step: 1734/1541, loss: 0.0050079915672540665 2023-01-23 21:01:33.470697: step: 1736/1541, loss: 0.009813745506107807 2023-01-23 21:01:34.397575: step: 1738/1541, loss: 0.03683032840490341 2023-01-23 21:01:35.315483: step: 1740/1541, loss: 0.06029695272445679 2023-01-23 21:01:36.173194: step: 1742/1541, loss: 0.0012909664073958993 2023-01-23 21:01:37.036394: step: 1744/1541, loss: 0.00486145680770278 2023-01-23 21:01:37.985888: step: 1746/1541, loss: 0.5976241230964661 2023-01-23 21:01:38.860241: step: 1748/1541, loss: 0.00066407211124897 2023-01-23 21:01:39.767422: step: 1750/1541, loss: 0.007758514955639839 2023-01-23 21:01:40.706089: step: 1752/1541, loss: 0.0007173014455474913 2023-01-23 21:01:41.601876: step: 1754/1541, loss: 0.01733381673693657 2023-01-23 21:01:42.523922: step: 1756/1541, loss: 0.0002887882583308965 2023-01-23 21:01:43.424668: step: 1758/1541, loss: 0.006539355963468552 2023-01-23 21:01:44.379139: step: 1760/1541, loss: 2.597957063699141e-05 2023-01-23 21:01:45.270655: step: 1762/1541, loss: 0.00022207427537068725 2023-01-23 21:01:46.159279: step: 1764/1541, loss: 0.9945812225341797 2023-01-23 21:01:47.039743: step: 1766/1541, loss: 0.00023584024165757 2023-01-23 21:01:47.979501: step: 1768/1541, loss: 0.00026030404842458665 2023-01-23 21:01:48.849314: step: 1770/1541, loss: 0.00033275631722062826 2023-01-23 21:01:49.715596: step: 1772/1541, loss: 0.20494475960731506 2023-01-23 21:01:50.640386: step: 1774/1541, loss: 0.007658238522708416 2023-01-23 21:01:51.490749: step: 1776/1541, loss: 0.00030373292975127697 2023-01-23 21:01:52.372112: step: 1778/1541, loss: 0.07282552123069763 2023-01-23 21:01:53.315744: step: 1780/1541, loss: 0.15300419926643372 2023-01-23 21:01:54.287448: step: 1782/1541, loss: 0.1541212499141693 2023-01-23 21:01:55.174972: step: 1784/1541, loss: 0.9785298705101013 2023-01-23 21:01:56.089715: step: 1786/1541, loss: 0.0015337734948843718 2023-01-23 21:01:56.966321: step: 1788/1541, loss: 0.00016373753896914423 2023-01-23 21:01:57.828048: step: 1790/1541, loss: 4.5534248783951625e-05 2023-01-23 21:01:58.727976: step: 1792/1541, loss: 0.0316341333091259 2023-01-23 21:01:59.643899: step: 1794/1541, loss: 0.004439865704625845 2023-01-23 21:02:00.604638: step: 1796/1541, loss: 0.07462546974420547 2023-01-23 21:02:01.509983: step: 1798/1541, loss: 0.0011100907577201724 2023-01-23 21:02:02.457177: step: 1800/1541, loss: 0.24819079041481018 2023-01-23 21:02:03.354330: step: 1802/1541, loss: 0.0007012019050307572 2023-01-23 21:02:04.222854: step: 1804/1541, loss: 0.0004474138841032982 2023-01-23 21:02:05.098061: step: 1806/1541, loss: 0.8186846375465393 2023-01-23 21:02:06.011955: step: 1808/1541, loss: 0.040161605924367905 2023-01-23 21:02:06.883261: step: 1810/1541, loss: 0.00016651309852022678 2023-01-23 21:02:07.867062: step: 1812/1541, loss: 0.029466889798641205 2023-01-23 21:02:08.722372: step: 1814/1541, loss: 0.002169986953958869 2023-01-23 21:02:09.577893: step: 1816/1541, loss: 0.003482372034341097 2023-01-23 21:02:10.458529: step: 1818/1541, loss: 0.06065352261066437 2023-01-23 21:02:11.340990: step: 1820/1541, loss: 0.0002511606435291469 2023-01-23 21:02:12.219600: step: 1822/1541, loss: 0.1567968875169754 2023-01-23 21:02:13.079339: step: 1824/1541, loss: 9.371392661705613e-05 2023-01-23 21:02:13.976145: step: 1826/1541, loss: 0.0007861290941946208 2023-01-23 21:02:14.833256: step: 1828/1541, loss: 0.370914489030838 2023-01-23 21:02:15.733162: step: 1830/1541, loss: 0.0023215822875499725 2023-01-23 21:02:16.671458: step: 1832/1541, loss: 0.00021748495055362582 2023-01-23 21:02:17.574679: step: 1834/1541, loss: -5.624311597784981e-06 2023-01-23 21:02:18.459174: step: 1836/1541, loss: 0.06562268733978271 2023-01-23 21:02:19.357073: step: 1838/1541, loss: 0.9809129238128662 2023-01-23 21:02:20.255634: step: 1840/1541, loss: 0.002516237087547779 2023-01-23 21:02:21.166098: step: 1842/1541, loss: 0.004097909666597843 2023-01-23 21:02:22.013737: step: 1844/1541, loss: 0.0035186256282031536 2023-01-23 21:02:22.887987: step: 1846/1541, loss: 0.0013754601823166013 2023-01-23 21:02:23.804222: step: 1848/1541, loss: 0.0009914773982018232 2023-01-23 21:02:24.689911: step: 1850/1541, loss: 0.08456594496965408 2023-01-23 21:02:25.596263: step: 1852/1541, loss: 0.018975213170051575 2023-01-23 21:02:26.506492: step: 1854/1541, loss: 0.0008047979208640754 2023-01-23 21:02:27.363557: step: 1856/1541, loss: 8.992476796265692e-05 2023-01-23 21:02:28.308665: step: 1858/1541, loss: 0.00021031191863585263 2023-01-23 21:02:29.216465: step: 1860/1541, loss: 0.029843563213944435 2023-01-23 21:02:30.152000: step: 1862/1541, loss: 0.006962975487112999 2023-01-23 21:02:31.038435: step: 1864/1541, loss: 0.010659502819180489 2023-01-23 21:02:31.947837: step: 1866/1541, loss: 4.430858098203316e-05 2023-01-23 21:02:32.810666: step: 1868/1541, loss: 0.0001798942539608106 2023-01-23 21:02:33.687913: step: 1870/1541, loss: 0.00040728889871388674 2023-01-23 21:02:34.568907: step: 1872/1541, loss: 0.0002473511267453432 2023-01-23 21:02:35.494376: step: 1874/1541, loss: 8.856115891830996e-05 2023-01-23 21:02:36.384488: step: 1876/1541, loss: 0.0008453624905087054 2023-01-23 21:02:37.290048: step: 1878/1541, loss: 0.14567382633686066 2023-01-23 21:02:38.162774: step: 1880/1541, loss: 0.008471604436635971 2023-01-23 21:02:39.025479: step: 1882/1541, loss: 0.0002660913742147386 2023-01-23 21:02:39.896325: step: 1884/1541, loss: 0.001817651791498065 2023-01-23 21:02:40.785442: step: 1886/1541, loss: 0.0004034265875816345 2023-01-23 21:02:41.682522: step: 1888/1541, loss: 4.791294122696854e-05 2023-01-23 21:02:42.595191: step: 1890/1541, loss: 0.0001608108141226694 2023-01-23 21:02:43.486700: step: 1892/1541, loss: 0.0007503569941036403 2023-01-23 21:02:44.415504: step: 1894/1541, loss: 0.9512805342674255 2023-01-23 21:02:45.329838: step: 1896/1541, loss: 0.0007506986730732024 2023-01-23 21:02:46.254370: step: 1898/1541, loss: 2.460958957672119 2023-01-23 21:02:47.172836: step: 1900/1541, loss: 0.004200738389045 2023-01-23 21:02:48.099068: step: 1902/1541, loss: 0.0003938743902835995 2023-01-23 21:02:49.001922: step: 1904/1541, loss: 0.33915504813194275 2023-01-23 21:02:49.943869: step: 1906/1541, loss: 0.003479970619082451 2023-01-23 21:02:50.826821: step: 1908/1541, loss: 0.0019129259744659066 2023-01-23 21:02:51.726761: step: 1910/1541, loss: 1.5435773093486205e-05 2023-01-23 21:02:52.723620: step: 1912/1541, loss: 0.002518652006983757 2023-01-23 21:02:53.645583: step: 1914/1541, loss: 0.003090082434937358 2023-01-23 21:02:54.519551: step: 1916/1541, loss: 0.0015233713202178478 2023-01-23 21:02:55.365527: step: 1918/1541, loss: 0.001931767095811665 2023-01-23 21:02:56.296133: step: 1920/1541, loss: 0.0011043979320675135 2023-01-23 21:02:57.181506: step: 1922/1541, loss: 0.000316602032398805 2023-01-23 21:02:58.065143: step: 1924/1541, loss: 0.0009148772223852575 2023-01-23 21:02:58.948339: step: 1926/1541, loss: 0.0008193131070584059 2023-01-23 21:02:59.838562: step: 1928/1541, loss: 0.0002368515415582806 2023-01-23 21:03:00.685901: step: 1930/1541, loss: 8.720999903744087e-05 2023-01-23 21:03:01.594936: step: 1932/1541, loss: 1.1112570762634277 2023-01-23 21:03:02.485826: step: 1934/1541, loss: 0.01879606954753399 2023-01-23 21:03:03.341473: step: 1936/1541, loss: 0.00011036574142053723 2023-01-23 21:03:04.241619: step: 1938/1541, loss: 0.3105608820915222 2023-01-23 21:03:05.185363: step: 1940/1541, loss: 0.08482371270656586 2023-01-23 21:03:06.090515: step: 1942/1541, loss: 0.006188218016177416 2023-01-23 21:03:07.109304: step: 1944/1541, loss: 0.0014182078884914517 2023-01-23 21:03:08.042967: step: 1946/1541, loss: 6.565015792148188e-05 2023-01-23 21:03:08.895189: step: 1948/1541, loss: 0.0003259098739363253 2023-01-23 21:03:09.801849: step: 1950/1541, loss: 0.00075714779086411 2023-01-23 21:03:10.706417: step: 1952/1541, loss: 0.1388889104127884 2023-01-23 21:03:11.584601: step: 1954/1541, loss: 0.0003543974307831377 2023-01-23 21:03:12.478112: step: 1956/1541, loss: 0.0003796659002546221 2023-01-23 21:03:13.343684: step: 1958/1541, loss: 8.20495915831998e-05 2023-01-23 21:03:14.296790: step: 1960/1541, loss: 0.009123330935835838 2023-01-23 21:03:15.235030: step: 1962/1541, loss: 0.17539751529693604 2023-01-23 21:03:16.112086: step: 1964/1541, loss: 0.00015333894407376647 2023-01-23 21:03:16.976228: step: 1966/1541, loss: 0.11763500422239304 2023-01-23 21:03:17.884254: step: 1968/1541, loss: 0.001443742774426937 2023-01-23 21:03:18.791583: step: 1970/1541, loss: 0.025830186903476715 2023-01-23 21:03:19.691456: step: 1972/1541, loss: 0.00016556071932427585 2023-01-23 21:03:20.570578: step: 1974/1541, loss: 0.0006654965691268444 2023-01-23 21:03:21.473470: step: 1976/1541, loss: 0.1091451644897461 2023-01-23 21:03:22.375201: step: 1978/1541, loss: 0.000360035861376673 2023-01-23 21:03:23.327711: step: 1980/1541, loss: 0.0035021991934627295 2023-01-23 21:03:24.193172: step: 1982/1541, loss: 0.00959630124270916 2023-01-23 21:03:25.078251: step: 1984/1541, loss: 0.008188823238015175 2023-01-23 21:03:25.995222: step: 1986/1541, loss: 0.013666758313775063 2023-01-23 21:03:26.862303: step: 1988/1541, loss: 0.11187081038951874 2023-01-23 21:03:27.737850: step: 1990/1541, loss: 0.1603558212518692 2023-01-23 21:03:28.690513: step: 1992/1541, loss: 0.05570610240101814 2023-01-23 21:03:29.585735: step: 1994/1541, loss: 0.012276445515453815 2023-01-23 21:03:30.574499: step: 1996/1541, loss: 0.003171505406498909 2023-01-23 21:03:31.459138: step: 1998/1541, loss: 0.020911922678351402 2023-01-23 21:03:32.362762: step: 2000/1541, loss: 0.0010310356738045812 2023-01-23 21:03:33.234286: step: 2002/1541, loss: 0.00079062842996791 2023-01-23 21:03:34.149474: step: 2004/1541, loss: 0.004404530860483646 2023-01-23 21:03:35.083873: step: 2006/1541, loss: 0.0004331833915784955 2023-01-23 21:03:35.953851: step: 2008/1541, loss: 7.527451816713437e-05 2023-01-23 21:03:36.831327: step: 2010/1541, loss: -1.8477980120223947e-05 2023-01-23 21:03:37.802909: step: 2012/1541, loss: 0.012478180229663849 2023-01-23 21:03:38.679593: step: 2014/1541, loss: 0.06490474939346313 2023-01-23 21:03:39.525752: step: 2016/1541, loss: 7.49010723666288e-05 2023-01-23 21:03:40.399246: step: 2018/1541, loss: 0.000660348916426301 2023-01-23 21:03:41.262527: step: 2020/1541, loss: 0.0008783364901319146 2023-01-23 21:03:42.169070: step: 2022/1541, loss: 0.06087355315685272 2023-01-23 21:03:43.063733: step: 2024/1541, loss: 0.00233806436881423 2023-01-23 21:03:43.919588: step: 2026/1541, loss: 0.022751929238438606 2023-01-23 21:03:44.832377: step: 2028/1541, loss: 0.00013488568947650492 2023-01-23 21:03:45.708840: step: 2030/1541, loss: 0.015605799853801727 2023-01-23 21:03:46.594700: step: 2032/1541, loss: 0.001699887216091156 2023-01-23 21:03:47.510291: step: 2034/1541, loss: 0.0015704373363405466 2023-01-23 21:03:48.396646: step: 2036/1541, loss: 2.9793049179716036e-05 2023-01-23 21:03:49.278115: step: 2038/1541, loss: 0.030211085453629494 2023-01-23 21:03:50.153273: step: 2040/1541, loss: 6.843359005870298e-05 2023-01-23 21:03:51.028821: step: 2042/1541, loss: 0.07397761940956116 2023-01-23 21:03:51.882881: step: 2044/1541, loss: 0.5517488121986389 2023-01-23 21:03:52.754646: step: 2046/1541, loss: 0.016056740656495094 2023-01-23 21:03:53.680060: step: 2048/1541, loss: 0.01880820281803608 2023-01-23 21:03:54.531013: step: 2050/1541, loss: 1.5727593563497066e-05 2023-01-23 21:03:55.427061: step: 2052/1541, loss: 0.0030301494989544153 2023-01-23 21:03:56.284696: step: 2054/1541, loss: 0.00037387709016911685 2023-01-23 21:03:57.157954: step: 2056/1541, loss: 0.0015086372150108218 2023-01-23 21:03:58.043903: step: 2058/1541, loss: 0.00012327710282988846 2023-01-23 21:03:58.919199: step: 2060/1541, loss: 0.0007210184121504426 2023-01-23 21:03:59.761317: step: 2062/1541, loss: 0.0002540885761845857 2023-01-23 21:04:00.639738: step: 2064/1541, loss: 0.0046814060769975185 2023-01-23 21:04:01.540839: step: 2066/1541, loss: 0.0011362534714862704 2023-01-23 21:04:02.401504: step: 2068/1541, loss: 1.773325857357122e-05 2023-01-23 21:04:03.294372: step: 2070/1541, loss: 5.816491830046289e-05 2023-01-23 21:04:04.149698: step: 2072/1541, loss: 0.02320931851863861 2023-01-23 21:04:05.034344: step: 2074/1541, loss: 0.21948282420635223 2023-01-23 21:04:05.915760: step: 2076/1541, loss: 0.004403292201459408 2023-01-23 21:04:06.814721: step: 2078/1541, loss: 0.16038832068443298 2023-01-23 21:04:07.711413: step: 2080/1541, loss: 0.010829001665115356 2023-01-23 21:04:08.577056: step: 2082/1541, loss: 0.000668759923428297 2023-01-23 21:04:09.453332: step: 2084/1541, loss: 0.021340588107705116 2023-01-23 21:04:10.348918: step: 2086/1541, loss: 0.11857849359512329 2023-01-23 21:04:11.256650: step: 2088/1541, loss: 0.24379077553749084 2023-01-23 21:04:12.151558: step: 2090/1541, loss: 0.0002697905001696199 2023-01-23 21:04:13.074108: step: 2092/1541, loss: 0.008390659466385841 2023-01-23 21:04:13.984332: step: 2094/1541, loss: 1.4164295862428844e-05 2023-01-23 21:04:14.905219: step: 2096/1541, loss: 0.07243356108665466 2023-01-23 21:04:15.794474: step: 2098/1541, loss: 5.5777276429580525e-05 2023-01-23 21:04:16.774824: step: 2100/1541, loss: -0.00011453368642833084 2023-01-23 21:04:17.653437: step: 2102/1541, loss: 0.059307098388671875 2023-01-23 21:04:18.534500: step: 2104/1541, loss: 0.0006901364540681243 2023-01-23 21:04:19.425295: step: 2106/1541, loss: 0.02261776477098465 2023-01-23 21:04:20.329020: step: 2108/1541, loss: 0.002083877567201853 2023-01-23 21:04:21.228791: step: 2110/1541, loss: 1.1624775652308017e-05 2023-01-23 21:04:22.153929: step: 2112/1541, loss: 0.00037051885738037527 2023-01-23 21:04:23.037705: step: 2114/1541, loss: 0.005758396349847317 2023-01-23 21:04:23.984661: step: 2116/1541, loss: 0.0004872425342909992 2023-01-23 21:04:24.855932: step: 2118/1541, loss: 0.00031841470627114177 2023-01-23 21:04:25.743586: step: 2120/1541, loss: 0.00026683375472202897 2023-01-23 21:04:26.627865: step: 2122/1541, loss: 0.005507184192538261 2023-01-23 21:04:27.605482: step: 2124/1541, loss: 0.006988778244704008 2023-01-23 21:04:28.520258: step: 2126/1541, loss: 0.00038740618037991226 2023-01-23 21:04:29.437554: step: 2128/1541, loss: 0.001533792121335864 2023-01-23 21:04:30.325807: step: 2130/1541, loss: 0.0001358166482532397 2023-01-23 21:04:31.237832: step: 2132/1541, loss: 9.69364627962932e-05 2023-01-23 21:04:32.220282: step: 2134/1541, loss: 0.00030074064852669835 2023-01-23 21:04:33.129677: step: 2136/1541, loss: 0.01943778805434704 2023-01-23 21:04:34.011900: step: 2138/1541, loss: 0.000770963728427887 2023-01-23 21:04:34.901166: step: 2140/1541, loss: 0.0004179503594059497 2023-01-23 21:04:35.831323: step: 2142/1541, loss: 0.01613067276775837 2023-01-23 21:04:36.717196: step: 2144/1541, loss: 4.4745203922502697e-05 2023-01-23 21:04:37.669695: step: 2146/1541, loss: 0.003916763700544834 2023-01-23 21:04:38.581677: step: 2148/1541, loss: 0.00011997098772553727 2023-01-23 21:04:39.472036: step: 2150/1541, loss: 0.002750799059867859 2023-01-23 21:04:40.349999: step: 2152/1541, loss: 0.012937474995851517 2023-01-23 21:04:41.253582: step: 2154/1541, loss: 0.00011043498670915142 2023-01-23 21:04:42.193719: step: 2156/1541, loss: 0.04316571727395058 2023-01-23 21:04:43.084599: step: 2158/1541, loss: 2.5044484573299997e-05 2023-01-23 21:04:44.061776: step: 2160/1541, loss: 0.01150202564895153 2023-01-23 21:04:44.971703: step: 2162/1541, loss: 0.3071189522743225 2023-01-23 21:04:45.854540: step: 2164/1541, loss: 0.01353787537664175 2023-01-23 21:04:46.723327: step: 2166/1541, loss: 0.0025922602508217096 2023-01-23 21:04:47.626091: step: 2168/1541, loss: 0.0004074626776855439 2023-01-23 21:04:48.472301: step: 2170/1541, loss: 0.06823848187923431 2023-01-23 21:04:49.389180: step: 2172/1541, loss: 0.5314806699752808 2023-01-23 21:04:50.288387: step: 2174/1541, loss: 0.008039591833949089 2023-01-23 21:04:51.204252: step: 2176/1541, loss: 0.0003611837746575475 2023-01-23 21:04:52.119334: step: 2178/1541, loss: 0.02042776346206665 2023-01-23 21:04:53.035331: step: 2180/1541, loss: 0.012895723804831505 2023-01-23 21:04:53.983998: step: 2182/1541, loss: 0.1439099907875061 2023-01-23 21:04:54.920565: step: 2184/1541, loss: 0.0004498461203183979 2023-01-23 21:04:55.819247: step: 2186/1541, loss: 0.06607680767774582 2023-01-23 21:04:56.733090: step: 2188/1541, loss: 0.0005918190581724048 2023-01-23 21:04:57.595328: step: 2190/1541, loss: 0.0009666761616244912 2023-01-23 21:04:58.517027: step: 2192/1541, loss: 0.19789694249629974 2023-01-23 21:04:59.456486: step: 2194/1541, loss: 0.055346108973026276 2023-01-23 21:05:00.388797: step: 2196/1541, loss: 0.3335101902484894 2023-01-23 21:05:01.309078: step: 2198/1541, loss: 0.00040953076677396894 2023-01-23 21:05:02.214301: step: 2200/1541, loss: 0.0017488303128629923 2023-01-23 21:05:03.116991: step: 2202/1541, loss: 0.0016627791337668896 2023-01-23 21:05:04.049276: step: 2204/1541, loss: 0.10583733022212982 2023-01-23 21:05:04.903211: step: 2206/1541, loss: 0.0010616210056468844 2023-01-23 21:05:05.759527: step: 2208/1541, loss: 0.012228269129991531 2023-01-23 21:05:06.609004: step: 2210/1541, loss: 0.013928323052823544 2023-01-23 21:05:07.493553: step: 2212/1541, loss: 0.0007826122455298901 2023-01-23 21:05:08.395809: step: 2214/1541, loss: 0.10801888257265091 2023-01-23 21:05:09.311965: step: 2216/1541, loss: 0.00014343632210511714 2023-01-23 21:05:10.253389: step: 2218/1541, loss: 0.00262240506708622 2023-01-23 21:05:11.212113: step: 2220/1541, loss: 0.011727680452167988 2023-01-23 21:05:12.141019: step: 2222/1541, loss: 0.0038162071723490953 2023-01-23 21:05:13.003914: step: 2224/1541, loss: 0.033762574195861816 2023-01-23 21:05:13.872009: step: 2226/1541, loss: 0.19650708138942719 2023-01-23 21:05:14.791449: step: 2228/1541, loss: 0.0003276894858572632 2023-01-23 21:05:15.725155: step: 2230/1541, loss: 0.09534882754087448 2023-01-23 21:05:16.604767: step: 2232/1541, loss: 0.08393392711877823 2023-01-23 21:05:17.510894: step: 2234/1541, loss: 0.24655017256736755 2023-01-23 21:05:18.437689: step: 2236/1541, loss: 0.5504010915756226 2023-01-23 21:05:19.347498: step: 2238/1541, loss: 0.44664138555526733 2023-01-23 21:05:20.272252: step: 2240/1541, loss: 0.00026305188657715917 2023-01-23 21:05:21.143015: step: 2242/1541, loss: 0.37502509355545044 2023-01-23 21:05:22.029875: step: 2244/1541, loss: 0.00012641203647945076 2023-01-23 21:05:22.949364: step: 2246/1541, loss: 0.000837436702568084 2023-01-23 21:05:23.841002: step: 2248/1541, loss: 0.00014597653353121132 2023-01-23 21:05:24.724665: step: 2250/1541, loss: 0.5290601849555969 2023-01-23 21:05:25.607267: step: 2252/1541, loss: 0.0009760178509168327 2023-01-23 21:05:26.505903: step: 2254/1541, loss: 0.06260788440704346 2023-01-23 21:05:27.495680: step: 2256/1541, loss: 0.06282661855220795 2023-01-23 21:05:28.371486: step: 2258/1541, loss: 0.1681995689868927 2023-01-23 21:05:29.278739: step: 2260/1541, loss: 0.007001316640526056 2023-01-23 21:05:30.178285: step: 2262/1541, loss: 0.00046893753460608423 2023-01-23 21:05:31.069649: step: 2264/1541, loss: 0.02999526634812355 2023-01-23 21:05:31.963154: step: 2266/1541, loss: 0.009075513109564781 2023-01-23 21:05:32.850294: step: 2268/1541, loss: 0.003473112592473626 2023-01-23 21:05:33.756149: step: 2270/1541, loss: 0.0019634347409009933 2023-01-23 21:05:34.659600: step: 2272/1541, loss: 0.10388544201850891 2023-01-23 21:05:35.545018: step: 2274/1541, loss: 0.006173858884721994 2023-01-23 21:05:36.420253: step: 2276/1541, loss: 0.0007474822923541069 2023-01-23 21:05:37.430259: step: 2278/1541, loss: 0.0997902974486351 2023-01-23 21:05:38.344345: step: 2280/1541, loss: 0.009997479617595673 2023-01-23 21:05:39.282089: step: 2282/1541, loss: 0.00021628840477205813 2023-01-23 21:05:40.217240: step: 2284/1541, loss: 0.07459178566932678 2023-01-23 21:05:41.167177: step: 2286/1541, loss: 0.0015949938679113984 2023-01-23 21:05:42.139615: step: 2288/1541, loss: 0.007228992413729429 2023-01-23 21:05:43.105976: step: 2290/1541, loss: 0.0806230753660202 2023-01-23 21:05:44.014400: step: 2292/1541, loss: 0.0027365086134523153 2023-01-23 21:05:44.914060: step: 2294/1541, loss: 2.3210350263980217e-05 2023-01-23 21:05:45.786567: step: 2296/1541, loss: 0.000378824450308457 2023-01-23 21:05:46.644120: step: 2298/1541, loss: 0.002179330214858055 2023-01-23 21:05:47.541165: step: 2300/1541, loss: 7.172275218181312e-05 2023-01-23 21:05:48.432181: step: 2302/1541, loss: 0.0008623221074230969 2023-01-23 21:05:49.289544: step: 2304/1541, loss: 0.2168785035610199 2023-01-23 21:05:50.170926: step: 2306/1541, loss: 0.0005627223872579634 2023-01-23 21:05:51.118136: step: 2308/1541, loss: 0.11932511627674103 2023-01-23 21:05:51.973675: step: 2310/1541, loss: 0.0594424307346344 2023-01-23 21:05:52.858288: step: 2312/1541, loss: 0.00011281970364507288 2023-01-23 21:05:53.786177: step: 2314/1541, loss: 0.25565677881240845 2023-01-23 21:05:54.789146: step: 2316/1541, loss: 0.04087074100971222 2023-01-23 21:05:55.711672: step: 2318/1541, loss: 0.00130305381026119 2023-01-23 21:05:56.671038: step: 2320/1541, loss: 0.0020811990834772587 2023-01-23 21:05:57.561937: step: 2322/1541, loss: 4.5022305130260065e-05 2023-01-23 21:05:58.440908: step: 2324/1541, loss: 0.6012044548988342 2023-01-23 21:05:59.325706: step: 2326/1541, loss: 0.0919756293296814 2023-01-23 21:06:00.256429: step: 2328/1541, loss: 0.0001954015315277502 2023-01-23 21:06:01.183485: step: 2330/1541, loss: 0.0041128541342914104 2023-01-23 21:06:02.214096: step: 2332/1541, loss: 0.006794918328523636 2023-01-23 21:06:03.183116: step: 2334/1541, loss: 0.03482368588447571 2023-01-23 21:06:04.126850: step: 2336/1541, loss: 0.002501293085515499 2023-01-23 21:06:05.028266: step: 2338/1541, loss: 7.005575025686994e-05 2023-01-23 21:06:05.905891: step: 2340/1541, loss: 0.0032455571927130222 2023-01-23 21:06:06.815162: step: 2342/1541, loss: 0.03058745339512825 2023-01-23 21:06:07.742285: step: 2344/1541, loss: 0.03851528465747833 2023-01-23 21:06:08.627243: step: 2346/1541, loss: 0.007906323298811913 2023-01-23 21:06:09.531337: step: 2348/1541, loss: 0.14289319515228271 2023-01-23 21:06:10.450208: step: 2350/1541, loss: 0.0002985708706546575 2023-01-23 21:06:11.365536: step: 2352/1541, loss: 0.00014425403787754476 2023-01-23 21:06:12.251524: step: 2354/1541, loss: 0.0014991668285802007 2023-01-23 21:06:13.209250: step: 2356/1541, loss: 1.62049473146908e-05 2023-01-23 21:06:14.085108: step: 2358/1541, loss: 0.001488035311922431 2023-01-23 21:06:14.972906: step: 2360/1541, loss: 0.16128070652484894 2023-01-23 21:06:15.852845: step: 2362/1541, loss: 0.002592226956039667 2023-01-23 21:06:16.789758: step: 2364/1541, loss: 0.10717310011386871 2023-01-23 21:06:17.732146: step: 2366/1541, loss: 0.6729015707969666 2023-01-23 21:06:18.658747: step: 2368/1541, loss: 0.0012045801850035787 2023-01-23 21:06:19.642331: step: 2370/1541, loss: 0.08240304887294769 2023-01-23 21:06:20.555905: step: 2372/1541, loss: 0.23594214022159576 2023-01-23 21:06:21.472835: step: 2374/1541, loss: 0.11580789089202881 2023-01-23 21:06:22.353266: step: 2376/1541, loss: 0.001207619672641158 2023-01-23 21:06:23.289988: step: 2378/1541, loss: 0.11355260014533997 2023-01-23 21:06:24.180165: step: 2380/1541, loss: 0.0002515214146114886 2023-01-23 21:06:25.069388: step: 2382/1541, loss: 0.0002498755347914994 2023-01-23 21:06:25.933663: step: 2384/1541, loss: 8.521004201611504e-05 2023-01-23 21:06:26.842154: step: 2386/1541, loss: 5.355870962375775e-05 2023-01-23 21:06:27.764025: step: 2388/1541, loss: 0.00016214547213166952 2023-01-23 21:06:28.667730: step: 2390/1541, loss: 0.0024789839517325163 2023-01-23 21:06:29.556498: step: 2392/1541, loss: 0.0003392633225303143 2023-01-23 21:06:30.425849: step: 2394/1541, loss: 0.04966026544570923 2023-01-23 21:06:31.339557: step: 2396/1541, loss: 0.0011886443244293332 2023-01-23 21:06:32.234821: step: 2398/1541, loss: 0.0003677163040265441 2023-01-23 21:06:33.122137: step: 2400/1541, loss: 0.1269700825214386 2023-01-23 21:06:34.003887: step: 2402/1541, loss: 0.0004154342459514737 2023-01-23 21:06:34.922198: step: 2404/1541, loss: 0.09001912921667099 2023-01-23 21:06:35.824499: step: 2406/1541, loss: 0.0001540578668937087 2023-01-23 21:06:36.706903: step: 2408/1541, loss: 0.01879260689020157 2023-01-23 21:06:37.595485: step: 2410/1541, loss: 1.2852407693862915 2023-01-23 21:06:38.454330: step: 2412/1541, loss: 0.030949806794524193 2023-01-23 21:06:39.308068: step: 2414/1541, loss: 0.0001211404669447802 2023-01-23 21:06:40.245773: step: 2416/1541, loss: 0.2854081094264984 2023-01-23 21:06:41.140346: step: 2418/1541, loss: 0.01355732511729002 2023-01-23 21:06:41.989408: step: 2420/1541, loss: 0.0004940758808515966 2023-01-23 21:06:42.912018: step: 2422/1541, loss: 0.1468416303396225 2023-01-23 21:06:43.834787: step: 2424/1541, loss: 0.03198198229074478 2023-01-23 21:06:44.706918: step: 2426/1541, loss: 0.00436908844858408 2023-01-23 21:06:45.613137: step: 2428/1541, loss: 0.00022086779063101858 2023-01-23 21:06:46.520276: step: 2430/1541, loss: 0.17720936238765717 2023-01-23 21:06:47.450917: step: 2432/1541, loss: 0.05978265032172203 2023-01-23 21:06:48.370547: step: 2434/1541, loss: 0.004481338430196047 2023-01-23 21:06:49.256615: step: 2436/1541, loss: 0.04645082727074623 2023-01-23 21:06:50.143415: step: 2438/1541, loss: 0.14725174009799957 2023-01-23 21:06:51.047149: step: 2440/1541, loss: 0.001391165773384273 2023-01-23 21:06:51.906053: step: 2442/1541, loss: 0.008051950484514236 2023-01-23 21:06:52.806597: step: 2444/1541, loss: 0.00012813393550459296 2023-01-23 21:06:53.674701: step: 2446/1541, loss: 4.7752258979016915e-05 2023-01-23 21:06:54.643287: step: 2448/1541, loss: 0.015238717198371887 2023-01-23 21:06:55.531642: step: 2450/1541, loss: 0.0075262682512402534 2023-01-23 21:06:56.488036: step: 2452/1541, loss: 0.16680128872394562 2023-01-23 21:06:57.372276: step: 2454/1541, loss: 0.0005400354275479913 2023-01-23 21:06:58.272279: step: 2456/1541, loss: 0.09911719709634781 2023-01-23 21:06:59.182034: step: 2458/1541, loss: 0.0033266693353652954 2023-01-23 21:07:00.105932: step: 2460/1541, loss: 0.007174333091825247 2023-01-23 21:07:01.039951: step: 2462/1541, loss: 0.006781228817999363 2023-01-23 21:07:01.882277: step: 2464/1541, loss: 3.0752176826354116e-05 2023-01-23 21:07:02.778060: step: 2466/1541, loss: 0.0031044050119817257 2023-01-23 21:07:03.657390: step: 2468/1541, loss: 0.010907551273703575 2023-01-23 21:07:04.587512: step: 2470/1541, loss: 0.0004295215767342597 2023-01-23 21:07:05.473229: step: 2472/1541, loss: 0.00443069264292717 2023-01-23 21:07:06.386782: step: 2474/1541, loss: 0.0562078133225441 2023-01-23 21:07:07.263045: step: 2476/1541, loss: 0.00037969418917782605 2023-01-23 21:07:08.165205: step: 2478/1541, loss: 0.00048089760821312666 2023-01-23 21:07:09.061501: step: 2480/1541, loss: 0.00011177048872923478 2023-01-23 21:07:10.018850: step: 2482/1541, loss: 0.000852194323670119 2023-01-23 21:07:10.975003: step: 2484/1541, loss: 0.0002881260879803449 2023-01-23 21:07:11.884682: step: 2486/1541, loss: 0.0029529475141316652 2023-01-23 21:07:12.764042: step: 2488/1541, loss: 0.001766597735695541 2023-01-23 21:07:13.652070: step: 2490/1541, loss: 1.042285680770874 2023-01-23 21:07:14.518480: step: 2492/1541, loss: 0.0016547031700611115 2023-01-23 21:07:15.507064: step: 2494/1541, loss: 2.4450810087728314e-05 2023-01-23 21:07:16.397419: step: 2496/1541, loss: 0.4270651042461395 2023-01-23 21:07:17.261716: step: 2498/1541, loss: 1.3559728860855103 2023-01-23 21:07:18.146407: step: 2500/1541, loss: 0.256335586309433 2023-01-23 21:07:19.044578: step: 2502/1541, loss: 0.00010064103844342753 2023-01-23 21:07:19.935406: step: 2504/1541, loss: 0.03977234661579132 2023-01-23 21:07:20.799291: step: 2506/1541, loss: 0.00012404992594383657 2023-01-23 21:07:21.642753: step: 2508/1541, loss: 0.28377804160118103 2023-01-23 21:07:22.615123: step: 2510/1541, loss: 1.56386067828862e-05 2023-01-23 21:07:23.505984: step: 2512/1541, loss: 6.660374492639676e-05 2023-01-23 21:07:24.382877: step: 2514/1541, loss: 0.002155384747311473 2023-01-23 21:07:25.240595: step: 2516/1541, loss: 0.0013415605062618852 2023-01-23 21:07:26.130617: step: 2518/1541, loss: 0.0007772239041514695 2023-01-23 21:07:27.024316: step: 2520/1541, loss: 1.69568374985829e-05 2023-01-23 21:07:27.923692: step: 2522/1541, loss: 0.12956960499286652 2023-01-23 21:07:28.791746: step: 2524/1541, loss: 0.0026100543327629566 2023-01-23 21:07:29.673849: step: 2526/1541, loss: 0.0006169543485157192 2023-01-23 21:07:30.552651: step: 2528/1541, loss: 6.53006718493998e-05 2023-01-23 21:07:31.438159: step: 2530/1541, loss: 0.018742263317108154 2023-01-23 21:07:32.390243: step: 2532/1541, loss: 0.0001024121738737449 2023-01-23 21:07:33.306350: step: 2534/1541, loss: 0.00020451383898034692 2023-01-23 21:07:34.195764: step: 2536/1541, loss: 0.00042990854126401246 2023-01-23 21:07:35.194192: step: 2538/1541, loss: 0.0009407623438164592 2023-01-23 21:07:36.132780: step: 2540/1541, loss: 0.1077786535024643 2023-01-23 21:07:37.017817: step: 2542/1541, loss: 0.00048753831651993096 2023-01-23 21:07:37.912860: step: 2544/1541, loss: 0.2828390598297119 2023-01-23 21:07:38.880305: step: 2546/1541, loss: 0.0017982058925554156 2023-01-23 21:07:39.789982: step: 2548/1541, loss: 0.00011059758253395557 2023-01-23 21:07:40.688283: step: 2550/1541, loss: 0.017348024994134903 2023-01-23 21:07:41.640842: step: 2552/1541, loss: 2.802051067352295 2023-01-23 21:07:42.509949: step: 2554/1541, loss: 0.0002257846063002944 2023-01-23 21:07:43.380239: step: 2556/1541, loss: 7.621585973538458e-05 2023-01-23 21:07:44.257214: step: 2558/1541, loss: 0.0009235755424015224 2023-01-23 21:07:45.137369: step: 2560/1541, loss: 0.00046979982289485633 2023-01-23 21:07:46.062957: step: 2562/1541, loss: 0.005293241236358881 2023-01-23 21:07:46.899595: step: 2564/1541, loss: 0.0005011188331991434 2023-01-23 21:07:47.768686: step: 2566/1541, loss: 0.00040773133514449 2023-01-23 21:07:48.701516: step: 2568/1541, loss: 0.935463011264801 2023-01-23 21:07:49.711559: step: 2570/1541, loss: 8.851669917930849e-06 2023-01-23 21:07:50.593779: step: 2572/1541, loss: 0.017468038946390152 2023-01-23 21:07:51.483774: step: 2574/1541, loss: 0.09926947951316833 2023-01-23 21:07:52.387997: step: 2576/1541, loss: 0.0007193209603428841 2023-01-23 21:07:53.268632: step: 2578/1541, loss: 0.2372235506772995 2023-01-23 21:07:54.145854: step: 2580/1541, loss: 0.010238533839583397 2023-01-23 21:07:54.997520: step: 2582/1541, loss: 0.00028338091215118766 2023-01-23 21:07:55.872674: step: 2584/1541, loss: 4.3486423237482086e-05 2023-01-23 21:07:56.779991: step: 2586/1541, loss: 0.000652642862405628 2023-01-23 21:07:57.673710: step: 2588/1541, loss: 0.0003701102687045932 2023-01-23 21:07:58.569060: step: 2590/1541, loss: 0.00015847221948206425 2023-01-23 21:07:59.467454: step: 2592/1541, loss: 0.0016709320480003953 2023-01-23 21:08:00.333160: step: 2594/1541, loss: 0.00010058964835479856 2023-01-23 21:08:01.203338: step: 2596/1541, loss: 0.0003083153860643506 2023-01-23 21:08:02.057543: step: 2598/1541, loss: 0.08015093207359314 2023-01-23 21:08:02.920811: step: 2600/1541, loss: 0.00012895578402094543 2023-01-23 21:08:03.872235: step: 2602/1541, loss: 7.261306018335745e-05 2023-01-23 21:08:04.812462: step: 2604/1541, loss: 0.004830003250390291 2023-01-23 21:08:05.704820: step: 2606/1541, loss: 4.951319715473801e-05 2023-01-23 21:08:06.540505: step: 2608/1541, loss: 7.76685556047596e-05 2023-01-23 21:08:07.440733: step: 2610/1541, loss: 0.00030864778091199696 2023-01-23 21:08:08.377463: step: 2612/1541, loss: 2.6302976608276367 2023-01-23 21:08:09.244423: step: 2614/1541, loss: 0.003052916843444109 2023-01-23 21:08:10.117341: step: 2616/1541, loss: 1.54313074745005e-05 2023-01-23 21:08:10.987463: step: 2618/1541, loss: 0.054256685078144073 2023-01-23 21:08:11.903613: step: 2620/1541, loss: 4.779112714459188e-05 2023-01-23 21:08:12.763701: step: 2622/1541, loss: 0.00016391401004511863 2023-01-23 21:08:13.690193: step: 2624/1541, loss: 0.009844600223004818 2023-01-23 21:08:14.612390: step: 2626/1541, loss: 0.09034863114356995 2023-01-23 21:08:15.441118: step: 2628/1541, loss: 0.016707992181181908 2023-01-23 21:08:16.341299: step: 2630/1541, loss: -2.6443603928782977e-05 2023-01-23 21:08:17.214085: step: 2632/1541, loss: 0.000541923101991415 2023-01-23 21:08:18.159189: step: 2634/1541, loss: 4.124223050894216e-05 2023-01-23 21:08:19.096789: step: 2636/1541, loss: 4.1138519009109586e-05 2023-01-23 21:08:19.973170: step: 2638/1541, loss: 0.00013514040620066226 2023-01-23 21:08:20.868977: step: 2640/1541, loss: 0.16666999459266663 2023-01-23 21:08:21.780811: step: 2642/1541, loss: 0.3440711498260498 2023-01-23 21:08:22.672084: step: 2644/1541, loss: 0.16549547016620636 2023-01-23 21:08:23.560141: step: 2646/1541, loss: 0.001289062318392098 2023-01-23 21:08:24.450921: step: 2648/1541, loss: 0.10224944353103638 2023-01-23 21:08:25.342937: step: 2650/1541, loss: 0.0024940744042396545 2023-01-23 21:08:26.225895: step: 2652/1541, loss: 0.0019471126142889261 2023-01-23 21:08:27.119783: step: 2654/1541, loss: 0.051015596836805344 2023-01-23 21:08:28.001084: step: 2656/1541, loss: 0.0018528869841247797 2023-01-23 21:08:28.935002: step: 2658/1541, loss: 0.003608746686950326 2023-01-23 21:08:29.815778: step: 2660/1541, loss: 2.9090431780787185e-05 2023-01-23 21:08:30.676499: step: 2662/1541, loss: 0.04290114343166351 2023-01-23 21:08:31.549416: step: 2664/1541, loss: 0.06214875727891922 2023-01-23 21:08:32.475522: step: 2666/1541, loss: 0.00020657983259297907 2023-01-23 21:08:33.382038: step: 2668/1541, loss: 5.8375248045194894e-05 2023-01-23 21:08:34.251526: step: 2670/1541, loss: 0.06389402598142624 2023-01-23 21:08:35.121941: step: 2672/1541, loss: 5.544234227272682e-05 2023-01-23 21:08:36.034563: step: 2674/1541, loss: 0.00028203317197039723 2023-01-23 21:08:36.991682: step: 2676/1541, loss: 0.04162124916911125 2023-01-23 21:08:37.884263: step: 2678/1541, loss: 0.0018761928658932447 2023-01-23 21:08:38.747167: step: 2680/1541, loss: 0.0014410587027668953 2023-01-23 21:08:39.624249: step: 2682/1541, loss: 0.012229742482304573 2023-01-23 21:08:40.488955: step: 2684/1541, loss: 0.06120047718286514 2023-01-23 21:08:41.364233: step: 2686/1541, loss: 0.0015778889646753669 2023-01-23 21:08:42.236567: step: 2688/1541, loss: 0.006253506056964397 2023-01-23 21:08:43.135886: step: 2690/1541, loss: 0.006264049559831619 2023-01-23 21:08:44.011927: step: 2692/1541, loss: 0.0019163216929882765 2023-01-23 21:08:44.898710: step: 2694/1541, loss: 7.804622873663902e-05 2023-01-23 21:08:45.832598: step: 2696/1541, loss: 0.13797913491725922 2023-01-23 21:08:46.720323: step: 2698/1541, loss: 0.2665035128593445 2023-01-23 21:08:47.665630: step: 2700/1541, loss: 0.0027160104364156723 2023-01-23 21:08:48.523471: step: 2702/1541, loss: 0.001237477408722043 2023-01-23 21:08:49.418994: step: 2704/1541, loss: 0.0488094761967659 2023-01-23 21:08:50.386543: step: 2706/1541, loss: 9.306866559199989e-05 2023-01-23 21:08:51.304363: step: 2708/1541, loss: 0.16168653964996338 2023-01-23 21:08:52.168949: step: 2710/1541, loss: 0.0019129280699416995 2023-01-23 21:08:53.047141: step: 2712/1541, loss: 0.001931157079525292 2023-01-23 21:08:53.980043: step: 2714/1541, loss: 0.001182659063488245 2023-01-23 21:08:54.838363: step: 2716/1541, loss: 0.006849738769233227 2023-01-23 21:08:55.715020: step: 2718/1541, loss: 3.055371416849084e-05 2023-01-23 21:08:56.593815: step: 2720/1541, loss: 4.5097411202732474e-05 2023-01-23 21:08:57.487108: step: 2722/1541, loss: 0.08981021493673325 2023-01-23 21:08:58.382703: step: 2724/1541, loss: 6.283397669903934e-05 2023-01-23 21:08:59.254824: step: 2726/1541, loss: 0.002591849770396948 2023-01-23 21:09:00.182019: step: 2728/1541, loss: 0.02600787952542305 2023-01-23 21:09:01.063352: step: 2730/1541, loss: 7.535868644481525e-05 2023-01-23 21:09:01.976333: step: 2732/1541, loss: 0.000277053564786911 2023-01-23 21:09:02.869016: step: 2734/1541, loss: 0.002329414477571845 2023-01-23 21:09:03.712196: step: 2736/1541, loss: 6.014033351675607e-05 2023-01-23 21:09:04.611854: step: 2738/1541, loss: 0.04742036759853363 2023-01-23 21:09:05.502814: step: 2740/1541, loss: 0.00457120593637228 2023-01-23 21:09:06.367627: step: 2742/1541, loss: 0.0002768808917608112 2023-01-23 21:09:07.353134: step: 2744/1541, loss: 0.00030292748124338686 2023-01-23 21:09:08.252936: step: 2746/1541, loss: 0.23549336194992065 2023-01-23 21:09:09.139785: step: 2748/1541, loss: 0.0003801451821345836 2023-01-23 21:09:10.017164: step: 2750/1541, loss: 0.07042240351438522 2023-01-23 21:09:10.855172: step: 2752/1541, loss: 0.02445468120276928 2023-01-23 21:09:11.762299: step: 2754/1541, loss: 0.058974504470825195 2023-01-23 21:09:12.658866: step: 2756/1541, loss: 0.05681410804390907 2023-01-23 21:09:13.551887: step: 2758/1541, loss: 0.009231599047780037 2023-01-23 21:09:14.418274: step: 2760/1541, loss: 0.1254548877477646 2023-01-23 21:09:15.277131: step: 2762/1541, loss: 0.00021401089907158166 2023-01-23 21:09:16.152290: step: 2764/1541, loss: 4.139624070376158e-05 2023-01-23 21:09:17.022781: step: 2766/1541, loss: 0.00035101501271128654 2023-01-23 21:09:17.883118: step: 2768/1541, loss: 0.0017802913207560778 2023-01-23 21:09:18.742604: step: 2770/1541, loss: 0.07396870106458664 2023-01-23 21:09:19.666556: step: 2772/1541, loss: 0.00015910679940134287 2023-01-23 21:09:20.558773: step: 2774/1541, loss: 0.015258323401212692 2023-01-23 21:09:21.441966: step: 2776/1541, loss: 0.0035357880406081676 2023-01-23 21:09:22.351266: step: 2778/1541, loss: 1.1265165085205808e-05 2023-01-23 21:09:23.247348: step: 2780/1541, loss: 0.00950718391686678 2023-01-23 21:09:24.199221: step: 2782/1541, loss: 0.03638943284749985 2023-01-23 21:09:25.051192: step: 2784/1541, loss: 0.23374058306217194 2023-01-23 21:09:25.951416: step: 2786/1541, loss: 0.000504431314766407 2023-01-23 21:09:26.876175: step: 2788/1541, loss: 6.0885318816872314e-05 2023-01-23 21:09:27.826860: step: 2790/1541, loss: 0.0015053142560645938 2023-01-23 21:09:28.703956: step: 2792/1541, loss: 0.18821148574352264 2023-01-23 21:09:29.671211: step: 2794/1541, loss: 0.22145560383796692 2023-01-23 21:09:30.555509: step: 2796/1541, loss: 0.05511907488107681 2023-01-23 21:09:31.425952: step: 2798/1541, loss: 0.0001320667943218723 2023-01-23 21:09:32.345342: step: 2800/1541, loss: 0.002017746213823557 2023-01-23 21:09:33.220731: step: 2802/1541, loss: 5.9051628340967e-05 2023-01-23 21:09:34.098428: step: 2804/1541, loss: 0.00020259128359612077 2023-01-23 21:09:34.985594: step: 2806/1541, loss: 0.0008042838308028877 2023-01-23 21:09:35.851602: step: 2808/1541, loss: 0.00020933107589371502 2023-01-23 21:09:36.741626: step: 2810/1541, loss: 0.0011389035498723388 2023-01-23 21:09:37.653903: step: 2812/1541, loss: 0.0016200067475438118 2023-01-23 21:09:38.533865: step: 2814/1541, loss: 0.0291590578854084 2023-01-23 21:09:39.441026: step: 2816/1541, loss: 1.3066477549728006e-05 2023-01-23 21:09:40.368108: step: 2818/1541, loss: 0.018017183989286423 2023-01-23 21:09:41.231255: step: 2820/1541, loss: 0.0003381737042218447 2023-01-23 21:09:42.117124: step: 2822/1541, loss: 0.052268318831920624 2023-01-23 21:09:42.993149: step: 2824/1541, loss: 0.0021241484209895134 2023-01-23 21:09:43.933829: step: 2826/1541, loss: 0.16814115643501282 2023-01-23 21:09:44.872024: step: 2828/1541, loss: 0.1865774244070053 2023-01-23 21:09:45.747057: step: 2830/1541, loss: 0.010711103677749634 2023-01-23 21:09:46.657311: step: 2832/1541, loss: 0.0010330314980819821 2023-01-23 21:09:47.630511: step: 2834/1541, loss: 0.0004659178084693849 2023-01-23 21:09:48.520037: step: 2836/1541, loss: 0.00040950244874693453 2023-01-23 21:09:49.460441: step: 2838/1541, loss: 0.0016792224487289786 2023-01-23 21:09:50.312816: step: 2840/1541, loss: 4.444548176252283e-05 2023-01-23 21:09:51.288372: step: 2842/1541, loss: 0.041635770350694656 2023-01-23 21:09:52.166421: step: 2844/1541, loss: 0.11870450526475906 2023-01-23 21:09:53.025135: step: 2846/1541, loss: 0.11739461123943329 2023-01-23 21:09:53.956300: step: 2848/1541, loss: 0.05649678036570549 2023-01-23 21:09:54.896352: step: 2850/1541, loss: 0.01985679753124714 2023-01-23 21:09:55.800594: step: 2852/1541, loss: 0.0006833644001744688 2023-01-23 21:09:56.686517: step: 2854/1541, loss: 0.00040299014654010534 2023-01-23 21:09:57.608221: step: 2856/1541, loss: 0.00266997586004436 2023-01-23 21:09:58.525925: step: 2858/1541, loss: 0.4582873284816742 2023-01-23 21:09:59.402689: step: 2860/1541, loss: 3.8622245483566076e-05 2023-01-23 21:10:00.235030: step: 2862/1541, loss: 0.015686744824051857 2023-01-23 21:10:01.158763: step: 2864/1541, loss: 0.031937941908836365 2023-01-23 21:10:02.055156: step: 2866/1541, loss: 0.02589832991361618 2023-01-23 21:10:02.928159: step: 2868/1541, loss: 7.3306291596964e-05 2023-01-23 21:10:03.808917: step: 2870/1541, loss: 0.0007317098788917065 2023-01-23 21:10:04.664083: step: 2872/1541, loss: 0.007043009158223867 2023-01-23 21:10:05.651907: step: 2874/1541, loss: 0.0568724200129509 2023-01-23 21:10:06.543233: step: 2876/1541, loss: 0.0009771635523065925 2023-01-23 21:10:07.422331: step: 2878/1541, loss: 0.006744437385350466 2023-01-23 21:10:08.323108: step: 2880/1541, loss: 5.2583323849830776e-05 2023-01-23 21:10:09.204302: step: 2882/1541, loss: 0.2793005704879761 2023-01-23 21:10:10.089988: step: 2884/1541, loss: 0.00722865667194128 2023-01-23 21:10:10.985392: step: 2886/1541, loss: 0.03577016294002533 2023-01-23 21:10:11.852732: step: 2888/1541, loss: 0.009334467351436615 2023-01-23 21:10:12.764992: step: 2890/1541, loss: 0.04438556730747223 2023-01-23 21:10:13.693276: step: 2892/1541, loss: 0.016495637595653534 2023-01-23 21:10:14.576571: step: 2894/1541, loss: 0.00016757614503148943 2023-01-23 21:10:15.493366: step: 2896/1541, loss: 0.09525258094072342 2023-01-23 21:10:16.356346: step: 2898/1541, loss: 0.08655298501253128 2023-01-23 21:10:17.247533: step: 2900/1541, loss: 0.00012055539264110848 2023-01-23 21:10:18.107621: step: 2902/1541, loss: 0.0002407838182989508 2023-01-23 21:10:18.989844: step: 2904/1541, loss: 0.00026207126211375 2023-01-23 21:10:19.893313: step: 2906/1541, loss: 3.05860121443402e-05 2023-01-23 21:10:20.780397: step: 2908/1541, loss: 0.15463517606258392 2023-01-23 21:10:21.681450: step: 2910/1541, loss: 0.0059377835132181644 2023-01-23 21:10:22.591549: step: 2912/1541, loss: 0.09830921143293381 2023-01-23 21:10:23.466282: step: 2914/1541, loss: 2.4782766558928415e-05 2023-01-23 21:10:24.478190: step: 2916/1541, loss: 0.041410963982343674 2023-01-23 21:10:25.413550: step: 2918/1541, loss: 0.11370977759361267 2023-01-23 21:10:26.302463: step: 2920/1541, loss: 3.0705683457199484e-05 2023-01-23 21:10:27.178304: step: 2922/1541, loss: 0.02397456392645836 2023-01-23 21:10:28.077248: step: 2924/1541, loss: 4.433036519913003e-05 2023-01-23 21:10:28.964085: step: 2926/1541, loss: 0.13971221446990967 2023-01-23 21:10:29.907407: step: 2928/1541, loss: 0.001170321716926992 2023-01-23 21:10:30.807535: step: 2930/1541, loss: 0.00018961096066050231 2023-01-23 21:10:31.784353: step: 2932/1541, loss: 0.08636212348937988 2023-01-23 21:10:32.680504: step: 2934/1541, loss: 0.0021049624774605036 2023-01-23 21:10:33.628599: step: 2936/1541, loss: 0.04282429441809654 2023-01-23 21:10:34.592560: step: 2938/1541, loss: 0.021696750074625015 2023-01-23 21:10:35.462888: step: 2940/1541, loss: 0.11276151239871979 2023-01-23 21:10:36.327674: step: 2942/1541, loss: 0.01230805553495884 2023-01-23 21:10:37.207935: step: 2944/1541, loss: 0.10877028852701187 2023-01-23 21:10:38.078381: step: 2946/1541, loss: 0.21354734897613525 2023-01-23 21:10:38.953734: step: 2948/1541, loss: 8.625326881883666e-05 2023-01-23 21:10:39.842779: step: 2950/1541, loss: 0.0012876959517598152 2023-01-23 21:10:40.799781: step: 2952/1541, loss: 0.26900994777679443 2023-01-23 21:10:41.682222: step: 2954/1541, loss: 0.28224408626556396 2023-01-23 21:10:42.604775: step: 2956/1541, loss: 0.0009110732353292406 2023-01-23 21:10:43.469375: step: 2958/1541, loss: 0.0013776092091575265 2023-01-23 21:10:44.310596: step: 2960/1541, loss: 0.012468051165342331 2023-01-23 21:10:45.252267: step: 2962/1541, loss: 5.1935232477262616e-05 2023-01-23 21:10:46.166699: step: 2964/1541, loss: 0.11005135625600815 2023-01-23 21:10:47.078821: step: 2966/1541, loss: 0.03753264248371124 2023-01-23 21:10:47.964218: step: 2968/1541, loss: 0.000135193724418059 2023-01-23 21:10:48.995070: step: 2970/1541, loss: 0.0007456349558196962 2023-01-23 21:10:49.931392: step: 2972/1541, loss: 0.21438156068325043 2023-01-23 21:10:50.798080: step: 2974/1541, loss: 0.0007722023874521255 2023-01-23 21:10:51.683252: step: 2976/1541, loss: 0.3110201358795166 2023-01-23 21:10:52.692502: step: 2978/1541, loss: 0.0013158792862668633 2023-01-23 21:10:53.604483: step: 2980/1541, loss: 0.002696351148188114 2023-01-23 21:10:54.488191: step: 2982/1541, loss: 0.0069328960962593555 2023-01-23 21:10:55.411382: step: 2984/1541, loss: 0.1317610889673233 2023-01-23 21:10:56.321179: step: 2986/1541, loss: 0.0008112884243018925 2023-01-23 21:10:57.194041: step: 2988/1541, loss: 0.0010444929357618093 2023-01-23 21:10:58.101399: step: 2990/1541, loss: 0.015576701611280441 2023-01-23 21:10:59.035202: step: 2992/1541, loss: 0.02817138470709324 2023-01-23 21:10:59.994935: step: 2994/1541, loss: 0.013314432464540005 2023-01-23 21:11:00.902107: step: 2996/1541, loss: 0.06799936294555664 2023-01-23 21:11:01.791866: step: 2998/1541, loss: 0.00012792788038495928 2023-01-23 21:11:02.697655: step: 3000/1541, loss: 0.04659964144229889 2023-01-23 21:11:03.657967: step: 3002/1541, loss: 0.023343399167060852 2023-01-23 21:11:04.539962: step: 3004/1541, loss: 0.00101692124735564 2023-01-23 21:11:05.427575: step: 3006/1541, loss: 0.15194572508335114 2023-01-23 21:11:06.364459: step: 3008/1541, loss: 0.001446771901100874 2023-01-23 21:11:07.269220: step: 3010/1541, loss: 0.008298045955598354 2023-01-23 21:11:08.160420: step: 3012/1541, loss: 0.005834381561726332 2023-01-23 21:11:09.044950: step: 3014/1541, loss: 0.0004222437855787575 2023-01-23 21:11:10.017781: step: 3016/1541, loss: 0.0006702836253680289 2023-01-23 21:11:10.926944: step: 3018/1541, loss: 0.013642421923577785 2023-01-23 21:11:11.767405: step: 3020/1541, loss: 0.1886254996061325 2023-01-23 21:11:12.690350: step: 3022/1541, loss: 0.30957862734794617 2023-01-23 21:11:13.571743: step: 3024/1541, loss: 0.004344746004790068 2023-01-23 21:11:14.431632: step: 3026/1541, loss: 0.00017742413911037147 2023-01-23 21:11:15.342342: step: 3028/1541, loss: 0.04955430328845978 2023-01-23 21:11:16.239422: step: 3030/1541, loss: 0.0027120602317154408 2023-01-23 21:11:17.120115: step: 3032/1541, loss: 0.14277514815330505 2023-01-23 21:11:18.018370: step: 3034/1541, loss: 0.009095238521695137 2023-01-23 21:11:18.879628: step: 3036/1541, loss: 0.1113634705543518 2023-01-23 21:11:19.801891: step: 3038/1541, loss: 0.003633879590779543 2023-01-23 21:11:20.683245: step: 3040/1541, loss: 0.0006040045409463346 2023-01-23 21:11:21.536675: step: 3042/1541, loss: 0.0018180360784754157 2023-01-23 21:11:22.454189: step: 3044/1541, loss: 0.014955981634557247 2023-01-23 21:11:23.322711: step: 3046/1541, loss: 0.0032303589396178722 2023-01-23 21:11:24.190180: step: 3048/1541, loss: 0.002369015710428357 2023-01-23 21:11:25.070678: step: 3050/1541, loss: 0.0012688733404502273 2023-01-23 21:11:25.967202: step: 3052/1541, loss: 0.0014561963034793735 2023-01-23 21:11:26.870087: step: 3054/1541, loss: 0.004607843700796366 2023-01-23 21:11:27.750290: step: 3056/1541, loss: 0.00020136054081376642 2023-01-23 21:11:28.624619: step: 3058/1541, loss: 0.0021387720480561256 2023-01-23 21:11:29.573736: step: 3060/1541, loss: 0.00019713962683454156 2023-01-23 21:11:30.481442: step: 3062/1541, loss: 0.17212603986263275 2023-01-23 21:11:31.358929: step: 3064/1541, loss: 7.66605298849754e-05 2023-01-23 21:11:32.293085: step: 3066/1541, loss: 0.0002858826774172485 2023-01-23 21:11:33.196187: step: 3068/1541, loss: 0.17651326954364777 2023-01-23 21:11:34.077624: step: 3070/1541, loss: 0.000656489166431129 2023-01-23 21:11:34.992421: step: 3072/1541, loss: 0.007724291179329157 2023-01-23 21:11:35.946006: step: 3074/1541, loss: 0.01047708373516798 2023-01-23 21:11:36.838333: step: 3076/1541, loss: 0.00016016469453461468 2023-01-23 21:11:37.699648: step: 3078/1541, loss: 0.0002513122744858265 2023-01-23 21:11:38.595623: step: 3080/1541, loss: 2.245898962020874 2023-01-23 21:11:39.506069: step: 3082/1541, loss: 0.00011509909381857142 2023-01-23 21:11:40.133598: step: 3084/1541, loss: 7.023812213446945e-05 ================================================== Loss: 0.062 -------------------- Dev Chinese: {'event': {'p': 0.6200716845878136, 'r': 0.6910785619174434, 'f1': 0.6536523929471032}, 'argument': {'p': 0.667344626189574, 'r': 0.6549192801627705, 'f1': 0.6610735725370487}, 'combined': 0.43211232260293225, 'epoch': 33} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7172937437299021, 'r': 0.7034393091753522, 'f1': 0.7102989749318}, 'combined': 0.46265903547784015, 'epoch': 33} Dev Korean: {'event': {'p': 0.5627450980392157, 'r': 0.7643142476697736, 'f1': 0.6482213438735178}, 'argument': {'p': 0.6636872750304308, 'r': 0.6608306583516772, 'f1': 0.6622558862273458}, 'combined': 0.4292884005584376, 'epoch': 33} Test Korean: {'event': {'p': 0.613441372735939, 'r': 0.7711204313960456, 'f1': 0.6833023626227768}, 'argument': {'p': 0.7113677591330053, 'r': 0.7181046688126234, 'f1': 0.7147203389020036}, 'combined': 0.4883700961862908, 'epoch': 33} Dev Russian: {'event': {'p': 0.6481042654028436, 'r': 0.7283621837549934, 'f1': 0.6858934169278997}, 'argument': {'p': 0.6738499304583855, 'r': 0.663701588132205, 'f1': 0.6687372602728135}, 'combined': 0.4586824844755223, 'epoch': 33} Test Russian: {'event': {'p': 0.6509680795395081, 'r': 0.7453565008987417, 'f1': 0.6949720670391061}, 'argument': {'p': 0.7144920197618627, 'r': 0.7057755760331569, 'f1': 0.710107050687512}, 'combined': 0.4935045648353435, 'epoch': 33} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6442550505050505, 'r': 0.6841646554035935, 'f1': 0.6636103524515542}, 'combined': 0.49912573517723735, 'epoch': 33} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5351696234049176, 'r': 0.35912698412698413, 'f1': 0.4298212723409574}, 'combined': 0.2504470405766208, 'epoch': 33} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5601886863442256, 'r': 0.612299261818107, 'f1': 0.585085961292858}, 'combined': 0.2841846097708167, 'epoch': 33} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 34 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 21:17:03.461207: step: 2/1541, loss: 5.040715404902585e-05 2023-01-23 21:17:04.315756: step: 4/1541, loss: 0.018505530431866646 2023-01-23 21:17:05.207398: step: 6/1541, loss: 0.0017848419956862926 2023-01-23 21:17:06.156699: step: 8/1541, loss: 0.005341160111129284 2023-01-23 21:17:07.044028: step: 10/1541, loss: 0.04373176768422127 2023-01-23 21:17:07.945509: step: 12/1541, loss: 0.003276577452197671 2023-01-23 21:17:08.813049: step: 14/1541, loss: 9.649758794694208e-06 2023-01-23 21:17:09.684549: step: 16/1541, loss: 0.0021386146545410156 2023-01-23 21:17:10.578420: step: 18/1541, loss: 5.609085928881541e-05 2023-01-23 21:17:11.453321: step: 20/1541, loss: 0.22236743569374084 2023-01-23 21:17:12.313780: step: 22/1541, loss: 0.03140844777226448 2023-01-23 21:17:13.199787: step: 24/1541, loss: 0.06833625584840775 2023-01-23 21:17:14.074556: step: 26/1541, loss: 0.004375885706394911 2023-01-23 21:17:14.920422: step: 28/1541, loss: 0.007194328121840954 2023-01-23 21:17:15.768580: step: 30/1541, loss: 1.3373284673434682e-05 2023-01-23 21:17:16.644581: step: 32/1541, loss: 0.00023818001500330865 2023-01-23 21:17:17.538490: step: 34/1541, loss: 0.32046574354171753 2023-01-23 21:17:18.450362: step: 36/1541, loss: 6.226669938769192e-05 2023-01-23 21:17:19.443252: step: 38/1541, loss: 0.0001287052291445434 2023-01-23 21:17:20.319514: step: 40/1541, loss: 0.00015076792624313384 2023-01-23 21:17:21.203084: step: 42/1541, loss: 0.00046371808275580406 2023-01-23 21:17:22.217560: step: 44/1541, loss: 0.059756338596343994 2023-01-23 21:17:23.166376: step: 46/1541, loss: 0.0046590277925133705 2023-01-23 21:17:24.087305: step: 48/1541, loss: 0.00023350681294687092 2023-01-23 21:17:24.959476: step: 50/1541, loss: 0.000516925472766161 2023-01-23 21:17:25.871514: step: 52/1541, loss: 0.13150878250598907 2023-01-23 21:17:26.784215: step: 54/1541, loss: 0.0002795705513563007 2023-01-23 21:17:27.686260: step: 56/1541, loss: -6.157166353659704e-06 2023-01-23 21:17:28.599886: step: 58/1541, loss: 0.4976353645324707 2023-01-23 21:17:29.521246: step: 60/1541, loss: 0.0778946802020073 2023-01-23 21:17:30.384500: step: 62/1541, loss: 0.04560811445116997 2023-01-23 21:17:31.259496: step: 64/1541, loss: 0.1592939794063568 2023-01-23 21:17:32.277469: step: 66/1541, loss: 0.30627524852752686 2023-01-23 21:17:33.163786: step: 68/1541, loss: 3.66427320841467e-06 2023-01-23 21:17:34.047284: step: 70/1541, loss: 0.0001710992946755141 2023-01-23 21:17:34.955566: step: 72/1541, loss: 0.01075070258229971 2023-01-23 21:17:35.864952: step: 74/1541, loss: 0.003094171639531851 2023-01-23 21:17:36.748007: step: 76/1541, loss: 9.255073382519186e-05 2023-01-23 21:17:37.617202: step: 78/1541, loss: 0.0007427495438605547 2023-01-23 21:17:38.452932: step: 80/1541, loss: 0.0008022294496186078 2023-01-23 21:17:39.349492: step: 82/1541, loss: 0.012246508151292801 2023-01-23 21:17:40.265786: step: 84/1541, loss: 0.00012759397213812917 2023-01-23 21:17:41.126536: step: 86/1541, loss: 0.0018710761796683073 2023-01-23 21:17:42.029588: step: 88/1541, loss: 5.9874128055525944e-05 2023-01-23 21:17:42.950082: step: 90/1541, loss: 0.2989628314971924 2023-01-23 21:17:43.886256: step: 92/1541, loss: 0.04855772480368614 2023-01-23 21:17:44.789452: step: 94/1541, loss: 0.10331562161445618 2023-01-23 21:17:45.660177: step: 96/1541, loss: 0.1135108470916748 2023-01-23 21:17:46.540331: step: 98/1541, loss: 0.00010394578566774726 2023-01-23 21:17:47.432582: step: 100/1541, loss: 1.2397369146347046 2023-01-23 21:17:48.322822: step: 102/1541, loss: 0.021916477009654045 2023-01-23 21:17:49.231251: step: 104/1541, loss: 0.0003291817265562713 2023-01-23 21:17:50.178188: step: 106/1541, loss: 0.0015961886383593082 2023-01-23 21:17:51.044968: step: 108/1541, loss: 0.001279426971450448 2023-01-23 21:17:51.893134: step: 110/1541, loss: 2.8707958335871808e-05 2023-01-23 21:17:52.819571: step: 112/1541, loss: 0.002885448979213834 2023-01-23 21:17:53.726981: step: 114/1541, loss: 0.04173992946743965 2023-01-23 21:17:54.652949: step: 116/1541, loss: 0.014141540974378586 2023-01-23 21:17:55.543562: step: 118/1541, loss: -2.494202590241912e-06 2023-01-23 21:17:56.402324: step: 120/1541, loss: 0.026257731020450592 2023-01-23 21:17:57.249992: step: 122/1541, loss: 4.970528607373126e-05 2023-01-23 21:17:58.072514: step: 124/1541, loss: 0.016446132212877274 2023-01-23 21:17:59.036268: step: 126/1541, loss: 0.00010774585098261014 2023-01-23 21:17:59.921377: step: 128/1541, loss: 0.001393011654727161 2023-01-23 21:18:00.784400: step: 130/1541, loss: 7.805947097949684e-05 2023-01-23 21:18:01.688648: step: 132/1541, loss: 0.026461057364940643 2023-01-23 21:18:02.651040: step: 134/1541, loss: 0.009213542565703392 2023-01-23 21:18:03.506628: step: 136/1541, loss: 0.0043035284616053104 2023-01-23 21:18:04.379350: step: 138/1541, loss: 0.00012160480400780216 2023-01-23 21:18:05.267915: step: 140/1541, loss: 0.29725855588912964 2023-01-23 21:18:06.200667: step: 142/1541, loss: 8.176781557267532e-05 2023-01-23 21:18:07.094055: step: 144/1541, loss: 1.4107276911090594e-05 2023-01-23 21:18:07.978482: step: 146/1541, loss: 0.17360761761665344 2023-01-23 21:18:08.934378: step: 148/1541, loss: 3.5706409107660875e-05 2023-01-23 21:18:09.836387: step: 150/1541, loss: 0.09509344398975372 2023-01-23 21:18:10.743814: step: 152/1541, loss: 0.00978798232972622 2023-01-23 21:18:11.622849: step: 154/1541, loss: 0.019168199971318245 2023-01-23 21:18:12.553218: step: 156/1541, loss: -3.255938054280705e-06 2023-01-23 21:18:13.478562: step: 158/1541, loss: 0.0008894376223906875 2023-01-23 21:18:14.415810: step: 160/1541, loss: 0.00011818084749393165 2023-01-23 21:18:15.281512: step: 162/1541, loss: 4.730284126708284e-05 2023-01-23 21:18:16.191029: step: 164/1541, loss: 0.30333372950553894 2023-01-23 21:18:17.052386: step: 166/1541, loss: 0.004237257409840822 2023-01-23 21:18:17.941949: step: 168/1541, loss: 0.22516866028308868 2023-01-23 21:18:18.833919: step: 170/1541, loss: 0.040211185812950134 2023-01-23 21:18:19.680533: step: 172/1541, loss: 0.0001579646923346445 2023-01-23 21:18:20.532071: step: 174/1541, loss: 0.0003406722971703857 2023-01-23 21:18:21.434886: step: 176/1541, loss: 8.416056516580284e-05 2023-01-23 21:18:22.339641: step: 178/1541, loss: 9.712576866149902e-05 2023-01-23 21:18:23.229370: step: 180/1541, loss: 0.00010122983803739771 2023-01-23 21:18:24.147642: step: 182/1541, loss: 0.00026474142214283347 2023-01-23 21:18:25.036947: step: 184/1541, loss: 0.15159468352794647 2023-01-23 21:18:25.958798: step: 186/1541, loss: 0.0020693151745945215 2023-01-23 21:18:26.845624: step: 188/1541, loss: 0.015993377193808556 2023-01-23 21:18:27.741910: step: 190/1541, loss: 0.48746153712272644 2023-01-23 21:18:28.630187: step: 192/1541, loss: 9.894218237604946e-05 2023-01-23 21:18:29.525160: step: 194/1541, loss: 0.00011775767779909074 2023-01-23 21:18:30.393292: step: 196/1541, loss: 0.24170690774917603 2023-01-23 21:18:31.269332: step: 198/1541, loss: 0.011717121116816998 2023-01-23 21:18:32.207681: step: 200/1541, loss: 7.64472788432613e-05 2023-01-23 21:18:33.097467: step: 202/1541, loss: 0.00010329822544008493 2023-01-23 21:18:34.003950: step: 204/1541, loss: 0.00046505682985298336 2023-01-23 21:18:34.916178: step: 206/1541, loss: 0.1863817721605301 2023-01-23 21:18:35.760967: step: 208/1541, loss: -3.9271639252547175e-05 2023-01-23 21:18:36.607976: step: 210/1541, loss: 0.0005437269574031234 2023-01-23 21:18:37.480676: step: 212/1541, loss: 0.2937021851539612 2023-01-23 21:18:38.364237: step: 214/1541, loss: 0.013867632485926151 2023-01-23 21:18:39.276684: step: 216/1541, loss: 0.005617933347821236 2023-01-23 21:18:40.145797: step: 218/1541, loss: 0.0013541651424020529 2023-01-23 21:18:41.038056: step: 220/1541, loss: 0.002349620684981346 2023-01-23 21:18:41.955836: step: 222/1541, loss: 0.644286572933197 2023-01-23 21:18:42.844542: step: 224/1541, loss: 0.068056121468544 2023-01-23 21:18:43.687814: step: 226/1541, loss: 5.173042882233858e-05 2023-01-23 21:18:44.589790: step: 228/1541, loss: 0.0053190793842077255 2023-01-23 21:18:45.481063: step: 230/1541, loss: 0.00010453701543156058 2023-01-23 21:18:46.403198: step: 232/1541, loss: 0.006564779672771692 2023-01-23 21:18:47.294175: step: 234/1541, loss: 0.0006095229182392359 2023-01-23 21:18:48.138354: step: 236/1541, loss: 0.1772049218416214 2023-01-23 21:18:49.013182: step: 238/1541, loss: 0.04677444323897362 2023-01-23 21:18:49.848563: step: 240/1541, loss: 0.13766151666641235 2023-01-23 21:18:50.835561: step: 242/1541, loss: 0.6752125024795532 2023-01-23 21:18:51.769412: step: 244/1541, loss: 0.007709369994699955 2023-01-23 21:18:52.653564: step: 246/1541, loss: 0.00425203749909997 2023-01-23 21:18:53.535056: step: 248/1541, loss: 1.5273630619049072 2023-01-23 21:18:54.411449: step: 250/1541, loss: 0.0005033991765230894 2023-01-23 21:18:55.375510: step: 252/1541, loss: 0.00830151792615652 2023-01-23 21:18:56.258484: step: 254/1541, loss: 0.06121102347970009 2023-01-23 21:18:57.146385: step: 256/1541, loss: 0.021714070811867714 2023-01-23 21:18:58.019312: step: 258/1541, loss: 0.003515866119414568 2023-01-23 21:18:58.897531: step: 260/1541, loss: 0.13137885928153992 2023-01-23 21:18:59.811699: step: 262/1541, loss: 0.0010222485288977623 2023-01-23 21:19:00.740952: step: 264/1541, loss: 0.007680822629481554 2023-01-23 21:19:01.667014: step: 266/1541, loss: 0.16227887570858002 2023-01-23 21:19:02.610402: step: 268/1541, loss: 0.005593752022832632 2023-01-23 21:19:03.510996: step: 270/1541, loss: 0.11369162052869797 2023-01-23 21:19:04.413590: step: 272/1541, loss: 0.024446990340948105 2023-01-23 21:19:05.336076: step: 274/1541, loss: 0.00028626047424040735 2023-01-23 21:19:06.223362: step: 276/1541, loss: 0.03135381266474724 2023-01-23 21:19:07.091892: step: 278/1541, loss: 0.0680716261267662 2023-01-23 21:19:07.987541: step: 280/1541, loss: 0.00037857372080907226 2023-01-23 21:19:08.891798: step: 282/1541, loss: 0.09950162470340729 2023-01-23 21:19:09.758042: step: 284/1541, loss: 0.011750330217182636 2023-01-23 21:19:10.668772: step: 286/1541, loss: 0.001045315875671804 2023-01-23 21:19:11.499970: step: 288/1541, loss: 0.026307780295610428 2023-01-23 21:19:12.385888: step: 290/1541, loss: 0.001108636031858623 2023-01-23 21:19:13.275700: step: 292/1541, loss: 0.06190035492181778 2023-01-23 21:19:14.192799: step: 294/1541, loss: 8.43090601847507e-05 2023-01-23 21:19:15.205879: step: 296/1541, loss: 0.08671167492866516 2023-01-23 21:19:16.066731: step: 298/1541, loss: 0.29463788866996765 2023-01-23 21:19:16.939168: step: 300/1541, loss: 6.174394366098568e-05 2023-01-23 21:19:17.829972: step: 302/1541, loss: 0.00017020244558807462 2023-01-23 21:19:18.749870: step: 304/1541, loss: 0.19145135581493378 2023-01-23 21:19:19.652449: step: 306/1541, loss: 0.14402362704277039 2023-01-23 21:19:20.566197: step: 308/1541, loss: 0.0004452541470527649 2023-01-23 21:19:21.450008: step: 310/1541, loss: 0.0012298966757953167 2023-01-23 21:19:22.322553: step: 312/1541, loss: 0.05998736619949341 2023-01-23 21:19:23.248047: step: 314/1541, loss: 0.0700465515255928 2023-01-23 21:19:24.133616: step: 316/1541, loss: 0.013119015842676163 2023-01-23 21:19:25.003512: step: 318/1541, loss: 0.28753069043159485 2023-01-23 21:19:25.888825: step: 320/1541, loss: 0.05834729224443436 2023-01-23 21:19:26.759514: step: 322/1541, loss: 3.085510979872197e-05 2023-01-23 21:19:27.589560: step: 324/1541, loss: 0.00024052336812019348 2023-01-23 21:19:28.519153: step: 326/1541, loss: 9.816958481678739e-05 2023-01-23 21:19:29.396898: step: 328/1541, loss: 0.0015138626331463456 2023-01-23 21:19:30.320957: step: 330/1541, loss: 0.0020144956652075052 2023-01-23 21:19:31.219162: step: 332/1541, loss: 0.14148786664009094 2023-01-23 21:19:32.102599: step: 334/1541, loss: 0.1589018851518631 2023-01-23 21:19:33.001628: step: 336/1541, loss: 0.06617190688848495 2023-01-23 21:19:33.887915: step: 338/1541, loss: 0.05682392790913582 2023-01-23 21:19:34.737110: step: 340/1541, loss: 0.0030315995682030916 2023-01-23 21:19:35.633683: step: 342/1541, loss: 0.00010912082507275045 2023-01-23 21:19:36.548188: step: 344/1541, loss: 0.05222265422344208 2023-01-23 21:19:37.440518: step: 346/1541, loss: 0.000718752620741725 2023-01-23 21:19:38.341744: step: 348/1541, loss: 0.005616058595478535 2023-01-23 21:19:39.204494: step: 350/1541, loss: 0.0004330810625106096 2023-01-23 21:19:40.112060: step: 352/1541, loss: 0.002085033804178238 2023-01-23 21:19:41.003089: step: 354/1541, loss: 0.001559335389174521 2023-01-23 21:19:41.887329: step: 356/1541, loss: 0.011627026833593845 2023-01-23 21:19:42.792684: step: 358/1541, loss: 0.0006919152801856399 2023-01-23 21:19:43.738353: step: 360/1541, loss: 0.0003937732253689319 2023-01-23 21:19:44.638837: step: 362/1541, loss: 0.002048514550551772 2023-01-23 21:19:45.568685: step: 364/1541, loss: 0.00012210161366965622 2023-01-23 21:19:46.521072: step: 366/1541, loss: 0.0016607261495664716 2023-01-23 21:19:47.430483: step: 368/1541, loss: 0.11589132249355316 2023-01-23 21:19:48.330592: step: 370/1541, loss: 0.0026508758310228586 2023-01-23 21:19:49.231024: step: 372/1541, loss: 0.0004993539187125862 2023-01-23 21:19:50.104161: step: 374/1541, loss: 0.0005357842310331762 2023-01-23 21:19:51.013472: step: 376/1541, loss: 0.09876315295696259 2023-01-23 21:19:51.929178: step: 378/1541, loss: 0.013152766972780228 2023-01-23 21:19:52.864144: step: 380/1541, loss: 0.02268177643418312 2023-01-23 21:19:53.746454: step: 382/1541, loss: 0.031009595841169357 2023-01-23 21:19:54.674953: step: 384/1541, loss: 0.029316818341612816 2023-01-23 21:19:55.533177: step: 386/1541, loss: 0.002809118712320924 2023-01-23 21:19:56.416385: step: 388/1541, loss: 0.37158986926078796 2023-01-23 21:19:57.388336: step: 390/1541, loss: 0.002765717450529337 2023-01-23 21:19:58.325362: step: 392/1541, loss: 0.0010853923158720136 2023-01-23 21:19:59.205166: step: 394/1541, loss: 0.0008582405280321836 2023-01-23 21:20:00.109204: step: 396/1541, loss: 0.13164788484573364 2023-01-23 21:20:00.996834: step: 398/1541, loss: 0.0018099654698744416 2023-01-23 21:20:01.938893: step: 400/1541, loss: 0.005663123447448015 2023-01-23 21:20:02.808362: step: 402/1541, loss: 0.02446013316512108 2023-01-23 21:20:03.751259: step: 404/1541, loss: 0.04479800909757614 2023-01-23 21:20:04.681356: step: 406/1541, loss: 3.6701414501294494e-05 2023-01-23 21:20:05.660374: step: 408/1541, loss: 0.04231713339686394 2023-01-23 21:20:06.587009: step: 410/1541, loss: 0.10109715163707733 2023-01-23 21:20:07.457247: step: 412/1541, loss: 8.471260662190616e-05 2023-01-23 21:20:08.338926: step: 414/1541, loss: 5.3919211495667696e-05 2023-01-23 21:20:09.256368: step: 416/1541, loss: 0.46648460626602173 2023-01-23 21:20:10.130959: step: 418/1541, loss: 0.025439061224460602 2023-01-23 21:20:10.991298: step: 420/1541, loss: 0.005794030614197254 2023-01-23 21:20:11.831597: step: 422/1541, loss: 0.0005299053736962378 2023-01-23 21:20:12.721155: step: 424/1541, loss: 0.41968756914138794 2023-01-23 21:20:13.627160: step: 426/1541, loss: 4.4586016883840784e-05 2023-01-23 21:20:14.547447: step: 428/1541, loss: 0.0018547051586210728 2023-01-23 21:20:15.452177: step: 430/1541, loss: 0.0001302634773310274 2023-01-23 21:20:16.290674: step: 432/1541, loss: 0.04809632524847984 2023-01-23 21:20:17.215101: step: 434/1541, loss: 0.7413986921310425 2023-01-23 21:20:18.082741: step: 436/1541, loss: 0.16094255447387695 2023-01-23 21:20:18.951744: step: 438/1541, loss: 0.07604566216468811 2023-01-23 21:20:19.852250: step: 440/1541, loss: 0.008612383157014847 2023-01-23 21:20:20.685115: step: 442/1541, loss: 0.00030031948699615896 2023-01-23 21:20:21.543323: step: 444/1541, loss: 0.22612902522087097 2023-01-23 21:20:22.442760: step: 446/1541, loss: 0.00039829552406445146 2023-01-23 21:20:23.360452: step: 448/1541, loss: 0.005415217485278845 2023-01-23 21:20:24.273361: step: 450/1541, loss: 0.07092544436454773 2023-01-23 21:20:25.205878: step: 452/1541, loss: 0.00018746792920865119 2023-01-23 21:20:26.136224: step: 454/1541, loss: 0.00042542393202893436 2023-01-23 21:20:27.047267: step: 456/1541, loss: 0.0001236469397554174 2023-01-23 21:20:27.946768: step: 458/1541, loss: 0.00029233447276055813 2023-01-23 21:20:28.860670: step: 460/1541, loss: 0.04245573654770851 2023-01-23 21:20:29.724666: step: 462/1541, loss: 0.0005416856147348881 2023-01-23 21:20:30.638225: step: 464/1541, loss: 0.0039555905386805534 2023-01-23 21:20:31.577165: step: 466/1541, loss: 0.08279909938573837 2023-01-23 21:20:32.436267: step: 468/1541, loss: 0.0002223778865300119 2023-01-23 21:20:33.356900: step: 470/1541, loss: 0.05274762958288193 2023-01-23 21:20:34.264363: step: 472/1541, loss: 0.00012576475273817778 2023-01-23 21:20:35.136778: step: 474/1541, loss: 0.0006338809616863728 2023-01-23 21:20:36.047785: step: 476/1541, loss: 0.0012019311543554068 2023-01-23 21:20:36.931409: step: 478/1541, loss: 0.34458160400390625 2023-01-23 21:20:37.797638: step: 480/1541, loss: 0.0003653598832897842 2023-01-23 21:20:38.685059: step: 482/1541, loss: 0.07652474194765091 2023-01-23 21:20:39.574308: step: 484/1541, loss: 0.31385159492492676 2023-01-23 21:20:40.513252: step: 486/1541, loss: 0.0005932151107117534 2023-01-23 21:20:41.413024: step: 488/1541, loss: 0.03676406666636467 2023-01-23 21:20:42.304136: step: 490/1541, loss: 0.0002499255642760545 2023-01-23 21:20:43.239619: step: 492/1541, loss: 0.00025184571859426796 2023-01-23 21:20:44.185799: step: 494/1541, loss: 0.14109010994434357 2023-01-23 21:20:45.090791: step: 496/1541, loss: 0.0018146807560697198 2023-01-23 21:20:45.982964: step: 498/1541, loss: 0.06956673413515091 2023-01-23 21:20:46.890868: step: 500/1541, loss: 0.023968670517206192 2023-01-23 21:20:47.756545: step: 502/1541, loss: 0.0010739403078332543 2023-01-23 21:20:48.643388: step: 504/1541, loss: 0.002055396093055606 2023-01-23 21:20:49.586510: step: 506/1541, loss: 0.019118716940283775 2023-01-23 21:20:50.435526: step: 508/1541, loss: 0.0468178391456604 2023-01-23 21:20:51.383488: step: 510/1541, loss: 0.0005404601106420159 2023-01-23 21:20:52.332597: step: 512/1541, loss: 0.0002660294121596962 2023-01-23 21:20:53.223419: step: 514/1541, loss: 0.0001907639525597915 2023-01-23 21:20:54.102884: step: 516/1541, loss: 0.006608129478991032 2023-01-23 21:20:55.017404: step: 518/1541, loss: 0.4560336768627167 2023-01-23 21:20:55.933108: step: 520/1541, loss: 0.14215479791164398 2023-01-23 21:20:56.801745: step: 522/1541, loss: 8.035069913603365e-05 2023-01-23 21:20:57.639682: step: 524/1541, loss: 0.00013589880836661905 2023-01-23 21:20:58.529861: step: 526/1541, loss: 0.1623097062110901 2023-01-23 21:20:59.452233: step: 528/1541, loss: 0.0009924453916028142 2023-01-23 21:21:00.287268: step: 530/1541, loss: 0.00019808208162430674 2023-01-23 21:21:01.268278: step: 532/1541, loss: 0.05773625522851944 2023-01-23 21:21:02.202966: step: 534/1541, loss: 0.034893982112407684 2023-01-23 21:21:03.106388: step: 536/1541, loss: 0.0004589080635923892 2023-01-23 21:21:03.976907: step: 538/1541, loss: 0.10047570616006851 2023-01-23 21:21:04.861952: step: 540/1541, loss: 0.3636973202228546 2023-01-23 21:21:05.743269: step: 542/1541, loss: 0.04720123112201691 2023-01-23 21:21:06.611510: step: 544/1541, loss: 0.0006011273362673819 2023-01-23 21:21:07.483990: step: 546/1541, loss: 0.0005160315777175128 2023-01-23 21:21:08.389246: step: 548/1541, loss: 0.0005277449963614345 2023-01-23 21:21:09.284172: step: 550/1541, loss: 0.010399050079286098 2023-01-23 21:21:10.154940: step: 552/1541, loss: 0.0024336064234375954 2023-01-23 21:21:11.112360: step: 554/1541, loss: 0.16448843479156494 2023-01-23 21:21:12.026351: step: 556/1541, loss: 0.00032488539000041783 2023-01-23 21:21:12.924326: step: 558/1541, loss: 0.006575283594429493 2023-01-23 21:21:13.845061: step: 560/1541, loss: 0.00029082002583891153 2023-01-23 21:21:14.759054: step: 562/1541, loss: 0.0005862418911419809 2023-01-23 21:21:15.605791: step: 564/1541, loss: 0.004440647084265947 2023-01-23 21:21:16.522934: step: 566/1541, loss: 0.0006565529038198292 2023-01-23 21:21:17.481242: step: 568/1541, loss: 0.06839170306921005 2023-01-23 21:21:18.386849: step: 570/1541, loss: 0.05310308188199997 2023-01-23 21:21:19.284826: step: 572/1541, loss: 5.310475899023004e-05 2023-01-23 21:21:20.161702: step: 574/1541, loss: 0.36523640155792236 2023-01-23 21:21:21.053390: step: 576/1541, loss: 3.1201288948068395e-05 2023-01-23 21:21:21.946279: step: 578/1541, loss: 0.008267570286989212 2023-01-23 21:21:22.819688: step: 580/1541, loss: 0.0006700222147628665 2023-01-23 21:21:23.724552: step: 582/1541, loss: 0.0006790286279283464 2023-01-23 21:21:24.560083: step: 584/1541, loss: 0.0006670407019555569 2023-01-23 21:21:25.483905: step: 586/1541, loss: 0.0001763903710525483 2023-01-23 21:21:26.412799: step: 588/1541, loss: 0.002496433909982443 2023-01-23 21:21:27.337436: step: 590/1541, loss: 0.11332675814628601 2023-01-23 21:21:28.205098: step: 592/1541, loss: 4.535669359029271e-05 2023-01-23 21:21:29.066776: step: 594/1541, loss: 0.002512263599783182 2023-01-23 21:21:30.005482: step: 596/1541, loss: 0.01301061362028122 2023-01-23 21:21:30.849312: step: 598/1541, loss: 0.00019307165348436683 2023-01-23 21:21:31.732616: step: 600/1541, loss: 0.16041290760040283 2023-01-23 21:21:32.640939: step: 602/1541, loss: 0.11080832779407501 2023-01-23 21:21:33.538407: step: 604/1541, loss: 6.319002750387881e-06 2023-01-23 21:21:34.468913: step: 606/1541, loss: -2.7547266654437408e-05 2023-01-23 21:21:35.371206: step: 608/1541, loss: 7.045641541481018e-05 2023-01-23 21:21:36.258441: step: 610/1541, loss: 0.0011467376025393605 2023-01-23 21:21:37.091019: step: 612/1541, loss: 0.0013852228876203299 2023-01-23 21:21:37.959294: step: 614/1541, loss: 1.683308437350206e-05 2023-01-23 21:21:38.847397: step: 616/1541, loss: 0.12414567917585373 2023-01-23 21:21:39.734115: step: 618/1541, loss: 0.0006240633083507419 2023-01-23 21:21:40.650833: step: 620/1541, loss: 0.0032144137658178806 2023-01-23 21:21:41.557466: step: 622/1541, loss: 5.6241533457068726e-05 2023-01-23 21:21:42.482901: step: 624/1541, loss: 9.663900709711015e-05 2023-01-23 21:21:43.430590: step: 626/1541, loss: 0.6114293932914734 2023-01-23 21:21:44.324844: step: 628/1541, loss: 1.5473067833227105e-05 2023-01-23 21:21:45.292981: step: 630/1541, loss: 0.055465053766965866 2023-01-23 21:21:46.166283: step: 632/1541, loss: 0.00010261982242809609 2023-01-23 21:21:47.044400: step: 634/1541, loss: 0.048096951097249985 2023-01-23 21:21:47.912427: step: 636/1541, loss: 3.526916407281533e-05 2023-01-23 21:21:48.815779: step: 638/1541, loss: 0.48152580857276917 2023-01-23 21:21:49.739712: step: 640/1541, loss: 0.030567379668354988 2023-01-23 21:21:50.680000: step: 642/1541, loss: 0.0002053887874353677 2023-01-23 21:21:51.607201: step: 644/1541, loss: 2.1844096409040503e-05 2023-01-23 21:21:52.516648: step: 646/1541, loss: -1.1967309546889737e-05 2023-01-23 21:21:53.399661: step: 648/1541, loss: 0.036174941807985306 2023-01-23 21:21:54.243213: step: 650/1541, loss: 0.00013704760931432247 2023-01-23 21:21:55.149671: step: 652/1541, loss: 4.242571594659239e-05 2023-01-23 21:21:56.079463: step: 654/1541, loss: 0.004053850192576647 2023-01-23 21:21:56.982917: step: 656/1541, loss: 0.021922757849097252 2023-01-23 21:21:57.902079: step: 658/1541, loss: 4.433810681803152e-05 2023-01-23 21:21:58.830875: step: 660/1541, loss: 0.0646040216088295 2023-01-23 21:21:59.700278: step: 662/1541, loss: 7.855928561184555e-05 2023-01-23 21:22:00.593255: step: 664/1541, loss: 0.0016732019139453769 2023-01-23 21:22:01.499332: step: 666/1541, loss: 0.08655521273612976 2023-01-23 21:22:02.381482: step: 668/1541, loss: 0.025649450719356537 2023-01-23 21:22:03.295304: step: 670/1541, loss: 0.0020972113125026226 2023-01-23 21:22:04.187039: step: 672/1541, loss: 2.1368257876019925e-05 2023-01-23 21:22:05.136188: step: 674/1541, loss: 0.0002024351997533813 2023-01-23 21:22:06.083945: step: 676/1541, loss: 0.00016996166959870607 2023-01-23 21:22:07.083884: step: 678/1541, loss: 0.15696917474269867 2023-01-23 21:22:07.951195: step: 680/1541, loss: 1.875996713351924e-05 2023-01-23 21:22:08.815680: step: 682/1541, loss: 0.003147277981042862 2023-01-23 21:22:09.680360: step: 684/1541, loss: 0.004075591452419758 2023-01-23 21:22:10.579659: step: 686/1541, loss: 2.0745894289575517e-05 2023-01-23 21:22:11.456324: step: 688/1541, loss: 7.254379306687042e-05 2023-01-23 21:22:12.401386: step: 690/1541, loss: 0.0006790816551074386 2023-01-23 21:22:13.309823: step: 692/1541, loss: 0.0016843769699335098 2023-01-23 21:22:14.164871: step: 694/1541, loss: 0.0004225785378366709 2023-01-23 21:22:15.048320: step: 696/1541, loss: 0.08339795470237732 2023-01-23 21:22:15.929776: step: 698/1541, loss: 8.920542313717306e-05 2023-01-23 21:22:16.812801: step: 700/1541, loss: 0.00014783309597987682 2023-01-23 21:22:17.706918: step: 702/1541, loss: 0.21695028245449066 2023-01-23 21:22:18.643906: step: 704/1541, loss: 0.09682732820510864 2023-01-23 21:22:19.518899: step: 706/1541, loss: 0.000907766749151051 2023-01-23 21:22:20.486406: step: 708/1541, loss: 6.120142643339932e-05 2023-01-23 21:22:21.377026: step: 710/1541, loss: 0.001836612937040627 2023-01-23 21:22:22.242168: step: 712/1541, loss: 7.928957347758114e-05 2023-01-23 21:22:23.095680: step: 714/1541, loss: 3.576149174477905e-05 2023-01-23 21:22:24.003215: step: 716/1541, loss: 0.706050455570221 2023-01-23 21:22:24.916444: step: 718/1541, loss: 0.011142626404762268 2023-01-23 21:22:25.862116: step: 720/1541, loss: 0.023333728313446045 2023-01-23 21:22:26.757095: step: 722/1541, loss: 0.002538189524784684 2023-01-23 21:22:27.622482: step: 724/1541, loss: 0.0014298786409199238 2023-01-23 21:22:28.528647: step: 726/1541, loss: 8.94183394848369e-05 2023-01-23 21:22:29.457788: step: 728/1541, loss: 0.0037844404578208923 2023-01-23 21:22:30.441397: step: 730/1541, loss: 4.228888428770006e-05 2023-01-23 21:22:31.326030: step: 732/1541, loss: 0.07052629441022873 2023-01-23 21:22:32.262521: step: 734/1541, loss: 0.00012358679668977857 2023-01-23 21:22:33.257966: step: 736/1541, loss: 0.00015631687710992992 2023-01-23 21:22:34.225997: step: 738/1541, loss: 0.10014073550701141 2023-01-23 21:22:35.095759: step: 740/1541, loss: 5.649352169712074e-05 2023-01-23 21:22:35.994581: step: 742/1541, loss: -1.985828384931665e-05 2023-01-23 21:22:36.890967: step: 744/1541, loss: 0.10020188987255096 2023-01-23 21:22:37.785536: step: 746/1541, loss: 4.849895049119368e-05 2023-01-23 21:22:38.678328: step: 748/1541, loss: 0.000533826183527708 2023-01-23 21:22:39.553162: step: 750/1541, loss: 4.2893385398201644e-05 2023-01-23 21:22:40.518898: step: 752/1541, loss: 5.587562918663025e-05 2023-01-23 21:22:41.393954: step: 754/1541, loss: 0.0005608075298368931 2023-01-23 21:22:42.267897: step: 756/1541, loss: 4.769165752804838e-05 2023-01-23 21:22:43.189696: step: 758/1541, loss: 0.0010788519866764545 2023-01-23 21:22:44.085643: step: 760/1541, loss: 4.3806430767290294e-05 2023-01-23 21:22:45.006068: step: 762/1541, loss: 0.015622054226696491 2023-01-23 21:22:45.894702: step: 764/1541, loss: 0.00017062197730410844 2023-01-23 21:22:46.744266: step: 766/1541, loss: 0.053488895297050476 2023-01-23 21:22:47.624447: step: 768/1541, loss: 0.007443512789905071 2023-01-23 21:22:48.531057: step: 770/1541, loss: 0.01980423368513584 2023-01-23 21:22:49.443633: step: 772/1541, loss: 0.03282457962632179 2023-01-23 21:22:50.356835: step: 774/1541, loss: 1.8615184671944007e-05 2023-01-23 21:22:51.228912: step: 776/1541, loss: 0.0002876253565773368 2023-01-23 21:22:52.110781: step: 778/1541, loss: 0.005794807802885771 2023-01-23 21:22:52.974632: step: 780/1541, loss: 0.10006265342235565 2023-01-23 21:22:53.868448: step: 782/1541, loss: 2.6078803784912452e-05 2023-01-23 21:22:54.756674: step: 784/1541, loss: 0.00023592972138430923 2023-01-23 21:22:55.613093: step: 786/1541, loss: 0.027071719989180565 2023-01-23 21:22:56.465709: step: 788/1541, loss: 0.21614806354045868 2023-01-23 21:22:57.348850: step: 790/1541, loss: 7.614339119754732e-05 2023-01-23 21:22:58.247958: step: 792/1541, loss: 0.016425026580691338 2023-01-23 21:22:59.222968: step: 794/1541, loss: 0.23562151193618774 2023-01-23 21:23:00.098482: step: 796/1541, loss: 1.215685551869683e-05 2023-01-23 21:23:00.983062: step: 798/1541, loss: 0.012102028355002403 2023-01-23 21:23:01.860675: step: 800/1541, loss: 1.4376987564901356e-05 2023-01-23 21:23:02.704384: step: 802/1541, loss: 0.00011544861627044156 2023-01-23 21:23:03.605018: step: 804/1541, loss: 0.0019237224478274584 2023-01-23 21:23:04.524947: step: 806/1541, loss: 0.0818520337343216 2023-01-23 21:23:05.420291: step: 808/1541, loss: 1.3700121144211153e-06 2023-01-23 21:23:06.321792: step: 810/1541, loss: 0.33432888984680176 2023-01-23 21:23:07.207103: step: 812/1541, loss: 0.07680332660675049 2023-01-23 21:23:08.118380: step: 814/1541, loss: 0.00031523086363449693 2023-01-23 21:23:09.107013: step: 816/1541, loss: -1.02296519344236e-06 2023-01-23 21:23:09.976522: step: 818/1541, loss: 0.0005355719476938248 2023-01-23 21:23:10.872092: step: 820/1541, loss: -8.32974365039263e-06 2023-01-23 21:23:11.783600: step: 822/1541, loss: 0.00043576108873821795 2023-01-23 21:23:12.691569: step: 824/1541, loss: 0.005302715580910444 2023-01-23 21:23:13.579492: step: 826/1541, loss: 0.009990931488573551 2023-01-23 21:23:14.477964: step: 828/1541, loss: 0.0005845763953402638 2023-01-23 21:23:15.396860: step: 830/1541, loss: 0.004062253050506115 2023-01-23 21:23:16.328194: step: 832/1541, loss: 0.00021083727187942713 2023-01-23 21:23:17.254884: step: 834/1541, loss: 0.05794781446456909 2023-01-23 21:23:18.138950: step: 836/1541, loss: 0.00048819996300153434 2023-01-23 21:23:19.079418: step: 838/1541, loss: 0.3031432032585144 2023-01-23 21:23:19.995068: step: 840/1541, loss: 0.03538389503955841 2023-01-23 21:23:20.852192: step: 842/1541, loss: 0.0008422869723290205 2023-01-23 21:23:21.695586: step: 844/1541, loss: 0.0026151929050683975 2023-01-23 21:23:22.578333: step: 846/1541, loss: 0.0159853994846344 2023-01-23 21:23:23.480382: step: 848/1541, loss: 0.0002390393492532894 2023-01-23 21:23:24.519554: step: 850/1541, loss: 0.16214029490947723 2023-01-23 21:23:25.381933: step: 852/1541, loss: 0.014317210763692856 2023-01-23 21:23:26.254768: step: 854/1541, loss: 0.00033439305843785405 2023-01-23 21:23:27.084763: step: 856/1541, loss: 0.0014097040984779596 2023-01-23 21:23:27.982208: step: 858/1541, loss: 0.12467937916517258 2023-01-23 21:23:28.917404: step: 860/1541, loss: 0.00038283882895484567 2023-01-23 21:23:29.815199: step: 862/1541, loss: 5.487891030497849e-05 2023-01-23 21:23:30.712013: step: 864/1541, loss: 0.010566442273557186 2023-01-23 21:23:31.639762: step: 866/1541, loss: 7.430223195115104e-05 2023-01-23 21:23:32.584128: step: 868/1541, loss: 6.702847895212471e-05 2023-01-23 21:23:33.495801: step: 870/1541, loss: 5.990718818793539e-06 2023-01-23 21:23:34.365080: step: 872/1541, loss: 0.00017199951980728656 2023-01-23 21:23:35.279719: step: 874/1541, loss: 0.04826049506664276 2023-01-23 21:23:36.140504: step: 876/1541, loss: 0.09976356476545334 2023-01-23 21:23:37.042135: step: 878/1541, loss: 0.07707572728395462 2023-01-23 21:23:37.923377: step: 880/1541, loss: 1.386732401442714e-05 2023-01-23 21:23:38.864035: step: 882/1541, loss: 0.04607364907860756 2023-01-23 21:23:39.759639: step: 884/1541, loss: 0.0026537240482866764 2023-01-23 21:23:40.717060: step: 886/1541, loss: 0.03078310750424862 2023-01-23 21:23:41.611351: step: 888/1541, loss: 0.0015870018396526575 2023-01-23 21:23:42.511436: step: 890/1541, loss: 0.00013020558981224895 2023-01-23 21:23:43.438972: step: 892/1541, loss: 0.5457150340080261 2023-01-23 21:23:44.374280: step: 894/1541, loss: 1.0838532447814941 2023-01-23 21:23:45.311404: step: 896/1541, loss: 0.0007406065124087036 2023-01-23 21:23:46.161207: step: 898/1541, loss: 0.002011805074289441 2023-01-23 21:23:47.046770: step: 900/1541, loss: 0.03260943666100502 2023-01-23 21:23:47.986908: step: 902/1541, loss: 0.0003903098404407501 2023-01-23 21:23:48.950355: step: 904/1541, loss: 0.11593426764011383 2023-01-23 21:23:49.876154: step: 906/1541, loss: 0.0017554680816829205 2023-01-23 21:23:50.781115: step: 908/1541, loss: 0.0011691447580233216 2023-01-23 21:23:51.689538: step: 910/1541, loss: 0.38245919346809387 2023-01-23 21:23:52.588905: step: 912/1541, loss: 0.0010596485808491707 2023-01-23 21:23:53.487706: step: 914/1541, loss: 0.0007213306962512434 2023-01-23 21:23:54.390304: step: 916/1541, loss: 0.0004430031985975802 2023-01-23 21:23:55.299779: step: 918/1541, loss: 0.0016508224653080106 2023-01-23 21:23:56.195871: step: 920/1541, loss: 0.00033910313504748046 2023-01-23 21:23:57.086829: step: 922/1541, loss: 0.014749373309314251 2023-01-23 21:23:57.952190: step: 924/1541, loss: 0.0004093990719411522 2023-01-23 21:23:58.868719: step: 926/1541, loss: 0.0005772198201157153 2023-01-23 21:23:59.746028: step: 928/1541, loss: 0.16470086574554443 2023-01-23 21:24:00.675125: step: 930/1541, loss: 1.9561707631510217e-06 2023-01-23 21:24:01.559571: step: 932/1541, loss: 0.0001510735892225057 2023-01-23 21:24:02.428760: step: 934/1541, loss: 0.00010250904597342014 2023-01-23 21:24:03.310511: step: 936/1541, loss: 0.0018438893603160977 2023-01-23 21:24:04.213973: step: 938/1541, loss: 2.9822236683685333e-05 2023-01-23 21:24:05.094313: step: 940/1541, loss: 0.08512765914201736 2023-01-23 21:24:05.995694: step: 942/1541, loss: 0.0014595644315704703 2023-01-23 21:24:06.869582: step: 944/1541, loss: 0.06237313523888588 2023-01-23 21:24:07.753954: step: 946/1541, loss: 0.00015781784895807505 2023-01-23 21:24:08.614922: step: 948/1541, loss: 0.0001193508505821228 2023-01-23 21:24:09.461859: step: 950/1541, loss: 0.043957654386758804 2023-01-23 21:24:10.358341: step: 952/1541, loss: 0.0015427330508828163 2023-01-23 21:24:11.250714: step: 954/1541, loss: 0.0010355281410738826 2023-01-23 21:24:12.132751: step: 956/1541, loss: 3.529101741150953e-05 2023-01-23 21:24:13.037579: step: 958/1541, loss: 0.0005831705057062209 2023-01-23 21:24:13.934152: step: 960/1541, loss: 0.4372755289077759 2023-01-23 21:24:14.789821: step: 962/1541, loss: 0.02115846425294876 2023-01-23 21:24:15.680109: step: 964/1541, loss: 0.0007613387424498796 2023-01-23 21:24:16.583441: step: 966/1541, loss: 0.007147602271288633 2023-01-23 21:24:17.493894: step: 968/1541, loss: 0.001220537698827684 2023-01-23 21:24:18.402356: step: 970/1541, loss: 0.08718756586313248 2023-01-23 21:24:19.298368: step: 972/1541, loss: 6.654794560745358e-05 2023-01-23 21:24:20.184142: step: 974/1541, loss: 0.014617500826716423 2023-01-23 21:24:21.061564: step: 976/1541, loss: 0.0008202884346246719 2023-01-23 21:24:21.940177: step: 978/1541, loss: 0.00010756528354249895 2023-01-23 21:24:22.832306: step: 980/1541, loss: 0.0006220383220352232 2023-01-23 21:24:23.730828: step: 982/1541, loss: 0.25898224115371704 2023-01-23 21:24:24.626594: step: 984/1541, loss: 0.009505251422524452 2023-01-23 21:24:25.475413: step: 986/1541, loss: 0.00031038152519613504 2023-01-23 21:24:26.423456: step: 988/1541, loss: 0.005755643825978041 2023-01-23 21:24:27.375454: step: 990/1541, loss: 0.01710854284465313 2023-01-23 21:24:28.247717: step: 992/1541, loss: 0.3936667740345001 2023-01-23 21:24:29.144964: step: 994/1541, loss: 0.0002315408637514338 2023-01-23 21:24:30.035129: step: 996/1541, loss: 0.023575132712721825 2023-01-23 21:24:30.892556: step: 998/1541, loss: 0.013290787115693092 2023-01-23 21:24:31.758436: step: 1000/1541, loss: 0.016789674758911133 2023-01-23 21:24:32.641066: step: 1002/1541, loss: 0.0005585234612226486 2023-01-23 21:24:33.598789: step: 1004/1541, loss: 0.051430538296699524 2023-01-23 21:24:34.526111: step: 1006/1541, loss: 0.10086911171674728 2023-01-23 21:24:35.386732: step: 1008/1541, loss: 0.00825825147330761 2023-01-23 21:24:36.287849: step: 1010/1541, loss: 0.013658207841217518 2023-01-23 21:24:37.192751: step: 1012/1541, loss: 0.013955331407487392 2023-01-23 21:24:38.085035: step: 1014/1541, loss: 4.143518162891269e-05 2023-01-23 21:24:38.941518: step: 1016/1541, loss: -8.332740435434971e-06 2023-01-23 21:24:39.793035: step: 1018/1541, loss: 0.011367381550371647 2023-01-23 21:24:40.674743: step: 1020/1541, loss: 6.917744758538902e-05 2023-01-23 21:24:41.526853: step: 1022/1541, loss: 0.0001554528425913304 2023-01-23 21:24:42.477047: step: 1024/1541, loss: 0.015601077117025852 2023-01-23 21:24:43.360862: step: 1026/1541, loss: 0.07471605390310287 2023-01-23 21:24:44.288097: step: 1028/1541, loss: 0.00923585519194603 2023-01-23 21:24:45.181349: step: 1030/1541, loss: 0.042330242693424225 2023-01-23 21:24:46.080561: step: 1032/1541, loss: 0.7103049755096436 2023-01-23 21:24:47.016326: step: 1034/1541, loss: 0.0005924063734710217 2023-01-23 21:24:47.977577: step: 1036/1541, loss: 0.11775357276201248 2023-01-23 21:24:48.854614: step: 1038/1541, loss: 0.009948993101716042 2023-01-23 21:24:49.759604: step: 1040/1541, loss: 0.07456641644239426 2023-01-23 21:24:50.650933: step: 1042/1541, loss: 7.1971944635151885e-06 2023-01-23 21:24:51.548094: step: 1044/1541, loss: 0.0019208085723221302 2023-01-23 21:24:52.466509: step: 1046/1541, loss: 0.008259681984782219 2023-01-23 21:24:53.345512: step: 1048/1541, loss: 0.00016917289758566767 2023-01-23 21:24:54.263372: step: 1050/1541, loss: 0.14789344370365143 2023-01-23 21:24:55.184221: step: 1052/1541, loss: 0.00017575290985405445 2023-01-23 21:24:56.050908: step: 1054/1541, loss: 0.0005670437240041792 2023-01-23 21:24:56.942860: step: 1056/1541, loss: 0.017930345609784126 2023-01-23 21:24:57.844536: step: 1058/1541, loss: 0.0017651154194027185 2023-01-23 21:24:58.707174: step: 1060/1541, loss: 4.145455750403926e-05 2023-01-23 21:24:59.586923: step: 1062/1541, loss: 0.0006333795608952641 2023-01-23 21:25:00.460173: step: 1064/1541, loss: 0.0008271755650639534 2023-01-23 21:25:01.325060: step: 1066/1541, loss: 0.0001998600346269086 2023-01-23 21:25:02.270966: step: 1068/1541, loss: 0.07839681953191757 2023-01-23 21:25:03.201148: step: 1070/1541, loss: 0.00034712281194515526 2023-01-23 21:25:04.071987: step: 1072/1541, loss: 0.0009212724398821592 2023-01-23 21:25:04.939031: step: 1074/1541, loss: 0.004799296148121357 2023-01-23 21:25:05.824385: step: 1076/1541, loss: 0.07115459442138672 2023-01-23 21:25:06.755824: step: 1078/1541, loss: 0.13338026404380798 2023-01-23 21:25:07.663542: step: 1080/1541, loss: 9.706090349936858e-05 2023-01-23 21:25:08.597461: step: 1082/1541, loss: 0.021728867664933205 2023-01-23 21:25:09.518165: step: 1084/1541, loss: 0.018001899123191833 2023-01-23 21:25:10.404789: step: 1086/1541, loss: 0.004708166234195232 2023-01-23 21:25:11.297210: step: 1088/1541, loss: 0.0006684497348032892 2023-01-23 21:25:12.196190: step: 1090/1541, loss: 0.00086885568452999 2023-01-23 21:25:13.142142: step: 1092/1541, loss: 2.4034081434365362e-05 2023-01-23 21:25:13.988883: step: 1094/1541, loss: 0.0001270316424779594 2023-01-23 21:25:14.860021: step: 1096/1541, loss: 0.004203286487609148 2023-01-23 21:25:15.714293: step: 1098/1541, loss: 0.06082536652684212 2023-01-23 21:25:16.584801: step: 1100/1541, loss: 0.0006645168177783489 2023-01-23 21:25:17.557657: step: 1102/1541, loss: 0.0002900836698245257 2023-01-23 21:25:18.484354: step: 1104/1541, loss: 0.01805184781551361 2023-01-23 21:25:19.449882: step: 1106/1541, loss: 0.008441613987088203 2023-01-23 21:25:20.374282: step: 1108/1541, loss: 0.00016721710562705994 2023-01-23 21:25:21.248245: step: 1110/1541, loss: 0.0747857317328453 2023-01-23 21:25:22.160438: step: 1112/1541, loss: 0.006227947771549225 2023-01-23 21:25:23.165040: step: 1114/1541, loss: 0.15131953358650208 2023-01-23 21:25:24.032681: step: 1116/1541, loss: 0.003419315442442894 2023-01-23 21:25:24.946115: step: 1118/1541, loss: 9.99983458314091e-05 2023-01-23 21:25:25.880988: step: 1120/1541, loss: 0.0016288070473819971 2023-01-23 21:25:26.775917: step: 1122/1541, loss: 0.01661446876823902 2023-01-23 21:25:27.668792: step: 1124/1541, loss: 0.0003887322382070124 2023-01-23 21:25:28.524654: step: 1126/1541, loss: 0.04790261760354042 2023-01-23 21:25:29.412748: step: 1128/1541, loss: 0.006051352247595787 2023-01-23 21:25:30.326907: step: 1130/1541, loss: 0.0013631333131343126 2023-01-23 21:25:31.229414: step: 1132/1541, loss: 0.00024933108943514526 2023-01-23 21:25:32.197423: step: 1134/1541, loss: 0.2855913043022156 2023-01-23 21:25:33.121096: step: 1136/1541, loss: 0.13964468240737915 2023-01-23 21:25:34.048006: step: 1138/1541, loss: 0.0038877499755471945 2023-01-23 21:25:34.941757: step: 1140/1541, loss: 0.0005005089333280921 2023-01-23 21:25:35.829230: step: 1142/1541, loss: 0.00034755392698571086 2023-01-23 21:25:36.743602: step: 1144/1541, loss: 0.1641596108675003 2023-01-23 21:25:37.625733: step: 1146/1541, loss: 0.009783290326595306 2023-01-23 21:25:38.525362: step: 1148/1541, loss: 6.226957339094952e-05 2023-01-23 21:25:39.382715: step: 1150/1541, loss: 8.039946806093212e-06 2023-01-23 21:25:40.243153: step: 1152/1541, loss: 0.00037954680738039315 2023-01-23 21:25:41.120798: step: 1154/1541, loss: 0.0003923986223526299 2023-01-23 21:25:41.982436: step: 1156/1541, loss: 0.03373786062002182 2023-01-23 21:25:42.868072: step: 1158/1541, loss: 0.09656938165426254 2023-01-23 21:25:43.754798: step: 1160/1541, loss: 0.0005567345069721341 2023-01-23 21:25:44.650390: step: 1162/1541, loss: 0.0020916692446917295 2023-01-23 21:25:45.547991: step: 1164/1541, loss: 0.0056694005616009235 2023-01-23 21:25:46.450158: step: 1166/1541, loss: 0.002018200932070613 2023-01-23 21:25:47.316625: step: 1168/1541, loss: 0.0017764234216883779 2023-01-23 21:25:48.211677: step: 1170/1541, loss: 0.0002831154561135918 2023-01-23 21:25:49.091945: step: 1172/1541, loss: 0.16107583045959473 2023-01-23 21:25:49.946369: step: 1174/1541, loss: 0.0003065984637942165 2023-01-23 21:25:50.807316: step: 1176/1541, loss: 0.00038889507413841784 2023-01-23 21:25:51.702381: step: 1178/1541, loss: 0.00043749806354753673 2023-01-23 21:25:52.597096: step: 1180/1541, loss: 0.03732014819979668 2023-01-23 21:25:53.499361: step: 1182/1541, loss: 0.0010792459361255169 2023-01-23 21:25:54.390250: step: 1184/1541, loss: 0.004342158790677786 2023-01-23 21:25:55.293466: step: 1186/1541, loss: 0.08832195401191711 2023-01-23 21:25:56.168848: step: 1188/1541, loss: 3.292121255071834e-05 2023-01-23 21:25:57.047514: step: 1190/1541, loss: 0.3250707983970642 2023-01-23 21:25:57.922567: step: 1192/1541, loss: 0.04102453589439392 2023-01-23 21:25:58.802446: step: 1194/1541, loss: -5.212680207478115e-06 2023-01-23 21:25:59.677971: step: 1196/1541, loss: 0.03238225355744362 2023-01-23 21:26:00.603680: step: 1198/1541, loss: 0.00020251386740710586 2023-01-23 21:26:01.526428: step: 1200/1541, loss: 0.14675091207027435 2023-01-23 21:26:02.435593: step: 1202/1541, loss: 0.13468289375305176 2023-01-23 21:26:03.362043: step: 1204/1541, loss: 0.00804906990379095 2023-01-23 21:26:04.222016: step: 1206/1541, loss: 0.0013367735082283616 2023-01-23 21:26:05.170915: step: 1208/1541, loss: 0.0009812351781874895 2023-01-23 21:26:06.080418: step: 1210/1541, loss: 0.0662287175655365 2023-01-23 21:26:06.978570: step: 1212/1541, loss: 0.005649674218147993 2023-01-23 21:26:07.852689: step: 1214/1541, loss: 0.023336365818977356 2023-01-23 21:26:08.756539: step: 1216/1541, loss: 0.16242925822734833 2023-01-23 21:26:09.652904: step: 1218/1541, loss: 0.047004345804452896 2023-01-23 21:26:10.627788: step: 1220/1541, loss: 0.02401815541088581 2023-01-23 21:26:11.608954: step: 1222/1541, loss: 0.028109362348914146 2023-01-23 21:26:12.513363: step: 1224/1541, loss: 0.06386903673410416 2023-01-23 21:26:13.444631: step: 1226/1541, loss: 0.19166485965251923 2023-01-23 21:26:14.332449: step: 1228/1541, loss: 0.0011608561035245657 2023-01-23 21:26:15.211017: step: 1230/1541, loss: 0.016035720705986023 2023-01-23 21:26:16.063800: step: 1232/1541, loss: 0.0022476210724562407 2023-01-23 21:26:16.997015: step: 1234/1541, loss: 0.011372881941497326 2023-01-23 21:26:17.844326: step: 1236/1541, loss: 0.005045575089752674 2023-01-23 21:26:18.848641: step: 1238/1541, loss: 0.08529044687747955 2023-01-23 21:26:19.747046: step: 1240/1541, loss: 0.0017671155510470271 2023-01-23 21:26:20.612259: step: 1242/1541, loss: 0.29827117919921875 2023-01-23 21:26:21.509713: step: 1244/1541, loss: 0.06952071189880371 2023-01-23 21:26:22.444548: step: 1246/1541, loss: 0.0014198602875694633 2023-01-23 21:26:23.358062: step: 1248/1541, loss: 0.008592631667852402 2023-01-23 21:26:24.297717: step: 1250/1541, loss: 0.025062672793865204 2023-01-23 21:26:25.166897: step: 1252/1541, loss: 0.00024387352459598333 2023-01-23 21:26:26.084643: step: 1254/1541, loss: 0.0027331237215548754 2023-01-23 21:26:26.956957: step: 1256/1541, loss: 0.2812616527080536 2023-01-23 21:26:27.906233: step: 1258/1541, loss: 0.2824479639530182 2023-01-23 21:26:28.815653: step: 1260/1541, loss: 0.04735898599028587 2023-01-23 21:26:29.727462: step: 1262/1541, loss: 0.0005647212965413928 2023-01-23 21:26:30.592255: step: 1264/1541, loss: 0.0002675054711289704 2023-01-23 21:26:31.456479: step: 1266/1541, loss: 0.0027897306717932224 2023-01-23 21:26:32.392986: step: 1268/1541, loss: 0.008287548087537289 2023-01-23 21:26:33.283311: step: 1270/1541, loss: 0.932135820388794 2023-01-23 21:26:34.223547: step: 1272/1541, loss: 0.0038025570102036 2023-01-23 21:26:35.164851: step: 1274/1541, loss: 0.01805569976568222 2023-01-23 21:26:36.176957: step: 1276/1541, loss: 0.0005820983205921948 2023-01-23 21:26:37.086053: step: 1278/1541, loss: 0.0010718200355768204 2023-01-23 21:26:37.983270: step: 1280/1541, loss: 0.0011349550914019346 2023-01-23 21:26:38.865478: step: 1282/1541, loss: 0.032157059758901596 2023-01-23 21:26:39.769076: step: 1284/1541, loss: 0.0004716313269454986 2023-01-23 21:26:40.686422: step: 1286/1541, loss: 0.00027583082555793226 2023-01-23 21:26:41.610999: step: 1288/1541, loss: 4.9001770094037056e-05 2023-01-23 21:26:42.534655: step: 1290/1541, loss: 0.0013777300482615829 2023-01-23 21:26:43.422510: step: 1292/1541, loss: 0.0035025295801460743 2023-01-23 21:26:44.397784: step: 1294/1541, loss: 0.22610867023468018 2023-01-23 21:26:45.271680: step: 1296/1541, loss: 0.012262584641575813 2023-01-23 21:26:46.173490: step: 1298/1541, loss: 0.0002296433231094852 2023-01-23 21:26:47.021103: step: 1300/1541, loss: 0.07742518931627274 2023-01-23 21:26:47.917655: step: 1302/1541, loss: 0.008355306461453438 2023-01-23 21:26:48.813540: step: 1304/1541, loss: 0.0855138748884201 2023-01-23 21:26:49.699501: step: 1306/1541, loss: 0.1397312432527542 2023-01-23 21:26:50.569600: step: 1308/1541, loss: 0.000519772875122726 2023-01-23 21:26:51.440134: step: 1310/1541, loss: 0.0001695108221611008 2023-01-23 21:26:52.364038: step: 1312/1541, loss: 0.00019352883100509644 2023-01-23 21:26:53.264795: step: 1314/1541, loss: 0.0024765136186033487 2023-01-23 21:26:54.112928: step: 1316/1541, loss: 0.00012633568258024752 2023-01-23 21:26:55.030776: step: 1318/1541, loss: 8.479969255859032e-05 2023-01-23 21:26:55.890395: step: 1320/1541, loss: 0.4525011479854584 2023-01-23 21:26:56.821634: step: 1322/1541, loss: 0.0400700643658638 2023-01-23 21:26:57.721042: step: 1324/1541, loss: 0.0003705956623889506 2023-01-23 21:26:58.634383: step: 1326/1541, loss: 0.0002649058878887445 2023-01-23 21:26:59.531204: step: 1328/1541, loss: 0.12283418327569962 2023-01-23 21:27:00.404913: step: 1330/1541, loss: 0.04326731711626053 2023-01-23 21:27:01.354155: step: 1332/1541, loss: 0.3629518449306488 2023-01-23 21:27:02.227025: step: 1334/1541, loss: 0.0001919014030136168 2023-01-23 21:27:03.203542: step: 1336/1541, loss: 0.0007827593362890184 2023-01-23 21:27:04.196391: step: 1338/1541, loss: 0.008031262084841728 2023-01-23 21:27:05.059696: step: 1340/1541, loss: 0.0003960242902394384 2023-01-23 21:27:06.021033: step: 1342/1541, loss: 0.00022399009321816266 2023-01-23 21:27:06.924623: step: 1344/1541, loss: 0.1467849463224411 2023-01-23 21:27:07.774137: step: 1346/1541, loss: 0.0002135884715244174 2023-01-23 21:27:08.657223: step: 1348/1541, loss: 0.021592536941170692 2023-01-23 21:27:09.547640: step: 1350/1541, loss: 0.0005978046683594584 2023-01-23 21:27:10.500921: step: 1352/1541, loss: 0.022705374285578728 2023-01-23 21:27:11.410284: step: 1354/1541, loss: 0.0651211142539978 2023-01-23 21:27:12.300773: step: 1356/1541, loss: 0.0007468494586646557 2023-01-23 21:27:13.213641: step: 1358/1541, loss: 0.003746546572074294 2023-01-23 21:27:14.109605: step: 1360/1541, loss: 0.08017078042030334 2023-01-23 21:27:14.971827: step: 1362/1541, loss: 0.00336240092292428 2023-01-23 21:27:15.889801: step: 1364/1541, loss: 0.0007511325529776514 2023-01-23 21:27:16.877487: step: 1366/1541, loss: 0.003285149345174432 2023-01-23 21:27:17.749328: step: 1368/1541, loss: 0.00047823405475355685 2023-01-23 21:27:18.639274: step: 1370/1541, loss: 0.008987884037196636 2023-01-23 21:27:19.523295: step: 1372/1541, loss: 0.2713070213794708 2023-01-23 21:27:20.429954: step: 1374/1541, loss: 0.047036491334438324 2023-01-23 21:27:21.351773: step: 1376/1541, loss: 0.06434527039527893 2023-01-23 21:27:22.263581: step: 1378/1541, loss: 0.30745676159858704 2023-01-23 21:27:23.156056: step: 1380/1541, loss: 0.25076186656951904 2023-01-23 21:27:24.081709: step: 1382/1541, loss: 0.015329199843108654 2023-01-23 21:27:25.015463: step: 1384/1541, loss: 4.8877082008402795e-05 2023-01-23 21:27:25.909779: step: 1386/1541, loss: 0.13766704499721527 2023-01-23 21:27:26.810519: step: 1388/1541, loss: 0.009541662409901619 2023-01-23 21:27:27.723605: step: 1390/1541, loss: 0.00032504729460924864 2023-01-23 21:27:28.616532: step: 1392/1541, loss: 0.0017496384680271149 2023-01-23 21:27:29.526642: step: 1394/1541, loss: 0.013480208814144135 2023-01-23 21:27:30.433942: step: 1396/1541, loss: 0.00047584023559466004 2023-01-23 21:27:31.336207: step: 1398/1541, loss: 0.14214152097702026 2023-01-23 21:27:32.247421: step: 1400/1541, loss: 0.0014177645789459348 2023-01-23 21:27:33.110857: step: 1402/1541, loss: 8.423230610787868e-05 2023-01-23 21:27:34.018660: step: 1404/1541, loss: 0.07026927918195724 2023-01-23 21:27:34.920310: step: 1406/1541, loss: 0.04148098826408386 2023-01-23 21:27:35.842115: step: 1408/1541, loss: 0.0001626332086743787 2023-01-23 21:27:36.742562: step: 1410/1541, loss: 0.10536157339811325 2023-01-23 21:27:37.642101: step: 1412/1541, loss: 0.0003379848785698414 2023-01-23 21:27:38.486499: step: 1414/1541, loss: 0.0004062287334818393 2023-01-23 21:27:39.413419: step: 1416/1541, loss: 0.0031001491006463766 2023-01-23 21:27:40.322935: step: 1418/1541, loss: 0.01370322611182928 2023-01-23 21:27:41.278058: step: 1420/1541, loss: 0.0001165812136605382 2023-01-23 21:27:42.206159: step: 1422/1541, loss: 0.015091982670128345 2023-01-23 21:27:43.083329: step: 1424/1541, loss: 0.011806885711848736 2023-01-23 21:27:44.013047: step: 1426/1541, loss: 0.000884472276084125 2023-01-23 21:27:44.962886: step: 1428/1541, loss: 0.031147053465247154 2023-01-23 21:27:45.860755: step: 1430/1541, loss: 0.07211069017648697 2023-01-23 21:27:46.825786: step: 1432/1541, loss: 0.003021837444975972 2023-01-23 21:27:47.683420: step: 1434/1541, loss: 0.00023999753466341645 2023-01-23 21:27:48.539537: step: 1436/1541, loss: 0.0002521495916880667 2023-01-23 21:27:49.445649: step: 1438/1541, loss: 0.002832709578797221 2023-01-23 21:27:50.314060: step: 1440/1541, loss: 0.048650436103343964 2023-01-23 21:27:51.216528: step: 1442/1541, loss: 0.0001587013539392501 2023-01-23 21:27:52.045827: step: 1444/1541, loss: 0.001792142866179347 2023-01-23 21:27:52.911918: step: 1446/1541, loss: 3.680308145703748e-05 2023-01-23 21:27:53.789361: step: 1448/1541, loss: 6.797090463805944e-05 2023-01-23 21:27:54.693784: step: 1450/1541, loss: 0.03820057213306427 2023-01-23 21:27:55.580147: step: 1452/1541, loss: 7.612762419739738e-05 2023-01-23 21:27:56.471621: step: 1454/1541, loss: 0.00018476390687283128 2023-01-23 21:27:57.423334: step: 1456/1541, loss: 0.00020030677842441946 2023-01-23 21:27:58.374340: step: 1458/1541, loss: 0.00017818951164372265 2023-01-23 21:27:59.252530: step: 1460/1541, loss: 4.363618427305482e-05 2023-01-23 21:28:00.182515: step: 1462/1541, loss: 7.559209916507825e-05 2023-01-23 21:28:01.048762: step: 1464/1541, loss: 0.0001423346548108384 2023-01-23 21:28:01.940472: step: 1466/1541, loss: 0.05947250872850418 2023-01-23 21:28:02.812993: step: 1468/1541, loss: 0.0016814202535897493 2023-01-23 21:28:03.705563: step: 1470/1541, loss: 0.04241000860929489 2023-01-23 21:28:04.599574: step: 1472/1541, loss: 0.0021031824871897697 2023-01-23 21:28:05.509230: step: 1474/1541, loss: 0.0003456751292105764 2023-01-23 21:28:06.369632: step: 1476/1541, loss: 0.00025778679992072284 2023-01-23 21:28:07.221640: step: 1478/1541, loss: 0.003201268846169114 2023-01-23 21:28:08.118767: step: 1480/1541, loss: 0.0027152474503964186 2023-01-23 21:28:09.070622: step: 1482/1541, loss: 0.0004174957284703851 2023-01-23 21:28:09.955973: step: 1484/1541, loss: 0.011631147935986519 2023-01-23 21:28:10.831043: step: 1486/1541, loss: 0.07058191299438477 2023-01-23 21:28:11.678366: step: 1488/1541, loss: 4.2527895857347175e-05 2023-01-23 21:28:12.573824: step: 1490/1541, loss: 0.04117574915289879 2023-01-23 21:28:13.485290: step: 1492/1541, loss: 0.00015736036584712565 2023-01-23 21:28:14.350336: step: 1494/1541, loss: 0.21917815506458282 2023-01-23 21:28:15.256488: step: 1496/1541, loss: 0.00012878054985776544 2023-01-23 21:28:16.204064: step: 1498/1541, loss: 0.000405122380470857 2023-01-23 21:28:17.119440: step: 1500/1541, loss: 0.012696499936282635 2023-01-23 21:28:18.010797: step: 1502/1541, loss: 6.447264604503289e-05 2023-01-23 21:28:18.879348: step: 1504/1541, loss: 0.0004185407015029341 2023-01-23 21:28:19.784467: step: 1506/1541, loss: 0.010921081528067589 2023-01-23 21:28:20.667392: step: 1508/1541, loss: 0.0004143388650845736 2023-01-23 21:28:21.638590: step: 1510/1541, loss: 0.0006466962513513863 2023-01-23 21:28:22.569422: step: 1512/1541, loss: 0.017477771267294884 2023-01-23 21:28:23.439034: step: 1514/1541, loss: 4.457636896404438e-05 2023-01-23 21:28:24.352055: step: 1516/1541, loss: 1.8970640667248517e-05 2023-01-23 21:28:25.268773: step: 1518/1541, loss: 0.06442176550626755 2023-01-23 21:28:26.154377: step: 1520/1541, loss: 0.7542723417282104 2023-01-23 21:28:27.040002: step: 1522/1541, loss: 8.972003706730902e-05 2023-01-23 21:28:27.909656: step: 1524/1541, loss: 0.012644385918974876 2023-01-23 21:28:28.827462: step: 1526/1541, loss: 4.774719491251744e-05 2023-01-23 21:28:29.697741: step: 1528/1541, loss: 0.0001218393153976649 2023-01-23 21:28:30.583914: step: 1530/1541, loss: 0.008241482079029083 2023-01-23 21:28:31.478530: step: 1532/1541, loss: -3.575415030354634e-05 2023-01-23 21:28:32.344852: step: 1534/1541, loss: 0.20866650342941284 2023-01-23 21:28:33.262085: step: 1536/1541, loss: 0.0001745297631714493 2023-01-23 21:28:34.147374: step: 1538/1541, loss: 0.00017927715089172125 2023-01-23 21:28:35.023464: step: 1540/1541, loss: 0.00695639057084918 2023-01-23 21:28:35.942200: step: 1542/1541, loss: 0.0005346855032257736 2023-01-23 21:28:36.802961: step: 1544/1541, loss: 0.007283213548362255 2023-01-23 21:28:37.706309: step: 1546/1541, loss: 0.0038199503906071186 2023-01-23 21:28:38.672190: step: 1548/1541, loss: 7.553248724434525e-05 2023-01-23 21:28:39.565472: step: 1550/1541, loss: 1.7561013009981252e-05 2023-01-23 21:28:40.433614: step: 1552/1541, loss: 0.0004943712847307324 2023-01-23 21:28:41.323242: step: 1554/1541, loss: 0.10909909754991531 2023-01-23 21:28:42.240779: step: 1556/1541, loss: 0.03003242425620556 2023-01-23 21:28:43.126876: step: 1558/1541, loss: 0.0005493597127497196 2023-01-23 21:28:44.004390: step: 1560/1541, loss: 0.1679629534482956 2023-01-23 21:28:44.885050: step: 1562/1541, loss: 0.10021799802780151 2023-01-23 21:28:45.816450: step: 1564/1541, loss: 0.06662347912788391 2023-01-23 21:28:46.705233: step: 1566/1541, loss: -3.354873842909001e-05 2023-01-23 21:28:47.601127: step: 1568/1541, loss: 0.003989907447248697 2023-01-23 21:28:48.453496: step: 1570/1541, loss: 9.863378363661468e-05 2023-01-23 21:28:49.305511: step: 1572/1541, loss: 0.005323279183357954 2023-01-23 21:28:50.219860: step: 1574/1541, loss: 1.4881044990033843e-05 2023-01-23 21:28:51.076134: step: 1576/1541, loss: 0.0044865417294204235 2023-01-23 21:28:51.932897: step: 1578/1541, loss: 0.001350673963315785 2023-01-23 21:28:52.846340: step: 1580/1541, loss: 0.06621578335762024 2023-01-23 21:28:53.791232: step: 1582/1541, loss: 0.08236528933048248 2023-01-23 21:28:54.641825: step: 1584/1541, loss: 9.192394645651802e-05 2023-01-23 21:28:55.570351: step: 1586/1541, loss: 0.003976586274802685 2023-01-23 21:28:56.489528: step: 1588/1541, loss: 0.008749638684093952 2023-01-23 21:28:57.387815: step: 1590/1541, loss: 0.033332981169223785 2023-01-23 21:28:58.269908: step: 1592/1541, loss: 3.345579170854762e-05 2023-01-23 21:28:59.182451: step: 1594/1541, loss: 0.0022853920236229897 2023-01-23 21:29:00.066907: step: 1596/1541, loss: 0.0003574399743229151 2023-01-23 21:29:00.953148: step: 1598/1541, loss: 0.0002447670267429203 2023-01-23 21:29:01.816195: step: 1600/1541, loss: 0.00013515487080439925 2023-01-23 21:29:02.672067: step: 1602/1541, loss: 0.0007805429049767554 2023-01-23 21:29:03.571313: step: 1604/1541, loss: 0.31007036566734314 2023-01-23 21:29:04.451589: step: 1606/1541, loss: 0.0006153195863589644 2023-01-23 21:29:05.336731: step: 1608/1541, loss: 0.03655456006526947 2023-01-23 21:29:06.272887: step: 1610/1541, loss: 0.0014608876081183553 2023-01-23 21:29:07.223852: step: 1612/1541, loss: 0.00013752595987170935 2023-01-23 21:29:08.052717: step: 1614/1541, loss: 5.19698514835909e-05 2023-01-23 21:29:08.980148: step: 1616/1541, loss: 0.0015192346181720495 2023-01-23 21:29:09.869219: step: 1618/1541, loss: 0.17047131061553955 2023-01-23 21:29:10.761154: step: 1620/1541, loss: 0.00011415958579163998 2023-01-23 21:29:11.643289: step: 1622/1541, loss: 0.0024147839285433292 2023-01-23 21:29:12.525609: step: 1624/1541, loss: 0.00041836334276013076 2023-01-23 21:29:13.371212: step: 1626/1541, loss: 0.00014713052951265126 2023-01-23 21:29:14.257936: step: 1628/1541, loss: 0.00022453695419244468 2023-01-23 21:29:15.194489: step: 1630/1541, loss: 0.031842831522226334 2023-01-23 21:29:16.051794: step: 1632/1541, loss: 5.4722841014154255e-05 2023-01-23 21:29:17.002472: step: 1634/1541, loss: 0.008253749459981918 2023-01-23 21:29:17.902217: step: 1636/1541, loss: 0.012264265678822994 2023-01-23 21:29:18.825850: step: 1638/1541, loss: 0.007796508725732565 2023-01-23 21:29:19.772994: step: 1640/1541, loss: 0.17838850617408752 2023-01-23 21:29:20.657864: step: 1642/1541, loss: 0.004120353609323502 2023-01-23 21:29:21.538572: step: 1644/1541, loss: 0.0030964561738073826 2023-01-23 21:29:22.376050: step: 1646/1541, loss: 0.0001612761989235878 2023-01-23 21:29:23.291662: step: 1648/1541, loss: 0.01178842131048441 2023-01-23 21:29:24.155179: step: 1650/1541, loss: 0.0005356431938707829 2023-01-23 21:29:25.047098: step: 1652/1541, loss: 0.0002788972924463451 2023-01-23 21:29:25.918455: step: 1654/1541, loss: 0.01978183351457119 2023-01-23 21:29:26.814047: step: 1656/1541, loss: 0.026201043277978897 2023-01-23 21:29:27.715168: step: 1658/1541, loss: 0.22230488061904907 2023-01-23 21:29:28.592131: step: 1660/1541, loss: 0.0026224751491099596 2023-01-23 21:29:29.503711: step: 1662/1541, loss: 0.00033561186864972115 2023-01-23 21:29:30.429823: step: 1664/1541, loss: 0.00026894762413576245 2023-01-23 21:29:31.339444: step: 1666/1541, loss: 0.02768559195101261 2023-01-23 21:29:32.254747: step: 1668/1541, loss: 0.006666642613708973 2023-01-23 21:29:33.139746: step: 1670/1541, loss: 0.012435897253453732 2023-01-23 21:29:34.032600: step: 1672/1541, loss: 0.003119496861472726 2023-01-23 21:29:34.918042: step: 1674/1541, loss: 0.0037532788701355457 2023-01-23 21:29:35.801889: step: 1676/1541, loss: 0.0038482267409563065 2023-01-23 21:29:36.690530: step: 1678/1541, loss: 0.0016710154013708234 2023-01-23 21:29:37.561920: step: 1680/1541, loss: 0.0012146540684625506 2023-01-23 21:29:38.450625: step: 1682/1541, loss: 1.7382426449330524e-06 2023-01-23 21:29:39.325891: step: 1684/1541, loss: 0.0003492433752398938 2023-01-23 21:29:40.200549: step: 1686/1541, loss: 0.05684801563620567 2023-01-23 21:29:41.102493: step: 1688/1541, loss: 0.03925158828496933 2023-01-23 21:29:42.041428: step: 1690/1541, loss: 0.03968118131160736 2023-01-23 21:29:42.919801: step: 1692/1541, loss: 0.0009609893895685673 2023-01-23 21:29:43.796162: step: 1694/1541, loss: 0.004059463273733854 2023-01-23 21:29:44.679313: step: 1696/1541, loss: -9.292363756685518e-06 2023-01-23 21:29:45.583484: step: 1698/1541, loss: 0.007250759284943342 2023-01-23 21:29:46.469775: step: 1700/1541, loss: -1.966172749234829e-05 2023-01-23 21:29:47.419462: step: 1702/1541, loss: 4.1650979255791754e-05 2023-01-23 21:29:48.285160: step: 1704/1541, loss: 0.04248516261577606 2023-01-23 21:29:49.137915: step: 1706/1541, loss: 5.470165888254996e-06 2023-01-23 21:29:49.996786: step: 1708/1541, loss: 2.919435610237997e-05 2023-01-23 21:29:50.906801: step: 1710/1541, loss: 0.0001809176173992455 2023-01-23 21:29:51.892763: step: 1712/1541, loss: 0.0019072602735832334 2023-01-23 21:29:52.833440: step: 1714/1541, loss: 0.20159250497817993 2023-01-23 21:29:53.716585: step: 1716/1541, loss: 0.001356662018224597 2023-01-23 21:29:54.664055: step: 1718/1541, loss: -1.4010035556566436e-05 2023-01-23 21:29:55.562632: step: 1720/1541, loss: 0.12246017158031464 2023-01-23 21:29:56.462428: step: 1722/1541, loss: 0.00028084678342565894 2023-01-23 21:29:57.397610: step: 1724/1541, loss: 0.06895890086889267 2023-01-23 21:29:58.267630: step: 1726/1541, loss: 0.18319018185138702 2023-01-23 21:29:59.141764: step: 1728/1541, loss: 0.00032119371462613344 2023-01-23 21:30:00.021745: step: 1730/1541, loss: 0.0009989315876737237 2023-01-23 21:30:00.907992: step: 1732/1541, loss: 0.0254492349922657 2023-01-23 21:30:01.870245: step: 1734/1541, loss: 0.001157125225290656 2023-01-23 21:30:02.749775: step: 1736/1541, loss: -3.43281717505306e-05 2023-01-23 21:30:03.669752: step: 1738/1541, loss: 0.00029354545404203236 2023-01-23 21:30:04.549378: step: 1740/1541, loss: 1.878843249869533e-05 2023-01-23 21:30:05.407875: step: 1742/1541, loss: 0.00010052020661532879 2023-01-23 21:30:06.305867: step: 1744/1541, loss: 0.006445671897381544 2023-01-23 21:30:07.258501: step: 1746/1541, loss: -1.7462607502238825e-06 2023-01-23 21:30:08.135605: step: 1748/1541, loss: 0.008139831945300102 2023-01-23 21:30:09.011894: step: 1750/1541, loss: 0.33189383149147034 2023-01-23 21:30:09.896299: step: 1752/1541, loss: 8.93932519829832e-05 2023-01-23 21:30:10.867346: step: 1754/1541, loss: 0.014227354899048805 2023-01-23 21:30:11.761525: step: 1756/1541, loss: 0.15807193517684937 2023-01-23 21:30:12.642550: step: 1758/1541, loss: 0.0033966773189604282 2023-01-23 21:30:13.481864: step: 1760/1541, loss: 2.6817619072971866e-05 2023-01-23 21:30:14.330953: step: 1762/1541, loss: 0.007872596383094788 2023-01-23 21:30:15.225543: step: 1764/1541, loss: 0.00016438103921245784 2023-01-23 21:30:16.220997: step: 1766/1541, loss: 7.628151070093736e-05 2023-01-23 21:30:17.104092: step: 1768/1541, loss: 0.000371795438695699 2023-01-23 21:30:17.963404: step: 1770/1541, loss: 5.396604319685139e-05 2023-01-23 21:30:18.852937: step: 1772/1541, loss: -0.00010865689546335489 2023-01-23 21:30:19.722798: step: 1774/1541, loss: 0.001015148009173572 2023-01-23 21:30:20.637099: step: 1776/1541, loss: 8.448240987490863e-05 2023-01-23 21:30:21.499446: step: 1778/1541, loss: 7.129987352527678e-05 2023-01-23 21:30:22.362624: step: 1780/1541, loss: 0.5511035919189453 2023-01-23 21:30:23.231493: step: 1782/1541, loss: 0.01220639143139124 2023-01-23 21:30:24.122710: step: 1784/1541, loss: 8.635664562461898e-05 2023-01-23 21:30:25.011526: step: 1786/1541, loss: 0.329463928937912 2023-01-23 21:30:25.874455: step: 1788/1541, loss: 0.10636680573225021 2023-01-23 21:30:26.730427: step: 1790/1541, loss: 0.0003288662701379508 2023-01-23 21:30:27.654211: step: 1792/1541, loss: -5.8053587963513564e-06 2023-01-23 21:30:28.526130: step: 1794/1541, loss: 0.18297313153743744 2023-01-23 21:30:29.417845: step: 1796/1541, loss: 0.08625808358192444 2023-01-23 21:30:30.271739: step: 1798/1541, loss: 0.1049361303448677 2023-01-23 21:30:31.156732: step: 1800/1541, loss: 0.003073473460972309 2023-01-23 21:30:32.143375: step: 1802/1541, loss: 0.009993337094783783 2023-01-23 21:30:33.033756: step: 1804/1541, loss: 0.00026127122691832483 2023-01-23 21:30:33.901615: step: 1806/1541, loss: 6.336424121400341e-05 2023-01-23 21:30:34.802957: step: 1808/1541, loss: 0.5757269859313965 2023-01-23 21:30:35.744045: step: 1810/1541, loss: 0.00977927353233099 2023-01-23 21:30:36.682334: step: 1812/1541, loss: 0.00011013258335879073 2023-01-23 21:30:37.574580: step: 1814/1541, loss: 0.06446537375450134 2023-01-23 21:30:38.476307: step: 1816/1541, loss: 0.0001365040661767125 2023-01-23 21:30:39.334426: step: 1818/1541, loss: 0.003961888607591391 2023-01-23 21:30:40.267298: step: 1820/1541, loss: 7.373337575700134e-05 2023-01-23 21:30:41.174873: step: 1822/1541, loss: 0.0015202990034595132 2023-01-23 21:30:42.048492: step: 1824/1541, loss: 0.09319940209388733 2023-01-23 21:30:42.926103: step: 1826/1541, loss: 0.10663436353206635 2023-01-23 21:30:43.805784: step: 1828/1541, loss: 0.07333007454872131 2023-01-23 21:30:44.815931: step: 1830/1541, loss: 7.486685353796929e-05 2023-01-23 21:30:45.691171: step: 1832/1541, loss: 0.027881020680069923 2023-01-23 21:30:46.627876: step: 1834/1541, loss: 0.049075089395046234 2023-01-23 21:30:47.514136: step: 1836/1541, loss: 0.005425292998552322 2023-01-23 21:30:48.388270: step: 1838/1541, loss: 0.001010849024169147 2023-01-23 21:30:49.243462: step: 1840/1541, loss: 0.00023874171893112361 2023-01-23 21:30:50.172282: step: 1842/1541, loss: 0.00011838166392408311 2023-01-23 21:30:51.173918: step: 1844/1541, loss: 0.575101375579834 2023-01-23 21:30:52.070523: step: 1846/1541, loss: 0.0002995878749061376 2023-01-23 21:30:52.952760: step: 1848/1541, loss: 3.447675408096984e-05 2023-01-23 21:30:53.845137: step: 1850/1541, loss: 0.019702445715665817 2023-01-23 21:30:54.734605: step: 1852/1541, loss: 0.0015578496968373656 2023-01-23 21:30:55.634278: step: 1854/1541, loss: 0.0446198396384716 2023-01-23 21:30:56.553735: step: 1856/1541, loss: 0.010713919997215271 2023-01-23 21:30:57.415071: step: 1858/1541, loss: 0.00024492197553627193 2023-01-23 21:30:58.345349: step: 1860/1541, loss: 0.016832606866955757 2023-01-23 21:30:59.241306: step: 1862/1541, loss: 0.012714210897684097 2023-01-23 21:31:00.145148: step: 1864/1541, loss: 0.001747169066220522 2023-01-23 21:31:01.044711: step: 1866/1541, loss: 0.0028737112879753113 2023-01-23 21:31:01.947353: step: 1868/1541, loss: 0.00797225907444954 2023-01-23 21:31:02.865916: step: 1870/1541, loss: 0.00040159773197956383 2023-01-23 21:31:03.751735: step: 1872/1541, loss: 0.0012908843345940113 2023-01-23 21:31:04.688517: step: 1874/1541, loss: 0.05065518990159035 2023-01-23 21:31:05.547111: step: 1876/1541, loss: 0.005039627198129892 2023-01-23 21:31:06.494859: step: 1878/1541, loss: 0.4860290288925171 2023-01-23 21:31:07.408159: step: 1880/1541, loss: 0.2858533263206482 2023-01-23 21:31:08.311798: step: 1882/1541, loss: 0.00015125380014069378 2023-01-23 21:31:09.192786: step: 1884/1541, loss: 0.042042076587677 2023-01-23 21:31:10.097978: step: 1886/1541, loss: 6.11922368989326e-05 2023-01-23 21:31:10.993889: step: 1888/1541, loss: -1.0255166671413463e-06 2023-01-23 21:31:11.877395: step: 1890/1541, loss: 0.0001533815375296399 2023-01-23 21:31:12.738577: step: 1892/1541, loss: 0.0005531618371605873 2023-01-23 21:31:13.653436: step: 1894/1541, loss: 0.05023377016186714 2023-01-23 21:31:14.533273: step: 1896/1541, loss: 3.404527888051234e-05 2023-01-23 21:31:15.479191: step: 1898/1541, loss: 0.005273704882711172 2023-01-23 21:31:16.345775: step: 1900/1541, loss: 3.4804739698302e-05 2023-01-23 21:31:17.273129: step: 1902/1541, loss: 0.0005336771137081087 2023-01-23 21:31:18.139290: step: 1904/1541, loss: 0.05822675675153732 2023-01-23 21:31:19.033394: step: 1906/1541, loss: -6.417572876671329e-05 2023-01-23 21:31:19.929678: step: 1908/1541, loss: 0.14640489220619202 2023-01-23 21:31:20.807937: step: 1910/1541, loss: 0.01386748906224966 2023-01-23 21:31:21.716949: step: 1912/1541, loss: 0.0016802221070975065 2023-01-23 21:31:22.707656: step: 1914/1541, loss: 0.000172849468071945 2023-01-23 21:31:23.629179: step: 1916/1541, loss: 4.210024781059474e-05 2023-01-23 21:31:24.548060: step: 1918/1541, loss: 6.636110629187897e-05 2023-01-23 21:31:25.417313: step: 1920/1541, loss: 0.07194633036851883 2023-01-23 21:31:26.319362: step: 1922/1541, loss: 0.010541846975684166 2023-01-23 21:31:27.236702: step: 1924/1541, loss: 0.021223850548267365 2023-01-23 21:31:28.133703: step: 1926/1541, loss: 0.0006843170849606395 2023-01-23 21:31:29.069165: step: 1928/1541, loss: -1.744164364936296e-05 2023-01-23 21:31:29.955104: step: 1930/1541, loss: 0.3190734386444092 2023-01-23 21:31:30.883056: step: 1932/1541, loss: 2.831369056366384e-05 2023-01-23 21:31:31.792826: step: 1934/1541, loss: 1.0989251677528955e-05 2023-01-23 21:31:32.734609: step: 1936/1541, loss: 0.0006658956408500671 2023-01-23 21:31:33.600472: step: 1938/1541, loss: 1.4641121197200846e-05 2023-01-23 21:31:34.531929: step: 1940/1541, loss: 0.001942059607245028 2023-01-23 21:31:35.442432: step: 1942/1541, loss: 0.04148418828845024 2023-01-23 21:31:36.409068: step: 1944/1541, loss: 8.11816644272767e-05 2023-01-23 21:31:37.254470: step: 1946/1541, loss: -7.036110218905378e-06 2023-01-23 21:31:38.165494: step: 1948/1541, loss: 0.00014208017091732472 2023-01-23 21:31:39.041093: step: 1950/1541, loss: 5.053087807027623e-05 2023-01-23 21:31:39.940939: step: 1952/1541, loss: 0.34477582573890686 2023-01-23 21:31:40.862823: step: 1954/1541, loss: 0.0001697055558906868 2023-01-23 21:31:41.742612: step: 1956/1541, loss: 0.19331401586532593 2023-01-23 21:31:42.616093: step: 1958/1541, loss: 0.13662762939929962 2023-01-23 21:31:43.470138: step: 1960/1541, loss: 0.0002826668496709317 2023-01-23 21:31:44.366361: step: 1962/1541, loss: 1.6207352018682286e-05 2023-01-23 21:31:45.245964: step: 1964/1541, loss: 0.00040672282921150327 2023-01-23 21:31:46.212500: step: 1966/1541, loss: 0.0006894769612699747 2023-01-23 21:31:47.112368: step: 1968/1541, loss: 0.005932700354605913 2023-01-23 21:31:48.009354: step: 1970/1541, loss: 0.000677387579344213 2023-01-23 21:31:48.946663: step: 1972/1541, loss: 0.0003685482370201498 2023-01-23 21:31:49.881439: step: 1974/1541, loss: 0.010946284048259258 2023-01-23 21:31:50.791492: step: 1976/1541, loss: 0.05049535632133484 2023-01-23 21:31:51.663309: step: 1978/1541, loss: 0.00028785475296899676 2023-01-23 21:31:52.584846: step: 1980/1541, loss: 0.013655713759362698 2023-01-23 21:31:53.448918: step: 1982/1541, loss: 0.00029402694781310856 2023-01-23 21:31:54.354870: step: 1984/1541, loss: 0.0031310797203332186 2023-01-23 21:31:55.246364: step: 1986/1541, loss: 0.00013999262591823936 2023-01-23 21:31:56.116287: step: 1988/1541, loss: 0.0020948590245097876 2023-01-23 21:31:57.011451: step: 1990/1541, loss: 0.09508789330720901 2023-01-23 21:31:57.877037: step: 1992/1541, loss: 0.008482463657855988 2023-01-23 21:31:58.839622: step: 1994/1541, loss: 0.23152245581150055 2023-01-23 21:31:59.714323: step: 1996/1541, loss: 6.501732423203066e-05 2023-01-23 21:32:00.565296: step: 1998/1541, loss: 0.0015913612442091107 2023-01-23 21:32:01.422355: step: 2000/1541, loss: 0.00028549745911732316 2023-01-23 21:32:02.338028: step: 2002/1541, loss: 0.0005991846555843949 2023-01-23 21:32:03.202420: step: 2004/1541, loss: 0.154060959815979 2023-01-23 21:32:04.096245: step: 2006/1541, loss: 0.010531683452427387 2023-01-23 21:32:04.997722: step: 2008/1541, loss: 0.08333012461662292 2023-01-23 21:32:05.864540: step: 2010/1541, loss: 0.00022618785442318767 2023-01-23 21:32:06.724094: step: 2012/1541, loss: 0.2345806062221527 2023-01-23 21:32:07.606770: step: 2014/1541, loss: 0.0034199648071080446 2023-01-23 21:32:08.533543: step: 2016/1541, loss: 0.00018662349611986428 2023-01-23 21:32:09.417456: step: 2018/1541, loss: 6.79055810905993e-05 2023-01-23 21:32:10.291385: step: 2020/1541, loss: 5.5236574553418905e-05 2023-01-23 21:32:11.191745: step: 2022/1541, loss: 0.012106459587812424 2023-01-23 21:32:12.057426: step: 2024/1541, loss: 0.00033633524435572326 2023-01-23 21:32:12.960637: step: 2026/1541, loss: 0.3650614321231842 2023-01-23 21:32:13.872295: step: 2028/1541, loss: 0.00043070089304819703 2023-01-23 21:32:14.791926: step: 2030/1541, loss: 0.0016590147279202938 2023-01-23 21:32:15.759834: step: 2032/1541, loss: 0.0027970110531896353 2023-01-23 21:32:16.713160: step: 2034/1541, loss: 0.011505176313221455 2023-01-23 21:32:17.618221: step: 2036/1541, loss: 0.0004563195980153978 2023-01-23 21:32:18.534670: step: 2038/1541, loss: 0.18113180994987488 2023-01-23 21:32:19.415199: step: 2040/1541, loss: 0.001565420301631093 2023-01-23 21:32:20.316848: step: 2042/1541, loss: 0.00040431160596199334 2023-01-23 21:32:21.199630: step: 2044/1541, loss: 0.027102677151560783 2023-01-23 21:32:22.148280: step: 2046/1541, loss: 0.17383795976638794 2023-01-23 21:32:23.048313: step: 2048/1541, loss: 0.0008397132041864097 2023-01-23 21:32:23.909835: step: 2050/1541, loss: 0.004183779004961252 2023-01-23 21:32:24.874648: step: 2052/1541, loss: 0.14496690034866333 2023-01-23 21:32:25.762174: step: 2054/1541, loss: 0.0014075624058023095 2023-01-23 21:32:26.664197: step: 2056/1541, loss: 0.0002783517411444336 2023-01-23 21:32:27.630934: step: 2058/1541, loss: 0.019855402410030365 2023-01-23 21:32:28.514778: step: 2060/1541, loss: 0.000327343208482489 2023-01-23 21:32:29.404316: step: 2062/1541, loss: 3.5193839721614495e-05 2023-01-23 21:32:30.301406: step: 2064/1541, loss: 3.096546424785629e-05 2023-01-23 21:32:31.198840: step: 2066/1541, loss: 0.0002749978157225996 2023-01-23 21:32:32.131473: step: 2068/1541, loss: 0.000139183975988999 2023-01-23 21:32:33.020134: step: 2070/1541, loss: 0.0004832289705518633 2023-01-23 21:32:33.897102: step: 2072/1541, loss: 4.113871545996517e-05 2023-01-23 21:32:34.770676: step: 2074/1541, loss: 0.011235369369387627 2023-01-23 21:32:35.682672: step: 2076/1541, loss: 0.00011079210526077077 2023-01-23 21:32:36.587793: step: 2078/1541, loss: 0.11759558320045471 2023-01-23 21:32:37.458363: step: 2080/1541, loss: 0.0016416236758232117 2023-01-23 21:32:38.360747: step: 2082/1541, loss: 7.300028664758429e-05 2023-01-23 21:32:39.392887: step: 2084/1541, loss: 0.0001634755899431184 2023-01-23 21:32:40.309466: step: 2086/1541, loss: 0.0026284796185791492 2023-01-23 21:32:41.184394: step: 2088/1541, loss: 4.212110070511699e-05 2023-01-23 21:32:42.093509: step: 2090/1541, loss: 3.6543573514791206e-05 2023-01-23 21:32:42.939554: step: 2092/1541, loss: 3.515710704959929e-05 2023-01-23 21:32:43.805513: step: 2094/1541, loss: 0.0003951129619963467 2023-01-23 21:32:44.685866: step: 2096/1541, loss: 0.019030027091503143 2023-01-23 21:32:45.545253: step: 2098/1541, loss: 0.00011600057769101113 2023-01-23 21:32:46.551547: step: 2100/1541, loss: -2.9355303468037164e-06 2023-01-23 21:32:47.437104: step: 2102/1541, loss: 0.01325700432062149 2023-01-23 21:32:48.353850: step: 2104/1541, loss: 0.0033796874340623617 2023-01-23 21:32:49.249550: step: 2106/1541, loss: 0.04873927682638168 2023-01-23 21:32:50.153916: step: 2108/1541, loss: 1.876593887573108e-05 2023-01-23 21:32:51.044926: step: 2110/1541, loss: 0.021024467423558235 2023-01-23 21:32:51.921851: step: 2112/1541, loss: 0.024855926632881165 2023-01-23 21:32:52.888254: step: 2114/1541, loss: 0.44331860542297363 2023-01-23 21:32:53.762286: step: 2116/1541, loss: -4.6145892156346235e-06 2023-01-23 21:32:54.641944: step: 2118/1541, loss: 1.885292113001924e-05 2023-01-23 21:32:55.496259: step: 2120/1541, loss: 0.0002531126083340496 2023-01-23 21:32:56.433436: step: 2122/1541, loss: 0.051907725632190704 2023-01-23 21:32:57.291314: step: 2124/1541, loss: 0.0296243354678154 2023-01-23 21:32:58.237151: step: 2126/1541, loss: 8.946043817559257e-05 2023-01-23 21:32:59.117966: step: 2128/1541, loss: 0.11908098310232162 2023-01-23 21:33:00.043587: step: 2130/1541, loss: 0.00862432923167944 2023-01-23 21:33:00.894571: step: 2132/1541, loss: 0.0017877721693366766 2023-01-23 21:33:01.777962: step: 2134/1541, loss: 0.008777660317718983 2023-01-23 21:33:02.688852: step: 2136/1541, loss: 0.024020561948418617 2023-01-23 21:33:03.626303: step: 2138/1541, loss: 0.14292198419570923 2023-01-23 21:33:04.496492: step: 2140/1541, loss: 0.0019254365470260382 2023-01-23 21:33:05.413004: step: 2142/1541, loss: 0.16827857494354248 2023-01-23 21:33:06.379221: step: 2144/1541, loss: 6.443064212799072 2023-01-23 21:33:07.257827: step: 2146/1541, loss: 0.13045039772987366 2023-01-23 21:33:08.210768: step: 2148/1541, loss: 0.2846686840057373 2023-01-23 21:33:09.164652: step: 2150/1541, loss: 0.00025454183924011886 2023-01-23 21:33:10.084831: step: 2152/1541, loss: 0.0003731547622010112 2023-01-23 21:33:10.923360: step: 2154/1541, loss: 0.00019706334569491446 2023-01-23 21:33:11.883664: step: 2156/1541, loss: 0.04109329730272293 2023-01-23 21:33:12.775141: step: 2158/1541, loss: 0.00016209720342885703 2023-01-23 21:33:13.661606: step: 2160/1541, loss: 0.0002534060913603753 2023-01-23 21:33:14.552308: step: 2162/1541, loss: 0.1523936241865158 2023-01-23 21:33:15.454589: step: 2164/1541, loss: 0.2101590782403946 2023-01-23 21:33:16.316210: step: 2166/1541, loss: 1.4824667232460342e-05 2023-01-23 21:33:17.184964: step: 2168/1541, loss: 0.0002505763550288975 2023-01-23 21:33:18.124232: step: 2170/1541, loss: 0.007204446010291576 2023-01-23 21:33:19.028791: step: 2172/1541, loss: 6.792514614062384e-05 2023-01-23 21:33:19.962485: step: 2174/1541, loss: 0.00020901621610391885 2023-01-23 21:33:20.918787: step: 2176/1541, loss: 0.04078347980976105 2023-01-23 21:33:21.808507: step: 2178/1541, loss: 8.795820758678019e-05 2023-01-23 21:33:22.737064: step: 2180/1541, loss: 0.0011309570400044322 2023-01-23 21:33:23.630315: step: 2182/1541, loss: 5.182746463106014e-05 2023-01-23 21:33:24.565281: step: 2184/1541, loss: 0.1221766397356987 2023-01-23 21:33:25.458031: step: 2186/1541, loss: 4.358999649411999e-05 2023-01-23 21:33:26.331299: step: 2188/1541, loss: 0.19214710593223572 2023-01-23 21:33:27.202086: step: 2190/1541, loss: 0.00014964453293941915 2023-01-23 21:33:28.048219: step: 2192/1541, loss: 0.008058828301727772 2023-01-23 21:33:28.947230: step: 2194/1541, loss: 0.0026607646141201258 2023-01-23 21:33:29.817052: step: 2196/1541, loss: 0.00013586439308710396 2023-01-23 21:33:30.762294: step: 2198/1541, loss: 0.0003695919585879892 2023-01-23 21:33:31.644426: step: 2200/1541, loss: 0.0002343436935916543 2023-01-23 21:33:32.550412: step: 2202/1541, loss: 3.4293818316655234e-05 2023-01-23 21:33:33.423471: step: 2204/1541, loss: 0.0006734238704666495 2023-01-23 21:33:34.314660: step: 2206/1541, loss: 0.0013615803327411413 2023-01-23 21:33:35.245336: step: 2208/1541, loss: 0.3082558214664459 2023-01-23 21:33:36.122040: step: 2210/1541, loss: 0.1522759646177292 2023-01-23 21:33:36.978099: step: 2212/1541, loss: 0.0022660023532807827 2023-01-23 21:33:37.813036: step: 2214/1541, loss: 0.000507825636304915 2023-01-23 21:33:38.679780: step: 2216/1541, loss: 5.584567770711146e-05 2023-01-23 21:33:39.550518: step: 2218/1541, loss: 0.00010021313210017979 2023-01-23 21:33:40.433737: step: 2220/1541, loss: 0.07302463054656982 2023-01-23 21:33:41.319062: step: 2222/1541, loss: 0.3920030891895294 2023-01-23 21:33:42.187306: step: 2224/1541, loss: 0.06135549768805504 2023-01-23 21:33:43.047779: step: 2226/1541, loss: 3.1999865313991904e-05 2023-01-23 21:33:43.876019: step: 2228/1541, loss: 5.9341953601688147e-05 2023-01-23 21:33:44.781250: step: 2230/1541, loss: 0.00030180037720128894 2023-01-23 21:33:45.634964: step: 2232/1541, loss: 0.002416664734482765 2023-01-23 21:33:46.506492: step: 2234/1541, loss: 0.04568244516849518 2023-01-23 21:33:47.348991: step: 2236/1541, loss: 0.0006285231211222708 2023-01-23 21:33:48.278857: step: 2238/1541, loss: 0.044299058616161346 2023-01-23 21:33:49.239179: step: 2240/1541, loss: 1.3195134401321411 2023-01-23 21:33:50.101600: step: 2242/1541, loss: 0.09858008474111557 2023-01-23 21:33:51.011062: step: 2244/1541, loss: 0.00015213544247671962 2023-01-23 21:33:51.899920: step: 2246/1541, loss: 0.07531605660915375 2023-01-23 21:33:52.819892: step: 2248/1541, loss: 0.002178683876991272 2023-01-23 21:33:53.701025: step: 2250/1541, loss: 0.0659288838505745 2023-01-23 21:33:54.625146: step: 2252/1541, loss: 0.024449286982417107 2023-01-23 21:33:55.541677: step: 2254/1541, loss: 0.11137179285287857 2023-01-23 21:33:56.460103: step: 2256/1541, loss: 0.00038825912633910775 2023-01-23 21:33:57.388189: step: 2258/1541, loss: 0.005676186643540859 2023-01-23 21:33:58.235767: step: 2260/1541, loss: 1.4900431324349483e-06 2023-01-23 21:33:59.170344: step: 2262/1541, loss: 0.0004565284471027553 2023-01-23 21:34:00.066556: step: 2264/1541, loss: 0.00015396038361359388 2023-01-23 21:34:00.986238: step: 2266/1541, loss: 0.00013275841774884611 2023-01-23 21:34:01.938670: step: 2268/1541, loss: 0.00017468238365836442 2023-01-23 21:34:02.820413: step: 2270/1541, loss: 0.038932546973228455 2023-01-23 21:34:03.689289: step: 2272/1541, loss: 0.10003826022148132 2023-01-23 21:34:04.569829: step: 2274/1541, loss: 0.5194870233535767 2023-01-23 21:34:05.468584: step: 2276/1541, loss: 8.85583067429252e-05 2023-01-23 21:34:06.342238: step: 2278/1541, loss: 0.0005773405428044498 2023-01-23 21:34:07.277857: step: 2280/1541, loss: 0.011296368204057217 2023-01-23 21:34:08.183759: step: 2282/1541, loss: 0.03452250733971596 2023-01-23 21:34:09.068350: step: 2284/1541, loss: 0.00030670015257783234 2023-01-23 21:34:09.926373: step: 2286/1541, loss: 0.033060453832149506 2023-01-23 21:34:10.801645: step: 2288/1541, loss: 0.006935239303857088 2023-01-23 21:34:11.734094: step: 2290/1541, loss: 0.0068789939396083355 2023-01-23 21:34:12.651900: step: 2292/1541, loss: 0.2366517335176468 2023-01-23 21:34:13.545506: step: 2294/1541, loss: 0.00851903110742569 2023-01-23 21:34:14.442097: step: 2296/1541, loss: 0.27001598477363586 2023-01-23 21:34:15.303411: step: 2298/1541, loss: 0.1339539885520935 2023-01-23 21:34:16.232592: step: 2300/1541, loss: 0.0007147688884288073 2023-01-23 21:34:17.136765: step: 2302/1541, loss: 0.0953621193766594 2023-01-23 21:34:18.084000: step: 2304/1541, loss: 0.0002802204980980605 2023-01-23 21:34:18.950907: step: 2306/1541, loss: 0.000853167672175914 2023-01-23 21:34:19.818650: step: 2308/1541, loss: 0.0002152322413166985 2023-01-23 21:34:20.667613: step: 2310/1541, loss: 0.14859382808208466 2023-01-23 21:34:21.527858: step: 2312/1541, loss: 5.5766431614756584e-05 2023-01-23 21:34:22.446223: step: 2314/1541, loss: 0.06725046038627625 2023-01-23 21:34:23.352552: step: 2316/1541, loss: 0.11406757682561874 2023-01-23 21:34:24.235840: step: 2318/1541, loss: 0.004039696417748928 2023-01-23 21:34:25.130020: step: 2320/1541, loss: 0.00012118909216951579 2023-01-23 21:34:26.105613: step: 2322/1541, loss: 0.00010979605576721951 2023-01-23 21:34:27.009400: step: 2324/1541, loss: 0.00010138002835446969 2023-01-23 21:34:27.882599: step: 2326/1541, loss: 5.03358241985552e-05 2023-01-23 21:34:28.822647: step: 2328/1541, loss: 0.516661524772644 2023-01-23 21:34:29.694989: step: 2330/1541, loss: 0.0033972314558923244 2023-01-23 21:34:30.716197: step: 2332/1541, loss: 0.0006765917059965432 2023-01-23 21:34:31.693200: step: 2334/1541, loss: 0.0011509163305163383 2023-01-23 21:34:32.607469: step: 2336/1541, loss: 0.001998973311856389 2023-01-23 21:34:33.480205: step: 2338/1541, loss: 0.0017022609245032072 2023-01-23 21:34:34.462836: step: 2340/1541, loss: 0.0045914906077086926 2023-01-23 21:34:35.344317: step: 2342/1541, loss: 8.945930312620476e-05 2023-01-23 21:34:36.262746: step: 2344/1541, loss: 0.13172784447669983 2023-01-23 21:34:37.112256: step: 2346/1541, loss: 0.10239026695489883 2023-01-23 21:34:38.031655: step: 2348/1541, loss: 0.01717602089047432 2023-01-23 21:34:38.914404: step: 2350/1541, loss: 0.00017566277529112995 2023-01-23 21:34:39.784261: step: 2352/1541, loss: 0.0009906780906021595 2023-01-23 21:34:40.632167: step: 2354/1541, loss: 0.0004166853614151478 2023-01-23 21:34:41.485702: step: 2356/1541, loss: 0.0025412621907889843 2023-01-23 21:34:42.356258: step: 2358/1541, loss: 0.00012976714060641825 2023-01-23 21:34:43.258970: step: 2360/1541, loss: 0.0001665633317315951 2023-01-23 21:34:44.145980: step: 2362/1541, loss: 0.21288956701755524 2023-01-23 21:34:45.046606: step: 2364/1541, loss: 0.00040591572178527713 2023-01-23 21:34:45.924784: step: 2366/1541, loss: 0.0847228467464447 2023-01-23 21:34:46.842749: step: 2368/1541, loss: 8.711643749848008e-05 2023-01-23 21:34:47.715857: step: 2370/1541, loss: 0.00017174839740619063 2023-01-23 21:34:48.591351: step: 2372/1541, loss: 0.00559903122484684 2023-01-23 21:34:49.552993: step: 2374/1541, loss: 0.0014836612390354276 2023-01-23 21:34:50.434292: step: 2376/1541, loss: 0.00013840023893862963 2023-01-23 21:34:51.304666: step: 2378/1541, loss: 0.12278325110673904 2023-01-23 21:34:52.167416: step: 2380/1541, loss: 0.06449806690216064 2023-01-23 21:34:53.056773: step: 2382/1541, loss: 1.2169253750471398e-05 2023-01-23 21:34:53.946052: step: 2384/1541, loss: 0.018898705020546913 2023-01-23 21:34:54.848781: step: 2386/1541, loss: 0.008915882557630539 2023-01-23 21:34:55.756570: step: 2388/1541, loss: 3.564879443729296e-05 2023-01-23 21:34:56.701350: step: 2390/1541, loss: 0.01734384521842003 2023-01-23 21:34:57.597503: step: 2392/1541, loss: 5.09567980770953e-05 2023-01-23 21:34:58.491277: step: 2394/1541, loss: 0.002969298278912902 2023-01-23 21:34:59.353649: step: 2396/1541, loss: -5.901783879380673e-06 2023-01-23 21:35:00.244590: step: 2398/1541, loss: 0.013956175185739994 2023-01-23 21:35:01.147904: step: 2400/1541, loss: 0.039061617106199265 2023-01-23 21:35:02.018307: step: 2402/1541, loss: 0.003425567178055644 2023-01-23 21:35:02.990753: step: 2404/1541, loss: -9.60273428063374e-06 2023-01-23 21:35:03.909619: step: 2406/1541, loss: 3.0604066978412447e-06 2023-01-23 21:35:04.869774: step: 2408/1541, loss: 0.0006033199606463313 2023-01-23 21:35:05.754486: step: 2410/1541, loss: 0.0008025679271668196 2023-01-23 21:35:06.645333: step: 2412/1541, loss: 6.962940460653044e-06 2023-01-23 21:35:07.576843: step: 2414/1541, loss: 0.00011078726674895734 2023-01-23 21:35:08.447346: step: 2416/1541, loss: 0.0007418569875881076 2023-01-23 21:35:09.346536: step: 2418/1541, loss: 0.0006546163931488991 2023-01-23 21:35:10.287441: step: 2420/1541, loss: 0.0021566820796579123 2023-01-23 21:35:11.173954: step: 2422/1541, loss: 0.00014705478679388762 2023-01-23 21:35:12.046148: step: 2424/1541, loss: 0.002654728479683399 2023-01-23 21:35:12.956133: step: 2426/1541, loss: 0.4382820725440979 2023-01-23 21:35:13.837980: step: 2428/1541, loss: 0.003700685454532504 2023-01-23 21:35:14.826312: step: 2430/1541, loss: 0.03191673755645752 2023-01-23 21:35:15.718739: step: 2432/1541, loss: 0.004124502185732126 2023-01-23 21:35:16.650133: step: 2434/1541, loss: 0.000176249144715257 2023-01-23 21:35:17.608330: step: 2436/1541, loss: 0.004120737314224243 2023-01-23 21:35:18.490773: step: 2438/1541, loss: 0.0002579287684056908 2023-01-23 21:35:19.410399: step: 2440/1541, loss: 6.563959323102608e-05 2023-01-23 21:35:20.329758: step: 2442/1541, loss: 7.899031334090978e-05 2023-01-23 21:35:21.205254: step: 2444/1541, loss: -1.635998341953382e-05 2023-01-23 21:35:22.107005: step: 2446/1541, loss: 0.0011791335418820381 2023-01-23 21:35:23.077764: step: 2448/1541, loss: 7.324368198169395e-05 2023-01-23 21:35:23.956568: step: 2450/1541, loss: 6.644189124926925e-05 2023-01-23 21:35:24.863466: step: 2452/1541, loss: -1.8978118532686494e-05 2023-01-23 21:35:25.798778: step: 2454/1541, loss: 0.06686105579137802 2023-01-23 21:35:26.681781: step: 2456/1541, loss: 4.40402545791585e-05 2023-01-23 21:35:27.577619: step: 2458/1541, loss: 0.02781241200864315 2023-01-23 21:35:28.460721: step: 2460/1541, loss: 0.013675667345523834 2023-01-23 21:35:29.310747: step: 2462/1541, loss: 0.0013257687678560615 2023-01-23 21:35:30.186033: step: 2464/1541, loss: 0.00022971691214479506 2023-01-23 21:35:31.080263: step: 2466/1541, loss: 0.0005348509876057506 2023-01-23 21:35:31.999750: step: 2468/1541, loss: 0.701015830039978 2023-01-23 21:35:32.933510: step: 2470/1541, loss: 0.0005111763020977378 2023-01-23 21:35:33.846984: step: 2472/1541, loss: 0.2680231034755707 2023-01-23 21:35:34.695313: step: 2474/1541, loss: 8.332322613568977e-05 2023-01-23 21:35:35.645582: step: 2476/1541, loss: 0.00017659754666965455 2023-01-23 21:35:36.557937: step: 2478/1541, loss: 0.008516231551766396 2023-01-23 21:35:37.435423: step: 2480/1541, loss: 0.23175159096717834 2023-01-23 21:35:38.285768: step: 2482/1541, loss: 0.013998027890920639 2023-01-23 21:35:39.150610: step: 2484/1541, loss: 0.023899098858237267 2023-01-23 21:35:40.075115: step: 2486/1541, loss: 0.001296044560149312 2023-01-23 21:35:40.919619: step: 2488/1541, loss: 5.645795681630261e-05 2023-01-23 21:35:41.825942: step: 2490/1541, loss: 0.00046579065383411944 2023-01-23 21:35:42.683164: step: 2492/1541, loss: 0.0021414950024336576 2023-01-23 21:35:43.589950: step: 2494/1541, loss: 6.266361015150324e-05 2023-01-23 21:35:44.438064: step: 2496/1541, loss: 0.0011638345895335078 2023-01-23 21:35:45.348622: step: 2498/1541, loss: 0.09610864520072937 2023-01-23 21:35:46.229610: step: 2500/1541, loss: 0.0011386467376723886 2023-01-23 21:35:47.124207: step: 2502/1541, loss: 0.1341967135667801 2023-01-23 21:35:48.001764: step: 2504/1541, loss: 0.21449267864227295 2023-01-23 21:35:48.904088: step: 2506/1541, loss: 0.05293000489473343 2023-01-23 21:35:49.772530: step: 2508/1541, loss: 0.00010175702482229099 2023-01-23 21:35:50.627056: step: 2510/1541, loss: 0.024256883189082146 2023-01-23 21:35:51.552282: step: 2512/1541, loss: 0.00043969028047285974 2023-01-23 21:35:52.410890: step: 2514/1541, loss: 0.0004883359069935977 2023-01-23 21:35:53.307163: step: 2516/1541, loss: 0.008684959262609482 2023-01-23 21:35:54.174437: step: 2518/1541, loss: 0.14202246069908142 2023-01-23 21:35:55.047189: step: 2520/1541, loss: 2.6968649763148278e-05 2023-01-23 21:35:55.971918: step: 2522/1541, loss: 0.005500837694853544 2023-01-23 21:35:56.856753: step: 2524/1541, loss: 0.008041844703257084 2023-01-23 21:35:57.722533: step: 2526/1541, loss: 0.07749363034963608 2023-01-23 21:35:58.636920: step: 2528/1541, loss: 0.0012738379882648587 2023-01-23 21:35:59.526308: step: 2530/1541, loss: 0.009696230292320251 2023-01-23 21:36:00.400134: step: 2532/1541, loss: 0.00042647417285479605 2023-01-23 21:36:01.314642: step: 2534/1541, loss: 1.914908534672577e-05 2023-01-23 21:36:02.217678: step: 2536/1541, loss: 0.00035363214556127787 2023-01-23 21:36:03.113123: step: 2538/1541, loss: 0.10938326269388199 2023-01-23 21:36:04.034571: step: 2540/1541, loss: -7.753078534733504e-08 2023-01-23 21:36:04.961207: step: 2542/1541, loss: 0.08255802094936371 2023-01-23 21:36:05.853507: step: 2544/1541, loss: 4.2106217733817175e-05 2023-01-23 21:36:06.742694: step: 2546/1541, loss: 0.06389123946428299 2023-01-23 21:36:07.646698: step: 2548/1541, loss: 0.009529558941721916 2023-01-23 21:36:08.535654: step: 2550/1541, loss: 0.0014385880203917623 2023-01-23 21:36:09.458778: step: 2552/1541, loss: 0.040161363780498505 2023-01-23 21:36:10.324389: step: 2554/1541, loss: 0.005886697210371494 2023-01-23 21:36:11.177859: step: 2556/1541, loss: 0.1315467804670334 2023-01-23 21:36:12.083248: step: 2558/1541, loss: 0.002201540395617485 2023-01-23 21:36:12.989457: step: 2560/1541, loss: 0.0003504323249217123 2023-01-23 21:36:13.888970: step: 2562/1541, loss: 0.00024171009135898203 2023-01-23 21:36:14.776396: step: 2564/1541, loss: 3.628410559031181e-05 2023-01-23 21:36:15.665180: step: 2566/1541, loss: 2.649485395522788e-05 2023-01-23 21:36:16.581222: step: 2568/1541, loss: 0.006584804970771074 2023-01-23 21:36:17.453991: step: 2570/1541, loss: 0.0017513467464596033 2023-01-23 21:36:18.340192: step: 2572/1541, loss: 0.025671081617474556 2023-01-23 21:36:19.195991: step: 2574/1541, loss: 0.0005398127250373363 2023-01-23 21:36:20.061423: step: 2576/1541, loss: 3.0834849894745275e-05 2023-01-23 21:36:20.979536: step: 2578/1541, loss: 6.872666563140228e-05 2023-01-23 21:36:21.841498: step: 2580/1541, loss: 0.09633888304233551 2023-01-23 21:36:22.720720: step: 2582/1541, loss: 0.00011583372543100268 2023-01-23 21:36:23.653867: step: 2584/1541, loss: 0.0026421642396599054 2023-01-23 21:36:24.557974: step: 2586/1541, loss: 0.4290817379951477 2023-01-23 21:36:25.437851: step: 2588/1541, loss: 0.0006407178589142859 2023-01-23 21:36:26.328970: step: 2590/1541, loss: 0.0008338213665410876 2023-01-23 21:36:27.223911: step: 2592/1541, loss: 0.0019408253720030189 2023-01-23 21:36:28.150046: step: 2594/1541, loss: 3.9779730286682025e-05 2023-01-23 21:36:29.090258: step: 2596/1541, loss: 0.00010016596934292465 2023-01-23 21:36:29.978578: step: 2598/1541, loss: 0.00024523891624994576 2023-01-23 21:36:30.873884: step: 2600/1541, loss: 0.0008664726046845317 2023-01-23 21:36:31.761872: step: 2602/1541, loss: 0.16543306410312653 2023-01-23 21:36:32.641474: step: 2604/1541, loss: 0.0013353435788303614 2023-01-23 21:36:33.524640: step: 2606/1541, loss: 0.46662625670433044 2023-01-23 21:36:34.402217: step: 2608/1541, loss: 0.0008671408868394792 2023-01-23 21:36:35.337084: step: 2610/1541, loss: 0.20085731148719788 2023-01-23 21:36:36.214268: step: 2612/1541, loss: 0.00015600894403178245 2023-01-23 21:36:37.159203: step: 2614/1541, loss: 0.20495013892650604 2023-01-23 21:36:38.042740: step: 2616/1541, loss: 0.11037760227918625 2023-01-23 21:36:38.954170: step: 2618/1541, loss: 0.06925712525844574 2023-01-23 21:36:39.869477: step: 2620/1541, loss: 0.006834788713604212 2023-01-23 21:36:40.717807: step: 2622/1541, loss: 0.002995555056259036 2023-01-23 21:36:41.605720: step: 2624/1541, loss: 0.3312494158744812 2023-01-23 21:36:42.506026: step: 2626/1541, loss: 0.0006501996540464461 2023-01-23 21:36:43.374815: step: 2628/1541, loss: 0.0010133711621165276 2023-01-23 21:36:44.252360: step: 2630/1541, loss: 0.05653497949242592 2023-01-23 21:36:45.145437: step: 2632/1541, loss: 0.0023154797963798046 2023-01-23 21:36:46.034143: step: 2634/1541, loss: 0.0022271526977419853 2023-01-23 21:36:46.956290: step: 2636/1541, loss: 0.004283569753170013 2023-01-23 21:36:47.820852: step: 2638/1541, loss: 0.07519231736660004 2023-01-23 21:36:48.724588: step: 2640/1541, loss: 1.4923354387283325 2023-01-23 21:36:49.603667: step: 2642/1541, loss: 0.008397018536925316 2023-01-23 21:36:50.487240: step: 2644/1541, loss: 0.07233992218971252 2023-01-23 21:36:51.415235: step: 2646/1541, loss: 0.05004410445690155 2023-01-23 21:36:52.272631: step: 2648/1541, loss: 5.34045830136165e-05 2023-01-23 21:36:53.169477: step: 2650/1541, loss: 0.0011391682783141732 2023-01-23 21:36:54.100916: step: 2652/1541, loss: 0.09721431136131287 2023-01-23 21:36:55.023790: step: 2654/1541, loss: 0.016552165150642395 2023-01-23 21:36:55.923847: step: 2656/1541, loss: 0.00026859669014811516 2023-01-23 21:36:56.795041: step: 2658/1541, loss: 0.0008251350955106318 2023-01-23 21:36:57.701524: step: 2660/1541, loss: 0.0044682445004582405 2023-01-23 21:36:58.559082: step: 2662/1541, loss: 0.014311134815216064 2023-01-23 21:36:59.443753: step: 2664/1541, loss: 0.0010381703032180667 2023-01-23 21:37:00.326037: step: 2666/1541, loss: 0.0005723139620386064 2023-01-23 21:37:01.252063: step: 2668/1541, loss: 0.002879119710996747 2023-01-23 21:37:02.175107: step: 2670/1541, loss: 0.0006118641467764974 2023-01-23 21:37:03.121884: step: 2672/1541, loss: 0.0005328195984475315 2023-01-23 21:37:03.991118: step: 2674/1541, loss: 0.0012446208856999874 2023-01-23 21:37:04.924788: step: 2676/1541, loss: 9.80435470410157e-06 2023-01-23 21:37:05.779070: step: 2678/1541, loss: 0.0013314586831256747 2023-01-23 21:37:06.668993: step: 2680/1541, loss: 0.008194266818463802 2023-01-23 21:37:07.558017: step: 2682/1541, loss: 9.057354873220902e-06 2023-01-23 21:37:08.439135: step: 2684/1541, loss: 0.0035088027361780405 2023-01-23 21:37:09.378823: step: 2686/1541, loss: 0.020693104714155197 2023-01-23 21:37:10.281769: step: 2688/1541, loss: 0.09953288733959198 2023-01-23 21:37:11.183953: step: 2690/1541, loss: 0.00019576025079004467 2023-01-23 21:37:12.124512: step: 2692/1541, loss: 0.00026676899869926274 2023-01-23 21:37:12.990432: step: 2694/1541, loss: 0.0007185761933214962 2023-01-23 21:37:13.860295: step: 2696/1541, loss: 0.0031299572438001633 2023-01-23 21:37:14.735056: step: 2698/1541, loss: -1.5257228369591758e-06 2023-01-23 21:37:15.600622: step: 2700/1541, loss: 0.0002925268781837076 2023-01-23 21:37:16.580954: step: 2702/1541, loss: 0.00114347820635885 2023-01-23 21:37:17.480637: step: 2704/1541, loss: 0.09077602624893188 2023-01-23 21:37:18.432928: step: 2706/1541, loss: 4.786089266417548e-05 2023-01-23 21:37:19.325836: step: 2708/1541, loss: 0.05790863558650017 2023-01-23 21:37:20.224045: step: 2710/1541, loss: 0.0006748943123966455 2023-01-23 21:37:21.096470: step: 2712/1541, loss: 0.009054097346961498 2023-01-23 21:37:21.972221: step: 2714/1541, loss: 0.0008677655132487416 2023-01-23 21:37:22.874875: step: 2716/1541, loss: 0.00010401179315522313 2023-01-23 21:37:23.831254: step: 2718/1541, loss: 3.6578636354533955e-05 2023-01-23 21:37:24.704388: step: 2720/1541, loss: 0.0071707540191709995 2023-01-23 21:37:25.570160: step: 2722/1541, loss: 0.0001239750417880714 2023-01-23 21:37:26.466615: step: 2724/1541, loss: 0.0018572225235402584 2023-01-23 21:37:27.338880: step: 2726/1541, loss: 0.0916808620095253 2023-01-23 21:37:28.240975: step: 2728/1541, loss: 0.20354880392551422 2023-01-23 21:37:29.099553: step: 2730/1541, loss: 1.0314361134078354e-05 2023-01-23 21:37:29.984869: step: 2732/1541, loss: 7.775066478643566e-05 2023-01-23 21:37:30.848624: step: 2734/1541, loss: 0.000260885659372434 2023-01-23 21:37:31.734894: step: 2736/1541, loss: 0.115012526512146 2023-01-23 21:37:32.650603: step: 2738/1541, loss: 3.82596263079904e-05 2023-01-23 21:37:33.534875: step: 2740/1541, loss: 0.0020861481316387653 2023-01-23 21:37:34.423221: step: 2742/1541, loss: 7.643006028956734e-06 2023-01-23 21:37:35.343639: step: 2744/1541, loss: 0.0055339038372039795 2023-01-23 21:37:36.269007: step: 2746/1541, loss: -9.764012793311849e-05 2023-01-23 21:37:37.187308: step: 2748/1541, loss: 0.0010145286796614528 2023-01-23 21:37:38.114039: step: 2750/1541, loss: 0.0001895124587463215 2023-01-23 21:37:39.002615: step: 2752/1541, loss: 0.1636124700307846 2023-01-23 21:37:39.923319: step: 2754/1541, loss: 0.030047105625271797 2023-01-23 21:37:40.769043: step: 2756/1541, loss: 0.00026478696963749826 2023-01-23 21:37:41.683028: step: 2758/1541, loss: 0.544425904750824 2023-01-23 21:37:42.567657: step: 2760/1541, loss: 0.0020296748261898756 2023-01-23 21:37:43.467624: step: 2762/1541, loss: 0.0016368411015719175 2023-01-23 21:37:44.399555: step: 2764/1541, loss: 0.0017639631405472755 2023-01-23 21:37:45.330975: step: 2766/1541, loss: 0.2840316593647003 2023-01-23 21:37:46.226871: step: 2768/1541, loss: 0.0008529129554517567 2023-01-23 21:37:47.161752: step: 2770/1541, loss: 0.34226295351982117 2023-01-23 21:37:48.100088: step: 2772/1541, loss: 0.0027220435440540314 2023-01-23 21:37:49.002349: step: 2774/1541, loss: 0.1341380625963211 2023-01-23 21:37:49.908014: step: 2776/1541, loss: 0.019030414521694183 2023-01-23 21:37:50.786407: step: 2778/1541, loss: 0.5754509568214417 2023-01-23 21:37:51.728249: step: 2780/1541, loss: 0.008493978530168533 2023-01-23 21:37:52.605578: step: 2782/1541, loss: 0.2169143408536911 2023-01-23 21:37:53.494782: step: 2784/1541, loss: 0.009645001962780952 2023-01-23 21:37:54.406204: step: 2786/1541, loss: 1.251846194267273 2023-01-23 21:37:55.402131: step: 2788/1541, loss: -2.7808549930341542e-05 2023-01-23 21:37:56.300769: step: 2790/1541, loss: 0.00010550901060923934 2023-01-23 21:37:57.178786: step: 2792/1541, loss: 0.012260095216333866 2023-01-23 21:37:58.057245: step: 2794/1541, loss: 0.20479801297187805 2023-01-23 21:37:58.947737: step: 2796/1541, loss: 0.00634554261341691 2023-01-23 21:37:59.849566: step: 2798/1541, loss: 0.0010539115173742175 2023-01-23 21:38:00.751477: step: 2800/1541, loss: 0.00048367801355198026 2023-01-23 21:38:01.637851: step: 2802/1541, loss: 6.055686026229523e-05 2023-01-23 21:38:02.609303: step: 2804/1541, loss: 0.38136276602745056 2023-01-23 21:38:03.481884: step: 2806/1541, loss: 0.003541575511917472 2023-01-23 21:38:04.369980: step: 2808/1541, loss: 0.027629701420664787 2023-01-23 21:38:05.313770: step: 2810/1541, loss: 1.0226665835944004e-05 2023-01-23 21:38:06.185963: step: 2812/1541, loss: 0.00015926011838018894 2023-01-23 21:38:07.040271: step: 2814/1541, loss: 0.024090329185128212 2023-01-23 21:38:08.014868: step: 2816/1541, loss: 0.09832914918661118 2023-01-23 21:38:08.906092: step: 2818/1541, loss: 0.0226501002907753 2023-01-23 21:38:09.751670: step: 2820/1541, loss: 0.10292334109544754 2023-01-23 21:38:10.638156: step: 2822/1541, loss: 0.001238064025528729 2023-01-23 21:38:11.498220: step: 2824/1541, loss: 0.0010350249940529466 2023-01-23 21:38:12.381854: step: 2826/1541, loss: 0.00011400604125810787 2023-01-23 21:38:13.248343: step: 2828/1541, loss: 0.10647591203451157 2023-01-23 21:38:14.145869: step: 2830/1541, loss: 5.169761061551981e-05 2023-01-23 21:38:15.125735: step: 2832/1541, loss: 0.05639069527387619 2023-01-23 21:38:16.027173: step: 2834/1541, loss: 0.007107700686901808 2023-01-23 21:38:16.885754: step: 2836/1541, loss: 0.36181411147117615 2023-01-23 21:38:17.784328: step: 2838/1541, loss: 0.04591454192996025 2023-01-23 21:38:18.721245: step: 2840/1541, loss: 0.006018301472067833 2023-01-23 21:38:19.602022: step: 2842/1541, loss: 0.00021474953973665833 2023-01-23 21:38:20.426510: step: 2844/1541, loss: 0.0039954353123903275 2023-01-23 21:38:21.296333: step: 2846/1541, loss: 0.07450257241725922 2023-01-23 21:38:22.191300: step: 2848/1541, loss: 0.003633934771642089 2023-01-23 21:38:23.086932: step: 2850/1541, loss: 3.15877769025974e-05 2023-01-23 21:38:23.935400: step: 2852/1541, loss: 0.0014350597048178315 2023-01-23 21:38:24.800251: step: 2854/1541, loss: 0.024834252893924713 2023-01-23 21:38:25.715603: step: 2856/1541, loss: 0.038783397525548935 2023-01-23 21:38:26.604705: step: 2858/1541, loss: 0.0006602359935641289 2023-01-23 21:38:27.474734: step: 2860/1541, loss: 6.798854155931622e-05 2023-01-23 21:38:28.395368: step: 2862/1541, loss: 0.6808887720108032 2023-01-23 21:38:29.280248: step: 2864/1541, loss: 0.0006893100799061358 2023-01-23 21:38:30.152264: step: 2866/1541, loss: 2.1734811525675468e-05 2023-01-23 21:38:31.027421: step: 2868/1541, loss: 0.00010590285819489509 2023-01-23 21:38:31.948701: step: 2870/1541, loss: 0.015612330287694931 2023-01-23 21:38:32.900109: step: 2872/1541, loss: 0.00018311299209017307 2023-01-23 21:38:33.806802: step: 2874/1541, loss: 0.006795779336243868 2023-01-23 21:38:34.704612: step: 2876/1541, loss: 0.006602097302675247 2023-01-23 21:38:35.603875: step: 2878/1541, loss: 0.062448494136333466 2023-01-23 21:38:36.476201: step: 2880/1541, loss: 7.528916466981173e-05 2023-01-23 21:38:37.366427: step: 2882/1541, loss: 0.02520497888326645 2023-01-23 21:38:38.328186: step: 2884/1541, loss: 0.32202407717704773 2023-01-23 21:38:39.230645: step: 2886/1541, loss: 0.003877073060721159 2023-01-23 21:38:40.124850: step: 2888/1541, loss: 3.2507454307051376e-05 2023-01-23 21:38:41.041096: step: 2890/1541, loss: 0.00032518457737751305 2023-01-23 21:38:41.909962: step: 2892/1541, loss: 0.00019753805827349424 2023-01-23 21:38:42.885575: step: 2894/1541, loss: 1.835077637224458e-05 2023-01-23 21:38:43.814347: step: 2896/1541, loss: 0.008842447772622108 2023-01-23 21:38:44.691483: step: 2898/1541, loss: 0.0002022263506660238 2023-01-23 21:38:45.603242: step: 2900/1541, loss: 0.017281709238886833 2023-01-23 21:38:46.508437: step: 2902/1541, loss: 0.00028119306080043316 2023-01-23 21:38:47.382276: step: 2904/1541, loss: 0.02677961438894272 2023-01-23 21:38:48.322313: step: 2906/1541, loss: 0.00016526394756510854 2023-01-23 21:38:49.230908: step: 2908/1541, loss: 0.014088964089751244 2023-01-23 21:38:50.138437: step: 2910/1541, loss: 0.11959237605333328 2023-01-23 21:38:51.160848: step: 2912/1541, loss: 8.828163117868826e-05 2023-01-23 21:38:52.092090: step: 2914/1541, loss: 0.39995160698890686 2023-01-23 21:38:52.974719: step: 2916/1541, loss: 0.024328099563717842 2023-01-23 21:38:53.926444: step: 2918/1541, loss: 0.009360560216009617 2023-01-23 21:38:54.851845: step: 2920/1541, loss: 0.0005357297486625612 2023-01-23 21:38:55.805529: step: 2922/1541, loss: 0.0021929487120360136 2023-01-23 21:38:56.708894: step: 2924/1541, loss: 0.4657498896121979 2023-01-23 21:38:57.574327: step: 2926/1541, loss: 0.35397735238075256 2023-01-23 21:38:58.580107: step: 2928/1541, loss: 0.3361801505088806 2023-01-23 21:38:59.548817: step: 2930/1541, loss: 0.16280022263526917 2023-01-23 21:39:00.476330: step: 2932/1541, loss: 0.00024204497458413243 2023-01-23 21:39:01.350185: step: 2934/1541, loss: 0.002770470455288887 2023-01-23 21:39:02.261506: step: 2936/1541, loss: 0.019786741584539413 2023-01-23 21:39:03.183246: step: 2938/1541, loss: 0.0008851408492773771 2023-01-23 21:39:04.075178: step: 2940/1541, loss: 0.06355790048837662 2023-01-23 21:39:04.943721: step: 2942/1541, loss: 0.2315392941236496 2023-01-23 21:39:05.805863: step: 2944/1541, loss: 5.169939322513528e-05 2023-01-23 21:39:06.669433: step: 2946/1541, loss: 0.0003643987001851201 2023-01-23 21:39:07.585306: step: 2948/1541, loss: 0.0009643564117141068 2023-01-23 21:39:08.584871: step: 2950/1541, loss: 0.00040220003575086594 2023-01-23 21:39:09.452128: step: 2952/1541, loss: 0.008941004052758217 2023-01-23 21:39:10.374156: step: 2954/1541, loss: 0.004339165985584259 2023-01-23 21:39:11.308048: step: 2956/1541, loss: 0.0002916785888373852 2023-01-23 21:39:12.207540: step: 2958/1541, loss: 0.14318065345287323 2023-01-23 21:39:13.102226: step: 2960/1541, loss: 0.18598617613315582 2023-01-23 21:39:14.001204: step: 2962/1541, loss: 0.44477468729019165 2023-01-23 21:39:14.855731: step: 2964/1541, loss: 0.0005758926272392273 2023-01-23 21:39:15.714691: step: 2966/1541, loss: 0.0004439751501195133 2023-01-23 21:39:16.560697: step: 2968/1541, loss: 0.03864666074514389 2023-01-23 21:39:17.449834: step: 2970/1541, loss: 0.045902036130428314 2023-01-23 21:39:18.317501: step: 2972/1541, loss: 0.0005481585394591093 2023-01-23 21:39:19.214103: step: 2974/1541, loss: 0.02618306502699852 2023-01-23 21:39:20.213064: step: 2976/1541, loss: 0.06802093982696533 2023-01-23 21:39:21.096881: step: 2978/1541, loss: 0.0003372339706402272 2023-01-23 21:39:21.978966: step: 2980/1541, loss: 0.03987034037709236 2023-01-23 21:39:22.850387: step: 2982/1541, loss: 5.033698835177347e-05 2023-01-23 21:39:23.765911: step: 2984/1541, loss: 0.007891098037362099 2023-01-23 21:39:24.643460: step: 2986/1541, loss: -2.693128408282064e-05 2023-01-23 21:39:25.523004: step: 2988/1541, loss: 0.0003342179988976568 2023-01-23 21:39:26.400806: step: 2990/1541, loss: 0.03318024426698685 2023-01-23 21:39:27.351135: step: 2992/1541, loss: 0.0010087568080052733 2023-01-23 21:39:28.246910: step: 2994/1541, loss: 0.0004884326481260359 2023-01-23 21:39:29.132836: step: 2996/1541, loss: 0.00019286313909105957 2023-01-23 21:39:29.993311: step: 2998/1541, loss: 0.3103654682636261 2023-01-23 21:39:30.969338: step: 3000/1541, loss: 0.0027597867883741856 2023-01-23 21:39:31.892412: step: 3002/1541, loss: 0.0032041214872151613 2023-01-23 21:39:32.776264: step: 3004/1541, loss: 0.0011140910210087895 2023-01-23 21:39:33.712409: step: 3006/1541, loss: 0.0018319926457479596 2023-01-23 21:39:34.583791: step: 3008/1541, loss: 0.0015577260637655854 2023-01-23 21:39:35.471113: step: 3010/1541, loss: 0.10999844968318939 2023-01-23 21:39:36.456169: step: 3012/1541, loss: 0.28245171904563904 2023-01-23 21:39:37.338932: step: 3014/1541, loss: 0.0006354727665893734 2023-01-23 21:39:38.232868: step: 3016/1541, loss: 0.001281008473597467 2023-01-23 21:39:39.090500: step: 3018/1541, loss: 0.00031398533610627055 2023-01-23 21:39:39.955658: step: 3020/1541, loss: 0.005755788646638393 2023-01-23 21:39:40.821473: step: 3022/1541, loss: 0.0006874944665469229 2023-01-23 21:39:41.781463: step: 3024/1541, loss: 0.005729160271584988 2023-01-23 21:39:42.682016: step: 3026/1541, loss: 0.07714920490980148 2023-01-23 21:39:43.644843: step: 3028/1541, loss: 0.0221499465405941 2023-01-23 21:39:44.544685: step: 3030/1541, loss: 8.356035687029362e-05 2023-01-23 21:39:45.381349: step: 3032/1541, loss: 0.01592499203979969 2023-01-23 21:39:46.320432: step: 3034/1541, loss: 0.0007011053967289627 2023-01-23 21:39:47.249764: step: 3036/1541, loss: 0.001839327858760953 2023-01-23 21:39:48.123503: step: 3038/1541, loss: 8.193400572054088e-05 2023-01-23 21:39:49.042094: step: 3040/1541, loss: 0.001328221638686955 2023-01-23 21:39:49.952257: step: 3042/1541, loss: 0.012190422043204308 2023-01-23 21:39:50.941613: step: 3044/1541, loss: 0.03553859144449234 2023-01-23 21:39:51.966825: step: 3046/1541, loss: 0.45419538021087646 2023-01-23 21:39:52.836272: step: 3048/1541, loss: 0.019541360437870026 2023-01-23 21:39:53.688741: step: 3050/1541, loss: 0.001107479678466916 2023-01-23 21:39:54.547045: step: 3052/1541, loss: 0.01566120982170105 2023-01-23 21:39:55.489778: step: 3054/1541, loss: 0.00017244418268091977 2023-01-23 21:39:56.403674: step: 3056/1541, loss: 0.0021266727708280087 2023-01-23 21:39:57.288968: step: 3058/1541, loss: 0.0010117321508005261 2023-01-23 21:39:58.207690: step: 3060/1541, loss: 0.0004704854218289256 2023-01-23 21:39:59.111162: step: 3062/1541, loss: 0.00017041787214111537 2023-01-23 21:40:00.028931: step: 3064/1541, loss: 0.00029931587050668895 2023-01-23 21:40:00.926341: step: 3066/1541, loss: 0.3270507752895355 2023-01-23 21:40:01.777494: step: 3068/1541, loss: 0.017923878505825996 2023-01-23 21:40:02.660318: step: 3070/1541, loss: 0.00031257420778274536 2023-01-23 21:40:03.540922: step: 3072/1541, loss: 0.0312928669154644 2023-01-23 21:40:04.421367: step: 3074/1541, loss: 0.0005937493406236172 2023-01-23 21:40:05.323831: step: 3076/1541, loss: 0.06125633418560028 2023-01-23 21:40:06.203867: step: 3078/1541, loss: 6.515049790323246e-06 2023-01-23 21:40:07.083961: step: 3080/1541, loss: 0.0006469248910434544 2023-01-23 21:40:07.965220: step: 3082/1541, loss: 0.2375534325838089 2023-01-23 21:40:08.563614: step: 3084/1541, loss: 5.828607754665427e-05 ================================================== Loss: 0.056 -------------------- Dev Chinese: {'event': {'p': 0.6236559139784946, 'r': 0.6950732356857523, 'f1': 0.6574307304785894}, 'argument': {'p': 0.6857984262495748, 'r': 0.6537517708173516, 'f1': 0.6693917653345131}, 'combined': 0.4400787172602214, 'epoch': 34} Test Chinese: {'event': {'p': 0.61198738170347, 'r': 0.6974236069502696, 'f1': 0.6519182301876226}, 'argument': {'p': 0.7289310758677934, 'r': 0.6912094239825067, 'f1': 0.7095692702611291}, 'combined': 0.4625811428641582, 'epoch': 34} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6797819022259629, 'r': 0.6563747779025437, 'f1': 0.6678733141577707}, 'combined': 0.4397178341659857, 'epoch': 34} Test Korean: {'event': {'p': 0.6129647283126788, 'r': 0.7705212702216896, 'f1': 0.6827714361560924}, 'argument': {'p': 0.7272593327512444, 'r': 0.7030003311320843, 'f1': 0.714924100362141}, 'combined': 0.4881297547468614, 'epoch': 34} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6884964799060531, 'r': 0.6610188342471519, 'f1': 0.6744779192318231}, 'combined': 0.4676944066899037, 'epoch': 34} Test Russian: {'event': {'p': 0.652014652014652, 'r': 0.7465548232474536, 'f1': 0.6960893854748604}, 'argument': {'p': 0.7251475848977346, 'r': 0.6962665723668237, 'f1': 0.7104136691145296}, 'combined': 0.4945114143668738, 'epoch': 34} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6703869047619045, 'r': 0.6644542772861355, 'f1': 0.6674074074074071}, 'combined': 0.5019816397594173, 'epoch': 34} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5878585352269563, 'r': 0.44089390142021717, 'f1': 0.5038787444802483}, 'combined': 0.29359863851605017, 'epoch': 34} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5669764957264958, 'r': 0.5274199960246473, 'f1': 0.5464833693749357}, 'combined': 0.2810485899642527, 'epoch': 34} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 35 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 21:45:30.466167: step: 2/1541, loss: 0.0003253375762142241 2023-01-23 21:45:31.351543: step: 4/1541, loss: 0.06308946758508682 2023-01-23 21:45:32.253119: step: 6/1541, loss: 0.0001258223201148212 2023-01-23 21:45:33.123440: step: 8/1541, loss: 0.00047231142525561154 2023-01-23 21:45:34.005648: step: 10/1541, loss: 0.003913385793566704 2023-01-23 21:45:34.878943: step: 12/1541, loss: 0.00017103554273489863 2023-01-23 21:45:35.737893: step: 14/1541, loss: 0.00011211669334443286 2023-01-23 21:45:36.599382: step: 16/1541, loss: 0.028908856213092804 2023-01-23 21:45:37.495950: step: 18/1541, loss: 0.0001469989656470716 2023-01-23 21:45:38.425194: step: 20/1541, loss: 2.9496843126253225e-05 2023-01-23 21:45:39.336623: step: 22/1541, loss: 0.0013272175565361977 2023-01-23 21:45:40.242735: step: 24/1541, loss: 7.821351027814671e-05 2023-01-23 21:45:41.135820: step: 26/1541, loss: 0.07069814205169678 2023-01-23 21:45:41.984880: step: 28/1541, loss: 0.0001641884446144104 2023-01-23 21:45:42.895328: step: 30/1541, loss: 0.0017412914894521236 2023-01-23 21:45:43.792806: step: 32/1541, loss: 0.02307557500898838 2023-01-23 21:45:44.664736: step: 34/1541, loss: 5.8088702644454315e-05 2023-01-23 21:45:45.531357: step: 36/1541, loss: 0.00015512167010456324 2023-01-23 21:45:46.428413: step: 38/1541, loss: 0.00024630501866340637 2023-01-23 21:45:47.304297: step: 40/1541, loss: 0.00011746111704269424 2023-01-23 21:45:48.213094: step: 42/1541, loss: 0.004545825533568859 2023-01-23 21:45:49.097591: step: 44/1541, loss: 0.004892605822533369 2023-01-23 21:45:50.015186: step: 46/1541, loss: 0.008391664363443851 2023-01-23 21:45:50.931430: step: 48/1541, loss: 0.23068493604660034 2023-01-23 21:45:51.825641: step: 50/1541, loss: 0.0068092732690274715 2023-01-23 21:45:52.710211: step: 52/1541, loss: 0.005082844290882349 2023-01-23 21:45:53.617884: step: 54/1541, loss: 0.000393041962524876 2023-01-23 21:45:54.496212: step: 56/1541, loss: 0.21453288197517395 2023-01-23 21:45:55.354230: step: 58/1541, loss: 5.592111119767651e-05 2023-01-23 21:45:56.296148: step: 60/1541, loss: 5.07555523654446e-05 2023-01-23 21:45:57.170497: step: 62/1541, loss: 8.10672645457089e-05 2023-01-23 21:45:58.014413: step: 64/1541, loss: 0.04336938634514809 2023-01-23 21:45:58.966315: step: 66/1541, loss: 0.00047232708311639726 2023-01-23 21:45:59.884432: step: 68/1541, loss: 0.00015526911010965705 2023-01-23 21:46:00.772565: step: 70/1541, loss: 0.0004193625063635409 2023-01-23 21:46:01.635909: step: 72/1541, loss: 0.0001439473417121917 2023-01-23 21:46:02.615367: step: 74/1541, loss: 0.12148822098970413 2023-01-23 21:46:03.467018: step: 76/1541, loss: 0.0009628627449274063 2023-01-23 21:46:04.387584: step: 78/1541, loss: 0.0002500890986993909 2023-01-23 21:46:05.245191: step: 80/1541, loss: 0.00615455349907279 2023-01-23 21:46:06.134192: step: 82/1541, loss: 0.17824234068393707 2023-01-23 21:46:07.028679: step: 84/1541, loss: 0.04792819917201996 2023-01-23 21:46:07.934540: step: 86/1541, loss: 0.00010125542030436918 2023-01-23 21:46:08.804161: step: 88/1541, loss: 0.005894065834581852 2023-01-23 21:46:09.714551: step: 90/1541, loss: 5.638703441945836e-05 2023-01-23 21:46:10.576577: step: 92/1541, loss: 0.001162716536782682 2023-01-23 21:46:11.465746: step: 94/1541, loss: 0.0025366288609802723 2023-01-23 21:46:12.355563: step: 96/1541, loss: -1.1334401278872974e-05 2023-01-23 21:46:13.248111: step: 98/1541, loss: 0.00035913140163756907 2023-01-23 21:46:14.138185: step: 100/1541, loss: 0.04348123073577881 2023-01-23 21:46:15.004998: step: 102/1541, loss: 0.006190251559019089 2023-01-23 21:46:15.869164: step: 104/1541, loss: 0.0010359446750953794 2023-01-23 21:46:16.771846: step: 106/1541, loss: 0.000886402209289372 2023-01-23 21:46:17.686808: step: 108/1541, loss: 0.013295607641339302 2023-01-23 21:46:18.603254: step: 110/1541, loss: 0.0007087074918672442 2023-01-23 21:46:19.530248: step: 112/1541, loss: 0.0039779809303581715 2023-01-23 21:46:20.405907: step: 114/1541, loss: 0.0003111975092906505 2023-01-23 21:46:21.274436: step: 116/1541, loss: 6.777804810553789e-05 2023-01-23 21:46:22.152903: step: 118/1541, loss: 8.079727194854058e-06 2023-01-23 21:46:23.059021: step: 120/1541, loss: 0.29324206709861755 2023-01-23 21:46:23.999477: step: 122/1541, loss: 0.11530908942222595 2023-01-23 21:46:24.917753: step: 124/1541, loss: 0.2106274664402008 2023-01-23 21:46:25.785047: step: 126/1541, loss: 0.00010750480578280985 2023-01-23 21:46:26.704543: step: 128/1541, loss: 8.354559395229444e-05 2023-01-23 21:46:27.612531: step: 130/1541, loss: 0.10855117440223694 2023-01-23 21:46:28.535951: step: 132/1541, loss: 1.600500945642125e-05 2023-01-23 21:46:29.453971: step: 134/1541, loss: 0.00037544319638982415 2023-01-23 21:46:30.348073: step: 136/1541, loss: 0.01409247238188982 2023-01-23 21:46:31.293695: step: 138/1541, loss: 0.0006590909324586391 2023-01-23 21:46:32.310465: step: 140/1541, loss: 0.0001236133393831551 2023-01-23 21:46:33.188517: step: 142/1541, loss: 0.0001678167755017057 2023-01-23 21:46:34.063046: step: 144/1541, loss: -7.097755769791547e-06 2023-01-23 21:46:34.976960: step: 146/1541, loss: 0.156316339969635 2023-01-23 21:46:35.851690: step: 148/1541, loss: 0.0004072432639077306 2023-01-23 21:46:36.736911: step: 150/1541, loss: 0.015650898218154907 2023-01-23 21:46:37.661803: step: 152/1541, loss: 4.462695505935699e-05 2023-01-23 21:46:38.565827: step: 154/1541, loss: 0.005916112568229437 2023-01-23 21:46:39.470989: step: 156/1541, loss: 0.007692730985581875 2023-01-23 21:46:40.387655: step: 158/1541, loss: 0.03449350222945213 2023-01-23 21:46:41.340122: step: 160/1541, loss: 0.00032897552591748536 2023-01-23 21:46:42.292561: step: 162/1541, loss: 0.08824287354946136 2023-01-23 21:46:43.176079: step: 164/1541, loss: 0.0012843706645071507 2023-01-23 21:46:44.075333: step: 166/1541, loss: 0.00019510037964209914 2023-01-23 21:46:45.028325: step: 168/1541, loss: 0.001372256432659924 2023-01-23 21:46:45.928926: step: 170/1541, loss: 0.0004960402147844434 2023-01-23 21:46:46.869287: step: 172/1541, loss: 0.11680440604686737 2023-01-23 21:46:47.742271: step: 174/1541, loss: 0.02181239239871502 2023-01-23 21:46:48.651220: step: 176/1541, loss: 0.03131614997982979 2023-01-23 21:46:49.572991: step: 178/1541, loss: 0.0002643604821059853 2023-01-23 21:46:50.448258: step: 180/1541, loss: 0.04029673710465431 2023-01-23 21:46:51.345376: step: 182/1541, loss: 0.001433409284800291 2023-01-23 21:46:52.264685: step: 184/1541, loss: 0.09455118328332901 2023-01-23 21:46:53.146972: step: 186/1541, loss: 0.0002109471388394013 2023-01-23 21:46:54.071427: step: 188/1541, loss: 0.00017150244093500078 2023-01-23 21:46:54.952006: step: 190/1541, loss: 0.0001597844820935279 2023-01-23 21:46:55.845130: step: 192/1541, loss: 2.2269177861744538e-05 2023-01-23 21:46:56.755353: step: 194/1541, loss: 0.0447319895029068 2023-01-23 21:46:57.786450: step: 196/1541, loss: 0.03459428995847702 2023-01-23 21:46:58.692659: step: 198/1541, loss: 0.0008920604013837874 2023-01-23 21:46:59.553004: step: 200/1541, loss: 0.03914058580994606 2023-01-23 21:47:00.479040: step: 202/1541, loss: 1.5641748177586123e-05 2023-01-23 21:47:01.477518: step: 204/1541, loss: 0.0007439710316248238 2023-01-23 21:47:02.319565: step: 206/1541, loss: 0.00036037713289260864 2023-01-23 21:47:03.276116: step: 208/1541, loss: 0.001961183501407504 2023-01-23 21:47:04.213618: step: 210/1541, loss: 0.0003097811422776431 2023-01-23 21:47:05.221412: step: 212/1541, loss: 0.2144431173801422 2023-01-23 21:47:06.101301: step: 214/1541, loss: 0.12056787312030792 2023-01-23 21:47:06.971858: step: 216/1541, loss: 0.00048306380631402135 2023-01-23 21:47:07.869961: step: 218/1541, loss: 0.00036703544901683927 2023-01-23 21:47:08.764993: step: 220/1541, loss: 0.43152132630348206 2023-01-23 21:47:09.684123: step: 222/1541, loss: 0.020580943673849106 2023-01-23 21:47:10.577556: step: 224/1541, loss: 0.05207117274403572 2023-01-23 21:47:11.465937: step: 226/1541, loss: -3.430009019211866e-05 2023-01-23 21:47:12.368870: step: 228/1541, loss: 2.2889176761964336e-05 2023-01-23 21:47:13.257202: step: 230/1541, loss: 6.219069473445415e-05 2023-01-23 21:47:14.255599: step: 232/1541, loss: 0.0027486695908010006 2023-01-23 21:47:15.115523: step: 234/1541, loss: 8.45121467136778e-05 2023-01-23 21:47:16.033371: step: 236/1541, loss: 0.00011194941180292517 2023-01-23 21:47:16.898888: step: 238/1541, loss: 0.2805124521255493 2023-01-23 21:47:17.806621: step: 240/1541, loss: 0.0699651911854744 2023-01-23 21:47:18.814714: step: 242/1541, loss: 0.013082695193588734 2023-01-23 21:47:19.734744: step: 244/1541, loss: 0.0007521912339143455 2023-01-23 21:47:20.641063: step: 246/1541, loss: 0.08294911682605743 2023-01-23 21:47:21.502680: step: 248/1541, loss: 0.09392952173948288 2023-01-23 21:47:22.474792: step: 250/1541, loss: 0.001031843712553382 2023-01-23 21:47:23.398578: step: 252/1541, loss: 9.385545854456723e-05 2023-01-23 21:47:24.316109: step: 254/1541, loss: 0.00023324397625401616 2023-01-23 21:47:25.215841: step: 256/1541, loss: 0.009847749955952168 2023-01-23 21:47:26.085138: step: 258/1541, loss: 8.074938705249224e-06 2023-01-23 21:47:26.992164: step: 260/1541, loss: 0.0007855340372771025 2023-01-23 21:47:27.864538: step: 262/1541, loss: 0.0004314780526328832 2023-01-23 21:47:28.728280: step: 264/1541, loss: 0.03572113439440727 2023-01-23 21:47:29.620052: step: 266/1541, loss: 0.2792757749557495 2023-01-23 21:47:30.564361: step: 268/1541, loss: 0.005808725953102112 2023-01-23 21:47:31.482488: step: 270/1541, loss: 0.00033393382909707725 2023-01-23 21:47:32.420921: step: 272/1541, loss: 0.0004072491137776524 2023-01-23 21:47:33.305536: step: 274/1541, loss: 0.00020290350948926061 2023-01-23 21:47:34.198859: step: 276/1541, loss: 0.10911814123392105 2023-01-23 21:47:35.071608: step: 278/1541, loss: 0.00032943737460300326 2023-01-23 21:47:35.995097: step: 280/1541, loss: 0.0001923395466292277 2023-01-23 21:47:36.907321: step: 282/1541, loss: 0.001427030423656106 2023-01-23 21:47:37.842018: step: 284/1541, loss: 0.9089934825897217 2023-01-23 21:47:38.707309: step: 286/1541, loss: 0.09454052150249481 2023-01-23 21:47:39.635926: step: 288/1541, loss: 0.0004455673915799707 2023-01-23 21:47:40.506333: step: 290/1541, loss: 4.9375739763490856e-05 2023-01-23 21:47:41.400257: step: 292/1541, loss: 0.003046320751309395 2023-01-23 21:47:42.343701: step: 294/1541, loss: 0.03604862466454506 2023-01-23 21:47:43.236782: step: 296/1541, loss: 0.0028020222671329975 2023-01-23 21:47:44.126962: step: 298/1541, loss: 0.001050814287737012 2023-01-23 21:47:45.016004: step: 300/1541, loss: 0.003297808114439249 2023-01-23 21:47:45.958368: step: 302/1541, loss: 0.07740121334791183 2023-01-23 21:47:46.840061: step: 304/1541, loss: 0.0008232265245169401 2023-01-23 21:47:47.743455: step: 306/1541, loss: 0.0002918252721428871 2023-01-23 21:47:48.631532: step: 308/1541, loss: 0.00024957762798294425 2023-01-23 21:47:49.548439: step: 310/1541, loss: 0.05421259626746178 2023-01-23 21:47:50.485795: step: 312/1541, loss: 0.0022864830680191517 2023-01-23 21:47:51.381711: step: 314/1541, loss: 0.011741068214178085 2023-01-23 21:47:52.263382: step: 316/1541, loss: 4.502087540458888e-05 2023-01-23 21:47:53.280822: step: 318/1541, loss: 0.24771134555339813 2023-01-23 21:47:54.164678: step: 320/1541, loss: 9.003431478049606e-05 2023-01-23 21:47:55.033977: step: 322/1541, loss: 0.03494542837142944 2023-01-23 21:47:55.953670: step: 324/1541, loss: 0.00021062881569378078 2023-01-23 21:47:56.892645: step: 326/1541, loss: 0.08313364535570145 2023-01-23 21:47:57.756737: step: 328/1541, loss: 0.00013383847544901073 2023-01-23 21:47:58.664970: step: 330/1541, loss: 0.6003732681274414 2023-01-23 21:47:59.540455: step: 332/1541, loss: 0.00013725654571317136 2023-01-23 21:48:00.442801: step: 334/1541, loss: 0.00019857088045682758 2023-01-23 21:48:01.368216: step: 336/1541, loss: 0.0016812452813610435 2023-01-23 21:48:02.270127: step: 338/1541, loss: 0.0002177858114009723 2023-01-23 21:48:03.149618: step: 340/1541, loss: 0.00872824341058731 2023-01-23 21:48:04.039256: step: 342/1541, loss: 0.0003993762074969709 2023-01-23 21:48:04.886564: step: 344/1541, loss: 0.001228643930517137 2023-01-23 21:48:05.873425: step: 346/1541, loss: 0.0019920975901186466 2023-01-23 21:48:06.750625: step: 348/1541, loss: 0.0022500259801745415 2023-01-23 21:48:07.605317: step: 350/1541, loss: 0.013402516022324562 2023-01-23 21:48:08.520989: step: 352/1541, loss: 0.08428235352039337 2023-01-23 21:48:09.426640: step: 354/1541, loss: 0.0038059642538428307 2023-01-23 21:48:10.343420: step: 356/1541, loss: 0.05313954874873161 2023-01-23 21:48:11.211184: step: 358/1541, loss: 0.055764228105545044 2023-01-23 21:48:12.107042: step: 360/1541, loss: 0.02828758955001831 2023-01-23 21:48:13.013565: step: 362/1541, loss: 0.000393692753277719 2023-01-23 21:48:13.905191: step: 364/1541, loss: 5.6460467021679506e-05 2023-01-23 21:48:14.772303: step: 366/1541, loss: 0.00010243197903037071 2023-01-23 21:48:15.665493: step: 368/1541, loss: 0.00020784384105354548 2023-01-23 21:48:16.623155: step: 370/1541, loss: 0.0037828958593308926 2023-01-23 21:48:17.554347: step: 372/1541, loss: 6.625075184274465e-05 2023-01-23 21:48:18.429268: step: 374/1541, loss: 0.09991299360990524 2023-01-23 21:48:19.304433: step: 376/1541, loss: 0.02958415076136589 2023-01-23 21:48:20.155380: step: 378/1541, loss: 0.00032432188163511455 2023-01-23 21:48:21.033301: step: 380/1541, loss: 0.002335391705855727 2023-01-23 21:48:21.886128: step: 382/1541, loss: 0.00031415472039952874 2023-01-23 21:48:22.790111: step: 384/1541, loss: 0.009489422664046288 2023-01-23 21:48:23.718146: step: 386/1541, loss: 0.049542251974344254 2023-01-23 21:48:24.576455: step: 388/1541, loss: 0.0010513365268707275 2023-01-23 21:48:25.486389: step: 390/1541, loss: 3.629177808761597e-05 2023-01-23 21:48:26.371344: step: 392/1541, loss: 0.0008574301027692854 2023-01-23 21:48:27.221278: step: 394/1541, loss: 0.00015274879115168005 2023-01-23 21:48:28.133219: step: 396/1541, loss: -1.0133365435649466e-07 2023-01-23 21:48:29.065311: step: 398/1541, loss: 0.005844548810273409 2023-01-23 21:48:29.911770: step: 400/1541, loss: 1.3149140613677446e-05 2023-01-23 21:48:30.868175: step: 402/1541, loss: 0.005189054179936647 2023-01-23 21:48:31.751049: step: 404/1541, loss: 0.0029459947254508734 2023-01-23 21:48:32.697763: step: 406/1541, loss: 3.0625462386524305e-05 2023-01-23 21:48:33.564194: step: 408/1541, loss: 0.023101646453142166 2023-01-23 21:48:34.494782: step: 410/1541, loss: -9.354971552966163e-06 2023-01-23 21:48:35.405308: step: 412/1541, loss: 0.08165550231933594 2023-01-23 21:48:36.265678: step: 414/1541, loss: 0.027312535792589188 2023-01-23 21:48:37.167246: step: 416/1541, loss: 0.019808974117040634 2023-01-23 21:48:38.006286: step: 418/1541, loss: 0.03002863936126232 2023-01-23 21:48:38.945781: step: 420/1541, loss: 0.00016833750123623759 2023-01-23 21:48:39.847190: step: 422/1541, loss: 0.5611726641654968 2023-01-23 21:48:40.809501: step: 424/1541, loss: 0.0003042106400243938 2023-01-23 21:48:41.710639: step: 426/1541, loss: 0.013106931000947952 2023-01-23 21:48:42.596472: step: 428/1541, loss: 0.1267240047454834 2023-01-23 21:48:43.487544: step: 430/1541, loss: 3.619136623456143e-05 2023-01-23 21:48:44.426058: step: 432/1541, loss: 5.0736220146063715e-05 2023-01-23 21:48:45.395830: step: 434/1541, loss: 0.026677044108510017 2023-01-23 21:48:46.268292: step: 436/1541, loss: 0.00018397685198578984 2023-01-23 21:48:47.147233: step: 438/1541, loss: 0.11028922349214554 2023-01-23 21:48:48.041025: step: 440/1541, loss: 0.000741193478461355 2023-01-23 21:48:48.939259: step: 442/1541, loss: 0.002847434487193823 2023-01-23 21:48:49.892737: step: 444/1541, loss: 0.0004796838038600981 2023-01-23 21:48:50.777569: step: 446/1541, loss: 0.0005237905425019562 2023-01-23 21:48:51.650163: step: 448/1541, loss: 0.0004700816934928298 2023-01-23 21:48:52.550048: step: 450/1541, loss: 0.05244027450680733 2023-01-23 21:48:53.472577: step: 452/1541, loss: 0.03736425191164017 2023-01-23 21:48:54.381947: step: 454/1541, loss: 0.00030756741762161255 2023-01-23 21:48:55.264604: step: 456/1541, loss: 0.00035247800406068563 2023-01-23 21:48:56.170656: step: 458/1541, loss: 0.004795540124177933 2023-01-23 21:48:57.043810: step: 460/1541, loss: 0.0006295350613072515 2023-01-23 21:48:57.933902: step: 462/1541, loss: 6.423511513276026e-05 2023-01-23 21:48:58.897073: step: 464/1541, loss: 0.028750231489539146 2023-01-23 21:48:59.769844: step: 466/1541, loss: 0.0007509369752369821 2023-01-23 21:49:00.680679: step: 468/1541, loss: 1.6864219903945923 2023-01-23 21:49:01.599399: step: 470/1541, loss: 3.0277666155598126e-05 2023-01-23 21:49:02.503721: step: 472/1541, loss: 0.014566791243851185 2023-01-23 21:49:03.424832: step: 474/1541, loss: 9.862814476946369e-05 2023-01-23 21:49:04.338631: step: 476/1541, loss: 1.3286709785461426 2023-01-23 21:49:05.238486: step: 478/1541, loss: 0.00017335114534944296 2023-01-23 21:49:06.198748: step: 480/1541, loss: 0.4707429111003876 2023-01-23 21:49:07.052290: step: 482/1541, loss: 0.01015438511967659 2023-01-23 21:49:07.949119: step: 484/1541, loss: 0.03334207087755203 2023-01-23 21:49:08.828089: step: 486/1541, loss: 0.08657515048980713 2023-01-23 21:49:09.708558: step: 488/1541, loss: 0.13740625977516174 2023-01-23 21:49:10.584995: step: 490/1541, loss: 0.00030946286278776824 2023-01-23 21:49:11.456882: step: 492/1541, loss: 0.0011834419565275311 2023-01-23 21:49:12.288964: step: 494/1541, loss: 0.0064265308901667595 2023-01-23 21:49:13.135755: step: 496/1541, loss: 0.0020670692902058363 2023-01-23 21:49:14.052502: step: 498/1541, loss: 0.0002923791471403092 2023-01-23 21:49:14.906594: step: 500/1541, loss: 0.0021483656018972397 2023-01-23 21:49:15.770039: step: 502/1541, loss: 0.0033134191762655973 2023-01-23 21:49:16.627855: step: 504/1541, loss: 0.1097782626748085 2023-01-23 21:49:17.522671: step: 506/1541, loss: 0.0004042654181830585 2023-01-23 21:49:18.412218: step: 508/1541, loss: 0.0004339792940299958 2023-01-23 21:49:19.266743: step: 510/1541, loss: 0.003971070051193237 2023-01-23 21:49:20.120837: step: 512/1541, loss: 0.0003640229406300932 2023-01-23 21:49:21.026246: step: 514/1541, loss: 0.000157417482114397 2023-01-23 21:49:21.986594: step: 516/1541, loss: 0.0022905298974364996 2023-01-23 21:49:22.877107: step: 518/1541, loss: 0.0014219727599993348 2023-01-23 21:49:23.777566: step: 520/1541, loss: 0.0027998564764857292 2023-01-23 21:49:24.660267: step: 522/1541, loss: 0.13110759854316711 2023-01-23 21:49:25.564583: step: 524/1541, loss: 0.06559653580188751 2023-01-23 21:49:26.461748: step: 526/1541, loss: 0.00022791426454205066 2023-01-23 21:49:27.332086: step: 528/1541, loss: 0.0005527040339075029 2023-01-23 21:49:28.200446: step: 530/1541, loss: 0.0003657893103081733 2023-01-23 21:49:29.076977: step: 532/1541, loss: 0.1470743864774704 2023-01-23 21:49:30.005897: step: 534/1541, loss: 0.12378004938364029 2023-01-23 21:49:30.929730: step: 536/1541, loss: 0.0003477476420812309 2023-01-23 21:49:31.793283: step: 538/1541, loss: 0.0009477721760049462 2023-01-23 21:49:32.666572: step: 540/1541, loss: 0.05406498163938522 2023-01-23 21:49:33.604761: step: 542/1541, loss: 0.014007355086505413 2023-01-23 21:49:34.563670: step: 544/1541, loss: 0.000323703046888113 2023-01-23 21:49:35.473070: step: 546/1541, loss: 0.0005730192642658949 2023-01-23 21:49:36.426411: step: 548/1541, loss: 0.00016063546354416758 2023-01-23 21:49:37.317196: step: 550/1541, loss: 1.078701376914978 2023-01-23 21:49:38.262611: step: 552/1541, loss: 0.0001599840761628002 2023-01-23 21:49:39.144015: step: 554/1541, loss: 1.9121887817163952e-05 2023-01-23 21:49:40.046528: step: 556/1541, loss: 0.04738171398639679 2023-01-23 21:49:40.950843: step: 558/1541, loss: 3.130215554847382e-05 2023-01-23 21:49:41.894519: step: 560/1541, loss: 0.02998659387230873 2023-01-23 21:49:42.823465: step: 562/1541, loss: 0.17487922310829163 2023-01-23 21:49:43.752715: step: 564/1541, loss: 0.11244629323482513 2023-01-23 21:49:44.641402: step: 566/1541, loss: 0.19414956867694855 2023-01-23 21:49:45.531616: step: 568/1541, loss: 0.004230341408401728 2023-01-23 21:49:46.413912: step: 570/1541, loss: 0.0002880171814467758 2023-01-23 21:49:47.388931: step: 572/1541, loss: 0.000840328517369926 2023-01-23 21:49:48.274746: step: 574/1541, loss: 0.007516920566558838 2023-01-23 21:49:49.145127: step: 576/1541, loss: 0.007790668867528439 2023-01-23 21:49:50.037323: step: 578/1541, loss: 0.00031798001145944 2023-01-23 21:49:50.952313: step: 580/1541, loss: 0.19503627717494965 2023-01-23 21:49:51.839245: step: 582/1541, loss: 0.0013885637745261192 2023-01-23 21:49:52.741667: step: 584/1541, loss: 0.08975837379693985 2023-01-23 21:49:53.698176: step: 586/1541, loss: 0.00024338164075743407 2023-01-23 21:49:54.553251: step: 588/1541, loss: 0.015340018086135387 2023-01-23 21:49:55.435159: step: 590/1541, loss: 0.10341282933950424 2023-01-23 21:49:56.322348: step: 592/1541, loss: 0.04500965029001236 2023-01-23 21:49:57.239602: step: 594/1541, loss: 0.00010938661580439657 2023-01-23 21:49:58.199921: step: 596/1541, loss: 0.008661310188472271 2023-01-23 21:49:59.138413: step: 598/1541, loss: 0.09025884419679642 2023-01-23 21:50:00.060792: step: 600/1541, loss: 0.08076345920562744 2023-01-23 21:50:01.008107: step: 602/1541, loss: 0.00021656454191543162 2023-01-23 21:50:01.934875: step: 604/1541, loss: 0.07340960949659348 2023-01-23 21:50:02.855871: step: 606/1541, loss: 0.001161634223535657 2023-01-23 21:50:03.729474: step: 608/1541, loss: 0.008326378650963306 2023-01-23 21:50:04.654690: step: 610/1541, loss: 0.0004481359792407602 2023-01-23 21:50:05.545691: step: 612/1541, loss: 0.00023310307005885988 2023-01-23 21:50:06.418563: step: 614/1541, loss: 0.0004757724527735263 2023-01-23 21:50:07.301499: step: 616/1541, loss: 0.00022386830823961645 2023-01-23 21:50:08.218798: step: 618/1541, loss: 0.0001074258834705688 2023-01-23 21:50:09.094419: step: 620/1541, loss: 0.0005142443114891648 2023-01-23 21:50:09.996693: step: 622/1541, loss: 0.6814780831336975 2023-01-23 21:50:10.860942: step: 624/1541, loss: 0.0011415202170610428 2023-01-23 21:50:11.741676: step: 626/1541, loss: 0.0028491010889410973 2023-01-23 21:50:12.652908: step: 628/1541, loss: 0.0015815987717360258 2023-01-23 21:50:13.564562: step: 630/1541, loss: 0.07934755086898804 2023-01-23 21:50:14.467042: step: 632/1541, loss: 7.534059113822877e-05 2023-01-23 21:50:15.361075: step: 634/1541, loss: 6.948358350200579e-05 2023-01-23 21:50:16.262447: step: 636/1541, loss: 0.02870023436844349 2023-01-23 21:50:17.124720: step: 638/1541, loss: 0.04253317788243294 2023-01-23 21:50:17.998113: step: 640/1541, loss: 0.00142385414801538 2023-01-23 21:50:18.945644: step: 642/1541, loss: 0.14672517776489258 2023-01-23 21:50:19.839716: step: 644/1541, loss: 0.3192439377307892 2023-01-23 21:50:20.738240: step: 646/1541, loss: 0.06463271379470825 2023-01-23 21:50:21.619018: step: 648/1541, loss: 0.0009139962494373322 2023-01-23 21:50:22.515499: step: 650/1541, loss: 0.04631732404232025 2023-01-23 21:50:23.459972: step: 652/1541, loss: 0.000901348830666393 2023-01-23 21:50:24.374797: step: 654/1541, loss: 0.00267788372002542 2023-01-23 21:50:25.292104: step: 656/1541, loss: 0.0001196322264149785 2023-01-23 21:50:26.207652: step: 658/1541, loss: 0.0004568012373056263 2023-01-23 21:50:27.081556: step: 660/1541, loss: 6.838906847406179e-05 2023-01-23 21:50:28.011918: step: 662/1541, loss: 0.0006728217122144997 2023-01-23 21:50:28.873323: step: 664/1541, loss: 0.11934168636798859 2023-01-23 21:50:29.723653: step: 666/1541, loss: 0.00021992699475958943 2023-01-23 21:50:30.603978: step: 668/1541, loss: 1.1774393320083618 2023-01-23 21:50:31.498599: step: 670/1541, loss: 0.00020248891087248921 2023-01-23 21:50:32.400545: step: 672/1541, loss: 0.024057794362306595 2023-01-23 21:50:33.283711: step: 674/1541, loss: 0.0017637351993471384 2023-01-23 21:50:34.209586: step: 676/1541, loss: 0.0002554072707425803 2023-01-23 21:50:35.123356: step: 678/1541, loss: 0.0008207054925151169 2023-01-23 21:50:35.984335: step: 680/1541, loss: 0.009636455215513706 2023-01-23 21:50:36.893854: step: 682/1541, loss: 0.0008009279263205826 2023-01-23 21:50:37.838078: step: 684/1541, loss: 0.0006232544546946883 2023-01-23 21:50:38.728146: step: 686/1541, loss: 0.0003936644352506846 2023-01-23 21:50:39.638383: step: 688/1541, loss: 0.0002196198038291186 2023-01-23 21:50:40.535713: step: 690/1541, loss: 3.895870395353995e-05 2023-01-23 21:50:41.463181: step: 692/1541, loss: 0.0824686735868454 2023-01-23 21:50:42.414474: step: 694/1541, loss: 0.17341484129428864 2023-01-23 21:50:43.334386: step: 696/1541, loss: 0.000795068743173033 2023-01-23 21:50:44.242904: step: 698/1541, loss: 4.202065974823199e-05 2023-01-23 21:50:45.107393: step: 700/1541, loss: 0.0006166882230900228 2023-01-23 21:50:45.991396: step: 702/1541, loss: 0.0017277784645557404 2023-01-23 21:50:46.998177: step: 704/1541, loss: 0.058423466980457306 2023-01-23 21:50:47.869932: step: 706/1541, loss: 0.0005131723592057824 2023-01-23 21:50:48.722373: step: 708/1541, loss: 0.0014534363290295005 2023-01-23 21:50:49.574255: step: 710/1541, loss: 0.00047778314910829067 2023-01-23 21:50:50.463462: step: 712/1541, loss: 9.650314495956991e-06 2023-01-23 21:50:51.384255: step: 714/1541, loss: 6.145656516309828e-05 2023-01-23 21:50:52.361764: step: 716/1541, loss: 0.00044023693772032857 2023-01-23 21:50:53.258991: step: 718/1541, loss: 0.07123418897390366 2023-01-23 21:50:54.118007: step: 720/1541, loss: 3.3424239518353716e-05 2023-01-23 21:50:55.018869: step: 722/1541, loss: 0.001392936334013939 2023-01-23 21:50:55.921833: step: 724/1541, loss: 0.00010759061115095392 2023-01-23 21:50:56.826009: step: 726/1541, loss: 0.00021924078464508057 2023-01-23 21:50:57.756657: step: 728/1541, loss: 0.00028094646404497325 2023-01-23 21:50:58.655974: step: 730/1541, loss: 0.04706385359168053 2023-01-23 21:50:59.523453: step: 732/1541, loss: 0.01818680204451084 2023-01-23 21:51:00.425661: step: 734/1541, loss: 0.025965385138988495 2023-01-23 21:51:01.301791: step: 736/1541, loss: 0.00047900265781208873 2023-01-23 21:51:02.223995: step: 738/1541, loss: 0.25405746698379517 2023-01-23 21:51:03.150248: step: 740/1541, loss: 0.00034578220220282674 2023-01-23 21:51:04.040344: step: 742/1541, loss: 0.0104064317420125 2023-01-23 21:51:04.923739: step: 744/1541, loss: 0.6351834535598755 2023-01-23 21:51:05.856699: step: 746/1541, loss: 0.014663567766547203 2023-01-23 21:51:06.773301: step: 748/1541, loss: 0.00032205163734033704 2023-01-23 21:51:07.645428: step: 750/1541, loss: 0.06989441812038422 2023-01-23 21:51:08.532549: step: 752/1541, loss: 0.08802127093076706 2023-01-23 21:51:09.413397: step: 754/1541, loss: 0.015212266705930233 2023-01-23 21:51:10.296402: step: 756/1541, loss: 4.144897684454918e-05 2023-01-23 21:51:11.152409: step: 758/1541, loss: 0.0013780189910903573 2023-01-23 21:51:12.013544: step: 760/1541, loss: 0.0005226157954894006 2023-01-23 21:51:12.896870: step: 762/1541, loss: 0.010305525735020638 2023-01-23 21:51:13.819315: step: 764/1541, loss: 0.0001157551960204728 2023-01-23 21:51:14.718836: step: 766/1541, loss: 0.00025126629043370485 2023-01-23 21:51:15.594134: step: 768/1541, loss: 0.0005468703457154334 2023-01-23 21:51:16.511468: step: 770/1541, loss: 0.13441090285778046 2023-01-23 21:51:17.396683: step: 772/1541, loss: 0.0001399201137246564 2023-01-23 21:51:18.256354: step: 774/1541, loss: 0.00034228709409944713 2023-01-23 21:51:19.172601: step: 776/1541, loss: 0.00033156084828078747 2023-01-23 21:51:20.019774: step: 778/1541, loss: 0.000744435063097626 2023-01-23 21:51:20.936169: step: 780/1541, loss: 0.001347607932984829 2023-01-23 21:51:21.838748: step: 782/1541, loss: 0.0003734548226930201 2023-01-23 21:51:22.728787: step: 784/1541, loss: 0.0005352504667825997 2023-01-23 21:51:23.628888: step: 786/1541, loss: 0.00011435974738560617 2023-01-23 21:51:24.517990: step: 788/1541, loss: 0.03433222323656082 2023-01-23 21:51:25.390288: step: 790/1541, loss: 0.003320933785289526 2023-01-23 21:51:26.358564: step: 792/1541, loss: 0.0008897141669876873 2023-01-23 21:51:27.324711: step: 794/1541, loss: 0.101331926882267 2023-01-23 21:51:28.232408: step: 796/1541, loss: 8.168369095074013e-05 2023-01-23 21:51:29.117676: step: 798/1541, loss: 0.0008467077277600765 2023-01-23 21:51:29.990626: step: 800/1541, loss: 0.0027116730343550444 2023-01-23 21:51:30.862608: step: 802/1541, loss: 0.0002328643895452842 2023-01-23 21:51:31.742436: step: 804/1541, loss: 0.008792913518846035 2023-01-23 21:51:32.692653: step: 806/1541, loss: 0.10674396902322769 2023-01-23 21:51:33.605151: step: 808/1541, loss: 0.004853473976254463 2023-01-23 21:51:34.491324: step: 810/1541, loss: 0.029320891946554184 2023-01-23 21:51:35.378901: step: 812/1541, loss: 4.355882265372202e-05 2023-01-23 21:51:36.269435: step: 814/1541, loss: 0.0003650221333373338 2023-01-23 21:51:37.163457: step: 816/1541, loss: 0.13763749599456787 2023-01-23 21:51:38.042102: step: 818/1541, loss: 0.3116951882839203 2023-01-23 21:51:38.918811: step: 820/1541, loss: 0.00040108978282660246 2023-01-23 21:51:39.790280: step: 822/1541, loss: 3.261242818553001e-05 2023-01-23 21:51:40.701286: step: 824/1541, loss: 0.0011912602931261063 2023-01-23 21:51:41.599480: step: 826/1541, loss: 0.0010978046339005232 2023-01-23 21:51:42.528550: step: 828/1541, loss: 0.4712655544281006 2023-01-23 21:51:43.398853: step: 830/1541, loss: 0.39122748374938965 2023-01-23 21:51:44.347692: step: 832/1541, loss: 2.172197103500366 2023-01-23 21:51:45.266524: step: 834/1541, loss: 5.723157755710417e-06 2023-01-23 21:51:46.142197: step: 836/1541, loss: 0.05340223014354706 2023-01-23 21:51:47.048162: step: 838/1541, loss: 0.001634308253414929 2023-01-23 21:51:47.925713: step: 840/1541, loss: 0.23517712950706482 2023-01-23 21:51:48.823921: step: 842/1541, loss: 0.09137792140245438 2023-01-23 21:51:49.706880: step: 844/1541, loss: 0.003965805750340223 2023-01-23 21:51:50.591987: step: 846/1541, loss: 0.020447751507163048 2023-01-23 21:51:51.518404: step: 848/1541, loss: 0.016809973865747452 2023-01-23 21:51:52.411430: step: 850/1541, loss: 0.056688349694013596 2023-01-23 21:51:53.284825: step: 852/1541, loss: 0.018250057473778725 2023-01-23 21:51:54.144794: step: 854/1541, loss: 0.0047908322885632515 2023-01-23 21:51:55.065565: step: 856/1541, loss: 0.0025515311863273382 2023-01-23 21:51:55.919044: step: 858/1541, loss: 0.0012283166870474815 2023-01-23 21:51:56.899266: step: 860/1541, loss: 0.455655574798584 2023-01-23 21:51:57.889960: step: 862/1541, loss: 0.0048952409997582436 2023-01-23 21:51:58.791658: step: 864/1541, loss: 0.06701548397541046 2023-01-23 21:51:59.663593: step: 866/1541, loss: 0.3829469680786133 2023-01-23 21:52:00.634068: step: 868/1541, loss: 0.0006747336592525244 2023-01-23 21:52:01.655664: step: 870/1541, loss: 0.2740583121776581 2023-01-23 21:52:02.567317: step: 872/1541, loss: 0.0009707877179607749 2023-01-23 21:52:03.473507: step: 874/1541, loss: 3.2799591281218454e-05 2023-01-23 21:52:04.337825: step: 876/1541, loss: 0.0015069436049088836 2023-01-23 21:52:05.345971: step: 878/1541, loss: 8.556028478778899e-05 2023-01-23 21:52:06.215804: step: 880/1541, loss: 0.3591572344303131 2023-01-23 21:52:07.066317: step: 882/1541, loss: 0.0009422270813956857 2023-01-23 21:52:07.949043: step: 884/1541, loss: 0.017849057912826538 2023-01-23 21:52:08.932638: step: 886/1541, loss: 0.00221505225636065 2023-01-23 21:52:09.821416: step: 888/1541, loss: 0.10558422654867172 2023-01-23 21:52:10.686912: step: 890/1541, loss: 0.1489919275045395 2023-01-23 21:52:11.579878: step: 892/1541, loss: 0.05942004173994064 2023-01-23 21:52:12.458461: step: 894/1541, loss: 0.057167161256074905 2023-01-23 21:52:13.316068: step: 896/1541, loss: 0.01919388212263584 2023-01-23 21:52:14.185065: step: 898/1541, loss: 0.2455124408006668 2023-01-23 21:52:15.040742: step: 900/1541, loss: 0.002440402517095208 2023-01-23 21:52:15.902019: step: 902/1541, loss: 0.02783210203051567 2023-01-23 21:52:16.759498: step: 904/1541, loss: 0.08794393390417099 2023-01-23 21:52:17.646305: step: 906/1541, loss: 0.006486113183200359 2023-01-23 21:52:18.522433: step: 908/1541, loss: 0.0021999464370310307 2023-01-23 21:52:19.366751: step: 910/1541, loss: 0.0009152228012681007 2023-01-23 21:52:20.271803: step: 912/1541, loss: 0.0004652442003134638 2023-01-23 21:52:21.161686: step: 914/1541, loss: 0.0015126116340979934 2023-01-23 21:52:22.112979: step: 916/1541, loss: 0.0010488488478586078 2023-01-23 21:52:23.007151: step: 918/1541, loss: 0.09706959128379822 2023-01-23 21:52:23.877938: step: 920/1541, loss: 0.16675378382205963 2023-01-23 21:52:24.814397: step: 922/1541, loss: 0.0012980238534510136 2023-01-23 21:52:25.734565: step: 924/1541, loss: 0.0009149761172011495 2023-01-23 21:52:26.661279: step: 926/1541, loss: 0.0013760756701231003 2023-01-23 21:52:27.514807: step: 928/1541, loss: 0.005495487712323666 2023-01-23 21:52:28.408419: step: 930/1541, loss: 0.020802486687898636 2023-01-23 21:52:29.314367: step: 932/1541, loss: 0.020132534205913544 2023-01-23 21:52:30.178122: step: 934/1541, loss: 0.00027342617977410555 2023-01-23 21:52:31.055308: step: 936/1541, loss: 0.001312194624915719 2023-01-23 21:52:31.964619: step: 938/1541, loss: 0.00025889230892062187 2023-01-23 21:52:32.919247: step: 940/1541, loss: 0.022694310173392296 2023-01-23 21:52:33.815530: step: 942/1541, loss: 2.3692780814599246e-05 2023-01-23 21:52:34.662919: step: 944/1541, loss: 1.760836676112376e-05 2023-01-23 21:52:35.592035: step: 946/1541, loss: 0.010748923756182194 2023-01-23 21:52:36.477539: step: 948/1541, loss: 0.025254571810364723 2023-01-23 21:52:37.366544: step: 950/1541, loss: 0.8239190578460693 2023-01-23 21:52:38.273900: step: 952/1541, loss: 0.05134524777531624 2023-01-23 21:52:39.174180: step: 954/1541, loss: 0.00792626105248928 2023-01-23 21:52:40.080002: step: 956/1541, loss: 6.578719330718741e-05 2023-01-23 21:52:41.004855: step: 958/1541, loss: 0.0013645467115566134 2023-01-23 21:52:42.014988: step: 960/1541, loss: 0.00015067309141159058 2023-01-23 21:52:42.887478: step: 962/1541, loss: 0.0032934679184108973 2023-01-23 21:52:43.832818: step: 964/1541, loss: 0.05235534533858299 2023-01-23 21:52:44.689768: step: 966/1541, loss: 0.36298230290412903 2023-01-23 21:52:45.614008: step: 968/1541, loss: 0.26943573355674744 2023-01-23 21:52:46.507848: step: 970/1541, loss: 0.062499091029167175 2023-01-23 21:52:47.377090: step: 972/1541, loss: 0.00022890330001246184 2023-01-23 21:52:48.274372: step: 974/1541, loss: 0.0008938861428759992 2023-01-23 21:52:49.138680: step: 976/1541, loss: 6.648131238762289e-05 2023-01-23 21:52:50.047420: step: 978/1541, loss: 0.4349713921546936 2023-01-23 21:52:50.914024: step: 980/1541, loss: 0.014087274670600891 2023-01-23 21:52:51.783151: step: 982/1541, loss: 0.01679684780538082 2023-01-23 21:52:52.646953: step: 984/1541, loss: 0.0008292710990644991 2023-01-23 21:52:53.501073: step: 986/1541, loss: 3.0463372240774333e-05 2023-01-23 21:52:54.368047: step: 988/1541, loss: 0.00011290689144516364 2023-01-23 21:52:55.285889: step: 990/1541, loss: 0.1287219226360321 2023-01-23 21:52:56.195210: step: 992/1541, loss: 0.0001439472835045308 2023-01-23 21:52:57.099759: step: 994/1541, loss: 0.023404214531183243 2023-01-23 21:52:58.003876: step: 996/1541, loss: 0.029245279729366302 2023-01-23 21:52:58.889700: step: 998/1541, loss: 0.03182407096028328 2023-01-23 21:52:59.793703: step: 1000/1541, loss: 0.20179405808448792 2023-01-23 21:53:00.675423: step: 1002/1541, loss: 0.006342434789985418 2023-01-23 21:53:01.601734: step: 1004/1541, loss: 0.12254463136196136 2023-01-23 21:53:02.473442: step: 1006/1541, loss: 0.0021260154899209738 2023-01-23 21:53:03.375086: step: 1008/1541, loss: 1.1453529623395298e-05 2023-01-23 21:53:04.351706: step: 1010/1541, loss: 0.060150425881147385 2023-01-23 21:53:05.242851: step: 1012/1541, loss: 0.23721711337566376 2023-01-23 21:53:06.114938: step: 1014/1541, loss: 0.05937950313091278 2023-01-23 21:53:06.975642: step: 1016/1541, loss: 7.285355968633667e-05 2023-01-23 21:53:07.903866: step: 1018/1541, loss: 5.41783265362028e-05 2023-01-23 21:53:08.783113: step: 1020/1541, loss: 7.77238528826274e-05 2023-01-23 21:53:09.680762: step: 1022/1541, loss: 0.00046561891213059425 2023-01-23 21:53:10.560311: step: 1024/1541, loss: 0.0009181065834127367 2023-01-23 21:53:11.459377: step: 1026/1541, loss: 0.25315961241722107 2023-01-23 21:53:12.348925: step: 1028/1541, loss: 0.41076749563217163 2023-01-23 21:53:13.212777: step: 1030/1541, loss: 6.915305857546628e-05 2023-01-23 21:53:14.146716: step: 1032/1541, loss: 0.0001019525807350874 2023-01-23 21:53:15.049761: step: 1034/1541, loss: 0.008008907549083233 2023-01-23 21:53:15.929419: step: 1036/1541, loss: 0.11250178515911102 2023-01-23 21:53:16.841412: step: 1038/1541, loss: 0.00028603526880033314 2023-01-23 21:53:17.757769: step: 1040/1541, loss: 0.013221762143075466 2023-01-23 21:53:18.724998: step: 1042/1541, loss: 0.0872233659029007 2023-01-23 21:53:19.609619: step: 1044/1541, loss: 0.008031466975808144 2023-01-23 21:53:20.483963: step: 1046/1541, loss: 0.0007717724656686187 2023-01-23 21:53:21.351830: step: 1048/1541, loss: 0.051204584538936615 2023-01-23 21:53:22.244988: step: 1050/1541, loss: 0.004267062991857529 2023-01-23 21:53:23.151164: step: 1052/1541, loss: 0.09437382221221924 2023-01-23 21:53:24.064449: step: 1054/1541, loss: 0.02063961699604988 2023-01-23 21:53:25.027627: step: 1056/1541, loss: 0.18653102219104767 2023-01-23 21:53:25.934136: step: 1058/1541, loss: 0.004291858524084091 2023-01-23 21:53:26.849340: step: 1060/1541, loss: 0.0035192754585295916 2023-01-23 21:53:27.789532: step: 1062/1541, loss: 0.0013589835725724697 2023-01-23 21:53:28.680616: step: 1064/1541, loss: 0.007373508997261524 2023-01-23 21:53:29.567190: step: 1066/1541, loss: 4.605796129908413e-05 2023-01-23 21:53:30.445715: step: 1068/1541, loss: 0.08459728211164474 2023-01-23 21:53:31.297375: step: 1070/1541, loss: 0.0025873754639178514 2023-01-23 21:53:32.215677: step: 1072/1541, loss: 0.3805241882801056 2023-01-23 21:53:33.103191: step: 1074/1541, loss: 2.7312900783726946e-05 2023-01-23 21:53:33.963022: step: 1076/1541, loss: 0.08284791558980942 2023-01-23 21:53:34.846951: step: 1078/1541, loss: 0.15513144433498383 2023-01-23 21:53:35.765484: step: 1080/1541, loss: -1.102255737350788e-05 2023-01-23 21:53:36.720666: step: 1082/1541, loss: 0.4086682200431824 2023-01-23 21:53:37.606947: step: 1084/1541, loss: 0.007312471978366375 2023-01-23 21:53:38.493897: step: 1086/1541, loss: 0.6748341917991638 2023-01-23 21:53:39.430149: step: 1088/1541, loss: 0.0035340336617082357 2023-01-23 21:53:40.284603: step: 1090/1541, loss: 7.408038800349459e-05 2023-01-23 21:53:41.221929: step: 1092/1541, loss: 0.00030749241705052555 2023-01-23 21:53:42.104192: step: 1094/1541, loss: 0.06378843635320663 2023-01-23 21:53:42.959769: step: 1096/1541, loss: 0.00014175500837154686 2023-01-23 21:53:43.902915: step: 1098/1541, loss: 0.0004183832788839936 2023-01-23 21:53:44.762870: step: 1100/1541, loss: 0.0001267808402189985 2023-01-23 21:53:45.637091: step: 1102/1541, loss: 0.03708496317267418 2023-01-23 21:53:46.510084: step: 1104/1541, loss: 0.09779825061559677 2023-01-23 21:53:47.498539: step: 1106/1541, loss: 0.001237946329638362 2023-01-23 21:53:48.465467: step: 1108/1541, loss: 0.011753721162676811 2023-01-23 21:53:49.373321: step: 1110/1541, loss: 0.1946430504322052 2023-01-23 21:53:50.261386: step: 1112/1541, loss: 0.008752661757171154 2023-01-23 21:53:51.213352: step: 1114/1541, loss: 0.0074776215478777885 2023-01-23 21:53:52.086254: step: 1116/1541, loss: 0.00810373667627573 2023-01-23 21:53:53.047740: step: 1118/1541, loss: 0.00016849266830831766 2023-01-23 21:53:53.934782: step: 1120/1541, loss: 0.1197756826877594 2023-01-23 21:53:54.809713: step: 1122/1541, loss: 0.005175714381039143 2023-01-23 21:53:55.772857: step: 1124/1541, loss: 0.15995824337005615 2023-01-23 21:53:56.711005: step: 1126/1541, loss: 0.0006824735319241881 2023-01-23 21:53:57.612826: step: 1128/1541, loss: 0.0022396736312657595 2023-01-23 21:53:58.466974: step: 1130/1541, loss: 0.03352467343211174 2023-01-23 21:53:59.351050: step: 1132/1541, loss: 0.0012940102024003863 2023-01-23 21:54:00.236933: step: 1134/1541, loss: 0.06360381096601486 2023-01-23 21:54:01.176483: step: 1136/1541, loss: 0.0016171603929251432 2023-01-23 21:54:02.097038: step: 1138/1541, loss: 0.0020095864310860634 2023-01-23 21:54:02.985696: step: 1140/1541, loss: 0.013811911456286907 2023-01-23 21:54:03.947914: step: 1142/1541, loss: 0.0002770675055216998 2023-01-23 21:54:04.819230: step: 1144/1541, loss: 0.43117597699165344 2023-01-23 21:54:05.739192: step: 1146/1541, loss: 0.08395009487867355 2023-01-23 21:54:06.627979: step: 1148/1541, loss: 0.008302765898406506 2023-01-23 21:54:07.597890: step: 1150/1541, loss: 0.0535389706492424 2023-01-23 21:54:08.438253: step: 1152/1541, loss: 0.10708393901586533 2023-01-23 21:54:09.340370: step: 1154/1541, loss: 0.0003108680248260498 2023-01-23 21:54:10.210051: step: 1156/1541, loss: 0.0008719006436876953 2023-01-23 21:54:11.100120: step: 1158/1541, loss: 0.0011761116329580545 2023-01-23 21:54:11.946939: step: 1160/1541, loss: 0.02726726047694683 2023-01-23 21:54:12.868155: step: 1162/1541, loss: 0.019565481692552567 2023-01-23 21:54:13.796590: step: 1164/1541, loss: 0.0021024360321462154 2023-01-23 21:54:14.684607: step: 1166/1541, loss: 0.06443029642105103 2023-01-23 21:54:15.574697: step: 1168/1541, loss: 0.06646610796451569 2023-01-23 21:54:16.467106: step: 1170/1541, loss: 0.0002337647893000394 2023-01-23 21:54:17.339887: step: 1172/1541, loss: 0.02690839394927025 2023-01-23 21:54:18.237022: step: 1174/1541, loss: 0.00016608637815807015 2023-01-23 21:54:19.108077: step: 1176/1541, loss: 0.01071836519986391 2023-01-23 21:54:20.039920: step: 1178/1541, loss: 1.272688627243042 2023-01-23 21:54:20.904262: step: 1180/1541, loss: 0.0022163530811667442 2023-01-23 21:54:21.802769: step: 1182/1541, loss: 0.06216079741716385 2023-01-23 21:54:22.735975: step: 1184/1541, loss: 0.02752615325152874 2023-01-23 21:54:23.659700: step: 1186/1541, loss: 0.0067029427736997604 2023-01-23 21:54:24.552191: step: 1188/1541, loss: 0.13760598003864288 2023-01-23 21:54:25.432542: step: 1190/1541, loss: 0.0011358254123479128 2023-01-23 21:54:26.421179: step: 1192/1541, loss: 0.0014621878508478403 2023-01-23 21:54:27.280217: step: 1194/1541, loss: 0.010383853688836098 2023-01-23 21:54:28.189972: step: 1196/1541, loss: 0.0002797231136355549 2023-01-23 21:54:29.059315: step: 1198/1541, loss: 0.00044670861097984016 2023-01-23 21:54:29.968168: step: 1200/1541, loss: 0.08510788530111313 2023-01-23 21:54:30.846720: step: 1202/1541, loss: 0.00044892667210660875 2023-01-23 21:54:31.747441: step: 1204/1541, loss: 0.13648198544979095 2023-01-23 21:54:32.709754: step: 1206/1541, loss: 0.0004400920297484845 2023-01-23 21:54:33.671493: step: 1208/1541, loss: 0.021318595856428146 2023-01-23 21:54:34.573893: step: 1210/1541, loss: 0.040539465844631195 2023-01-23 21:54:35.488011: step: 1212/1541, loss: 0.026759376749396324 2023-01-23 21:54:36.362169: step: 1214/1541, loss: 0.00026217836420983076 2023-01-23 21:54:37.228685: step: 1216/1541, loss: 0.06014246493577957 2023-01-23 21:54:38.139112: step: 1218/1541, loss: 0.00020028610015287995 2023-01-23 21:54:39.079731: step: 1220/1541, loss: 0.06685745716094971 2023-01-23 21:54:40.008380: step: 1222/1541, loss: 0.0017439615912735462 2023-01-23 21:54:40.931388: step: 1224/1541, loss: 0.000563859473913908 2023-01-23 21:54:41.789784: step: 1226/1541, loss: 0.9317457675933838 2023-01-23 21:54:42.755462: step: 1228/1541, loss: 0.06950722634792328 2023-01-23 21:54:43.622487: step: 1230/1541, loss: 0.002249593148007989 2023-01-23 21:54:44.496570: step: 1232/1541, loss: 0.00023359766055364162 2023-01-23 21:54:45.485792: step: 1234/1541, loss: 0.3220636546611786 2023-01-23 21:54:46.402605: step: 1236/1541, loss: 0.07808058708906174 2023-01-23 21:54:47.341034: step: 1238/1541, loss: 0.0001984226837521419 2023-01-23 21:54:48.273467: step: 1240/1541, loss: 0.011137071996927261 2023-01-23 21:54:49.188408: step: 1242/1541, loss: 0.00019286219321656972 2023-01-23 21:54:50.090824: step: 1244/1541, loss: 0.000166078403708525 2023-01-23 21:54:50.964073: step: 1246/1541, loss: 0.000140193325933069 2023-01-23 21:54:51.873979: step: 1248/1541, loss: 0.12496864795684814 2023-01-23 21:54:52.752502: step: 1250/1541, loss: 0.00030374806374311447 2023-01-23 21:54:53.613316: step: 1252/1541, loss: 0.0030254009179770947 2023-01-23 21:54:54.560477: step: 1254/1541, loss: 0.02476808801293373 2023-01-23 21:54:55.416354: step: 1256/1541, loss: 0.010232657194137573 2023-01-23 21:54:56.310030: step: 1258/1541, loss: 0.18679426610469818 2023-01-23 21:54:57.203712: step: 1260/1541, loss: 0.20759087800979614 2023-01-23 21:54:58.042256: step: 1262/1541, loss: 4.5045753722661175e-06 2023-01-23 21:54:58.934837: step: 1264/1541, loss: 0.012496608309447765 2023-01-23 21:54:59.813790: step: 1266/1541, loss: 0.001641980605199933 2023-01-23 21:55:00.739023: step: 1268/1541, loss: 0.008970163762569427 2023-01-23 21:55:01.612699: step: 1270/1541, loss: 0.00675378181040287 2023-01-23 21:55:02.524110: step: 1272/1541, loss: 0.003892479231581092 2023-01-23 21:55:03.427159: step: 1274/1541, loss: 0.25893521308898926 2023-01-23 21:55:04.273618: step: 1276/1541, loss: 0.0001954090257640928 2023-01-23 21:55:05.151843: step: 1278/1541, loss: 0.0005820273654535413 2023-01-23 21:55:06.013496: step: 1280/1541, loss: 0.0058868275955319405 2023-01-23 21:55:06.893479: step: 1282/1541, loss: 0.0018592100823298097 2023-01-23 21:55:07.765174: step: 1284/1541, loss: 0.001829745015129447 2023-01-23 21:55:08.659838: step: 1286/1541, loss: 9.659864008426666e-05 2023-01-23 21:55:09.524067: step: 1288/1541, loss: 0.00027557986322790384 2023-01-23 21:55:10.384919: step: 1290/1541, loss: 0.15405693650245667 2023-01-23 21:55:11.302896: step: 1292/1541, loss: 0.05928824469447136 2023-01-23 21:55:12.197091: step: 1294/1541, loss: 0.003887430066242814 2023-01-23 21:55:13.089670: step: 1296/1541, loss: 0.024683497846126556 2023-01-23 21:55:14.013862: step: 1298/1541, loss: 0.01986081525683403 2023-01-23 21:55:14.899601: step: 1300/1541, loss: 0.00030402702395804226 2023-01-23 21:55:15.750293: step: 1302/1541, loss: 0.00019935224554501474 2023-01-23 21:55:16.607159: step: 1304/1541, loss: 0.00029117707163095474 2023-01-23 21:55:17.487407: step: 1306/1541, loss: 0.11273915320634842 2023-01-23 21:55:18.400062: step: 1308/1541, loss: 0.00012387133028823882 2023-01-23 21:55:19.297758: step: 1310/1541, loss: 0.0007742486777715385 2023-01-23 21:55:20.137761: step: 1312/1541, loss: 0.6333755254745483 2023-01-23 21:55:21.041110: step: 1314/1541, loss: 0.015306136570870876 2023-01-23 21:55:21.970875: step: 1316/1541, loss: 0.005254029296338558 2023-01-23 21:55:22.854461: step: 1318/1541, loss: 0.07223089784383774 2023-01-23 21:55:23.712828: step: 1320/1541, loss: 0.011559306643903255 2023-01-23 21:55:24.575946: step: 1322/1541, loss: 0.094668909907341 2023-01-23 21:55:25.503965: step: 1324/1541, loss: 0.0012520329328253865 2023-01-23 21:55:26.416140: step: 1326/1541, loss: 0.002905172063037753 2023-01-23 21:55:27.290699: step: 1328/1541, loss: 0.004664214793592691 2023-01-23 21:55:28.171093: step: 1330/1541, loss: 7.113714673323557e-05 2023-01-23 21:55:29.058179: step: 1332/1541, loss: 0.0018688878044486046 2023-01-23 21:55:29.934847: step: 1334/1541, loss: 0.0003537411103025079 2023-01-23 21:55:30.865435: step: 1336/1541, loss: 0.0283790435642004 2023-01-23 21:55:31.808076: step: 1338/1541, loss: 0.010263021104037762 2023-01-23 21:55:32.696411: step: 1340/1541, loss: 0.027128351852297783 2023-01-23 21:55:33.575442: step: 1342/1541, loss: 0.19030584394931793 2023-01-23 21:55:34.455917: step: 1344/1541, loss: 0.002467838814482093 2023-01-23 21:55:35.437593: step: 1346/1541, loss: 0.0017535550286993384 2023-01-23 21:55:36.374079: step: 1348/1541, loss: 0.002115654293447733 2023-01-23 21:55:37.237059: step: 1350/1541, loss: 0.005270020570605993 2023-01-23 21:55:38.103985: step: 1352/1541, loss: 0.006251528859138489 2023-01-23 21:55:38.988437: step: 1354/1541, loss: 0.0015117330476641655 2023-01-23 21:55:39.911124: step: 1356/1541, loss: 0.0003215586766600609 2023-01-23 21:55:40.838465: step: 1358/1541, loss: 0.010692480020225048 2023-01-23 21:55:41.874251: step: 1360/1541, loss: 0.1360832154750824 2023-01-23 21:55:42.783655: step: 1362/1541, loss: 0.9807928204536438 2023-01-23 21:55:43.687595: step: 1364/1541, loss: 0.0003068479709327221 2023-01-23 21:55:44.616063: step: 1366/1541, loss: 0.0005001688841730356 2023-01-23 21:55:45.508596: step: 1368/1541, loss: 0.18093198537826538 2023-01-23 21:55:46.418501: step: 1370/1541, loss: 0.08064626157283783 2023-01-23 21:55:47.292558: step: 1372/1541, loss: 0.0009535592398606241 2023-01-23 21:55:48.264212: step: 1374/1541, loss: 0.98198401927948 2023-01-23 21:55:49.258829: step: 1376/1541, loss: 0.015613309107720852 2023-01-23 21:55:50.183545: step: 1378/1541, loss: 0.001650013611651957 2023-01-23 21:55:51.138261: step: 1380/1541, loss: 0.010076272301375866 2023-01-23 21:55:52.068010: step: 1382/1541, loss: 0.047127459198236465 2023-01-23 21:55:52.928278: step: 1384/1541, loss: 0.023896964266896248 2023-01-23 21:55:53.858202: step: 1386/1541, loss: 0.013234966434538364 2023-01-23 21:55:54.764484: step: 1388/1541, loss: 0.23731550574302673 2023-01-23 21:55:55.639518: step: 1390/1541, loss: 0.0015670665306970477 2023-01-23 21:55:56.595275: step: 1392/1541, loss: 0.0005120598361827433 2023-01-23 21:55:57.446158: step: 1394/1541, loss: 0.0012771100737154484 2023-01-23 21:55:58.312166: step: 1396/1541, loss: 0.0010852789273485541 2023-01-23 21:55:59.243123: step: 1398/1541, loss: 0.02150077186524868 2023-01-23 21:56:00.099716: step: 1400/1541, loss: 0.002449506428092718 2023-01-23 21:56:01.003564: step: 1402/1541, loss: 0.0002169263461837545 2023-01-23 21:56:01.903556: step: 1404/1541, loss: 0.30097636580467224 2023-01-23 21:56:02.813935: step: 1406/1541, loss: 7.441751949954778e-05 2023-01-23 21:56:03.700721: step: 1408/1541, loss: 0.0010816839057952166 2023-01-23 21:56:04.630362: step: 1410/1541, loss: 0.0024956215638667345 2023-01-23 21:56:05.583080: step: 1412/1541, loss: 0.27019423246383667 2023-01-23 21:56:06.450711: step: 1414/1541, loss: 0.00016030887491069734 2023-01-23 21:56:07.331979: step: 1416/1541, loss: 0.14621466398239136 2023-01-23 21:56:08.237458: step: 1418/1541, loss: 0.15067940950393677 2023-01-23 21:56:09.137718: step: 1420/1541, loss: 0.02852839045226574 2023-01-23 21:56:10.045060: step: 1422/1541, loss: 0.0028704279102385044 2023-01-23 21:56:10.928706: step: 1424/1541, loss: 0.15757936239242554 2023-01-23 21:56:11.836309: step: 1426/1541, loss: 0.002233341569080949 2023-01-23 21:56:12.700492: step: 1428/1541, loss: 0.937044620513916 2023-01-23 21:56:13.612291: step: 1430/1541, loss: 0.0006877468549646437 2023-01-23 21:56:14.512013: step: 1432/1541, loss: 0.00037027886719442904 2023-01-23 21:56:15.401223: step: 1434/1541, loss: 0.0001917662302730605 2023-01-23 21:56:16.333058: step: 1436/1541, loss: 0.06159747764468193 2023-01-23 21:56:17.222053: step: 1438/1541, loss: 0.00020074369967915118 2023-01-23 21:56:18.152150: step: 1440/1541, loss: 0.007470440585166216 2023-01-23 21:56:19.026980: step: 1442/1541, loss: 0.198761448264122 2023-01-23 21:56:19.909006: step: 1444/1541, loss: 0.06909658014774323 2023-01-23 21:56:20.782153: step: 1446/1541, loss: 0.6132206320762634 2023-01-23 21:56:21.692750: step: 1448/1541, loss: 0.017244726419448853 2023-01-23 21:56:22.610973: step: 1450/1541, loss: 0.030167438089847565 2023-01-23 21:56:23.478714: step: 1452/1541, loss: 0.039685361087322235 2023-01-23 21:56:24.343223: step: 1454/1541, loss: 0.0009467705967836082 2023-01-23 21:56:25.328217: step: 1456/1541, loss: 0.009882968850433826 2023-01-23 21:56:26.299118: step: 1458/1541, loss: 0.14739789068698883 2023-01-23 21:56:27.185539: step: 1460/1541, loss: -8.062008419074118e-06 2023-01-23 21:56:28.106131: step: 1462/1541, loss: 0.0018938046414405107 2023-01-23 21:56:29.008281: step: 1464/1541, loss: 0.006014090031385422 2023-01-23 21:56:29.931066: step: 1466/1541, loss: 0.0002956946555059403 2023-01-23 21:56:30.802902: step: 1468/1541, loss: 0.0073993755504488945 2023-01-23 21:56:31.660702: step: 1470/1541, loss: 0.0003650230064522475 2023-01-23 21:56:32.599664: step: 1472/1541, loss: 2.1519281290238723e-07 2023-01-23 21:56:33.491415: step: 1474/1541, loss: 0.0014673892874270678 2023-01-23 21:56:34.392636: step: 1476/1541, loss: 0.0004669454647228122 2023-01-23 21:56:35.386824: step: 1478/1541, loss: 0.0003136733139399439 2023-01-23 21:56:36.331531: step: 1480/1541, loss: 0.08494125306606293 2023-01-23 21:56:37.221397: step: 1482/1541, loss: 4.683520455728285e-05 2023-01-23 21:56:38.104429: step: 1484/1541, loss: 0.0010144731495529413 2023-01-23 21:56:39.023927: step: 1486/1541, loss: 0.12157569080591202 2023-01-23 21:56:39.858809: step: 1488/1541, loss: 0.00012185543891973794 2023-01-23 21:56:40.731232: step: 1490/1541, loss: 5.5973418056964874e-05 2023-01-23 21:56:41.643755: step: 1492/1541, loss: 0.025785893201828003 2023-01-23 21:56:42.490616: step: 1494/1541, loss: 0.01996072381734848 2023-01-23 21:56:43.379873: step: 1496/1541, loss: 0.002445501508191228 2023-01-23 21:56:44.288521: step: 1498/1541, loss: 0.00032889118301682174 2023-01-23 21:56:45.182308: step: 1500/1541, loss: 0.027773896232247353 2023-01-23 21:56:46.126158: step: 1502/1541, loss: 0.0004634583310689777 2023-01-23 21:56:47.028989: step: 1504/1541, loss: 0.00012037689157295972 2023-01-23 21:56:47.981087: step: 1506/1541, loss: 0.009333514608442783 2023-01-23 21:56:48.838469: step: 1508/1541, loss: 0.00758745614439249 2023-01-23 21:56:49.773908: step: 1510/1541, loss: 0.0005378100904636085 2023-01-23 21:56:50.649194: step: 1512/1541, loss: 0.0025011925026774406 2023-01-23 21:56:51.554570: step: 1514/1541, loss: 0.00332347908988595 2023-01-23 21:56:52.455351: step: 1516/1541, loss: 0.0001467110269004479 2023-01-23 21:56:53.345479: step: 1518/1541, loss: 0.00012371805496513844 2023-01-23 21:56:54.226662: step: 1520/1541, loss: 0.0011379681527614594 2023-01-23 21:56:55.088211: step: 1522/1541, loss: 0.00046604376984760165 2023-01-23 21:56:55.988616: step: 1524/1541, loss: 0.0010683273430913687 2023-01-23 21:56:56.921646: step: 1526/1541, loss: 9.806156595004722e-05 2023-01-23 21:56:57.795873: step: 1528/1541, loss: 0.00028969577397219837 2023-01-23 21:56:58.691858: step: 1530/1541, loss: 0.051462575793266296 2023-01-23 21:56:59.645878: step: 1532/1541, loss: 0.00014634088438469917 2023-01-23 21:57:00.514824: step: 1534/1541, loss: 0.0003876126720570028 2023-01-23 21:57:01.413511: step: 1536/1541, loss: 0.10169696062803268 2023-01-23 21:57:02.291037: step: 1538/1541, loss: 0.0011793748708441854 2023-01-23 21:57:03.189958: step: 1540/1541, loss: 0.037823114544153214 2023-01-23 21:57:04.074864: step: 1542/1541, loss: 6.591623241547495e-05 2023-01-23 21:57:04.938276: step: 1544/1541, loss: 0.16039308905601501 2023-01-23 21:57:05.886792: step: 1546/1541, loss: 0.0016390183009207249 2023-01-23 21:57:06.806113: step: 1548/1541, loss: 0.014932136051356792 2023-01-23 21:57:07.675045: step: 1550/1541, loss: 0.3398846685886383 2023-01-23 21:57:08.576119: step: 1552/1541, loss: 0.00026053344481624663 2023-01-23 21:57:09.478675: step: 1554/1541, loss: 0.008884605951607227 2023-01-23 21:57:10.425245: step: 1556/1541, loss: 0.0005222360487096012 2023-01-23 21:57:11.284947: step: 1558/1541, loss: 2.8236212529009208e-05 2023-01-23 21:57:12.185585: step: 1560/1541, loss: 0.0013549539726227522 2023-01-23 21:57:13.069624: step: 1562/1541, loss: 0.06504102051258087 2023-01-23 21:57:13.961366: step: 1564/1541, loss: 0.0021517493296414614 2023-01-23 21:57:14.860629: step: 1566/1541, loss: 0.0025749006308615208 2023-01-23 21:57:15.754420: step: 1568/1541, loss: 0.00022547022672370076 2023-01-23 21:57:16.653060: step: 1570/1541, loss: 0.00026245234766975045 2023-01-23 21:57:17.554530: step: 1572/1541, loss: 1.5769013166427612 2023-01-23 21:57:18.453155: step: 1574/1541, loss: 0.010003695264458656 2023-01-23 21:57:19.343393: step: 1576/1541, loss: 0.04124677553772926 2023-01-23 21:57:20.234287: step: 1578/1541, loss: 0.07020694762468338 2023-01-23 21:57:21.143391: step: 1580/1541, loss: 0.0005359978531487286 2023-01-23 21:57:22.108430: step: 1582/1541, loss: -8.530387276550755e-06 2023-01-23 21:57:22.982625: step: 1584/1541, loss: 0.00014688109513372183 2023-01-23 21:57:23.897094: step: 1586/1541, loss: 0.0022979534696787596 2023-01-23 21:57:24.811682: step: 1588/1541, loss: 0.20805498957633972 2023-01-23 21:57:25.755455: step: 1590/1541, loss: 0.12622429430484772 2023-01-23 21:57:26.673372: step: 1592/1541, loss: 0.003054555505514145 2023-01-23 21:57:27.548975: step: 1594/1541, loss: 0.00018253239977639169 2023-01-23 21:57:28.422876: step: 1596/1541, loss: 0.032441698014736176 2023-01-23 21:57:29.335574: step: 1598/1541, loss: 0.03510451689362526 2023-01-23 21:57:30.297550: step: 1600/1541, loss: 0.09290680289268494 2023-01-23 21:57:31.188446: step: 1602/1541, loss: 0.0025333680678159 2023-01-23 21:57:32.068580: step: 1604/1541, loss: 0.0013758321292698383 2023-01-23 21:57:32.988839: step: 1606/1541, loss: 0.00010533750173635781 2023-01-23 21:57:33.854140: step: 1608/1541, loss: 4.1108338336925954e-05 2023-01-23 21:57:34.842767: step: 1610/1541, loss: 0.001998860388994217 2023-01-23 21:57:35.742904: step: 1612/1541, loss: 0.0018891486106440425 2023-01-23 21:57:36.641102: step: 1614/1541, loss: 0.6623410582542419 2023-01-23 21:57:37.507973: step: 1616/1541, loss: 0.0014001113595440984 2023-01-23 21:57:38.395893: step: 1618/1541, loss: 0.005568246357142925 2023-01-23 21:57:39.334524: step: 1620/1541, loss: 0.0005001720855943859 2023-01-23 21:57:40.229585: step: 1622/1541, loss: 0.0004468628903850913 2023-01-23 21:57:41.106252: step: 1624/1541, loss: 9.053647954715416e-05 2023-01-23 21:57:42.017305: step: 1626/1541, loss: 0.001583322649821639 2023-01-23 21:57:42.962251: step: 1628/1541, loss: 0.36984482407569885 2023-01-23 21:57:43.898286: step: 1630/1541, loss: 0.004115539602935314 2023-01-23 21:57:44.773884: step: 1632/1541, loss: 0.00015445600729435682 2023-01-23 21:57:45.653414: step: 1634/1541, loss: 0.009623764082789421 2023-01-23 21:57:46.552668: step: 1636/1541, loss: 0.0030856484081596136 2023-01-23 21:57:47.508824: step: 1638/1541, loss: 0.006776516325771809 2023-01-23 21:57:48.420191: step: 1640/1541, loss: 0.0034846263006329536 2023-01-23 21:57:49.299298: step: 1642/1541, loss: 0.06677161157131195 2023-01-23 21:57:50.219811: step: 1644/1541, loss: 0.011769406497478485 2023-01-23 21:57:51.062522: step: 1646/1541, loss: 0.000798261258751154 2023-01-23 21:57:52.034698: step: 1648/1541, loss: 0.0036633601412177086 2023-01-23 21:57:52.896371: step: 1650/1541, loss: 0.0005609892541542649 2023-01-23 21:57:53.766516: step: 1652/1541, loss: 0.015408660285174847 2023-01-23 21:57:54.675115: step: 1654/1541, loss: 0.00829560961574316 2023-01-23 21:57:55.551336: step: 1656/1541, loss: 0.0002778207417577505 2023-01-23 21:57:56.443225: step: 1658/1541, loss: 0.00044081208761781454 2023-01-23 21:57:57.324118: step: 1660/1541, loss: 0.0002489840844646096 2023-01-23 21:57:58.222913: step: 1662/1541, loss: 0.0015232511796057224 2023-01-23 21:57:59.112833: step: 1664/1541, loss: 0.13159729540348053 2023-01-23 21:58:00.030938: step: 1666/1541, loss: 0.011407481506466866 2023-01-23 21:58:00.920433: step: 1668/1541, loss: 0.010528108105063438 2023-01-23 21:58:01.890756: step: 1670/1541, loss: 0.37745222449302673 2023-01-23 21:58:02.815705: step: 1672/1541, loss: 0.002178743714466691 2023-01-23 21:58:03.695983: step: 1674/1541, loss: 0.0012032523518428206 2023-01-23 21:58:04.566394: step: 1676/1541, loss: 0.10544602572917938 2023-01-23 21:58:05.406906: step: 1678/1541, loss: 0.0005858839140273631 2023-01-23 21:58:06.319189: step: 1680/1541, loss: 0.0320434495806694 2023-01-23 21:58:07.228819: step: 1682/1541, loss: 0.15700219571590424 2023-01-23 21:58:08.101908: step: 1684/1541, loss: 0.004443218000233173 2023-01-23 21:58:08.963701: step: 1686/1541, loss: 0.00021405122242867947 2023-01-23 21:58:09.911926: step: 1688/1541, loss: 0.01023491658270359 2023-01-23 21:58:10.765398: step: 1690/1541, loss: 0.0003000847646035254 2023-01-23 21:58:11.630636: step: 1692/1541, loss: 0.001260452438145876 2023-01-23 21:58:12.573574: step: 1694/1541, loss: 0.0024016648530960083 2023-01-23 21:58:13.442177: step: 1696/1541, loss: 0.008056639693677425 2023-01-23 21:58:14.328935: step: 1698/1541, loss: 0.05438961461186409 2023-01-23 21:58:15.263955: step: 1700/1541, loss: 0.00025599519722163677 2023-01-23 21:58:16.141133: step: 1702/1541, loss: 0.00011559783888515085 2023-01-23 21:58:17.043337: step: 1704/1541, loss: 5.9766840422526e-05 2023-01-23 21:58:17.903717: step: 1706/1541, loss: 0.00014960794942453504 2023-01-23 21:58:18.768481: step: 1708/1541, loss: 0.004324294626712799 2023-01-23 21:58:19.618993: step: 1710/1541, loss: 0.0021611200645565987 2023-01-23 21:58:20.593647: step: 1712/1541, loss: 0.04853370413184166 2023-01-23 21:58:21.477752: step: 1714/1541, loss: 0.00022016564616933465 2023-01-23 21:58:22.393091: step: 1716/1541, loss: 2.2798278223490342e-05 2023-01-23 21:58:23.377350: step: 1718/1541, loss: 0.0039016816299408674 2023-01-23 21:58:24.269805: step: 1720/1541, loss: 0.0006920396699570119 2023-01-23 21:58:25.144227: step: 1722/1541, loss: 0.00011818046914413571 2023-01-23 21:58:26.026013: step: 1724/1541, loss: 0.0013060863129794598 2023-01-23 21:58:26.930549: step: 1726/1541, loss: -2.7038655389333144e-05 2023-01-23 21:58:27.814218: step: 1728/1541, loss: 0.0054998318664729595 2023-01-23 21:58:28.773964: step: 1730/1541, loss: 0.0022907413076609373 2023-01-23 21:58:29.645943: step: 1732/1541, loss: 0.012405249290168285 2023-01-23 21:58:30.535916: step: 1734/1541, loss: 0.006021430715918541 2023-01-23 21:58:31.379920: step: 1736/1541, loss: 0.003983908798545599 2023-01-23 21:58:32.278057: step: 1738/1541, loss: 0.0010565943084657192 2023-01-23 21:58:33.189228: step: 1740/1541, loss: 0.0026759295724332333 2023-01-23 21:58:34.082898: step: 1742/1541, loss: 7.173866470111534e-05 2023-01-23 21:58:35.020384: step: 1744/1541, loss: 0.021464582532644272 2023-01-23 21:58:35.878866: step: 1746/1541, loss: 0.017116250470280647 2023-01-23 21:58:36.783665: step: 1748/1541, loss: 0.022896597161889076 2023-01-23 21:58:37.683024: step: 1750/1541, loss: 0.12277346104383469 2023-01-23 21:58:38.636285: step: 1752/1541, loss: 0.001978851156309247 2023-01-23 21:58:39.528190: step: 1754/1541, loss: 0.0009742790716700256 2023-01-23 21:58:40.381238: step: 1756/1541, loss: 0.00010268144978908822 2023-01-23 21:58:41.260720: step: 1758/1541, loss: 0.0002776911132968962 2023-01-23 21:58:42.235868: step: 1760/1541, loss: 0.0019992466550320387 2023-01-23 21:58:43.131775: step: 1762/1541, loss: 0.012492765672504902 2023-01-23 21:58:44.015092: step: 1764/1541, loss: 0.001467997208237648 2023-01-23 21:58:44.913279: step: 1766/1541, loss: 0.003522083628922701 2023-01-23 21:58:45.784796: step: 1768/1541, loss: 0.0017094019567593932 2023-01-23 21:58:46.646023: step: 1770/1541, loss: 0.15308570861816406 2023-01-23 21:58:47.589248: step: 1772/1541, loss: 0.295093297958374 2023-01-23 21:58:48.486812: step: 1774/1541, loss: -1.7260068489122204e-05 2023-01-23 21:58:49.373444: step: 1776/1541, loss: 0.12325053662061691 2023-01-23 21:58:50.237296: step: 1778/1541, loss: 0.0931362584233284 2023-01-23 21:58:51.109484: step: 1780/1541, loss: 1.2192142094136216e-05 2023-01-23 21:58:51.957415: step: 1782/1541, loss: -1.215265911014285e-05 2023-01-23 21:58:52.809826: step: 1784/1541, loss: 4.432798050402198e-06 2023-01-23 21:58:53.679337: step: 1786/1541, loss: 0.009984053671360016 2023-01-23 21:58:54.512210: step: 1788/1541, loss: 0.05237088352441788 2023-01-23 21:58:55.399042: step: 1790/1541, loss: 0.1819593757390976 2023-01-23 21:58:56.276899: step: 1792/1541, loss: 0.00012245739344507456 2023-01-23 21:58:57.210138: step: 1794/1541, loss: 0.028790485113859177 2023-01-23 21:58:58.049333: step: 1796/1541, loss: 5.0566348363645375e-05 2023-01-23 21:58:58.962078: step: 1798/1541, loss: 0.10501952469348907 2023-01-23 21:58:59.865362: step: 1800/1541, loss: 0.05699004977941513 2023-01-23 21:59:00.746436: step: 1802/1541, loss: 8.60479412949644e-05 2023-01-23 21:59:01.650569: step: 1804/1541, loss: 0.0002012089971685782 2023-01-23 21:59:02.574512: step: 1806/1541, loss: 6.882542948005721e-05 2023-01-23 21:59:03.485232: step: 1808/1541, loss: -5.310773303790484e-06 2023-01-23 21:59:04.478733: step: 1810/1541, loss: 3.630768333096057e-05 2023-01-23 21:59:05.334046: step: 1812/1541, loss: 3.172427386743948e-05 2023-01-23 21:59:06.246431: step: 1814/1541, loss: 0.000341635022778064 2023-01-23 21:59:07.141635: step: 1816/1541, loss: 0.15558035671710968 2023-01-23 21:59:07.996036: step: 1818/1541, loss: 0.00012130737013649195 2023-01-23 21:59:08.888900: step: 1820/1541, loss: 0.010026618838310242 2023-01-23 21:59:09.830444: step: 1822/1541, loss: 0.0027085975743830204 2023-01-23 21:59:10.706456: step: 1824/1541, loss: 1.5709123545093462e-05 2023-01-23 21:59:11.625075: step: 1826/1541, loss: 0.0010540126822888851 2023-01-23 21:59:12.509634: step: 1828/1541, loss: 0.004321194253861904 2023-01-23 21:59:13.440988: step: 1830/1541, loss: 0.08414081484079361 2023-01-23 21:59:14.283964: step: 1832/1541, loss: 0.0015347665175795555 2023-01-23 21:59:15.168487: step: 1834/1541, loss: 0.24368298053741455 2023-01-23 21:59:16.050475: step: 1836/1541, loss: 0.000486447912408039 2023-01-23 21:59:16.989362: step: 1838/1541, loss: 0.0010297419503331184 2023-01-23 21:59:17.863700: step: 1840/1541, loss: 3.155317608616315e-05 2023-01-23 21:59:18.727345: step: 1842/1541, loss: -1.5353285789387883e-06 2023-01-23 21:59:19.594658: step: 1844/1541, loss: 0.006489648949354887 2023-01-23 21:59:20.520649: step: 1846/1541, loss: 0.0007148048025555909 2023-01-23 21:59:21.419061: step: 1848/1541, loss: 0.0006811365601606667 2023-01-23 21:59:22.295628: step: 1850/1541, loss: 0.00011925250146305189 2023-01-23 21:59:23.198450: step: 1852/1541, loss: 7.033527799649164e-05 2023-01-23 21:59:24.074923: step: 1854/1541, loss: 0.00022861655452288687 2023-01-23 21:59:25.004186: step: 1856/1541, loss: 0.0037183016538619995 2023-01-23 21:59:25.880961: step: 1858/1541, loss: 0.0008493923814967275 2023-01-23 21:59:26.775988: step: 1860/1541, loss: 0.0637759268283844 2023-01-23 21:59:27.663015: step: 1862/1541, loss: 0.01004908885806799 2023-01-23 21:59:28.567281: step: 1864/1541, loss: 0.0004171222390141338 2023-01-23 21:59:29.424126: step: 1866/1541, loss: 0.2664875388145447 2023-01-23 21:59:30.288968: step: 1868/1541, loss: 0.00013261890853755176 2023-01-23 21:59:31.185116: step: 1870/1541, loss: 0.00290169520303607 2023-01-23 21:59:32.089477: step: 1872/1541, loss: 3.3869196158775594e-07 2023-01-23 21:59:32.977771: step: 1874/1541, loss: 0.41510868072509766 2023-01-23 21:59:33.850671: step: 1876/1541, loss: 0.0016300728311762214 2023-01-23 21:59:34.704477: step: 1878/1541, loss: 0.0010234036017209291 2023-01-23 21:59:35.554572: step: 1880/1541, loss: 0.0003516891738399863 2023-01-23 21:59:36.431677: step: 1882/1541, loss: 0.0005366533878259361 2023-01-23 21:59:37.323143: step: 1884/1541, loss: 0.034204933792352676 2023-01-23 21:59:38.229452: step: 1886/1541, loss: 0.0005825315020047128 2023-01-23 21:59:39.224804: step: 1888/1541, loss: 0.0005613237735815346 2023-01-23 21:59:40.192166: step: 1890/1541, loss: 0.04973332956433296 2023-01-23 21:59:41.119682: step: 1892/1541, loss: 0.0027888850308954716 2023-01-23 21:59:41.989824: step: 1894/1541, loss: 8.929595060180873e-05 2023-01-23 21:59:42.846151: step: 1896/1541, loss: 0.0009561377228237689 2023-01-23 21:59:43.745101: step: 1898/1541, loss: 0.14810319244861603 2023-01-23 21:59:44.664531: step: 1900/1541, loss: 0.08001255989074707 2023-01-23 21:59:45.541303: step: 1902/1541, loss: 0.05941452458500862 2023-01-23 21:59:46.451201: step: 1904/1541, loss: 0.16531872749328613 2023-01-23 21:59:47.296465: step: 1906/1541, loss: 0.0003147817915305495 2023-01-23 21:59:48.165216: step: 1908/1541, loss: 0.04427232965826988 2023-01-23 21:59:49.043537: step: 1910/1541, loss: 8.402318053413182e-05 2023-01-23 21:59:49.922449: step: 1912/1541, loss: 6.62974125589244e-05 2023-01-23 21:59:50.800756: step: 1914/1541, loss: 7.061790529405698e-05 2023-01-23 21:59:51.703893: step: 1916/1541, loss: 0.006740734912455082 2023-01-23 21:59:52.579606: step: 1918/1541, loss: 4.056155739817768e-05 2023-01-23 21:59:53.478824: step: 1920/1541, loss: 0.048359472304582596 2023-01-23 21:59:54.353082: step: 1922/1541, loss: 0.0001714178069960326 2023-01-23 21:59:55.268193: step: 1924/1541, loss: 9.382566349813715e-05 2023-01-23 21:59:56.218949: step: 1926/1541, loss: 0.000976690324023366 2023-01-23 21:59:57.106516: step: 1928/1541, loss: 0.20760247111320496 2023-01-23 21:59:58.019257: step: 1930/1541, loss: 0.24109312891960144 2023-01-23 21:59:58.863213: step: 1932/1541, loss: 0.14096832275390625 2023-01-23 21:59:59.742035: step: 1934/1541, loss: 0.00013931569992564619 2023-01-23 22:00:00.634570: step: 1936/1541, loss: 0.0009208604460582137 2023-01-23 22:00:01.504742: step: 1938/1541, loss: 0.0013724350137636065 2023-01-23 22:00:02.375288: step: 1940/1541, loss: 0.010532736778259277 2023-01-23 22:00:03.263341: step: 1942/1541, loss: 0.05470457300543785 2023-01-23 22:00:04.165861: step: 1944/1541, loss: 0.0009526514331810176 2023-01-23 22:00:05.079015: step: 1946/1541, loss: 0.00011320020712446421 2023-01-23 22:00:05.987767: step: 1948/1541, loss: 0.18657085299491882 2023-01-23 22:00:06.882560: step: 1950/1541, loss: 0.002930668881163001 2023-01-23 22:00:07.745283: step: 1952/1541, loss: 0.005931820720434189 2023-01-23 22:00:08.623766: step: 1954/1541, loss: 1.698905907687731e-05 2023-01-23 22:00:09.513690: step: 1956/1541, loss: 1.4997073776612524e-05 2023-01-23 22:00:10.396654: step: 1958/1541, loss: 0.0005917088128626347 2023-01-23 22:00:11.378452: step: 1960/1541, loss: 0.0006680372171103954 2023-01-23 22:00:12.287110: step: 1962/1541, loss: 3.9825725252740085e-05 2023-01-23 22:00:13.296939: step: 1964/1541, loss: 0.42991623282432556 2023-01-23 22:00:14.233699: step: 1966/1541, loss: 0.00019438590970821679 2023-01-23 22:00:15.227053: step: 1968/1541, loss: 0.007987048476934433 2023-01-23 22:00:16.154688: step: 1970/1541, loss: 0.0005020698881708086 2023-01-23 22:00:17.087011: step: 1972/1541, loss: 0.0194336399435997 2023-01-23 22:00:17.933250: step: 1974/1541, loss: 8.010327292140573e-05 2023-01-23 22:00:18.873521: step: 1976/1541, loss: 0.0013254092773422599 2023-01-23 22:00:19.744641: step: 1978/1541, loss: 0.0004449232656043023 2023-01-23 22:00:20.645668: step: 1980/1541, loss: 0.001165103865787387 2023-01-23 22:00:21.538777: step: 1982/1541, loss: 9.538531594444066e-05 2023-01-23 22:00:22.431416: step: 1984/1541, loss: 0.09321723878383636 2023-01-23 22:00:23.334905: step: 1986/1541, loss: 0.1367533951997757 2023-01-23 22:00:24.193724: step: 1988/1541, loss: 0.0019100748468190432 2023-01-23 22:00:25.073010: step: 1990/1541, loss: 0.007902204990386963 2023-01-23 22:00:25.958869: step: 1992/1541, loss: 0.010506223887205124 2023-01-23 22:00:26.851889: step: 1994/1541, loss: 0.0026509142480790615 2023-01-23 22:00:27.738903: step: 1996/1541, loss: 0.12563951313495636 2023-01-23 22:00:28.612665: step: 1998/1541, loss: 0.02203267253935337 2023-01-23 22:00:29.489137: step: 2000/1541, loss: 0.11083034425973892 2023-01-23 22:00:30.384271: step: 2002/1541, loss: 0.0006868519121780992 2023-01-23 22:00:31.245568: step: 2004/1541, loss: 0.001610645093023777 2023-01-23 22:00:32.195583: step: 2006/1541, loss: 0.00021440339332912117 2023-01-23 22:00:33.065992: step: 2008/1541, loss: 0.274516761302948 2023-01-23 22:00:33.935052: step: 2010/1541, loss: 0.029965976253151894 2023-01-23 22:00:34.868008: step: 2012/1541, loss: 0.04734114557504654 2023-01-23 22:00:35.754128: step: 2014/1541, loss: 0.0019355567637830973 2023-01-23 22:00:36.607807: step: 2016/1541, loss: 0.2146480530500412 2023-01-23 22:00:37.489418: step: 2018/1541, loss: 0.010568100959062576 2023-01-23 22:00:38.414753: step: 2020/1541, loss: 0.019075695425271988 2023-01-23 22:00:39.369100: step: 2022/1541, loss: 0.14965525269508362 2023-01-23 22:00:40.249812: step: 2024/1541, loss: 0.058990977704524994 2023-01-23 22:00:41.222089: step: 2026/1541, loss: 0.02251172997057438 2023-01-23 22:00:42.117958: step: 2028/1541, loss: 0.00031189178116619587 2023-01-23 22:00:43.009333: step: 2030/1541, loss: 0.0038187294267117977 2023-01-23 22:00:43.860954: step: 2032/1541, loss: 0.0038615793455392122 2023-01-23 22:00:44.752020: step: 2034/1541, loss: 0.0012347889132797718 2023-01-23 22:00:45.631433: step: 2036/1541, loss: 0.004273836966603994 2023-01-23 22:00:46.526743: step: 2038/1541, loss: 0.00032777839805930853 2023-01-23 22:00:47.435018: step: 2040/1541, loss: 0.0012247629929333925 2023-01-23 22:00:48.366614: step: 2042/1541, loss: 0.0008589060744270682 2023-01-23 22:00:49.261978: step: 2044/1541, loss: 0.00016931904247030616 2023-01-23 22:00:50.187849: step: 2046/1541, loss: 0.0004148632287979126 2023-01-23 22:00:51.058189: step: 2048/1541, loss: 0.00566515326499939 2023-01-23 22:00:51.982928: step: 2050/1541, loss: 0.1076783612370491 2023-01-23 22:00:52.851000: step: 2052/1541, loss: 0.00026889037690125406 2023-01-23 22:00:53.721942: step: 2054/1541, loss: 5.643120675813407e-05 2023-01-23 22:00:54.619535: step: 2056/1541, loss: 7.059751078486443e-05 2023-01-23 22:00:55.510760: step: 2058/1541, loss: 0.02332102693617344 2023-01-23 22:00:56.453324: step: 2060/1541, loss: 0.03976212441921234 2023-01-23 22:00:57.355633: step: 2062/1541, loss: 0.0006774663925170898 2023-01-23 22:00:58.255054: step: 2064/1541, loss: 0.00033632959821261466 2023-01-23 22:00:59.130302: step: 2066/1541, loss: 0.037671931087970734 2023-01-23 22:00:59.994482: step: 2068/1541, loss: 9.348131425213069e-05 2023-01-23 22:01:00.874313: step: 2070/1541, loss: 5.1780396461253986e-05 2023-01-23 22:01:01.789108: step: 2072/1541, loss: 0.009274626150727272 2023-01-23 22:01:02.677120: step: 2074/1541, loss: 5.676741784554906e-05 2023-01-23 22:01:03.538755: step: 2076/1541, loss: 0.001807976863346994 2023-01-23 22:01:04.367668: step: 2078/1541, loss: 3.7349732338043395e-06 2023-01-23 22:01:05.267295: step: 2080/1541, loss: 5.095750020700507e-05 2023-01-23 22:01:06.130858: step: 2082/1541, loss: 3.9832055335864425e-05 2023-01-23 22:01:07.027659: step: 2084/1541, loss: 0.001012371969409287 2023-01-23 22:01:07.914487: step: 2086/1541, loss: 0.0014123960863798857 2023-01-23 22:01:08.826004: step: 2088/1541, loss: 0.0013385595520958304 2023-01-23 22:01:09.683271: step: 2090/1541, loss: 0.0030788162257522345 2023-01-23 22:01:10.626239: step: 2092/1541, loss: 6.381010462064296e-05 2023-01-23 22:01:11.624824: step: 2094/1541, loss: 0.0023691190872341394 2023-01-23 22:01:12.481185: step: 2096/1541, loss: 2.7811351174023002e-05 2023-01-23 22:01:13.377665: step: 2098/1541, loss: 1.5901028746156953e-05 2023-01-23 22:01:14.311019: step: 2100/1541, loss: 0.0024685459211468697 2023-01-23 22:01:15.241384: step: 2102/1541, loss: 0.0005306885577738285 2023-01-23 22:01:16.161338: step: 2104/1541, loss: 0.047774359583854675 2023-01-23 22:01:17.041943: step: 2106/1541, loss: 0.00028500944608822465 2023-01-23 22:01:17.969513: step: 2108/1541, loss: 0.200821191072464 2023-01-23 22:01:18.880449: step: 2110/1541, loss: 0.0007355449488386512 2023-01-23 22:01:19.796785: step: 2112/1541, loss: 0.00014269290841184556 2023-01-23 22:01:20.647567: step: 2114/1541, loss: 0.3890622556209564 2023-01-23 22:01:21.539079: step: 2116/1541, loss: 0.35817280411720276 2023-01-23 22:01:22.453293: step: 2118/1541, loss: 0.038320161402225494 2023-01-23 22:01:23.389163: step: 2120/1541, loss: 3.8380170735763386e-05 2023-01-23 22:01:24.280510: step: 2122/1541, loss: 0.028328292071819305 2023-01-23 22:01:25.174622: step: 2124/1541, loss: 0.12073297798633575 2023-01-23 22:01:26.108200: step: 2126/1541, loss: 0.003009319771081209 2023-01-23 22:01:26.959908: step: 2128/1541, loss: 4.827499287785031e-05 2023-01-23 22:01:27.849369: step: 2130/1541, loss: 0.00128963403403759 2023-01-23 22:01:28.730837: step: 2132/1541, loss: 0.07854825258255005 2023-01-23 22:01:29.616560: step: 2134/1541, loss: 2.320056955795735e-05 2023-01-23 22:01:30.504791: step: 2136/1541, loss: -4.578333755489439e-05 2023-01-23 22:01:31.419560: step: 2138/1541, loss: -4.811408871319145e-07 2023-01-23 22:01:32.290150: step: 2140/1541, loss: 9.135380969382823e-05 2023-01-23 22:01:33.190932: step: 2142/1541, loss: 0.02044564113020897 2023-01-23 22:01:34.131195: step: 2144/1541, loss: 0.00036717503098770976 2023-01-23 22:01:35.049463: step: 2146/1541, loss: 0.00690447399392724 2023-01-23 22:01:35.965846: step: 2148/1541, loss: 0.06874766945838928 2023-01-23 22:01:36.841398: step: 2150/1541, loss: 0.00013206950097810477 2023-01-23 22:01:37.762178: step: 2152/1541, loss: 0.24821580946445465 2023-01-23 22:01:38.674669: step: 2154/1541, loss: 0.19167371094226837 2023-01-23 22:01:39.537393: step: 2156/1541, loss: 0.0007991504389792681 2023-01-23 22:01:40.411370: step: 2158/1541, loss: 6.986260268604383e-05 2023-01-23 22:01:41.277162: step: 2160/1541, loss: 5.3006344387540594e-05 2023-01-23 22:01:42.163235: step: 2162/1541, loss: 0.00031157254124991596 2023-01-23 22:01:43.085779: step: 2164/1541, loss: 0.000703396974131465 2023-01-23 22:01:43.955231: step: 2166/1541, loss: 6.438625860027969e-05 2023-01-23 22:01:44.846039: step: 2168/1541, loss: 0.004789919592440128 2023-01-23 22:01:45.801992: step: 2170/1541, loss: 0.004010527860373259 2023-01-23 22:01:46.783662: step: 2172/1541, loss: 0.057167258113622665 2023-01-23 22:01:47.666628: step: 2174/1541, loss: 0.0002994602546095848 2023-01-23 22:01:48.523507: step: 2176/1541, loss: 0.00020373162988107651 2023-01-23 22:01:49.410707: step: 2178/1541, loss: 0.05820731818675995 2023-01-23 22:01:50.313634: step: 2180/1541, loss: 0.008382903411984444 2023-01-23 22:01:51.218200: step: 2182/1541, loss: 0.003639346919953823 2023-01-23 22:01:52.104405: step: 2184/1541, loss: 0.0001865130034275353 2023-01-23 22:01:52.999231: step: 2186/1541, loss: 0.007506978698074818 2023-01-23 22:01:53.871220: step: 2188/1541, loss: 0.00022293149959295988 2023-01-23 22:01:54.798290: step: 2190/1541, loss: 0.00030422062263824046 2023-01-23 22:01:55.685828: step: 2192/1541, loss: 5.058432725491002e-05 2023-01-23 22:01:56.545368: step: 2194/1541, loss: 0.00044060262735001743 2023-01-23 22:01:57.419475: step: 2196/1541, loss: 0.000887254485860467 2023-01-23 22:01:58.271616: step: 2198/1541, loss: 0.1493789106607437 2023-01-23 22:01:59.174988: step: 2200/1541, loss: 0.0005370908183977008 2023-01-23 22:02:00.076018: step: 2202/1541, loss: 0.00017447947175242007 2023-01-23 22:02:00.981921: step: 2204/1541, loss: 0.000539728906005621 2023-01-23 22:02:01.936525: step: 2206/1541, loss: 0.14223264157772064 2023-01-23 22:02:02.841928: step: 2208/1541, loss: 0.00026266431086696684 2023-01-23 22:02:03.767028: step: 2210/1541, loss: 0.03592958301305771 2023-01-23 22:02:04.613813: step: 2212/1541, loss: 0.0010953706223517656 2023-01-23 22:02:05.471298: step: 2214/1541, loss: 0.00029898821958340704 2023-01-23 22:02:06.376163: step: 2216/1541, loss: 6.710260640829802e-05 2023-01-23 22:02:07.320529: step: 2218/1541, loss: 0.07628444582223892 2023-01-23 22:02:08.291192: step: 2220/1541, loss: 0.32020246982574463 2023-01-23 22:02:09.182882: step: 2222/1541, loss: 4.050786083098501e-05 2023-01-23 22:02:10.069540: step: 2224/1541, loss: 0.34622329473495483 2023-01-23 22:02:11.006012: step: 2226/1541, loss: 6.445318285841495e-05 2023-01-23 22:02:11.900100: step: 2228/1541, loss: 0.012748192064464092 2023-01-23 22:02:12.769828: step: 2230/1541, loss: 0.00014329551777336746 2023-01-23 22:02:13.690409: step: 2232/1541, loss: 0.4298342168331146 2023-01-23 22:02:14.576193: step: 2234/1541, loss: 0.0004474230809137225 2023-01-23 22:02:15.521650: step: 2236/1541, loss: 0.004660641308873892 2023-01-23 22:02:16.445073: step: 2238/1541, loss: 0.00761550385504961 2023-01-23 22:02:17.361325: step: 2240/1541, loss: 3.13296914100647e-05 2023-01-23 22:02:18.257649: step: 2242/1541, loss: 0.00015889316273387522 2023-01-23 22:02:19.162414: step: 2244/1541, loss: 0.0005055552464909852 2023-01-23 22:02:20.081693: step: 2246/1541, loss: 0.17436042428016663 2023-01-23 22:02:20.969154: step: 2248/1541, loss: 0.056394610553979874 2023-01-23 22:02:21.880149: step: 2250/1541, loss: 0.00016119374777190387 2023-01-23 22:02:22.749292: step: 2252/1541, loss: 1.062935552909039e-06 2023-01-23 22:02:23.668363: step: 2254/1541, loss: 0.04933404549956322 2023-01-23 22:02:24.594288: step: 2256/1541, loss: 0.00010341157030779868 2023-01-23 22:02:25.486497: step: 2258/1541, loss: 0.0006500979070551693 2023-01-23 22:02:26.403096: step: 2260/1541, loss: 0.00010929405834758654 2023-01-23 22:02:27.299394: step: 2262/1541, loss: 0.12114597111940384 2023-01-23 22:02:28.197859: step: 2264/1541, loss: -4.934158641844988e-05 2023-01-23 22:02:29.129432: step: 2266/1541, loss: 0.0007524094544351101 2023-01-23 22:02:29.999783: step: 2268/1541, loss: 0.06752176582813263 2023-01-23 22:02:30.941464: step: 2270/1541, loss: 0.00016517886251676828 2023-01-23 22:02:31.787300: step: 2272/1541, loss: 0.23799516260623932 2023-01-23 22:02:32.766367: step: 2274/1541, loss: 6.085029963287525e-05 2023-01-23 22:02:33.674517: step: 2276/1541, loss: 3.661730443127453e-05 2023-01-23 22:02:34.578977: step: 2278/1541, loss: 0.025441095232963562 2023-01-23 22:02:35.431256: step: 2280/1541, loss: 0.003920901566743851 2023-01-23 22:02:36.323856: step: 2282/1541, loss: 0.011842354200780392 2023-01-23 22:02:37.197108: step: 2284/1541, loss: 0.0031175557523965836 2023-01-23 22:02:38.095605: step: 2286/1541, loss: 6.847456097602844e-05 2023-01-23 22:02:39.009195: step: 2288/1541, loss: 0.003843576181679964 2023-01-23 22:02:39.924756: step: 2290/1541, loss: 0.00011086962331319228 2023-01-23 22:02:40.808075: step: 2292/1541, loss: 0.00012199617776786909 2023-01-23 22:02:41.679307: step: 2294/1541, loss: 0.00044470562716014683 2023-01-23 22:02:42.555441: step: 2296/1541, loss: 1.5598528989357874e-05 2023-01-23 22:02:43.459277: step: 2298/1541, loss: 0.007089586928486824 2023-01-23 22:02:44.339078: step: 2300/1541, loss: 0.2540923058986664 2023-01-23 22:02:45.221673: step: 2302/1541, loss: 0.035031601786613464 2023-01-23 22:02:46.088497: step: 2304/1541, loss: 6.132567796157673e-05 2023-01-23 22:02:46.977163: step: 2306/1541, loss: 0.0009189974516630173 2023-01-23 22:02:47.900944: step: 2308/1541, loss: 0.0005279863835312426 2023-01-23 22:02:48.797399: step: 2310/1541, loss: 0.0016870375256985426 2023-01-23 22:02:49.707888: step: 2312/1541, loss: 0.1966637670993805 2023-01-23 22:02:50.605258: step: 2314/1541, loss: 0.4183582365512848 2023-01-23 22:02:51.471098: step: 2316/1541, loss: 0.0005597309209406376 2023-01-23 22:02:52.353723: step: 2318/1541, loss: -8.94617824087618e-06 2023-01-23 22:02:53.220393: step: 2320/1541, loss: 1.0837052286660764e-05 2023-01-23 22:02:54.104442: step: 2322/1541, loss: 0.044121693819761276 2023-01-23 22:02:55.020046: step: 2324/1541, loss: 0.0006741088582202792 2023-01-23 22:02:55.920471: step: 2326/1541, loss: 0.0037520425394177437 2023-01-23 22:02:56.799164: step: 2328/1541, loss: 0.005707327276468277 2023-01-23 22:02:57.677829: step: 2330/1541, loss: 0.03054976835846901 2023-01-23 22:02:58.585953: step: 2332/1541, loss: 0.00034747610334306955 2023-01-23 22:02:59.532524: step: 2334/1541, loss: 0.011185914278030396 2023-01-23 22:03:00.454725: step: 2336/1541, loss: 8.771196007728577e-06 2023-01-23 22:03:01.433939: step: 2338/1541, loss: 0.01788564957678318 2023-01-23 22:03:02.367521: step: 2340/1541, loss: 0.03194107487797737 2023-01-23 22:03:03.380711: step: 2342/1541, loss: 0.07858928292989731 2023-01-23 22:03:04.263783: step: 2344/1541, loss: 0.08922194689512253 2023-01-23 22:03:05.108231: step: 2346/1541, loss: 0.001525932690128684 2023-01-23 22:03:05.994907: step: 2348/1541, loss: 0.001865425962023437 2023-01-23 22:03:06.876400: step: 2350/1541, loss: 0.0956285297870636 2023-01-23 22:03:07.785990: step: 2352/1541, loss: 0.004056594334542751 2023-01-23 22:03:08.683592: step: 2354/1541, loss: 1.1567225456237793 2023-01-23 22:03:09.602902: step: 2356/1541, loss: 7.833137351553887e-05 2023-01-23 22:03:10.467699: step: 2358/1541, loss: 0.006079387851059437 2023-01-23 22:03:11.358934: step: 2360/1541, loss: 0.35161685943603516 2023-01-23 22:03:12.256407: step: 2362/1541, loss: 0.0021399115212261677 2023-01-23 22:03:13.157074: step: 2364/1541, loss: 0.04560929164290428 2023-01-23 22:03:14.026407: step: 2366/1541, loss: 0.3430701494216919 2023-01-23 22:03:14.931577: step: 2368/1541, loss: 0.0016974479658529162 2023-01-23 22:03:15.786474: step: 2370/1541, loss: 0.00010608480806695297 2023-01-23 22:03:16.685128: step: 2372/1541, loss: 0.0042635537683963776 2023-01-23 22:03:17.600461: step: 2374/1541, loss: 0.020922604948282242 2023-01-23 22:03:18.462428: step: 2376/1541, loss: 0.0005378610221669078 2023-01-23 22:03:19.338153: step: 2378/1541, loss: 0.024950508028268814 2023-01-23 22:03:20.243324: step: 2380/1541, loss: 3.128051321255043e-05 2023-01-23 22:03:21.132733: step: 2382/1541, loss: 0.03497547283768654 2023-01-23 22:03:21.981010: step: 2384/1541, loss: 5.610939842881635e-05 2023-01-23 22:03:22.897992: step: 2386/1541, loss: 0.35875144600868225 2023-01-23 22:03:23.814098: step: 2388/1541, loss: 0.003252067370340228 2023-01-23 22:03:24.718377: step: 2390/1541, loss: 0.00016186980064958334 2023-01-23 22:03:25.569462: step: 2392/1541, loss: 0.00026160129345953465 2023-01-23 22:03:26.478794: step: 2394/1541, loss: 0.013583173044025898 2023-01-23 22:03:27.400183: step: 2396/1541, loss: 0.008886812254786491 2023-01-23 22:03:28.275750: step: 2398/1541, loss: 0.007418809924274683 2023-01-23 22:03:29.196330: step: 2400/1541, loss: 0.0001051468716468662 2023-01-23 22:03:30.131043: step: 2402/1541, loss: 0.004998447373509407 2023-01-23 22:03:30.972873: step: 2404/1541, loss: 0.12555190920829773 2023-01-23 22:03:31.906220: step: 2406/1541, loss: 0.07804296165704727 2023-01-23 22:03:32.857267: step: 2408/1541, loss: 0.003179053543135524 2023-01-23 22:03:33.733745: step: 2410/1541, loss: 2.148323255823925e-05 2023-01-23 22:03:34.623601: step: 2412/1541, loss: 0.001956630265340209 2023-01-23 22:03:35.600333: step: 2414/1541, loss: 0.10926549881696701 2023-01-23 22:03:36.467111: step: 2416/1541, loss: 0.00010028158430941403 2023-01-23 22:03:37.337626: step: 2418/1541, loss: 0.37903574109077454 2023-01-23 22:03:38.227509: step: 2420/1541, loss: 0.13857711851596832 2023-01-23 22:03:39.112381: step: 2422/1541, loss: 0.0025275133084505796 2023-01-23 22:03:40.047924: step: 2424/1541, loss: 0.13272103667259216 2023-01-23 22:03:40.905985: step: 2426/1541, loss: 0.00010328459757147357 2023-01-23 22:03:41.816931: step: 2428/1541, loss: 0.00015752227045595646 2023-01-23 22:03:42.725720: step: 2430/1541, loss: 0.001520770601928234 2023-01-23 22:03:43.599227: step: 2432/1541, loss: 0.0010364140616729856 2023-01-23 22:03:44.445851: step: 2434/1541, loss: 0.000496910244692117 2023-01-23 22:03:45.343327: step: 2436/1541, loss: 0.028998980298638344 2023-01-23 22:03:46.253435: step: 2438/1541, loss: 0.00010397208825452253 2023-01-23 22:03:47.153611: step: 2440/1541, loss: 0.00017430176376365125 2023-01-23 22:03:48.044868: step: 2442/1541, loss: 0.021617254242300987 2023-01-23 22:03:48.928087: step: 2444/1541, loss: 0.11612191796302795 2023-01-23 22:03:49.790655: step: 2446/1541, loss: 0.0003556250303518027 2023-01-23 22:03:50.645734: step: 2448/1541, loss: 0.015002826228737831 2023-01-23 22:03:51.543204: step: 2450/1541, loss: 0.0004142269608564675 2023-01-23 22:03:52.438606: step: 2452/1541, loss: 0.00721911434084177 2023-01-23 22:03:53.305055: step: 2454/1541, loss: 0.05685054510831833 2023-01-23 22:03:54.189852: step: 2456/1541, loss: 0.00042056868551298976 2023-01-23 22:03:55.069038: step: 2458/1541, loss: 0.00017421423399355263 2023-01-23 22:03:55.955907: step: 2460/1541, loss: 0.10434535890817642 2023-01-23 22:03:56.825777: step: 2462/1541, loss: 0.00010814132110681385 2023-01-23 22:03:57.757974: step: 2464/1541, loss: 0.0024626674130558968 2023-01-23 22:03:58.690759: step: 2466/1541, loss: 0.0013477364555001259 2023-01-23 22:03:59.577599: step: 2468/1541, loss: 0.0658196359872818 2023-01-23 22:04:00.499698: step: 2470/1541, loss: 0.0014453600160777569 2023-01-23 22:04:01.389601: step: 2472/1541, loss: 0.0011151530779898167 2023-01-23 22:04:02.354129: step: 2474/1541, loss: 0.16076849400997162 2023-01-23 22:04:03.232325: step: 2476/1541, loss: 0.00045616738498210907 2023-01-23 22:04:04.107992: step: 2478/1541, loss: 0.0008328810799866915 2023-01-23 22:04:04.956939: step: 2480/1541, loss: 0.00035219843266531825 2023-01-23 22:04:05.850022: step: 2482/1541, loss: 0.0008937357924878597 2023-01-23 22:04:06.739762: step: 2484/1541, loss: 0.0880758985877037 2023-01-23 22:04:07.581771: step: 2486/1541, loss: 0.008465582504868507 2023-01-23 22:04:08.461528: step: 2488/1541, loss: 0.0015241022920235991 2023-01-23 22:04:09.370133: step: 2490/1541, loss: 5.279145989334211e-05 2023-01-23 22:04:10.201382: step: 2492/1541, loss: 0.24602022767066956 2023-01-23 22:04:11.109345: step: 2494/1541, loss: 0.04701373726129532 2023-01-23 22:04:11.956993: step: 2496/1541, loss: 0.10750529170036316 2023-01-23 22:04:12.841479: step: 2498/1541, loss: 0.015254323370754719 2023-01-23 22:04:13.743239: step: 2500/1541, loss: 0.0005216538556851447 2023-01-23 22:04:14.656544: step: 2502/1541, loss: 0.00817252229899168 2023-01-23 22:04:15.536425: step: 2504/1541, loss: 0.00047472515143454075 2023-01-23 22:04:16.480846: step: 2506/1541, loss: 1.3849300146102905 2023-01-23 22:04:17.366615: step: 2508/1541, loss: 0.00014368393749464303 2023-01-23 22:04:18.288208: step: 2510/1541, loss: 0.00043253821786493063 2023-01-23 22:04:19.151714: step: 2512/1541, loss: 0.03108859620988369 2023-01-23 22:04:20.050429: step: 2514/1541, loss: 0.16198977828025818 2023-01-23 22:04:20.985699: step: 2516/1541, loss: 0.0002845108392648399 2023-01-23 22:04:21.865918: step: 2518/1541, loss: 0.0008612257661297917 2023-01-23 22:04:22.765115: step: 2520/1541, loss: 0.03551581874489784 2023-01-23 22:04:23.675585: step: 2522/1541, loss: 0.0029981096740812063 2023-01-23 22:04:24.560477: step: 2524/1541, loss: 0.030045997351408005 2023-01-23 22:04:25.483964: step: 2526/1541, loss: 8.264381176559255e-05 2023-01-23 22:04:26.379646: step: 2528/1541, loss: 0.3199867606163025 2023-01-23 22:04:27.321808: step: 2530/1541, loss: 0.0010030069388449192 2023-01-23 22:04:28.214055: step: 2532/1541, loss: 0.0003024604811798781 2023-01-23 22:04:29.124339: step: 2534/1541, loss: 0.00981091521680355 2023-01-23 22:04:30.035908: step: 2536/1541, loss: 0.055391956120729446 2023-01-23 22:04:30.918406: step: 2538/1541, loss: 0.002036556601524353 2023-01-23 22:04:31.753321: step: 2540/1541, loss: 0.0010598561493679881 2023-01-23 22:04:32.701331: step: 2542/1541, loss: 0.00021413623471744359 2023-01-23 22:04:33.578315: step: 2544/1541, loss: 0.0205441415309906 2023-01-23 22:04:34.474598: step: 2546/1541, loss: 0.17105194926261902 2023-01-23 22:04:35.344249: step: 2548/1541, loss: 0.00015815695223864168 2023-01-23 22:04:36.228933: step: 2550/1541, loss: 0.02962198480963707 2023-01-23 22:04:37.107973: step: 2552/1541, loss: -2.5400990125490353e-05 2023-01-23 22:04:37.999008: step: 2554/1541, loss: 0.032592061907052994 2023-01-23 22:04:38.917671: step: 2556/1541, loss: 0.0002746175741776824 2023-01-23 22:04:39.768894: step: 2558/1541, loss: 0.0033703322988003492 2023-01-23 22:04:40.667145: step: 2560/1541, loss: 0.00024217517056968063 2023-01-23 22:04:41.565786: step: 2562/1541, loss: 0.0002788621059153229 2023-01-23 22:04:42.441404: step: 2564/1541, loss: 2.8350921638775617e-05 2023-01-23 22:04:43.378484: step: 2566/1541, loss: 0.007003528531640768 2023-01-23 22:04:44.345862: step: 2568/1541, loss: 0.05624701455235481 2023-01-23 22:04:45.165513: step: 2570/1541, loss: 0.0001948480203282088 2023-01-23 22:04:46.053915: step: 2572/1541, loss: 0.00024359324015676975 2023-01-23 22:04:46.905121: step: 2574/1541, loss: 0.3603677451610565 2023-01-23 22:04:47.769028: step: 2576/1541, loss: 0.15603189170360565 2023-01-23 22:04:48.658598: step: 2578/1541, loss: 0.03452206030488014 2023-01-23 22:04:49.517835: step: 2580/1541, loss: 0.0014821324730291963 2023-01-23 22:04:50.408647: step: 2582/1541, loss: 0.017690733075141907 2023-01-23 22:04:51.387775: step: 2584/1541, loss: 0.001106490846723318 2023-01-23 22:04:52.316846: step: 2586/1541, loss: 0.11582418531179428 2023-01-23 22:04:53.176897: step: 2588/1541, loss: 3.7778165278723463e-05 2023-01-23 22:04:54.091351: step: 2590/1541, loss: 0.034097228199243546 2023-01-23 22:04:54.968584: step: 2592/1541, loss: 0.000445595127530396 2023-01-23 22:04:55.864044: step: 2594/1541, loss: 0.0043688686564564705 2023-01-23 22:04:56.762774: step: 2596/1541, loss: 0.0001728531060507521 2023-01-23 22:04:57.609729: step: 2598/1541, loss: 0.007360911462455988 2023-01-23 22:04:58.503869: step: 2600/1541, loss: 0.031991589814424515 2023-01-23 22:04:59.370889: step: 2602/1541, loss: 0.00019460530893411487 2023-01-23 22:05:00.310595: step: 2604/1541, loss: 0.0006410708301700652 2023-01-23 22:05:01.194231: step: 2606/1541, loss: 0.018469301983714104 2023-01-23 22:05:02.054216: step: 2608/1541, loss: 0.0453878752887249 2023-01-23 22:05:02.940353: step: 2610/1541, loss: 0.0001816697622416541 2023-01-23 22:05:03.818751: step: 2612/1541, loss: 0.0004558000073302537 2023-01-23 22:05:04.807221: step: 2614/1541, loss: 0.000133661218569614 2023-01-23 22:05:05.710785: step: 2616/1541, loss: 0.023418845608830452 2023-01-23 22:05:06.617458: step: 2618/1541, loss: 0.0008727509994059801 2023-01-23 22:05:07.511955: step: 2620/1541, loss: 7.089673454174772e-05 2023-01-23 22:05:08.400512: step: 2622/1541, loss: 0.0013469872064888477 2023-01-23 22:05:09.284278: step: 2624/1541, loss: 0.0004302930901758373 2023-01-23 22:05:10.158385: step: 2626/1541, loss: 0.0010568778961896896 2023-01-23 22:05:11.042980: step: 2628/1541, loss: 3.7527632230194286e-05 2023-01-23 22:05:11.967818: step: 2630/1541, loss: 0.00024172534176614136 2023-01-23 22:05:12.884098: step: 2632/1541, loss: 0.00025058817118406296 2023-01-23 22:05:13.737961: step: 2634/1541, loss: 0.00020937171939294785 2023-01-23 22:05:14.634881: step: 2636/1541, loss: 0.00012577467714436352 2023-01-23 22:05:15.569159: step: 2638/1541, loss: 0.006020988337695599 2023-01-23 22:05:16.468437: step: 2640/1541, loss: 0.00018171645933762193 2023-01-23 22:05:17.340090: step: 2642/1541, loss: 0.00029486860148608685 2023-01-23 22:05:18.261760: step: 2644/1541, loss: 0.0007430591504089534 2023-01-23 22:05:19.276809: step: 2646/1541, loss: 0.18891312181949615 2023-01-23 22:05:20.193913: step: 2648/1541, loss: 0.007023844867944717 2023-01-23 22:05:21.134512: step: 2650/1541, loss: 0.20283055305480957 2023-01-23 22:05:22.017391: step: 2652/1541, loss: 0.0017959220567718148 2023-01-23 22:05:22.937314: step: 2654/1541, loss: 0.010407458990812302 2023-01-23 22:05:23.823733: step: 2656/1541, loss: 0.014381791464984417 2023-01-23 22:05:24.780445: step: 2658/1541, loss: 0.00018616943270899355 2023-01-23 22:05:25.637437: step: 2660/1541, loss: 3.304540950921364e-05 2023-01-23 22:05:26.510707: step: 2662/1541, loss: 0.0012514773989096284 2023-01-23 22:05:27.404960: step: 2664/1541, loss: 0.027248186990618706 2023-01-23 22:05:28.288011: step: 2666/1541, loss: 0.00022776082914788276 2023-01-23 22:05:29.189194: step: 2668/1541, loss: 0.48594459891319275 2023-01-23 22:05:30.071386: step: 2670/1541, loss: 0.0005461490945890546 2023-01-23 22:05:31.003118: step: 2672/1541, loss: 0.008000862784683704 2023-01-23 22:05:31.924476: step: 2674/1541, loss: 0.021853206679224968 2023-01-23 22:05:32.808955: step: 2676/1541, loss: 0.00165216822642833 2023-01-23 22:05:33.721018: step: 2678/1541, loss: 0.0003407131298445165 2023-01-23 22:05:34.586306: step: 2680/1541, loss: 4.7381643526023254e-05 2023-01-23 22:05:35.468633: step: 2682/1541, loss: 0.0009278933284804225 2023-01-23 22:05:36.363871: step: 2684/1541, loss: 6.014307655277662e-05 2023-01-23 22:05:37.267120: step: 2686/1541, loss: 0.006223183590918779 2023-01-23 22:05:38.169542: step: 2688/1541, loss: 0.3775002956390381 2023-01-23 22:05:39.079445: step: 2690/1541, loss: 0.00015701763913966715 2023-01-23 22:05:40.022417: step: 2692/1541, loss: 0.10078498721122742 2023-01-23 22:05:40.886177: step: 2694/1541, loss: 0.0001263270532945171 2023-01-23 22:05:41.749535: step: 2696/1541, loss: 2.7421208869782276e-05 2023-01-23 22:05:42.659861: step: 2698/1541, loss: 0.004979076329618692 2023-01-23 22:05:43.525762: step: 2700/1541, loss: 0.27597174048423767 2023-01-23 22:05:44.401910: step: 2702/1541, loss: 0.00015535939019173384 2023-01-23 22:05:45.307235: step: 2704/1541, loss: 0.001352847320958972 2023-01-23 22:05:46.221301: step: 2706/1541, loss: -6.56258544040611e-06 2023-01-23 22:05:47.149024: step: 2708/1541, loss: 0.009591053240001202 2023-01-23 22:05:48.023938: step: 2710/1541, loss: -2.226793731097132e-05 2023-01-23 22:05:48.953930: step: 2712/1541, loss: 0.00016687563038431108 2023-01-23 22:05:49.864568: step: 2714/1541, loss: 0.00024738823412917554 2023-01-23 22:05:50.773988: step: 2716/1541, loss: 0.061746105551719666 2023-01-23 22:05:51.666206: step: 2718/1541, loss: 0.03973909467458725 2023-01-23 22:05:52.578264: step: 2720/1541, loss: 6.279128137975931e-05 2023-01-23 22:05:53.433327: step: 2722/1541, loss: 0.00034814770333468914 2023-01-23 22:05:54.334902: step: 2724/1541, loss: 0.00015211105346679688 2023-01-23 22:05:55.227283: step: 2726/1541, loss: 0.020184455439448357 2023-01-23 22:05:56.112698: step: 2728/1541, loss: 0.00016705639427527785 2023-01-23 22:05:57.005152: step: 2730/1541, loss: 9.56423464231193e-05 2023-01-23 22:05:57.895302: step: 2732/1541, loss: 0.09444639086723328 2023-01-23 22:05:58.805385: step: 2734/1541, loss: 3.7139841879252344e-05 2023-01-23 22:05:59.693141: step: 2736/1541, loss: 0.005013088695704937 2023-01-23 22:06:00.570064: step: 2738/1541, loss: 0.0008066302980296314 2023-01-23 22:06:01.440130: step: 2740/1541, loss: 0.4990866780281067 2023-01-23 22:06:02.348550: step: 2742/1541, loss: 0.0029682519380003214 2023-01-23 22:06:03.208890: step: 2744/1541, loss: 0.001078995643183589 2023-01-23 22:06:04.075453: step: 2746/1541, loss: 0.22785034775733948 2023-01-23 22:06:05.027010: step: 2748/1541, loss: 0.010130112990736961 2023-01-23 22:06:05.893686: step: 2750/1541, loss: 0.178532212972641 2023-01-23 22:06:06.815571: step: 2752/1541, loss: 0.03984030708670616 2023-01-23 22:06:07.710895: step: 2754/1541, loss: 0.01670350693166256 2023-01-23 22:06:08.616045: step: 2756/1541, loss: 3.218999700038694e-05 2023-01-23 22:06:09.491471: step: 2758/1541, loss: 0.00673210434615612 2023-01-23 22:06:10.337955: step: 2760/1541, loss: 0.001493147574365139 2023-01-23 22:06:11.212361: step: 2762/1541, loss: 0.0007457794272340834 2023-01-23 22:06:12.075763: step: 2764/1541, loss: 0.0001980549277504906 2023-01-23 22:06:12.937301: step: 2766/1541, loss: 0.0004067328409291804 2023-01-23 22:06:13.847759: step: 2768/1541, loss: 0.0022239401005208492 2023-01-23 22:06:14.782581: step: 2770/1541, loss: 0.0003048975777346641 2023-01-23 22:06:15.692195: step: 2772/1541, loss: 0.0005470752948895097 2023-01-23 22:06:16.574947: step: 2774/1541, loss: 0.000986285274848342 2023-01-23 22:06:17.431263: step: 2776/1541, loss: 0.03697298839688301 2023-01-23 22:06:18.313847: step: 2778/1541, loss: 0.00010013394057750702 2023-01-23 22:06:19.230065: step: 2780/1541, loss: 0.0022140247747302055 2023-01-23 22:06:20.125859: step: 2782/1541, loss: 0.00012351683108136058 2023-01-23 22:06:21.007460: step: 2784/1541, loss: 0.0005913127097301185 2023-01-23 22:06:21.940067: step: 2786/1541, loss: 0.146463081240654 2023-01-23 22:06:22.822466: step: 2788/1541, loss: 0.004718639887869358 2023-01-23 22:06:23.752137: step: 2790/1541, loss: 0.03886730596423149 2023-01-23 22:06:24.655803: step: 2792/1541, loss: 0.0025842078030109406 2023-01-23 22:06:25.531256: step: 2794/1541, loss: 0.030401812866330147 2023-01-23 22:06:26.451775: step: 2796/1541, loss: 0.021216347813606262 2023-01-23 22:06:27.370662: step: 2798/1541, loss: 0.0006953456322662532 2023-01-23 22:06:28.303602: step: 2800/1541, loss: 0.03500879183411598 2023-01-23 22:06:29.231061: step: 2802/1541, loss: 0.04321516305208206 2023-01-23 22:06:30.092491: step: 2804/1541, loss: 0.00029358439496718347 2023-01-23 22:06:30.947402: step: 2806/1541, loss: 0.0009337039664387703 2023-01-23 22:06:31.873647: step: 2808/1541, loss: 6.060022133169696e-05 2023-01-23 22:06:32.764033: step: 2810/1541, loss: 0.00042668948299251497 2023-01-23 22:06:33.677905: step: 2812/1541, loss: -3.0231412893044762e-06 2023-01-23 22:06:34.570554: step: 2814/1541, loss: 0.0010835160501301289 2023-01-23 22:06:35.476954: step: 2816/1541, loss: 0.0020058543886989355 2023-01-23 22:06:36.343918: step: 2818/1541, loss: 6.981700425967574e-05 2023-01-23 22:06:37.212346: step: 2820/1541, loss: 4.617365630110726e-05 2023-01-23 22:06:38.097864: step: 2822/1541, loss: 0.00019911072740796953 2023-01-23 22:06:38.988377: step: 2824/1541, loss: 0.0002573407837189734 2023-01-23 22:06:39.870228: step: 2826/1541, loss: 6.530979590024799e-05 2023-01-23 22:06:40.795565: step: 2828/1541, loss: 0.0008546771132387221 2023-01-23 22:06:41.674554: step: 2830/1541, loss: 0.008643425069749355 2023-01-23 22:06:42.568787: step: 2832/1541, loss: 8.373019227292389e-05 2023-01-23 22:06:43.483725: step: 2834/1541, loss: 9.943058103090152e-05 2023-01-23 22:06:44.424255: step: 2836/1541, loss: 7.399198511848226e-05 2023-01-23 22:06:45.330997: step: 2838/1541, loss: 0.08897028118371964 2023-01-23 22:06:46.215853: step: 2840/1541, loss: 0.0010030139237642288 2023-01-23 22:06:47.095400: step: 2842/1541, loss: 5.3167932492215186e-05 2023-01-23 22:06:48.017113: step: 2844/1541, loss: 0.0001674178201938048 2023-01-23 22:06:48.912933: step: 2846/1541, loss: 1.9512312064762227e-05 2023-01-23 22:06:49.823045: step: 2848/1541, loss: 0.2544457018375397 2023-01-23 22:06:50.720246: step: 2850/1541, loss: 0.00013259791012387723 2023-01-23 22:06:51.660989: step: 2852/1541, loss: 0.01850077696144581 2023-01-23 22:06:52.593092: step: 2854/1541, loss: 0.20591045916080475 2023-01-23 22:06:53.518267: step: 2856/1541, loss: 0.0022346312180161476 2023-01-23 22:06:54.475877: step: 2858/1541, loss: 0.0001855206210166216 2023-01-23 22:06:55.335822: step: 2860/1541, loss: 3.9072707295417786e-05 2023-01-23 22:06:56.205559: step: 2862/1541, loss: 0.027178417891263962 2023-01-23 22:06:57.083247: step: 2864/1541, loss: 0.0026774937286973 2023-01-23 22:06:57.987890: step: 2866/1541, loss: 0.001481899176724255 2023-01-23 22:06:58.959407: step: 2868/1541, loss: 0.0265927966684103 2023-01-23 22:06:59.887809: step: 2870/1541, loss: 0.013028019107878208 2023-01-23 22:07:00.723301: step: 2872/1541, loss: 5.201127351028845e-06 2023-01-23 22:07:01.663553: step: 2874/1541, loss: 0.029165929183363914 2023-01-23 22:07:02.605970: step: 2876/1541, loss: 0.00011285378423053771 2023-01-23 22:07:03.527461: step: 2878/1541, loss: 0.000590125157032162 2023-01-23 22:07:04.451224: step: 2880/1541, loss: 0.03002028726041317 2023-01-23 22:07:05.329663: step: 2882/1541, loss: 3.337252564961091e-05 2023-01-23 22:07:06.226092: step: 2884/1541, loss: 5.021018296247348e-05 2023-01-23 22:07:07.095826: step: 2886/1541, loss: 0.0294200349599123 2023-01-23 22:07:07.992718: step: 2888/1541, loss: 0.009101065807044506 2023-01-23 22:07:08.854560: step: 2890/1541, loss: 6.384099106071517e-05 2023-01-23 22:07:09.737673: step: 2892/1541, loss: 4.333809920353815e-05 2023-01-23 22:07:10.617929: step: 2894/1541, loss: 0.31360501050949097 2023-01-23 22:07:11.483068: step: 2896/1541, loss: 2.3821172362659127e-05 2023-01-23 22:07:12.366097: step: 2898/1541, loss: 0.0051511358469724655 2023-01-23 22:07:13.264769: step: 2900/1541, loss: 0.006713321898132563 2023-01-23 22:07:14.127000: step: 2902/1541, loss: 0.0024379189126193523 2023-01-23 22:07:15.016944: step: 2904/1541, loss: 0.0002754867309704423 2023-01-23 22:07:15.926171: step: 2906/1541, loss: 0.0004341627936810255 2023-01-23 22:07:16.830462: step: 2908/1541, loss: 1.8086398995365016e-05 2023-01-23 22:07:17.728043: step: 2910/1541, loss: 0.7070914506912231 2023-01-23 22:07:18.624561: step: 2912/1541, loss: 9.199198393616825e-05 2023-01-23 22:07:19.539807: step: 2914/1541, loss: 0.13714700937271118 2023-01-23 22:07:20.408184: step: 2916/1541, loss: 0.00011442758841440082 2023-01-23 22:07:21.298157: step: 2918/1541, loss: 0.11344889551401138 2023-01-23 22:07:22.227152: step: 2920/1541, loss: 1.3224631402408704e-05 2023-01-23 22:07:23.121855: step: 2922/1541, loss: 0.0006130983238108456 2023-01-23 22:07:23.984778: step: 2924/1541, loss: 4.249736593919806e-05 2023-01-23 22:07:24.864185: step: 2926/1541, loss: 0.00107492262031883 2023-01-23 22:07:25.733710: step: 2928/1541, loss: 0.10534854233264923 2023-01-23 22:07:26.703962: step: 2930/1541, loss: 0.0107411565259099 2023-01-23 22:07:27.633666: step: 2932/1541, loss: -1.426814833394019e-05 2023-01-23 22:07:28.611758: step: 2934/1541, loss: 0.13585802912712097 2023-01-23 22:07:29.495912: step: 2936/1541, loss: 0.5413711667060852 2023-01-23 22:07:30.405856: step: 2938/1541, loss: 0.00037291605258360505 2023-01-23 22:07:31.286003: step: 2940/1541, loss: 0.0004168607119936496 2023-01-23 22:07:32.223806: step: 2942/1541, loss: 0.3373461365699768 2023-01-23 22:07:33.078100: step: 2944/1541, loss: 0.006047649774700403 2023-01-23 22:07:33.976446: step: 2946/1541, loss: -1.1921197256015148e-06 2023-01-23 22:07:34.864224: step: 2948/1541, loss: 0.0004052415315527469 2023-01-23 22:07:35.749483: step: 2950/1541, loss: 0.021466756239533424 2023-01-23 22:07:36.647585: step: 2952/1541, loss: 0.31029945611953735 2023-01-23 22:07:37.530903: step: 2954/1541, loss: 0.00015313203039113432 2023-01-23 22:07:38.437048: step: 2956/1541, loss: 9.75299408310093e-05 2023-01-23 22:07:39.315118: step: 2958/1541, loss: 1.3113401109876577e-05 2023-01-23 22:07:40.161967: step: 2960/1541, loss: 6.289034354267642e-05 2023-01-23 22:07:41.011134: step: 2962/1541, loss: 0.050025008618831635 2023-01-23 22:07:41.952864: step: 2964/1541, loss: 3.7634097679983824e-05 2023-01-23 22:07:42.866184: step: 2966/1541, loss: 0.0005988865741528571 2023-01-23 22:07:43.781868: step: 2968/1541, loss: 0.01456229668110609 2023-01-23 22:07:44.698491: step: 2970/1541, loss: 8.751671703066677e-05 2023-01-23 22:07:45.556959: step: 2972/1541, loss: 0.03285932168364525 2023-01-23 22:07:46.424183: step: 2974/1541, loss: 0.006559044122695923 2023-01-23 22:07:47.321050: step: 2976/1541, loss: 0.056223466992378235 2023-01-23 22:07:48.216725: step: 2978/1541, loss: 0.0052852933295071125 2023-01-23 22:07:49.061745: step: 2980/1541, loss: 0.0001661235437495634 2023-01-23 22:07:49.940403: step: 2982/1541, loss: 0.0016315209213644266 2023-01-23 22:07:50.856881: step: 2984/1541, loss: 0.10995081812143326 2023-01-23 22:07:51.732705: step: 2986/1541, loss: 6.670757284155115e-05 2023-01-23 22:07:52.599209: step: 2988/1541, loss: 0.02030481956899166 2023-01-23 22:07:53.507059: step: 2990/1541, loss: 0.00011127491598017514 2023-01-23 22:07:54.378982: step: 2992/1541, loss: 0.20208069682121277 2023-01-23 22:07:55.288970: step: 2994/1541, loss: 0.10522051155567169 2023-01-23 22:07:56.178191: step: 2996/1541, loss: 0.0008355328463949263 2023-01-23 22:07:57.042486: step: 2998/1541, loss: 0.25533372163772583 2023-01-23 22:07:57.962333: step: 3000/1541, loss: 0.002533651888370514 2023-01-23 22:07:58.858358: step: 3002/1541, loss: 0.0185361597687006 2023-01-23 22:07:59.744411: step: 3004/1541, loss: 0.23105502128601074 2023-01-23 22:08:00.631565: step: 3006/1541, loss: 8.819230424705893e-05 2023-01-23 22:08:01.504079: step: 3008/1541, loss: 0.009635884314775467 2023-01-23 22:08:02.423482: step: 3010/1541, loss: 0.0003724170965142548 2023-01-23 22:08:03.277826: step: 3012/1541, loss: 0.00014017685316503048 2023-01-23 22:08:04.268597: step: 3014/1541, loss: 0.0004435106529854238 2023-01-23 22:08:05.199274: step: 3016/1541, loss: 0.027558594942092896 2023-01-23 22:08:06.146605: step: 3018/1541, loss: 0.017209909856319427 2023-01-23 22:08:07.070700: step: 3020/1541, loss: 0.00018437430844642222 2023-01-23 22:08:07.924009: step: 3022/1541, loss: 0.04907499626278877 2023-01-23 22:08:08.797530: step: 3024/1541, loss: 0.02687055617570877 2023-01-23 22:08:09.679519: step: 3026/1541, loss: 0.0450243316590786 2023-01-23 22:08:10.571315: step: 3028/1541, loss: 0.0016621325630694628 2023-01-23 22:08:11.446201: step: 3030/1541, loss: 0.0018584049539640546 2023-01-23 22:08:12.358759: step: 3032/1541, loss: 0.0020319162867963314 2023-01-23 22:08:13.252235: step: 3034/1541, loss: 4.477800757740624e-05 2023-01-23 22:08:14.179749: step: 3036/1541, loss: 0.024969954043626785 2023-01-23 22:08:15.076479: step: 3038/1541, loss: 0.0003178988117724657 2023-01-23 22:08:15.952965: step: 3040/1541, loss: 0.0005339828203432262 2023-01-23 22:08:16.900189: step: 3042/1541, loss: 0.3860003352165222 2023-01-23 22:08:17.773345: step: 3044/1541, loss: 0.007321869023144245 2023-01-23 22:08:18.619677: step: 3046/1541, loss: 0.189614400267601 2023-01-23 22:08:19.517688: step: 3048/1541, loss: 0.45220378041267395 2023-01-23 22:08:20.433413: step: 3050/1541, loss: 0.0005173876415938139 2023-01-23 22:08:21.365746: step: 3052/1541, loss: 0.007946720346808434 2023-01-23 22:08:22.307929: step: 3054/1541, loss: 0.0028107790276408195 2023-01-23 22:08:23.166097: step: 3056/1541, loss: 0.035941388458013535 2023-01-23 22:08:24.107612: step: 3058/1541, loss: 0.063249871134758 2023-01-23 22:08:25.000507: step: 3060/1541, loss: 7.152745092753321e-05 2023-01-23 22:08:25.861222: step: 3062/1541, loss: 0.11345978081226349 2023-01-23 22:08:26.723036: step: 3064/1541, loss: 0.00031880554161034524 2023-01-23 22:08:27.614057: step: 3066/1541, loss: 0.019233014434576035 2023-01-23 22:08:28.515915: step: 3068/1541, loss: 0.0003972281119786203 2023-01-23 22:08:29.395167: step: 3070/1541, loss: 0.02862180583178997 2023-01-23 22:08:30.324257: step: 3072/1541, loss: 0.0002613654942251742 2023-01-23 22:08:31.210223: step: 3074/1541, loss: 0.0008835738408379257 2023-01-23 22:08:32.087820: step: 3076/1541, loss: 0.0032507674768567085 2023-01-23 22:08:32.961787: step: 3078/1541, loss: 0.0008463131380267441 2023-01-23 22:08:33.828292: step: 3080/1541, loss: 0.00048805694677866995 2023-01-23 22:08:34.718235: step: 3082/1541, loss: 0.0016154232434928417 2023-01-23 22:08:35.336271: step: 3084/1541, loss: 0.0002987682819366455 ================================================== Loss: 0.052 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6873583196168412, 'r': 0.6505926420559405, 'f1': 0.6684703371184095}, 'combined': 0.4419986485984446, 'epoch': 35} Test Chinese: {'event': {'p': 0.6114616193480547, 'r': 0.6968244457759137, 'f1': 0.6513581629795576}, 'argument': {'p': 0.7268315472720632, 'r': 0.6966241843887183, 'f1': 0.7114073483910106}, 'combined': 0.4633809835781268, 'epoch': 35} Dev Korean: {'event': {'p': 0.5725490196078431, 'r': 0.7776298268974701, 'f1': 0.6595143986448334}, 'argument': {'p': 0.6815636922678625, 'r': 0.6571173618421859, 'f1': 0.6691173136654546}, 'combined': 0.44129250274491866, 'epoch': 35} Test Korean: {'event': {'p': 0.6124880838894184, 'r': 0.7699221090473337, 'f1': 0.6822405096894081}, 'argument': {'p': 0.7243608933407139, 'r': 0.7081002489444298, 'f1': 0.7161382794397195}, 'combined': 0.48857854477305, 'epoch': 35} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.687408505687428, 'r': 0.6599949234440617, 'f1': 0.673422843191676}, 'combined': 0.46696279910344435, 'epoch': 35} Test Russian: {'event': {'p': 0.6493982208267922, 'r': 0.7435590173756741, 'f1': 0.6932960893854749}, 'argument': {'p': 0.725344132470029, 'r': 0.7032379111505768, 'f1': 0.7141199833238517}, 'combined': 0.49509659179044696, 'epoch': 35} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6881789594053744, 'r': 0.6455484043979618, 'f1': 0.6661823716618236}, 'combined': 0.4896725125035626, 'epoch': 35} Korean: {'event': {'p': 0.5625, 'r': 0.5714285714285714, 'f1': 0.5669291338582677}, 'argument': {'p': 0.6058730158730159, 'r': 0.3883801383801384, 'f1': 0.47333829365079366}, 'combined': 0.2683492688413948, 'epoch': 35} Russian: {'event': {'p': 0.5, 'r': 0.4722222222222222, 'f1': 0.4857142857142857}, 'argument': {'p': 0.5553464266230224, 'r': 0.6070065593321408, 'f1': 0.58002849002849}, 'combined': 0.2817281237281237, 'epoch': 35} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6260454002389486, 'r': 0.6977363515312917, 'f1': 0.659949622166247}, 'argument': {'p': 0.6809153876089666, 'r': 0.6491957266955055, 'f1': 0.6646773417995319}, 'combined': 0.4386535605830664, 'epoch': 28} Test for Chinese: {'event': {'p': 0.610410094637224, 'r': 0.6956261234272019, 'f1': 0.6502380285634276}, 'argument': {'p': 0.716394349621888, 'r': 0.6884285723838208, 'f1': 0.7021331039642987}, 'combined': 0.4565536453108658, 'epoch': 28} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.6800847457627118, 'r': 0.7101769911504425, 'f1': 0.6948051948051949}, 'combined': 0.5225885225885226, 'epoch': 28} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 36 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 22:13:56.897337: step: 2/1541, loss: 0.0030942445155233145 2023-01-23 22:13:57.847419: step: 4/1541, loss: 0.0004353972617536783 2023-01-23 22:13:58.698587: step: 6/1541, loss: 0.0002446018042974174 2023-01-23 22:13:59.586447: step: 8/1541, loss: 0.037282444536685944 2023-01-23 22:14:00.445593: step: 10/1541, loss: 0.18438273668289185 2023-01-23 22:14:01.346646: step: 12/1541, loss: 0.002627922222018242 2023-01-23 22:14:02.252730: step: 14/1541, loss: 0.016929352656006813 2023-01-23 22:14:03.151737: step: 16/1541, loss: 0.33366867899894714 2023-01-23 22:14:04.006819: step: 18/1541, loss: 0.0039239367470145226 2023-01-23 22:14:04.875624: step: 20/1541, loss: 0.0005712618003599346 2023-01-23 22:14:05.773999: step: 22/1541, loss: 0.013661211356520653 2023-01-23 22:14:06.667356: step: 24/1541, loss: 0.019761983305215836 2023-01-23 22:14:07.540215: step: 26/1541, loss: 0.0016843359917402267 2023-01-23 22:14:08.427450: step: 28/1541, loss: 0.0006814348744228482 2023-01-23 22:14:09.324644: step: 30/1541, loss: 0.0014863632386550307 2023-01-23 22:14:10.239130: step: 32/1541, loss: 0.00018827708845492452 2023-01-23 22:14:11.119187: step: 34/1541, loss: 0.00048827784485183656 2023-01-23 22:14:12.013094: step: 36/1541, loss: 0.002342923078685999 2023-01-23 22:14:12.920805: step: 38/1541, loss: 0.007260311394929886 2023-01-23 22:14:13.784037: step: 40/1541, loss: 9.395271626999602e-05 2023-01-23 22:14:14.644378: step: 42/1541, loss: 0.00010440209007356316 2023-01-23 22:14:15.543968: step: 44/1541, loss: 0.008676208555698395 2023-01-23 22:14:16.397460: step: 46/1541, loss: 0.0036195889115333557 2023-01-23 22:14:17.321580: step: 48/1541, loss: 0.003855935065075755 2023-01-23 22:14:18.286402: step: 50/1541, loss: 0.015926942229270935 2023-01-23 22:14:19.159482: step: 52/1541, loss: 0.0003967075317632407 2023-01-23 22:14:20.018044: step: 54/1541, loss: 0.11282509565353394 2023-01-23 22:14:20.908250: step: 56/1541, loss: 0.009137988090515137 2023-01-23 22:14:21.796150: step: 58/1541, loss: 0.02586805634200573 2023-01-23 22:14:22.709389: step: 60/1541, loss: 1.7664391634752974e-05 2023-01-23 22:14:23.597588: step: 62/1541, loss: 0.00015018670819699764 2023-01-23 22:14:24.482704: step: 64/1541, loss: 0.02402229607105255 2023-01-23 22:14:25.349445: step: 66/1541, loss: 4.293773235986009e-05 2023-01-23 22:14:26.226088: step: 68/1541, loss: 0.010279816575348377 2023-01-23 22:14:27.143384: step: 70/1541, loss: 0.027775729075074196 2023-01-23 22:14:28.040323: step: 72/1541, loss: 0.3264687955379486 2023-01-23 22:14:28.911565: step: 74/1541, loss: 0.0006745383725501597 2023-01-23 22:14:29.782332: step: 76/1541, loss: 0.0479138046503067 2023-01-23 22:14:30.670516: step: 78/1541, loss: 0.0003293622867204249 2023-01-23 22:14:31.548172: step: 80/1541, loss: 0.0014886759454384446 2023-01-23 22:14:32.492909: step: 82/1541, loss: 0.0002595327387098223 2023-01-23 22:14:33.361675: step: 84/1541, loss: 0.0009127259254455566 2023-01-23 22:14:34.333195: step: 86/1541, loss: 0.00019137287745252252 2023-01-23 22:14:35.253455: step: 88/1541, loss: 0.07665479928255081 2023-01-23 22:14:36.117623: step: 90/1541, loss: 0.1256372630596161 2023-01-23 22:14:37.027551: step: 92/1541, loss: 0.01851024478673935 2023-01-23 22:14:37.920524: step: 94/1541, loss: 0.0006056020502001047 2023-01-23 22:14:38.811520: step: 96/1541, loss: 0.0017419731011614203 2023-01-23 22:14:39.688936: step: 98/1541, loss: 0.000435469119111076 2023-01-23 22:14:40.693876: step: 100/1541, loss: 0.05678626522421837 2023-01-23 22:14:41.584378: step: 102/1541, loss: 8.089244511211291e-05 2023-01-23 22:14:42.411902: step: 104/1541, loss: 0.0001474278251407668 2023-01-23 22:14:43.277114: step: 106/1541, loss: 0.0003790709306485951 2023-01-23 22:14:44.214037: step: 108/1541, loss: 0.04334495589137077 2023-01-23 22:14:45.097223: step: 110/1541, loss: 0.058276791125535965 2023-01-23 22:14:46.008455: step: 112/1541, loss: 0.00032276156707666814 2023-01-23 22:14:46.923418: step: 114/1541, loss: 0.0012041166191920638 2023-01-23 22:14:47.810697: step: 116/1541, loss: 0.00027757964562624693 2023-01-23 22:14:48.762667: step: 118/1541, loss: 0.00030599351157434285 2023-01-23 22:14:49.630565: step: 120/1541, loss: 0.003590412437915802 2023-01-23 22:14:50.528397: step: 122/1541, loss: 2.0184570530545898e-05 2023-01-23 22:14:51.428012: step: 124/1541, loss: 0.00015543607878498733 2023-01-23 22:14:52.304170: step: 126/1541, loss: 0.0013586969580501318 2023-01-23 22:14:53.152153: step: 128/1541, loss: 4.8076512030093e-05 2023-01-23 22:14:54.066366: step: 130/1541, loss: 0.0006584798684343696 2023-01-23 22:14:54.969853: step: 132/1541, loss: 0.0010193901835009456 2023-01-23 22:14:55.851411: step: 134/1541, loss: 0.0006639219936914742 2023-01-23 22:14:56.843732: step: 136/1541, loss: -3.6418828130990732e-06 2023-01-23 22:14:57.761589: step: 138/1541, loss: 0.00013659772230312228 2023-01-23 22:14:58.660665: step: 140/1541, loss: 0.002298168372362852 2023-01-23 22:14:59.578329: step: 142/1541, loss: 0.07901471108198166 2023-01-23 22:15:00.440382: step: 144/1541, loss: 0.025281835347414017 2023-01-23 22:15:01.366954: step: 146/1541, loss: 3.710496821440756e-05 2023-01-23 22:15:02.283179: step: 148/1541, loss: 3.2861524232430384e-05 2023-01-23 22:15:03.164572: step: 150/1541, loss: 0.029877236112952232 2023-01-23 22:15:04.031736: step: 152/1541, loss: 7.844468200346455e-05 2023-01-23 22:15:04.939352: step: 154/1541, loss: 0.10012571513652802 2023-01-23 22:15:05.810596: step: 156/1541, loss: 3.804897642112337e-05 2023-01-23 22:15:06.710319: step: 158/1541, loss: 0.0006196210160851479 2023-01-23 22:15:07.631050: step: 160/1541, loss: 6.700069206999615e-05 2023-01-23 22:15:08.562686: step: 162/1541, loss: 0.0015839914558455348 2023-01-23 22:15:09.488659: step: 164/1541, loss: 0.00938186515122652 2023-01-23 22:15:10.354993: step: 166/1541, loss: 2.7419253456173465e-05 2023-01-23 22:15:11.250907: step: 168/1541, loss: 9.732714534038678e-05 2023-01-23 22:15:12.146062: step: 170/1541, loss: 3.6181812902214006e-05 2023-01-23 22:15:12.983168: step: 172/1541, loss: 2.2130212528281845e-05 2023-01-23 22:15:13.865440: step: 174/1541, loss: 0.10957687348127365 2023-01-23 22:15:14.725868: step: 176/1541, loss: -5.980097739666235e-06 2023-01-23 22:15:15.634701: step: 178/1541, loss: 0.030443184077739716 2023-01-23 22:15:16.572526: step: 180/1541, loss: 0.02193119004368782 2023-01-23 22:15:17.450472: step: 182/1541, loss: 0.18140728771686554 2023-01-23 22:15:18.402132: step: 184/1541, loss: 0.0008679084712639451 2023-01-23 22:15:19.309375: step: 186/1541, loss: 0.17810115218162537 2023-01-23 22:15:20.203952: step: 188/1541, loss: 0.013510193675756454 2023-01-23 22:15:21.115784: step: 190/1541, loss: 0.00041349243838340044 2023-01-23 22:15:22.048012: step: 192/1541, loss: 0.000262884161202237 2023-01-23 22:15:22.930929: step: 194/1541, loss: 0.005691021215170622 2023-01-23 22:15:23.831212: step: 196/1541, loss: 8.097886893665418e-05 2023-01-23 22:15:24.704058: step: 198/1541, loss: 4.683130100602284e-05 2023-01-23 22:15:25.619974: step: 200/1541, loss: 0.10116809606552124 2023-01-23 22:15:26.509959: step: 202/1541, loss: 0.19413655996322632 2023-01-23 22:15:27.417196: step: 204/1541, loss: 0.00020906458667013794 2023-01-23 22:15:28.364622: step: 206/1541, loss: 0.24671141803264618 2023-01-23 22:15:29.260902: step: 208/1541, loss: 6.535172724397853e-05 2023-01-23 22:15:30.179400: step: 210/1541, loss: 3.237924101995304e-05 2023-01-23 22:15:31.083614: step: 212/1541, loss: 0.099099300801754 2023-01-23 22:15:32.040532: step: 214/1541, loss: 0.13294151425361633 2023-01-23 22:15:32.954386: step: 216/1541, loss: 8.315609011333436e-05 2023-01-23 22:15:33.809602: step: 218/1541, loss: 0.00028853744152002037 2023-01-23 22:15:34.707940: step: 220/1541, loss: 0.0009988570818677545 2023-01-23 22:15:35.591178: step: 222/1541, loss: 0.0004036329919472337 2023-01-23 22:15:36.485411: step: 224/1541, loss: -2.090718044200912e-05 2023-01-23 22:15:37.392832: step: 226/1541, loss: 0.03373953700065613 2023-01-23 22:15:38.267501: step: 228/1541, loss: 0.0038307791110128164 2023-01-23 22:15:39.152120: step: 230/1541, loss: 5.227802830631845e-05 2023-01-23 22:15:40.039333: step: 232/1541, loss: 5.545578824239783e-05 2023-01-23 22:15:40.915351: step: 234/1541, loss: 2.4534263502573594e-05 2023-01-23 22:15:41.851883: step: 236/1541, loss: 0.013393130153417587 2023-01-23 22:15:42.727698: step: 238/1541, loss: 0.14388258755207062 2023-01-23 22:15:43.600460: step: 240/1541, loss: 0.0006198324845172465 2023-01-23 22:15:44.528848: step: 242/1541, loss: 0.00010441539052408189 2023-01-23 22:15:45.448660: step: 244/1541, loss: 0.05995125323534012 2023-01-23 22:15:46.358598: step: 246/1541, loss: 0.014686857350170612 2023-01-23 22:15:47.228577: step: 248/1541, loss: 0.00014221889432519674 2023-01-23 22:15:48.101357: step: 250/1541, loss: 0.0012109453091397882 2023-01-23 22:15:48.973682: step: 252/1541, loss: 0.01894582249224186 2023-01-23 22:15:49.854133: step: 254/1541, loss: 0.13978198170661926 2023-01-23 22:15:50.745349: step: 256/1541, loss: 0.0013805602211505175 2023-01-23 22:15:51.654090: step: 258/1541, loss: 0.0013496350729838014 2023-01-23 22:15:52.641000: step: 260/1541, loss: 0.000377546064555645 2023-01-23 22:15:53.541770: step: 262/1541, loss: 0.00036588654620572925 2023-01-23 22:15:54.438421: step: 264/1541, loss: 0.00010129263682756573 2023-01-23 22:15:55.350338: step: 266/1541, loss: 0.06805410236120224 2023-01-23 22:15:56.263071: step: 268/1541, loss: 0.0439826063811779 2023-01-23 22:15:57.137487: step: 270/1541, loss: 0.0693521648645401 2023-01-23 22:15:58.031032: step: 272/1541, loss: 0.0010621682740747929 2023-01-23 22:15:59.008248: step: 274/1541, loss: 0.0002184978366130963 2023-01-23 22:15:59.888679: step: 276/1541, loss: 0.0006104329950176179 2023-01-23 22:16:00.794916: step: 278/1541, loss: 0.000545141811016947 2023-01-23 22:16:01.710150: step: 280/1541, loss: 0.3966052532196045 2023-01-23 22:16:02.659104: step: 282/1541, loss: 0.00031983188819140196 2023-01-23 22:16:03.584474: step: 284/1541, loss: 0.5707733631134033 2023-01-23 22:16:04.461203: step: 286/1541, loss: 0.0004047142865601927 2023-01-23 22:16:05.406711: step: 288/1541, loss: 0.01477112341672182 2023-01-23 22:16:06.308953: step: 290/1541, loss: 6.563588976860046e-05 2023-01-23 22:16:07.182430: step: 292/1541, loss: 0.00596232945099473 2023-01-23 22:16:08.069771: step: 294/1541, loss: 0.0048187472857534885 2023-01-23 22:16:08.941585: step: 296/1541, loss: 0.0001125522903748788 2023-01-23 22:16:09.818250: step: 298/1541, loss: 0.12379853427410126 2023-01-23 22:16:10.700926: step: 300/1541, loss: 9.12714604055509e-05 2023-01-23 22:16:11.552005: step: 302/1541, loss: 0.0006981216138228774 2023-01-23 22:16:12.479562: step: 304/1541, loss: 0.002099081641063094 2023-01-23 22:16:13.388706: step: 306/1541, loss: 5.164964750292711e-05 2023-01-23 22:16:14.241989: step: 308/1541, loss: -9.35753996600397e-06 2023-01-23 22:16:15.227042: step: 310/1541, loss: 3.4606116969371215e-05 2023-01-23 22:16:16.119607: step: 312/1541, loss: 0.0026263422332704067 2023-01-23 22:16:17.034181: step: 314/1541, loss: 0.002453223103657365 2023-01-23 22:16:17.941162: step: 316/1541, loss: 0.04077836498618126 2023-01-23 22:16:18.990772: step: 318/1541, loss: 0.0957692489027977 2023-01-23 22:16:19.870321: step: 320/1541, loss: 0.015586542896926403 2023-01-23 22:16:20.838471: step: 322/1541, loss: 0.008024788461625576 2023-01-23 22:16:21.662269: step: 324/1541, loss: 4.576415449264459e-05 2023-01-23 22:16:22.536381: step: 326/1541, loss: 0.13900119066238403 2023-01-23 22:16:23.473946: step: 328/1541, loss: 0.0011710058897733688 2023-01-23 22:16:24.331775: step: 330/1541, loss: 0.00010453402501298115 2023-01-23 22:16:25.217868: step: 332/1541, loss: 0.08828803896903992 2023-01-23 22:16:26.080788: step: 334/1541, loss: 0.0001893043954623863 2023-01-23 22:16:26.971321: step: 336/1541, loss: 0.00019611764582805336 2023-01-23 22:16:27.962272: step: 338/1541, loss: 0.009160865098237991 2023-01-23 22:16:28.855168: step: 340/1541, loss: 2.4922192096710205e-05 2023-01-23 22:16:29.812427: step: 342/1541, loss: 0.21351110935211182 2023-01-23 22:16:30.725866: step: 344/1541, loss: -4.713237558462424e-06 2023-01-23 22:16:31.607402: step: 346/1541, loss: 0.04918675497174263 2023-01-23 22:16:32.551280: step: 348/1541, loss: 0.0005532609648071229 2023-01-23 22:16:33.418014: step: 350/1541, loss: 5.648285150527954e-05 2023-01-23 22:16:34.272610: step: 352/1541, loss: 0.0002598908613435924 2023-01-23 22:16:35.181840: step: 354/1541, loss: 0.0003134733415208757 2023-01-23 22:16:36.151571: step: 356/1541, loss: 0.12009566277265549 2023-01-23 22:16:37.031901: step: 358/1541, loss: 0.00014351359277497977 2023-01-23 22:16:37.923080: step: 360/1541, loss: 8.675497520016506e-05 2023-01-23 22:16:38.790788: step: 362/1541, loss: 0.22681789100170135 2023-01-23 22:16:39.674589: step: 364/1541, loss: 5.1946906751254573e-05 2023-01-23 22:16:40.593616: step: 366/1541, loss: 4.413722490426153e-05 2023-01-23 22:16:41.458698: step: 368/1541, loss: 0.00814900640398264 2023-01-23 22:16:42.324311: step: 370/1541, loss: 0.4385072588920593 2023-01-23 22:16:43.183553: step: 372/1541, loss: 0.0754951685667038 2023-01-23 22:16:44.065680: step: 374/1541, loss: 0.00024874447262845933 2023-01-23 22:16:44.947820: step: 376/1541, loss: 2.4735883926041424e-05 2023-01-23 22:16:45.861853: step: 378/1541, loss: 0.0009653851739130914 2023-01-23 22:16:46.754677: step: 380/1541, loss: 0.2968807816505432 2023-01-23 22:16:47.614170: step: 382/1541, loss: 3.634400127339177e-05 2023-01-23 22:16:48.479688: step: 384/1541, loss: 3.104889765381813e-05 2023-01-23 22:16:49.350868: step: 386/1541, loss: 0.0019612866453826427 2023-01-23 22:16:50.264798: step: 388/1541, loss: 0.0004997906507924199 2023-01-23 22:16:51.288566: step: 390/1541, loss: 0.1978285163640976 2023-01-23 22:16:52.164547: step: 392/1541, loss: 0.00024477599072270095 2023-01-23 22:16:53.042538: step: 394/1541, loss: 0.012311588041484356 2023-01-23 22:16:53.968309: step: 396/1541, loss: 0.0010196760995313525 2023-01-23 22:16:54.850637: step: 398/1541, loss: 0.1705726832151413 2023-01-23 22:16:55.694582: step: 400/1541, loss: 0.00019544063252396882 2023-01-23 22:16:56.581066: step: 402/1541, loss: 0.0011592654045671225 2023-01-23 22:16:57.472037: step: 404/1541, loss: 0.0006515805143862963 2023-01-23 22:16:58.353787: step: 406/1541, loss: 0.07760295271873474 2023-01-23 22:16:59.269491: step: 408/1541, loss: 8.030991011764854e-05 2023-01-23 22:17:00.219213: step: 410/1541, loss: 0.0008331958670169115 2023-01-23 22:17:01.117791: step: 412/1541, loss: 0.002133532427251339 2023-01-23 22:17:02.036075: step: 414/1541, loss: 0.00018378109962213784 2023-01-23 22:17:02.982144: step: 416/1541, loss: 0.0015752000035718083 2023-01-23 22:17:03.902334: step: 418/1541, loss: 0.0009177647298201919 2023-01-23 22:17:04.816430: step: 420/1541, loss: 0.005798266269266605 2023-01-23 22:17:05.703400: step: 422/1541, loss: 0.00020413976744748652 2023-01-23 22:17:06.589047: step: 424/1541, loss: 0.1502528339624405 2023-01-23 22:17:07.474428: step: 426/1541, loss: 0.0012474283576011658 2023-01-23 22:17:08.376407: step: 428/1541, loss: 0.0025407050270587206 2023-01-23 22:17:09.305369: step: 430/1541, loss: 3.190858842572197e-05 2023-01-23 22:17:10.229881: step: 432/1541, loss: 0.09211205691099167 2023-01-23 22:17:11.099756: step: 434/1541, loss: 0.25007960200309753 2023-01-23 22:17:11.983712: step: 436/1541, loss: 0.02376442402601242 2023-01-23 22:17:12.959829: step: 438/1541, loss: 0.0012015699176117778 2023-01-23 22:17:13.847655: step: 440/1541, loss: 0.00030782067915424705 2023-01-23 22:17:14.742624: step: 442/1541, loss: 0.0005204221233725548 2023-01-23 22:17:15.624207: step: 444/1541, loss: 1.9093418814009055e-05 2023-01-23 22:17:16.553444: step: 446/1541, loss: 0.0001191243136418052 2023-01-23 22:17:17.436923: step: 448/1541, loss: 0.0009479852160438895 2023-01-23 22:17:18.382947: step: 450/1541, loss: 0.00034780462738126516 2023-01-23 22:17:19.283223: step: 452/1541, loss: 0.00034789100755006075 2023-01-23 22:17:20.156566: step: 454/1541, loss: 0.0016230238834396005 2023-01-23 22:17:21.168944: step: 456/1541, loss: 0.004781294614076614 2023-01-23 22:17:22.048855: step: 458/1541, loss: 2.2130607248982415e-05 2023-01-23 22:17:22.987489: step: 460/1541, loss: 8.102849824354053e-05 2023-01-23 22:17:23.939091: step: 462/1541, loss: 0.008479660376906395 2023-01-23 22:17:24.861776: step: 464/1541, loss: 0.0003694143088068813 2023-01-23 22:17:25.731734: step: 466/1541, loss: 0.00024671037681400776 2023-01-23 22:17:26.595109: step: 468/1541, loss: 7.672006904613227e-05 2023-01-23 22:17:27.512690: step: 470/1541, loss: 0.05269874259829521 2023-01-23 22:17:28.395135: step: 472/1541, loss: 0.01074562780559063 2023-01-23 22:17:29.261756: step: 474/1541, loss: 0.07294701039791107 2023-01-23 22:17:30.133330: step: 476/1541, loss: 1.655931191635318e-05 2023-01-23 22:17:30.990132: step: 478/1541, loss: 4.10582433687523e-05 2023-01-23 22:17:31.869291: step: 480/1541, loss: 1.5103259102033917e-05 2023-01-23 22:17:32.738185: step: 482/1541, loss: 9.777716331882402e-05 2023-01-23 22:17:33.665537: step: 484/1541, loss: 0.00022694015933666378 2023-01-23 22:17:34.556794: step: 486/1541, loss: 0.00010980366641888395 2023-01-23 22:17:35.455919: step: 488/1541, loss: 0.41880157589912415 2023-01-23 22:17:36.412629: step: 490/1541, loss: 0.00478026457130909 2023-01-23 22:17:37.290576: step: 492/1541, loss: 1.8687258958816528 2023-01-23 22:17:38.189838: step: 494/1541, loss: 0.0006052628741599619 2023-01-23 22:17:39.055981: step: 496/1541, loss: 8.305557275889441e-05 2023-01-23 22:17:39.966260: step: 498/1541, loss: 3.117973028565757e-05 2023-01-23 22:17:40.854026: step: 500/1541, loss: 0.0006684152758680284 2023-01-23 22:17:41.756034: step: 502/1541, loss: -5.934238652116619e-05 2023-01-23 22:17:42.672118: step: 504/1541, loss: 0.00024938813294284046 2023-01-23 22:17:43.549992: step: 506/1541, loss: 0.0002242291229777038 2023-01-23 22:17:44.452651: step: 508/1541, loss: 0.0067351749166846275 2023-01-23 22:17:45.354975: step: 510/1541, loss: 0.000450534833362326 2023-01-23 22:17:46.206318: step: 512/1541, loss: 5.981985304970294e-05 2023-01-23 22:17:47.119977: step: 514/1541, loss: 0.01063147746026516 2023-01-23 22:17:47.993606: step: 516/1541, loss: 2.331983341719024e-05 2023-01-23 22:17:48.852173: step: 518/1541, loss: 0.03459830582141876 2023-01-23 22:17:49.754222: step: 520/1541, loss: 0.07388073205947876 2023-01-23 22:17:50.648056: step: 522/1541, loss: 5.687290467903949e-05 2023-01-23 22:17:51.515840: step: 524/1541, loss: -1.2488294487411622e-05 2023-01-23 22:17:52.393025: step: 526/1541, loss: 0.0020656990818679333 2023-01-23 22:17:53.255351: step: 528/1541, loss: 0.009402934461832047 2023-01-23 22:17:54.242007: step: 530/1541, loss: 0.00035131184267811477 2023-01-23 22:17:55.137400: step: 532/1541, loss: 3.0706331017427146e-05 2023-01-23 22:17:56.011938: step: 534/1541, loss: 1.299744144489523e-05 2023-01-23 22:17:56.878326: step: 536/1541, loss: 0.0017233967082574964 2023-01-23 22:17:57.760521: step: 538/1541, loss: 0.0005612742970697582 2023-01-23 22:17:58.632753: step: 540/1541, loss: 0.0013120145304128528 2023-01-23 22:17:59.495545: step: 542/1541, loss: 0.0002567481715232134 2023-01-23 22:18:00.445493: step: 544/1541, loss: 7.835846190573648e-05 2023-01-23 22:18:01.337934: step: 546/1541, loss: 0.03839794546365738 2023-01-23 22:18:02.222350: step: 548/1541, loss: 0.00036713507142849267 2023-01-23 22:18:03.157748: step: 550/1541, loss: 0.0008293962455354631 2023-01-23 22:18:04.043908: step: 552/1541, loss: 0.006691355723887682 2023-01-23 22:18:04.960118: step: 554/1541, loss: 0.0004304725443944335 2023-01-23 22:18:05.923311: step: 556/1541, loss: 0.006455791182816029 2023-01-23 22:18:06.787470: step: 558/1541, loss: 0.004078097175806761 2023-01-23 22:18:07.681677: step: 560/1541, loss: 0.0003421824658289552 2023-01-23 22:18:08.647550: step: 562/1541, loss: 8.168046770151705e-05 2023-01-23 22:18:09.519866: step: 564/1541, loss: 0.053099457174539566 2023-01-23 22:18:10.382141: step: 566/1541, loss: 2.5535660824971274e-05 2023-01-23 22:18:11.276935: step: 568/1541, loss: 0.004415187519043684 2023-01-23 22:18:12.248575: step: 570/1541, loss: 0.033038776367902756 2023-01-23 22:18:13.162643: step: 572/1541, loss: 0.016510436311364174 2023-01-23 22:18:14.011516: step: 574/1541, loss: 0.013387331739068031 2023-01-23 22:18:14.940985: step: 576/1541, loss: 0.3063281178474426 2023-01-23 22:18:15.867309: step: 578/1541, loss: 0.0031703528948128223 2023-01-23 22:18:16.793849: step: 580/1541, loss: 0.14028459787368774 2023-01-23 22:18:17.723991: step: 582/1541, loss: 0.00013133510947227478 2023-01-23 22:18:18.576948: step: 584/1541, loss: 0.003307253122329712 2023-01-23 22:18:19.415292: step: 586/1541, loss: 0.008144237101078033 2023-01-23 22:18:20.289296: step: 588/1541, loss: 1.6946863979683258e-05 2023-01-23 22:18:21.181427: step: 590/1541, loss: 0.28563445806503296 2023-01-23 22:18:22.051088: step: 592/1541, loss: 0.00018390451441518962 2023-01-23 22:18:22.945734: step: 594/1541, loss: 0.002198125934228301 2023-01-23 22:18:23.891803: step: 596/1541, loss: 1.1301058293611277e-06 2023-01-23 22:18:24.771101: step: 598/1541, loss: 5.276230876916088e-05 2023-01-23 22:18:25.638276: step: 600/1541, loss: 0.0003384526935406029 2023-01-23 22:18:26.558841: step: 602/1541, loss: 0.3029119074344635 2023-01-23 22:18:27.457468: step: 604/1541, loss: 1.6695948943379335e-05 2023-01-23 22:18:28.339822: step: 606/1541, loss: 0.055536024272441864 2023-01-23 22:18:29.221198: step: 608/1541, loss: 8.392876770813018e-05 2023-01-23 22:18:30.130783: step: 610/1541, loss: 0.02017275057733059 2023-01-23 22:18:30.992259: step: 612/1541, loss: 0.006102729123085737 2023-01-23 22:18:31.889864: step: 614/1541, loss: 0.00027908748597837985 2023-01-23 22:18:32.763919: step: 616/1541, loss: 0.04635491967201233 2023-01-23 22:18:33.637954: step: 618/1541, loss: 0.0001301486190641299 2023-01-23 22:18:34.515627: step: 620/1541, loss: 0.0006885413895361125 2023-01-23 22:18:35.454486: step: 622/1541, loss: 0.007600751239806414 2023-01-23 22:18:36.363800: step: 624/1541, loss: 0.003359511960297823 2023-01-23 22:18:37.332329: step: 626/1541, loss: 5.153289748704992e-05 2023-01-23 22:18:38.234165: step: 628/1541, loss: 0.00010434693831484765 2023-01-23 22:18:39.158772: step: 630/1541, loss: 0.00685358140617609 2023-01-23 22:18:40.128785: step: 632/1541, loss: 0.05722792446613312 2023-01-23 22:18:41.030274: step: 634/1541, loss: 0.00018660727073438466 2023-01-23 22:18:41.928396: step: 636/1541, loss: 0.05682096630334854 2023-01-23 22:18:42.823069: step: 638/1541, loss: 8.714570867596194e-05 2023-01-23 22:18:43.850210: step: 640/1541, loss: 0.03693367913365364 2023-01-23 22:18:44.767484: step: 642/1541, loss: 4.133299444220029e-05 2023-01-23 22:18:45.632687: step: 644/1541, loss: 0.011415764689445496 2023-01-23 22:18:46.539014: step: 646/1541, loss: 0.00021595203725155443 2023-01-23 22:18:47.418927: step: 648/1541, loss: 8.431597234448418e-05 2023-01-23 22:18:48.349285: step: 650/1541, loss: 4.305918264435604e-05 2023-01-23 22:18:49.246517: step: 652/1541, loss: 0.12997819483280182 2023-01-23 22:18:50.095902: step: 654/1541, loss: 0.0011246216017752886 2023-01-23 22:18:50.963312: step: 656/1541, loss: 0.0011388695565983653 2023-01-23 22:18:51.845897: step: 658/1541, loss: 0.0002334569435333833 2023-01-23 22:18:52.690332: step: 660/1541, loss: 0.01348966732621193 2023-01-23 22:18:53.545325: step: 662/1541, loss: 0.00024779673549346626 2023-01-23 22:18:54.428277: step: 664/1541, loss: 0.0007933693705126643 2023-01-23 22:18:55.306473: step: 666/1541, loss: 0.22623921930789948 2023-01-23 22:18:56.197899: step: 668/1541, loss: 0.2663314640522003 2023-01-23 22:18:57.098949: step: 670/1541, loss: 1.0112516974913888e-05 2023-01-23 22:18:57.966074: step: 672/1541, loss: 4.673898365581408e-05 2023-01-23 22:18:58.855220: step: 674/1541, loss: 0.004201028496026993 2023-01-23 22:18:59.745610: step: 676/1541, loss: 0.0005579007556661963 2023-01-23 22:19:00.629912: step: 678/1541, loss: 0.011418940499424934 2023-01-23 22:19:01.568764: step: 680/1541, loss: 0.0007760326843708754 2023-01-23 22:19:02.461677: step: 682/1541, loss: 0.19561754167079926 2023-01-23 22:19:03.400314: step: 684/1541, loss: 8.483526471536607e-05 2023-01-23 22:19:04.297031: step: 686/1541, loss: 0.006851082667708397 2023-01-23 22:19:05.246590: step: 688/1541, loss: 0.00011364892270648852 2023-01-23 22:19:06.182492: step: 690/1541, loss: 1.030564362736186e-05 2023-01-23 22:19:07.067470: step: 692/1541, loss: 0.0003202153602614999 2023-01-23 22:19:07.937179: step: 694/1541, loss: 0.00019026472000405192 2023-01-23 22:19:08.846128: step: 696/1541, loss: 0.00010914563608821481 2023-01-23 22:19:09.726941: step: 698/1541, loss: -3.4881239116657525e-05 2023-01-23 22:19:10.674233: step: 700/1541, loss: 0.00011414591426728293 2023-01-23 22:19:11.532355: step: 702/1541, loss: 0.06723960489034653 2023-01-23 22:19:12.506909: step: 704/1541, loss: 0.0009939582087099552 2023-01-23 22:19:13.448158: step: 706/1541, loss: 4.559702938422561e-05 2023-01-23 22:19:14.316573: step: 708/1541, loss: 3.0681490898132324e-05 2023-01-23 22:19:15.247289: step: 710/1541, loss: 0.0015009664930403233 2023-01-23 22:19:16.193101: step: 712/1541, loss: 0.0013061913195997477 2023-01-23 22:19:17.092335: step: 714/1541, loss: 0.0018136849394068122 2023-01-23 22:19:17.971706: step: 716/1541, loss: 0.0012909675715491176 2023-01-23 22:19:18.843618: step: 718/1541, loss: 0.00013200224202591926 2023-01-23 22:19:19.706182: step: 720/1541, loss: 0.012430118396878242 2023-01-23 22:19:20.613471: step: 722/1541, loss: -2.2013486159266904e-05 2023-01-23 22:19:21.487997: step: 724/1541, loss: 4.014372734673088e-06 2023-01-23 22:19:22.410102: step: 726/1541, loss: 0.00010818683222169057 2023-01-23 22:19:23.311360: step: 728/1541, loss: 0.8116453886032104 2023-01-23 22:19:24.170474: step: 730/1541, loss: 0.016643593087792397 2023-01-23 22:19:25.095242: step: 732/1541, loss: 0.1548808068037033 2023-01-23 22:19:25.996197: step: 734/1541, loss: 9.696260531200096e-05 2023-01-23 22:19:26.863534: step: 736/1541, loss: 0.3117617070674896 2023-01-23 22:19:27.717371: step: 738/1541, loss: 0.0018226687097921968 2023-01-23 22:19:28.673472: step: 740/1541, loss: 0.09838071465492249 2023-01-23 22:19:29.542270: step: 742/1541, loss: 0.56045001745224 2023-01-23 22:19:30.429638: step: 744/1541, loss: 0.16139842569828033 2023-01-23 22:19:31.295665: step: 746/1541, loss: 0.0006027171039022505 2023-01-23 22:19:32.171945: step: 748/1541, loss: 3.952340193791315e-05 2023-01-23 22:19:33.035712: step: 750/1541, loss: 2.3484097255277447e-05 2023-01-23 22:19:33.933087: step: 752/1541, loss: 2.0899102310067974e-05 2023-01-23 22:19:34.803236: step: 754/1541, loss: 0.0033495991956442595 2023-01-23 22:19:35.699146: step: 756/1541, loss: 0.01886448822915554 2023-01-23 22:19:36.610639: step: 758/1541, loss: 0.0002567334449850023 2023-01-23 22:19:37.524450: step: 760/1541, loss: 0.0014776034513488412 2023-01-23 22:19:38.422450: step: 762/1541, loss: 0.0009444922325201333 2023-01-23 22:19:39.270064: step: 764/1541, loss: 0.0009113856358453631 2023-01-23 22:19:40.120287: step: 766/1541, loss: 4.333813922130503e-05 2023-01-23 22:19:40.979059: step: 768/1541, loss: 0.0006362903513945639 2023-01-23 22:19:41.872470: step: 770/1541, loss: 0.18420396745204926 2023-01-23 22:19:42.715108: step: 772/1541, loss: 0.000886674621142447 2023-01-23 22:19:43.621327: step: 774/1541, loss: 0.08312159031629562 2023-01-23 22:19:44.516916: step: 776/1541, loss: 5.5071304814191535e-05 2023-01-23 22:19:45.444065: step: 778/1541, loss: 0.29041773080825806 2023-01-23 22:19:46.394050: step: 780/1541, loss: 0.6376892328262329 2023-01-23 22:19:47.393593: step: 782/1541, loss: 0.00020060014503542334 2023-01-23 22:19:48.332698: step: 784/1541, loss: 0.6163436770439148 2023-01-23 22:19:49.235660: step: 786/1541, loss: 0.06360181421041489 2023-01-23 22:19:50.091540: step: 788/1541, loss: 0.001110604964196682 2023-01-23 22:19:51.000121: step: 790/1541, loss: 0.0001477005280321464 2023-01-23 22:19:51.861689: step: 792/1541, loss: 0.0037509091198444366 2023-01-23 22:19:52.747155: step: 794/1541, loss: 0.0031985188834369183 2023-01-23 22:19:53.618907: step: 796/1541, loss: 0.00018747565627563745 2023-01-23 22:19:54.537498: step: 798/1541, loss: 0.13693839311599731 2023-01-23 22:19:55.394405: step: 800/1541, loss: 0.00044343585614115 2023-01-23 22:19:56.263608: step: 802/1541, loss: 0.033768180757761 2023-01-23 22:19:57.216711: step: 804/1541, loss: 0.0162053145468235 2023-01-23 22:19:58.101933: step: 806/1541, loss: 0.004842561203986406 2023-01-23 22:19:58.975764: step: 808/1541, loss: 0.00016148433496709913 2023-01-23 22:19:59.805042: step: 810/1541, loss: 0.00033498535049147904 2023-01-23 22:20:00.651154: step: 812/1541, loss: 0.33569619059562683 2023-01-23 22:20:01.556010: step: 814/1541, loss: 0.008114511147141457 2023-01-23 22:20:02.480934: step: 816/1541, loss: 0.010900693014264107 2023-01-23 22:20:03.403246: step: 818/1541, loss: 0.11149270087480545 2023-01-23 22:20:04.272468: step: 820/1541, loss: 0.00620864937081933 2023-01-23 22:20:05.244786: step: 822/1541, loss: 0.02255169115960598 2023-01-23 22:20:06.205068: step: 824/1541, loss: 0.00037472290568985045 2023-01-23 22:20:07.057573: step: 826/1541, loss: 0.04959241300821304 2023-01-23 22:20:07.922181: step: 828/1541, loss: 0.001073227496817708 2023-01-23 22:20:08.815054: step: 830/1541, loss: 0.01281682774424553 2023-01-23 22:20:09.679847: step: 832/1541, loss: 0.02146848663687706 2023-01-23 22:20:10.528256: step: 834/1541, loss: 0.0007212322088889778 2023-01-23 22:20:11.408838: step: 836/1541, loss: 0.22043193876743317 2023-01-23 22:20:12.299346: step: 838/1541, loss: 0.002545783529058099 2023-01-23 22:20:13.162615: step: 840/1541, loss: 0.00012725083797704428 2023-01-23 22:20:14.069672: step: 842/1541, loss: 0.26428353786468506 2023-01-23 22:20:14.998708: step: 844/1541, loss: 0.014048783108592033 2023-01-23 22:20:15.906124: step: 846/1541, loss: 0.0014548785984516144 2023-01-23 22:20:16.838710: step: 848/1541, loss: 0.027895692735910416 2023-01-23 22:20:17.727203: step: 850/1541, loss: 0.01348648127168417 2023-01-23 22:20:18.634906: step: 852/1541, loss: 0.009175792336463928 2023-01-23 22:20:19.567197: step: 854/1541, loss: 0.038325924426317215 2023-01-23 22:20:20.471770: step: 856/1541, loss: 0.006472236011177301 2023-01-23 22:20:21.343035: step: 858/1541, loss: 0.000405202794354409 2023-01-23 22:20:22.231490: step: 860/1541, loss: 9.092901018448174e-05 2023-01-23 22:20:23.124004: step: 862/1541, loss: 0.000571677868720144 2023-01-23 22:20:24.051611: step: 864/1541, loss: 8.651462121633813e-05 2023-01-23 22:20:25.001586: step: 866/1541, loss: 0.0026188804768025875 2023-01-23 22:20:25.880305: step: 868/1541, loss: -4.930815521220211e-06 2023-01-23 22:20:26.746973: step: 870/1541, loss: 9.854107338469476e-05 2023-01-23 22:20:27.594609: step: 872/1541, loss: 0.00023453794710803777 2023-01-23 22:20:28.500546: step: 874/1541, loss: 0.00019773017265833914 2023-01-23 22:20:29.393156: step: 876/1541, loss: 0.0038007237017154694 2023-01-23 22:20:30.276882: step: 878/1541, loss: 0.0005953052313998342 2023-01-23 22:20:31.194734: step: 880/1541, loss: 1.8073758837999776e-05 2023-01-23 22:20:32.158777: step: 882/1541, loss: 0.06074139103293419 2023-01-23 22:20:33.037656: step: 884/1541, loss: 0.09322599321603775 2023-01-23 22:20:33.950012: step: 886/1541, loss: 0.11933501064777374 2023-01-23 22:20:34.851152: step: 888/1541, loss: 0.17472092807292938 2023-01-23 22:20:35.839868: step: 890/1541, loss: 0.040809597820043564 2023-01-23 22:20:36.723897: step: 892/1541, loss: 6.486218626378104e-05 2023-01-23 22:20:37.733710: step: 894/1541, loss: 0.0005263849161565304 2023-01-23 22:20:38.625702: step: 896/1541, loss: 0.01236933283507824 2023-01-23 22:20:39.501923: step: 898/1541, loss: 0.36676907539367676 2023-01-23 22:20:40.412636: step: 900/1541, loss: 0.010088615119457245 2023-01-23 22:20:41.336384: step: 902/1541, loss: 0.0001911535655381158 2023-01-23 22:20:42.236935: step: 904/1541, loss: 0.18243259191513062 2023-01-23 22:20:43.123428: step: 906/1541, loss: 0.06914026290178299 2023-01-23 22:20:43.977414: step: 908/1541, loss: 2.5371935407747515e-05 2023-01-23 22:20:44.869996: step: 910/1541, loss: 0.0008957866812124848 2023-01-23 22:20:45.766895: step: 912/1541, loss: 0.00010926348477369174 2023-01-23 22:20:46.698769: step: 914/1541, loss: 8.341902866959572e-05 2023-01-23 22:20:47.570098: step: 916/1541, loss: 0.005079899448901415 2023-01-23 22:20:48.455519: step: 918/1541, loss: 0.00014478372759185731 2023-01-23 22:20:49.374360: step: 920/1541, loss: 0.017727592960000038 2023-01-23 22:20:50.243078: step: 922/1541, loss: 0.015255071222782135 2023-01-23 22:20:51.130317: step: 924/1541, loss: 8.213165710913017e-05 2023-01-23 22:20:52.022158: step: 926/1541, loss: 0.000878066464792937 2023-01-23 22:20:52.911890: step: 928/1541, loss: 0.0017385841347277164 2023-01-23 22:20:53.781050: step: 930/1541, loss: 0.002143292687833309 2023-01-23 22:20:54.727663: step: 932/1541, loss: 6.79892473272048e-05 2023-01-23 22:20:55.634648: step: 934/1541, loss: 0.17931988835334778 2023-01-23 22:20:56.520791: step: 936/1541, loss: 4.0033388359006494e-05 2023-01-23 22:20:57.413690: step: 938/1541, loss: 3.7242585904095904e-08 2023-01-23 22:20:58.257444: step: 940/1541, loss: 0.0006838943809270859 2023-01-23 22:20:59.142157: step: 942/1541, loss: 1.6892670828383416e-05 2023-01-23 22:21:00.047177: step: 944/1541, loss: 0.12180227041244507 2023-01-23 22:21:00.935635: step: 946/1541, loss: 0.32719624042510986 2023-01-23 22:21:01.872113: step: 948/1541, loss: 0.00020072954066563398 2023-01-23 22:21:02.851104: step: 950/1541, loss: 0.000506295298691839 2023-01-23 22:21:03.789194: step: 952/1541, loss: -3.7878158764215186e-05 2023-01-23 22:21:04.715297: step: 954/1541, loss: 0.00014192654634825885 2023-01-23 22:21:05.600094: step: 956/1541, loss: 0.10017079859972 2023-01-23 22:21:06.510928: step: 958/1541, loss: 0.0023644347675144672 2023-01-23 22:21:07.361862: step: 960/1541, loss: 3.3039352274499834e-05 2023-01-23 22:21:08.269885: step: 962/1541, loss: 0.001026816782541573 2023-01-23 22:21:09.153720: step: 964/1541, loss: 0.19254685938358307 2023-01-23 22:21:10.022162: step: 966/1541, loss: 0.018703347072005272 2023-01-23 22:21:10.899754: step: 968/1541, loss: 0.00031623058021068573 2023-01-23 22:21:11.783050: step: 970/1541, loss: -9.136647349805571e-06 2023-01-23 22:21:12.705172: step: 972/1541, loss: 0.0014493847265839577 2023-01-23 22:21:13.575055: step: 974/1541, loss: 0.0006306596333160996 2023-01-23 22:21:14.533577: step: 976/1541, loss: 8.968546171672642e-05 2023-01-23 22:21:15.459620: step: 978/1541, loss: 0.00025345690664835274 2023-01-23 22:21:16.316933: step: 980/1541, loss: 4.7702673327876255e-05 2023-01-23 22:21:17.201104: step: 982/1541, loss: 0.09900213032960892 2023-01-23 22:21:18.112891: step: 984/1541, loss: -6.8918452598154545e-06 2023-01-23 22:21:19.070928: step: 986/1541, loss: 0.002007010392844677 2023-01-23 22:21:19.952028: step: 988/1541, loss: -9.182098438031971e-06 2023-01-23 22:21:20.819555: step: 990/1541, loss: 6.0765119997086003e-05 2023-01-23 22:21:21.729281: step: 992/1541, loss: 4.7039684432093054e-05 2023-01-23 22:21:22.657049: step: 994/1541, loss: 7.512830052291974e-05 2023-01-23 22:21:23.535001: step: 996/1541, loss: 0.06445179879665375 2023-01-23 22:21:24.402017: step: 998/1541, loss: 0.0001951707963598892 2023-01-23 22:21:25.254795: step: 1000/1541, loss: -7.590132736368105e-05 2023-01-23 22:21:26.164703: step: 1002/1541, loss: 9.5780698757153e-05 2023-01-23 22:21:27.106830: step: 1004/1541, loss: 0.3995164930820465 2023-01-23 22:21:28.004816: step: 1006/1541, loss: 0.022738998755812645 2023-01-23 22:21:28.944658: step: 1008/1541, loss: 0.00022730836644768715 2023-01-23 22:21:29.805627: step: 1010/1541, loss: -2.402454811090138e-06 2023-01-23 22:21:30.749079: step: 1012/1541, loss: 0.049203209578990936 2023-01-23 22:21:31.647285: step: 1014/1541, loss: 0.05489727482199669 2023-01-23 22:21:32.709903: step: 1016/1541, loss: -6.934009775250161e-07 2023-01-23 22:21:33.620249: step: 1018/1541, loss: 0.0004315738333389163 2023-01-23 22:21:34.508563: step: 1020/1541, loss: 0.0039092483930289745 2023-01-23 22:21:35.376550: step: 1022/1541, loss: 6.608664989471436e-05 2023-01-23 22:21:36.265900: step: 1024/1541, loss: 0.00010453373397467658 2023-01-23 22:21:37.220902: step: 1026/1541, loss: 0.0002868771553039551 2023-01-23 22:21:38.159664: step: 1028/1541, loss: 0.00017542182467877865 2023-01-23 22:21:39.053618: step: 1030/1541, loss: -2.5549384190526325e-06 2023-01-23 22:21:39.906631: step: 1032/1541, loss: 0.0001334600237896666 2023-01-23 22:21:40.755567: step: 1034/1541, loss: 4.624961729859933e-05 2023-01-23 22:21:41.726188: step: 1036/1541, loss: 1.1486650691949762e-05 2023-01-23 22:21:42.622696: step: 1038/1541, loss: 0.16546715795993805 2023-01-23 22:21:43.521147: step: 1040/1541, loss: 0.00013811032113153487 2023-01-23 22:21:44.439354: step: 1042/1541, loss: 0.09803931415081024 2023-01-23 22:21:45.388431: step: 1044/1541, loss: 0.1688183844089508 2023-01-23 22:21:46.277117: step: 1046/1541, loss: 3.589928019209765e-05 2023-01-23 22:21:47.135511: step: 1048/1541, loss: 8.04915907792747e-05 2023-01-23 22:21:48.043949: step: 1050/1541, loss: 0.000305209745420143 2023-01-23 22:21:48.927209: step: 1052/1541, loss: 2.4244025098596467e-06 2023-01-23 22:21:49.793707: step: 1054/1541, loss: 0.056207817047834396 2023-01-23 22:21:50.681690: step: 1056/1541, loss: 0.05652298778295517 2023-01-23 22:21:51.572499: step: 1058/1541, loss: 0.0013931655557826161 2023-01-23 22:21:52.459411: step: 1060/1541, loss: 0.10402600467205048 2023-01-23 22:21:53.315890: step: 1062/1541, loss: 0.0008706366643309593 2023-01-23 22:21:54.240796: step: 1064/1541, loss: 3.8531419704668224e-05 2023-01-23 22:21:55.141845: step: 1066/1541, loss: -1.293743662245106e-05 2023-01-23 22:21:56.013552: step: 1068/1541, loss: -9.70735300143133e-07 2023-01-23 22:21:56.920845: step: 1070/1541, loss: 0.16718217730522156 2023-01-23 22:21:57.807196: step: 1072/1541, loss: 8.746619278099388e-05 2023-01-23 22:21:58.681800: step: 1074/1541, loss: 0.00010641260450938717 2023-01-23 22:21:59.563827: step: 1076/1541, loss: 6.267422577366233e-05 2023-01-23 22:22:00.482780: step: 1078/1541, loss: 0.1707247793674469 2023-01-23 22:22:01.377736: step: 1080/1541, loss: 0.018964271992444992 2023-01-23 22:22:02.244096: step: 1082/1541, loss: -2.066433626168873e-05 2023-01-23 22:22:03.096636: step: 1084/1541, loss: 1.419056934537366e-05 2023-01-23 22:22:04.009057: step: 1086/1541, loss: 0.09566459059715271 2023-01-23 22:22:04.888776: step: 1088/1541, loss: 0.0002828152209986001 2023-01-23 22:22:05.808682: step: 1090/1541, loss: 0.04762093722820282 2023-01-23 22:22:06.782954: step: 1092/1541, loss: 0.0002991688088513911 2023-01-23 22:22:07.662044: step: 1094/1541, loss: 0.05813460797071457 2023-01-23 22:22:08.641184: step: 1096/1541, loss: 0.0001933724561240524 2023-01-23 22:22:09.548726: step: 1098/1541, loss: 0.00798703171312809 2023-01-23 22:22:10.413611: step: 1100/1541, loss: 7.798890146659687e-05 2023-01-23 22:22:11.303739: step: 1102/1541, loss: 8.927236922318116e-05 2023-01-23 22:22:12.179957: step: 1104/1541, loss: 0.08149067312479019 2023-01-23 22:22:13.072373: step: 1106/1541, loss: 0.00019573819008655846 2023-01-23 22:22:13.962239: step: 1108/1541, loss: 6.865360774099827e-05 2023-01-23 22:22:14.850860: step: 1110/1541, loss: 0.11839938908815384 2023-01-23 22:22:15.702014: step: 1112/1541, loss: 0.0006539012538269162 2023-01-23 22:22:16.592229: step: 1114/1541, loss: 0.00010685474262572825 2023-01-23 22:22:17.515829: step: 1116/1541, loss: 0.003482577158138156 2023-01-23 22:22:18.388253: step: 1118/1541, loss: -7.015470373517019e-07 2023-01-23 22:22:19.383578: step: 1120/1541, loss: 0.00012382265413179994 2023-01-23 22:22:20.285982: step: 1122/1541, loss: 0.01731858402490616 2023-01-23 22:22:21.178610: step: 1124/1541, loss: 0.00012686065747402608 2023-01-23 22:22:22.161705: step: 1126/1541, loss: 0.11316022276878357 2023-01-23 22:22:23.118977: step: 1128/1541, loss: 0.18829669058322906 2023-01-23 22:22:24.067046: step: 1130/1541, loss: 9.098138980334625e-05 2023-01-23 22:22:24.962085: step: 1132/1541, loss: 0.8126175403594971 2023-01-23 22:22:25.857693: step: 1134/1541, loss: 0.0002473242348060012 2023-01-23 22:22:26.764317: step: 1136/1541, loss: 0.0025745187886059284 2023-01-23 22:22:27.682548: step: 1138/1541, loss: 0.14863570034503937 2023-01-23 22:22:28.662235: step: 1140/1541, loss: 0.002268917392939329 2023-01-23 22:22:29.550927: step: 1142/1541, loss: 0.0007201128755696118 2023-01-23 22:22:30.428180: step: 1144/1541, loss: 0.0005892319604754448 2023-01-23 22:22:31.317647: step: 1146/1541, loss: 0.22707107663154602 2023-01-23 22:22:32.212322: step: 1148/1541, loss: 1.1736895430658478e-05 2023-01-23 22:22:33.134923: step: 1150/1541, loss: 0.0009523335029371083 2023-01-23 22:22:34.010368: step: 1152/1541, loss: 0.00018796916992869228 2023-01-23 22:22:34.876969: step: 1154/1541, loss: 0.0004998788936063647 2023-01-23 22:22:35.805522: step: 1156/1541, loss: 0.05137161538004875 2023-01-23 22:22:36.788758: step: 1158/1541, loss: 0.6175495386123657 2023-01-23 22:22:37.646334: step: 1160/1541, loss: 0.00024120343732647598 2023-01-23 22:22:38.549977: step: 1162/1541, loss: 0.00013589013542514294 2023-01-23 22:22:39.417747: step: 1164/1541, loss: 0.004012762103229761 2023-01-23 22:22:40.291944: step: 1166/1541, loss: 0.23477908968925476 2023-01-23 22:22:41.168933: step: 1168/1541, loss: 0.024065617471933365 2023-01-23 22:22:42.028874: step: 1170/1541, loss: 0.003007266204804182 2023-01-23 22:22:42.908781: step: 1172/1541, loss: 0.048788685351610184 2023-01-23 22:22:43.783648: step: 1174/1541, loss: 0.0009203987428918481 2023-01-23 22:22:44.694256: step: 1176/1541, loss: 0.0019953004084527493 2023-01-23 22:22:45.594557: step: 1178/1541, loss: 0.011456732638180256 2023-01-23 22:22:46.483202: step: 1180/1541, loss: 0.02284804731607437 2023-01-23 22:22:47.369690: step: 1182/1541, loss: 0.0452093780040741 2023-01-23 22:22:48.271941: step: 1184/1541, loss: 0.00043090677354484797 2023-01-23 22:22:49.144980: step: 1186/1541, loss: 0.7038396000862122 2023-01-23 22:22:49.992280: step: 1188/1541, loss: 0.00038991044857539237 2023-01-23 22:22:50.893814: step: 1190/1541, loss: 0.0002563926682341844 2023-01-23 22:22:51.777063: step: 1192/1541, loss: 1.0552936792373657 2023-01-23 22:22:52.659501: step: 1194/1541, loss: 0.011412358842790127 2023-01-23 22:22:53.541510: step: 1196/1541, loss: 0.021483441814780235 2023-01-23 22:22:54.479525: step: 1198/1541, loss: 0.004980762489140034 2023-01-23 22:22:55.356794: step: 1200/1541, loss: 0.0004066119436174631 2023-01-23 22:22:56.252847: step: 1202/1541, loss: 0.0003667826531454921 2023-01-23 22:22:57.166929: step: 1204/1541, loss: 0.00020766288798768073 2023-01-23 22:22:58.033583: step: 1206/1541, loss: 0.0015531760873273015 2023-01-23 22:22:58.909956: step: 1208/1541, loss: 0.00037453442928381264 2023-01-23 22:22:59.857574: step: 1210/1541, loss: 0.0008015601779334247 2023-01-23 22:23:00.684765: step: 1212/1541, loss: 0.005880589596927166 2023-01-23 22:23:01.586553: step: 1214/1541, loss: 0.05721070617437363 2023-01-23 22:23:02.450106: step: 1216/1541, loss: 0.00010147368448087946 2023-01-23 22:23:03.354114: step: 1218/1541, loss: 0.049431975930929184 2023-01-23 22:23:04.281120: step: 1220/1541, loss: 0.009073100052773952 2023-01-23 22:23:05.140099: step: 1222/1541, loss: 0.023520294576883316 2023-01-23 22:23:06.065330: step: 1224/1541, loss: 0.029426511377096176 2023-01-23 22:23:06.982187: step: 1226/1541, loss: 0.0005399219808168709 2023-01-23 22:23:07.853075: step: 1228/1541, loss: 0.001191378803923726 2023-01-23 22:23:08.729770: step: 1230/1541, loss: 0.06300029158592224 2023-01-23 22:23:09.697233: step: 1232/1541, loss: 0.0005020652315579355 2023-01-23 22:23:10.585352: step: 1234/1541, loss: 0.03890267014503479 2023-01-23 22:23:11.506610: step: 1236/1541, loss: 9.29740272113122e-05 2023-01-23 22:23:12.425899: step: 1238/1541, loss: 0.0010939912172034383 2023-01-23 22:23:13.295679: step: 1240/1541, loss: 0.0008303782087750733 2023-01-23 22:23:14.223025: step: 1242/1541, loss: 0.0003530600224621594 2023-01-23 22:23:15.179691: step: 1244/1541, loss: 0.028975097462534904 2023-01-23 22:23:16.079690: step: 1246/1541, loss: 0.3711473345756531 2023-01-23 22:23:16.977272: step: 1248/1541, loss: 0.000757482135668397 2023-01-23 22:23:17.932761: step: 1250/1541, loss: 2.751933061517775e-05 2023-01-23 22:23:18.834601: step: 1252/1541, loss: 0.00011393222666811198 2023-01-23 22:23:19.788174: step: 1254/1541, loss: 5.7952442148234695e-05 2023-01-23 22:23:20.657211: step: 1256/1541, loss: 0.00032223688322119415 2023-01-23 22:23:21.547431: step: 1258/1541, loss: 0.0017960191471502185 2023-01-23 22:23:22.438595: step: 1260/1541, loss: 8.294134750030935e-05 2023-01-23 22:23:23.318515: step: 1262/1541, loss: 0.24340787529945374 2023-01-23 22:23:24.207542: step: 1264/1541, loss: 0.038714416325092316 2023-01-23 22:23:25.105039: step: 1266/1541, loss: 0.003229820868000388 2023-01-23 22:23:25.978977: step: 1268/1541, loss: 0.0003294009657111019 2023-01-23 22:23:26.840443: step: 1270/1541, loss: 0.00021462493168655783 2023-01-23 22:23:27.708343: step: 1272/1541, loss: 0.00029024790273979306 2023-01-23 22:23:28.614493: step: 1274/1541, loss: 0.005430690478533506 2023-01-23 22:23:29.502835: step: 1276/1541, loss: 0.0002907687739934772 2023-01-23 22:23:30.429225: step: 1278/1541, loss: 0.0013033709255978465 2023-01-23 22:23:31.321245: step: 1280/1541, loss: 0.00031555595342069864 2023-01-23 22:23:32.265319: step: 1282/1541, loss: 7.236251258291304e-05 2023-01-23 22:23:33.126119: step: 1284/1541, loss: 0.00027883797883987427 2023-01-23 22:23:33.999964: step: 1286/1541, loss: 0.40042349696159363 2023-01-23 22:23:34.878736: step: 1288/1541, loss: 0.11657173931598663 2023-01-23 22:23:35.827385: step: 1290/1541, loss: -9.890905857901089e-06 2023-01-23 22:23:36.697733: step: 1292/1541, loss: 0.019954530522227287 2023-01-23 22:23:37.591214: step: 1294/1541, loss: -8.514062756148633e-06 2023-01-23 22:23:38.460201: step: 1296/1541, loss: 0.00021187504171393812 2023-01-23 22:23:39.379573: step: 1298/1541, loss: 0.000775942753534764 2023-01-23 22:23:40.223572: step: 1300/1541, loss: 1.8867152903112583e-05 2023-01-23 22:23:41.158345: step: 1302/1541, loss: 0.0035076707135885954 2023-01-23 22:23:42.024563: step: 1304/1541, loss: 0.0002416847419226542 2023-01-23 22:23:42.936001: step: 1306/1541, loss: 3.760528488783166e-05 2023-01-23 22:23:43.854919: step: 1308/1541, loss: 0.0006156449089758098 2023-01-23 22:23:44.799474: step: 1310/1541, loss: 0.04920291155576706 2023-01-23 22:23:45.704828: step: 1312/1541, loss: 0.013932316564023495 2023-01-23 22:23:46.585073: step: 1314/1541, loss: 0.006126480642706156 2023-01-23 22:23:47.477316: step: 1316/1541, loss: 0.0005240626051090658 2023-01-23 22:23:48.323289: step: 1318/1541, loss: 0.00767867173999548 2023-01-23 22:23:49.211765: step: 1320/1541, loss: 0.00021473650122061372 2023-01-23 22:23:50.109097: step: 1322/1541, loss: 3.5592314816312864e-05 2023-01-23 22:23:50.996447: step: 1324/1541, loss: 0.0021633245050907135 2023-01-23 22:23:51.929528: step: 1326/1541, loss: 0.03512080758810043 2023-01-23 22:23:52.851364: step: 1328/1541, loss: -1.0667577953427099e-05 2023-01-23 22:23:53.737092: step: 1330/1541, loss: 0.0039053780492395163 2023-01-23 22:23:54.607128: step: 1332/1541, loss: 0.14623910188674927 2023-01-23 22:23:55.549689: step: 1334/1541, loss: 0.16655661165714264 2023-01-23 22:23:56.431291: step: 1336/1541, loss: 0.00029600728885270655 2023-01-23 22:23:57.355065: step: 1338/1541, loss: 0.009518825449049473 2023-01-23 22:23:58.204283: step: 1340/1541, loss: 9.357069939142093e-05 2023-01-23 22:23:59.106662: step: 1342/1541, loss: 0.21878062188625336 2023-01-23 22:23:59.963400: step: 1344/1541, loss: 8.471481851302087e-05 2023-01-23 22:24:00.878498: step: 1346/1541, loss: 0.05921998247504234 2023-01-23 22:24:01.805376: step: 1348/1541, loss: -2.9395137971732765e-05 2023-01-23 22:24:02.715736: step: 1350/1541, loss: 0.07680027931928635 2023-01-23 22:24:03.605878: step: 1352/1541, loss: 0.00013884784129913896 2023-01-23 22:24:04.503220: step: 1354/1541, loss: 0.02525121159851551 2023-01-23 22:24:05.427114: step: 1356/1541, loss: 0.1994038075208664 2023-01-23 22:24:06.300674: step: 1358/1541, loss: 0.07372588664293289 2023-01-23 22:24:07.189234: step: 1360/1541, loss: 0.021936625242233276 2023-01-23 22:24:08.094019: step: 1362/1541, loss: 0.28429117798805237 2023-01-23 22:24:09.038573: step: 1364/1541, loss: 7.705018106207717e-06 2023-01-23 22:24:09.939282: step: 1366/1541, loss: 0.0013167060678824782 2023-01-23 22:24:10.830849: step: 1368/1541, loss: 2.7207695893594064e-05 2023-01-23 22:24:11.740151: step: 1370/1541, loss: 0.024099253118038177 2023-01-23 22:24:12.647411: step: 1372/1541, loss: 0.0004033918376080692 2023-01-23 22:24:13.513601: step: 1374/1541, loss: 1.3321032383828424e-05 2023-01-23 22:24:14.421567: step: 1376/1541, loss: 0.0013238525716587901 2023-01-23 22:24:15.262678: step: 1378/1541, loss: 0.0022363720927387476 2023-01-23 22:24:16.224985: step: 1380/1541, loss: 0.0085947560146451 2023-01-23 22:24:17.131351: step: 1382/1541, loss: 0.5290251970291138 2023-01-23 22:24:18.043561: step: 1384/1541, loss: 9.672613668954e-05 2023-01-23 22:24:18.918692: step: 1386/1541, loss: 1.0432428121566772 2023-01-23 22:24:19.785738: step: 1388/1541, loss: 0.0019602919928729534 2023-01-23 22:24:20.700245: step: 1390/1541, loss: 0.0008137297700159252 2023-01-23 22:24:21.558739: step: 1392/1541, loss: 0.0003729361342266202 2023-01-23 22:24:22.455640: step: 1394/1541, loss: 0.0005139596178196371 2023-01-23 22:24:23.345770: step: 1396/1541, loss: 0.00016084402159322053 2023-01-23 22:24:24.229755: step: 1398/1541, loss: 0.0959293320775032 2023-01-23 22:24:25.111591: step: 1400/1541, loss: 0.03273635730147362 2023-01-23 22:24:25.976153: step: 1402/1541, loss: 0.0002554670500103384 2023-01-23 22:24:26.867034: step: 1404/1541, loss: 1.2362301276880316e-05 2023-01-23 22:24:27.857802: step: 1406/1541, loss: 0.0004902324872091413 2023-01-23 22:24:28.794888: step: 1408/1541, loss: 0.025087283924221992 2023-01-23 22:24:29.676317: step: 1410/1541, loss: 1.2087076356692705e-05 2023-01-23 22:24:30.549645: step: 1412/1541, loss: 0.011330698616802692 2023-01-23 22:24:31.491649: step: 1414/1541, loss: 0.00020022485114168376 2023-01-23 22:24:32.378182: step: 1416/1541, loss: 0.039001692086458206 2023-01-23 22:24:33.257749: step: 1418/1541, loss: 0.020518027245998383 2023-01-23 22:24:34.161768: step: 1420/1541, loss: 0.00011663884652080014 2023-01-23 22:24:35.025063: step: 1422/1541, loss: 1.8882154108723626e-05 2023-01-23 22:24:35.869209: step: 1424/1541, loss: 0.00017702282639220357 2023-01-23 22:24:36.730346: step: 1426/1541, loss: 4.701258148998022e-05 2023-01-23 22:24:37.609456: step: 1428/1541, loss: 2.034940553130582e-05 2023-01-23 22:24:38.511765: step: 1430/1541, loss: 0.0036716237664222717 2023-01-23 22:24:39.390365: step: 1432/1541, loss: 0.0008798846392892301 2023-01-23 22:24:40.269420: step: 1434/1541, loss: 4.245492891641334e-05 2023-01-23 22:24:41.204268: step: 1436/1541, loss: 0.5561798214912415 2023-01-23 22:24:42.069074: step: 1438/1541, loss: 0.0004503344534896314 2023-01-23 22:24:42.959180: step: 1440/1541, loss: -5.703866463591112e-06 2023-01-23 22:24:43.822043: step: 1442/1541, loss: 0.00393383763730526 2023-01-23 22:24:44.717015: step: 1444/1541, loss: 5.653191328747198e-05 2023-01-23 22:24:45.663750: step: 1446/1541, loss: 0.3137199878692627 2023-01-23 22:24:46.541864: step: 1448/1541, loss: 3.527483931975439e-05 2023-01-23 22:24:47.418501: step: 1450/1541, loss: 1.775800956238527e-05 2023-01-23 22:24:48.283567: step: 1452/1541, loss: 0.0008653028053231537 2023-01-23 22:24:49.230960: step: 1454/1541, loss: 0.0010622866684570909 2023-01-23 22:24:50.128585: step: 1456/1541, loss: 0.03417595475912094 2023-01-23 22:24:51.005159: step: 1458/1541, loss: 0.017566563561558723 2023-01-23 22:24:51.962752: step: 1460/1541, loss: 2.5764835299924016e-05 2023-01-23 22:24:52.838738: step: 1462/1541, loss: 1.84744167199824e-05 2023-01-23 22:24:53.757622: step: 1464/1541, loss: 0.10494501888751984 2023-01-23 22:24:54.611110: step: 1466/1541, loss: 0.0001336060813628137 2023-01-23 22:24:55.485134: step: 1468/1541, loss: 2.356567347305827e-05 2023-01-23 22:24:56.426769: step: 1470/1541, loss: 0.0037513261195272207 2023-01-23 22:24:57.347380: step: 1472/1541, loss: 0.04521724954247475 2023-01-23 22:24:58.218416: step: 1474/1541, loss: 0.00026355471345596015 2023-01-23 22:24:59.106157: step: 1476/1541, loss: 0.00036368027213029563 2023-01-23 22:25:00.003093: step: 1478/1541, loss: 0.007173973135650158 2023-01-23 22:25:00.872351: step: 1480/1541, loss: 0.0013116637710481882 2023-01-23 22:25:01.734704: step: 1482/1541, loss: 2.5561948859831318e-05 2023-01-23 22:25:02.627135: step: 1484/1541, loss: 0.0383729450404644 2023-01-23 22:25:03.558507: step: 1486/1541, loss: 0.8796068429946899 2023-01-23 22:25:04.457266: step: 1488/1541, loss: 0.0003505136410240084 2023-01-23 22:25:05.326720: step: 1490/1541, loss: 0.09250536561012268 2023-01-23 22:25:06.209783: step: 1492/1541, loss: 0.006570470053702593 2023-01-23 22:25:07.094225: step: 1494/1541, loss: 0.015305353328585625 2023-01-23 22:25:07.985875: step: 1496/1541, loss: -3.2375883165514097e-06 2023-01-23 22:25:08.856529: step: 1498/1541, loss: 0.0018462493317201734 2023-01-23 22:25:09.722428: step: 1500/1541, loss: 4.609293682733551e-05 2023-01-23 22:25:10.583365: step: 1502/1541, loss: 0.00023953721392899752 2023-01-23 22:25:11.532038: step: 1504/1541, loss: 0.003135440405458212 2023-01-23 22:25:12.461503: step: 1506/1541, loss: 6.90758679411374e-05 2023-01-23 22:25:13.345475: step: 1508/1541, loss: 0.0005950826453045011 2023-01-23 22:25:14.231160: step: 1510/1541, loss: 0.00047652170178480446 2023-01-23 22:25:15.166077: step: 1512/1541, loss: 0.0003665864351205528 2023-01-23 22:25:16.055898: step: 1514/1541, loss: -1.5800940673216246e-05 2023-01-23 22:25:16.968268: step: 1516/1541, loss: 1.2550968676805496e-05 2023-01-23 22:25:17.952711: step: 1518/1541, loss: 0.00012510879605542868 2023-01-23 22:25:18.835473: step: 1520/1541, loss: 0.0006481200689449906 2023-01-23 22:25:19.721657: step: 1522/1541, loss: 0.0005073142237961292 2023-01-23 22:25:20.666581: step: 1524/1541, loss: 0.0020026806741952896 2023-01-23 22:25:21.522049: step: 1526/1541, loss: 0.07592944800853729 2023-01-23 22:25:22.456567: step: 1528/1541, loss: 0.03111339919269085 2023-01-23 22:25:23.365468: step: 1530/1541, loss: 0.00036454424844123423 2023-01-23 22:25:24.305385: step: 1532/1541, loss: 0.3007708489894867 2023-01-23 22:25:25.239321: step: 1534/1541, loss: 0.0001091839512810111 2023-01-23 22:25:26.164710: step: 1536/1541, loss: 0.00012511272507254034 2023-01-23 22:25:27.074428: step: 1538/1541, loss: 0.0001653901708777994 2023-01-23 22:25:27.999800: step: 1540/1541, loss: 0.06257597357034683 2023-01-23 22:25:28.900911: step: 1542/1541, loss: 0.4468078315258026 2023-01-23 22:25:29.818603: step: 1544/1541, loss: 2.6845906177186407e-05 2023-01-23 22:25:30.679203: step: 1546/1541, loss: 0.001908288337290287 2023-01-23 22:25:31.576314: step: 1548/1541, loss: -2.0180932551738806e-06 2023-01-23 22:25:32.479770: step: 1550/1541, loss: 0.0003628695267252624 2023-01-23 22:25:33.367470: step: 1552/1541, loss: 1.0482592188054696e-05 2023-01-23 22:25:34.224670: step: 1554/1541, loss: 0.0033324866089969873 2023-01-23 22:25:35.077273: step: 1556/1541, loss: 0.047731950879096985 2023-01-23 22:25:35.980817: step: 1558/1541, loss: 0.14917631447315216 2023-01-23 22:25:36.828128: step: 1560/1541, loss: 1.7955899238586426e-05 2023-01-23 22:25:37.747430: step: 1562/1541, loss: 9.74364229477942e-06 2023-01-23 22:25:38.636343: step: 1564/1541, loss: 1.0393932825536467e-05 2023-01-23 22:25:39.562101: step: 1566/1541, loss: 0.05914732813835144 2023-01-23 22:25:40.442745: step: 1568/1541, loss: 0.039662934839725494 2023-01-23 22:25:41.373176: step: 1570/1541, loss: 0.21271052956581116 2023-01-23 22:25:42.291163: step: 1572/1541, loss: 9.605288141756319e-06 2023-01-23 22:25:43.151337: step: 1574/1541, loss: 9.595025039743632e-05 2023-01-23 22:25:44.033404: step: 1576/1541, loss: 0.0011741103371605277 2023-01-23 22:25:44.950092: step: 1578/1541, loss: 0.0013996415073052049 2023-01-23 22:25:45.835400: step: 1580/1541, loss: 0.5159559845924377 2023-01-23 22:25:46.774246: step: 1582/1541, loss: 0.11803144961595535 2023-01-23 22:25:47.711749: step: 1584/1541, loss: 0.003438659245148301 2023-01-23 22:25:48.634700: step: 1586/1541, loss: 0.0241871178150177 2023-01-23 22:25:49.536657: step: 1588/1541, loss: 6.796121306251734e-05 2023-01-23 22:25:50.422881: step: 1590/1541, loss: 2.7110618248116225e-05 2023-01-23 22:25:51.295814: step: 1592/1541, loss: 1.8771179384202696e-05 2023-01-23 22:25:52.161338: step: 1594/1541, loss: 0.0006363482098095119 2023-01-23 22:25:53.033342: step: 1596/1541, loss: 0.08271567523479462 2023-01-23 22:25:53.931861: step: 1598/1541, loss: -1.8971235476783477e-05 2023-01-23 22:25:54.812792: step: 1600/1541, loss: 0.01263805665075779 2023-01-23 22:25:55.707126: step: 1602/1541, loss: 0.1989239901304245 2023-01-23 22:25:56.591517: step: 1604/1541, loss: 0.001483057625591755 2023-01-23 22:25:57.467146: step: 1606/1541, loss: 0.0007195390062406659 2023-01-23 22:25:58.413789: step: 1608/1541, loss: 0.2177351713180542 2023-01-23 22:25:59.327982: step: 1610/1541, loss: 0.00017412948363926262 2023-01-23 22:26:00.203924: step: 1612/1541, loss: 4.611327312886715e-05 2023-01-23 22:26:01.082579: step: 1614/1541, loss: 0.15465544164180756 2023-01-23 22:26:01.955039: step: 1616/1541, loss: 0.10660356283187866 2023-01-23 22:26:02.910557: step: 1618/1541, loss: 0.36429500579833984 2023-01-23 22:26:03.816881: step: 1620/1541, loss: 0.00033199763856828213 2023-01-23 22:26:04.671115: step: 1622/1541, loss: 3.278380972915329e-05 2023-01-23 22:26:05.544470: step: 1624/1541, loss: 0.010922206565737724 2023-01-23 22:26:06.409835: step: 1626/1541, loss: 9.018558193929493e-05 2023-01-23 22:26:07.294375: step: 1628/1541, loss: 0.00029306067153811455 2023-01-23 22:26:08.176771: step: 1630/1541, loss: 4.770220402861014e-05 2023-01-23 22:26:09.038146: step: 1632/1541, loss: -1.348903970210813e-05 2023-01-23 22:26:09.922459: step: 1634/1541, loss: -7.182540139183402e-05 2023-01-23 22:26:10.822538: step: 1636/1541, loss: 0.03507348150014877 2023-01-23 22:26:11.686168: step: 1638/1541, loss: 0.0007765335030853748 2023-01-23 22:26:12.568711: step: 1640/1541, loss: 2.6104822609340772e-05 2023-01-23 22:26:13.470072: step: 1642/1541, loss: -1.788394365576096e-05 2023-01-23 22:26:14.346538: step: 1644/1541, loss: 0.08048360794782639 2023-01-23 22:26:15.190019: step: 1646/1541, loss: 0.00217191269621253 2023-01-23 22:26:16.048662: step: 1648/1541, loss: 0.0002653400006238371 2023-01-23 22:26:16.934809: step: 1650/1541, loss: 0.0006907336646690965 2023-01-23 22:26:17.850685: step: 1652/1541, loss: 0.0001644253497943282 2023-01-23 22:26:18.741585: step: 1654/1541, loss: 0.007684723008424044 2023-01-23 22:26:19.648625: step: 1656/1541, loss: 0.000624955166131258 2023-01-23 22:26:20.581352: step: 1658/1541, loss: 0.07370737940073013 2023-01-23 22:26:21.543579: step: 1660/1541, loss: 0.0011051062028855085 2023-01-23 22:26:22.522421: step: 1662/1541, loss: 0.02635819837450981 2023-01-23 22:26:23.405924: step: 1664/1541, loss: 9.903820682666264e-06 2023-01-23 22:26:24.266175: step: 1666/1541, loss: 0.008458772674202919 2023-01-23 22:26:25.143682: step: 1668/1541, loss: 0.0007766926428303123 2023-01-23 22:26:26.043684: step: 1670/1541, loss: 0.03449952229857445 2023-01-23 22:26:26.935927: step: 1672/1541, loss: 0.003526601940393448 2023-01-23 22:26:27.845843: step: 1674/1541, loss: 0.030815741047263145 2023-01-23 22:26:28.737498: step: 1676/1541, loss: 0.001236498006619513 2023-01-23 22:26:29.634144: step: 1678/1541, loss: 0.0472971647977829 2023-01-23 22:26:30.563763: step: 1680/1541, loss: 0.0014709144597873092 2023-01-23 22:26:31.454921: step: 1682/1541, loss: 0.0014730369439348578 2023-01-23 22:26:32.346402: step: 1684/1541, loss: 0.00048746285028755665 2023-01-23 22:26:33.225944: step: 1686/1541, loss: 0.00012111890828236938 2023-01-23 22:26:34.119942: step: 1688/1541, loss: 0.022171007469296455 2023-01-23 22:26:35.044905: step: 1690/1541, loss: 0.0001549336884636432 2023-01-23 22:26:35.963870: step: 1692/1541, loss: 0.0004077971098013222 2023-01-23 22:26:36.891974: step: 1694/1541, loss: 0.0013177270302549005 2023-01-23 22:26:37.755792: step: 1696/1541, loss: 5.126807081978768e-05 2023-01-23 22:26:38.635416: step: 1698/1541, loss: 0.00019531857105903327 2023-01-23 22:26:39.538456: step: 1700/1541, loss: 0.01062665693461895 2023-01-23 22:26:40.391144: step: 1702/1541, loss: 0.02021198347210884 2023-01-23 22:26:41.391582: step: 1704/1541, loss: 0.06047173589468002 2023-01-23 22:26:42.309531: step: 1706/1541, loss: 4.834835635847412e-06 2023-01-23 22:26:43.170482: step: 1708/1541, loss: 0.0004417781892698258 2023-01-23 22:26:44.057407: step: 1710/1541, loss: -1.4490777175524272e-05 2023-01-23 22:26:44.948275: step: 1712/1541, loss: 0.03567296266555786 2023-01-23 22:26:45.910540: step: 1714/1541, loss: 0.13000237941741943 2023-01-23 22:26:46.799216: step: 1716/1541, loss: 0.0001957345666596666 2023-01-23 22:26:47.690032: step: 1718/1541, loss: 0.12292908132076263 2023-01-23 22:26:48.579824: step: 1720/1541, loss: 0.00018322154937777668 2023-01-23 22:26:49.480841: step: 1722/1541, loss: 0.000499582034535706 2023-01-23 22:26:50.444290: step: 1724/1541, loss: 3.577966708689928e-05 2023-01-23 22:26:51.440534: step: 1726/1541, loss: 0.007933170534670353 2023-01-23 22:26:52.335958: step: 1728/1541, loss: 8.951865311246365e-06 2023-01-23 22:26:53.213126: step: 1730/1541, loss: 5.509351467480883e-05 2023-01-23 22:26:54.092095: step: 1732/1541, loss: 9.082299948204309e-05 2023-01-23 22:26:54.947971: step: 1734/1541, loss: -1.191347791973385e-06 2023-01-23 22:26:55.848635: step: 1736/1541, loss: 7.460021151928231e-05 2023-01-23 22:26:56.746024: step: 1738/1541, loss: 0.0008565367897972465 2023-01-23 22:26:57.635644: step: 1740/1541, loss: 0.019245974719524384 2023-01-23 22:26:58.568624: step: 1742/1541, loss: 3.1129249691730365e-05 2023-01-23 22:26:59.442109: step: 1744/1541, loss: 4.184493809589185e-05 2023-01-23 22:27:00.350437: step: 1746/1541, loss: 8.504597644787282e-05 2023-01-23 22:27:01.251405: step: 1748/1541, loss: 0.0005082045681774616 2023-01-23 22:27:02.212106: step: 1750/1541, loss: 0.00030066314502619207 2023-01-23 22:27:03.123545: step: 1752/1541, loss: 2.028290145972278e-05 2023-01-23 22:27:04.033501: step: 1754/1541, loss: 0.0238395556807518 2023-01-23 22:27:04.947632: step: 1756/1541, loss: 0.002405778504908085 2023-01-23 22:27:05.854040: step: 1758/1541, loss: 0.03495974838733673 2023-01-23 22:27:06.754996: step: 1760/1541, loss: 0.19530989229679108 2023-01-23 22:27:07.639122: step: 1762/1541, loss: 0.007015295326709747 2023-01-23 22:27:08.570668: step: 1764/1541, loss: 0.05599288269877434 2023-01-23 22:27:09.498616: step: 1766/1541, loss: 0.00011960689153056592 2023-01-23 22:27:10.393659: step: 1768/1541, loss: 0.25332364439964294 2023-01-23 22:27:11.261069: step: 1770/1541, loss: 0.0027486849576234818 2023-01-23 22:27:12.148643: step: 1772/1541, loss: 0.0002797929337248206 2023-01-23 22:27:13.035828: step: 1774/1541, loss: 0.011884817853569984 2023-01-23 22:27:13.900367: step: 1776/1541, loss: 0.0005506055313162506 2023-01-23 22:27:14.777510: step: 1778/1541, loss: 0.0003839468117803335 2023-01-23 22:27:15.639611: step: 1780/1541, loss: 0.060851383954286575 2023-01-23 22:27:16.555085: step: 1782/1541, loss: 0.10665690898895264 2023-01-23 22:27:17.452578: step: 1784/1541, loss: 0.005234244279563427 2023-01-23 22:27:18.309306: step: 1786/1541, loss: 2.0822764781769365e-05 2023-01-23 22:27:19.203475: step: 1788/1541, loss: 0.00047300654114224017 2023-01-23 22:27:20.107735: step: 1790/1541, loss: 0.05447503551840782 2023-01-23 22:27:20.972550: step: 1792/1541, loss: 0.21457786858081818 2023-01-23 22:27:21.874844: step: 1794/1541, loss: 0.00019100314239040017 2023-01-23 22:27:22.742309: step: 1796/1541, loss: 0.00013961177319288254 2023-01-23 22:27:23.646081: step: 1798/1541, loss: 1.6361470898118569e-06 2023-01-23 22:27:24.517589: step: 1800/1541, loss: 0.001704515889286995 2023-01-23 22:27:25.410466: step: 1802/1541, loss: 4.810392783838324e-05 2023-01-23 22:27:26.261286: step: 1804/1541, loss: 3.54156582034193e-05 2023-01-23 22:27:27.161228: step: 1806/1541, loss: 0.0014555498491972685 2023-01-23 22:27:28.089845: step: 1808/1541, loss: 0.030650677159428596 2023-01-23 22:27:28.980839: step: 1810/1541, loss: 0.09478617459535599 2023-01-23 22:27:29.896885: step: 1812/1541, loss: 0.0012957592261955142 2023-01-23 22:27:30.793242: step: 1814/1541, loss: 0.003150645876303315 2023-01-23 22:27:31.670365: step: 1816/1541, loss: 0.28608444333076477 2023-01-23 22:27:32.623109: step: 1818/1541, loss: 0.01727987267076969 2023-01-23 22:27:33.512004: step: 1820/1541, loss: 0.006205531302839518 2023-01-23 22:27:34.434354: step: 1822/1541, loss: 5.349021012079902e-05 2023-01-23 22:27:35.400396: step: 1824/1541, loss: 3.131516132270917e-05 2023-01-23 22:27:36.280171: step: 1826/1541, loss: 0.02089187502861023 2023-01-23 22:27:37.184692: step: 1828/1541, loss: 2.942541141237598e-05 2023-01-23 22:27:38.102355: step: 1830/1541, loss: 5.949019032414071e-05 2023-01-23 22:27:38.983326: step: 1832/1541, loss: 0.0002718949690461159 2023-01-23 22:27:39.858073: step: 1834/1541, loss: 8.453843474853784e-05 2023-01-23 22:27:40.768829: step: 1836/1541, loss: 0.00034238473745062947 2023-01-23 22:27:41.703032: step: 1838/1541, loss: 0.00020199516438879073 2023-01-23 22:27:42.568398: step: 1840/1541, loss: 0.029134338721632957 2023-01-23 22:27:43.455936: step: 1842/1541, loss: 0.0684163048863411 2023-01-23 22:27:44.361843: step: 1844/1541, loss: -7.009952969383448e-05 2023-01-23 22:27:45.280203: step: 1846/1541, loss: 1.7576814570929855e-05 2023-01-23 22:27:46.141391: step: 1848/1541, loss: 5.720878834836185e-05 2023-01-23 22:27:47.029010: step: 1850/1541, loss: 0.00019071451970376074 2023-01-23 22:27:47.928218: step: 1852/1541, loss: 0.0001082069065887481 2023-01-23 22:27:48.787172: step: 1854/1541, loss: 8.64421235746704e-05 2023-01-23 22:27:49.726000: step: 1856/1541, loss: 0.0001583226548973471 2023-01-23 22:27:50.631838: step: 1858/1541, loss: 0.4200441539287567 2023-01-23 22:27:51.526307: step: 1860/1541, loss: 0.00019318012346047908 2023-01-23 22:27:52.402353: step: 1862/1541, loss: 0.0013104762183502316 2023-01-23 22:27:53.310247: step: 1864/1541, loss: 0.0001361939503112808 2023-01-23 22:27:54.172741: step: 1866/1541, loss: 6.485381891252473e-05 2023-01-23 22:27:55.043992: step: 1868/1541, loss: 7.813270349288359e-05 2023-01-23 22:27:56.018071: step: 1870/1541, loss: 0.024399418383836746 2023-01-23 22:27:56.999204: step: 1872/1541, loss: 9.311587200500071e-05 2023-01-23 22:27:57.883293: step: 1874/1541, loss: 0.00010246402234770358 2023-01-23 22:27:58.798672: step: 1876/1541, loss: 0.00011190010263817385 2023-01-23 22:27:59.726713: step: 1878/1541, loss: 0.0010020824847742915 2023-01-23 22:28:00.571629: step: 1880/1541, loss: 0.001257247058674693 2023-01-23 22:28:01.468110: step: 1882/1541, loss: 6.53787446935894e-06 2023-01-23 22:28:02.387992: step: 1884/1541, loss: 3.475284393061884e-05 2023-01-23 22:28:03.321992: step: 1886/1541, loss: 0.5545393824577332 2023-01-23 22:28:04.262584: step: 1888/1541, loss: 0.6112414002418518 2023-01-23 22:28:05.188990: step: 1890/1541, loss: 0.0032346509397029877 2023-01-23 22:28:06.103260: step: 1892/1541, loss: 0.00010699781705625355 2023-01-23 22:28:07.049767: step: 1894/1541, loss: 5.1851271564373747e-05 2023-01-23 22:28:07.923714: step: 1896/1541, loss: 0.0003450100193731487 2023-01-23 22:28:08.820218: step: 1898/1541, loss: 0.00010267782636219636 2023-01-23 22:28:09.702906: step: 1900/1541, loss: 0.015810102224349976 2023-01-23 22:28:10.602835: step: 1902/1541, loss: 0.054009564220905304 2023-01-23 22:28:11.484361: step: 1904/1541, loss: 0.11492250114679337 2023-01-23 22:28:12.396950: step: 1906/1541, loss: 4.850956247537397e-05 2023-01-23 22:28:13.304739: step: 1908/1541, loss: 0.2457752227783203 2023-01-23 22:28:14.258502: step: 1910/1541, loss: 0.46676716208457947 2023-01-23 22:28:15.179972: step: 1912/1541, loss: 0.0004975329502485693 2023-01-23 22:28:16.076444: step: 1914/1541, loss: 0.00015359398094005883 2023-01-23 22:28:16.958129: step: 1916/1541, loss: 0.02379615232348442 2023-01-23 22:28:17.857648: step: 1918/1541, loss: 5.220559614826925e-06 2023-01-23 22:28:18.737162: step: 1920/1541, loss: 1.5463931049453095e-05 2023-01-23 22:28:19.618894: step: 1922/1541, loss: 9.448363016417716e-06 2023-01-23 22:28:20.537922: step: 1924/1541, loss: 0.026415422558784485 2023-01-23 22:28:21.487959: step: 1926/1541, loss: 0.0017632169183343649 2023-01-23 22:28:22.383357: step: 1928/1541, loss: 0.2748775780200958 2023-01-23 22:28:23.247265: step: 1930/1541, loss: 7.858946628402919e-05 2023-01-23 22:28:24.131647: step: 1932/1541, loss: -1.063819490809692e-05 2023-01-23 22:28:25.087992: step: 1934/1541, loss: 0.9002385139465332 2023-01-23 22:28:25.954373: step: 1936/1541, loss: 2.2034138964954764e-05 2023-01-23 22:28:26.816670: step: 1938/1541, loss: 0.0016940804198384285 2023-01-23 22:28:27.695706: step: 1940/1541, loss: 0.0014713311102241278 2023-01-23 22:28:28.593392: step: 1942/1541, loss: 0.0014659164007753134 2023-01-23 22:28:29.500081: step: 1944/1541, loss: 1.0527967788220849e-05 2023-01-23 22:28:30.360214: step: 1946/1541, loss: 0.038947977125644684 2023-01-23 22:28:31.250391: step: 1948/1541, loss: 8.749315747991204e-05 2023-01-23 22:28:32.161746: step: 1950/1541, loss: 0.0004784971533808857 2023-01-23 22:28:33.079235: step: 1952/1541, loss: 0.3149845600128174 2023-01-23 22:28:33.984182: step: 1954/1541, loss: 5.9350088122300804e-05 2023-01-23 22:28:34.937395: step: 1956/1541, loss: 1.1592600458243396e-05 2023-01-23 22:28:35.892299: step: 1958/1541, loss: 0.0005273577990010381 2023-01-23 22:28:36.800544: step: 1960/1541, loss: 0.04676007851958275 2023-01-23 22:28:37.794093: step: 1962/1541, loss: 0.0005413571489043534 2023-01-23 22:28:38.672941: step: 1964/1541, loss: 0.08505251258611679 2023-01-23 22:28:39.608273: step: 1966/1541, loss: 0.006386688444763422 2023-01-23 22:28:40.490309: step: 1968/1541, loss: 0.0004280083521734923 2023-01-23 22:28:41.369550: step: 1970/1541, loss: 0.0002113006921717897 2023-01-23 22:28:42.290566: step: 1972/1541, loss: 0.007391661871224642 2023-01-23 22:28:43.249793: step: 1974/1541, loss: 5.326049722498283e-05 2023-01-23 22:28:44.110909: step: 1976/1541, loss: 6.279249646468088e-05 2023-01-23 22:28:45.018573: step: 1978/1541, loss: 4.741747761727311e-05 2023-01-23 22:28:45.931828: step: 1980/1541, loss: 0.003456337843090296 2023-01-23 22:28:46.861993: step: 1982/1541, loss: 0.00028995866887271404 2023-01-23 22:28:47.767882: step: 1984/1541, loss: -2.677656084415503e-06 2023-01-23 22:28:48.698386: step: 1986/1541, loss: 6.753328489139676e-05 2023-01-23 22:28:49.604307: step: 1988/1541, loss: 0.08548744022846222 2023-01-23 22:28:50.513684: step: 1990/1541, loss: 6.422486330848187e-05 2023-01-23 22:28:51.445308: step: 1992/1541, loss: 9.421457434655167e-06 2023-01-23 22:28:52.365899: step: 1994/1541, loss: 0.10599222779273987 2023-01-23 22:28:53.290832: step: 1996/1541, loss: 0.000622594088781625 2023-01-23 22:28:54.178464: step: 1998/1541, loss: 0.0011498077074065804 2023-01-23 22:28:55.093205: step: 2000/1541, loss: 4.072780484420946e-06 2023-01-23 22:28:55.990806: step: 2002/1541, loss: 0.00031146343098953366 2023-01-23 22:28:56.853424: step: 2004/1541, loss: 0.16298261284828186 2023-01-23 22:28:57.760567: step: 2006/1541, loss: 0.028472624719142914 2023-01-23 22:28:58.658571: step: 2008/1541, loss: 9.534574928693473e-05 2023-01-23 22:28:59.528382: step: 2010/1541, loss: 0.0013471473939716816 2023-01-23 22:29:00.450964: step: 2012/1541, loss: -3.598926923586987e-05 2023-01-23 22:29:01.483642: step: 2014/1541, loss: 0.0011133183725178242 2023-01-23 22:29:02.335697: step: 2016/1541, loss: -3.761319021577947e-06 2023-01-23 22:29:03.252096: step: 2018/1541, loss: 0.01664905995130539 2023-01-23 22:29:04.142157: step: 2020/1541, loss: -2.0561310520861298e-05 2023-01-23 22:29:05.084494: step: 2022/1541, loss: 0.10340553522109985 2023-01-23 22:29:06.029110: step: 2024/1541, loss: 0.014467085711658001 2023-01-23 22:29:06.885158: step: 2026/1541, loss: 0.010569676756858826 2023-01-23 22:29:07.812811: step: 2028/1541, loss: 0.5131539702415466 2023-01-23 22:29:08.708092: step: 2030/1541, loss: 0.0727466493844986 2023-01-23 22:29:09.657102: step: 2032/1541, loss: 0.0001474790769862011 2023-01-23 22:29:10.543763: step: 2034/1541, loss: 0.00013017700985074043 2023-01-23 22:29:11.413476: step: 2036/1541, loss: 6.119307363405824e-05 2023-01-23 22:29:12.327276: step: 2038/1541, loss: 0.04933863878250122 2023-01-23 22:29:13.213568: step: 2040/1541, loss: 0.08447014540433884 2023-01-23 22:29:14.115426: step: 2042/1541, loss: 0.00040056530269794166 2023-01-23 22:29:14.986946: step: 2044/1541, loss: 5.718387183151208e-05 2023-01-23 22:29:15.863518: step: 2046/1541, loss: 0.0005486158188432455 2023-01-23 22:29:16.765617: step: 2048/1541, loss: 0.0035414223093539476 2023-01-23 22:29:17.659310: step: 2050/1541, loss: 0.1205073744058609 2023-01-23 22:29:18.538653: step: 2052/1541, loss: 0.0003238207718823105 2023-01-23 22:29:19.427955: step: 2054/1541, loss: 0.08955094963312149 2023-01-23 22:29:20.331604: step: 2056/1541, loss: 0.027177387848496437 2023-01-23 22:29:21.216335: step: 2058/1541, loss: 0.001038435846567154 2023-01-23 22:29:22.068588: step: 2060/1541, loss: 0.0007811072282493114 2023-01-23 22:29:22.998252: step: 2062/1541, loss: 0.013505665585398674 2023-01-23 22:29:23.870874: step: 2064/1541, loss: 0.0004682947474066168 2023-01-23 22:29:24.779732: step: 2066/1541, loss: 8.784249803284183e-05 2023-01-23 22:29:25.656021: step: 2068/1541, loss: 0.258734792470932 2023-01-23 22:29:26.579653: step: 2070/1541, loss: 0.2529490888118744 2023-01-23 22:29:27.452823: step: 2072/1541, loss: 0.009444543160498142 2023-01-23 22:29:28.307654: step: 2074/1541, loss: 3.0389395760721527e-05 2023-01-23 22:29:29.206201: step: 2076/1541, loss: 0.0003240599180571735 2023-01-23 22:29:30.055331: step: 2078/1541, loss: 2.2349373466568068e-05 2023-01-23 22:29:30.937005: step: 2080/1541, loss: 0.003690871875733137 2023-01-23 22:29:31.855944: step: 2082/1541, loss: 4.046022513648495e-05 2023-01-23 22:29:32.730679: step: 2084/1541, loss: 0.00016959029017016292 2023-01-23 22:29:33.611454: step: 2086/1541, loss: 5.896141010452993e-05 2023-01-23 22:29:34.519364: step: 2088/1541, loss: 0.0058884103782474995 2023-01-23 22:29:35.384021: step: 2090/1541, loss: 0.0005088407197035849 2023-01-23 22:29:36.277270: step: 2092/1541, loss: 0.0008949295151978731 2023-01-23 22:29:37.162949: step: 2094/1541, loss: 0.024945585057139397 2023-01-23 22:29:38.037584: step: 2096/1541, loss: 2.796132321236655e-07 2023-01-23 22:29:38.915108: step: 2098/1541, loss: 2.5386361812707037e-05 2023-01-23 22:29:39.841791: step: 2100/1541, loss: 5.3979856602381915e-05 2023-01-23 22:29:40.753145: step: 2102/1541, loss: -1.9775827240664512e-05 2023-01-23 22:29:41.646158: step: 2104/1541, loss: 0.1697358638048172 2023-01-23 22:29:42.516081: step: 2106/1541, loss: 1.5596060620737262e-05 2023-01-23 22:29:43.400742: step: 2108/1541, loss: -4.042685031890869e-05 2023-01-23 22:29:44.248113: step: 2110/1541, loss: 3.681453745230101e-05 2023-01-23 22:29:45.180035: step: 2112/1541, loss: 3.7261841498548165e-05 2023-01-23 22:29:46.061021: step: 2114/1541, loss: 0.00010251534695271403 2023-01-23 22:29:46.977960: step: 2116/1541, loss: 2.5007100703078322e-05 2023-01-23 22:29:47.908138: step: 2118/1541, loss: 1.9742561562452465e-05 2023-01-23 22:29:48.824833: step: 2120/1541, loss: 0.0002439659583615139 2023-01-23 22:29:49.712863: step: 2122/1541, loss: 7.70285478211008e-05 2023-01-23 22:29:50.590310: step: 2124/1541, loss: 0.009115003049373627 2023-01-23 22:29:51.501396: step: 2126/1541, loss: 0.00022586286650039256 2023-01-23 22:29:52.376600: step: 2128/1541, loss: 0.012080841697752476 2023-01-23 22:29:53.263209: step: 2130/1541, loss: 0.013730035163462162 2023-01-23 22:29:54.139107: step: 2132/1541, loss: 0.21258273720741272 2023-01-23 22:29:55.050790: step: 2134/1541, loss: -2.6610494387568906e-05 2023-01-23 22:29:56.025818: step: 2136/1541, loss: 0.00013051502173766494 2023-01-23 22:29:56.902805: step: 2138/1541, loss: -7.402152277791174e-06 2023-01-23 22:29:57.826852: step: 2140/1541, loss: 0.0004653192008845508 2023-01-23 22:29:58.737108: step: 2142/1541, loss: 0.00020329069229774177 2023-01-23 22:29:59.611778: step: 2144/1541, loss: 0.0006376464734785259 2023-01-23 22:30:00.550256: step: 2146/1541, loss: 0.007099049631506205 2023-01-23 22:30:01.435934: step: 2148/1541, loss: 0.022823786363005638 2023-01-23 22:30:02.354537: step: 2150/1541, loss: 7.894522241258528e-06 2023-01-23 22:30:03.271784: step: 2152/1541, loss: 0.0001021559874061495 2023-01-23 22:30:04.191239: step: 2154/1541, loss: 0.0006314925849437714 2023-01-23 22:30:05.074854: step: 2156/1541, loss: 2.872255390684586e-05 2023-01-23 22:30:05.956105: step: 2158/1541, loss: 0.0006349379546009004 2023-01-23 22:30:06.812109: step: 2160/1541, loss: 0.007965217344462872 2023-01-23 22:30:07.704408: step: 2162/1541, loss: 0.000133601002744399 2023-01-23 22:30:08.661218: step: 2164/1541, loss: 0.22453747689723969 2023-01-23 22:30:09.568137: step: 2166/1541, loss: 0.0020243460312485695 2023-01-23 22:30:10.480638: step: 2168/1541, loss: 0.0001489438145654276 2023-01-23 22:30:11.398242: step: 2170/1541, loss: 9.770307224243879e-05 2023-01-23 22:30:12.312816: step: 2172/1541, loss: -1.4805795217398554e-05 2023-01-23 22:30:13.225410: step: 2174/1541, loss: 0.01007198914885521 2023-01-23 22:30:14.142611: step: 2176/1541, loss: 0.06787361949682236 2023-01-23 22:30:15.033747: step: 2178/1541, loss: 0.006789712235331535 2023-01-23 22:30:15.937159: step: 2180/1541, loss: 6.977732118684798e-05 2023-01-23 22:30:16.798276: step: 2182/1541, loss: 0.042401738464832306 2023-01-23 22:30:17.707713: step: 2184/1541, loss: 0.0009404790471307933 2023-01-23 22:30:18.596950: step: 2186/1541, loss: 0.06191227212548256 2023-01-23 22:30:19.464484: step: 2188/1541, loss: 0.09564824402332306 2023-01-23 22:30:20.359263: step: 2190/1541, loss: 2.295782178407535e-05 2023-01-23 22:30:21.280846: step: 2192/1541, loss: 0.007783511187881231 2023-01-23 22:30:22.224648: step: 2194/1541, loss: 0.00013274062075652182 2023-01-23 22:30:23.120705: step: 2196/1541, loss: 0.00042128225322812796 2023-01-23 22:30:24.012685: step: 2198/1541, loss: 0.5400719046592712 2023-01-23 22:30:24.993974: step: 2200/1541, loss: 7.571953574370127e-06 2023-01-23 22:30:25.874467: step: 2202/1541, loss: 9.351711923955008e-05 2023-01-23 22:30:26.734313: step: 2204/1541, loss: 0.0005425597191788256 2023-01-23 22:30:27.613347: step: 2206/1541, loss: 0.0007050956482999027 2023-01-23 22:30:28.489426: step: 2208/1541, loss: 0.3014461100101471 2023-01-23 22:30:29.372701: step: 2210/1541, loss: 0.15596230328083038 2023-01-23 22:30:30.341176: step: 2212/1541, loss: 0.0017560984706506133 2023-01-23 22:30:31.248120: step: 2214/1541, loss: 0.5365473628044128 2023-01-23 22:30:32.152664: step: 2216/1541, loss: 0.001720926957204938 2023-01-23 22:30:33.013982: step: 2218/1541, loss: 8.200799493351951e-05 2023-01-23 22:30:33.899933: step: 2220/1541, loss: 0.09469674527645111 2023-01-23 22:30:34.793138: step: 2222/1541, loss: 0.3025040328502655 2023-01-23 22:30:35.712301: step: 2224/1541, loss: 0.0718548372387886 2023-01-23 22:30:36.577171: step: 2226/1541, loss: 4.7198936954373494e-05 2023-01-23 22:30:37.474418: step: 2228/1541, loss: 0.004919333849102259 2023-01-23 22:30:38.418952: step: 2230/1541, loss: 0.0007872513378970325 2023-01-23 22:30:39.254850: step: 2232/1541, loss: 0.00043973885476589203 2023-01-23 22:30:40.195639: step: 2234/1541, loss: 0.002755194902420044 2023-01-23 22:30:41.084202: step: 2236/1541, loss: 8.420143785770051e-06 2023-01-23 22:30:41.986896: step: 2238/1541, loss: 9.008208144223318e-05 2023-01-23 22:30:42.900412: step: 2240/1541, loss: 5.01972135680262e-05 2023-01-23 22:30:43.770113: step: 2242/1541, loss: 0.029019979760050774 2023-01-23 22:30:44.642307: step: 2244/1541, loss: 0.00012838974362239242 2023-01-23 22:30:45.533215: step: 2246/1541, loss: 4.764248660649173e-05 2023-01-23 22:30:46.443990: step: 2248/1541, loss: 0.00043101684423163533 2023-01-23 22:30:47.363154: step: 2250/1541, loss: 0.0002775554603431374 2023-01-23 22:30:48.227854: step: 2252/1541, loss: 7.978081703186035e-05 2023-01-23 22:30:49.122232: step: 2254/1541, loss: 0.001604599878191948 2023-01-23 22:30:50.102090: step: 2256/1541, loss: 8.361531945411116e-05 2023-01-23 22:30:50.959114: step: 2258/1541, loss: 8.206169150071219e-05 2023-01-23 22:30:51.838404: step: 2260/1541, loss: 0.0007330164080485702 2023-01-23 22:30:52.736824: step: 2262/1541, loss: 0.34756433963775635 2023-01-23 22:30:53.637830: step: 2264/1541, loss: 0.02778281271457672 2023-01-23 22:30:54.592959: step: 2266/1541, loss: 8.625015470897779e-05 2023-01-23 22:30:55.470506: step: 2268/1541, loss: 0.0005955058732070029 2023-01-23 22:30:56.401532: step: 2270/1541, loss: 0.0010829012608155608 2023-01-23 22:30:57.292687: step: 2272/1541, loss: 0.00014201257727108896 2023-01-23 22:30:58.187749: step: 2274/1541, loss: 0.00014248579100240022 2023-01-23 22:30:59.098024: step: 2276/1541, loss: 0.0001736422418616712 2023-01-23 22:31:00.014054: step: 2278/1541, loss: 0.00021462904987856746 2023-01-23 22:31:00.942601: step: 2280/1541, loss: 0.013637052848935127 2023-01-23 22:31:01.806378: step: 2282/1541, loss: 0.1140907034277916 2023-01-23 22:31:02.681411: step: 2284/1541, loss: 1.7982717963604955e-06 2023-01-23 22:31:03.623646: step: 2286/1541, loss: 3.430662036407739e-05 2023-01-23 22:31:04.525738: step: 2288/1541, loss: 9.036450501298532e-05 2023-01-23 22:31:05.417281: step: 2290/1541, loss: 0.7477787137031555 2023-01-23 22:31:06.279321: step: 2292/1541, loss: -2.12875002034707e-05 2023-01-23 22:31:07.149892: step: 2294/1541, loss: 0.0038900403305888176 2023-01-23 22:31:08.049077: step: 2296/1541, loss: 0.650916337966919 2023-01-23 22:31:08.952811: step: 2298/1541, loss: 0.0001713705132715404 2023-01-23 22:31:09.804413: step: 2300/1541, loss: 0.001175837591290474 2023-01-23 22:31:10.653280: step: 2302/1541, loss: 0.0025146922562271357 2023-01-23 22:31:11.497356: step: 2304/1541, loss: 0.013761797919869423 2023-01-23 22:31:12.468269: step: 2306/1541, loss: 0.00014618052227888256 2023-01-23 22:31:13.398389: step: 2308/1541, loss: 0.001566962106153369 2023-01-23 22:31:14.335638: step: 2310/1541, loss: 0.01522433664649725 2023-01-23 22:31:15.232963: step: 2312/1541, loss: 0.12841370701789856 2023-01-23 22:31:16.162815: step: 2314/1541, loss: 0.059782348573207855 2023-01-23 22:31:17.062052: step: 2316/1541, loss: 6.389642658177763e-05 2023-01-23 22:31:17.955875: step: 2318/1541, loss: 0.0004963663523085415 2023-01-23 22:31:18.869850: step: 2320/1541, loss: 0.004832261707633734 2023-01-23 22:31:19.770818: step: 2322/1541, loss: 9.414006490260363e-05 2023-01-23 22:31:20.621827: step: 2324/1541, loss: 0.01817714236676693 2023-01-23 22:31:21.531587: step: 2326/1541, loss: 0.2284332662820816 2023-01-23 22:31:22.480005: step: 2328/1541, loss: 0.00025442230980843306 2023-01-23 22:31:23.371488: step: 2330/1541, loss: 0.0014721560291945934 2023-01-23 22:31:24.276548: step: 2332/1541, loss: 0.0008403299143537879 2023-01-23 22:31:25.171980: step: 2334/1541, loss: 0.00010387425572844222 2023-01-23 22:31:26.043310: step: 2336/1541, loss: 0.0006141698104329407 2023-01-23 22:31:27.023567: step: 2338/1541, loss: 0.05104599893093109 2023-01-23 22:31:27.917683: step: 2340/1541, loss: 0.3162803053855896 2023-01-23 22:31:28.818392: step: 2342/1541, loss: 0.19909852743148804 2023-01-23 22:31:29.679236: step: 2344/1541, loss: 0.019854238256812096 2023-01-23 22:31:30.564645: step: 2346/1541, loss: -2.23517560016262e-07 2023-01-23 22:31:31.573367: step: 2348/1541, loss: 0.11486835032701492 2023-01-23 22:31:32.456022: step: 2350/1541, loss: -1.985033850360196e-05 2023-01-23 22:31:33.342108: step: 2352/1541, loss: 0.45071741938591003 2023-01-23 22:31:34.232884: step: 2354/1541, loss: 0.0031753787770867348 2023-01-23 22:31:35.168375: step: 2356/1541, loss: 0.0018440396524965763 2023-01-23 22:31:36.119289: step: 2358/1541, loss: 0.014878218993544579 2023-01-23 22:31:36.977463: step: 2360/1541, loss: -3.2531256692891475e-06 2023-01-23 22:31:37.816046: step: 2362/1541, loss: 0.013058479875326157 2023-01-23 22:31:38.663861: step: 2364/1541, loss: 1.3746809599979315e-05 2023-01-23 22:31:39.568329: step: 2366/1541, loss: 0.0007526411209255457 2023-01-23 22:31:40.472434: step: 2368/1541, loss: 0.10384602099657059 2023-01-23 22:31:41.338844: step: 2370/1541, loss: 0.041265543550252914 2023-01-23 22:31:42.315808: step: 2372/1541, loss: 0.09272107481956482 2023-01-23 22:31:43.169278: step: 2374/1541, loss: 0.00016747546033002436 2023-01-23 22:31:44.095621: step: 2376/1541, loss: 0.10894312709569931 2023-01-23 22:31:44.951684: step: 2378/1541, loss: 9.110513201449066e-05 2023-01-23 22:31:45.844641: step: 2380/1541, loss: 0.2635439336299896 2023-01-23 22:31:46.804226: step: 2382/1541, loss: 0.028182992711663246 2023-01-23 22:31:47.707617: step: 2384/1541, loss: 0.0017762086354196072 2023-01-23 22:31:48.587160: step: 2386/1541, loss: 0.000130934888147749 2023-01-23 22:31:49.449513: step: 2388/1541, loss: 0.011303690262138844 2023-01-23 22:31:50.316250: step: 2390/1541, loss: 0.019398558884859085 2023-01-23 22:31:51.189884: step: 2392/1541, loss: 0.0004917398910038173 2023-01-23 22:31:52.130955: step: 2394/1541, loss: 0.0003731252800207585 2023-01-23 22:31:52.997996: step: 2396/1541, loss: 0.000545885821338743 2023-01-23 22:31:53.850677: step: 2398/1541, loss: 0.0013216197257861495 2023-01-23 22:31:54.723121: step: 2400/1541, loss: 0.0034395402763038874 2023-01-23 22:31:55.614943: step: 2402/1541, loss: 0.2562445104122162 2023-01-23 22:31:56.487784: step: 2404/1541, loss: 0.022625405341386795 2023-01-23 22:31:57.400843: step: 2406/1541, loss: 0.00060011021560058 2023-01-23 22:31:58.263708: step: 2408/1541, loss: 0.0758417472243309 2023-01-23 22:31:59.188943: step: 2410/1541, loss: 0.009414280764758587 2023-01-23 22:32:00.030903: step: 2412/1541, loss: 0.00017665603081695735 2023-01-23 22:32:00.917096: step: 2414/1541, loss: 0.0011262597981840372 2023-01-23 22:32:01.819068: step: 2416/1541, loss: 0.09684573858976364 2023-01-23 22:32:02.716554: step: 2418/1541, loss: 9.824557491810992e-05 2023-01-23 22:32:03.621113: step: 2420/1541, loss: 0.4819212555885315 2023-01-23 22:32:04.501771: step: 2422/1541, loss: 0.005911757703870535 2023-01-23 22:32:05.392812: step: 2424/1541, loss: 0.0001179465907625854 2023-01-23 22:32:06.287521: step: 2426/1541, loss: 0.00012074944243067876 2023-01-23 22:32:07.182576: step: 2428/1541, loss: 0.001172687509097159 2023-01-23 22:32:08.123900: step: 2430/1541, loss: 0.0900534987449646 2023-01-23 22:32:09.005909: step: 2432/1541, loss: 0.000752258871216327 2023-01-23 22:32:09.896251: step: 2434/1541, loss: 0.0025191267486661673 2023-01-23 22:32:10.780432: step: 2436/1541, loss: 0.0019359519938006997 2023-01-23 22:32:11.704591: step: 2438/1541, loss: 0.09083770960569382 2023-01-23 22:32:12.587370: step: 2440/1541, loss: 6.1124569583626e-06 2023-01-23 22:32:13.549896: step: 2442/1541, loss: 0.00046015973202884197 2023-01-23 22:32:14.472133: step: 2444/1541, loss: 0.04722238704562187 2023-01-23 22:32:15.432986: step: 2446/1541, loss: 0.00014841556549072266 2023-01-23 22:32:16.327919: step: 2448/1541, loss: 4.925429311697371e-05 2023-01-23 22:32:17.228610: step: 2450/1541, loss: 0.003350636688992381 2023-01-23 22:32:18.093936: step: 2452/1541, loss: 0.0024757871869951487 2023-01-23 22:32:19.022666: step: 2454/1541, loss: 0.029020506888628006 2023-01-23 22:32:19.940839: step: 2456/1541, loss: 0.12026183307170868 2023-01-23 22:32:20.852764: step: 2458/1541, loss: 2.2999952307145577e-06 2023-01-23 22:32:21.824314: step: 2460/1541, loss: 0.06532294303178787 2023-01-23 22:32:22.689161: step: 2462/1541, loss: 1.575465103087481e-05 2023-01-23 22:32:23.617363: step: 2464/1541, loss: 0.46408674120903015 2023-01-23 22:32:24.506492: step: 2466/1541, loss: 0.00022296150564216077 2023-01-23 22:32:25.410448: step: 2468/1541, loss: 0.010025566443800926 2023-01-23 22:32:26.284413: step: 2470/1541, loss: 0.025433797389268875 2023-01-23 22:32:27.164060: step: 2472/1541, loss: 0.000850990938488394 2023-01-23 22:32:28.035332: step: 2474/1541, loss: 0.03123455122113228 2023-01-23 22:32:28.924304: step: 2476/1541, loss: 0.008140613324940205 2023-01-23 22:32:29.782010: step: 2478/1541, loss: 0.00046463386388495564 2023-01-23 22:32:30.653651: step: 2480/1541, loss: 0.0021824350114911795 2023-01-23 22:32:31.522757: step: 2482/1541, loss: 0.3095216155052185 2023-01-23 22:32:32.409168: step: 2484/1541, loss: 1.9787716155406088e-05 2023-01-23 22:32:33.325282: step: 2486/1541, loss: 0.17332765460014343 2023-01-23 22:32:34.296013: step: 2488/1541, loss: 0.47511202096939087 2023-01-23 22:32:35.225017: step: 2490/1541, loss: -1.666237585595809e-05 2023-01-23 22:32:36.094142: step: 2492/1541, loss: 0.032741986215114594 2023-01-23 22:32:36.991582: step: 2494/1541, loss: 0.016777321696281433 2023-01-23 22:32:37.985155: step: 2496/1541, loss: 0.0016833861591294408 2023-01-23 22:32:38.867100: step: 2498/1541, loss: 0.0004155601200181991 2023-01-23 22:32:39.722648: step: 2500/1541, loss: 0.0026535755023360252 2023-01-23 22:32:40.609694: step: 2502/1541, loss: 0.0018827226012945175 2023-01-23 22:32:41.491711: step: 2504/1541, loss: 1.9196637367713265e-05 2023-01-23 22:32:42.362966: step: 2506/1541, loss: 0.0004677141550928354 2023-01-23 22:32:43.235794: step: 2508/1541, loss: 0.007374242413789034 2023-01-23 22:32:44.121655: step: 2510/1541, loss: 0.00011200681910850108 2023-01-23 22:32:44.985558: step: 2512/1541, loss: 0.08773643523454666 2023-01-23 22:32:45.920736: step: 2514/1541, loss: 0.274566650390625 2023-01-23 22:32:46.797599: step: 2516/1541, loss: 2.348819725739304e-05 2023-01-23 22:32:47.659184: step: 2518/1541, loss: 1.4517618183162995e-05 2023-01-23 22:32:48.520677: step: 2520/1541, loss: 0.00023438215430360287 2023-01-23 22:32:49.417605: step: 2522/1541, loss: 0.0003853722882922739 2023-01-23 22:32:50.288686: step: 2524/1541, loss: 3.491242750897072e-05 2023-01-23 22:32:51.208204: step: 2526/1541, loss: 0.00013052983558736742 2023-01-23 22:32:52.121619: step: 2528/1541, loss: 0.0011878430377691984 2023-01-23 22:32:53.020109: step: 2530/1541, loss: 0.0726211816072464 2023-01-23 22:32:53.898246: step: 2532/1541, loss: 4.8578804126009345e-06 2023-01-23 22:32:54.783645: step: 2534/1541, loss: 0.0793897733092308 2023-01-23 22:32:55.724595: step: 2536/1541, loss: 7.136729982448742e-05 2023-01-23 22:32:56.636687: step: 2538/1541, loss: 8.840664668241516e-05 2023-01-23 22:32:57.562977: step: 2540/1541, loss: 1.712190714897588e-05 2023-01-23 22:32:58.432592: step: 2542/1541, loss: 0.11703172326087952 2023-01-23 22:32:59.330664: step: 2544/1541, loss: 6.894170837767888e-06 2023-01-23 22:33:00.228156: step: 2546/1541, loss: 0.09031716734170914 2023-01-23 22:33:01.142245: step: 2548/1541, loss: 0.001362306298688054 2023-01-23 22:33:02.037356: step: 2550/1541, loss: 2.8719334295601584e-05 2023-01-23 22:33:02.932999: step: 2552/1541, loss: 0.03004583716392517 2023-01-23 22:33:03.826840: step: 2554/1541, loss: 8.659482409711927e-05 2023-01-23 22:33:04.669994: step: 2556/1541, loss: 0.00014096114318817854 2023-01-23 22:33:05.626925: step: 2558/1541, loss: 6.41257647657767e-05 2023-01-23 22:33:06.518723: step: 2560/1541, loss: -1.6391677490901202e-05 2023-01-23 22:33:07.488946: step: 2562/1541, loss: 0.004819874186068773 2023-01-23 22:33:08.388490: step: 2564/1541, loss: 0.001020062598399818 2023-01-23 22:33:09.233005: step: 2566/1541, loss: 2.0101650079595856e-05 2023-01-23 22:33:10.242487: step: 2568/1541, loss: -2.0198523998260498e-05 2023-01-23 22:33:11.090988: step: 2570/1541, loss: 0.006313717924058437 2023-01-23 22:33:11.935767: step: 2572/1541, loss: 0.12539416551589966 2023-01-23 22:33:12.826322: step: 2574/1541, loss: 0.021526547148823738 2023-01-23 22:33:13.683593: step: 2576/1541, loss: 6.885074981255457e-05 2023-01-23 22:33:14.548617: step: 2578/1541, loss: 0.031168146058917046 2023-01-23 22:33:15.459807: step: 2580/1541, loss: 4.578143489197828e-05 2023-01-23 22:33:16.378765: step: 2582/1541, loss: 0.004560110624879599 2023-01-23 22:33:17.274655: step: 2584/1541, loss: 0.0003404925810173154 2023-01-23 22:33:18.135737: step: 2586/1541, loss: 0.01480330154299736 2023-01-23 22:33:19.023416: step: 2588/1541, loss: 5.001425961381756e-05 2023-01-23 22:33:19.907426: step: 2590/1541, loss: 0.026586655527353287 2023-01-23 22:33:20.768185: step: 2592/1541, loss: 3.21945954055991e-05 2023-01-23 22:33:21.702627: step: 2594/1541, loss: 0.004187673795968294 2023-01-23 22:33:22.569088: step: 2596/1541, loss: 0.08143740892410278 2023-01-23 22:33:23.446424: step: 2598/1541, loss: 0.021853547543287277 2023-01-23 22:33:24.362103: step: 2600/1541, loss: 0.00369580602273345 2023-01-23 22:33:25.275885: step: 2602/1541, loss: 0.08601178973913193 2023-01-23 22:33:26.156749: step: 2604/1541, loss: 0.0002889176248572767 2023-01-23 22:33:27.084004: step: 2606/1541, loss: 0.016154099255800247 2023-01-23 22:33:28.015220: step: 2608/1541, loss: 0.000865040288772434 2023-01-23 22:33:28.979717: step: 2610/1541, loss: 0.28340962529182434 2023-01-23 22:33:29.881068: step: 2612/1541, loss: 0.0007909398409537971 2023-01-23 22:33:30.742856: step: 2614/1541, loss: 0.0038983544800430536 2023-01-23 22:33:31.628117: step: 2616/1541, loss: 0.14779551327228546 2023-01-23 22:33:32.569747: step: 2618/1541, loss: 0.0045385053381323814 2023-01-23 22:33:33.478657: step: 2620/1541, loss: 0.004814913030713797 2023-01-23 22:33:34.365365: step: 2622/1541, loss: 0.004665516782552004 2023-01-23 22:33:35.247208: step: 2624/1541, loss: 0.1869255006313324 2023-01-23 22:33:36.126262: step: 2626/1541, loss: 0.00011495115177240223 2023-01-23 22:33:37.073104: step: 2628/1541, loss: 0.0019317739643156528 2023-01-23 22:33:38.010959: step: 2630/1541, loss: 0.0009733967017382383 2023-01-23 22:33:38.958731: step: 2632/1541, loss: 0.020302195101976395 2023-01-23 22:33:39.870743: step: 2634/1541, loss: 0.023897558450698853 2023-01-23 22:33:40.817987: step: 2636/1541, loss: 0.8704795837402344 2023-01-23 22:33:41.792128: step: 2638/1541, loss: 9.003221202874556e-05 2023-01-23 22:33:42.694882: step: 2640/1541, loss: 0.2783878445625305 2023-01-23 22:33:43.604024: step: 2642/1541, loss: 0.006357214413583279 2023-01-23 22:33:44.482729: step: 2644/1541, loss: 0.05859101563692093 2023-01-23 22:33:45.334591: step: 2646/1541, loss: 0.008097464218735695 2023-01-23 22:33:46.181624: step: 2648/1541, loss: 5.9219193644821644e-05 2023-01-23 22:33:47.079471: step: 2650/1541, loss: 0.011282616294920444 2023-01-23 22:33:47.966058: step: 2652/1541, loss: 0.25553497672080994 2023-01-23 22:33:48.881159: step: 2654/1541, loss: 0.030781488865613937 2023-01-23 22:33:49.716652: step: 2656/1541, loss: 9.267024870496243e-05 2023-01-23 22:33:50.615336: step: 2658/1541, loss: 0.22813533246517181 2023-01-23 22:33:51.566144: step: 2660/1541, loss: 0.11243055760860443 2023-01-23 22:33:52.433543: step: 2662/1541, loss: 0.251596599817276 2023-01-23 22:33:53.283797: step: 2664/1541, loss: 0.5608062148094177 2023-01-23 22:33:54.156183: step: 2666/1541, loss: 0.002756627043709159 2023-01-23 22:33:55.098053: step: 2668/1541, loss: 0.0009569954127073288 2023-01-23 22:33:56.006828: step: 2670/1541, loss: 0.0005004793056286871 2023-01-23 22:33:56.882643: step: 2672/1541, loss: 0.0007396779838018119 2023-01-23 22:33:57.761686: step: 2674/1541, loss: 0.00013897240569349378 2023-01-23 22:33:58.646610: step: 2676/1541, loss: 0.008283890783786774 2023-01-23 22:33:59.536167: step: 2678/1541, loss: 0.00037564270314760506 2023-01-23 22:34:00.406919: step: 2680/1541, loss: 0.002276524668559432 2023-01-23 22:34:01.292275: step: 2682/1541, loss: 0.0011656263377517462 2023-01-23 22:34:02.202742: step: 2684/1541, loss: 0.0023418653290718794 2023-01-23 22:34:03.113982: step: 2686/1541, loss: 0.021938825026154518 2023-01-23 22:34:03.975082: step: 2688/1541, loss: 0.0005008485168218613 2023-01-23 22:34:04.877529: step: 2690/1541, loss: 0.0001217999670188874 2023-01-23 22:34:05.787157: step: 2692/1541, loss: 0.0024085072800517082 2023-01-23 22:34:06.667141: step: 2694/1541, loss: -1.1141231880174018e-05 2023-01-23 22:34:07.560641: step: 2696/1541, loss: 0.0005295570008456707 2023-01-23 22:34:08.446710: step: 2698/1541, loss: 0.001126863993704319 2023-01-23 22:34:09.338016: step: 2700/1541, loss: 0.0019448952516540885 2023-01-23 22:34:10.258572: step: 2702/1541, loss: 0.00012673158198595047 2023-01-23 22:34:11.199754: step: 2704/1541, loss: 0.0031111272983253 2023-01-23 22:34:12.171533: step: 2706/1541, loss: 0.003159115556627512 2023-01-23 22:34:13.033976: step: 2708/1541, loss: 0.002574312500655651 2023-01-23 22:34:13.948982: step: 2710/1541, loss: 0.12695962190628052 2023-01-23 22:34:14.888083: step: 2712/1541, loss: 0.00019552795856725425 2023-01-23 22:34:15.827915: step: 2714/1541, loss: 0.00037817080738022923 2023-01-23 22:34:16.702009: step: 2716/1541, loss: 4.915644240099937e-05 2023-01-23 22:34:17.618278: step: 2718/1541, loss: 0.0004361324245110154 2023-01-23 22:34:18.509916: step: 2720/1541, loss: -1.5974788766470738e-05 2023-01-23 22:34:19.429590: step: 2722/1541, loss: 0.0009085829951800406 2023-01-23 22:34:20.352760: step: 2724/1541, loss: 0.0008652337710373104 2023-01-23 22:34:21.212125: step: 2726/1541, loss: 0.02542891539633274 2023-01-23 22:34:22.105992: step: 2728/1541, loss: 8.926913869800046e-05 2023-01-23 22:34:23.044175: step: 2730/1541, loss: 0.6594635248184204 2023-01-23 22:34:23.930960: step: 2732/1541, loss: 0.7935692667961121 2023-01-23 22:34:24.793122: step: 2734/1541, loss: 4.5698503527091816e-05 2023-01-23 22:34:25.653668: step: 2736/1541, loss: 0.003727525472640991 2023-01-23 22:34:26.553310: step: 2738/1541, loss: 1.687106123426929e-05 2023-01-23 22:34:27.447930: step: 2740/1541, loss: 9.155521547654644e-05 2023-01-23 22:34:28.379841: step: 2742/1541, loss: 0.0001535276387585327 2023-01-23 22:34:29.247387: step: 2744/1541, loss: 0.0466717854142189 2023-01-23 22:34:30.143677: step: 2746/1541, loss: 0.007267594803124666 2023-01-23 22:34:31.043904: step: 2748/1541, loss: 0.012964251451194286 2023-01-23 22:34:31.938418: step: 2750/1541, loss: 7.20127864042297e-05 2023-01-23 22:34:32.826009: step: 2752/1541, loss: 0.003947909455746412 2023-01-23 22:34:33.671969: step: 2754/1541, loss: -1.5159451322688255e-05 2023-01-23 22:34:34.542535: step: 2756/1541, loss: 0.11784641444683075 2023-01-23 22:34:35.519855: step: 2758/1541, loss: -4.6716151700820774e-05 2023-01-23 22:34:36.375901: step: 2760/1541, loss: 0.0007181338733062148 2023-01-23 22:34:37.281631: step: 2762/1541, loss: 5.274398426990956e-05 2023-01-23 22:34:38.169076: step: 2764/1541, loss: 0.00020742845663335174 2023-01-23 22:34:39.069202: step: 2766/1541, loss: 0.005658628419041634 2023-01-23 22:34:39.967966: step: 2768/1541, loss: 6.237028628675034e-06 2023-01-23 22:34:40.903788: step: 2770/1541, loss: 0.007609610445797443 2023-01-23 22:34:41.809191: step: 2772/1541, loss: 0.28842687606811523 2023-01-23 22:34:42.683736: step: 2774/1541, loss: 8.993842129711993e-06 2023-01-23 22:34:43.576987: step: 2776/1541, loss: 8.380901999771595e-05 2023-01-23 22:34:44.482215: step: 2778/1541, loss: -6.503648819489172e-06 2023-01-23 22:34:45.465991: step: 2780/1541, loss: 0.0006912194658070803 2023-01-23 22:34:46.339722: step: 2782/1541, loss: -2.66715887846658e-05 2023-01-23 22:34:47.216579: step: 2784/1541, loss: 0.009130042046308517 2023-01-23 22:34:48.101393: step: 2786/1541, loss: 0.012748733162879944 2023-01-23 22:34:48.973192: step: 2788/1541, loss: 0.007863115519285202 2023-01-23 22:34:49.876658: step: 2790/1541, loss: 0.1417107731103897 2023-01-23 22:34:50.781261: step: 2792/1541, loss: 4.158206138527021e-05 2023-01-23 22:34:51.663737: step: 2794/1541, loss: 0.004054918419569731 2023-01-23 22:34:52.526845: step: 2796/1541, loss: -3.6921723221894354e-05 2023-01-23 22:34:53.420441: step: 2798/1541, loss: 0.00017433165339753032 2023-01-23 22:34:54.262124: step: 2800/1541, loss: 0.012091828510165215 2023-01-23 22:34:55.153848: step: 2802/1541, loss: 2.5177003408316523e-05 2023-01-23 22:34:56.011186: step: 2804/1541, loss: 0.00010625395952956751 2023-01-23 22:34:56.876480: step: 2806/1541, loss: 0.004242708906531334 2023-01-23 22:34:57.755831: step: 2808/1541, loss: 0.00046997220488265157 2023-01-23 22:34:58.673032: step: 2810/1541, loss: 0.6214064955711365 2023-01-23 22:34:59.561083: step: 2812/1541, loss: 0.0004646798479370773 2023-01-23 22:35:00.428064: step: 2814/1541, loss: 0.0013626208528876305 2023-01-23 22:35:01.284111: step: 2816/1541, loss: 0.0010145221604034305 2023-01-23 22:35:02.201689: step: 2818/1541, loss: 3.5101671528536826e-05 2023-01-23 22:35:03.093424: step: 2820/1541, loss: 0.13564300537109375 2023-01-23 22:35:03.984693: step: 2822/1541, loss: 3.4421686905261595e-06 2023-01-23 22:35:04.884285: step: 2824/1541, loss: 0.00010076910257339478 2023-01-23 22:35:05.788141: step: 2826/1541, loss: 5.380585935199633e-05 2023-01-23 22:35:06.696758: step: 2828/1541, loss: 0.00030368566513061523 2023-01-23 22:35:07.573800: step: 2830/1541, loss: 9.6857536391326e-07 2023-01-23 22:35:08.427097: step: 2832/1541, loss: -1.4110548363532871e-05 2023-01-23 22:35:09.367416: step: 2834/1541, loss: 9.21220489544794e-05 2023-01-23 22:35:10.242282: step: 2836/1541, loss: -1.4495848517981358e-05 2023-01-23 22:35:11.152488: step: 2838/1541, loss: 0.17665469646453857 2023-01-23 22:35:12.106538: step: 2840/1541, loss: 0.6343620419502258 2023-01-23 22:35:12.985636: step: 2842/1541, loss: 2.1072724848636426e-05 2023-01-23 22:35:13.852113: step: 2844/1541, loss: 1.922746605487191e-06 2023-01-23 22:35:14.699924: step: 2846/1541, loss: 0.00010515550820855424 2023-01-23 22:35:15.622070: step: 2848/1541, loss: 0.00043062420445494354 2023-01-23 22:35:16.506016: step: 2850/1541, loss: 0.12627656757831573 2023-01-23 22:35:17.374869: step: 2852/1541, loss: 0.6621265411376953 2023-01-23 22:35:18.292814: step: 2854/1541, loss: 0.23198723793029785 2023-01-23 22:35:19.149906: step: 2856/1541, loss: 0.0014560980489477515 2023-01-23 22:35:20.047766: step: 2858/1541, loss: 1.2633240222930908 2023-01-23 22:35:20.953307: step: 2860/1541, loss: 0.0008917806553654373 2023-01-23 22:35:21.826773: step: 2862/1541, loss: 0.06741928309202194 2023-01-23 22:35:22.726774: step: 2864/1541, loss: 0.3513329327106476 2023-01-23 22:35:23.635734: step: 2866/1541, loss: 1.9941202481277287e-05 2023-01-23 22:35:24.576116: step: 2868/1541, loss: 0.01767195388674736 2023-01-23 22:35:25.450407: step: 2870/1541, loss: 0.008311783894896507 2023-01-23 22:35:26.340306: step: 2872/1541, loss: 0.019049759954214096 2023-01-23 22:35:27.227325: step: 2874/1541, loss: 0.0001217007520608604 2023-01-23 22:35:28.103270: step: 2876/1541, loss: 0.018235161900520325 2023-01-23 22:35:28.951998: step: 2878/1541, loss: 0.0006433548405766487 2023-01-23 22:35:29.846068: step: 2880/1541, loss: 0.1620931774377823 2023-01-23 22:35:30.780874: step: 2882/1541, loss: 0.08312371373176575 2023-01-23 22:35:31.662960: step: 2884/1541, loss: 0.2658223509788513 2023-01-23 22:35:32.561899: step: 2886/1541, loss: 0.0068293348886072636 2023-01-23 22:35:33.445050: step: 2888/1541, loss: 0.0003009891661349684 2023-01-23 22:35:34.383067: step: 2890/1541, loss: 0.14984731376171112 2023-01-23 22:35:35.258595: step: 2892/1541, loss: 0.005298467352986336 2023-01-23 22:35:36.197041: step: 2894/1541, loss: 0.01223302073776722 2023-01-23 22:35:37.094554: step: 2896/1541, loss: -3.700213710544631e-05 2023-01-23 22:35:38.002731: step: 2898/1541, loss: 0.04901650547981262 2023-01-23 22:35:38.898994: step: 2900/1541, loss: -7.50203980715014e-05 2023-01-23 22:35:39.800961: step: 2902/1541, loss: -1.5491248632315546e-05 2023-01-23 22:35:40.678712: step: 2904/1541, loss: 0.0005424824194051325 2023-01-23 22:35:41.610533: step: 2906/1541, loss: 0.03761947154998779 2023-01-23 22:35:42.519661: step: 2908/1541, loss: 0.08289078623056412 2023-01-23 22:35:43.454040: step: 2910/1541, loss: 0.0017889788141474128 2023-01-23 22:35:44.341328: step: 2912/1541, loss: 0.0429394356906414 2023-01-23 22:35:45.220264: step: 2914/1541, loss: 7.815059507265687e-05 2023-01-23 22:35:46.129627: step: 2916/1541, loss: 0.00015717739006504416 2023-01-23 22:35:47.027361: step: 2918/1541, loss: 0.012175790034234524 2023-01-23 22:35:47.990287: step: 2920/1541, loss: 5.0967501010745764e-05 2023-01-23 22:35:48.891069: step: 2922/1541, loss: 0.03269575908780098 2023-01-23 22:35:49.766502: step: 2924/1541, loss: 0.0027694746386259794 2023-01-23 22:35:50.615822: step: 2926/1541, loss: 5.347952537704259e-05 2023-01-23 22:35:51.472093: step: 2928/1541, loss: 0.19107010960578918 2023-01-23 22:35:52.353579: step: 2930/1541, loss: 1.9958613847848028e-05 2023-01-23 22:35:53.195256: step: 2932/1541, loss: 0.00023294240236282349 2023-01-23 22:35:54.193093: step: 2934/1541, loss: 0.18018077313899994 2023-01-23 22:35:55.040063: step: 2936/1541, loss: 2.5565923351678066e-05 2023-01-23 22:35:55.899440: step: 2938/1541, loss: 0.00041428732220083475 2023-01-23 22:35:56.781927: step: 2940/1541, loss: 1.0245447158813477 2023-01-23 22:35:57.637441: step: 2942/1541, loss: 0.04995384439826012 2023-01-23 22:35:58.545241: step: 2944/1541, loss: 0.018274245783686638 2023-01-23 22:35:59.400433: step: 2946/1541, loss: 2.2910040570423007e-05 2023-01-23 22:36:00.344577: step: 2948/1541, loss: 3.593652218114585e-05 2023-01-23 22:36:01.301101: step: 2950/1541, loss: 0.10651949048042297 2023-01-23 22:36:02.226488: step: 2952/1541, loss: 0.1755829155445099 2023-01-23 22:36:03.110739: step: 2954/1541, loss: 0.02308073826134205 2023-01-23 22:36:03.997379: step: 2956/1541, loss: 6.607353134313598e-05 2023-01-23 22:36:04.866354: step: 2958/1541, loss: 0.000657646800391376 2023-01-23 22:36:05.829188: step: 2960/1541, loss: 0.00017085540457628667 2023-01-23 22:36:06.723249: step: 2962/1541, loss: 0.3305518627166748 2023-01-23 22:36:07.574594: step: 2964/1541, loss: 0.41446247696876526 2023-01-23 22:36:08.478171: step: 2966/1541, loss: 0.0010422515915706754 2023-01-23 22:36:09.412077: step: 2968/1541, loss: 0.01491482462733984 2023-01-23 22:36:10.294278: step: 2970/1541, loss: 0.07263872027397156 2023-01-23 22:36:11.204506: step: 2972/1541, loss: 0.006973541807383299 2023-01-23 22:36:12.097911: step: 2974/1541, loss: 0.002491349121555686 2023-01-23 22:36:12.983627: step: 2976/1541, loss: 0.0003845544997602701 2023-01-23 22:36:13.868078: step: 2978/1541, loss: 0.0004873457073699683 2023-01-23 22:36:14.737982: step: 2980/1541, loss: 0.3432813584804535 2023-01-23 22:36:15.700515: step: 2982/1541, loss: 0.000626221823040396 2023-01-23 22:36:16.712901: step: 2984/1541, loss: 0.00024365111312363297 2023-01-23 22:36:17.599633: step: 2986/1541, loss: 0.00025731039931997657 2023-01-23 22:36:18.514327: step: 2988/1541, loss: 0.04340824857354164 2023-01-23 22:36:19.396271: step: 2990/1541, loss: 0.08352386951446533 2023-01-23 22:36:20.353320: step: 2992/1541, loss: 0.2752270996570587 2023-01-23 22:36:21.264959: step: 2994/1541, loss: 0.0037283366546034813 2023-01-23 22:36:22.129095: step: 2996/1541, loss: 0.3745122253894806 2023-01-23 22:36:23.005166: step: 2998/1541, loss: 0.18491266667842865 2023-01-23 22:36:23.921874: step: 3000/1541, loss: 0.007564629428088665 2023-01-23 22:36:24.831863: step: 3002/1541, loss: 0.0006670001894235611 2023-01-23 22:36:25.735874: step: 3004/1541, loss: 0.00026005174731835723 2023-01-23 22:36:26.649740: step: 3006/1541, loss: 0.03512326627969742 2023-01-23 22:36:27.563871: step: 3008/1541, loss: 0.0003692289174068719 2023-01-23 22:36:28.501782: step: 3010/1541, loss: 0.00042460227268747985 2023-01-23 22:36:29.397265: step: 3012/1541, loss: 0.0920020267367363 2023-01-23 22:36:30.263595: step: 3014/1541, loss: 0.00011826306581497192 2023-01-23 22:36:31.160246: step: 3016/1541, loss: 0.011825773864984512 2023-01-23 22:36:32.098333: step: 3018/1541, loss: 0.026677682995796204 2023-01-23 22:36:32.982807: step: 3020/1541, loss: 0.02262372337281704 2023-01-23 22:36:33.929393: step: 3022/1541, loss: 0.08146253228187561 2023-01-23 22:36:34.889860: step: 3024/1541, loss: 0.001182736363261938 2023-01-23 22:36:35.757304: step: 3026/1541, loss: 0.0007861251360736787 2023-01-23 22:36:36.631974: step: 3028/1541, loss: 0.0024823476560413837 2023-01-23 22:36:37.531379: step: 3030/1541, loss: 0.0003123211790807545 2023-01-23 22:36:38.523279: step: 3032/1541, loss: 0.00015389276086352766 2023-01-23 22:36:39.391864: step: 3034/1541, loss: 0.002299412153661251 2023-01-23 22:36:40.298620: step: 3036/1541, loss: 0.2178746461868286 2023-01-23 22:36:41.166358: step: 3038/1541, loss: 0.00017990991182159632 2023-01-23 22:36:42.026825: step: 3040/1541, loss: 0.0004690464411396533 2023-01-23 22:36:42.885753: step: 3042/1541, loss: 0.0035830545239150524 2023-01-23 22:36:43.844776: step: 3044/1541, loss: 0.15356777608394623 2023-01-23 22:36:44.750291: step: 3046/1541, loss: 0.05506231263279915 2023-01-23 22:36:45.652653: step: 3048/1541, loss: 8.986317698145285e-05 2023-01-23 22:36:46.562428: step: 3050/1541, loss: 0.002249170560389757 2023-01-23 22:36:47.456346: step: 3052/1541, loss: 0.0002295613376190886 2023-01-23 22:36:48.456843: step: 3054/1541, loss: 0.2238631397485733 2023-01-23 22:36:49.351417: step: 3056/1541, loss: 0.5505971312522888 2023-01-23 22:36:50.251126: step: 3058/1541, loss: 0.00018754236225504428 2023-01-23 22:36:51.121620: step: 3060/1541, loss: 0.00016444838547613472 2023-01-23 22:36:52.021311: step: 3062/1541, loss: 0.00020610817591659725 2023-01-23 22:36:52.890997: step: 3064/1541, loss: 0.0010295688407495618 2023-01-23 22:36:53.830282: step: 3066/1541, loss: 0.00026711635291576385 2023-01-23 22:36:54.697211: step: 3068/1541, loss: 0.035340528935194016 2023-01-23 22:36:55.580561: step: 3070/1541, loss: 0.007890244014561176 2023-01-23 22:36:56.454221: step: 3072/1541, loss: 0.00029292379622347653 2023-01-23 22:36:57.398256: step: 3074/1541, loss: 0.0016635264037176967 2023-01-23 22:36:58.301802: step: 3076/1541, loss: 0.6167073249816895 2023-01-23 22:36:59.264923: step: 3078/1541, loss: 0.2525503933429718 2023-01-23 22:37:00.173305: step: 3080/1541, loss: 0.1361447125673294 2023-01-23 22:37:01.045998: step: 3082/1541, loss: 0.1357831358909607 2023-01-23 22:37:01.682074: step: 3084/1541, loss: 9.774533964446164e-07 ================================================== Loss: 0.049 -------------------- Dev Chinese: {'event': {'p': 0.6308243727598566, 'r': 0.7030625832223701, 'f1': 0.6649874055415615}, 'argument': {'p': 0.6788322431410229, 'r': 0.6497771665863487, 'f1': 0.6639870052194498}, 'combined': 0.44154299591419316, 'epoch': 36} Test Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7303126454040291, 'r': 0.7023771890224542, 'f1': 0.7160725659671817}, 'combined': 0.46722180865402785, 'epoch': 36} Dev Korean: {'event': {'p': 0.5764705882352941, 'r': 0.7829560585885486, 'f1': 0.6640316205533597}, 'argument': {'p': 0.675707796918868, 'r': 0.6543798607176985, 'f1': 0.6648728322452419}, 'combined': 0.44149658425771005, 'epoch': 36} Test Korean: {'event': {'p': 0.6153479504289799, 'r': 0.7735170760934691, 'f1': 0.6854260684895142}, 'argument': {'p': 0.724640180117913, 'r': 0.7159831997510792, 'f1': 0.720285679212291}, 'combined': 0.49370258129178, 'epoch': 36} Dev Russian: {'event': {'p': 0.6552132701421801, 'r': 0.7363515312916112, 'f1': 0.6934169278996866}, 'argument': {'p': 0.6855655882974819, 'r': 0.6551303966424394, 'f1': 0.6700025372357077}, 'combined': 0.4645911010549798, 'epoch': 36} Test Russian: {'event': {'p': 0.6562009419152276, 'r': 0.7513481126423007, 'f1': 0.7005586592178771}, 'argument': {'p': 0.7269449017502853, 'r': 0.7050190950188584, 'f1': 0.7158141376695298}, 'combined': 0.5014697925349667, 'epoch': 36} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.7283362470862471, 'r': 0.6703271654599089, 'f1': 0.6981287529674627}, 'combined': 0.5250882928302283, 'epoch': 36} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5849631519274376, 'r': 0.4310254803675856, 'f1': 0.49633237133237135}, 'combined': 0.2892015392015392, 'epoch': 36} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.5348601934464002, 'r': 0.6219304574958143, 'f1': 0.5751184875767745}, 'combined': 0.2957752221823412, 'epoch': 36} New best chinese model... ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6308243727598566, 'r': 0.7030625832223701, 'f1': 0.6649874055415615}, 'argument': {'p': 0.6788322431410229, 'r': 0.6497771665863487, 'f1': 0.6639870052194498}, 'combined': 0.44154299591419316, 'epoch': 36} Test for Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7303126454040291, 'r': 0.7023771890224542, 'f1': 0.7160725659671817}, 'combined': 0.46722180865402785, 'epoch': 36} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.7283362470862471, 'r': 0.6703271654599089, 'f1': 0.6981287529674627}, 'combined': 0.5250882928302283, 'epoch': 36} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 37 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 22:42:31.007114: step: 2/1541, loss: 0.08168140053749084 2023-01-23 22:42:31.892358: step: 4/1541, loss: 0.001194793381728232 2023-01-23 22:42:32.762074: step: 6/1541, loss: 0.0009901244193315506 2023-01-23 22:42:33.686378: step: 8/1541, loss: 0.11503743380308151 2023-01-23 22:42:34.552023: step: 10/1541, loss: 0.00016735809913370758 2023-01-23 22:42:35.484200: step: 12/1541, loss: 0.000192052117199637 2023-01-23 22:42:36.388647: step: 14/1541, loss: 5.0492581067373976e-05 2023-01-23 22:42:37.254252: step: 16/1541, loss: 0.006606415845453739 2023-01-23 22:42:38.141935: step: 18/1541, loss: 1.3920906782150269 2023-01-23 22:42:39.025829: step: 20/1541, loss: 4.54056280432269e-05 2023-01-23 22:42:39.953512: step: 22/1541, loss: 0.014920373447239399 2023-01-23 22:42:40.837789: step: 24/1541, loss: 0.0007303135353140533 2023-01-23 22:42:41.707639: step: 26/1541, loss: 3.6418306990526617e-05 2023-01-23 22:42:42.624175: step: 28/1541, loss: 0.0014057373628020287 2023-01-23 22:42:43.520056: step: 30/1541, loss: 0.00011792883015004918 2023-01-23 22:42:44.417084: step: 32/1541, loss: 0.0013014437863603234 2023-01-23 22:42:45.309578: step: 34/1541, loss: 3.1393763492815197e-05 2023-01-23 22:42:46.178233: step: 36/1541, loss: 0.12027733772993088 2023-01-23 22:42:47.142151: step: 38/1541, loss: 5.380064521887107e-06 2023-01-23 22:42:48.027812: step: 40/1541, loss: 0.09912224858999252 2023-01-23 22:42:48.914923: step: 42/1541, loss: 0.0007113716565072536 2023-01-23 22:42:49.849324: step: 44/1541, loss: 0.0031762283761054277 2023-01-23 22:42:50.689329: step: 46/1541, loss: 0.05964170768857002 2023-01-23 22:42:51.630524: step: 48/1541, loss: 3.417168773012236e-05 2023-01-23 22:42:52.513447: step: 50/1541, loss: 0.00029148903558962047 2023-01-23 22:42:53.381336: step: 52/1541, loss: 5.419652006821707e-05 2023-01-23 22:42:54.356657: step: 54/1541, loss: 4.15970862377435e-05 2023-01-23 22:42:55.241023: step: 56/1541, loss: 5.98924707446713e-05 2023-01-23 22:42:56.158414: step: 58/1541, loss: -7.361918960668845e-06 2023-01-23 22:42:57.082931: step: 60/1541, loss: 3.643830859800801e-05 2023-01-23 22:42:57.938663: step: 62/1541, loss: 0.0001926576078403741 2023-01-23 22:42:58.819465: step: 64/1541, loss: 0.30263984203338623 2023-01-23 22:42:59.677539: step: 66/1541, loss: 6.448738713515922e-05 2023-01-23 22:43:00.537593: step: 68/1541, loss: 0.0021526406053453684 2023-01-23 22:43:01.450942: step: 70/1541, loss: 0.001787574146874249 2023-01-23 22:43:02.323502: step: 72/1541, loss: -6.645918801950756e-07 2023-01-23 22:43:03.279232: step: 74/1541, loss: 0.0002131374494638294 2023-01-23 22:43:04.163043: step: 76/1541, loss: 0.0018266488332301378 2023-01-23 22:43:05.077932: step: 78/1541, loss: 0.08881720155477524 2023-01-23 22:43:06.022962: step: 80/1541, loss: 0.00024178955936804414 2023-01-23 22:43:06.883462: step: 82/1541, loss: 0.0007672019419260323 2023-01-23 22:43:07.765970: step: 84/1541, loss: 0.00010528911661822349 2023-01-23 22:43:08.707708: step: 86/1541, loss: 2.5956333047361113e-05 2023-01-23 22:43:09.581313: step: 88/1541, loss: 0.011511590331792831 2023-01-23 22:43:10.488427: step: 90/1541, loss: 0.0003618115733843297 2023-01-23 22:43:11.366873: step: 92/1541, loss: 0.0012406677706167102 2023-01-23 22:43:12.284012: step: 94/1541, loss: 0.0006455305847339332 2023-01-23 22:43:13.177213: step: 96/1541, loss: 0.0020768402609974146 2023-01-23 22:43:14.040713: step: 98/1541, loss: 0.00022338918643072248 2023-01-23 22:43:14.978314: step: 100/1541, loss: 0.00025634976918809116 2023-01-23 22:43:15.893034: step: 102/1541, loss: 0.0004496673645917326 2023-01-23 22:43:16.815766: step: 104/1541, loss: 0.020564036443829536 2023-01-23 22:43:17.741795: step: 106/1541, loss: 0.22794972360134125 2023-01-23 22:43:18.713652: step: 108/1541, loss: 0.00011527500464580953 2023-01-23 22:43:19.628742: step: 110/1541, loss: 0.015824100002646446 2023-01-23 22:43:20.537559: step: 112/1541, loss: 0.0012326819123700261 2023-01-23 22:43:21.400398: step: 114/1541, loss: 0.10488276183605194 2023-01-23 22:43:22.301770: step: 116/1541, loss: 0.10469169169664383 2023-01-23 22:43:23.216852: step: 118/1541, loss: 0.0014902607072144747 2023-01-23 22:43:24.125538: step: 120/1541, loss: 0.007904847152531147 2023-01-23 22:43:25.033736: step: 122/1541, loss: 0.016526225954294205 2023-01-23 22:43:25.936847: step: 124/1541, loss: 0.06521851569414139 2023-01-23 22:43:26.814210: step: 126/1541, loss: 0.00015418528346344829 2023-01-23 22:43:27.835451: step: 128/1541, loss: 0.00023233768297359347 2023-01-23 22:43:28.741243: step: 130/1541, loss: 0.0022628854494541883 2023-01-23 22:43:29.630518: step: 132/1541, loss: 0.00028144617681391537 2023-01-23 22:43:30.492526: step: 134/1541, loss: 0.007729202974587679 2023-01-23 22:43:31.467015: step: 136/1541, loss: 0.00010795832349685952 2023-01-23 22:43:32.380912: step: 138/1541, loss: 0.4505132734775543 2023-01-23 22:43:33.287020: step: 140/1541, loss: 0.0028245688881725073 2023-01-23 22:43:34.143905: step: 142/1541, loss: 0.1369674801826477 2023-01-23 22:43:35.087889: step: 144/1541, loss: 0.003338209120556712 2023-01-23 22:43:35.974201: step: 146/1541, loss: 0.24867232143878937 2023-01-23 22:43:36.894247: step: 148/1541, loss: 0.000128948682686314 2023-01-23 22:43:37.800232: step: 150/1541, loss: 0.00176079326774925 2023-01-23 22:43:38.750097: step: 152/1541, loss: 0.010269770398736 2023-01-23 22:43:39.658957: step: 154/1541, loss: 0.05472565442323685 2023-01-23 22:43:40.499257: step: 156/1541, loss: 0.0038955712225288153 2023-01-23 22:43:41.394166: step: 158/1541, loss: 0.0069572520442306995 2023-01-23 22:43:42.290484: step: 160/1541, loss: 0.0018768302397802472 2023-01-23 22:43:43.174394: step: 162/1541, loss: 5.6916771427495405e-05 2023-01-23 22:43:44.203185: step: 164/1541, loss: 0.0002859812811948359 2023-01-23 22:43:45.074280: step: 166/1541, loss: 0.022341547533869743 2023-01-23 22:43:45.982172: step: 168/1541, loss: 0.003551251022145152 2023-01-23 22:43:46.883743: step: 170/1541, loss: 0.00307351048104465 2023-01-23 22:43:47.734046: step: 172/1541, loss: 0.000778078509029001 2023-01-23 22:43:48.680489: step: 174/1541, loss: 3.128948446828872e-05 2023-01-23 22:43:49.547709: step: 176/1541, loss: 0.06750454008579254 2023-01-23 22:43:50.428520: step: 178/1541, loss: 0.007625236641615629 2023-01-23 22:43:51.315077: step: 180/1541, loss: 0.0024341579992324114 2023-01-23 22:43:52.186947: step: 182/1541, loss: 0.3757275938987732 2023-01-23 22:43:53.067415: step: 184/1541, loss: 0.0966043546795845 2023-01-23 22:43:53.938714: step: 186/1541, loss: 2.842068533936981e-05 2023-01-23 22:43:54.817293: step: 188/1541, loss: 0.06170567870140076 2023-01-23 22:43:55.813812: step: 190/1541, loss: 5.1880382670788094e-05 2023-01-23 22:43:56.716183: step: 192/1541, loss: 0.0004736005503218621 2023-01-23 22:43:57.616517: step: 194/1541, loss: 0.0001163586784969084 2023-01-23 22:43:58.471655: step: 196/1541, loss: 0.00198946800082922 2023-01-23 22:43:59.338223: step: 198/1541, loss: 0.02103196084499359 2023-01-23 22:44:00.230890: step: 200/1541, loss: 0.00011924088175874203 2023-01-23 22:44:01.187403: step: 202/1541, loss: 0.06592424213886261 2023-01-23 22:44:02.095342: step: 204/1541, loss: 0.005648423917591572 2023-01-23 22:44:03.059467: step: 206/1541, loss: 8.282039198093116e-05 2023-01-23 22:44:03.945315: step: 208/1541, loss: 0.0006020794389769435 2023-01-23 22:44:04.837826: step: 210/1541, loss: 0.007766172289848328 2023-01-23 22:44:05.748851: step: 212/1541, loss: 0.0011827257694676518 2023-01-23 22:44:06.749810: step: 214/1541, loss: 0.0006591954152099788 2023-01-23 22:44:07.669751: step: 216/1541, loss: 0.0001036754110828042 2023-01-23 22:44:08.564046: step: 218/1541, loss: 0.00022590378648601472 2023-01-23 22:44:09.530559: step: 220/1541, loss: 0.006509061437100172 2023-01-23 22:44:10.446576: step: 222/1541, loss: 0.0018648378318175673 2023-01-23 22:44:11.321324: step: 224/1541, loss: 0.037798620760440826 2023-01-23 22:44:12.220180: step: 226/1541, loss: 0.14020706713199615 2023-01-23 22:44:13.111345: step: 228/1541, loss: 0.00072479248046875 2023-01-23 22:44:13.969774: step: 230/1541, loss: 0.0006080731982365251 2023-01-23 22:44:14.876136: step: 232/1541, loss: 0.0005146416369825602 2023-01-23 22:44:15.780379: step: 234/1541, loss: 0.0007033305009827018 2023-01-23 22:44:16.713881: step: 236/1541, loss: 0.00017127762839663774 2023-01-23 22:44:17.619482: step: 238/1541, loss: 0.07033156603574753 2023-01-23 22:44:18.464372: step: 240/1541, loss: 0.00019678428361658007 2023-01-23 22:44:19.302212: step: 242/1541, loss: 0.004482572432607412 2023-01-23 22:44:20.232346: step: 244/1541, loss: 0.12830820679664612 2023-01-23 22:44:21.117703: step: 246/1541, loss: 5.501933992491104e-05 2023-01-23 22:44:21.985227: step: 248/1541, loss: 0.0005105388117954135 2023-01-23 22:44:22.894410: step: 250/1541, loss: 0.01902465894818306 2023-01-23 22:44:23.774001: step: 252/1541, loss: 0.0003563103382475674 2023-01-23 22:44:24.667909: step: 254/1541, loss: 0.0001821388868847862 2023-01-23 22:44:25.558409: step: 256/1541, loss: -4.308174175093882e-05 2023-01-23 22:44:26.461771: step: 258/1541, loss: 0.11125985532999039 2023-01-23 22:44:27.404611: step: 260/1541, loss: 0.02236694097518921 2023-01-23 22:44:28.277499: step: 262/1541, loss: 0.0038387554232031107 2023-01-23 22:44:29.172396: step: 264/1541, loss: 0.014338012784719467 2023-01-23 22:44:30.109364: step: 266/1541, loss: 3.1293930078390986e-05 2023-01-23 22:44:31.048548: step: 268/1541, loss: 0.35714590549468994 2023-01-23 22:44:31.945037: step: 270/1541, loss: 0.0054482086561620235 2023-01-23 22:44:32.860969: step: 272/1541, loss: 0.008075331337749958 2023-01-23 22:44:33.799027: step: 274/1541, loss: 0.055497609078884125 2023-01-23 22:44:34.717305: step: 276/1541, loss: 0.30851009488105774 2023-01-23 22:44:35.685367: step: 278/1541, loss: 0.0009548364905640483 2023-01-23 22:44:36.527572: step: 280/1541, loss: 0.00010503033990971744 2023-01-23 22:44:37.405853: step: 282/1541, loss: 0.03636530041694641 2023-01-23 22:44:38.301381: step: 284/1541, loss: 0.00045711558777838945 2023-01-23 22:44:39.177973: step: 286/1541, loss: 0.00011253431148361415 2023-01-23 22:44:40.130512: step: 288/1541, loss: 0.003197666257619858 2023-01-23 22:44:41.047126: step: 290/1541, loss: 1.7934044080902822e-05 2023-01-23 22:44:41.948452: step: 292/1541, loss: 2.988576852658298e-05 2023-01-23 22:44:42.850328: step: 294/1541, loss: -3.89307715522591e-05 2023-01-23 22:44:43.746147: step: 296/1541, loss: 9.594311995897442e-05 2023-01-23 22:44:44.624563: step: 298/1541, loss: 0.004301580134779215 2023-01-23 22:44:45.528343: step: 300/1541, loss: 0.4043545424938202 2023-01-23 22:44:46.412560: step: 302/1541, loss: 0.01977480761706829 2023-01-23 22:44:47.297134: step: 304/1541, loss: 0.04350131005048752 2023-01-23 22:44:48.187438: step: 306/1541, loss: 0.1260351687669754 2023-01-23 22:44:49.144286: step: 308/1541, loss: 0.013863944448530674 2023-01-23 22:44:50.034378: step: 310/1541, loss: 2.6054678528453223e-05 2023-01-23 22:44:50.907113: step: 312/1541, loss: 0.00047945629921741784 2023-01-23 22:44:51.806991: step: 314/1541, loss: 0.014047305099666119 2023-01-23 22:44:52.710944: step: 316/1541, loss: 0.20554402470588684 2023-01-23 22:44:53.609584: step: 318/1541, loss: 0.00016919175686780363 2023-01-23 22:44:54.457691: step: 320/1541, loss: 0.00028828001813963056 2023-01-23 22:44:55.335831: step: 322/1541, loss: 0.003691795514896512 2023-01-23 22:44:56.210040: step: 324/1541, loss: 0.00010299764107912779 2023-01-23 22:44:57.096935: step: 326/1541, loss: 0.0009736900683492422 2023-01-23 22:44:57.958808: step: 328/1541, loss: 0.0035603116266429424 2023-01-23 22:44:58.913946: step: 330/1541, loss: 0.012086588889360428 2023-01-23 22:44:59.799856: step: 332/1541, loss: 0.07826609164476395 2023-01-23 22:45:00.725943: step: 334/1541, loss: 0.0006968684610910714 2023-01-23 22:45:01.624326: step: 336/1541, loss: 0.004692480433732271 2023-01-23 22:45:02.507038: step: 338/1541, loss: 0.08697344362735748 2023-01-23 22:45:03.385513: step: 340/1541, loss: 0.010892488062381744 2023-01-23 22:45:04.269072: step: 342/1541, loss: 0.0005903436103835702 2023-01-23 22:45:05.149983: step: 344/1541, loss: 0.00029451921000145376 2023-01-23 22:45:06.032851: step: 346/1541, loss: 8.18922053440474e-05 2023-01-23 22:45:06.930153: step: 348/1541, loss: 8.335908205481246e-05 2023-01-23 22:45:07.829495: step: 350/1541, loss: 0.20457753539085388 2023-01-23 22:45:08.721022: step: 352/1541, loss: 0.0005222500767558813 2023-01-23 22:45:09.672040: step: 354/1541, loss: 0.0004885006928816438 2023-01-23 22:45:10.582792: step: 356/1541, loss: 0.0007057934999465942 2023-01-23 22:45:11.457228: step: 358/1541, loss: 0.001329075894318521 2023-01-23 22:45:12.347991: step: 360/1541, loss: 0.0009844987653195858 2023-01-23 22:45:13.290325: step: 362/1541, loss: 0.07621056586503983 2023-01-23 22:45:14.215400: step: 364/1541, loss: 0.1757373958826065 2023-01-23 22:45:15.186895: step: 366/1541, loss: 0.00034975053858943284 2023-01-23 22:45:16.133678: step: 368/1541, loss: 0.04838482663035393 2023-01-23 22:45:17.037602: step: 370/1541, loss: 0.002292296849191189 2023-01-23 22:45:17.944617: step: 372/1541, loss: 8.241136674769223e-05 2023-01-23 22:45:18.848491: step: 374/1541, loss: 5.141483779880218e-05 2023-01-23 22:45:19.757740: step: 376/1541, loss: 0.3209441900253296 2023-01-23 22:45:20.771837: step: 378/1541, loss: 0.02771584503352642 2023-01-23 22:45:21.690792: step: 380/1541, loss: 0.009748903103172779 2023-01-23 22:45:22.592676: step: 382/1541, loss: 0.004586013965308666 2023-01-23 22:45:23.498048: step: 384/1541, loss: 0.00027741550002247095 2023-01-23 22:45:24.424431: step: 386/1541, loss: 0.0007409995305351913 2023-01-23 22:45:25.315340: step: 388/1541, loss: 0.08160095661878586 2023-01-23 22:45:26.181775: step: 390/1541, loss: 0.00018806755542755127 2023-01-23 22:45:27.092188: step: 392/1541, loss: 0.4392924904823303 2023-01-23 22:45:27.979882: step: 394/1541, loss: 0.00019299158884678036 2023-01-23 22:45:28.898996: step: 396/1541, loss: 0.0895976647734642 2023-01-23 22:45:29.830350: step: 398/1541, loss: 0.0016822076868265867 2023-01-23 22:45:30.777473: step: 400/1541, loss: -2.9561117116827518e-05 2023-01-23 22:45:31.656407: step: 402/1541, loss: 0.0007884249207563698 2023-01-23 22:45:32.632205: step: 404/1541, loss: 0.00018476306286174804 2023-01-23 22:45:33.580132: step: 406/1541, loss: 0.002072148257866502 2023-01-23 22:45:34.569326: step: 408/1541, loss: 0.006487892009317875 2023-01-23 22:45:35.497514: step: 410/1541, loss: 0.00012827299360651523 2023-01-23 22:45:36.419905: step: 412/1541, loss: 0.3340933918952942 2023-01-23 22:45:37.358068: step: 414/1541, loss: 0.005358163267374039 2023-01-23 22:45:38.391475: step: 416/1541, loss: 0.04188626632094383 2023-01-23 22:45:39.307845: step: 418/1541, loss: 9.514148405287415e-05 2023-01-23 22:45:40.237720: step: 420/1541, loss: 0.04000239446759224 2023-01-23 22:45:41.144679: step: 422/1541, loss: 0.004256249405443668 2023-01-23 22:45:42.014324: step: 424/1541, loss: 0.0003414571692701429 2023-01-23 22:45:42.914002: step: 426/1541, loss: 0.2011704444885254 2023-01-23 22:45:43.816926: step: 428/1541, loss: 0.0017388828564435244 2023-01-23 22:45:44.645267: step: 430/1541, loss: 8.779373456491157e-05 2023-01-23 22:45:45.575365: step: 432/1541, loss: 0.00010314306564396247 2023-01-23 22:45:46.463795: step: 434/1541, loss: 0.00020876413327641785 2023-01-23 22:45:47.317515: step: 436/1541, loss: 1.6786654668976553e-05 2023-01-23 22:45:48.182938: step: 438/1541, loss: 2.3426600819220766e-05 2023-01-23 22:45:49.098476: step: 440/1541, loss: 0.0013649652246385813 2023-01-23 22:45:49.987097: step: 442/1541, loss: 0.0009442612645216286 2023-01-23 22:45:50.895076: step: 444/1541, loss: 0.0019011644180864096 2023-01-23 22:45:51.721925: step: 446/1541, loss: 0.0001440911873942241 2023-01-23 22:45:52.614444: step: 448/1541, loss: 0.0027130721136927605 2023-01-23 22:45:53.519787: step: 450/1541, loss: 0.00015198886103462428 2023-01-23 22:45:54.436764: step: 452/1541, loss: 0.0009397302055731416 2023-01-23 22:45:55.353158: step: 454/1541, loss: 0.00011411586456233636 2023-01-23 22:45:56.305065: step: 456/1541, loss: 0.00024650065461173654 2023-01-23 22:45:57.233911: step: 458/1541, loss: 0.027856620028614998 2023-01-23 22:45:58.173377: step: 460/1541, loss: 0.0012183067156001925 2023-01-23 22:45:59.056797: step: 462/1541, loss: 0.00015182793140411377 2023-01-23 22:45:59.931442: step: 464/1541, loss: 0.005501324776560068 2023-01-23 22:46:00.799001: step: 466/1541, loss: 0.00013697572285309434 2023-01-23 22:46:01.733255: step: 468/1541, loss: 0.0003823688020929694 2023-01-23 22:46:02.655688: step: 470/1541, loss: 4.897007966064848e-05 2023-01-23 22:46:03.542632: step: 472/1541, loss: 0.00039585778722539544 2023-01-23 22:46:04.418687: step: 474/1541, loss: 4.355669079814106e-05 2023-01-23 22:46:05.304002: step: 476/1541, loss: 0.0002233367267763242 2023-01-23 22:46:06.234639: step: 478/1541, loss: 0.45843371748924255 2023-01-23 22:46:07.164794: step: 480/1541, loss: 3.0761955258640228e-06 2023-01-23 22:46:08.114978: step: 482/1541, loss: 0.0002748338447418064 2023-01-23 22:46:09.055743: step: 484/1541, loss: 0.0009854797972366214 2023-01-23 22:46:09.915801: step: 486/1541, loss: 0.0002533065271563828 2023-01-23 22:46:10.790369: step: 488/1541, loss: 0.0014441371895372868 2023-01-23 22:46:11.672506: step: 490/1541, loss: 0.007706669624894857 2023-01-23 22:46:12.600651: step: 492/1541, loss: 0.0032816517632454634 2023-01-23 22:46:13.505726: step: 494/1541, loss: 0.0005735227605327964 2023-01-23 22:46:14.377303: step: 496/1541, loss: 0.023748410865664482 2023-01-23 22:46:15.361532: step: 498/1541, loss: 0.0008645261405035853 2023-01-23 22:46:16.307756: step: 500/1541, loss: 2.7685346140060574e-05 2023-01-23 22:46:17.215895: step: 502/1541, loss: 0.00017822532390709966 2023-01-23 22:46:18.091375: step: 504/1541, loss: 0.004964139778167009 2023-01-23 22:46:18.950964: step: 506/1541, loss: 0.00012035814870614558 2023-01-23 22:46:19.854060: step: 508/1541, loss: 0.004014940001070499 2023-01-23 22:46:20.760845: step: 510/1541, loss: 0.06859414279460907 2023-01-23 22:46:21.648763: step: 512/1541, loss: 0.004546003881841898 2023-01-23 22:46:22.569339: step: 514/1541, loss: 2.3827125914976932e-05 2023-01-23 22:46:23.489800: step: 516/1541, loss: 0.014944463036954403 2023-01-23 22:46:24.383419: step: 518/1541, loss: 0.00020790868438780308 2023-01-23 22:46:25.287318: step: 520/1541, loss: 0.00013193605991546065 2023-01-23 22:46:26.161926: step: 522/1541, loss: 0.0001445770321879536 2023-01-23 22:46:27.047659: step: 524/1541, loss: 0.05132763087749481 2023-01-23 22:46:27.965104: step: 526/1541, loss: 0.0001121904788305983 2023-01-23 22:46:28.840538: step: 528/1541, loss: 0.00022333359811455011 2023-01-23 22:46:29.715181: step: 530/1541, loss: 4.179631650913507e-05 2023-01-23 22:46:30.620021: step: 532/1541, loss: 0.0026439994107931852 2023-01-23 22:46:31.555499: step: 534/1541, loss: 0.255734920501709 2023-01-23 22:46:32.439102: step: 536/1541, loss: 0.03456607460975647 2023-01-23 22:46:33.374494: step: 538/1541, loss: 9.207180482917465e-06 2023-01-23 22:46:34.249955: step: 540/1541, loss: -2.123915692209266e-07 2023-01-23 22:46:35.159530: step: 542/1541, loss: 0.001040914561599493 2023-01-23 22:46:36.108830: step: 544/1541, loss: 2.7171376132173464e-05 2023-01-23 22:46:36.988306: step: 546/1541, loss: 0.11919441819190979 2023-01-23 22:46:37.869244: step: 548/1541, loss: 0.004822550341486931 2023-01-23 22:46:38.753326: step: 550/1541, loss: 0.010364077985286713 2023-01-23 22:46:39.692326: step: 552/1541, loss: 9.653965389588848e-05 2023-01-23 22:46:40.610856: step: 554/1541, loss: 0.00035032874438911676 2023-01-23 22:46:41.490550: step: 556/1541, loss: 4.286095281713642e-05 2023-01-23 22:46:42.374868: step: 558/1541, loss: 0.00011752590944524854 2023-01-23 22:46:43.240461: step: 560/1541, loss: 0.578635036945343 2023-01-23 22:46:44.143798: step: 562/1541, loss: 4.9152085921377875e-06 2023-01-23 22:46:45.033281: step: 564/1541, loss: 0.04357423633337021 2023-01-23 22:46:45.980272: step: 566/1541, loss: 0.0021759155206382275 2023-01-23 22:46:46.880658: step: 568/1541, loss: 0.01689044199883938 2023-01-23 22:46:47.776764: step: 570/1541, loss: 2.6869298380916007e-05 2023-01-23 22:46:48.680427: step: 572/1541, loss: 0.0019013647688552737 2023-01-23 22:46:49.611227: step: 574/1541, loss: 0.396761417388916 2023-01-23 22:46:50.535907: step: 576/1541, loss: 2.586606979370117 2023-01-23 22:46:51.385608: step: 578/1541, loss: 0.00047380675096064806 2023-01-23 22:46:52.311072: step: 580/1541, loss: 3.671745434985496e-05 2023-01-23 22:46:53.260782: step: 582/1541, loss: 0.034529805183410645 2023-01-23 22:46:54.138870: step: 584/1541, loss: 0.01008809544146061 2023-01-23 22:46:55.014129: step: 586/1541, loss: 3.6273002478992566e-05 2023-01-23 22:46:55.975628: step: 588/1541, loss: -7.415813456645992e-07 2023-01-23 22:46:56.879562: step: 590/1541, loss: 0.975903332233429 2023-01-23 22:46:57.843305: step: 592/1541, loss: 0.0008148399647325277 2023-01-23 22:46:58.771308: step: 594/1541, loss: 0.00025889286189340055 2023-01-23 22:46:59.688204: step: 596/1541, loss: 0.033907800912857056 2023-01-23 22:47:00.573142: step: 598/1541, loss: 8.534728112863377e-05 2023-01-23 22:47:01.435405: step: 600/1541, loss: 0.02217808924615383 2023-01-23 22:47:02.323261: step: 602/1541, loss: 0.0014681634493172169 2023-01-23 22:47:03.206112: step: 604/1541, loss: 0.00015595355944242328 2023-01-23 22:47:04.101811: step: 606/1541, loss: 3.524373096297495e-05 2023-01-23 22:47:05.029228: step: 608/1541, loss: 0.0723349079489708 2023-01-23 22:47:05.913851: step: 610/1541, loss: -1.859665098891128e-05 2023-01-23 22:47:06.795454: step: 612/1541, loss: 6.041348024155013e-05 2023-01-23 22:47:07.696574: step: 614/1541, loss: 0.034073591232299805 2023-01-23 22:47:08.593028: step: 616/1541, loss: 0.0862564668059349 2023-01-23 22:47:09.468660: step: 618/1541, loss: 0.03219207748770714 2023-01-23 22:47:10.337294: step: 620/1541, loss: 0.0024414961226284504 2023-01-23 22:47:11.205753: step: 622/1541, loss: 0.00015517266001552343 2023-01-23 22:47:12.213927: step: 624/1541, loss: 0.0011658868752419949 2023-01-23 22:47:13.115392: step: 626/1541, loss: 2.3919608793221414e-05 2023-01-23 22:47:14.032835: step: 628/1541, loss: 0.0014366238610818982 2023-01-23 22:47:14.914986: step: 630/1541, loss: 0.0005863647093065083 2023-01-23 22:47:15.800983: step: 632/1541, loss: 6.143661448732018e-05 2023-01-23 22:47:16.726898: step: 634/1541, loss: 5.928414793743286e-06 2023-01-23 22:47:17.641072: step: 636/1541, loss: 0.0003252141468692571 2023-01-23 22:47:18.505663: step: 638/1541, loss: 0.0118066081777215 2023-01-23 22:47:19.387888: step: 640/1541, loss: 0.07607119530439377 2023-01-23 22:47:20.281849: step: 642/1541, loss: 0.003274309914559126 2023-01-23 22:47:21.202880: step: 644/1541, loss: 0.0001541979581816122 2023-01-23 22:47:22.109294: step: 646/1541, loss: 0.00047173709026537836 2023-01-23 22:47:22.962822: step: 648/1541, loss: 5.739181870012544e-05 2023-01-23 22:47:23.880916: step: 650/1541, loss: 0.00023805077944416553 2023-01-23 22:47:24.781090: step: 652/1541, loss: 0.033018626272678375 2023-01-23 22:47:25.684179: step: 654/1541, loss: 0.000581376429181546 2023-01-23 22:47:26.565902: step: 656/1541, loss: -4.0664858715899754e-06 2023-01-23 22:47:27.462801: step: 658/1541, loss: -1.670344136073254e-05 2023-01-23 22:47:28.331152: step: 660/1541, loss: 0.0004337879363447428 2023-01-23 22:47:29.230495: step: 662/1541, loss: 0.0001864242076408118 2023-01-23 22:47:30.122596: step: 664/1541, loss: 0.00501672737300396 2023-01-23 22:47:31.037710: step: 666/1541, loss: 0.00018220227502752095 2023-01-23 22:47:31.949217: step: 668/1541, loss: 0.003972590435296297 2023-01-23 22:47:32.820875: step: 670/1541, loss: 0.00021446889149956405 2023-01-23 22:47:33.700098: step: 672/1541, loss: 0.08070871233940125 2023-01-23 22:47:34.627073: step: 674/1541, loss: 6.959608435863629e-05 2023-01-23 22:47:35.479498: step: 676/1541, loss: 0.00026106880977749825 2023-01-23 22:47:36.367059: step: 678/1541, loss: 0.026010597124695778 2023-01-23 22:47:37.278133: step: 680/1541, loss: 0.0023612240329384804 2023-01-23 22:47:38.164765: step: 682/1541, loss: 0.34117159247398376 2023-01-23 22:47:39.065311: step: 684/1541, loss: 0.07593011856079102 2023-01-23 22:47:40.038933: step: 686/1541, loss: 0.012839593924582005 2023-01-23 22:47:40.969817: step: 688/1541, loss: 1.2307381439313758e-05 2023-01-23 22:47:41.849967: step: 690/1541, loss: 0.0001594651839695871 2023-01-23 22:47:42.749244: step: 692/1541, loss: 0.040187813341617584 2023-01-23 22:47:43.665645: step: 694/1541, loss: 0.003448895178735256 2023-01-23 22:47:44.560315: step: 696/1541, loss: 0.0013736624969169497 2023-01-23 22:47:45.468878: step: 698/1541, loss: 0.0016687549650669098 2023-01-23 22:47:46.377692: step: 700/1541, loss: 0.00010562533134361729 2023-01-23 22:47:47.283220: step: 702/1541, loss: 0.0756981149315834 2023-01-23 22:47:48.186317: step: 704/1541, loss: 2.5592371457605623e-05 2023-01-23 22:47:49.076107: step: 706/1541, loss: 0.0682571530342102 2023-01-23 22:47:49.972298: step: 708/1541, loss: 0.0001669007760938257 2023-01-23 22:47:50.925301: step: 710/1541, loss: 0.00018019787967205048 2023-01-23 22:47:51.851869: step: 712/1541, loss: 0.05905849486589432 2023-01-23 22:47:52.746877: step: 714/1541, loss: 0.3238048553466797 2023-01-23 22:47:53.658667: step: 716/1541, loss: 0.0003667272685561329 2023-01-23 22:47:54.602708: step: 718/1541, loss: 0.26922550797462463 2023-01-23 22:47:55.525755: step: 720/1541, loss: 7.941073272377253e-05 2023-01-23 22:47:56.368607: step: 722/1541, loss: 0.06024238467216492 2023-01-23 22:47:57.287476: step: 724/1541, loss: 6.008148557157256e-05 2023-01-23 22:47:58.233846: step: 726/1541, loss: 0.012514444999396801 2023-01-23 22:47:59.200851: step: 728/1541, loss: 1.409214735031128 2023-01-23 22:48:00.130987: step: 730/1541, loss: 0.16534945368766785 2023-01-23 22:48:01.073844: step: 732/1541, loss: 1.843075006036088e-05 2023-01-23 22:48:01.967697: step: 734/1541, loss: -3.509335147100501e-05 2023-01-23 22:48:02.883582: step: 736/1541, loss: -3.070648017455824e-05 2023-01-23 22:48:03.768623: step: 738/1541, loss: 0.0009950889507308602 2023-01-23 22:48:04.629764: step: 740/1541, loss: 0.17936107516288757 2023-01-23 22:48:05.517061: step: 742/1541, loss: 0.08760679513216019 2023-01-23 22:48:06.400085: step: 744/1541, loss: 9.64189093792811e-05 2023-01-23 22:48:07.308845: step: 746/1541, loss: 0.00012144247011747211 2023-01-23 22:48:08.254894: step: 748/1541, loss: 0.0882793515920639 2023-01-23 22:48:09.201464: step: 750/1541, loss: 0.00035569569445215166 2023-01-23 22:48:10.061703: step: 752/1541, loss: -3.162473512929864e-05 2023-01-23 22:48:10.974332: step: 754/1541, loss: 0.0009397334652021527 2023-01-23 22:48:11.858685: step: 756/1541, loss: -2.7368278097128496e-05 2023-01-23 22:48:12.849393: step: 758/1541, loss: 0.00011694998102029786 2023-01-23 22:48:13.844424: step: 760/1541, loss: 0.1336497813463211 2023-01-23 22:48:14.740771: step: 762/1541, loss: 0.0004211466293781996 2023-01-23 22:48:15.635411: step: 764/1541, loss: 0.006579115986824036 2023-01-23 22:48:16.568814: step: 766/1541, loss: 0.034249886870384216 2023-01-23 22:48:17.468783: step: 768/1541, loss: 0.15750820934772491 2023-01-23 22:48:18.389927: step: 770/1541, loss: 0.011604461818933487 2023-01-23 22:48:19.253388: step: 772/1541, loss: 0.027384568005800247 2023-01-23 22:48:20.155685: step: 774/1541, loss: 2.1159648895263672e-05 2023-01-23 22:48:21.014562: step: 776/1541, loss: 1.67990419868147e-05 2023-01-23 22:48:21.889705: step: 778/1541, loss: 6.917720020283014e-05 2023-01-23 22:48:22.774490: step: 780/1541, loss: 0.8024245500564575 2023-01-23 22:48:23.676186: step: 782/1541, loss: 4.365427594166249e-05 2023-01-23 22:48:24.603736: step: 784/1541, loss: 0.04606278985738754 2023-01-23 22:48:25.498632: step: 786/1541, loss: 0.001992426346987486 2023-01-23 22:48:26.373764: step: 788/1541, loss: 0.05492207407951355 2023-01-23 22:48:27.252908: step: 790/1541, loss: 0.014604221098124981 2023-01-23 22:48:28.143712: step: 792/1541, loss: 1.8960103261633776e-05 2023-01-23 22:48:29.035145: step: 794/1541, loss: 6.895099068060517e-05 2023-01-23 22:48:29.945567: step: 796/1541, loss: 0.002102539874613285 2023-01-23 22:48:30.866808: step: 798/1541, loss: 0.09724655002355576 2023-01-23 22:48:31.771315: step: 800/1541, loss: 0.003353794338181615 2023-01-23 22:48:32.666260: step: 802/1541, loss: 0.09174118936061859 2023-01-23 22:48:33.574306: step: 804/1541, loss: 0.26946377754211426 2023-01-23 22:48:34.483463: step: 806/1541, loss: 0.0018516555428504944 2023-01-23 22:48:35.438039: step: 808/1541, loss: -6.149314867798239e-05 2023-01-23 22:48:36.402929: step: 810/1541, loss: 0.0001833550340961665 2023-01-23 22:48:37.354917: step: 812/1541, loss: 1.6629684296276537e-06 2023-01-23 22:48:38.253106: step: 814/1541, loss: 4.3372681830078363e-05 2023-01-23 22:48:39.140026: step: 816/1541, loss: 8.015364437596872e-05 2023-01-23 22:48:40.049701: step: 818/1541, loss: 2.449403291393537e-05 2023-01-23 22:48:40.958814: step: 820/1541, loss: 8.317053288919851e-05 2023-01-23 22:48:41.898197: step: 822/1541, loss: 5.960017369943671e-05 2023-01-23 22:48:42.776790: step: 824/1541, loss: 0.013566854409873486 2023-01-23 22:48:43.706849: step: 826/1541, loss: 1.7508085647932603e-08 2023-01-23 22:48:44.620781: step: 828/1541, loss: 8.319354674313217e-05 2023-01-23 22:48:45.504008: step: 830/1541, loss: 2.0902163669234142e-05 2023-01-23 22:48:46.426312: step: 832/1541, loss: 0.0006566749070771039 2023-01-23 22:48:47.342097: step: 834/1541, loss: 0.09257408231496811 2023-01-23 22:48:48.289584: step: 836/1541, loss: 4.252970029483549e-05 2023-01-23 22:48:49.202115: step: 838/1541, loss: 8.558571425965056e-05 2023-01-23 22:48:50.087286: step: 840/1541, loss: 0.1223648265004158 2023-01-23 22:48:51.011368: step: 842/1541, loss: 0.007824542000889778 2023-01-23 22:48:51.906908: step: 844/1541, loss: 0.00012958126899320632 2023-01-23 22:48:52.839506: step: 846/1541, loss: 0.00027297978522256017 2023-01-23 22:48:53.759329: step: 848/1541, loss: 7.580853707622737e-05 2023-01-23 22:48:54.617293: step: 850/1541, loss: 0.0002566402836237103 2023-01-23 22:48:55.527424: step: 852/1541, loss: -1.3700128874916118e-05 2023-01-23 22:48:56.422427: step: 854/1541, loss: 0.009724346920847893 2023-01-23 22:48:57.319263: step: 856/1541, loss: 0.03331826254725456 2023-01-23 22:48:58.209114: step: 858/1541, loss: 0.0017184149473905563 2023-01-23 22:48:59.092758: step: 860/1541, loss: 0.0051023815758526325 2023-01-23 22:49:00.026553: step: 862/1541, loss: 0.003201413433998823 2023-01-23 22:49:00.884866: step: 864/1541, loss: 0.00012019260611850768 2023-01-23 22:49:01.769329: step: 866/1541, loss: 8.312013233080506e-05 2023-01-23 22:49:02.643097: step: 868/1541, loss: 0.0026142492424696684 2023-01-23 22:49:03.516799: step: 870/1541, loss: 0.017515506595373154 2023-01-23 22:49:04.412413: step: 872/1541, loss: 0.0080482866615057 2023-01-23 22:49:05.352771: step: 874/1541, loss: 0.04298063740134239 2023-01-23 22:49:06.237831: step: 876/1541, loss: 3.504552296362817e-05 2023-01-23 22:49:07.182448: step: 878/1541, loss: 4.5580494770547375e-05 2023-01-23 22:49:08.125508: step: 880/1541, loss: 0.4494125247001648 2023-01-23 22:49:08.986973: step: 882/1541, loss: 1.835214061429724e-05 2023-01-23 22:49:09.862084: step: 884/1541, loss: 0.004866561386734247 2023-01-23 22:49:10.734144: step: 886/1541, loss: 0.04327693209052086 2023-01-23 22:49:11.634535: step: 888/1541, loss: 0.0018704199464991689 2023-01-23 22:49:12.559837: step: 890/1541, loss: 0.20713746547698975 2023-01-23 22:49:13.495889: step: 892/1541, loss: 8.5463747382164e-05 2023-01-23 22:49:14.435066: step: 894/1541, loss: 2.676710209925659e-05 2023-01-23 22:49:15.317801: step: 896/1541, loss: 0.00017544474394526333 2023-01-23 22:49:16.230593: step: 898/1541, loss: 0.10675840824842453 2023-01-23 22:49:17.129252: step: 900/1541, loss: 0.07489978522062302 2023-01-23 22:49:17.996843: step: 902/1541, loss: -5.2319785027066246e-05 2023-01-23 22:49:18.900706: step: 904/1541, loss: 0.000526415416970849 2023-01-23 22:49:19.779747: step: 906/1541, loss: 0.08843496441841125 2023-01-23 22:49:20.686042: step: 908/1541, loss: 4.762931348523125e-05 2023-01-23 22:49:21.558548: step: 910/1541, loss: 0.005001133307814598 2023-01-23 22:49:22.458700: step: 912/1541, loss: 1.7530721379444003e-05 2023-01-23 22:49:23.371868: step: 914/1541, loss: 0.016336839646100998 2023-01-23 22:49:24.265005: step: 916/1541, loss: 0.8214920163154602 2023-01-23 22:49:25.160603: step: 918/1541, loss: 0.04787491634488106 2023-01-23 22:49:26.072347: step: 920/1541, loss: 0.0008906308794394135 2023-01-23 22:49:26.978541: step: 922/1541, loss: 0.0013385695638135076 2023-01-23 22:49:27.862399: step: 924/1541, loss: 0.004054566845297813 2023-01-23 22:49:28.717971: step: 926/1541, loss: 0.00031348157790489495 2023-01-23 22:49:29.655558: step: 928/1541, loss: 7.863774953875691e-05 2023-01-23 22:49:30.572146: step: 930/1541, loss: 2.3990139961242676 2023-01-23 22:49:31.454468: step: 932/1541, loss: 9.119496098719537e-05 2023-01-23 22:49:32.429132: step: 934/1541, loss: 0.13647860288619995 2023-01-23 22:49:33.345045: step: 936/1541, loss: 0.0013832658296450973 2023-01-23 22:49:34.262481: step: 938/1541, loss: 5.665546632371843e-05 2023-01-23 22:49:35.151577: step: 940/1541, loss: 0.5931114554405212 2023-01-23 22:49:36.056606: step: 942/1541, loss: 0.0008156347903423011 2023-01-23 22:49:36.952403: step: 944/1541, loss: 0.027632350102066994 2023-01-23 22:49:37.921388: step: 946/1541, loss: 0.5220401287078857 2023-01-23 22:49:38.871530: step: 948/1541, loss: 0.011741142719984055 2023-01-23 22:49:39.779737: step: 950/1541, loss: 0.010403333231806755 2023-01-23 22:49:40.674012: step: 952/1541, loss: 0.00858028419315815 2023-01-23 22:49:41.588259: step: 954/1541, loss: 0.00017014393233694136 2023-01-23 22:49:42.505009: step: 956/1541, loss: 0.07489054650068283 2023-01-23 22:49:43.418510: step: 958/1541, loss: 0.03969340771436691 2023-01-23 22:49:44.342960: step: 960/1541, loss: 0.00012149015674367547 2023-01-23 22:49:45.232696: step: 962/1541, loss: 0.18015868961811066 2023-01-23 22:49:46.125261: step: 964/1541, loss: 0.2935854196548462 2023-01-23 22:49:47.036016: step: 966/1541, loss: 0.00043868107604794204 2023-01-23 22:49:48.006091: step: 968/1541, loss: 0.19993546605110168 2023-01-23 22:49:48.883680: step: 970/1541, loss: 0.018049687147140503 2023-01-23 22:49:49.804802: step: 972/1541, loss: 0.09648364037275314 2023-01-23 22:49:50.714329: step: 974/1541, loss: 0.0019915562588721514 2023-01-23 22:49:51.608907: step: 976/1541, loss: 0.08228101581335068 2023-01-23 22:49:52.484919: step: 978/1541, loss: 0.0017848334973677993 2023-01-23 22:49:53.391489: step: 980/1541, loss: 0.0010374508565291762 2023-01-23 22:49:54.306120: step: 982/1541, loss: 0.0014674862613901496 2023-01-23 22:49:55.172792: step: 984/1541, loss: 0.0004173099878244102 2023-01-23 22:49:56.094094: step: 986/1541, loss: 0.0175547506660223 2023-01-23 22:49:56.987196: step: 988/1541, loss: 0.0067068240605294704 2023-01-23 22:49:57.896262: step: 990/1541, loss: 0.00032789845135994256 2023-01-23 22:49:58.781373: step: 992/1541, loss: 0.0003728712326847017 2023-01-23 22:49:59.695626: step: 994/1541, loss: 0.04255283623933792 2023-01-23 22:50:00.600497: step: 996/1541, loss: 0.14362956583499908 2023-01-23 22:50:01.530024: step: 998/1541, loss: 0.002036500023677945 2023-01-23 22:50:02.475552: step: 1000/1541, loss: 0.488483190536499 2023-01-23 22:50:03.372034: step: 1002/1541, loss: 5.2634124585893005e-05 2023-01-23 22:50:04.263991: step: 1004/1541, loss: 1.3063320693618152e-05 2023-01-23 22:50:05.131469: step: 1006/1541, loss: 0.0014003318501636386 2023-01-23 22:50:06.099336: step: 1008/1541, loss: 0.01857098378241062 2023-01-23 22:50:07.007735: step: 1010/1541, loss: 0.01645631156861782 2023-01-23 22:50:07.903272: step: 1012/1541, loss: 0.001948903314769268 2023-01-23 22:50:08.779125: step: 1014/1541, loss: 0.00018954367260448635 2023-01-23 22:50:09.662538: step: 1016/1541, loss: 0.0025142489466816187 2023-01-23 22:50:10.553411: step: 1018/1541, loss: 0.00037220490048639476 2023-01-23 22:50:11.479490: step: 1020/1541, loss: 1.6997915736283176e-05 2023-01-23 22:50:12.366054: step: 1022/1541, loss: 0.01852790266275406 2023-01-23 22:50:13.295519: step: 1024/1541, loss: 0.0002407752035651356 2023-01-23 22:50:14.265288: step: 1026/1541, loss: 0.00017249950906261802 2023-01-23 22:50:15.117883: step: 1028/1541, loss: 0.014318176545202732 2023-01-23 22:50:16.045255: step: 1030/1541, loss: 9.642081568017602e-05 2023-01-23 22:50:16.960297: step: 1032/1541, loss: 0.533453643321991 2023-01-23 22:50:17.861221: step: 1034/1541, loss: 0.04847683757543564 2023-01-23 22:50:18.777780: step: 1036/1541, loss: 0.0008220569579862058 2023-01-23 22:50:19.665721: step: 1038/1541, loss: 0.0001717299164738506 2023-01-23 22:50:20.552537: step: 1040/1541, loss: 0.001576335052959621 2023-01-23 22:50:21.505907: step: 1042/1541, loss: 0.008929832838475704 2023-01-23 22:50:22.399991: step: 1044/1541, loss: 0.0020402998197823763 2023-01-23 22:50:23.312667: step: 1046/1541, loss: 0.019651038572192192 2023-01-23 22:50:24.181222: step: 1048/1541, loss: 0.0003422250447329134 2023-01-23 22:50:25.035565: step: 1050/1541, loss: 0.002925478620454669 2023-01-23 22:50:25.916187: step: 1052/1541, loss: 0.0016888079699128866 2023-01-23 22:50:26.799041: step: 1054/1541, loss: 0.0004195371293462813 2023-01-23 22:50:27.663588: step: 1056/1541, loss: 0.23491494357585907 2023-01-23 22:50:28.630948: step: 1058/1541, loss: 0.1453927457332611 2023-01-23 22:50:29.529818: step: 1060/1541, loss: -7.187330993474461e-06 2023-01-23 22:50:30.415655: step: 1062/1541, loss: 2.180515366489999e-05 2023-01-23 22:50:31.322959: step: 1064/1541, loss: 0.003186196321621537 2023-01-23 22:50:32.275907: step: 1066/1541, loss: 4.6804027078906074e-05 2023-01-23 22:50:33.219662: step: 1068/1541, loss: 0.6957592964172363 2023-01-23 22:50:34.152399: step: 1070/1541, loss: 0.00038917147321626544 2023-01-23 22:50:35.009378: step: 1072/1541, loss: 0.10619102418422699 2023-01-23 22:50:35.896011: step: 1074/1541, loss: 1.9834740669466555e-05 2023-01-23 22:50:36.805161: step: 1076/1541, loss: 0.0023089954629540443 2023-01-23 22:50:37.794680: step: 1078/1541, loss: 0.001489268965087831 2023-01-23 22:50:38.723733: step: 1080/1541, loss: 0.010156037285923958 2023-01-23 22:50:39.697340: step: 1082/1541, loss: 0.0046770223416388035 2023-01-23 22:50:40.597359: step: 1084/1541, loss: 0.10388627648353577 2023-01-23 22:50:41.518022: step: 1086/1541, loss: 7.106699194991961e-05 2023-01-23 22:50:42.435853: step: 1088/1541, loss: 0.0031482905615121126 2023-01-23 22:50:43.302865: step: 1090/1541, loss: 0.009158563800156116 2023-01-23 22:50:44.202293: step: 1092/1541, loss: 8.983207226265222e-05 2023-01-23 22:50:45.076566: step: 1094/1541, loss: 4.228889156365767e-05 2023-01-23 22:50:46.051867: step: 1096/1541, loss: 0.0008339366759173572 2023-01-23 22:50:47.008692: step: 1098/1541, loss: 0.04653355851769447 2023-01-23 22:50:47.909157: step: 1100/1541, loss: 0.0055053806863725185 2023-01-23 22:50:48.822085: step: 1102/1541, loss: 0.004043568857014179 2023-01-23 22:50:49.700423: step: 1104/1541, loss: 1.8328428268432617e-05 2023-01-23 22:50:50.603754: step: 1106/1541, loss: 3.9916878449730575e-05 2023-01-23 22:50:51.602387: step: 1108/1541, loss: 0.004336139187216759 2023-01-23 22:50:52.500311: step: 1110/1541, loss: 0.03648204356431961 2023-01-23 22:50:53.423892: step: 1112/1541, loss: 0.0019410875393077731 2023-01-23 22:50:54.303205: step: 1114/1541, loss: 8.75223267939873e-05 2023-01-23 22:50:55.160722: step: 1116/1541, loss: 0.05574142932891846 2023-01-23 22:50:56.030802: step: 1118/1541, loss: 4.4951066229259595e-06 2023-01-23 22:50:56.950822: step: 1120/1541, loss: 0.00015927341883070767 2023-01-23 22:50:57.842135: step: 1122/1541, loss: 0.1787637621164322 2023-01-23 22:50:58.787494: step: 1124/1541, loss: 0.00021426111925393343 2023-01-23 22:50:59.673082: step: 1126/1541, loss: 0.060032255947589874 2023-01-23 22:51:00.589130: step: 1128/1541, loss: 0.013037320226430893 2023-01-23 22:51:01.528608: step: 1130/1541, loss: 0.2009333223104477 2023-01-23 22:51:02.439744: step: 1132/1541, loss: 0.00034775727544911206 2023-01-23 22:51:03.321492: step: 1134/1541, loss: 0.00036130668013356626 2023-01-23 22:51:04.196558: step: 1136/1541, loss: 0.0023986061569303274 2023-01-23 22:51:05.065825: step: 1138/1541, loss: 0.00011448920849943534 2023-01-23 22:51:05.954174: step: 1140/1541, loss: 0.0002576761180534959 2023-01-23 22:51:06.901875: step: 1142/1541, loss: 0.09240764379501343 2023-01-23 22:51:07.783420: step: 1144/1541, loss: 7.659196853637695e-06 2023-01-23 22:51:08.641402: step: 1146/1541, loss: 0.0015222590882331133 2023-01-23 22:51:09.598324: step: 1148/1541, loss: 0.30221298336982727 2023-01-23 22:51:10.500623: step: 1150/1541, loss: 0.7148653268814087 2023-01-23 22:51:11.389045: step: 1152/1541, loss: -2.9407449346763315e-06 2023-01-23 22:51:12.324680: step: 1154/1541, loss: 0.00789168756455183 2023-01-23 22:51:13.176014: step: 1156/1541, loss: 4.936318873660639e-05 2023-01-23 22:51:14.112072: step: 1158/1541, loss: 2.104700615745969e-05 2023-01-23 22:51:15.020736: step: 1160/1541, loss: 0.1437813639640808 2023-01-23 22:51:15.932053: step: 1162/1541, loss: 0.20115646719932556 2023-01-23 22:51:16.812488: step: 1164/1541, loss: 0.0005082730203866959 2023-01-23 22:51:17.710494: step: 1166/1541, loss: 0.024901917204260826 2023-01-23 22:51:18.619613: step: 1168/1541, loss: 1.498297478974564e-06 2023-01-23 22:51:19.524029: step: 1170/1541, loss: 1.3061623576504644e-05 2023-01-23 22:51:20.423611: step: 1172/1541, loss: 0.05222228541970253 2023-01-23 22:51:21.334180: step: 1174/1541, loss: 0.000835948099847883 2023-01-23 22:51:22.223603: step: 1176/1541, loss: -6.217198460944928e-06 2023-01-23 22:51:23.118638: step: 1178/1541, loss: 0.0014062548289075494 2023-01-23 22:51:24.041413: step: 1180/1541, loss: 0.02411504089832306 2023-01-23 22:51:24.962208: step: 1182/1541, loss: 0.0012352312915027142 2023-01-23 22:51:25.872797: step: 1184/1541, loss: 0.26074960827827454 2023-01-23 22:51:26.770065: step: 1186/1541, loss: 0.010466041974723339 2023-01-23 22:51:27.647407: step: 1188/1541, loss: 1.594622699485626e-05 2023-01-23 22:51:28.543360: step: 1190/1541, loss: 0.02000991441309452 2023-01-23 22:51:29.424591: step: 1192/1541, loss: 4.8896672524278983e-05 2023-01-23 22:51:30.310776: step: 1194/1541, loss: 0.00021525747433770448 2023-01-23 22:51:31.193748: step: 1196/1541, loss: 0.4037455916404724 2023-01-23 22:51:32.090455: step: 1198/1541, loss: 0.00012537701695691794 2023-01-23 22:51:32.991849: step: 1200/1541, loss: 0.007547901943325996 2023-01-23 22:51:33.853474: step: 1202/1541, loss: 0.0005485511501319706 2023-01-23 22:51:34.823937: step: 1204/1541, loss: 0.0005117511027492583 2023-01-23 22:51:35.730350: step: 1206/1541, loss: 0.008348723873496056 2023-01-23 22:51:36.675685: step: 1208/1541, loss: 1.8135306163458154e-05 2023-01-23 22:51:37.562268: step: 1210/1541, loss: 0.1246078684926033 2023-01-23 22:51:38.473066: step: 1212/1541, loss: 0.00016134169709403068 2023-01-23 22:51:39.405773: step: 1214/1541, loss: 0.00044364246423356235 2023-01-23 22:51:40.273273: step: 1216/1541, loss: 0.001479721162468195 2023-01-23 22:51:41.154653: step: 1218/1541, loss: 0.0006258988287299871 2023-01-23 22:51:42.029933: step: 1220/1541, loss: 0.1972605139017105 2023-01-23 22:51:42.943623: step: 1222/1541, loss: 0.000982494791969657 2023-01-23 22:51:43.866668: step: 1224/1541, loss: 0.006175162270665169 2023-01-23 22:51:44.789909: step: 1226/1541, loss: 0.0011961079435423017 2023-01-23 22:51:45.654792: step: 1228/1541, loss: 0.004695126786828041 2023-01-23 22:51:46.540261: step: 1230/1541, loss: 2.3718475858913735e-05 2023-01-23 22:51:47.501032: step: 1232/1541, loss: 0.07181821018457413 2023-01-23 22:51:48.379005: step: 1234/1541, loss: 0.03597007319331169 2023-01-23 22:51:49.321936: step: 1236/1541, loss: 0.0036694130394607782 2023-01-23 22:51:50.230742: step: 1238/1541, loss: 0.0011513317003846169 2023-01-23 22:51:51.142108: step: 1240/1541, loss: 0.007527205161750317 2023-01-23 22:51:52.018292: step: 1242/1541, loss: 4.081962106283754e-05 2023-01-23 22:51:52.921301: step: 1244/1541, loss: 0.0002725647937040776 2023-01-23 22:51:53.842634: step: 1246/1541, loss: 0.025843992829322815 2023-01-23 22:51:54.705134: step: 1248/1541, loss: 0.00022304290905594826 2023-01-23 22:51:55.582830: step: 1250/1541, loss: 3.3879368857014924e-05 2023-01-23 22:51:56.501008: step: 1252/1541, loss: 0.00017574131197761744 2023-01-23 22:51:57.375060: step: 1254/1541, loss: 0.004614542238414288 2023-01-23 22:51:58.244865: step: 1256/1541, loss: 3.463454959273804e-06 2023-01-23 22:51:59.166102: step: 1258/1541, loss: 0.0002694833092391491 2023-01-23 22:52:00.034793: step: 1260/1541, loss: 0.0009111433755606413 2023-01-23 22:52:00.916922: step: 1262/1541, loss: 0.0037529910914599895 2023-01-23 22:52:01.920177: step: 1264/1541, loss: 0.03137117251753807 2023-01-23 22:52:02.819528: step: 1266/1541, loss: 0.0034307490568608046 2023-01-23 22:52:03.765553: step: 1268/1541, loss: 0.6205224394798279 2023-01-23 22:52:04.658507: step: 1270/1541, loss: 0.1546649932861328 2023-01-23 22:52:05.567796: step: 1272/1541, loss: 2.4733939426369034e-05 2023-01-23 22:52:06.447898: step: 1274/1541, loss: 0.001985769486054778 2023-01-23 22:52:07.354372: step: 1276/1541, loss: 0.15551471710205078 2023-01-23 22:52:08.286264: step: 1278/1541, loss: 1.9279121261206456e-05 2023-01-23 22:52:09.241287: step: 1280/1541, loss: 0.48056140542030334 2023-01-23 22:52:10.118843: step: 1282/1541, loss: 0.0028301971033215523 2023-01-23 22:52:11.058013: step: 1284/1541, loss: 0.00041065822006203234 2023-01-23 22:52:11.983667: step: 1286/1541, loss: 0.003407874144613743 2023-01-23 22:52:12.850031: step: 1288/1541, loss: 0.050455011427402496 2023-01-23 22:52:13.771967: step: 1290/1541, loss: 0.19612811505794525 2023-01-23 22:52:14.674252: step: 1292/1541, loss: 3.308369196020067e-05 2023-01-23 22:52:15.550151: step: 1294/1541, loss: 0.00010226390440948308 2023-01-23 22:52:16.506878: step: 1296/1541, loss: 0.00019445120415184647 2023-01-23 22:52:17.378232: step: 1298/1541, loss: 2.9636166800628416e-05 2023-01-23 22:52:18.308981: step: 1300/1541, loss: 0.2248460203409195 2023-01-23 22:52:19.207893: step: 1302/1541, loss: 0.0007253614021465182 2023-01-23 22:52:20.096962: step: 1304/1541, loss: 8.17350810393691e-05 2023-01-23 22:52:20.992981: step: 1306/1541, loss: 6.403376755770296e-05 2023-01-23 22:52:21.901346: step: 1308/1541, loss: 0.0007820149185135961 2023-01-23 22:52:22.793983: step: 1310/1541, loss: 0.018441779538989067 2023-01-23 22:52:23.686544: step: 1312/1541, loss: 0.0012970343232154846 2023-01-23 22:52:24.585788: step: 1314/1541, loss: 0.017185153439641 2023-01-23 22:52:25.493743: step: 1316/1541, loss: 0.004872195888310671 2023-01-23 22:52:26.375427: step: 1318/1541, loss: -2.024770037678536e-05 2023-01-23 22:52:27.259531: step: 1320/1541, loss: 0.0013844706118106842 2023-01-23 22:52:28.161250: step: 1322/1541, loss: 0.00028503587236627936 2023-01-23 22:52:29.087094: step: 1324/1541, loss: 2.013493394770194e-05 2023-01-23 22:52:29.991784: step: 1326/1541, loss: 0.0002159310388378799 2023-01-23 22:52:30.868929: step: 1328/1541, loss: 0.00017419080541003495 2023-01-23 22:52:31.785465: step: 1330/1541, loss: 2.857544313883409e-05 2023-01-23 22:52:32.673602: step: 1332/1541, loss: 0.007487850729376078 2023-01-23 22:52:33.533517: step: 1334/1541, loss: 0.005996706895530224 2023-01-23 22:52:34.439686: step: 1336/1541, loss: 0.002182774944230914 2023-01-23 22:52:35.341733: step: 1338/1541, loss: 0.002326616086065769 2023-01-23 22:52:36.222450: step: 1340/1541, loss: 0.003148800227791071 2023-01-23 22:52:37.115845: step: 1342/1541, loss: 0.0023803547956049442 2023-01-23 22:52:37.978179: step: 1344/1541, loss: 0.002057902282103896 2023-01-23 22:52:38.866152: step: 1346/1541, loss: 0.0021741713862866163 2023-01-23 22:52:39.765027: step: 1348/1541, loss: 0.35124513506889343 2023-01-23 22:52:40.721856: step: 1350/1541, loss: 0.2621447741985321 2023-01-23 22:52:41.645902: step: 1352/1541, loss: 0.0022894353605806828 2023-01-23 22:52:42.526360: step: 1354/1541, loss: 0.00019569197320379317 2023-01-23 22:52:43.433163: step: 1356/1541, loss: 0.00023724828497506678 2023-01-23 22:52:44.311678: step: 1358/1541, loss: 0.0016171949682757258 2023-01-23 22:52:45.282352: step: 1360/1541, loss: 0.5592731833457947 2023-01-23 22:52:46.248175: step: 1362/1541, loss: 0.00032310932874679565 2023-01-23 22:52:47.112793: step: 1364/1541, loss: 8.474588685203344e-05 2023-01-23 22:52:47.951209: step: 1366/1541, loss: 1.7972770365304314e-05 2023-01-23 22:52:48.869282: step: 1368/1541, loss: 0.007285454776138067 2023-01-23 22:52:49.802709: step: 1370/1541, loss: 0.019745416939258575 2023-01-23 22:52:50.766322: step: 1372/1541, loss: 0.0002713486028369516 2023-01-23 22:52:51.649282: step: 1374/1541, loss: 0.0005898872623220086 2023-01-23 22:52:52.620456: step: 1376/1541, loss: 5.961855276837014e-05 2023-01-23 22:52:53.479513: step: 1378/1541, loss: 0.019508108496665955 2023-01-23 22:52:54.332137: step: 1380/1541, loss: 0.023620789870619774 2023-01-23 22:52:55.253852: step: 1382/1541, loss: 7.292622467502952e-05 2023-01-23 22:52:56.159907: step: 1384/1541, loss: 0.24296344816684723 2023-01-23 22:52:57.052683: step: 1386/1541, loss: 5.817472992930561e-05 2023-01-23 22:52:57.941852: step: 1388/1541, loss: 0.0040107956156134605 2023-01-23 22:52:58.891511: step: 1390/1541, loss: 7.780391570122447e-06 2023-01-23 22:52:59.768186: step: 1392/1541, loss: 0.0008164426544681191 2023-01-23 22:53:00.668977: step: 1394/1541, loss: 1.373589020658983e-05 2023-01-23 22:53:01.551974: step: 1396/1541, loss: 3.892034510499798e-05 2023-01-23 22:53:02.549125: step: 1398/1541, loss: 0.00016481579223182052 2023-01-23 22:53:03.389929: step: 1400/1541, loss: 0.0002957341494038701 2023-01-23 22:53:04.326868: step: 1402/1541, loss: 9.002785373013467e-05 2023-01-23 22:53:05.207925: step: 1404/1541, loss: 3.901128366123885e-05 2023-01-23 22:53:06.109321: step: 1406/1541, loss: 0.0073481360450387 2023-01-23 22:53:07.009180: step: 1408/1541, loss: 0.00022239683312363923 2023-01-23 22:53:07.913481: step: 1410/1541, loss: 0.09622956812381744 2023-01-23 22:53:08.810000: step: 1412/1541, loss: 9.231567673850805e-05 2023-01-23 22:53:09.718467: step: 1414/1541, loss: 0.16475994884967804 2023-01-23 22:53:10.606393: step: 1416/1541, loss: 0.0002447297447361052 2023-01-23 22:53:11.483303: step: 1418/1541, loss: 2.2134930986794643e-05 2023-01-23 22:53:12.387500: step: 1420/1541, loss: 0.0007592628244310617 2023-01-23 22:53:13.268909: step: 1422/1541, loss: 0.006271905731409788 2023-01-23 22:53:14.144379: step: 1424/1541, loss: 0.2909272611141205 2023-01-23 22:53:15.002579: step: 1426/1541, loss: 4.781782263307832e-05 2023-01-23 22:53:15.913133: step: 1428/1541, loss: -2.284348056491581e-06 2023-01-23 22:53:16.812421: step: 1430/1541, loss: 0.01789579540491104 2023-01-23 22:53:17.701039: step: 1432/1541, loss: 2.9757371521554887e-05 2023-01-23 22:53:18.632916: step: 1434/1541, loss: 0.00017384767124895006 2023-01-23 22:53:19.529544: step: 1436/1541, loss: 0.00010795892012538388 2023-01-23 22:53:20.365819: step: 1438/1541, loss: 0.011162349954247475 2023-01-23 22:53:21.258463: step: 1440/1541, loss: 0.03333889693021774 2023-01-23 22:53:22.172305: step: 1442/1541, loss: 0.13451021909713745 2023-01-23 22:53:23.115073: step: 1444/1541, loss: 0.0241409782320261 2023-01-23 22:53:24.051561: step: 1446/1541, loss: 7.630884283571504e-06 2023-01-23 22:53:24.928965: step: 1448/1541, loss: 3.281037061242387e-05 2023-01-23 22:53:25.848912: step: 1450/1541, loss: 8.677370715304278e-06 2023-01-23 22:53:26.757639: step: 1452/1541, loss: 0.08195573091506958 2023-01-23 22:53:27.649685: step: 1454/1541, loss: 0.0011718790046870708 2023-01-23 22:53:28.692455: step: 1456/1541, loss: 6.791092891944572e-05 2023-01-23 22:53:29.595014: step: 1458/1541, loss: 0.18358419835567474 2023-01-23 22:53:30.487944: step: 1460/1541, loss: 4.730224463855848e-05 2023-01-23 22:53:31.392517: step: 1462/1541, loss: 0.2746323347091675 2023-01-23 22:53:32.285846: step: 1464/1541, loss: 0.017048940062522888 2023-01-23 22:53:33.227057: step: 1466/1541, loss: 0.031143857166171074 2023-01-23 22:53:34.132173: step: 1468/1541, loss: 2.7465821403893642e-05 2023-01-23 22:53:35.012537: step: 1470/1541, loss: 0.013094332069158554 2023-01-23 22:53:36.020990: step: 1472/1541, loss: 6.134153954917565e-05 2023-01-23 22:53:36.896230: step: 1474/1541, loss: 2.443392986606341e-05 2023-01-23 22:53:37.819548: step: 1476/1541, loss: 5.324184530763887e-05 2023-01-23 22:53:38.691942: step: 1478/1541, loss: 6.475559348473325e-05 2023-01-23 22:53:39.608187: step: 1480/1541, loss: 0.0026235003024339676 2023-01-23 22:53:40.503470: step: 1482/1541, loss: 0.0005682029295712709 2023-01-23 22:53:41.396964: step: 1484/1541, loss: 0.00022856210125610232 2023-01-23 22:53:42.290483: step: 1486/1541, loss: 0.002282226225361228 2023-01-23 22:53:43.185076: step: 1488/1541, loss: 0.166422501206398 2023-01-23 22:53:44.041418: step: 1490/1541, loss: 0.00018033946980722249 2023-01-23 22:53:44.943543: step: 1492/1541, loss: 0.00010607913282001391 2023-01-23 22:53:45.875151: step: 1494/1541, loss: 0.06498648226261139 2023-01-23 22:53:46.789636: step: 1496/1541, loss: 0.07922925055027008 2023-01-23 22:53:47.649694: step: 1498/1541, loss: 1.0026661157608032 2023-01-23 22:53:48.553846: step: 1500/1541, loss: 0.00011597037519095466 2023-01-23 22:53:49.453969: step: 1502/1541, loss: 5.7983401347883046e-05 2023-01-23 22:53:50.361007: step: 1504/1541, loss: 0.07755699753761292 2023-01-23 22:53:51.331099: step: 1506/1541, loss: 0.31067124009132385 2023-01-23 22:53:52.178985: step: 1508/1541, loss: 2.6293593691661954e-05 2023-01-23 22:53:53.027978: step: 1510/1541, loss: 0.14560188353061676 2023-01-23 22:53:53.939168: step: 1512/1541, loss: 0.00031204320839606225 2023-01-23 22:53:54.789300: step: 1514/1541, loss: 0.004083097446709871 2023-01-23 22:53:55.713633: step: 1516/1541, loss: 0.012077512219548225 2023-01-23 22:53:56.593429: step: 1518/1541, loss: 0.000593186414334923 2023-01-23 22:53:57.470492: step: 1520/1541, loss: -1.2980152860109229e-05 2023-01-23 22:53:58.324050: step: 1522/1541, loss: 0.02674419991672039 2023-01-23 22:53:59.232814: step: 1524/1541, loss: 4.959166108164936e-05 2023-01-23 22:54:00.149028: step: 1526/1541, loss: 0.0008549471967853606 2023-01-23 22:54:01.045299: step: 1528/1541, loss: 0.00013962238153908402 2023-01-23 22:54:01.933177: step: 1530/1541, loss: 0.04703598469495773 2023-01-23 22:54:02.852864: step: 1532/1541, loss: 0.0004546433629002422 2023-01-23 22:54:03.765117: step: 1534/1541, loss: 0.1347368061542511 2023-01-23 22:54:04.686934: step: 1536/1541, loss: 0.0016568247228860855 2023-01-23 22:54:05.531814: step: 1538/1541, loss: 0.003923443146049976 2023-01-23 22:54:06.442286: step: 1540/1541, loss: 9.692609455669299e-05 2023-01-23 22:54:07.399444: step: 1542/1541, loss: 0.0002903986314777285 2023-01-23 22:54:08.308767: step: 1544/1541, loss: 0.0002971211215481162 2023-01-23 22:54:09.181037: step: 1546/1541, loss: 7.092502346495166e-05 2023-01-23 22:54:10.036474: step: 1548/1541, loss: 0.20206090807914734 2023-01-23 22:54:10.956012: step: 1550/1541, loss: 0.2844730317592621 2023-01-23 22:54:11.832097: step: 1552/1541, loss: 0.0007705906173214316 2023-01-23 22:54:12.773836: step: 1554/1541, loss: 0.06438218057155609 2023-01-23 22:54:13.688084: step: 1556/1541, loss: 0.00011827707203337923 2023-01-23 22:54:14.638621: step: 1558/1541, loss: 0.0006317154038697481 2023-01-23 22:54:15.493988: step: 1560/1541, loss: 0.0006443105521611869 2023-01-23 22:54:16.358277: step: 1562/1541, loss: 0.0003955847059842199 2023-01-23 22:54:17.225424: step: 1564/1541, loss: 0.00016682129353284836 2023-01-23 22:54:18.094659: step: 1566/1541, loss: 9.366869926452637e-05 2023-01-23 22:54:18.960205: step: 1568/1541, loss: 0.04043145477771759 2023-01-23 22:54:19.821757: step: 1570/1541, loss: 0.03685314580798149 2023-01-23 22:54:20.700318: step: 1572/1541, loss: 0.011167636141180992 2023-01-23 22:54:21.567814: step: 1574/1541, loss: 0.00013046563253737986 2023-01-23 22:54:22.505561: step: 1576/1541, loss: 0.0042348587885499 2023-01-23 22:54:23.388243: step: 1578/1541, loss: 2.3273232727660798e-05 2023-01-23 22:54:24.290966: step: 1580/1541, loss: 0.22903352975845337 2023-01-23 22:54:25.144056: step: 1582/1541, loss: 2.621211388031952e-05 2023-01-23 22:54:26.015498: step: 1584/1541, loss: 0.00016289233462885022 2023-01-23 22:54:26.926128: step: 1586/1541, loss: -3.0982769203546923e-06 2023-01-23 22:54:27.759017: step: 1588/1541, loss: 0.0017944441642612219 2023-01-23 22:54:28.618270: step: 1590/1541, loss: 0.00018099596491083503 2023-01-23 22:54:29.524563: step: 1592/1541, loss: 0.003054244676604867 2023-01-23 22:54:30.412760: step: 1594/1541, loss: 0.006941622123122215 2023-01-23 22:54:31.427187: step: 1596/1541, loss: 0.014987397938966751 2023-01-23 22:54:32.339857: step: 1598/1541, loss: 0.0008880712557584047 2023-01-23 22:54:33.183106: step: 1600/1541, loss: 3.8285528717096895e-05 2023-01-23 22:54:34.043808: step: 1602/1541, loss: 0.0005258937017060816 2023-01-23 22:54:34.922629: step: 1604/1541, loss: 0.00020688324002549052 2023-01-23 22:54:35.790840: step: 1606/1541, loss: 4.972914757672697e-05 2023-01-23 22:54:36.690449: step: 1608/1541, loss: 4.972159877070226e-05 2023-01-23 22:54:37.618209: step: 1610/1541, loss: 9.518166916677728e-05 2023-01-23 22:54:38.475027: step: 1612/1541, loss: 3.36125485773664e-05 2023-01-23 22:54:39.358752: step: 1614/1541, loss: 0.02306884527206421 2023-01-23 22:54:40.263031: step: 1616/1541, loss: 0.00039678215398453176 2023-01-23 22:54:41.275864: step: 1618/1541, loss: 3.590225969674066e-05 2023-01-23 22:54:42.240853: step: 1620/1541, loss: 0.00043759524123743176 2023-01-23 22:54:43.104538: step: 1622/1541, loss: 0.06111231446266174 2023-01-23 22:54:43.991212: step: 1624/1541, loss: 0.33503884077072144 2023-01-23 22:54:44.857606: step: 1626/1541, loss: 0.0006900353473611176 2023-01-23 22:54:45.738066: step: 1628/1541, loss: 0.036899566650390625 2023-01-23 22:54:46.640014: step: 1630/1541, loss: 0.00020065350690856576 2023-01-23 22:54:47.576534: step: 1632/1541, loss: 4.6168566768756136e-05 2023-01-23 22:54:48.441472: step: 1634/1541, loss: 0.053013309836387634 2023-01-23 22:54:49.339140: step: 1636/1541, loss: 2.529355879232753e-05 2023-01-23 22:54:50.250036: step: 1638/1541, loss: 0.12470782548189163 2023-01-23 22:54:51.107321: step: 1640/1541, loss: 0.16520756483078003 2023-01-23 22:54:52.083289: step: 1642/1541, loss: 0.425096720457077 2023-01-23 22:54:52.979430: step: 1644/1541, loss: 0.0029243764001876116 2023-01-23 22:54:53.884959: step: 1646/1541, loss: 0.05079081654548645 2023-01-23 22:54:54.815557: step: 1648/1541, loss: 0.17707495391368866 2023-01-23 22:54:55.690138: step: 1650/1541, loss: 0.08556895703077316 2023-01-23 22:54:56.654579: step: 1652/1541, loss: 4.686138709075749e-06 2023-01-23 22:54:57.564900: step: 1654/1541, loss: 4.6158336772350594e-05 2023-01-23 22:54:58.437754: step: 1656/1541, loss: 0.0024824123829603195 2023-01-23 22:54:59.314797: step: 1658/1541, loss: 0.001007464132271707 2023-01-23 22:55:00.190298: step: 1660/1541, loss: 0.08776132762432098 2023-01-23 22:55:01.099617: step: 1662/1541, loss: 0.008320892229676247 2023-01-23 22:55:01.999504: step: 1664/1541, loss: 2.069969013973605e-05 2023-01-23 22:55:02.979970: step: 1666/1541, loss: 0.00014874424959998578 2023-01-23 22:55:03.883502: step: 1668/1541, loss: 0.054839134216308594 2023-01-23 22:55:04.835699: step: 1670/1541, loss: 5.802703890367411e-05 2023-01-23 22:55:05.789467: step: 1672/1541, loss: 0.00045586004853248596 2023-01-23 22:55:06.670828: step: 1674/1541, loss: 0.02563159167766571 2023-01-23 22:55:07.614659: step: 1676/1541, loss: 0.0006943434709683061 2023-01-23 22:55:08.485123: step: 1678/1541, loss: 0.00035630897036753595 2023-01-23 22:55:09.360790: step: 1680/1541, loss: 0.02618674747645855 2023-01-23 22:55:10.232541: step: 1682/1541, loss: 0.1548122614622116 2023-01-23 22:55:11.120848: step: 1684/1541, loss: 0.00020928446610923856 2023-01-23 22:55:12.062503: step: 1686/1541, loss: 0.01237251702696085 2023-01-23 22:55:12.965119: step: 1688/1541, loss: 0.0024581083562225103 2023-01-23 22:55:13.847521: step: 1690/1541, loss: 0.00018160889158025384 2023-01-23 22:55:14.718621: step: 1692/1541, loss: 0.0072504268027842045 2023-01-23 22:55:15.609396: step: 1694/1541, loss: 0.0005275020957924426 2023-01-23 22:55:16.541261: step: 1696/1541, loss: 0.0008960723644122481 2023-01-23 22:55:17.413256: step: 1698/1541, loss: 0.4678599238395691 2023-01-23 22:55:18.337191: step: 1700/1541, loss: 8.020296809263527e-05 2023-01-23 22:55:19.213102: step: 1702/1541, loss: 0.07889038324356079 2023-01-23 22:55:20.143839: step: 1704/1541, loss: 0.36424508690834045 2023-01-23 22:55:20.995574: step: 1706/1541, loss: 0.0020480803214013577 2023-01-23 22:55:21.891224: step: 1708/1541, loss: 0.007347768638283014 2023-01-23 22:55:22.755667: step: 1710/1541, loss: 0.06222893297672272 2023-01-23 22:55:23.710673: step: 1712/1541, loss: 0.05061595141887665 2023-01-23 22:55:24.567652: step: 1714/1541, loss: 0.3137431740760803 2023-01-23 22:55:25.458671: step: 1716/1541, loss: 0.1513175070285797 2023-01-23 22:55:26.327740: step: 1718/1541, loss: 0.0007183600682765245 2023-01-23 22:55:27.202841: step: 1720/1541, loss: 0.010364050976932049 2023-01-23 22:55:28.123238: step: 1722/1541, loss: 0.0003919172449968755 2023-01-23 22:55:29.018483: step: 1724/1541, loss: 0.07982976734638214 2023-01-23 22:55:29.952808: step: 1726/1541, loss: 0.012017196044325829 2023-01-23 22:55:30.827191: step: 1728/1541, loss: 0.03388016298413277 2023-01-23 22:55:31.679051: step: 1730/1541, loss: 0.0016355231637135148 2023-01-23 22:55:32.642708: step: 1732/1541, loss: 0.0014181557344272733 2023-01-23 22:55:33.537991: step: 1734/1541, loss: 0.22249209880828857 2023-01-23 22:55:34.405827: step: 1736/1541, loss: 0.011589828878641129 2023-01-23 22:55:35.259094: step: 1738/1541, loss: 0.0005040823016315699 2023-01-23 22:55:36.118354: step: 1740/1541, loss: 0.003778439946472645 2023-01-23 22:55:37.054228: step: 1742/1541, loss: 0.003822455182671547 2023-01-23 22:55:37.986413: step: 1744/1541, loss: 0.004831945523619652 2023-01-23 22:55:38.926994: step: 1746/1541, loss: 0.16397546231746674 2023-01-23 22:55:39.808799: step: 1748/1541, loss: 0.20160500705242157 2023-01-23 22:55:40.683002: step: 1750/1541, loss: 7.744815229671076e-05 2023-01-23 22:55:41.539945: step: 1752/1541, loss: 0.0012470788788050413 2023-01-23 22:55:42.406473: step: 1754/1541, loss: 0.001414877362549305 2023-01-23 22:55:43.251205: step: 1756/1541, loss: 0.00021709279099013656 2023-01-23 22:55:44.113874: step: 1758/1541, loss: 0.0020109936594963074 2023-01-23 22:55:45.128571: step: 1760/1541, loss: 0.00039059098344296217 2023-01-23 22:55:46.026409: step: 1762/1541, loss: 0.05565520375967026 2023-01-23 22:55:46.896368: step: 1764/1541, loss: 0.06699026376008987 2023-01-23 22:55:47.779439: step: 1766/1541, loss: 0.0004210897604934871 2023-01-23 22:55:48.683657: step: 1768/1541, loss: 0.09505964070558548 2023-01-23 22:55:49.619116: step: 1770/1541, loss: 0.03580080717802048 2023-01-23 22:55:50.533083: step: 1772/1541, loss: 0.0002864986308850348 2023-01-23 22:55:51.433548: step: 1774/1541, loss: 0.0007754738908261061 2023-01-23 22:55:52.287407: step: 1776/1541, loss: 0.004057671409100294 2023-01-23 22:55:53.195573: step: 1778/1541, loss: 3.991714402218349e-05 2023-01-23 22:55:54.064133: step: 1780/1541, loss: 0.008423671126365662 2023-01-23 22:55:54.931377: step: 1782/1541, loss: 0.0001586971920914948 2023-01-23 22:55:55.898550: step: 1784/1541, loss: 0.0011145059252157807 2023-01-23 22:55:56.758667: step: 1786/1541, loss: 5.764700108557008e-05 2023-01-23 22:55:57.632055: step: 1788/1541, loss: 0.003940111957490444 2023-01-23 22:55:58.510924: step: 1790/1541, loss: 0.01683802157640457 2023-01-23 22:55:59.418827: step: 1792/1541, loss: 0.00014554026711266488 2023-01-23 22:56:00.306110: step: 1794/1541, loss: 0.015414105728268623 2023-01-23 22:56:01.219484: step: 1796/1541, loss: 0.5435144305229187 2023-01-23 22:56:02.134935: step: 1798/1541, loss: 0.0001550147426314652 2023-01-23 22:56:03.022679: step: 1800/1541, loss: 2.6148896722588688e-05 2023-01-23 22:56:03.914449: step: 1802/1541, loss: 0.0002878043451346457 2023-01-23 22:56:04.798082: step: 1804/1541, loss: 0.03822506219148636 2023-01-23 22:56:05.657701: step: 1806/1541, loss: 7.791308598825708e-05 2023-01-23 22:56:06.524860: step: 1808/1541, loss: 2.8208227377035655e-05 2023-01-23 22:56:07.395611: step: 1810/1541, loss: 0.0002182191237807274 2023-01-23 22:56:08.253243: step: 1812/1541, loss: 0.0004238859110046178 2023-01-23 22:56:09.173302: step: 1814/1541, loss: 2.5372706659254618e-05 2023-01-23 22:56:10.101447: step: 1816/1541, loss: 0.2090653032064438 2023-01-23 22:56:10.993342: step: 1818/1541, loss: 0.004556646104902029 2023-01-23 22:56:11.930941: step: 1820/1541, loss: 0.020152412354946136 2023-01-23 22:56:12.861298: step: 1822/1541, loss: 0.00017538535757921636 2023-01-23 22:56:13.716010: step: 1824/1541, loss: 0.00022018104209564626 2023-01-23 22:56:14.588063: step: 1826/1541, loss: 0.00017442247190047055 2023-01-23 22:56:15.495291: step: 1828/1541, loss: 6.417482654796913e-05 2023-01-23 22:56:16.387558: step: 1830/1541, loss: 0.00016132592281792313 2023-01-23 22:56:17.293564: step: 1832/1541, loss: 0.00013725219469051808 2023-01-23 22:56:18.166120: step: 1834/1541, loss: 0.5720453262329102 2023-01-23 22:56:19.000997: step: 1836/1541, loss: 0.03577864170074463 2023-01-23 22:56:19.972844: step: 1838/1541, loss: 0.0001258561824215576 2023-01-23 22:56:20.875141: step: 1840/1541, loss: 0.002182826166972518 2023-01-23 22:56:21.755196: step: 1842/1541, loss: 0.0002450312895234674 2023-01-23 22:56:22.689300: step: 1844/1541, loss: 2.459175266267266e-05 2023-01-23 22:56:23.567352: step: 1846/1541, loss: 0.00017900977400131524 2023-01-23 22:56:24.450379: step: 1848/1541, loss: 5.175704063731246e-05 2023-01-23 22:56:25.454496: step: 1850/1541, loss: 1.5346055030822754 2023-01-23 22:56:26.332610: step: 1852/1541, loss: 0.0031200821977108717 2023-01-23 22:56:27.280220: step: 1854/1541, loss: 0.009462342597544193 2023-01-23 22:56:28.185857: step: 1856/1541, loss: 0.172743022441864 2023-01-23 22:56:29.104263: step: 1858/1541, loss: 0.0005029026651754975 2023-01-23 22:56:29.983264: step: 1860/1541, loss: 0.00915640126913786 2023-01-23 22:56:30.874181: step: 1862/1541, loss: 0.0004950300208292902 2023-01-23 22:56:31.731049: step: 1864/1541, loss: 0.0011653141118586063 2023-01-23 22:56:32.625835: step: 1866/1541, loss: 0.00038695696275681257 2023-01-23 22:56:33.604078: step: 1868/1541, loss: 0.15288542211055756 2023-01-23 22:56:34.527139: step: 1870/1541, loss: 0.07259257137775421 2023-01-23 22:56:35.413573: step: 1872/1541, loss: 0.00016754059470258653 2023-01-23 22:56:36.291003: step: 1874/1541, loss: 0.0008303644717670977 2023-01-23 22:56:37.189856: step: 1876/1541, loss: -2.54318956649513e-06 2023-01-23 22:56:38.100282: step: 1878/1541, loss: 0.3622312843799591 2023-01-23 22:56:39.046966: step: 1880/1541, loss: 0.08591324090957642 2023-01-23 22:56:39.948612: step: 1882/1541, loss: 0.025970740243792534 2023-01-23 22:56:40.881582: step: 1884/1541, loss: -1.833084024838172e-05 2023-01-23 22:56:41.770185: step: 1886/1541, loss: 9.69046086538583e-05 2023-01-23 22:56:42.777548: step: 1888/1541, loss: 0.051405612379312515 2023-01-23 22:56:43.677411: step: 1890/1541, loss: 0.050683796405792236 2023-01-23 22:56:44.661472: step: 1892/1541, loss: 0.00023456574126612395 2023-01-23 22:56:45.549588: step: 1894/1541, loss: 0.000845023023430258 2023-01-23 22:56:46.489011: step: 1896/1541, loss: -1.3729903002968058e-05 2023-01-23 22:56:47.393482: step: 1898/1541, loss: 1.7330027048956254e-06 2023-01-23 22:56:48.378608: step: 1900/1541, loss: 0.0013607689179480076 2023-01-23 22:56:49.319588: step: 1902/1541, loss: 0.2958414852619171 2023-01-23 22:56:50.187367: step: 1904/1541, loss: 0.0003712474717758596 2023-01-23 22:56:51.089639: step: 1906/1541, loss: 0.00020194947137497365 2023-01-23 22:56:51.994630: step: 1908/1541, loss: 0.0007779441657476127 2023-01-23 22:56:52.890096: step: 1910/1541, loss: 0.10315205156803131 2023-01-23 22:56:53.820819: step: 1912/1541, loss: 0.00020294111163821071 2023-01-23 22:56:54.690612: step: 1914/1541, loss: 0.00043191295117139816 2023-01-23 22:56:55.614599: step: 1916/1541, loss: 0.1815972626209259 2023-01-23 22:56:56.530781: step: 1918/1541, loss: 3.0630801575171063e-06 2023-01-23 22:56:57.427573: step: 1920/1541, loss: 0.0014641210436820984 2023-01-23 22:56:58.358424: step: 1922/1541, loss: 0.030409377068281174 2023-01-23 22:56:59.292711: step: 1924/1541, loss: 0.023381441831588745 2023-01-23 22:57:00.199186: step: 1926/1541, loss: 0.00021987184300087392 2023-01-23 22:57:01.182140: step: 1928/1541, loss: 0.00033058744156733155 2023-01-23 22:57:02.032597: step: 1930/1541, loss: 0.00013447599485516548 2023-01-23 22:57:02.933158: step: 1932/1541, loss: 0.043069202452898026 2023-01-23 22:57:03.833989: step: 1934/1541, loss: 0.0005050711333751678 2023-01-23 22:57:04.717687: step: 1936/1541, loss: 0.0024102835450321436 2023-01-23 22:57:05.632192: step: 1938/1541, loss: 0.0009351747576147318 2023-01-23 22:57:06.582060: step: 1940/1541, loss: 0.00025244057178497314 2023-01-23 22:57:07.512006: step: 1942/1541, loss: 0.007468939293175936 2023-01-23 22:57:08.422254: step: 1944/1541, loss: 0.08069542050361633 2023-01-23 22:57:09.333135: step: 1946/1541, loss: 0.04447748139500618 2023-01-23 22:57:10.215793: step: 1948/1541, loss: 0.0002382164675509557 2023-01-23 22:57:11.071559: step: 1950/1541, loss: 3.8056401535868645e-05 2023-01-23 22:57:11.969432: step: 1952/1541, loss: 2.3899145162431523e-05 2023-01-23 22:57:12.880404: step: 1954/1541, loss: 0.00016344837786164135 2023-01-23 22:57:13.821879: step: 1956/1541, loss: 0.0074877385050058365 2023-01-23 22:57:14.734909: step: 1958/1541, loss: 0.15013398230075836 2023-01-23 22:57:15.693340: step: 1960/1541, loss: 0.026021702215075493 2023-01-23 22:57:16.642121: step: 1962/1541, loss: 1.8079856090480462e-05 2023-01-23 22:57:17.578612: step: 1964/1541, loss: 0.00569685734808445 2023-01-23 22:57:18.485176: step: 1966/1541, loss: 0.18211333453655243 2023-01-23 22:57:19.344462: step: 1968/1541, loss: 0.0026757700834423304 2023-01-23 22:57:20.260771: step: 1970/1541, loss: 0.00022534315939992666 2023-01-23 22:57:21.162575: step: 1972/1541, loss: 4.778716538567096e-05 2023-01-23 22:57:22.076649: step: 1974/1541, loss: 2.5083892978727818e-05 2023-01-23 22:57:22.972780: step: 1976/1541, loss: 0.0007923011435195804 2023-01-23 22:57:23.842392: step: 1978/1541, loss: 0.00015157987945713103 2023-01-23 22:57:24.790263: step: 1980/1541, loss: 0.00017956890224013478 2023-01-23 22:57:25.765446: step: 1982/1541, loss: 0.00015632541908416897 2023-01-23 22:57:26.652434: step: 1984/1541, loss: 0.00015355377399828285 2023-01-23 22:57:27.538290: step: 1986/1541, loss: 0.0046203602105379105 2023-01-23 22:57:28.426910: step: 1988/1541, loss: 0.0003061486640945077 2023-01-23 22:57:29.322714: step: 1990/1541, loss: 0.0010806896025314927 2023-01-23 22:57:30.229317: step: 1992/1541, loss: 0.17378388345241547 2023-01-23 22:57:31.114514: step: 1994/1541, loss: 0.00012641244393307716 2023-01-23 22:57:31.986172: step: 1996/1541, loss: 0.048274293541908264 2023-01-23 22:57:32.867513: step: 1998/1541, loss: 0.005531008820980787 2023-01-23 22:57:33.753165: step: 2000/1541, loss: 0.011723848059773445 2023-01-23 22:57:34.638517: step: 2002/1541, loss: 0.0012231863802298903 2023-01-23 22:57:35.556944: step: 2004/1541, loss: 0.0016106459079310298 2023-01-23 22:57:36.491536: step: 2006/1541, loss: 8.356838952749968e-05 2023-01-23 22:57:37.389598: step: 2008/1541, loss: 0.0008877075742930174 2023-01-23 22:57:38.318208: step: 2010/1541, loss: 0.002053363248705864 2023-01-23 22:57:39.206542: step: 2012/1541, loss: 0.002467654412612319 2023-01-23 22:57:40.076490: step: 2014/1541, loss: 0.013606840744614601 2023-01-23 22:57:41.022097: step: 2016/1541, loss: 0.023046892136335373 2023-01-23 22:57:41.878643: step: 2018/1541, loss: 0.16969990730285645 2023-01-23 22:57:42.801301: step: 2020/1541, loss: 0.0015147349331527948 2023-01-23 22:57:43.729349: step: 2022/1541, loss: 0.00019913316646125168 2023-01-23 22:57:44.680825: step: 2024/1541, loss: 0.011862615123391151 2023-01-23 22:57:45.570507: step: 2026/1541, loss: 0.00023045549460221082 2023-01-23 22:57:46.447245: step: 2028/1541, loss: 0.10491589456796646 2023-01-23 22:57:47.344151: step: 2030/1541, loss: 7.615530921611935e-05 2023-01-23 22:57:48.263949: step: 2032/1541, loss: 0.0005207766080275178 2023-01-23 22:57:49.155120: step: 2034/1541, loss: 0.00039272676804102957 2023-01-23 22:57:50.067808: step: 2036/1541, loss: 3.05363537336234e-05 2023-01-23 22:57:50.946121: step: 2038/1541, loss: 0.000228295277338475 2023-01-23 22:57:51.834191: step: 2040/1541, loss: 0.04055594280362129 2023-01-23 22:57:52.728740: step: 2042/1541, loss: 0.0006267691496759653 2023-01-23 22:57:53.603014: step: 2044/1541, loss: 0.0022852506954222918 2023-01-23 22:57:54.447739: step: 2046/1541, loss: 0.0011557356920093298 2023-01-23 22:57:55.347851: step: 2048/1541, loss: 0.00016944907838478684 2023-01-23 22:57:56.239407: step: 2050/1541, loss: 0.00024614218273200095 2023-01-23 22:57:57.114663: step: 2052/1541, loss: 0.01556900329887867 2023-01-23 22:57:58.059668: step: 2054/1541, loss: 0.018768656998872757 2023-01-23 22:57:58.981407: step: 2056/1541, loss: 0.0002961377031169832 2023-01-23 22:57:59.865948: step: 2058/1541, loss: 0.0020128453616052866 2023-01-23 22:58:00.768528: step: 2060/1541, loss: 0.01819218322634697 2023-01-23 22:58:01.631221: step: 2062/1541, loss: 0.0006858400884084404 2023-01-23 22:58:02.536538: step: 2064/1541, loss: 0.0025203030090779066 2023-01-23 22:58:03.399914: step: 2066/1541, loss: 5.704156592400977e-06 2023-01-23 22:58:04.343814: step: 2068/1541, loss: 0.00685732439160347 2023-01-23 22:58:05.218791: step: 2070/1541, loss: 0.000395709095755592 2023-01-23 22:58:06.167538: step: 2072/1541, loss: 0.32302695512771606 2023-01-23 22:58:07.049682: step: 2074/1541, loss: 0.004060009494423866 2023-01-23 22:58:07.917093: step: 2076/1541, loss: 5.65608061151579e-05 2023-01-23 22:58:08.804523: step: 2078/1541, loss: 8.869730663718656e-05 2023-01-23 22:58:09.701401: step: 2080/1541, loss: 5.008378138882108e-05 2023-01-23 22:58:10.633135: step: 2082/1541, loss: 0.09243263304233551 2023-01-23 22:58:11.507775: step: 2084/1541, loss: 5.4767249821452424e-05 2023-01-23 22:58:12.433537: step: 2086/1541, loss: 2.7962698368355632e-05 2023-01-23 22:58:13.379620: step: 2088/1541, loss: 0.0003835854004137218 2023-01-23 22:58:14.288031: step: 2090/1541, loss: 0.26075488328933716 2023-01-23 22:58:15.192449: step: 2092/1541, loss: 9.113063424592838e-05 2023-01-23 22:58:16.109174: step: 2094/1541, loss: 0.05319974571466446 2023-01-23 22:58:16.992848: step: 2096/1541, loss: 0.0017356901662424207 2023-01-23 22:58:17.885492: step: 2098/1541, loss: 0.01939370296895504 2023-01-23 22:58:18.794058: step: 2100/1541, loss: 0.014441936276853085 2023-01-23 22:58:19.759775: step: 2102/1541, loss: 7.08686129655689e-05 2023-01-23 22:58:20.636217: step: 2104/1541, loss: 0.0007433007704094052 2023-01-23 22:58:21.533931: step: 2106/1541, loss: 0.005665728356689215 2023-01-23 22:58:22.403746: step: 2108/1541, loss: 0.00017655562260188162 2023-01-23 22:58:23.282467: step: 2110/1541, loss: 0.8809811472892761 2023-01-23 22:58:24.228246: step: 2112/1541, loss: 0.009052104316651821 2023-01-23 22:58:25.156081: step: 2114/1541, loss: 0.000676410214509815 2023-01-23 22:58:26.034756: step: 2116/1541, loss: 0.00023196487745735794 2023-01-23 22:58:26.917416: step: 2118/1541, loss: 0.0004729978390969336 2023-01-23 22:58:27.795352: step: 2120/1541, loss: 0.0004184332210570574 2023-01-23 22:58:28.696836: step: 2122/1541, loss: 0.0033881887793540955 2023-01-23 22:58:29.547770: step: 2124/1541, loss: 5.9037309256382287e-05 2023-01-23 22:58:30.402469: step: 2126/1541, loss: -5.498543941939715e-06 2023-01-23 22:58:31.305097: step: 2128/1541, loss: 0.00016718627011869103 2023-01-23 22:58:32.231740: step: 2130/1541, loss: 0.0007340125739574432 2023-01-23 22:58:33.158002: step: 2132/1541, loss: 8.703084313310683e-05 2023-01-23 22:58:34.031649: step: 2134/1541, loss: 0.00012171865091659129 2023-01-23 22:58:34.938613: step: 2136/1541, loss: 8.771911234362051e-05 2023-01-23 22:58:35.886141: step: 2138/1541, loss: 0.015137235634028912 2023-01-23 22:58:36.789232: step: 2140/1541, loss: 0.000696046743541956 2023-01-23 22:58:37.743758: step: 2142/1541, loss: 0.0007619907264597714 2023-01-23 22:58:38.676053: step: 2144/1541, loss: 0.0005137086845934391 2023-01-23 22:58:39.572937: step: 2146/1541, loss: -1.8662600268726237e-05 2023-01-23 22:58:40.499997: step: 2148/1541, loss: 0.034073133021593094 2023-01-23 22:58:41.466597: step: 2150/1541, loss: 0.026246801018714905 2023-01-23 22:58:42.404375: step: 2152/1541, loss: 0.006423416547477245 2023-01-23 22:58:43.296767: step: 2154/1541, loss: 0.038022466003894806 2023-01-23 22:58:44.161008: step: 2156/1541, loss: 2.6395318855065852e-05 2023-01-23 22:58:45.123968: step: 2158/1541, loss: 0.00025599924265407026 2023-01-23 22:58:46.026345: step: 2160/1541, loss: 2.062895509880036e-05 2023-01-23 22:58:46.910828: step: 2162/1541, loss: 0.0015534210251644254 2023-01-23 22:58:47.909057: step: 2164/1541, loss: -8.504614015691914e-06 2023-01-23 22:58:48.784947: step: 2166/1541, loss: 1.0621518413245212e-05 2023-01-23 22:58:49.660854: step: 2168/1541, loss: 0.003020164556801319 2023-01-23 22:58:50.552509: step: 2170/1541, loss: 1.9427820006967522e-05 2023-01-23 22:58:51.480277: step: 2172/1541, loss: 0.2692652940750122 2023-01-23 22:58:52.326400: step: 2174/1541, loss: 0.0011371106375008821 2023-01-23 22:58:53.187951: step: 2176/1541, loss: 0.000803595467004925 2023-01-23 22:58:54.076837: step: 2178/1541, loss: 4.962012098985724e-05 2023-01-23 22:58:54.957830: step: 2180/1541, loss: 0.00036930464557372034 2023-01-23 22:58:55.869689: step: 2182/1541, loss: 0.0011919107055291533 2023-01-23 22:58:56.710508: step: 2184/1541, loss: 0.13054145872592926 2023-01-23 22:58:57.708353: step: 2186/1541, loss: 0.000723944918718189 2023-01-23 22:58:58.657411: step: 2188/1541, loss: 0.0012353794882073998 2023-01-23 22:58:59.576016: step: 2190/1541, loss: 0.3871205747127533 2023-01-23 22:59:00.445902: step: 2192/1541, loss: 0.0149642089381814 2023-01-23 22:59:01.348324: step: 2194/1541, loss: 0.0002213669358752668 2023-01-23 22:59:02.203882: step: 2196/1541, loss: 0.3338276743888855 2023-01-23 22:59:03.101018: step: 2198/1541, loss: 0.051656030118465424 2023-01-23 22:59:04.036056: step: 2200/1541, loss: 0.048216260969638824 2023-01-23 22:59:04.894097: step: 2202/1541, loss: 0.0007387668010778725 2023-01-23 22:59:05.776310: step: 2204/1541, loss: 0.0019079542253166437 2023-01-23 22:59:06.640164: step: 2206/1541, loss: 0.0015938280848786235 2023-01-23 22:59:07.564237: step: 2208/1541, loss: 0.10139337927103043 2023-01-23 22:59:08.456010: step: 2210/1541, loss: 0.013196541927754879 2023-01-23 22:59:09.356592: step: 2212/1541, loss: 0.00619938550516963 2023-01-23 22:59:10.240465: step: 2214/1541, loss: 0.001087538548745215 2023-01-23 22:59:11.171078: step: 2216/1541, loss: 0.07423637062311172 2023-01-23 22:59:12.087277: step: 2218/1541, loss: 0.04312645643949509 2023-01-23 22:59:12.951002: step: 2220/1541, loss: 0.0002598453138489276 2023-01-23 22:59:13.835386: step: 2222/1541, loss: 0.0861194059252739 2023-01-23 22:59:14.703213: step: 2224/1541, loss: 0.00010752870730357245 2023-01-23 22:59:15.604063: step: 2226/1541, loss: 0.009347119368612766 2023-01-23 22:59:16.481720: step: 2228/1541, loss: 0.00029344111680984497 2023-01-23 22:59:17.382518: step: 2230/1541, loss: 0.0001868873368948698 2023-01-23 22:59:18.357333: step: 2232/1541, loss: 0.0005889983149245381 2023-01-23 22:59:19.309138: step: 2234/1541, loss: 0.011721054092049599 2023-01-23 22:59:20.155588: step: 2236/1541, loss: 0.627841591835022 2023-01-23 22:59:21.042254: step: 2238/1541, loss: 0.2644434869289398 2023-01-23 22:59:21.904138: step: 2240/1541, loss: 0.14488130807876587 2023-01-23 22:59:22.741337: step: 2242/1541, loss: 0.0006869256612844765 2023-01-23 22:59:23.603442: step: 2244/1541, loss: 0.04940327629446983 2023-01-23 22:59:24.512868: step: 2246/1541, loss: 0.006359035614877939 2023-01-23 22:59:25.413945: step: 2248/1541, loss: 0.00015689508290961385 2023-01-23 22:59:26.282748: step: 2250/1541, loss: 0.013092461973428726 2023-01-23 22:59:27.275929: step: 2252/1541, loss: 0.00021353643387556076 2023-01-23 22:59:28.158992: step: 2254/1541, loss: 0.0007786676869727671 2023-01-23 22:59:29.045822: step: 2256/1541, loss: 0.00039345360710285604 2023-01-23 22:59:29.941139: step: 2258/1541, loss: 0.018880564719438553 2023-01-23 22:59:30.798740: step: 2260/1541, loss: 0.12650124728679657 2023-01-23 22:59:31.695772: step: 2262/1541, loss: 0.8774988055229187 2023-01-23 22:59:32.583332: step: 2264/1541, loss: 0.05708898976445198 2023-01-23 22:59:33.495010: step: 2266/1541, loss: 0.0021183756180107594 2023-01-23 22:59:34.367058: step: 2268/1541, loss: 0.0004970398149453104 2023-01-23 22:59:35.252307: step: 2270/1541, loss: 0.016982002183794975 2023-01-23 22:59:36.185101: step: 2272/1541, loss: 0.004942007828503847 2023-01-23 22:59:37.074578: step: 2274/1541, loss: 0.12159078568220139 2023-01-23 22:59:37.974055: step: 2276/1541, loss: 0.07251086086034775 2023-01-23 22:59:38.872150: step: 2278/1541, loss: 0.002621241845190525 2023-01-23 22:59:39.748427: step: 2280/1541, loss: 0.0012477643322199583 2023-01-23 22:59:40.643043: step: 2282/1541, loss: 0.004480249248445034 2023-01-23 22:59:41.559011: step: 2284/1541, loss: 0.014762995764613152 2023-01-23 22:59:42.419023: step: 2286/1541, loss: 0.00486206728965044 2023-01-23 22:59:43.328610: step: 2288/1541, loss: 0.0006209515267983079 2023-01-23 22:59:44.229245: step: 2290/1541, loss: 0.002511188155040145 2023-01-23 22:59:45.124822: step: 2292/1541, loss: 0.052539221942424774 2023-01-23 22:59:46.034328: step: 2294/1541, loss: 0.0018334976630285382 2023-01-23 22:59:47.013656: step: 2296/1541, loss: 0.007070107851177454 2023-01-23 22:59:47.896714: step: 2298/1541, loss: 0.03167552873492241 2023-01-23 22:59:48.780879: step: 2300/1541, loss: 0.00923969503492117 2023-01-23 22:59:49.657330: step: 2302/1541, loss: 0.004070992581546307 2023-01-23 22:59:50.553766: step: 2304/1541, loss: 0.00011252128024352714 2023-01-23 22:59:51.420389: step: 2306/1541, loss: 0.0011625945335254073 2023-01-23 22:59:52.305851: step: 2308/1541, loss: 0.0016424369532614946 2023-01-23 22:59:53.213005: step: 2310/1541, loss: 0.0005887096631340683 2023-01-23 22:59:54.130786: step: 2312/1541, loss: 0.25057196617126465 2023-01-23 22:59:55.085827: step: 2314/1541, loss: 7.912582077551633e-05 2023-01-23 22:59:55.950101: step: 2316/1541, loss: 0.34355491399765015 2023-01-23 22:59:56.889622: step: 2318/1541, loss: 0.0001756772107910365 2023-01-23 22:59:57.810199: step: 2320/1541, loss: 0.10050946474075317 2023-01-23 22:59:58.692176: step: 2322/1541, loss: 0.2957567572593689 2023-01-23 22:59:59.585624: step: 2324/1541, loss: 0.0004341535095591098 2023-01-23 23:00:00.479978: step: 2326/1541, loss: 0.20728819072246552 2023-01-23 23:00:01.446631: step: 2328/1541, loss: 0.05576340854167938 2023-01-23 23:00:02.338571: step: 2330/1541, loss: 9.155770385405049e-05 2023-01-23 23:00:03.235852: step: 2332/1541, loss: 0.000821654568426311 2023-01-23 23:00:04.152479: step: 2334/1541, loss: 0.0028712882194668055 2023-01-23 23:00:05.033484: step: 2336/1541, loss: 0.000658813922200352 2023-01-23 23:00:05.927331: step: 2338/1541, loss: 0.03158692270517349 2023-01-23 23:00:06.850558: step: 2340/1541, loss: 0.001161114894784987 2023-01-23 23:00:07.730916: step: 2342/1541, loss: 0.0008896207436919212 2023-01-23 23:00:08.593006: step: 2344/1541, loss: 0.0012890025973320007 2023-01-23 23:00:09.491608: step: 2346/1541, loss: 0.0006271651946008205 2023-01-23 23:00:10.395969: step: 2348/1541, loss: 0.03851201757788658 2023-01-23 23:00:11.276900: step: 2350/1541, loss: 0.005262993276119232 2023-01-23 23:00:12.154543: step: 2352/1541, loss: 0.013501138426363468 2023-01-23 23:00:13.075966: step: 2354/1541, loss: 0.0004129305307287723 2023-01-23 23:00:13.956577: step: 2356/1541, loss: 0.005476463586091995 2023-01-23 23:00:14.849589: step: 2358/1541, loss: 0.001473152544349432 2023-01-23 23:00:15.710368: step: 2360/1541, loss: 0.004892335273325443 2023-01-23 23:00:16.615696: step: 2362/1541, loss: 0.11992213875055313 2023-01-23 23:00:17.548906: step: 2364/1541, loss: 0.052258215844631195 2023-01-23 23:00:18.415532: step: 2366/1541, loss: 0.0003364672011230141 2023-01-23 23:00:19.331681: step: 2368/1541, loss: 6.899569416418672e-05 2023-01-23 23:00:20.283343: step: 2370/1541, loss: 0.0001486167893745005 2023-01-23 23:00:21.211600: step: 2372/1541, loss: 0.003541677026078105 2023-01-23 23:00:22.129434: step: 2374/1541, loss: 0.0008117175893858075 2023-01-23 23:00:23.006709: step: 2376/1541, loss: 0.000848570023663342 2023-01-23 23:00:23.926767: step: 2378/1541, loss: 0.2647918462753296 2023-01-23 23:00:24.805031: step: 2380/1541, loss: 0.003062885021790862 2023-01-23 23:00:25.713177: step: 2382/1541, loss: 0.00015166867524385452 2023-01-23 23:00:26.596787: step: 2384/1541, loss: 0.16160738468170166 2023-01-23 23:00:27.486219: step: 2386/1541, loss: 3.150697739329189e-05 2023-01-23 23:00:28.390525: step: 2388/1541, loss: 5.546807005885057e-05 2023-01-23 23:00:29.278873: step: 2390/1541, loss: 7.609128806507215e-05 2023-01-23 23:00:30.187896: step: 2392/1541, loss: 0.04525792971253395 2023-01-23 23:00:31.092124: step: 2394/1541, loss: 0.00016853067791089416 2023-01-23 23:00:31.941848: step: 2396/1541, loss: 8.03396906121634e-05 2023-01-23 23:00:32.835948: step: 2398/1541, loss: 0.00039550213841721416 2023-01-23 23:00:33.728445: step: 2400/1541, loss: 0.14229179918766022 2023-01-23 23:00:34.616611: step: 2402/1541, loss: 0.000604666827712208 2023-01-23 23:00:35.526669: step: 2404/1541, loss: 0.0070034596137702465 2023-01-23 23:00:36.398976: step: 2406/1541, loss: 0.0007554706535302103 2023-01-23 23:00:37.286988: step: 2408/1541, loss: 0.0002590533113107085 2023-01-23 23:00:38.210902: step: 2410/1541, loss: 0.002926809014752507 2023-01-23 23:00:39.162052: step: 2412/1541, loss: 0.000992420013062656 2023-01-23 23:00:40.106221: step: 2414/1541, loss: 0.0017566883470863104 2023-01-23 23:00:41.074117: step: 2416/1541, loss: 0.00043699733214452863 2023-01-23 23:00:41.969049: step: 2418/1541, loss: 0.007709618657827377 2023-01-23 23:00:42.916757: step: 2420/1541, loss: 0.0019036595476791263 2023-01-23 23:00:43.797780: step: 2422/1541, loss: 0.004959757439792156 2023-01-23 23:00:44.726456: step: 2424/1541, loss: 0.009638046845793724 2023-01-23 23:00:45.591540: step: 2426/1541, loss: 0.0005710831028409302 2023-01-23 23:00:46.486321: step: 2428/1541, loss: 0.2499656230211258 2023-01-23 23:00:47.446130: step: 2430/1541, loss: 0.005364397540688515 2023-01-23 23:00:48.336097: step: 2432/1541, loss: 0.2871406376361847 2023-01-23 23:00:49.230868: step: 2434/1541, loss: 0.2297966480255127 2023-01-23 23:00:50.131535: step: 2436/1541, loss: 0.027972880750894547 2023-01-23 23:00:51.042387: step: 2438/1541, loss: 0.00010655206278897822 2023-01-23 23:00:51.913933: step: 2440/1541, loss: 0.00022866312065161765 2023-01-23 23:00:52.877026: step: 2442/1541, loss: 0.01153346337378025 2023-01-23 23:00:53.788661: step: 2444/1541, loss: 0.023890478536486626 2023-01-23 23:00:54.679402: step: 2446/1541, loss: 0.0016180594684556127 2023-01-23 23:00:55.589581: step: 2448/1541, loss: 0.019196348264813423 2023-01-23 23:00:56.470851: step: 2450/1541, loss: 0.000846502254717052 2023-01-23 23:00:57.419142: step: 2452/1541, loss: 0.0025225540157407522 2023-01-23 23:00:58.360954: step: 2454/1541, loss: 0.00036144210025668144 2023-01-23 23:00:59.314258: step: 2456/1541, loss: 0.0001783696934580803 2023-01-23 23:01:00.225249: step: 2458/1541, loss: 0.00016351547674275935 2023-01-23 23:01:01.068622: step: 2460/1541, loss: 0.003412215504795313 2023-01-23 23:01:02.017585: step: 2462/1541, loss: 0.0030929362401366234 2023-01-23 23:01:02.881129: step: 2464/1541, loss: 0.00179427326656878 2023-01-23 23:01:03.814841: step: 2466/1541, loss: 0.09597449004650116 2023-01-23 23:01:04.713535: step: 2468/1541, loss: 0.00036909442860633135 2023-01-23 23:01:05.646662: step: 2470/1541, loss: 0.0005467106238938868 2023-01-23 23:01:06.504490: step: 2472/1541, loss: 0.0006176250171847641 2023-01-23 23:01:07.429250: step: 2474/1541, loss: 0.01153662521392107 2023-01-23 23:01:08.309342: step: 2476/1541, loss: 0.0002225573844043538 2023-01-23 23:01:09.192486: step: 2478/1541, loss: 0.0007635549991391599 2023-01-23 23:01:10.075676: step: 2480/1541, loss: 0.0005015567294321954 2023-01-23 23:01:10.958989: step: 2482/1541, loss: 0.059150081127882004 2023-01-23 23:01:11.837170: step: 2484/1541, loss: 8.706151857040823e-05 2023-01-23 23:01:12.719101: step: 2486/1541, loss: 0.004517469089478254 2023-01-23 23:01:13.601180: step: 2488/1541, loss: 0.002953020390123129 2023-01-23 23:01:14.473197: step: 2490/1541, loss: 0.019031750038266182 2023-01-23 23:01:15.349294: step: 2492/1541, loss: 0.020948780700564384 2023-01-23 23:01:16.242893: step: 2494/1541, loss: 0.0005529611953534186 2023-01-23 23:01:17.132069: step: 2496/1541, loss: 0.06938519328832626 2023-01-23 23:01:18.010516: step: 2498/1541, loss: 0.00028901660698466003 2023-01-23 23:01:18.894269: step: 2500/1541, loss: 5.2734059863723814e-05 2023-01-23 23:01:19.829066: step: 2502/1541, loss: 5.225258428254165e-05 2023-01-23 23:01:20.709211: step: 2504/1541, loss: 0.003463480155915022 2023-01-23 23:01:21.589442: step: 2506/1541, loss: 0.006803775206208229 2023-01-23 23:01:22.478948: step: 2508/1541, loss: 0.0006285079289227724 2023-01-23 23:01:23.350740: step: 2510/1541, loss: 0.00018746676505543292 2023-01-23 23:01:24.220785: step: 2512/1541, loss: 0.0010065640090033412 2023-01-23 23:01:25.086078: step: 2514/1541, loss: 0.0010148674482479692 2023-01-23 23:01:25.986385: step: 2516/1541, loss: 8.624634938314557e-05 2023-01-23 23:01:26.862981: step: 2518/1541, loss: 7.15719215804711e-05 2023-01-23 23:01:27.779303: step: 2520/1541, loss: 0.0009549576207064092 2023-01-23 23:01:28.647927: step: 2522/1541, loss: 1.0952368256766931e-06 2023-01-23 23:01:29.510265: step: 2524/1541, loss: 7.36901129130274e-05 2023-01-23 23:01:30.396868: step: 2526/1541, loss: 0.00010610843310132623 2023-01-23 23:01:31.300384: step: 2528/1541, loss: -1.5766176147735678e-05 2023-01-23 23:01:32.248973: step: 2530/1541, loss: 0.0037031436804682016 2023-01-23 23:01:33.115639: step: 2532/1541, loss: 0.0003441813460085541 2023-01-23 23:01:33.997610: step: 2534/1541, loss: 0.0015288286376744509 2023-01-23 23:01:34.966780: step: 2536/1541, loss: 0.005806132685393095 2023-01-23 23:01:35.827364: step: 2538/1541, loss: 0.0016253716312348843 2023-01-23 23:01:36.733108: step: 2540/1541, loss: 0.0007782293832860887 2023-01-23 23:01:37.599664: step: 2542/1541, loss: 5.3161955293035135e-05 2023-01-23 23:01:38.460505: step: 2544/1541, loss: 0.14281389117240906 2023-01-23 23:01:39.361001: step: 2546/1541, loss: 0.0003644815587904304 2023-01-23 23:01:40.213309: step: 2548/1541, loss: 0.09487461298704147 2023-01-23 23:01:41.116792: step: 2550/1541, loss: 0.02442656271159649 2023-01-23 23:01:42.015796: step: 2552/1541, loss: 0.0036229544784873724 2023-01-23 23:01:42.917829: step: 2554/1541, loss: -3.1366978419100633e-06 2023-01-23 23:01:43.897568: step: 2556/1541, loss: 0.1664234697818756 2023-01-23 23:01:44.769798: step: 2558/1541, loss: 4.275620085536502e-05 2023-01-23 23:01:45.670859: step: 2560/1541, loss: -7.508396265620831e-06 2023-01-23 23:01:46.570167: step: 2562/1541, loss: -6.275660416577011e-05 2023-01-23 23:01:47.445827: step: 2564/1541, loss: 0.1214991956949234 2023-01-23 23:01:48.413579: step: 2566/1541, loss: 0.04650595784187317 2023-01-23 23:01:49.293119: step: 2568/1541, loss: 0.00011200853623449802 2023-01-23 23:01:50.218466: step: 2570/1541, loss: 0.12043032050132751 2023-01-23 23:01:51.078002: step: 2572/1541, loss: 0.0008457265212200582 2023-01-23 23:01:51.951030: step: 2574/1541, loss: 0.00041632636566646397 2023-01-23 23:01:52.831324: step: 2576/1541, loss: 0.1572836935520172 2023-01-23 23:01:53.769236: step: 2578/1541, loss: 0.001898355782032013 2023-01-23 23:01:54.677452: step: 2580/1541, loss: 0.003395537845790386 2023-01-23 23:01:55.557934: step: 2582/1541, loss: 0.00014679599553346634 2023-01-23 23:01:56.490626: step: 2584/1541, loss: 2.5138222554232925e-05 2023-01-23 23:01:57.365543: step: 2586/1541, loss: 0.006009490229189396 2023-01-23 23:01:58.387505: step: 2588/1541, loss: 0.09232606738805771 2023-01-23 23:01:59.302605: step: 2590/1541, loss: 0.0010017998283728957 2023-01-23 23:02:00.173956: step: 2592/1541, loss: 0.0005519295809790492 2023-01-23 23:02:01.082730: step: 2594/1541, loss: 2.297914761584252e-06 2023-01-23 23:02:01.991065: step: 2596/1541, loss: 1.3418495655059814e-05 2023-01-23 23:02:02.857511: step: 2598/1541, loss: 7.844980427762493e-05 2023-01-23 23:02:03.781878: step: 2600/1541, loss: 0.25350549817085266 2023-01-23 23:02:04.732524: step: 2602/1541, loss: 0.0002481359988451004 2023-01-23 23:02:05.674177: step: 2604/1541, loss: 0.00027112459065392613 2023-01-23 23:02:06.566247: step: 2606/1541, loss: 0.05856044217944145 2023-01-23 23:02:07.455881: step: 2608/1541, loss: 0.003990527708083391 2023-01-23 23:02:08.338074: step: 2610/1541, loss: -3.56075179297477e-05 2023-01-23 23:02:09.199804: step: 2612/1541, loss: 0.001985840732231736 2023-01-23 23:02:10.081834: step: 2614/1541, loss: 0.11592238396406174 2023-01-23 23:02:10.965807: step: 2616/1541, loss: 2.25647234916687 2023-01-23 23:02:11.909550: step: 2618/1541, loss: 9.027807391248643e-05 2023-01-23 23:02:12.821803: step: 2620/1541, loss: 0.14565353095531464 2023-01-23 23:02:13.740778: step: 2622/1541, loss: 6.699791265418753e-05 2023-01-23 23:02:14.610664: step: 2624/1541, loss: 0.0006413019727915525 2023-01-23 23:02:15.481150: step: 2626/1541, loss: 0.003650117665529251 2023-01-23 23:02:16.395706: step: 2628/1541, loss: 0.009418786503374577 2023-01-23 23:02:17.284987: step: 2630/1541, loss: 0.06077725812792778 2023-01-23 23:02:18.158551: step: 2632/1541, loss: 0.0004354826523922384 2023-01-23 23:02:19.058340: step: 2634/1541, loss: 0.00031672106706537306 2023-01-23 23:02:20.005287: step: 2636/1541, loss: 0.056038256734609604 2023-01-23 23:02:20.858014: step: 2638/1541, loss: 0.0007982325041666627 2023-01-23 23:02:21.703724: step: 2640/1541, loss: 0.25702205300331116 2023-01-23 23:02:22.589988: step: 2642/1541, loss: 0.07058453559875488 2023-01-23 23:02:23.478856: step: 2644/1541, loss: 0.0002430286112939939 2023-01-23 23:02:24.414479: step: 2646/1541, loss: 0.00018992296827491373 2023-01-23 23:02:25.281135: step: 2648/1541, loss: 6.288680015131831e-05 2023-01-23 23:02:26.181836: step: 2650/1541, loss: 0.008409143425524235 2023-01-23 23:02:27.172525: step: 2652/1541, loss: 0.00028832326643168926 2023-01-23 23:02:28.030578: step: 2654/1541, loss: 0.017628774046897888 2023-01-23 23:02:28.918398: step: 2656/1541, loss: 0.1652597188949585 2023-01-23 23:02:29.821388: step: 2658/1541, loss: 0.0005206650239415467 2023-01-23 23:02:30.711240: step: 2660/1541, loss: 0.00010291599028278142 2023-01-23 23:02:31.607561: step: 2662/1541, loss: 0.007161811459809542 2023-01-23 23:02:32.494850: step: 2664/1541, loss: 0.09178223460912704 2023-01-23 23:02:33.379828: step: 2666/1541, loss: 0.3200998306274414 2023-01-23 23:02:34.240761: step: 2668/1541, loss: 0.005311672575771809 2023-01-23 23:02:35.135316: step: 2670/1541, loss: 0.0033432627096772194 2023-01-23 23:02:36.007731: step: 2672/1541, loss: 0.1461172103881836 2023-01-23 23:02:36.867779: step: 2674/1541, loss: 0.001439868938177824 2023-01-23 23:02:37.740746: step: 2676/1541, loss: 3.874148751492612e-05 2023-01-23 23:02:38.609221: step: 2678/1541, loss: 0.004806823562830687 2023-01-23 23:02:39.519662: step: 2680/1541, loss: 0.005279433913528919 2023-01-23 23:02:40.420528: step: 2682/1541, loss: 0.004324617329984903 2023-01-23 23:02:41.300949: step: 2684/1541, loss: -2.6036518647742923e-06 2023-01-23 23:02:42.192788: step: 2686/1541, loss: 1.918893030961044e-05 2023-01-23 23:02:43.056597: step: 2688/1541, loss: 0.02725786529481411 2023-01-23 23:02:43.909154: step: 2690/1541, loss: 0.0003532475093379617 2023-01-23 23:02:44.817273: step: 2692/1541, loss: -8.468268788419664e-06 2023-01-23 23:02:45.733424: step: 2694/1541, loss: 1.0347082614898682 2023-01-23 23:02:46.634802: step: 2696/1541, loss: 0.2979260981082916 2023-01-23 23:02:47.492542: step: 2698/1541, loss: 0.0002712339337449521 2023-01-23 23:02:48.365381: step: 2700/1541, loss: 0.0019592628814280033 2023-01-23 23:02:49.273466: step: 2702/1541, loss: 0.035365957766771317 2023-01-23 23:02:50.161072: step: 2704/1541, loss: 0.09101156145334244 2023-01-23 23:02:51.089498: step: 2706/1541, loss: 3.971606201957911e-05 2023-01-23 23:02:51.955424: step: 2708/1541, loss: 0.00034359251731075346 2023-01-23 23:02:52.827504: step: 2710/1541, loss: 1.91360704775434e-05 2023-01-23 23:02:53.777375: step: 2712/1541, loss: 5.619344301521778e-05 2023-01-23 23:02:54.640925: step: 2714/1541, loss: 4.122089740121737e-05 2023-01-23 23:02:55.543889: step: 2716/1541, loss: 0.0005630672094412148 2023-01-23 23:02:56.452800: step: 2718/1541, loss: 0.0001547061838209629 2023-01-23 23:02:57.329238: step: 2720/1541, loss: 0.006231214385479689 2023-01-23 23:02:58.195893: step: 2722/1541, loss: 2.9201852157711983e-05 2023-01-23 23:02:59.033906: step: 2724/1541, loss: 0.1750098615884781 2023-01-23 23:02:59.870756: step: 2726/1541, loss: 0.002981262281537056 2023-01-23 23:03:00.785944: step: 2728/1541, loss: 0.0014572353102266788 2023-01-23 23:03:01.678827: step: 2730/1541, loss: 0.00015924868057481945 2023-01-23 23:03:02.609782: step: 2732/1541, loss: 0.014257999137043953 2023-01-23 23:03:03.549164: step: 2734/1541, loss: 0.00021671947615686804 2023-01-23 23:03:04.393880: step: 2736/1541, loss: 0.06613842397928238 2023-01-23 23:03:05.320373: step: 2738/1541, loss: 0.00783485360443592 2023-01-23 23:03:06.222099: step: 2740/1541, loss: 0.00037951284321025014 2023-01-23 23:03:07.159895: step: 2742/1541, loss: 0.04639870673418045 2023-01-23 23:03:08.061395: step: 2744/1541, loss: 0.003602518467232585 2023-01-23 23:03:08.988901: step: 2746/1541, loss: 0.039221975952386856 2023-01-23 23:03:09.889701: step: 2748/1541, loss: 5.3603198466589674e-05 2023-01-23 23:03:10.769759: step: 2750/1541, loss: 0.0016496374737471342 2023-01-23 23:03:11.832153: step: 2752/1541, loss: 0.06017075851559639 2023-01-23 23:03:12.790340: step: 2754/1541, loss: 0.0006817529210820794 2023-01-23 23:03:13.725338: step: 2756/1541, loss: 0.023311372846364975 2023-01-23 23:03:14.576496: step: 2758/1541, loss: 0.002876145066693425 2023-01-23 23:03:15.460930: step: 2760/1541, loss: 0.02706991322338581 2023-01-23 23:03:16.324039: step: 2762/1541, loss: 0.00025593978352844715 2023-01-23 23:03:17.242965: step: 2764/1541, loss: 0.11798303574323654 2023-01-23 23:03:18.188348: step: 2766/1541, loss: 0.053100988268852234 2023-01-23 23:03:19.098108: step: 2768/1541, loss: 0.021211661398410797 2023-01-23 23:03:19.974127: step: 2770/1541, loss: 0.00209166225977242 2023-01-23 23:03:20.932334: step: 2772/1541, loss: 0.0026296391151845455 2023-01-23 23:03:21.808642: step: 2774/1541, loss: 0.00014652237587142736 2023-01-23 23:03:22.733079: step: 2776/1541, loss: 8.671639807289466e-05 2023-01-23 23:03:23.696686: step: 2778/1541, loss: 0.00033807597355917096 2023-01-23 23:03:24.603868: step: 2780/1541, loss: 0.00019110558787360787 2023-01-23 23:03:25.470683: step: 2782/1541, loss: 0.00016250637418124825 2023-01-23 23:03:26.323227: step: 2784/1541, loss: 0.0009060577722266316 2023-01-23 23:03:27.202020: step: 2786/1541, loss: 0.007635063491761684 2023-01-23 23:03:28.127697: step: 2788/1541, loss: 0.051856983453035355 2023-01-23 23:03:29.005716: step: 2790/1541, loss: 0.026635296642780304 2023-01-23 23:03:29.887470: step: 2792/1541, loss: 0.5128422379493713 2023-01-23 23:03:30.762680: step: 2794/1541, loss: 0.013879184611141682 2023-01-23 23:03:31.597553: step: 2796/1541, loss: 0.19462017714977264 2023-01-23 23:03:32.493029: step: 2798/1541, loss: 0.0003230652364436537 2023-01-23 23:03:33.393121: step: 2800/1541, loss: 0.00043559595360420644 2023-01-23 23:03:34.307307: step: 2802/1541, loss: 0.0012509021908044815 2023-01-23 23:03:35.261178: step: 2804/1541, loss: 0.029872220009565353 2023-01-23 23:03:36.210969: step: 2806/1541, loss: 0.66985023021698 2023-01-23 23:03:37.157698: step: 2808/1541, loss: 0.02198808826506138 2023-01-23 23:03:38.093867: step: 2810/1541, loss: 0.034888025373220444 2023-01-23 23:03:38.991512: step: 2812/1541, loss: 0.069404236972332 2023-01-23 23:03:39.875792: step: 2814/1541, loss: 0.00011093653301941231 2023-01-23 23:03:40.730127: step: 2816/1541, loss: 0.04462577775120735 2023-01-23 23:03:41.592790: step: 2818/1541, loss: 0.003409123048186302 2023-01-23 23:03:42.497523: step: 2820/1541, loss: -1.0283788469678257e-05 2023-01-23 23:03:43.426862: step: 2822/1541, loss: 0.0003061212773900479 2023-01-23 23:03:44.375478: step: 2824/1541, loss: 0.00015116340364329517 2023-01-23 23:03:45.244328: step: 2826/1541, loss: 0.003143249312415719 2023-01-23 23:03:46.131799: step: 2828/1541, loss: 0.06617316603660583 2023-01-23 23:03:47.028265: step: 2830/1541, loss: 0.04707273468375206 2023-01-23 23:03:47.915016: step: 2832/1541, loss: 0.004762793891131878 2023-01-23 23:03:48.795650: step: 2834/1541, loss: 0.0010564548429101706 2023-01-23 23:03:49.696099: step: 2836/1541, loss: 3.129839569737669e-06 2023-01-23 23:03:50.600687: step: 2838/1541, loss: 0.7061739563941956 2023-01-23 23:03:51.460235: step: 2840/1541, loss: 0.0006188189727254212 2023-01-23 23:03:52.335807: step: 2842/1541, loss: 2.362934719712939e-05 2023-01-23 23:03:53.205774: step: 2844/1541, loss: 0.06198452040553093 2023-01-23 23:03:54.128353: step: 2846/1541, loss: 1.4192612070473842e-05 2023-01-23 23:03:55.064234: step: 2848/1541, loss: 0.0005004432168789208 2023-01-23 23:03:55.914375: step: 2850/1541, loss: 0.0014809801941737533 2023-01-23 23:03:56.780173: step: 2852/1541, loss: 0.01849971152842045 2023-01-23 23:03:57.660458: step: 2854/1541, loss: 0.0004003867506980896 2023-01-23 23:03:58.662720: step: 2856/1541, loss: 0.00017087871674448252 2023-01-23 23:03:59.567386: step: 2858/1541, loss: 5.45216198588605e-06 2023-01-23 23:04:00.494096: step: 2860/1541, loss: 0.0004707648477051407 2023-01-23 23:04:01.364968: step: 2862/1541, loss: 0.10963411629199982 2023-01-23 23:04:02.263157: step: 2864/1541, loss: 5.2713970944751054e-05 2023-01-23 23:04:03.178060: step: 2866/1541, loss: 0.0003921543247997761 2023-01-23 23:04:04.098622: step: 2868/1541, loss: 0.0007351051899604499 2023-01-23 23:04:04.978544: step: 2870/1541, loss: 0.00018537277355790138 2023-01-23 23:04:05.886235: step: 2872/1541, loss: 0.00047320855082944036 2023-01-23 23:04:06.741730: step: 2874/1541, loss: 0.0004470600688364357 2023-01-23 23:04:07.644155: step: 2876/1541, loss: 0.0003727900329977274 2023-01-23 23:04:08.548316: step: 2878/1541, loss: 0.001571385539136827 2023-01-23 23:04:09.438519: step: 2880/1541, loss: 0.0064346762374043465 2023-01-23 23:04:10.309122: step: 2882/1541, loss: 0.0003000617725774646 2023-01-23 23:04:11.259668: step: 2884/1541, loss: 0.027447186410427094 2023-01-23 23:04:12.142773: step: 2886/1541, loss: 0.00026257283752784133 2023-01-23 23:04:13.028085: step: 2888/1541, loss: 0.0024021631106734276 2023-01-23 23:04:13.977829: step: 2890/1541, loss: 0.045016322284936905 2023-01-23 23:04:14.865356: step: 2892/1541, loss: 0.005390141159296036 2023-01-23 23:04:15.779713: step: 2894/1541, loss: 0.005384753458201885 2023-01-23 23:04:16.665776: step: 2896/1541, loss: 0.00016400395543314517 2023-01-23 23:04:17.545717: step: 2898/1541, loss: 1.9899085600627586e-05 2023-01-23 23:04:18.438539: step: 2900/1541, loss: 0.0005763206863775849 2023-01-23 23:04:19.351609: step: 2902/1541, loss: 0.030929844826459885 2023-01-23 23:04:20.230797: step: 2904/1541, loss: 7.178873056545854e-05 2023-01-23 23:04:21.132163: step: 2906/1541, loss: 0.0028418598230928183 2023-01-23 23:04:22.051654: step: 2908/1541, loss: 0.16329997777938843 2023-01-23 23:04:22.909201: step: 2910/1541, loss: 0.0006158144678920507 2023-01-23 23:04:23.857749: step: 2912/1541, loss: 0.001195825869217515 2023-01-23 23:04:24.739789: step: 2914/1541, loss: 0.00019571930170059204 2023-01-23 23:04:25.635806: step: 2916/1541, loss: 0.01823274791240692 2023-01-23 23:04:26.515308: step: 2918/1541, loss: 0.04588202014565468 2023-01-23 23:04:27.452277: step: 2920/1541, loss: 5.353142478270456e-05 2023-01-23 23:04:28.296589: step: 2922/1541, loss: 0.0005059768445789814 2023-01-23 23:04:29.195453: step: 2924/1541, loss: 0.0014338685432448983 2023-01-23 23:04:30.098281: step: 2926/1541, loss: 0.0004267118638381362 2023-01-23 23:04:31.032654: step: 2928/1541, loss: 0.04106041043996811 2023-01-23 23:04:31.986432: step: 2930/1541, loss: 0.07162600755691528 2023-01-23 23:04:32.883869: step: 2932/1541, loss: 0.0008359297644346952 2023-01-23 23:04:33.797322: step: 2934/1541, loss: 0.002755397465080023 2023-01-23 23:04:34.671638: step: 2936/1541, loss: 0.0010999746154993773 2023-01-23 23:04:35.549512: step: 2938/1541, loss: 2.3859973225626163e-05 2023-01-23 23:04:36.427591: step: 2940/1541, loss: 0.028268083930015564 2023-01-23 23:04:37.326928: step: 2942/1541, loss: 0.0016832536784932017 2023-01-23 23:04:38.250667: step: 2944/1541, loss: -6.160590419312939e-05 2023-01-23 23:04:39.098054: step: 2946/1541, loss: 2.678433338587638e-05 2023-01-23 23:04:39.966536: step: 2948/1541, loss: 6.870999641250819e-05 2023-01-23 23:04:40.890630: step: 2950/1541, loss: 0.12176738679409027 2023-01-23 23:04:41.826051: step: 2952/1541, loss: 1.7724517583847046 2023-01-23 23:04:42.706933: step: 2954/1541, loss: 0.009925463236868382 2023-01-23 23:04:43.598641: step: 2956/1541, loss: 0.08957140892744064 2023-01-23 23:04:44.532590: step: 2958/1541, loss: 5.236863944446668e-05 2023-01-23 23:04:45.414032: step: 2960/1541, loss: 0.00011907948646694422 2023-01-23 23:04:46.297079: step: 2962/1541, loss: 0.03563906252384186 2023-01-23 23:04:47.221808: step: 2964/1541, loss: 0.0026500700041651726 2023-01-23 23:04:48.165861: step: 2966/1541, loss: 0.05910339578986168 2023-01-23 23:04:49.054396: step: 2968/1541, loss: 7.967641431605443e-05 2023-01-23 23:04:49.911256: step: 2970/1541, loss: 7.268647095770575e-06 2023-01-23 23:04:50.816639: step: 2972/1541, loss: 0.1485016793012619 2023-01-23 23:04:51.732589: step: 2974/1541, loss: 0.08965735137462616 2023-01-23 23:04:52.603301: step: 2976/1541, loss: 0.0007878479082137346 2023-01-23 23:04:53.502986: step: 2978/1541, loss: 0.39148637652397156 2023-01-23 23:04:54.410329: step: 2980/1541, loss: 4.19256066379603e-05 2023-01-23 23:04:55.319898: step: 2982/1541, loss: 0.015115004032850266 2023-01-23 23:04:56.226446: step: 2984/1541, loss: -5.971268819848774e-06 2023-01-23 23:04:57.115569: step: 2986/1541, loss: 0.0004358025034889579 2023-01-23 23:04:58.022204: step: 2988/1541, loss: 6.900918378960341e-05 2023-01-23 23:04:58.915332: step: 2990/1541, loss: 0.050251156091690063 2023-01-23 23:04:59.813692: step: 2992/1541, loss: 0.0014675055863335729 2023-01-23 23:05:00.685614: step: 2994/1541, loss: 0.0006926840869709849 2023-01-23 23:05:01.577043: step: 2996/1541, loss: 0.018766634166240692 2023-01-23 23:05:02.544781: step: 2998/1541, loss: 0.00046976417070254683 2023-01-23 23:05:03.420901: step: 3000/1541, loss: 0.00031375227263197303 2023-01-23 23:05:04.330916: step: 3002/1541, loss: 8.951225026976317e-06 2023-01-23 23:05:05.209137: step: 3004/1541, loss: 0.00023009502911008894 2023-01-23 23:05:06.101450: step: 3006/1541, loss: 0.002884338377043605 2023-01-23 23:05:07.044818: step: 3008/1541, loss: 7.253883813973516e-05 2023-01-23 23:05:07.947811: step: 3010/1541, loss: 0.00013704568846151233 2023-01-23 23:05:08.828954: step: 3012/1541, loss: 0.0026492795441299677 2023-01-23 23:05:09.679823: step: 3014/1541, loss: 0.0027529082726687193 2023-01-23 23:05:10.609025: step: 3016/1541, loss: 6.424482853617519e-05 2023-01-23 23:05:11.594247: step: 3018/1541, loss: 0.07983792573213577 2023-01-23 23:05:12.493608: step: 3020/1541, loss: 0.01978643238544464 2023-01-23 23:05:13.513060: step: 3022/1541, loss: 0.04386165365576744 2023-01-23 23:05:14.380294: step: 3024/1541, loss: 0.0042038727551698685 2023-01-23 23:05:15.262369: step: 3026/1541, loss: 1.6925194358918816e-05 2023-01-23 23:05:16.125072: step: 3028/1541, loss: 0.003112801117822528 2023-01-23 23:05:17.045177: step: 3030/1541, loss: 0.00013970458530820906 2023-01-23 23:05:17.919658: step: 3032/1541, loss: 6.259878864511847e-05 2023-01-23 23:05:18.836963: step: 3034/1541, loss: 0.00022101103968452662 2023-01-23 23:05:19.713583: step: 3036/1541, loss: 0.0014740494079887867 2023-01-23 23:05:20.735160: step: 3038/1541, loss: 8.47334522404708e-05 2023-01-23 23:05:21.626798: step: 3040/1541, loss: 0.006649949587881565 2023-01-23 23:05:22.495994: step: 3042/1541, loss: -2.614795448607765e-06 2023-01-23 23:05:23.362264: step: 3044/1541, loss: 2.0602346921805292e-05 2023-01-23 23:05:24.254897: step: 3046/1541, loss: 1.992136094486341e-05 2023-01-23 23:05:25.152485: step: 3048/1541, loss: -7.865525731176604e-06 2023-01-23 23:05:26.007901: step: 3050/1541, loss: 0.0005577083211392164 2023-01-23 23:05:26.891634: step: 3052/1541, loss: 0.0005375455366447568 2023-01-23 23:05:27.806963: step: 3054/1541, loss: 0.09671058505773544 2023-01-23 23:05:28.668184: step: 3056/1541, loss: 8.971783972810954e-05 2023-01-23 23:05:29.549474: step: 3058/1541, loss: 0.00043323932914063334 2023-01-23 23:05:30.422631: step: 3060/1541, loss: 0.0014474622439593077 2023-01-23 23:05:31.331540: step: 3062/1541, loss: 0.2589079439640045 2023-01-23 23:05:32.241420: step: 3064/1541, loss: 0.3021204471588135 2023-01-23 23:05:33.147824: step: 3066/1541, loss: -1.387894280924229e-05 2023-01-23 23:05:34.072982: step: 3068/1541, loss: 0.0001234650262631476 2023-01-23 23:05:34.895380: step: 3070/1541, loss: 1.1274940334260464e-05 2023-01-23 23:05:35.805689: step: 3072/1541, loss: 0.011107423342764378 2023-01-23 23:05:36.772830: step: 3074/1541, loss: 0.008822828531265259 2023-01-23 23:05:37.704990: step: 3076/1541, loss: 9.289561421610415e-05 2023-01-23 23:05:38.624903: step: 3078/1541, loss: 0.0003521311446093023 2023-01-23 23:05:39.470308: step: 3080/1541, loss: 4.108202483621426e-05 2023-01-23 23:05:40.345567: step: 3082/1541, loss: 9.531783143756911e-05 2023-01-23 23:05:40.948384: step: 3084/1541, loss: 0.0014171141665428877 ================================================== Loss: 0.049 -------------------- Dev Chinese: {'event': {'p': 0.6272401433691757, 'r': 0.6990679094540613, 'f1': 0.6612090680100756}, 'argument': {'p': 0.6727793514338454, 'r': 0.6586760575761482, 'f1': 0.6656530107164214}, 'combined': 0.44013580683390585, 'epoch': 37} Test Chinese: {'event': {'p': 0.6098843322818086, 'r': 0.695026962252846, 'f1': 0.6496779613553626}, 'argument': {'p': 0.7182761838136468, 'r': 0.7101585893192163, 'f1': 0.7141943209908744}, 'combined': 0.4639963104729288, 'epoch': 37} Dev Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6586894230381533, 'r': 0.6657771858398551, 'f1': 0.6622143396794563}, 'combined': 0.435992049726847, 'epoch': 37} Test Korean: {'event': {'p': 0.605815061963775, 'r': 0.7615338526063511, 'f1': 0.6748075391558269}, 'argument': {'p': 0.7170915247015461, 'r': 0.7216141762238405, 'f1': 0.7193457418591828}, 'combined': 0.48541992986621785, 'epoch': 37} Dev Russian: {'event': {'p': 0.6528436018957346, 'r': 0.7336884154460719, 'f1': 0.6909090909090909}, 'argument': {'p': 0.6694168528800274, 'r': 0.6663923791471358, 'f1': 0.6679011920810538}, 'combined': 0.461459005437819, 'epoch': 37} Test Russian: {'event': {'p': 0.6488749345892203, 'r': 0.7429598562013181, 'f1': 0.6927374301675977}, 'argument': {'p': 0.7169875215279904, 'r': 0.7121030601048883, 'f1': 0.7145369435627033}, 'combined': 0.49498648604343687, 'epoch': 37} Chinese: {'event': {'p': 0.6825396825396826, 'r': 0.7962962962962963, 'f1': 0.735042735042735}, 'argument': {'p': 0.6756258234519104, 'r': 0.6875838026280503, 'f1': 0.681552365762892}, 'combined': 0.5009701150052026, 'epoch': 37} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.5475570436507936, 'r': 0.44927757427757425, 'f1': 0.4935725463894478}, 'combined': 0.28759345222692234, 'epoch': 37} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.533530809639184, 'r': 0.6079769691237213, 'f1': 0.5683262972243481}, 'combined': 0.29228209571537905, 'epoch': 37} ================================================== Current best result: -------------------- Dev for Chinese: {'event': {'p': 0.6308243727598566, 'r': 0.7030625832223701, 'f1': 0.6649874055415615}, 'argument': {'p': 0.6788322431410229, 'r': 0.6497771665863487, 'f1': 0.6639870052194498}, 'combined': 0.44154299591419316, 'epoch': 36} Test for Chinese: {'event': {'p': 0.6125131440588854, 'r': 0.6980227681246255, 'f1': 0.6524782973956875}, 'argument': {'p': 0.7303126454040291, 'r': 0.7023771890224542, 'f1': 0.7160725659671817}, 'combined': 0.46722180865402785, 'epoch': 36} Chinese: {'event': {'p': 0.6984126984126984, 'r': 0.8148148148148148, 'f1': 0.7521367521367521}, 'argument': {'p': 0.7283362470862471, 'r': 0.6703271654599089, 'f1': 0.6981287529674627}, 'combined': 0.5250882928302283, 'epoch': 36} -------------------- Dev for Korean: {'event': {'p': 0.5715686274509804, 'r': 0.7762982689747004, 'f1': 0.6583850931677019}, 'argument': {'p': 0.6716276506286475, 'r': 0.6576454525897165, 'f1': 0.6645630144776751}, 'combined': 0.4375383822026929, 'epoch': 29} Test for Korean: {'event': {'p': 0.6072449952335558, 'r': 0.7633313361294188, 'f1': 0.67640031855588}, 'argument': {'p': 0.716767199678779, 'r': 0.7094763372478128, 'f1': 0.7131031332605374}, 'combined': 0.4823431865006237, 'epoch': 29} Korean: {'event': {'p': 0.578125, 'r': 0.5873015873015873, 'f1': 0.5826771653543307}, 'argument': {'p': 0.6237824675324675, 'r': 0.4536599763872491, 'f1': 0.5252904989747095}, 'combined': 0.3060747789301457, 'epoch': 29} -------------------- Dev for Russian: {'event': {'p': 0.6516587677725119, 'r': 0.7323568575233023, 'f1': 0.6896551724137931}, 'argument': {'p': 0.6690746367862752, 'r': 0.6640364241598725, 'f1': 0.6665460100410511}, 'combined': 0.45968690347658697, 'epoch': 16} Test for Russian: {'event': {'p': 0.6504447933019362, 'r': 0.7447573397243858, 'f1': 0.6944134078212291}, 'argument': {'p': 0.7205532683201716, 'r': 0.7065719975730321, 'f1': 0.7134941470412731}, 'combined': 0.49545990210743157, 'epoch': 16} Russian: {'event': {'p': 0.5294117647058824, 'r': 0.5, 'f1': 0.5142857142857143}, 'argument': {'p': 0.6021764497133955, 'r': 0.6862010706036367, 'f1': 0.6414488268686168}, 'combined': 0.3298879681038601, 'epoch': 16} ****************************** Epoch: 38 command: python train.py --model_name arg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --max_epoch 40 --p1_data_weight 0.4 --event_hidden_num 450 --role_hidden_num 350 --learning_rate 2e-4 2023-01-23 23:11:05.002816: step: 2/1541, loss: 0.0005544863524846733 2023-01-23 23:11:05.876626: step: 4/1541, loss: 0.040039874613285065 2023-01-23 23:11:06.752133: step: 6/1541, loss: 0.0032641286961734295 2023-01-23 23:11:07.670786: step: 8/1541, loss: 0.026470676064491272 2023-01-23 23:11:08.549877: step: 10/1541, loss: 0.033650048077106476 2023-01-23 23:11:09.427466: step: 12/1541, loss: 0.1155926063656807 2023-01-23 23:11:10.399501: step: 14/1541, loss: 0.0007065642857924104 2023-01-23 23:11:11.278035: step: 16/1541, loss: 0.20511825382709503 2023-01-23 23:11:12.169593: step: 18/1541, loss: 0.0021129222586750984 2023-01-23 23:11:13.055939: step: 20/1541, loss: 0.00011997626279480755 2023-01-23 23:11:13.959537: step: 22/1541, loss: 0.009408717975020409 2023-01-23 23:11:14.872908: step: 24/1541, loss: 0.002586783142760396 2023-01-23 23:11:15.763798: step: 26/1541, loss: 0.0001081264708773233 2023-01-23 23:11:16.669768: step: 28/1541, loss: 0.0005305563099682331 2023-01-23 23:11:17.542644: step: 30/1541, loss: 0.0008583274320699275 2023-01-23 23:11:18.420629: step: 32/1541, loss: 0.00018736663332674652 2023-01-23 23:11:19.349788: step: 34/1541, loss: 0.19743306934833527 2023-01-23 23:11:20.219856: step: 36/1541, loss: 0.0004255215171724558 2023-01-23 23:11:21.094655: step: 38/1541, loss: 0.0005818484933115542 2023-01-23 23:11:21.977475: step: 40/1541, loss: 4.5343636884354055e-05 2023-01-23 23:11:22.872885: step: 42/1541, loss: -1.8496313714422286e-05 2023-01-23 23:11:23.792820: step: 44/1541, loss: 0.00015411406639032066 2023-01-23 23:11:24.772917: step: 46/1541, loss: 0.0001438464387319982 2023-01-23 23:11:25.624808: step: 48/1541, loss: 0.3910493850708008 2023-01-23 23:11:26.498992: step: 50/1541, loss: 0.23116129636764526 2023-01-23 23:11:27.405349: step: 52/1541, loss: 2.3402786609949544e-05 2023-01-23 23:11:28.325169: step: 54/1541, loss: 0.01426095049828291 2023-01-23 23:11:29.208783: step: 56/1541, loss: 0.1299830973148346 2023-01-23 23:11:30.083201: step: 58/1541, loss: 0.001982170855626464 2023-01-23 23:11:31.030458: step: 60/1541, loss: 0.010530306026339531 2023-01-23 23:11:31.936119: step: 62/1541, loss: 0.007889796048402786 2023-01-23 23:11:32.782377: step: 64/1541, loss: 0.001466768211685121 2023-01-23 23:11:33.662995: step: 66/1541, loss: 0.003467484610155225 2023-01-23 23:11:34.572212: step: 68/1541, loss: 0.056701481342315674 2023-01-23 23:11:35.433632: step: 70/1541, loss: 0.03901088982820511 2023-01-23 23:11:36.320420: step: 72/1541, loss: 0.0007509809220209718 2023-01-23 23:11:37.172847: step: 74/1541, loss: 0.009768510237336159 2023-01-23 23:11:38.055514: step: 76/1541, loss: 0.30711737275123596 2023-01-23 23:11:38.949335: step: 78/1541, loss: -4.0061280742520466e-05 2023-01-23 23:11:39.840920: step: 80/1541, loss: 0.04991772398352623 2023-01-23 23:11:40.718599: step: 82/1541, loss: 8.626183262094855e-05 2023-01-23 23:11:41.624181: step: 84/1541, loss: 1.4076766092330217e-05 2023-01-23 23:11:42.515805: step: 86/1541, loss: 2.4476645194226876e-05 2023-01-23 23:11:43.416321: step: 88/1541, loss: 0.00017851445591077209 2023-01-23 23:11:44.290766: step: 90/1541, loss: 0.01702279970049858 2023-01-23 23:11:45.149181: step: 92/1541, loss: 0.02701621688902378 2023-01-23 23:11:46.080916: step: 94/1541, loss: 0.0003784179862122983 2023-01-23 23:11:46.983642: step: 96/1541, loss: 0.0012018728302791715 2023-01-23 23:11:47.918000: step: 98/1541, loss: 0.08076869696378708 2023-01-23 23:11:48.830238: step: 100/1541, loss: 3.8325186324073e-05 2023-01-23 23:11:49.737924: step: 102/1541, loss: 0.32812339067459106 2023-01-23 23:11:50.617236: step: 104/1541, loss: 2.861663597286679e-05 2023-01-23 23:11:51.537791: step: 106/1541, loss: 0.0018917287234216928 2023-01-23 23:11:52.446495: step: 108/1541, loss: 0.1151973158121109 2023-01-23 23:11:53.330512: step: 110/1541, loss: 0.0002677424345165491 2023-01-23 23:11:54.191579: step: 112/1541, loss: 0.5140216946601868 2023-01-23 23:11:55.124846: step: 114/1541, loss: 0.03831491991877556 2023-01-23 23:11:55.978385: step: 116/1541, loss: 0.00024621933698654175 2023-01-23 23:11:56.951280: step: 118/1541, loss: 0.0007653050124645233 2023-01-23 23:11:57.786003: step: 120/1541, loss: 0.00015197991160675883 2023-01-23 23:11:58.688173: step: 122/1541, loss: 6.592646241188049e-05 2023-01-23 23:11:59.577969: step: 124/1541, loss: 4.7389272367581725e-05 2023-01-23 23:12:00.508590: step: 126/1541, loss: 0.000500602531246841 2023-01-23 23:12:01.396655: step: 128/1541, loss: 0.07524003088474274 2023-01-23 23:12:02.324034: step: 130/1541, loss: 0.0016547442646697164 2023-01-23 23:12:03.144945: step: 132/1541, loss: 0.00010293178638676181 2023-01-23 23:12:04.036926: step: 134/1541, loss: 0.01898656040430069 2023-01-23 23:12:05.043551: step: 136/1541, loss: -6.901607412146404e-05 2023-01-23 23:12:05.891871: step: 138/1541, loss: 2.6988556783180684e-05 2023-01-23 23:12:06.803427: step: 140/1541, loss: 2.0375473468448035e-05 2023-01-23 23:12:07.687369: step: 142/1541, loss: 1.265704668185208e-05 2023-01-23 23:12:08.656498: step: 144/1541, loss: 0.5741065144538879 2023-01-23 23:12:09.546554: step: 146/1541, loss: 0.04397979751229286 2023-01-23 23:12:10.423695: step: 148/1541, loss: 5.656143912347034e-06 2023-01-23 23:12:11.304571: step: 150/1541, loss: 0.006420718040317297 2023-01-23 23:12:12.225775: step: 152/1541, loss: 0.17421825230121613 2023-01-23 23:12:13.101952: step: 154/1541, loss: 0.0735226571559906 2023-01-23 23:12:14.064795: step: 156/1541, loss: 4.04500933655072e-05 2023-01-23 23:12:14.925500: step: 158/1541, loss: 0.05688071995973587 2023-01-23 23:12:15.789240: step: 160/1541, loss: -1.4194295090419473e-06 2023-01-23 23:12:16.700511: step: 162/1541, loss: 0.027074065059423447 2023-01-23 23:12:17.623486: step: 164/1541, loss: 2.8236698199179955e-05 2023-01-23 23:12:18.458386: step: 166/1541, loss: 0.007379342336207628 2023-01-23 23:12:19.302854: step: 168/1541, loss: 0.0002773432061076164 2023-01-23 23:12:20.195901: step: 170/1541, loss: -2.5916100639733486e-05 2023-01-23 23:12:21.094979: step: 172/1541, loss: 0.00017196914996020496 2023-01-23 23:12:22.031994: step: 174/1541, loss: 3.973742059315555e-05 2023-01-23 23:12:22.912769: step: 176/1541, loss: 0.09586659073829651 2023-01-23 23:12:23.822639: step: 178/1541, loss: 0.00016952156147453934 2023-01-23 23:12:24.730336: step: 180/1541, loss: 1.908317244669888e-05 2023-01-23 23:12:25.596313: step: 182/1541, loss: 0.0005909662577323616 2023-01-23 23:12:26.492172: step: 184/1541, loss: 0.022042322903871536 2023-01-23 23:12:27.412020: step: 186/1541, loss: 0.028673987835645676 2023-01-23 23:12:28.346013: step: 188/1541, loss: 0.26163601875305176 2023-01-23 23:12:29.210851: step: 190/1541, loss: 0.10982796549797058 2023-01-23 23:12:30.067834: step: 192/1541, loss: 0.028025541454553604 2023-01-23 23:12:30.998517: step: 194/1541, loss: 0.0007866323576308787 2023-01-23 23:12:31.895136: step: 196/1541, loss: 5.033349953009747e-05 2023-01-23 23:12:32.781772: step: 198/1541, loss: 2.67393033936969e-06 2023-01-23 23:12:33.682106: step: 200/1541, loss: 0.0021134314592927694 2023-01-23 23:12:34.545861: step: 202/1541, loss: 0.00030147895449772477 2023-01-23 23:12:35.447041: step: 204/1541, loss: -5.721524212276563e-05 2023-01-23 23:12:36.345684: step: 206/1541, loss: 0.09911556541919708 2023-01-23 23:12:37.245159: step: 208/1541, loss: 4.735873153549619e-05 2023-01-23 23:12:38.129211: step: 210/1541, loss: 0.0024888929910957813 2023-01-23 23:12:39.026988: step: 212/1541, loss: 2.2811920642852783 2023-01-23 23:12:39.928831: step: 214/1541, loss: 0.0002567841438576579 2023-01-23 23:12:40.822042: step: 216/1541, loss: 0.003165594534948468 2023-01-23 23:12:41.708715: step: 218/1541, loss: 0.32438933849334717 2023-01-23 23:12:42.639054: step: 220/1541, loss: 2.4134911654982716e-05 2023-01-23 23:12:43.547512: step: 222/1541, loss: 0.0001125418275478296 2023-01-23 23:12:44.441035: step: 224/1541, loss: 0.00027855171356350183 2023-01-23 23:12:45.287701: step: 226/1541, loss: 5.399331348598935e-05 2023-01-23 23:12:46.168162: step: 228/1541, loss: 0.0009783577406778932 2023-01-23 23:12:47.070636: step: 230/1541, loss: -6.482502158178249e-06 2023-01-23 23:12:47.957324: step: 232/1541, loss: 0.00015668511332478374 2023-01-23 23:12:48.822539: step: 234/1541, loss: 2.9950588213978335e-05 2023-01-23 23:12:49.705452: step: 236/1541, loss: 0.0016761244041845202 2023-01-23 23:12:50.597451: step: 238/1541, loss: 5.3862222557654604e-05 2023-01-23 23:12:51.471462: step: 240/1541, loss: 0.050658468157052994 2023-01-23 23:12:52.401819: step: 242/1541, loss: 0.0005996721447445452 2023-01-23 23:12:53.355368: step: 244/1541, loss: 0.000161349555128254 2023-01-23 23:12:54.335481: step: 246/1541, loss: 0.00039484960143454373 2023-01-23 23:12:55.208929: step: 248/1541, loss: 0.0011442741379141808 2023-01-23 23:12:56.110530: step: 250/1541, loss: 8.701081242179498e-05 2023-01-23 23:12:56.996891: step: 252/1541, loss: 0.00034750430495478213 2023-01-23 23:12:57.969941: step: 254/1541, loss: 0.0658249706029892 2023-01-23 23:12:58.924749: step: 256/1541, loss: 0.0001604709104867652 2023-01-23 23:12:59.855240: step: 258/1541, loss: 0.0793047770857811 2023-01-23 23:13:00.757069: step: 260/1541, loss: 0.00017242974718101323 2023-01-23 23:13:01.627296: step: 262/1541, loss: 0.02355083078145981 2023-01-23 23:13:02.534759: step: 264/1541, loss: 0.00011256982543272898 2023-01-23 23:13:03.443017: step: 266/1541, loss: 0.0004313853569328785 2023-01-23 23:13:04.426479: step: 268/1541, loss: 0.0015051993541419506 2023-01-23 23:13:05.316122: step: 270/1541, loss: 0.1379927545785904 2023-01-23 23:13:06.274999: step: 272/1541, loss: 0.17928878962993622 2023-01-23 23:13:07.182679: step: 274/1541, loss: 0.00019007486116606742 2023-01-23 23:13:08.084744: step: 276/1541, loss: 0.18364080786705017 2023-01-23 23:13:08.994312: step: 278/1541, loss: 0.061600878834724426 2023-01-23 23:13:09.871448: step: 280/1541, loss: 4.2485979065531865e-05 2023-01-23 23:13:10.766748: step: 282/1541, loss: 0.013138361275196075 2023-01-23 23:13:11.646107: step: 284/1541, loss: 0.001626731944270432 2023-01-23 23:13:12.614171: step: 286/1541, loss: 0.002317843958735466 2023-01-23 23:13:13.507672: step: 288/1541, loss: 0.003885416081175208 2023-01-23 23:13:14.391119: step: 290/1541, loss: 0.03895903006196022 2023-01-23 23:13:15.261136: step: 292/1541, loss: 0.049289677292108536 2023-01-23 23:13:16.190668: step: 294/1541, loss: 0.0016057523898780346 2023-01-23 23:13:17.082065: step: 296/1541, loss: 0.0015829206677153707 2023-01-23 23:13:17.995736: step: 298/1541, loss: 0.08072610199451447 2023-01-23 23:13:18.939629: step: 300/1541, loss: 0.00022966241522226483 2023-01-23 23:13:19.931897: step: 302/1541, loss: 0.0017461145762354136 2023-01-23 23:13:20.836560: step: 304/1541, loss: 0.05854072421789169 2023-01-23 23:13:21.690829: step: 306/1541, loss: 0.0028370185755193233 2023-01-23 23:13:22.548931: step: 308/1541, loss: 0.00034420337760820985 2023-01-23 23:13:23.468825: step: 310/1541, loss: 0.0006466877530328929 2023-01-23 23:13:24.370460: step: 312/1541, loss: 0.0018195451702922583 2023-01-23 23:13:25.286316: step: 314/1541, loss: 0.2934971749782562 2023-01-23 23:13:26.213394: step: 316/1541, loss: 0.0008774411398917437 2023-01-23 23:13:27.111122: step: 318/1541, loss: 0.01277634035795927 2023-01-23 23:13:28.002450: step: 320/1541, loss: 0.007637536618858576 2023-01-23 23:13:28.945739: step: 322/1541, loss: 0.000889140646904707 2023-01-23 23:13:29.836075: step: 324/1541, loss: 0.00030876448727212846 2023-01-23 23:13:30.758452: step: 326/1541, loss: 0.48471975326538086 2023-01-23 23:13:31.704041: step: 328/1541, loss: 0.18071293830871582 2023-01-23 23:13:32.627893: step: 330/1541, loss: 0.0007381379837170243 2023-01-23 23:13:33.496288: step: 332/1541, loss: 0.04278886690735817 2023-01-23 23:13:34.359083: step: 334/1541, loss: 0.0016482193022966385 2023-01-23 23:13:35.251863: step: 336/1541, loss: 0.024512501433491707 2023-01-23 23:13:36.106845: step: 338/1541, loss: 4.282287409296259e-05 2023-01-23 23:13:37.060921: step: 340/1541, loss: 0.042379386723041534 2023-01-23 23:13:37.950129: step: 342/1541, loss: 0.00013989555009175092 2023-01-23 23:13:38.860893: step: 344/1541, loss: 0.001277281902730465 2023-01-23 23:13:39.759503: step: 346/1541, loss: 0.00020280400349292904 2023-01-23 23:13:40.677192: step: 348/1541, loss: 0.0014023075345903635 2023-01-23 23:13:41.574186: step: 350/1541, loss: 0.0003403087321203202 2023-01-23 23:13:42.447574: step: 352/1541, loss: 0.002532664919272065 2023-01-23 23:13:43.346455: step: 354/1541, loss: 0.05885617062449455 2023-01-23 23:13:44.335813: step: 356/1541, loss: 0.4930408298969269 2023-01-23 23:13:45.350763: step: 358/1541, loss: 0.025741057470440865 2023-01-23 23:13:46.337217: step: 360/1541, loss: 8.99334263522178e-05 2023-01-23 23:13:47.260499: step: 362/1541, loss: 8.419672667514533e-05 2023-01-23 23:13:48.221791: step: 364/1541, loss: 2.1848805772606283e-05 2023-01-23 23:13:49.140717: step: 366/1541, loss: 0.2719789445400238 2023-01-23 23:13:49.996503: step: 368/1541, loss: 0.0004527355486061424 2023-01-23 23:13:50.915101: step: 370/1541, loss: 0.0007036934257484972 2023-01-23 23:13:51.775738: step: 372/1541, loss: 5.483795121108415e-06 2023-01-23 23:13:52.678882: step: 374/1541, loss: 0.03519265353679657 2023-01-23 23:13:53.630337: step: 376/1541, loss: 0.0016006302321329713 2023-01-23 23:13:54.524122: step: 378/1541, loss: 7.948267011670396e-06 2023-01-23 23:13:55.450702: step: 380/1541, loss: -5.6213790230685845e-05 2023-01-23 23:13:56.466179: step: 382/1541, loss: 0.016003452241420746 2023-01-23 23:13:57.395196: step: 384/1541, loss: 6.449065404012799e-05 2023-01-23 23:13:58.322309: step: 386/1541, loss: 0.0060904850251972675 2023-01-23 23:13:59.222503: step: 388/1541, loss: 8.972032810561359e-05 2023-01-23 23:14:00.157959: step: 390/1541, loss: 0.01988106779754162 2023-01-23 23:14:01.039403: step: 392/1541, loss: 0.00028624129481613636 2023-01-23 23:14:01.965451: step: 394/1541, loss: 6.559677422046661e-05 2023-01-23 23:14:02.880955: step: 396/1541, loss: 0.0001922607480082661 2023-01-23 23:14:03.758231: step: 398/1541, loss: 0.00027476309332996607 2023-01-23 23:14:04.648426: step: 400/1541, loss: 0.10545522719621658 2023-01-23 23:14:05.559617: step: 402/1541, loss: 0.00013293146912474185 2023-01-23 23:14:06.503956: step: 404/1541, loss: 0.24241986870765686 2023-01-23 23:14:07.407865: step: 406/1541, loss: 0.03842881694436073 2023-01-23 23:14:08.325681: step: 408/1541, loss: 0.00150984525680542 2023-01-23 23:14:09.224501: step: 410/1541, loss: 5.2748619054909796e-05 2023-01-23 23:14:10.138037: step: 412/1541, loss: 0.0026180774439126253 2023-01-23 23:14:11.036007: step: 414/1541, loss: 6.926179048605263e-05 2023-01-23 23:14:11.904318: step: 416/1541, loss: 0.003249014727771282 2023-01-23 23:14:12.785562: step: 418/1541, loss: 0.0029324726201593876 2023-01-23 23:14:13.634156: step: 420/1541, loss: 0.21494251489639282 2023-01-23 23:14:14.497246: step: 422/1541, loss: 0.0007621715776622295 2023-01-23 23:14:15.381903: step: 424/1541, loss: 0.00011310048284940422 2023-01-23 23:14:16.321133: step: 426/1541, loss: 0.005459685809910297 2023-01-23 23:14:17.174725: step: 428/1541, loss: 0.08789266645908356 2023-01-23 23:14:18.043895: step: 430/1541, loss: 0.0024748491123318672 2023-01-23 23:14:19.037318: step: 432/1541, loss: 0.0329325869679451 2023-01-23 23:14:19.927613: step: 434/1541, loss: 0.00012013117520837113 2023-01-23 23:14:20.843569: step: 436/1541, loss: 0.04840170592069626 2023-01-23 23:14:21.730288: step: 438/1541, loss: 0.0434127114713192 2023-01-23 23:14:22.678517: step: 440/1541, loss: 2.140775541192852e-05 2023-01-23 23:14:23.581624: step: 442/1541, loss: 0.00033162979525513947 2023-01-23 23:14:24.517838: step: 444/1541, loss: 0.0021151273977011442 2023-01-23 23:14:25.414852: step: 446/1541, loss: 0.000915513897780329 2023-01-23 23:14:26.310580: step: 448/1541, loss: 0.0006853099330328405 2023-01-23 23:14:27.220254: step: 450/1541, loss: 0.005777362734079361 2023-01-23 23:14:28.122488: step: 452/1541, loss: 0.00038038159254938364 2023-01-23 23:14:29.014835: step: 454/1541, loss: 1.8896642359322868e-05 2023-01-23 23:14:29.895221: step: 456/1541, loss: 3.79379798687296e-06 2023-01-23 23:14:30.811446: step: 458/1541, loss: 4.929243368678726e-05 2023-01-23 23:14:31.679433: step: 460/1541, loss: 0.04692414030432701 2023-01-23 23:14:32.582704: step: 462/1541, loss: 0.018455706536769867 2023-01-23 23:14:33.460938: step: 464/1541, loss: 0.7305776476860046 2023-01-23 23:14:34.335453: step: 466/1541, loss: 0.0003853917005471885 2023-01-23 23:14:35.243889: step: 468/1541, loss: 0.0006027668714523315 2023-01-23 23:14:36.196107: step: 470/1541, loss: 2.594262514321599e-05 2023-01-23 23:14:37.167478: step: 472/1541, loss: 0.0007599177770316601 2023-01-23 23:14:38.084868: step: 474/1541, loss: 0.0001285607722820714 2023-01-23 23:14:38.976280: step: 476/1541, loss: 0.09989594668149948 2023-01-23 23:14:39.860086: step: 478/1541, loss: 0.0026462774258106947 2023-01-23 23:14:40.807466: step: 480/1541, loss: 0.0002747985708992928 2023-01-23 23:14:41.711842: step: 482/1541, loss: 0.008307092823088169 2023-01-23 23:14:42.609829: step: 484/1541, loss: 0.01995166763663292 2023-01-23 23:14:43.582777: step: 486/1541, loss: 0.22063356637954712 2023-01-23 23:14:44.510237: step: 488/1541, loss: 0.5589557886123657 2023-01-23 23:14:45.498788: step: 490/1541, loss: 0.0014278338057920337 2023-01-23 23:14:46.422222: step: 492/1541, loss: 5.95729798078537e-05 2023-01-23 23:14:47.294929: step: 494/1541, loss: 0.007641630247235298 2023-01-23 23:14:48.169519: step: 496/1541, loss: 3.584076330298558e-05 2023-01-23 23:14:49.063106: step: 498/1541, loss: 0.0002761870564427227 2023-01-23 23:14:49.971240: step: 500/1541, loss: 0.05295497551560402 2023-01-23 23:14:50.863139: step: 502/1541, loss: 0.00023490190505981445 2023-01-23 23:14:51.832776: step: 504/1541, loss: 0.11446351557970047 2023-01-23 23:14:52.708111: step: 506/1541, loss: 0.3269645869731903 2023-01-23 23:14:53.583006: step: 508/1541, loss: 0.02537771873176098 2023-01-23 23:14:54.455983: step: 510/1541, loss: 0.0033695115707814693 2023-01-23 23:14:55.406042: step: 512/1541, loss: 0.04498519003391266 2023-01-23 23:14:56.308593: step: 514/1541, loss: 8.410136615566444e-06 2023-01-23 23:14:57.217324: step: 516/1541, loss: 0.0004543459217529744 2023-01-23 23:14:58.113845: step: 518/1541, loss: 0.000814757018815726 2023-01-23 23:14:59.018903: step: 520/1541, loss: 3.206953260814771e-05 2023-01-23 23:14:59.917717: step: 522/1541, loss: 0.0072860075160861015 2023-01-23 23:15:00.850796: step: 524/1541, loss: 0.00045909936306998134 2023-01-23 23:15:01.707842: step: 526/1541, loss: 0.002234324347227812 2023-01-23 23:15:02.621701: step: 528/1541, loss: -1.6286971003864892e-05 2023-01-23 23:15:03.499837: step: 530/1541, loss: 0.0007006334490142763 2023-01-23 23:15:04.373339: step: 532/1541, loss: 0.0052201515063643456 2023-01-23 23:15:05.298797: step: 534/1541, loss: 0.0006389807676896453 2023-01-23 23:15:06.208669: step: 536/1541, loss: 0.0068031917326152325 2023-01-23 23:15:07.072219: step: 538/1541, loss: 0.0006900587468408048 2023-01-23 23:15:07.989813: step: 540/1541, loss: 0.018848346546292305 2023-01-23 23:15:08.964170: step: 542/1541, loss: 0.00011854934564325958 2023-01-23 23:15:09.857393: step: 544/1541, loss: 0.486664742231369 2023-01-23 23:15:10.800520: step: 546/1541, loss: 6.218946509761736e-05 2023-01-23 23:15:11.674757: step: 548/1541, loss: 0.030545759946107864 2023-01-23 23:15:12.559850: step: 550/1541, loss: 9.844004671322182e-05 2023-01-23 23:15:13.448375: step: 552/1541, loss: 0.03842518851161003 2023-01-23 23:15:14.376355: step: 554/1541, loss: 0.14612871408462524 2023-01-23 23:15:15.267553: step: 556/1541, loss: -7.719400855421554e-06 2023-01-23 23:15:16.121297: step: 558/1541, loss: 4.510083090281114e-05 2023-01-23 23:15:17.016809: step: 560/1541, loss: 0.03792256489396095 2023-01-23 23:15:17.911833: step: 562/1541, loss: 0.0130156883969903 2023-01-23 23:15:18.780312: step: 564/1541, loss: 0.0025280064437538385 2023-01-23 23:15:19.672342: step: 566/1541, loss: 0.000608226575423032 2023-01-23 23:15:20.542590: step: 568/1541, loss: 7.254481170093641e-05 2023-01-23 23:15:21.496604: step: 570/1541, loss: 0.0013363019097596407 2023-01-23 23:15:22.420801: step: 572/1541, loss: 0.000759418006055057 2023-01-23 23:15:23.281016: step: 574/1541, loss: 0.0003418380510993302 2023-01-23 23:15:24.144295: step: 576/1541, loss: 0.012167001143097878 2023-01-23 23:15:25.042716: step: 578/1541, loss: 0.00029422951047308743 2023-01-23 23:15:25.901671: step: 580/1541, loss: 0.18988735973834991 2023-01-23 23:15:26.812653: step: 582/1541, loss: 0.00044145379797555506 2023-01-23 23:15:27.769558: step: 584/1541, loss: 0.0008114724769257009 2023-01-23 23:15:28.643933: step: 586/1541, loss: 0.33428630232810974 2023-01-23 23:15:29.514039: step: 588/1541, loss: 8.872738544596359e-05 2023-01-23 23:15:30.436838: step: 590/1541, loss: 0.00044147251173853874 2023-01-23 23:15:31.320684: step: 592/1541, loss: 5.5586762755410746e-05 2023-01-23 23:15:32.237484: step: 594/1541, loss: 5.270115070743486e-05 2023-01-23 23:15:33.161828: step: 596/1541, loss: 2.4155164283001795e-05 2023-01-23 23:15:34.073598: step: 598/1541, loss: 0.07448158413171768 2023-01-23 23:15:35.003150: step: 600/1541, loss: 0.00014948827447369695 2023-01-23 23:15:35.878028: step: 602/1541, loss: 0.2435876429080963 2023-01-23 23:15:36.746626: step: 604/1541, loss: 1.3767086267471313 2023-01-23 23:15:37.648516: step: 606/1541, loss: 0.015136688016355038 2023-01-23 23:15:38.585676: step: 608/1541, loss: 0.031104108318686485 2023-01-23 23:15:39.442646: step: 610/1541, loss: 0.0008225251222029328 2023-01-23 23:15:40.365833: step: 612/1541, loss: 8.92504831426777e-05 2023-01-23 23:15:41.239369: step: 614/1541, loss: 0.0033918514382094145 2023-01-23 23:15:42.133281: step: 616/1541, loss: 0.00023564726870972663 2023-01-23 23:15:43.059897: step: 618/1541, loss: 7.309151988010854e-05 2023-01-23 23:15:43.949934: step: 620/1541, loss: 0.00017417859635315835 2023-01-23 23:15:44.815001: step: 622/1541, loss: 0.00010317464330000803 2023-01-23 23:15:45.718915: step: 624/1541, loss: 0.00033355195773765445 2023-01-23 23:15:46.592655: step: 626/1541, loss: 0.09730972349643707 2023-01-23 23:15:47.499615: step: 628/1541, loss: 3.8817684981040657e-05 2023-01-23 23:15:48.446533: step: 630/1541, loss: 0.112160325050354 2023-01-23 23:15:49.327200: step: 632/1541, loss: 0.002013968303799629 2023-01-23 23:15:50.188030: step: 634/1541, loss: 0.0016782886814326048 2023-01-23 23:15:51.076651: step: 636/1541, loss: -4.0350960262003355e-07 2023-01-23 23:15:52.018525: step: 638/1541, loss: 0.0002916883095167577 2023-01-23 23:15:52.937390: step: 640/1541, loss: 7.097671914380044e-05 2023-01-23 23:15:53.820873: step: 642/1541, loss: 0.01087779551744461 2023-01-23 23:15:54.689803: step: 644/1541, loss: 0.01652814820408821 2023-01-23 23:15:55.628445: step: 646/1541, loss: 0.00022676686057820916 2023-01-23 23:15:56.553722: step: 648/1541, loss: 0.0001337620778940618 2023-01-23 23:15:57.492866: step: 650/1541, loss: 0.0030461284331977367 2023-01-23 23:15:58.386416: step: 652/1541, loss: 0.0003442864981479943 2023-01-23 23:15:59.233759: step: 654/1541, loss: 0.00015716595225967467 2023-01-23 23:16:00.216132: step: 656/1541, loss: 0.0081619992852211 2023-01-23 23:16:01.163091: step: 658/1541, loss: 0.0001193522839457728 2023-01-23 23:16:02.081147: step: 660/1541, loss: 0.0704255923628807 2023-01-23 23:16:02.964162: step: 662/1541, loss: 0.0002281526249134913 2023-01-23 23:16:03.867480: step: 664/1541, loss: 0.0001992532197618857 2023-01-23 23:16:04.787668: step: 666/1541, loss: 0.0026043264660984278 2023-01-23 23:16:05.768230: step: 668/1541, loss: 0.0029634935781359673 2023-01-23 23:16:06.642022: step: 670/1541, loss: 5.307197352522053e-05 2023-01-23 23:16:07.493605: step: 672/1541, loss: -3.634362656157464e-05 2023-01-23 23:16:08.345234: step: 674/1541, loss: 0.0006274566985666752 2023-01-23 23:16:09.246351: step: 676/1541, loss: 0.08987633138895035 2023-01-23 23:16:10.162576: step: 678/1541, loss: 0.12128796428442001 2023-01-23 23:16:11.062696: step: 680/1541, loss: 9.038857388077304e-05 2023-01-23 23:16:11.990255: step: 682/1541, loss: 0.14608332514762878 2023-01-23 23:16:12.887323: step: 684/1541, loss: 0.12407292425632477 2023-01-23 23:16:13.781651: step: 686/1541, loss: -7.904822268756106e-06 2023-01-23 23:16:14.766997: step: 688/1541, loss: 9.166121162706986e-05 2023-01-23 23:16:15.660434: step: 690/1541, loss: 0.09150836616754532 2023-01-23 23:16:16.558465: step: 692/1541, loss: 0.027956441044807434 2023-01-23 23:16:17.444792: step: 694/1541, loss: 0.0029485770501196384 2023-01-23 23:16:18.440511: step: 696/1541, loss: 1.66605806350708 2023-01-23 23:16:19.321438: step: 698/1541, loss: 0.015344574116170406 2023-01-23 23:16:20.213623: step: 700/1541, loss: 0.00017706098151393235 2023-01-23 23:16:21.076814: step: 702/1541, loss: 0.00018505280604586005 2023-01-23 23:16:21.992926: step: 704/1541, loss: 0.0005256551085039973 2023-01-23 23:16:22.874921: step: 706/1541, loss: 0.012077228166162968 2023-01-23 23:16:23.736148: step: 708/1541, loss: 7.285841275006533e-05 2023-01-23 23:16:24.604772: step: 710/1541, loss: 0.00017534842481836677 2023-01-23 23:16:25.487651: step: 712/1541, loss: 0.10197419673204422 2023-01-23 23:16:26.401613: step: 714/1541, loss: 1.6093274552986259e-06 2023-01-23 23:16:27.305161: step: 716/1541, loss: 1.4550240848620888e-05 2023-01-23 23:16:28.218783: step: 718/1541, loss: 0.0006132276612333953 2023-01-23 23:16:29.079113: step: 720/1541, loss: 0.00018491892842575908 2023-01-23 23:16:29.955689: step: 722/1541, loss: 3.587575338315219e-05 2023-01-23 23:16:30.855396: step: 724/1541, loss: 0.00014000652299728245 2023-01-23 23:16:31.744838: step: 726/1541, loss: 0.0009752705809660256 2023-01-23 23:16:32.635447: step: 728/1541, loss: 0.011991827748715878 2023-01-23 23:16:33.486556: step: 730/1541, loss: 6.334712088573724e-05 2023-01-23 23:16:34.355315: step: 732/1541, loss: 0.00012996252917218953 2023-01-23 23:16:35.289582: step: 734/1541, loss: 0.00021306757116690278 2023-01-23 23:16:36.178550: step: 736/1541, loss: 0.0018258296186104417 2023-01-23 23:16:37.095295: step: 738/1541, loss: 0.0991835743188858 2023-01-23 23:16:38.013809: step: 740/1541, loss: 0.0615103617310524 2023-01-23 23:16:38.899468: step: 742/1541, loss: 0.00024536438286304474 2023-01-23 23:16:39.814861: step: 744/1541, loss: 2.524857882235665e-05 2023-01-23 23:16:40.730460: step: 746/1541, loss: 1.3352433597901836e-05 2023-01-23 23:16:41.638283: step: 748/1541, loss: 0.02514168620109558 2023-01-23 23:16:42.528898: step: 750/1541, loss: 0.0002656400320120156 2023-01-23 23:16:43.384878: step: 752/1541, loss: 0.0014125467278063297 2023-01-23 23:16:44.280588: step: 754/1541, loss: 0.009655091911554337 2023-01-23 23:16:45.212676: step: 756/1541, loss: 0.0006083173793740571 2023-01-23 23:16:46.194536: step: 758/1541, loss: 0.01242858823388815 2023-01-23 23:16:47.080428: step: 760/1541, loss: 4.431799607118592e-05 2023-01-23 23:16:47.953829: step: 762/1541, loss: 0.00010694344382500276 2023-01-23 23:16:48.858262: step: 764/1541, loss: 0.0003201753133907914 2023-01-23 23:16:49.776970: step: 766/1541, loss: 0.00011943660501856357 2023-01-23 23:16:50.687247: step: 768/1541, loss: 0.006668989080935717 2023-01-23 23:16:51.615577: step: 770/1541, loss: 0.0004893724108114839 2023-01-23 23:16:52.482412: step: 772/1541, loss: 0.018146909773349762 2023-01-23 23:16:53.397040: step: 774/1541, loss: 0.0003856904513668269 2023-01-23 23:16:54.300967: step: 776/1541, loss: 2.1981759346090257e-05 2023-01-23 23:16:55.190854: step: 778/1541, loss: 0.20149879157543182 2023-01-23 23:16:56.064942: step: 780/1541, loss: 0.0010316523257642984 2023-01-23 23:16:56.968700: step: 782/1541, loss: 0.14563493430614471 2023-01-23 23:16:57.850315: step: 784/1541, loss: 0.029897868633270264 2023-01-23 23:16:58.779831: step: 786/1541, loss: 0.027396397665143013 2023-01-23 23:16:59.657534: step: 788/1541, loss: 7.631481275893748e-05 2023-01-23 23:17:00.536845: step: 790/1541, loss: 0.005540752783417702 2023-01-23 23:17:01.444496: step: 792/1541, loss: 0.004279549233615398 2023-01-23 23:17:02.339766: step: 794/1541, loss: 0.0789875015616417 2023-01-23 23:17:03.217771: step: 796/1541, loss: 0.0013157713692635298 2023-01-23 23:17:04.125825: step: 798/1541, loss: 0.008133446797728539 2023-01-23 23:17:04.989258: step: 800/1541, loss: 0.22786380350589752 2023-01-23 23:17:05.912983: step: 802/1541, loss: 0.00020455275080166757 2023-01-23 23:17:06.815869: step: 804/1541, loss: 4.0878047002479434e-05 2023-01-23 23:17:07.732034: step: 806/1541, loss: 0.00045387327554635704 2023-01-23 23:17:08.652435: step: 808/1541, loss: -1.720918226055801e-05 2023-01-23 23:17:09.526617: step: 810/1541, loss: 7.088034180924296e-05 2023-01-23 23:17:10.505689: step: 812/1541, loss: 0.00965604092925787 2023-01-23 23:17:11.390957: step: 814/1541, loss: 0.004952907096594572 2023-01-23 23:17:12.296487: step: 816/1541, loss: 0.046203091740608215 2023-01-23 23:17:13.164623: step: 818/1541, loss: 3.356634988449514e-05 2023-01-23 23:17:14.068614: step: 820/1541, loss: 0.008286496624350548 2023-01-23 23:17:15.017000: step: 822/1541, loss: 0.006829055026173592 2023-01-23 23:17:15.854872: step: 824/1541, loss: 0.00018795434152707458 2023-01-23 23:17:16.787562: step: 826/1541, loss: 0.005884368903934956 2023-01-23 23:17:17.732972: step: 828/1541, loss: 0.00015561631880700588 2023-01-23 23:17:18.594038: step: 830/1541, loss: 4.361833998700604e-05 2023-01-23 23:17:19.518691: step: 832/1541, loss: 0.014432844705879688 2023-01-23 23:17:20.428034: step: 834/1541, loss: 0.050536125898361206 2023-01-23 23:17:21.331689: step: 836/1541, loss: 8.059052197495475e-05 2023-01-23 23:17:22.229938: step: 838/1541, loss: 2.378495628363453e-05 2023-01-23 23:17:23.108840: step: 840/1541, loss: 0.001071159727871418 2023-01-23 23:17:24.028629: step: 842/1541, loss: 0.0012079065199941397 2023-01-23 23:17:24.925378: step: 844/1541, loss: 0.00038990957546047866 2023-01-23 23:17:25.786600: step: 846/1541, loss: 0.00017202751769218594 2023-01-23 23:17:26.698601: step: 848/1541, loss: 1.8518854631111026e-05 2023-01-23 23:17:27.579528: step: 850/1541, loss: 0.002532223705202341 2023-01-23 23:17:28.500112: step: 852/1541, loss: 0.00020934551139362156 2023-01-23 23:17:29.419038: step: 854/1541, loss: 3.764976645470597e-05 2023-01-23 23:17:30.285063: step: 856/1541, loss: 0.00011086212180089206 2023-01-23 23:17:31.169690: step: 858/1541, loss: 0.9113470911979675 2023-01-23 23:17:32.083527: step: 860/1541, loss: 0.031245633959770203 2023-01-23 23:17:33.041844: step: 862/1541, loss: 5.181422602618113e-05 2023-01-23 23:17:33.962864: step: 864/1541, loss: 0.001941016293130815 2023-01-23 23:17:34.812900: step: 866/1541, loss: 2.8833806936745532e-05 2023-01-23 23:17:35.701228: step: 868/1541, loss: 0.03239027038216591 2023-01-23 23:17:36.612575: step: 870/1541, loss: 0.001935672014951706 2023-01-23 23:17:37.541613: step: 872/1541, loss: 0.0008383505628444254 2023-01-23 23:17:38.478825: step: 874/1541, loss: 6.230014696484432e-05 2023-01-23 23:17:39.374397: step: 876/1541, loss: 0.0016525952378287911 2023-01-23 23:17:40.262290: step: 878/1541, loss: 0.0003154992882627994 2023-01-23 23:17:41.229259: step: 880/1541, loss: 0.02105969376862049 2023-01-23 23:17:42.108418: step: 882/1541, loss: 0.48261716961860657 2023-01-23 23:17:43.037348: step: 884/1541, loss: 0.011179283261299133 2023-01-23 23:17:43.930328: step: 886/1541, loss: 0.333313912153244 2023-01-23 23:17:44.779345: step: 888/1541, loss: 0.1223953515291214 2023-01-23 23:17:45.668906: step: 890/1541, loss: 0.003176187863573432 2023-01-23 23:17:46.527109: step: 892/1541, loss: 5.191990567254834e-06 2023-01-23 23:17:47.451222: step: 894/1541, loss: 9.637480252422392e-05 2023-01-23 23:17:48.312022: step: 896/1541, loss: 0.00019671424524858594 2023-01-23 23:17:49.232018: step: 898/1541, loss: 0.00023475180205423385 2023-01-23 23:17:50.083871: step: 900/1541, loss: 1.1133657608297653e-05 2023-01-23 23:17:50.971434: step: 902/1541, loss: 0.00029061242821626365 2023-01-23 23:17:51.857343: step: 904/1541, loss: 0.0003159618645440787 2023-01-23 23:17:52.733284: step: 906/1541, loss: 0.0007865946390666068 2023-01-23 23:17:53.604585: step: 908/1541, loss: 0.0003303836565464735 2023-01-23 23:17:54.475759: step: 910/1541, loss: 0.08154893666505814 2023-01-23 23:17:55.397395: step: 912/1541, loss: 0.0006738010561093688 2023-01-23 23:17:56.273289: step: 914/1541, loss: 0.32589349150657654 2023-01-23 23:17:57.130981: step: 916/1541, loss: 0.006795390043407679 2023-01-23 23:17:58.017650: step: 918/1541, loss: 0.0005709499237127602 2023-01-23 23:17:58.912260: step: 920/1541, loss: 0.002857001032680273 2023-01-23 23:17:59.792819: step: 922/1541, loss: 0.18801742792129517 2023-01-23 23:18:00.650752: step: 924/1541, loss: 0.0010698109399527311 2023-01-23 23:18:01.525476: step: 926/1541, loss: 0.0005794886383228004 2023-01-23 23:18:02.435892: step: 928/1541, loss: 0.04505966603755951 2023-01-23 23:18:03.400550: step: 930/1541, loss: 7.33568231225945e-05 2023-01-23 23:18:04.291276: step: 932/1541, loss: 0.008093195967376232 2023-01-23 23:18:05.198087: step: 934/1541, loss: 5.0551963795442134e-05 2023-01-23 23:18:06.088634: step: 936/1541, loss: 7.252175419125706e-05 2023-01-23 23:18:06.989298: step: 938/1541, loss: 1.7821150322561152e-05 2023-01-23 23:18:07.867473: step: 940/1541, loss: 0.08307071030139923 2023-01-23 23:18:08.724637: step: 942/1541, loss: 0.396837055683136 2023-01-23 23:18:09.722277: step: 944/1541, loss: 0.009642388671636581 2023-01-23 23:18:10.610434: step: 946/1541, loss: 0.0006082613253965974 2023-01-23 23:18:11.495811: step: 948/1541, loss: 0.001166454516351223 2023-01-23 23:18:12.409143: step: 950/1541, loss: 0.0001578511146362871 2023-01-23 23:18:13.279574: step: 952/1541, loss: 0.020803265273571014 2023-01-23 23:18:14.177229: step: 954/1541, loss: 0.0025673110503703356 2023-01-23 23:18:15.076567: step: 956/1541, loss: 6.72742462484166e-05 2023-01-23 23:18:15.929331: step: 958/1541, loss: 0.0009558533201925457 2023-01-23 23:18:16.840688: step: 960/1541, loss: 3.073730230331421 2023-01-23 23:18:17.699552: step: 962/1541, loss: -2.0414592654560693e-05 2023-01-23 23:18:18.594392: step: 964/1541, loss: 9.38809389481321e-05 2023-01-23 23:18:19.541203: step: 966/1541, loss: 8.480449469061568e-05 2023-01-23 23:18:20.468166: step: 968/1541, loss: 0.012475418858230114 2023-01-23 23:18:21.384264: step: 970/1541, loss: 0.019863717257976532 2023-01-23 23:18:22.302835: step: 972/1541, loss: 0.027739210054278374 2023-01-23 23:18:23.155138: step: 974/1541, loss: 0.031500622630119324 2023-01-23 23:18:24.047395: step: 976/1541, loss: 0.00022089743288233876 2023-01-23 23:18:24.941400: step: 978/1541, loss: 0.7608073949813843 2023-01-23 23:18:25.844505: step: 980/1541, loss: 0.0019286054885014892 2023-01-23 23:18:26.715583: step: 982/1541, loss: 1.1207067473151255e-05 2023-01-23 23:18:27.610021: step: 984/1541, loss: 0.008746681734919548 2023-01-23 23:18:28.465521: step: 986/1541, loss: 0.00010717643453972414 2023-01-23 23:18:29.364531: step: 988/1541, loss: 0.00035262256278656423 2023-01-23 23:18:30.336961: step: 990/1541, loss: 9.666756341175642e-06 2023-01-23 23:18:31.198556: step: 992/1541, loss: 0.001241197343915701 2023-01-23 23:18:32.057669: step: 994/1541, loss: 0.004075057804584503 2023-01-23 23:18:32.954252: step: 996/1541, loss: 8.42150802782271e-06 2023-01-23 23:18:33.826288: step: 998/1541, loss: 3.949765596189536e-05 2023-01-23 23:18:34.833301: step: 1000/1541, loss: 0.0001279302960028872 2023-01-23 23:18:35.715778: step: 1002/1541, loss: 0.0002472955093253404 2023-01-23 23:18:36.611055: step: 1004/1541, loss: 0.035564493387937546 2023-01-23 23:18:37.496559: step: 1006/1541, loss: 0.0008061391417868435 2023-01-23 23:18:38.363592: step: 1008/1541, loss: 0.36446231603622437 2023-01-23 23:18:39.241535: step: 1010/1541, loss: 0.0001439181505702436 2023-01-23 23:18:40.111023: step: 1012/1541, loss: 0.2322818636894226 2023-01-23 23:18:40.974582: step: 1014/1541, loss: 0.01515495777130127 2023-01-23 23:18:41.880800: step: 1016/1541, loss: 0.003936184570193291 2023-01-23 23:18:42.796546: step: 1018/1541, loss: 0.0002695493458304554 2023-01-23 23:18:43.670476: step: 1020/1541, loss: 3.2174779335036874e-05 2023-01-23 23:18:44.509829: step: 1022/1541, loss: 6.771409243810922e-05 2023-01-23 23:18:45.397984: step: 1024/1541, loss: 0.46710166335105896 2023-01-23 23:18:46.283818: step: 1026/1541, loss: 0.1007230207324028 2023-01-23 23:18:47.191127: step: 1028/1541, loss: 0.0004731813969556242 2023-01-23 23:18:48.084420: step: 1030/1541, loss: 4.682900180341676e-05 2023-01-23 23:18:48.959325: step: 1032/1541, loss: 0.03184763342142105 2023-01-23 23:18:49.847252: step: 1034/1541, loss: 0.0010945126414299011 2023-01-23 23:18:50.790055: step: 1036/1541, loss: 0.0003458150604274124 2023-01-23 23:18:51.762668: step: 1038/1541, loss: 0.00010977423517033458 2023-01-23 23:18:52.642767: step: 1040/1541, loss: 9.72366160567617e-06 2023-01-23 23:18:53.512190: step: 1042/1541, loss: 0.061515625566244125 2023-01-23 23:18:54.423138: step: 1044/1541, loss: 0.017198430374264717 2023-01-23 23:18:55.335181: step: 1046/1541, loss: 5.460007014335133e-05 2023-01-23 23:18:56.249974: step: 1048/1541, loss: 0.04043157771229744 2023-01-23 23:18:57.170277: step: 1050/1541, loss: 0.011033846065402031 2023-01-23 23:18:58.076431: step: 1052/1541, loss: 0.00024232006398960948 2023-01-23 23:18:59.012352: step: 1054/1541, loss: 0.00975308008491993 2023-01-23 23:18:59.875809: step: 1056/1541, loss: 1.1413559150241781e-05 2023-01-23 23:19:00.795695: step: 1058/1541, loss: 0.08509288728237152 2023-01-23 23:19:01.708288: step: 1060/1541, loss: 2.9689072107430547e-05 2023-01-23 23:19:02.665410: step: 1062/1541, loss: 0.0002653987903613597 2023-01-23 23:19:03.600777: step: 1064/1541, loss: 0.0010730839567258954 2023-01-23 23:19:04.453814: step: 1066/1541, loss: 2.892705742851831e-05 2023-01-23 23:19:05.351987: step: 1068/1541, loss: 0.006335433106869459 2023-01-23 23:19:06.251507: step: 1070/1541, loss: 0.00014503540296573192 2023-01-23 23:19:07.132600: step: 1072/1541, loss: 0.0004820289323106408 2023-01-23 23:19:08.057386: step: 1074/1541, loss: 8.392143354285508e-05 2023-01-23 23:19:08.971461: step: 1076/1541, loss: 0.0003961560141760856 2023-01-23 23:19:09.815841: step: 1078/1541, loss: 7.047240069368854e-05 2023-01-23 23:19:10.708422: step: 1080/1541, loss: 0.0001756950659910217 2023-01-23 23:19:11.591599: step: 1082/1541, loss: 0.0015619359910488129 2023-01-23 23:19:12.501043: step: 1084/1541, loss: 5.884989150217734e-05 2023-01-23 23:19:13.401865: step: 1086/1541, loss: 0.0001456752943340689 2023-01-23 23:19:14.290305: step: 1088/1541, loss: 2.788007077469956e-05 2023-01-23 23:19:15.201478: step: 1090/1541, loss: 0.00035671962541528046 2023-01-23 23:19:16.107577: step: 1092/1541, loss: 0.00017088497406803071 2023-01-23 23:19:17.003396: step: 1094/1541, loss: 0.0024421680718660355 2023-01-23 23:19:17.884207: step: 1096/1541, loss: 0.013625536113977432 2023-01-23 23:19:18.805704: step: 1098/1541, loss: 0.0005600297590717673 2023-01-23 23:19:19.700599: step: 1100/1541, loss: 0.0031259041279554367 2023-01-23 23:19:20.607190: step: 1102/1541, loss: 0.0006127567030489445 2023-01-23 23:19:21.499829: step: 1104/1541, loss: 0.0032549239695072174 2023-01-23 23:19:22.366025: step: 1106/1541, loss: 0.000639872218016535 2023-01-23 23:19:23.307589: step: 1108/1541, loss: 4.5522501750383526e-05 2023-01-23 23:19:24.158250: step: 1110/1541, loss: 0.00017167568148579448 2023-01-23 23:19:25.102342: step: 1112/1541, loss: 1.61647803906817e-05 2023-01-23 23:19:26.004680: step: 1114/1541, loss: 0.0002834321931004524 2023-01-23 23:19:26.903992: step: 1116/1541, loss: 0.1281093806028366 2023-01-23 23:19:27.831840: step: 1118/1541, loss: 0.001080419053323567 2023-01-23 23:19:28.717331: step: 1120/1541, loss: 0.0023979726247489452 2023-01-23 23:19:29.605657: step: 1122/1541, loss: 0.0032144400756806135 2023-01-23 23:19:30.445137: step: 1124/1541, loss: 0.001475206227041781 2023-01-23 23:19:31.393841: step: 1126/1541, loss: 9.476541890762746e-05 2023-01-23 23:19:32.282359: step: 1128/1541, loss: -1.3768170902039856e-05 2023-01-23 23:19:33.208389: step: 1130/1541, loss: 0.0003127139061689377 2023-01-23 23:19:34.079833: step: 1132/1541, loss: -4.712380905402824e-05 2023-01-23 23:19:34.974006: step: 1134/1541, loss: 0.008416580967605114 2023-01-23 23:19:35.910343: step: 1136/1541, loss: 0.0459078773856163 2023-01-23 23:19:36.852417: step: 1138/1541, loss: -2.585067522886675e-05 2023-01-23 23:19:37.771191: step: 1140/1541, loss: 2.2923701180843636e-05 2023-01-23 23:19:38.649719: step: 1142/1541, loss: 0.00980948843061924 2023-01-23 23:19:39.531532: step: 1144/1541, loss: 3.775267759920098e-05 2023-01-23 23:19:40.425252: step: 1146/1541, loss: -2.8057962481398135e-05 2023-01-23 23:19:41.297295: step: 1148/1541, loss: 6.721759564243257e-05 2023-01-23 23:19:42.230199: step: 1150/1541, loss: 3.7639336369466037e-05 2023-01-23 23:19:43.148351: step: 1152/1541, loss: 0.00228042877279222 2023-01-23 23:19:44.038265: step: 1154/1541, loss: 2.1827599994139746e-05 2023-01-23 23:19:44.896289: step: 1156/1541, loss: 1.738552236929536e-05 2023-01-23 23:19:45.795837: step: 1158/1541, loss: -4.5605815103044733e-05 2023-01-23 23:19:46.662457: step: 1160/1541, loss: 0.00015262290253303945 2023-01-23 23:19:47.565520: step: 1162/1541, loss: 0.44163888692855835 2023-01-23 23:19:48.460458: step: 1164/1541, loss: 0.007370688486844301 2023-01-23 23:19:49.311681: step: 1166/1541, loss: 0.00021153736452106386 2023-01-23 23:19:50.257631: step: 1168/1541, loss: 0.4436362087726593 2023-01-23 23:19:51.155371: step: 1170/1541, loss: -6.015399776515551e-06 2023-01-23 23:19:52.024764: step: 1172/1541, loss: 7.789288065396249e-05 2023-01-23 23:19:52.877851: step: 1174/1541, loss: 2.3862250600359403e-05 2023-01-23 23:19:53.794694: step: 1176/1541, loss: 0.09776165336370468 2023-01-23 23:19:54.710021: step: 1178/1541, loss: 0.00012694280303549021 2023-01-23 23:19:55.627739: step: 1180/1541, loss: -5.126458199811168e-05 2023-01-23 23:19:56.541530: step: 1182/1541, loss: -3.6197452573105693e-06 2023-01-23 23:19:57.462465: step: 1184/1541, loss: 0.006615462247282267 2023-01-23 23:19:58.364571: step: 1186/1541, loss: 0.14903131127357483 2023-01-23 23:19:59.293083: step: 1188/1541, loss: 0.16790233552455902 2023-01-23 23:20:00.249927: step: 1190/1541, loss: -3.627971454989165e-05 2023-01-23 23:20:01.133323: step: 1192/1541, loss: 0.00012431122013367712 2023-01-23 23:20:02.025769: step: 1194/1541, loss: 0.012395311146974564 2023-01-23 23:20:02.937902: step: 1196/1541, loss: 6.0251673858147115e-05 2023-01-23 23:20:03.820673: step: 1198/1541, loss: 0.00022940825147088617 2023-01-23 23:20:04.678135: step: 1200/1541, loss: 0.0008206405909731984 2023-01-23 23:20:05.588010: step: 1202/1541, loss: -9.11566530703567e-06 2023-01-23 23:20:06.498999: step: 1204/1541, loss: 0.012723899446427822 2023-01-23 23:20:07.442164: step: 1206/1541, loss: 7.349593215622008e-05 2023-01-23 23:20:08.317089: step: 1208/1541, loss: 0.013337470591068268 2023-01-23 23:20:09.200004: step: 1210/1541, loss: 0.14751788973808289 2023-01-23 23:20:10.064190: step: 1212/1541, loss: 0.0002577665145508945 2023-01-23 23:20:10.983296: step: 1214/1541, loss: 0.020335087552666664 2023-01-23 23:20:11.902779: step: 1216/1541, loss: 0.0002029992756433785 2023-01-23 23:20:12.796367: step: 1218/1541, loss: 0.00033488054759800434 2023-01-23 23:20:13.652603: step: 1220/1541, loss: 0.0003413766098674387 2023-01-23 23:20:14.595967: step: 1222/1541, loss: 4.703497688751668e-05 2023-01-23 23:20:15.522768: step: 1224/1541, loss: 0.00024565146304666996 2023-01-23 23:20:16.443892: step: 1226/1541, loss: 0.009193478152155876 2023-01-23 23:20:17.317754: step: 1228/1541, loss: 5.107774995849468e-05 2023-01-23 23:20:18.243592: step: 1230/1541, loss: 0.0001654598891036585 2023-01-23 23:20:19.131631: step: 1232/1541, loss: 0.0017105777515098453 2023-01-23 23:20:20.001116: step: 1234/1541, loss: 0.0008438744698651135 2023-01-23 23:20:20.927501: step: 1236/1541, loss: 0.08520038425922394 2023-01-23 23:20:21.822963: step: 1238/1541, loss: 0.010467546992003918 2023-01-23 23:20:22.740623: step: 1240/1541, loss: 0.0016248010797426105 2023-01-23 23:20:23.658588: step: 1242/1541, loss: 0.000320752413244918 2023-01-23 23:20:24.552139: step: 1244/1541, loss: 0.0004976756172254682 2023-01-23 23:20:25.457617: step: 1246/1541, loss: 0.000718690047506243 2023-01-23 23:20:26.321937: step: 1248/1541, loss: 1.082746712199878e-05 2023-01-23 23:20:27.228199: step: 1250/1541, loss: 0.000486335571622476 2023-01-23 23:20:28.139089: step: 1252/1541, loss: 0.0014835410984233022 2023-01-23 23:20:29.111130: step: 1254/1541, loss: 0.00018308138533029705 2023-01-23 23:20:30.018089: step: 1256/1541, loss: 0.0039021449629217386 2023-01-23 23:20:30.893041: step: 1258/1541, loss: 0.00013334518007468432 2023-01-23 23:20:31.818002: step: 1260/1541, loss: 0.00017196785483974963 2023-01-23 23:20:32.719245: step: 1262/1541, loss: 0.00015226202958729118 2023-01-23 23:20:33.637688: step: 1264/1541, loss: 9.372977365273982e-05 2023-01-23 23:20:34.529321: step: 1266/1541, loss: 0.00024793980992399156 2023-01-23 23:20:35.481132: step: 1268/1541, loss: 7.625337457284331e-05 2023-01-23 23:20:36.355804: step: 1270/1541, loss: 0.15359991788864136 2023-01-23 23:20:37.239859: step: 1272/1541, loss: 0.00016968422278296202 2023-01-23 23:20:38.094932: step: 1274/1541, loss: 0.0019946040119975805 2023-01-23 23:20:39.041277: step: 1276/1541, loss: 0.004709294997155666 2023-01-23 23:20:39.989378: step: 1278/1541, loss: 0.15441815555095673 2023-01-23 23:20:40.891192: step: 1280/1541, loss: 0.006164147984236479 2023-01-23 23:20:41.791735: step: 1282/1541, loss: 0.6832258701324463 2023-01-23 23:20:42.720343: step: 1284/1541, loss: 0.00877291988581419 2023-01-23 23:20:43.602515: step: 1286/1541, loss: 0.012176446616649628 2023-01-23 23:20:44.504825: step: 1288/1541, loss: 0.00012223588419146836 2023-01-23 23:20:45.422175: step: 1290/1541, loss: 0.1429075449705124 2023-01-23 23:20:46.330259: step: 1292/1541, loss: 0.00145555823110044 2023-01-23 23:20:47.241158: step: 1294/1541, loss: 8.892341429600492e-05 2023-01-23 23:20:48.228713: step: 1296/1541, loss: 0.008408591151237488 2023-01-23 23:20:49.120759: step: 1298/1541, loss: 0.00026684702606871724 2023-01-23 23:20:50.107628: step: 1300/1541, loss: 0.0005867369472980499 2023-01-23 23:20:50.953131: step: 1302/1541, loss: 0.0025045110378414392 2023-01-23 23:20:51.857659: step: 1304/1541, loss: 0.07011941075325012 2023-01-23 23:20:52.741437: step: 1306/1541, loss: 0.07346322387456894 2023-01-23 23:20:53.620104: step: 1308/1541, loss: 6.632765871472657e-05 2023-01-23 23:20:54.597590: step: 1310/1541, loss: 0.0006779285613447428 2023-01-23 23:20:55.501971: step: 1312/1541, loss: 2.4904386009438895e-05 2023-01-23 23:20:56.415793: step: 1314/1541, loss: 0.0003691858146339655 2023-01-23 23:20:57.374556: step: 1316/1541, loss: 0.4373658001422882 2023-01-23 23:20:58.249287: step: 1318/1541, loss: 0.0011209914227947593 2023-01-23 23:20:59.136385: step: 1320/1541, loss: 0.0007658074609935284 2023-01-23 23:21:00.012340: step: 1322/1541, loss: 0.0005019631353206933 2023-01-23 23:21:00.908568: step: 1324/1541, loss: 0.00011322923091938719 2023-01-23 23:21:01.895506: step: 1326/1541, loss: 0.00018512354290578514 2023-01-23 23:21:02.777230: step: 1328/1541, loss: 0.099505715072155 2023-01-23 23:21:03.740421: step: 1330/1541, loss: 0.02002234198153019 2023-01-23 23:21:04.723732: step: 1332/1541, loss: 0.009574335999786854 2023-01-23 23:21:05.642467: step: 1334/1541, loss: 3.1580952054355294e-05 2023-01-23 23:21:06.516936: step: 1336/1541, loss: 3.0412080377573147e-05 2023-01-23 23:21:07.387639: step: 1338/1541, loss: 0.0019115753239020705 2023-01-23 23:21:08.266176: step: 1340/1541, loss: 0.0014591850340366364 2023-01-23 23:21:09.154222: step: 1342/1541, loss: 0.00022111165162641555 2023-01-23 23:21:10.053951: step: 1344/1541, loss: 0.0010449390392750502 2023-01-23 23:21:10.955993: step: 1346/1541, loss: 0.007665424607694149 2023-01-23 23:21:11.934841: step: 1348/1541, loss: 3.9794664189685136e-05 2023-01-23 23:21:12.840210: step: 1350/1541, loss: 0.00041548951412551105 2023-01-23 23:21:13.701407: step: 1352/1541, loss: 0.10326722264289856 2023-01-23 23:21:14.627349: step: 1354/1541, loss: 0.0002569803036749363 2023-01-23 23:21:15.527027: step: 1356/1541, loss: 0.31407785415649414 2023-01-23 23:21:16.417704: step: 1358/1541, loss: 0.00017466972349211574 2023-01-23 23:21:17.351407: step: 1360/1541, loss: 0.08664098381996155 2023-01-23 23:21:18.296326: step: 1362/1541, loss: 0.0016289131017401814 2023-01-23 23:21:19.169579: step: 1364/1541, loss: 0.0002569587668403983 2023-01-23 23:21:20.090649: step: 1366/1541, loss: 3.4868235161411576e-06 2023-01-23 23:21:20.978876: step: 1368/1541, loss: 0.0006865178584121168 2023-01-23 23:21:21.895588: step: 1370/1541, loss: 0.013625345192849636 2023-01-23 23:21:22.771853: step: 1372/1541, loss: 0.00017046158609446138 2023-01-23 23:21:23.686354: step: 1374/1541, loss: -2.7835540095111355e-06 2023-01-23 23:21:24.592456: step: 1376/1541, loss: 0.012731609866023064 2023-01-23 23:21:25.528111: step: 1378/1541, loss: 0.00031038999441079795 2023-01-23 23:21:26.407601: step: 1380/1541, loss: 0.25847327709198 2023-01-23 23:21:27.288888: step: 1382/1541, loss: -1.0366669812356122e-05 2023-01-23 23:21:28.257964: step: 1384/1541, loss: 0.0006233559106476605 2023-01-23 23:21:29.134642: step: 1386/1541, loss: 1.110018014907837 2023-01-23 23:21:30.030744: step: 1388/1541, loss: 0.0008824053220450878 2023-01-23 23:21:30.910775: step: 1390/1541, loss: 0.12322859466075897 2023-01-23 23:21:31.804879: step: 1392/1541, loss: 2.344123276998289e-05 2023-01-23 23:21:32.739243: step: 1394/1541, loss: 0.00085889344336465 2023-01-23 23:21:33.660183: step: 1396/1541, loss: 0.020191414281725883 2023-01-23 23:21:34.577194: step: 1398/1541, loss: 9.395334927830845e-05 2023-01-23 23:21:35.510175: step: 1400/1541, loss: 0.05429186671972275 2023-01-23 23:21:36.394817: step: 1402/1541, loss: 0.002393385861068964 2023-01-23 23:21:37.281063: step: 1404/1541, loss: 0.01573224365711212 2023-01-23 23:21:38.188961: step: 1406/1541, loss: 0.01166532188653946 2023-01-23 23:21:39.097236: step: 1408/1541, loss: 0.05815805494785309 2023-01-23 23:21:39.982585: step: 1410/1541, loss: 0.00022427325893659145 2023-01-23 23:21:40.914179: step: 1412/1541, loss: 0.00011033991177100688 2023-01-23 23:21:41.783052: step: 1414/1541, loss: 0.089873306453228 2023-01-23 23:21:42.693547: step: 1416/1541, loss: 0.25924691557884216 2023-01-23 23:21:43.569006: step: 1418/1541, loss: -8.976767276180908e-07 2023-01-23 23:21:44.462163: step: 1420/1541, loss: 0.0012932732934132218 2023-01-23 23:21:45.337495: step: 1422/1541, loss: 0.004917492158710957 2023-01-23 23:21:46.292392: step: 1424/1541, loss: 0.06252686679363251 2023-01-23 23:21:47.145163: step: 1426/1541, loss: 0.00016495661111548543 2023-01-23 23:21:48.032884: step: 1428/1541, loss: 0.1898050159215927 2023-01-23 23:21:48.920572: step: 1430/1541, loss: 0.0037380342837423086 2023-01-23 23:21:49.829581: step: 1432/1541, loss: 0.3151609003543854 2023-01-23 23:21:50.750627: step: 1434/1541, loss: 0.014119129627943039 2023-01-23 23:21:51.660848: step: 1436/1541, loss: 0.012555301189422607 2023-01-23 23:21:52.530019: step: 1438/1541, loss: 0.000206824392080307 2023-01-23 23:21:53.440445: step: 1440/1541, loss: 0.00023453921312466264 2023-01-23 23:21:54.348361: step: 1442/1541, loss: 0.00019310742209199816 2023-01-23 23:21:55.262437: step: 1444/1541, loss: 0.0002573485835455358 2023-01-23 23:21:56.146701: step: 1446/1541, loss: 0.03727623075246811 2023-01-23 23:21:56.990235: step: 1448/1541, loss: 0.0010880277259275317 2023-01-23 23:21:57.866872: step: 1450/1541, loss: 8.942947897594422e-05 2023-01-23 23:21:58.825705: step: 1452/1541, loss: 0.09461329877376556 2023-01-23 23:21:59.701012: step: 1454/1541, loss: 0.001067301258444786 2023-01-23 23:22:00.589869: step: 1456/1541, loss: 0.01118718832731247 2023-01-23 23:22:01.469945: step: 1458/1541, loss: 1.647862336540129e-05 2023-01-23 23:22:02.399354: step: 1460/1541, loss: 0.02384033426642418 2023-01-23 23:22:03.327201: step: 1462/1541, loss: 0.00774290319532156 2023-01-23 23:22:04.214670: step: 1464/1541, loss: 0.18102696537971497 2023-01-23 23:22:05.093148: step: 1466/1541, loss: 0.00835395511239767 2023-01-23 23:22:05.961950: step: 1468/1541, loss: 0.010798094794154167 2023-01-23 23:22:06.857218: step: 1470/1541, loss: 3.051293606404215e-05 2023-01-23 23:22:07.723495: step: 1472/1541, loss: -2.2969170458964072e-05 2023-01-23 23:22:08.655830: step: 1474/1541, loss: 0.0013025341322645545 2023-01-23 23:22:09.609356: step: 1476/1541, loss: 4.959097168466542e-06 2023-01-23 23:22:10.524266: step: 1478/1541, loss: 0.00029489980079233646 2023-01-23 23:22:11.455198: step: 1480/1541, loss: 0.00011454698687884957 2023-01-23 23:22:12.331590: step: 1482/1541, loss: 0.00016479217447340488 2023-01-23 23:22:13.262528: step: 1484/1541, loss: 0.01689946837723255 2023-01-23 23:22:14.150019: step: 1486/1541, loss: 0.07645824551582336 2023-01-23 23:22:15.017666: step: 1488/1541, loss: 0.0011186787160113454 2023-01-23 23:22:15.934971: step: 1490/1541, loss: 0.004420311655849218 2023-01-23 23:22:16.800021: step: 1492/1541, loss: 0.00041496005724184215 2023-01-23 23:22:17.720521: step: 1494/1541, loss: 8.834418258629739e-05 2023-01-23 23:22:18.677274: step: 1496/1541, loss: 0.00010207774175796658 2023-01-23 23:22:19.596319: step: 1498/1541, loss: 0.0024047752376645803 2023-01-23 23:22:20.452560: step: 1500/1541, loss: 0.0001912954612635076 2023-01-23 23:22:21.298145: step: 1502/1541, loss: 5.586370753007941e-05 2023-01-23 23:22:22.184342: step: 1504/1541, loss: 0.0007951264851726592 2023-01-23 23:22:23.118999: step: 1506/1541, loss: 0.00026033230824396014 2023-01-23 23:22:24.033640: step: 1508/1541, loss: 7.644652941962704e-05 2023-01-23 23:22:24.941947: step: 1510/1541, loss: 0.0002471495245117694 2023-01-23 23:22:25.810222: step: 1512/1541, loss: 7.06165792507818e-06 2023-01-23 23:22:26.745433: step: 1514/1541, loss: 0.004473135806620121 2023-01-23 23:22:27.713965: step: 1516/1541, loss: 0.10308840125799179 2023-01-23 23:22:28.596108: step: 1518/1541, loss: 0.01076761819422245 2023-01-23 23:22:29.482789: step: 1520/1541, loss: 0.6326578855514526 2023-01-23 23:22:30.346551: step: 1522/1541, loss: 0.0006370650371536613 2023-01-23 23:22:31.254226: step: 1524/1541, loss: 0.00877540186047554 2023-01-23 23:22:32.148885: step: 1526/1541, loss: 0.00036755026667378843 2023-01-23 23:22:33.023273: step: 1528/1541, loss: 4.103803803445771e-05 2023-01-23 23:22:33.880126: step: 1530/1541, loss: 0.00026767642702907324 2023-01-23 23:22:34.791232: step: 1532/1541, loss: 0.0008823426906019449 2023-01-23 23:22:35.629170: step: 1534/1541, loss: 0.0003236968186683953 2023-01-23 23:22:36.519750: step: 1536/1541, loss: 0.0010716404067352414 2023-01-23 23:22:37.429994: step: 1538/1541, loss: 0.0017723421333357692 2023-01-23 23:22:38.374017: step: 1540/1541, loss: 0.04294489696621895 2023-01-23 23:22:39.224538: step: 1542/1541, loss: 3.2396037568105385e-05 2023-01-23 23:22:40.140391: step: 1544/1541, loss: 3.6498742701951414e-05 2023-01-23 23:22:41.031691: step: 1546/1541, loss: 0.006286080460995436 2023-01-23 23:22:41.882634: step: 1548/1541, loss: 0.051482830196619034 2023-01-23 23:22:42.804876: step: 1550/1541, loss: -2.163909448427148e-05 2023-01-23 23:22:43.744788: step: 1552/1541, loss: 0.007454304955899715 2023-01-23 23:22:44.607430: step: 1554/1541, loss: 0.00017491966718807817 2023-01-23 23:22:45.555278: step: 1556/1541, loss: 0.001432691584341228 2023-01-23 23:22:46.482632: step: 1558/1541, loss: 0.0002008667797781527 2023-01-23 23:22:47.371933: step: 1560/1541, loss: 0.007685945834964514 2023-01-23 23:22:48.262406: step: 1562/1541, loss: 0.0009318428346887231 2023-01-23 23:22:49.167442: step: 1564/1541, loss: 0.1164410188794136 2023-01-23 23:22:50.090210: step: 1566/1541, loss: 0.009080462157726288 2023-01-23 23:22:51.021416: step: 1568/1541, loss: 0.05903146415948868 2023-01-23 23:22:51.924676: step: 1570/1541, loss: 0.0009142250055447221 2023-01-23 23:22:52.827588: step: 1572/1541, loss: 0.0029070039745420218 2023-01-23 23:22:53.730846: step: 1574/1541, loss: 6.73945905873552e-05 2023-01-23 23:22:54.640474: step: 1576/1541, loss: 6.9566071942972485e-06 2023-01-23 23:22:55.527950: step: 1578/1541, loss: 0.0021763662807643414 2023-01-23 23:22:56.451999: step: 1580/1541, loss: -6.713868060614914e-06 2023-01-23 23:22:57.327339: step: 1582/1541, loss: 0.0002155104448320344 2023-01-23 23:22:58.194428: step: 1584/1541, loss: 5.466580478241667e-05 2023-01-23 23:22:59.105993: step: 1586/1541, loss: -1.8039350834442303e-05 2023-01-23 23:23:00.013885: step: 1588/1541, loss: 0.05560435727238655 2023-01-23 23:23:00.904896: step: 1590/1541, loss: 2.0929410311509855e-05 2023-01-23 23:23:01.872394: step: 1592/1541, loss: -1.1438853107392788e-05 2023-01-23 23:23:02.777446: step: 1594/1541, loss: 0.002216099062934518 2023-01-23 23:23:03.654929: step: 1596/1541, loss: 0.09001214802265167 2023-01-23 23:23:04.530738: step: 1598/1541, loss: 0.003850087756291032 2023-01-23 23:23:05.451028: step: 1600/1541, loss: 0.0015953414840623736 2023-01-23 23:23:06.317385: step: 1602/1541, loss: 4.178671588306315e-06 2023-01-23 23:23:07.200625: step: 1604/1541, loss: 5.868871448910795e-05 2023-01-23 23:23:08.101813: step: 1606/1541, loss: 0.0005592746310867369 2023-01-23 23:23:09.010428: step: 1608/1541, loss: 0.007319100201129913 2023-01-23 23:23:09.938931: step: 1610/1541, loss: -2.490008910172037e-06 2023-01-23 23:23:10.871513: step: 1612/1541, loss: 0.00042807889985851943 2023-01-23 23:23:11.773228: step: 1614/1541, loss: 0.031872719526290894 2023-01-23 23:23:12.611070: step: 1616/1541, loss: 0.00013150280574336648 2023-01-23 23:23:13.474897: step: 1618/1541, loss: 0.00012065176269970834 2023-01-23 23:23:14.349791: step: 1620/1541, loss: 0.0017716243164613843 2023-01-23 23:23:15.211231: step: 1622/1541, loss: 2.9868735509808175e-05 2023-01-23 23:23:16.158380: step: 1624/1541, loss: 0.00028456299332901835 2023-01-23 23:23:17.049492: step: 1626/1541, loss: 4.288250420358963e-05 2023-01-23 23:23:17.926744: step: 1628/1541, loss: 0.001762824016623199 2023-01-23 23:23:18.834368: step: 1630/1541, loss: 0.08119149506092072 2023-01-23 23:23:19.704222: step: 1632/1541, loss: 0.00011208220530534163 2023-01-23 23:23:20.570013: step: 1634/1541, loss: 6.139905963209458e-06 2023-01-23 23:23:21.471427: step: 1636/1541, loss: 0.0002448865561746061 2023-01-23 23:23:22.347627: step: 1638/1541, loss: 1.4462306353379972e-05 2023-01-23 23:23:23.203252: step: 1640/1541, loss: 2.7701219096343266e-06 2023-01-23 23:23:24.083132: step: 1642/1541, loss: 4.9129113904200494e-05 2023-01-23 23:23:24.942087: step: 1644/1541, loss: 5.0992428441531956e-05 2023-01-23 23:23:25.805463: step: 1646/1541, loss: 0.00011711700062733144 2023-01-23 23:23:26.690071: step: 1648/1541, loss: 1.1342945981596131e-05 2023-01-23 23:23:27.563083: step: 1650/1541, loss: 0.18342788517475128 2023-01-23 23:23:28.444947: step: 1652/1541, loss: 0.014594536274671555 2023-01-23 23:23:29.341256: step: 1654/1541, loss: 0.4035676121711731 2023-01-23 23:23:30.256022: step: 1656/1541, loss: 0.012597743421792984 2023-01-23 23:23:31.206441: step: 1658/1541, loss: 2.111418143613264e-05 2023-01-23 23:23:32.221542: step: 1660/1541, loss: 0.00011226834612898529 2023-01-23 23:23:33.096767: step: 1662/1541, loss: 0.0102132149040699 2023-01-23 23:23:34.030271: step: 1664/1541, loss: 0.005604890175163746 2023-01-23 23:23:34.880720: step: 1666/1541, loss: -2.3139442419051193e-05 2023-01-23 23:23:35.812713: step: 1668/1541, loss: 7.062416261760518e-05 2023-01-23 23:23:36.686820: step: 1670/1541, loss: 0.02511010318994522 2023-01-23 23:23:37.545241: step: 1672/1541, loss: 0.0003399997658561915 2023-01-23 23:23:38.408400: step: 1674/1541, loss: -1.1802930202975404e-05 2023-01-23 23:23:39.324510: step: 1676/1541, loss: 4.5228443923406303e-07 2023-01-23 23:23:40.186377: step: 1678/1541, loss: 0.0009138754103332758 2023-01-23 23:23:41.124736: step: 1680/1541, loss: 1.7139060219051316e-05 2023-01-23 23:23:42.001569: step: 1682/1541, loss: 0.08328001201152802 2023-01-23 23:23:42.893054: step: 1684/1541, loss: 2.9641150831594132e-05 2023-01-23 23:23:43.770609: step: 1686/1541, loss: 4.913059456157498e-05 2023-01-23 23:23:44.640929: step: 1688/1541, loss: 0.0007785613415762782 2023-01-23 23:23:45.528443: step: 1690/1541, loss: 0.0001933759485837072 2023-01-23 23:23:46.429938: step: 1692/1541, loss: 8.837186032906175e-05 2023-01-23 23:23:47.456077: step: 1694/1541, loss: 0.003932422026991844 2023-01-23 23:23:48.345522: step: 1696/1541, loss: 1.8174576325691305e-05 2023-01-23 23:23:49.235704: step: 1698/1541, loss: 0.025634273886680603 2023-01-23 23:23:50.131604: step: 1700/1541, loss: 0.0009679583017714322 2023-01-23 23:23:51.042100: step: 1702/1541, loss: 0.034973543137311935 2023-01-23 23:23:51.904134: step: 1704/1541, loss: 0.00035252782981842756 2023-01-23 23:23:52.867785: step: 1706/1541, loss: 1.904346618175623e-06 2023-01-23 23:23:53.744703: step: 1708/1541, loss: 4.997702126274817e-05 2023-01-23 23:23:54.657473: step: 1710/1541, loss: 0.0001822549820644781 2023-01-23 23:23:55.501589: step: 1712/1541, loss: 5.1584032917162403e-05 2023-01-23 23:23:56.385790: step: 1714/1541, loss: 0.014695188030600548 2023-01-23 23:23:57.446294: step: 1716/1541, loss: 0.03854398801922798 2023-01-23 23:23:58.332114: step: 1718/1541, loss: 0.0023666296619921923 2023-01-23 23:23:59.245654: step: 1720/1541, loss: 0.39180222153663635 2023-01-23 23:24:00.118539: step: 1722/1541, loss: 0.033057745546102524 2023-01-23 23:24:01.021075: step: 1724/1541, loss: 0.014677220955491066 2023-01-23 23:24:01.915896: step: 1726/1541, loss: 0.00031516639865003526 2023-01-23 23:24:02.838026: step: 1728/1541, loss: 0.38182440400123596 2023-01-23 23:24:03.748741: step: 1730/1541, loss: 0.015043728053569794 2023-01-23 23:24:04.624817: step: 1732/1541, loss: 0.0015964286867529154 2023-01-23 23:24:05.514590: step: 1734/1541, loss: 0.00013869404210709035 2023-01-23 23:24:06.432045: step: 1736/1541, loss: 0.003910878673195839 2023-01-23 23:24:07.303921: step: 1738/1541, loss: 6.850146746728569e-06 2023-01-23 23:24:08.210354: step: 1740/1541, loss: 0.0012241165386512876 2023-01-23 23:24:09.111882: step: 1742/1541, loss: -7.3268747655674815e-06 2023-01-23 23:24:09.963532: step: 1744/1541, loss: 9.071272870642133e-06 2023-01-23 23:24:10.829405: step: 1746/1541, loss: 0.13366256654262543 2023-01-23 23:24:11.709158: step: 1748/1541, loss: 0.0001198914396809414 2023-01-23 23:24:12.564494: step: 1750/1541, loss: 4.395765245135408e-06 2023-01-23 23:24:13.455159: step: 1752/1541, loss: 0.0003312987682875246 2023-01-23 23:24:14.365794: step: 1754/1541, loss: 0.00044209370389580727 2023-01-23 23:24:15.253010: step: 1756/1541, loss: 0.0001857514725998044 2023-01-23 23:24:16.128144: step: 1758/1541, loss: 3.787601599469781e-05 2023-01-23 23:24:17.017787: step: 1760/1541, loss: 0.010029638186097145 2023-01-23 23:24:17.908078: step: 1762/1541, loss: 0.07620652765035629 2023-01-23 23:24:18.835566: step: 1764/1541, loss: 0.00867624394595623 2023-01-23 23:24:19.768740: step: 1766/1541, loss: 0.001023659948259592 2023-01-23 23:24:20.656315: step: 1768/1541, loss: 0.4886671304702759 2023-01-23 23:24:21.555424: step: 1770/1541, loss: -4.524172254605219e-05 2023-01-23 23:24:22.414813: step: 1772/1541, loss: 0.00274690892547369 2023-01-23 23:24:23.298556: step: 1774/1541, loss: 0.021289749071002007 2023-01-23 23:24:24.195726: step: 1776/1541, loss: 1.6480678368679946e-06 2023-01-23 23:24:25.114460: step: 1778/1541, loss: 0.06950525939464569 2023-01-23 23:24:26.079442: step: 1780/1541, loss: 9.319117816630751e-05 2023-01-23 23:24:26.958521: step: 1782/1541, loss: 0.001149495947174728 2023-01-23 23:24:27.897504: step: 1784/1541, loss: 0.00020140456035733223 2023-01-23 23:24:28.826795: step: 1786/1541, loss: 3.0758455977775156e-05 2023-01-23 23:24:29.795473: step: 1788/1541, loss: 0.0008897650404833257 2023-01-23 23:24:30.693090: step: 1790/1541, loss: 0.021903395652770996 2023-01-23 23:24:31.550683: step: 1792/1541, loss: 0.013640856370329857 2023-01-23 23:24:32.449931: step: 1794/1541, loss: 0.008083419874310493 2023-01-23 23:24:33.366526: step: 1796/1541, loss: 0.0010533835738897324 2023-01-23 23:24:34.246821: step: 1798/1541, loss: 0.09247612208127975 2023-01-23 23:24:35.191888: step: 1800/1541, loss: -1.7993817891692743e-05 2023-01-23 23:24:36.187225: step: 1802/1541, loss: 0.19209490716457367 2023-01-23 23:24:37.108106: step: 1804/1541, loss: 2.6819428967428394e-05 2023-01-23 23:24:37.954153: step: 1806/1541, loss: 1.986274219234474e-05 2023-01-23 23:24:38.835746: step: 1808/1541, loss: 2.5845565687632188e-05 2023-01-23 23:24:39.766116: step: 1810/1541, loss: 0.27551767230033875 2023-01-23 23:24:40.682709: step: 1812/1541, loss: 0.004067688714712858 2023-01-23 23:24:41.584963: step: 1814/1541, loss: 1.695445280347485e-05 2023-01-23 23:24:42.471392: step: 1816/1541, loss: 0.00031489855609834194 2023-01-23 23:24:43.431540: step: 1818/1541, loss: 0.024230558425188065 2023-01-23 23:24:44.375149: step: 1820/1541, loss: 0.044046126306056976 2023-01-23 23:24:45.272560: step: 1822/1541, loss: 0.00016528744890820235 2023-01-23 23:24:46.239158: step: 1824/1541, loss: 0.0008551981300115585 2023-01-23 23:24:47.105070: step: 1826/1541, loss: 0.027243303135037422 2023-01-23 23:24:47.993028: step: 1828/1541, loss: 0.15214695036411285 2023-01-23 23:24:48.865112: step: 1830/1541, loss: 0.0001725725014694035 2023-01-23 23:24:49.755299: step: 1832/1541, loss: 0.05083664879202843 2023-01-23 23:24:50.637957: step: 1834/1541, loss: 0.0002983357117045671 2023-01-23 23:24:51.554810: step: 1836/1541, loss: 3.747369919437915e-05 2023-01-23 23:24:52.473264: step: 1838/1541, loss: -4.16731636505574e-05 2023-01-23 23:24:53.375662: step: 1840/1541, loss: 5.674438216374256e-05 2023-01-23 23:24:54.280422: step: 1842/1541, loss: 0.22483468055725098 2023-01-23 23:24:55.162213: step: 1844/1541, loss: 0.1079176589846611 2023-01-23 23:24:56.079841: step: 1846/1541, loss: 0.00014455259952228516 2023-01-23 23:24:56.960848: step: 1848/1541, loss: 0.00041965529089793563 2023-01-23 23:24:57.839616: step: 1850/1541, loss: 7.95240412116982e-05 2023-01-23 23:24:58.727004: step: 1852/1541, loss: 0.07371863722801208 2023-01-23 23:24:59.634382: step: 1854/1541, loss: 0.2203746885061264 2023-01-23 23:25:00.550509: step: 1856/1541, loss: 4.6246499550761655e-05 2023-01-23 23:25:01.478448: step: 1858/1541, loss: 0.012485128827393055 2023-01-23 23:25:02.374622: step: 1860/1541, loss: 0.00031223802943713963 2023-01-23 23:25:03.226395: step: 1862/1541, loss: 0.0012718127109110355 2023-01-23 23:25:04.094431: step: 1864/1541, loss: 0.00022549362620338798 2023-01-23 23:25:04.971116: step: 1866/1541, loss: 0.06195307895541191 2023-01-23 23:25:05.864369: step: 1868/1541, loss: 0.002408993197605014 2023-01-23 23:25:06.762129: step: 1870/1541, loss: 0.00962394941598177 2023-01-23 23:25:07.677398: step: 1872/1541, loss: 0.0014342361828312278 2023-01-23 23:25:08.643333: step: 1874/1541, loss: 0.0016543305246159434 2023-01-23 23:25:09.496955: step: 1876/1541, loss: 0.0005012764595448971 2023-01-23 23:25:10.373849: step: 1878/1541, loss: 0.03928004205226898 2023-01-23 23:25:11.230205: step: 1880/1541, loss: 0.0013250544434413314 2023-01-23 23:25:12.206404: step: 1882/1541, loss: 0.00030164120835252106 2023-01-23 23:25:13.065919: step: 1884/1541, loss: 0.005739914253354073 2023-01-23 23:25:13.960547: step: 1886/1541, loss: 0.04848109185695648 2023-01-23 23:25:14.913966: step: 1888/1541, loss: 0.001030609244480729 2023-01-23 23:25:15.858345: step: 1890/1541, loss: 0.0023173599038273096 2023-01-23 23:25:16.815611: step: 1892/1541, loss: 0.03909137472510338 2023-01-23 23:25:17.824800: step: 1894/1541, loss: 0.004746714141219854 2023-01-23 23:25:18.672551: step: 1896/1541, loss: 0.009114900603890419 2023-01-23 23:25:19.551250: step: 1898/1541, loss: 0.0013685326557606459 2023-01-23 23:25:20.451589: step: 1900/1541, loss: 0.001146871130913496 2023-01-23 23:25:21.332854: step: 1902/1541, loss: 0.6746564507484436 2023-01-23 23:25:22.245444: step: 1904/1541, loss: 8.639181760372594e-05 2023-01-23 23:25:23.137180: step: 1906/1541, loss: 1.6146152120199986e-05 2023-01-23 23:25:24.062854: step: 1908/1541, loss: 0.3126709759235382 2023-01-23 23:25:24.960103: step: 1910/1541, loss: 0.00040407379856333137 2023-01-23 23:25:25.825528: step: 1912/1541, loss: 0.00010623357957229018 2023-01-23 23:25:26.699225: step: 1914/1541, loss: 0.006209523882716894 2023-01-23 23:25:27.556873: step: 1916/1541, loss: 0.0010276020038872957 2023-01-23 23:25:28.419891: step: 1918/1541, loss: 0.06313072890043259 2023-01-23 23:25:29.294870: step: 1920/1541, loss: 3.2688873034203425e-05 2023-01-23 23:25:30.202198: step: 1922/1541, loss: 0.004709710832685232 2023-01-23 23:25:31.091907: step: 1924/1541, loss: 2.1272375306580216e-05 2023-01-23 23:25:32.036957: step: 1926/1541, loss: 0.012531022541224957 2023-01-23 23:25:32.903187: step: 1928/1541, loss: 9.339147800346836e-05 2023-01-23 23:25:33.806736: step: 1930/1541, loss: 0.00012349797179922462 2023-01-23 23:25:34.692466: step: 1932/1541, loss: 0.0030636147130280733 2023-01-23 23:25:35.711670: step: 1934/1541, loss: 0.36733317375183105 2023-01-23 23:25:36.656241: step: 1936/1541, loss: 0.0013269835617393255 2023-01-23 23:25:37.589814: step: 1938/1541, loss: 2.7329235308570787e-05 2023-01-23 23:25:38.511929: step: 1940/1541, loss: 0.009192743338644505 2023-01-23 23:25:39.370536: step: 1942/1541, loss: 4.89420308440458e-05 2023-01-23 23:25:40.239932: step: 1944/1541, loss: 0.0005202734027989209 2023-01-23 23:25:41.098300: step: 1946/1541, loss: 7.898011972429231e-05 2023-01-23 23:25:41.974306: step: 1948/1541, loss: 0.0005762290093116462 2023-01-23 23:25:42.984645: step: 1950/1541, loss: 0.00016890227561816573 2023-01-23 23:25:43.872091: step: 1952/1541, loss: 8.610563963884488e-05 2023-01-23 23:25:44.754762: step: 1954/1541, loss: 0.0495326966047287 2023-01-23 23:25:45.649261: step: 1956/1541, loss: 0.10316745191812515 2023-01-23 23:25:46.516013: step: 1958/1541, loss: 0.0001057224435498938 2023-01-23 23:25:47.454748: step: 1960/1541, loss: 0.9185552597045898 2023-01-23 23:25:48.338319: step: 1962/1541, loss: 0.022369040176272392 2023-01-23 23:25:49.203975: step: 1964/1541, loss: -5.150511424290016e-05 2023-01-23 23:25:50.086771: step: 1966/1541, loss: 5.432129000837449e-06 2023-01-23 23:25:50.999769: step: 1968/1541, loss: 0.00025090205599553883 2023-01-23 23:25:51.915334: step: 1970/1541, loss: 0.0004075395700056106 2023-01-23 23:25:52.832267: step: 1972/1541, loss: 0.0783097967505455 2023-01-23 23:25:53.698099: step: 1974/1541, loss: 0.0015316993230953813 2023-01-23 23:25:54.554421: step: 1976/1541, loss: 0.00023666978813707829 2023-01-23 23:25:55.404833: step: 1978/1541, loss: 0.0004744518082588911 2023-01-23 23:25:56.351401: step: 1980/1541, loss: 0.0019377163844183087 2023-01-23 23:25:57.285521: step: 1982/1541, loss: -3.312486660433933e-05 2023-01-23 23:25:58.189531: step: 1984/1541, loss: 0.05077964439988136 2023-01-23 23:25:59.078376: step: 1986/1541, loss: 0.0011337127070873976 2023-01-23 23:25:59.950199: step: 1988/1541, loss: 0.00037200230872258544 2023-01-23 23:26:00.923904: step: 1990/1541, loss: 7.664971235499252e-06 2023-01-23 23:26:01.871836: step: 1992/1541, loss: -3.189924609614536e-05 2023-01-23 23:26:02.795908: step: 1994/1541, loss: 0.07301231473684311 2023-01-23 23:26:03.698395: step: 1996/1541, loss: 0.1437646597623825 2023-01-23 23:26:04.602900: step: 1998/1541, loss: 8.752424037083983e-06 2023-01-23 23:26:05.485437: step: 2000/1541, loss: 0.0007498273625969887 2023-01-23 23:26:06.353412: step: 2002/1541, loss: 0.07008591294288635 2023-01-23 23:26:07.241922: step: 2004/1541, loss: 6.0259990277700126e-05 2023-01-23 23:26:08.125766: step: 2006/1541, loss: 0.0002450219471938908 2023-01-23 23:26:09.037648: step: 2008/1541, loss: 0.00012412334035616368 2023-01-23 23:26:10.034129: step: 2010/1541, loss: 0.08043722808361053 2023-01-23 23:26:10.917165: step: 2012/1541, loss: 0.00013395525456871837 2023-01-23 23:26:11.779769: step: 2014/1541, loss: 0.0024820331018418074 2023-01-23 23:26:12.665195: step: 2016/1541, loss: 0.0008486578008159995 2023-01-23 23:26:13.523432: step: 2018/1541, loss: 0.014951150864362717 2023-01-23 23:26:14.434639: step: 2020/1541, loss: 0.0006466421182267368 2023-01-23 23:26:15.333983: step: 2022/1541, loss: 0.0004136962816119194 2023-01-23 23:26:16.223270: step: 2024/1541, loss: 8.894711936591193e-05 2023-01-23 23:26:17.154938: step: 2026/1541, loss: 0.11921858787536621 2023-01-23 23:26:18.043537: step: 2028/1541, loss: 0.00313932984136045 2023-01-23 23:26:18.986720: step: 2030/1541, loss: 0.0018991122487932444 2023-01-23 23:26:19.873541: step: 2032/1541, loss: 0.0026681581512093544 2023-01-23 23:26:20.757813: step: 2034/1541, loss: 1.1035230159759521 2023-01-23 23:26:21.638954: step: 2036/1541, loss: 0.22438964247703552 2023-01-23 23:26:22.498691: step: 2038/1541, loss: 0.07907050102949142 2023-01-23 23:26:23.357314: step: 2040/1541, loss: 0.0010988981230184436