Command that produces this log: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 ---------------------------------------------------------------------------------------------------- > trainable params: >>> 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]) >>> 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]) >>> trigger_crf.transition: torch.Size([235, 235]) n_trainable_params: 560511990, n_nontrainable_params: 0 ---------------------------------------------------------------------------------------------------- ****************************** Epoch: 0 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:36:15.688056: step: 4/526, loss: 3.6318392753601074 2023-01-21 08:36:16.844772: step: 8/526, loss: 28.890989303588867 2023-01-21 08:36:17.968591: step: 12/526, loss: 18.407047271728516 2023-01-21 08:36:19.108346: step: 16/526, loss: 25.845109939575195 2023-01-21 08:36:20.217487: step: 20/526, loss: 9.149889945983887 2023-01-21 08:36:21.348018: step: 24/526, loss: 3.810722827911377 2023-01-21 08:36:22.522087: step: 28/526, loss: 11.724077224731445 2023-01-21 08:36:23.668311: step: 32/526, loss: 14.911371231079102 2023-01-21 08:36:24.799034: step: 36/526, loss: 8.573379516601562 2023-01-21 08:36:25.928404: step: 40/526, loss: 4.186492919921875 2023-01-21 08:36:27.069676: step: 44/526, loss: 14.293676376342773 2023-01-21 08:36:28.189150: step: 48/526, loss: 3.38279128074646 2023-01-21 08:36:29.331496: step: 52/526, loss: 20.770090103149414 2023-01-21 08:36:30.482725: step: 56/526, loss: 11.954267501831055 2023-01-21 08:36:31.606204: step: 60/526, loss: 3.2696142196655273 2023-01-21 08:36:32.771449: step: 64/526, loss: 3.5716280937194824 2023-01-21 08:36:33.921258: step: 68/526, loss: 3.261629581451416 2023-01-21 08:36:35.086687: step: 72/526, loss: 13.102941513061523 2023-01-21 08:36:36.242454: step: 76/526, loss: 33.936065673828125 2023-01-21 08:36:37.407205: step: 80/526, loss: 3.9718658924102783 2023-01-21 08:36:38.554603: step: 84/526, loss: 9.024330139160156 2023-01-21 08:36:39.676146: step: 88/526, loss: 3.0318570137023926 2023-01-21 08:36:40.833738: step: 92/526, loss: 9.598344802856445 2023-01-21 08:36:41.973822: step: 96/526, loss: 7.868564605712891 2023-01-21 08:36:43.105714: step: 100/526, loss: 20.54723358154297 2023-01-21 08:36:44.247598: step: 104/526, loss: 7.756442070007324 2023-01-21 08:36:45.371913: step: 108/526, loss: 3.0901856422424316 2023-01-21 08:36:46.512206: step: 112/526, loss: 3.561924934387207 2023-01-21 08:36:47.628078: step: 116/526, loss: 7.165226936340332 2023-01-21 08:36:48.788882: step: 120/526, loss: 23.627790451049805 2023-01-21 08:36:49.914090: step: 124/526, loss: 4.433967113494873 2023-01-21 08:36:51.034639: step: 128/526, loss: 13.168096542358398 2023-01-21 08:36:52.181035: step: 132/526, loss: 15.751042366027832 2023-01-21 08:36:53.302804: step: 136/526, loss: 23.246429443359375 2023-01-21 08:36:54.445334: step: 140/526, loss: 16.421527862548828 2023-01-21 08:36:55.565835: step: 144/526, loss: 3.2233738899230957 2023-01-21 08:36:56.712405: step: 148/526, loss: 16.15756607055664 2023-01-21 08:36:57.825031: step: 152/526, loss: 3.4238474369049072 2023-01-21 08:36:58.939578: step: 156/526, loss: 10.485424041748047 2023-01-21 08:37:00.066978: step: 160/526, loss: 9.064422607421875 2023-01-21 08:37:01.221857: step: 164/526, loss: 4.762728214263916 2023-01-21 08:37:02.369101: step: 168/526, loss: 30.001049041748047 2023-01-21 08:37:03.511657: step: 172/526, loss: 4.155209541320801 2023-01-21 08:37:04.636081: step: 176/526, loss: 2.397127151489258 2023-01-21 08:37:05.776246: step: 180/526, loss: 2.4548416137695312 2023-01-21 08:37:06.964456: step: 184/526, loss: 3.7789855003356934 2023-01-21 08:37:08.099615: step: 188/526, loss: 14.679725646972656 2023-01-21 08:37:09.244694: step: 192/526, loss: 11.192453384399414 2023-01-21 08:37:10.360159: step: 196/526, loss: 2.4589171409606934 2023-01-21 08:37:11.486205: step: 200/526, loss: 22.41259765625 2023-01-21 08:37:12.614502: step: 204/526, loss: 7.1796698570251465 2023-01-21 08:37:13.733541: step: 208/526, loss: 3.5711936950683594 2023-01-21 08:37:14.880968: step: 212/526, loss: 20.40176773071289 2023-01-21 08:37:16.020478: step: 216/526, loss: 28.562137603759766 2023-01-21 08:37:17.225038: step: 220/526, loss: 9.086514472961426 2023-01-21 08:37:18.338043: step: 224/526, loss: 9.796311378479004 2023-01-21 08:37:19.461499: step: 228/526, loss: 7.405766487121582 2023-01-21 08:37:20.601975: step: 232/526, loss: 17.26175308227539 2023-01-21 08:37:21.700708: step: 236/526, loss: 2.8710148334503174 2023-01-21 08:37:22.853190: step: 240/526, loss: 4.286104679107666 2023-01-21 08:37:23.966689: step: 244/526, loss: 9.266624450683594 2023-01-21 08:37:25.098058: step: 248/526, loss: 2.771188259124756 2023-01-21 08:37:26.230160: step: 252/526, loss: 3.888857126235962 2023-01-21 08:37:27.342374: step: 256/526, loss: 13.929132461547852 2023-01-21 08:37:28.477564: step: 260/526, loss: 4.139479160308838 2023-01-21 08:37:29.591483: step: 264/526, loss: 13.014013290405273 2023-01-21 08:37:30.725080: step: 268/526, loss: 3.2820091247558594 2023-01-21 08:37:31.898762: step: 272/526, loss: 12.483892440795898 2023-01-21 08:37:33.035683: step: 276/526, loss: 3.6978845596313477 2023-01-21 08:37:34.166833: step: 280/526, loss: 17.864959716796875 2023-01-21 08:37:35.316564: step: 284/526, loss: 1.820019245147705 2023-01-21 08:37:36.470826: step: 288/526, loss: 26.590755462646484 2023-01-21 08:37:37.596906: step: 292/526, loss: 4.36468505859375 2023-01-21 08:37:38.759048: step: 296/526, loss: 25.427658081054688 2023-01-21 08:37:39.918548: step: 300/526, loss: 7.74377965927124 2023-01-21 08:37:41.055988: step: 304/526, loss: 10.137595176696777 2023-01-21 08:37:42.186221: step: 308/526, loss: 9.967369079589844 2023-01-21 08:37:43.393898: step: 312/526, loss: 15.322711944580078 2023-01-21 08:37:44.548407: step: 316/526, loss: 3.326475143432617 2023-01-21 08:37:45.687715: step: 320/526, loss: 20.173986434936523 2023-01-21 08:37:46.840807: step: 324/526, loss: 7.19556999206543 2023-01-21 08:37:47.969090: step: 328/526, loss: 3.106248378753662 2023-01-21 08:37:49.122477: step: 332/526, loss: 14.252914428710938 2023-01-21 08:37:50.271683: step: 336/526, loss: 15.54052448272705 2023-01-21 08:37:51.411668: step: 340/526, loss: 10.941143989562988 2023-01-21 08:37:52.525066: step: 344/526, loss: 3.1308562755584717 2023-01-21 08:37:53.657197: step: 348/526, loss: 12.181527137756348 2023-01-21 08:37:54.806267: step: 352/526, loss: 21.351825714111328 2023-01-21 08:37:55.943207: step: 356/526, loss: 3.507929801940918 2023-01-21 08:37:57.093936: step: 360/526, loss: 3.4611740112304688 2023-01-21 08:37:58.222042: step: 364/526, loss: 8.646406173706055 2023-01-21 08:37:59.347239: step: 368/526, loss: 2.7202072143554688 2023-01-21 08:38:00.499982: step: 372/526, loss: 11.316137313842773 2023-01-21 08:38:01.637283: step: 376/526, loss: 6.047792434692383 2023-01-21 08:38:02.760597: step: 380/526, loss: 9.796491622924805 2023-01-21 08:38:03.918694: step: 384/526, loss: 8.999422073364258 2023-01-21 08:38:05.026529: step: 388/526, loss: 3.275467872619629 2023-01-21 08:38:06.178696: step: 392/526, loss: 2.560004949569702 2023-01-21 08:38:07.332585: step: 396/526, loss: 14.490044593811035 2023-01-21 08:38:08.527601: step: 400/526, loss: 3.7974331378936768 2023-01-21 08:38:09.645075: step: 404/526, loss: 8.72499942779541 2023-01-21 08:38:10.796956: step: 408/526, loss: 8.436064720153809 2023-01-21 08:38:11.912492: step: 412/526, loss: 2.135772705078125 2023-01-21 08:38:13.019217: step: 416/526, loss: 2.2120108604431152 2023-01-21 08:38:14.161821: step: 420/526, loss: 11.856246948242188 2023-01-21 08:38:15.310935: step: 424/526, loss: 3.0674328804016113 2023-01-21 08:38:16.454475: step: 428/526, loss: 10.840810775756836 2023-01-21 08:38:17.609559: step: 432/526, loss: 4.878076553344727 2023-01-21 08:38:18.725038: step: 436/526, loss: 3.2610654830932617 2023-01-21 08:38:19.861134: step: 440/526, loss: 6.676850318908691 2023-01-21 08:38:20.982861: step: 444/526, loss: 2.6051454544067383 2023-01-21 08:38:22.094192: step: 448/526, loss: 1.9870959520339966 2023-01-21 08:38:23.208166: step: 452/526, loss: 1.5810116529464722 2023-01-21 08:38:24.333163: step: 456/526, loss: 12.493521690368652 2023-01-21 08:38:25.471912: step: 460/526, loss: 10.646100997924805 2023-01-21 08:38:26.624360: step: 464/526, loss: 13.667482376098633 2023-01-21 08:38:27.767733: step: 468/526, loss: 1.5309419631958008 2023-01-21 08:38:28.884581: step: 472/526, loss: 15.322973251342773 2023-01-21 08:38:29.994564: step: 476/526, loss: 1.8751522302627563 2023-01-21 08:38:31.116101: step: 480/526, loss: 4.105733871459961 2023-01-21 08:38:32.262877: step: 484/526, loss: 8.971606254577637 2023-01-21 08:38:33.378805: step: 488/526, loss: 5.2371320724487305 2023-01-21 08:38:34.493485: step: 492/526, loss: 1.2573308944702148 2023-01-21 08:38:35.611210: step: 496/526, loss: 6.7426252365112305 2023-01-21 08:38:36.727239: step: 500/526, loss: 7.719080924987793 2023-01-21 08:38:37.851414: step: 504/526, loss: 1.5554404258728027 2023-01-21 08:38:38.980485: step: 508/526, loss: 1.9200359582901 2023-01-21 08:38:40.116605: step: 512/526, loss: 8.222696304321289 2023-01-21 08:38:41.231524: step: 516/526, loss: 1.8079607486724854 2023-01-21 08:38:42.357304: step: 520/526, loss: 1.5630905628204346 2023-01-21 08:38:43.488814: step: 524/526, loss: 3.2997095584869385 2023-01-21 08:38:44.639744: step: 528/526, loss: 1.4388279914855957 2023-01-21 08:38:45.774885: step: 532/526, loss: 1.338409662246704 2023-01-21 08:38:46.911560: step: 536/526, loss: 1.5269181728363037 2023-01-21 08:38:48.062493: step: 540/526, loss: 2.7201998233795166 2023-01-21 08:38:49.198968: step: 544/526, loss: 3.768095016479492 2023-01-21 08:38:50.319987: step: 548/526, loss: 3.130605936050415 2023-01-21 08:38:51.456795: step: 552/526, loss: 2.2645978927612305 2023-01-21 08:38:52.567170: step: 556/526, loss: 1.4041845798492432 2023-01-21 08:38:53.723059: step: 560/526, loss: 9.178227424621582 2023-01-21 08:38:54.866624: step: 564/526, loss: 1.1238369941711426 2023-01-21 08:38:56.033104: step: 568/526, loss: 1.3538631200790405 2023-01-21 08:38:57.172601: step: 572/526, loss: 1.1342747211456299 2023-01-21 08:38:58.330171: step: 576/526, loss: 0.610762894153595 2023-01-21 08:38:59.480708: step: 580/526, loss: 0.8239659667015076 2023-01-21 08:39:00.625636: step: 584/526, loss: 2.9253759384155273 2023-01-21 08:39:01.764435: step: 588/526, loss: 2.825378894805908 2023-01-21 08:39:02.887588: step: 592/526, loss: 3.6761865615844727 2023-01-21 08:39:04.038049: step: 596/526, loss: 2.2604973316192627 2023-01-21 08:39:05.177134: step: 600/526, loss: 1.8053219318389893 2023-01-21 08:39:06.303205: step: 604/526, loss: 1.5901368856430054 2023-01-21 08:39:07.432854: step: 608/526, loss: 0.6941806674003601 2023-01-21 08:39:08.564861: step: 612/526, loss: 0.5261372923851013 2023-01-21 08:39:09.688770: step: 616/526, loss: 0.7327295541763306 2023-01-21 08:39:10.842453: step: 620/526, loss: 0.6329482793807983 2023-01-21 08:39:11.966347: step: 624/526, loss: 2.6085376739501953 2023-01-21 08:39:13.103589: step: 628/526, loss: 1.348832130432129 2023-01-21 08:39:14.243597: step: 632/526, loss: 0.6164334416389465 2023-01-21 08:39:15.377581: step: 636/526, loss: 0.8436727523803711 2023-01-21 08:39:16.482397: step: 640/526, loss: 1.7077959775924683 2023-01-21 08:39:17.623289: step: 644/526, loss: 0.9112004637718201 2023-01-21 08:39:18.754237: step: 648/526, loss: 0.9629526138305664 2023-01-21 08:39:19.887733: step: 652/526, loss: 3.245166778564453 2023-01-21 08:39:21.015755: step: 656/526, loss: 0.32384079694747925 2023-01-21 08:39:22.141716: step: 660/526, loss: 1.9293735027313232 2023-01-21 08:39:23.248666: step: 664/526, loss: 0.3393259048461914 2023-01-21 08:39:24.373692: step: 668/526, loss: 1.1617307662963867 2023-01-21 08:39:25.536931: step: 672/526, loss: 8.011707305908203 2023-01-21 08:39:26.663252: step: 676/526, loss: 2.6159844398498535 2023-01-21 08:39:27.785740: step: 680/526, loss: 1.7561559677124023 2023-01-21 08:39:28.899758: step: 684/526, loss: 0.5061033964157104 2023-01-21 08:39:30.051785: step: 688/526, loss: 1.9517956972122192 2023-01-21 08:39:31.159346: step: 692/526, loss: 2.0456690788269043 2023-01-21 08:39:32.312168: step: 696/526, loss: 1.0951050519943237 2023-01-21 08:39:33.431932: step: 700/526, loss: 1.2494298219680786 2023-01-21 08:39:34.543939: step: 704/526, loss: 0.6463273167610168 2023-01-21 08:39:35.669362: step: 708/526, loss: 0.38262584805488586 2023-01-21 08:39:36.807755: step: 712/526, loss: 3.2256932258605957 2023-01-21 08:39:37.921229: step: 716/526, loss: 4.036816596984863 2023-01-21 08:39:39.097746: step: 720/526, loss: 0.7773375511169434 2023-01-21 08:39:40.211129: step: 724/526, loss: 0.984978973865509 2023-01-21 08:39:41.336420: step: 728/526, loss: 2.834669589996338 2023-01-21 08:39:42.496259: step: 732/526, loss: 1.0841844081878662 2023-01-21 08:39:43.623212: step: 736/526, loss: 1.4302542209625244 2023-01-21 08:39:44.761364: step: 740/526, loss: 2.4996962547302246 2023-01-21 08:39:45.908891: step: 744/526, loss: 1.054992914199829 2023-01-21 08:39:47.043962: step: 748/526, loss: 0.9370424747467041 2023-01-21 08:39:48.166111: step: 752/526, loss: 2.3080639839172363 2023-01-21 08:39:49.318097: step: 756/526, loss: 6.626799583435059 2023-01-21 08:39:50.431928: step: 760/526, loss: 0.841479480266571 2023-01-21 08:39:51.568964: step: 764/526, loss: 9.992548942565918 2023-01-21 08:39:52.683494: step: 768/526, loss: 1.978711485862732 2023-01-21 08:39:53.821115: step: 772/526, loss: 0.7774362564086914 2023-01-21 08:39:54.956289: step: 776/526, loss: 1.2682746648788452 2023-01-21 08:39:56.102490: step: 780/526, loss: 2.283411979675293 2023-01-21 08:39:57.223117: step: 784/526, loss: 2.58374285697937 2023-01-21 08:39:58.350265: step: 788/526, loss: 0.33996516466140747 2023-01-21 08:39:59.490794: step: 792/526, loss: 0.7127280235290527 2023-01-21 08:40:00.625398: step: 796/526, loss: 2.485874652862549 2023-01-21 08:40:01.746309: step: 800/526, loss: 0.5513519048690796 2023-01-21 08:40:02.850045: step: 804/526, loss: 0.2581776976585388 2023-01-21 08:40:04.002595: step: 808/526, loss: 0.9015848636627197 2023-01-21 08:40:05.089819: step: 812/526, loss: 1.814470887184143 2023-01-21 08:40:06.214707: step: 816/526, loss: 0.8109980821609497 2023-01-21 08:40:07.340838: step: 820/526, loss: 0.8053920269012451 2023-01-21 08:40:08.459400: step: 824/526, loss: 0.6061736345291138 2023-01-21 08:40:09.578025: step: 828/526, loss: 0.5200212597846985 2023-01-21 08:40:10.696222: step: 832/526, loss: 0.5605607032775879 2023-01-21 08:40:11.820143: step: 836/526, loss: 2.9807093143463135 2023-01-21 08:40:12.970462: step: 840/526, loss: 3.461522340774536 2023-01-21 08:40:14.089327: step: 844/526, loss: 2.4613826274871826 2023-01-21 08:40:15.259474: step: 848/526, loss: 0.5394300222396851 2023-01-21 08:40:16.395552: step: 852/526, loss: 2.1785902976989746 2023-01-21 08:40:17.521587: step: 856/526, loss: 1.9200830459594727 2023-01-21 08:40:18.649304: step: 860/526, loss: 0.44600868225097656 2023-01-21 08:40:19.778370: step: 864/526, loss: 5.11099910736084 2023-01-21 08:40:20.879776: step: 868/526, loss: 0.9174662828445435 2023-01-21 08:40:21.999762: step: 872/526, loss: 0.5322202444076538 2023-01-21 08:40:23.111533: step: 876/526, loss: 0.2971850037574768 2023-01-21 08:40:24.235323: step: 880/526, loss: 3.78743314743042 2023-01-21 08:40:25.369076: step: 884/526, loss: 4.316065788269043 2023-01-21 08:40:26.494707: step: 888/526, loss: 3.725107192993164 2023-01-21 08:40:27.675934: step: 892/526, loss: 1.078489899635315 2023-01-21 08:40:28.828937: step: 896/526, loss: 1.4396113157272339 2023-01-21 08:40:29.972136: step: 900/526, loss: 1.194507360458374 2023-01-21 08:40:31.090172: step: 904/526, loss: 0.6034525632858276 2023-01-21 08:40:32.250166: step: 908/526, loss: 0.8795521259307861 2023-01-21 08:40:33.403019: step: 912/526, loss: 1.3526854515075684 2023-01-21 08:40:34.538377: step: 916/526, loss: 1.0072823762893677 2023-01-21 08:40:35.661111: step: 920/526, loss: 0.21270331740379333 2023-01-21 08:40:36.815402: step: 924/526, loss: 0.41289713978767395 2023-01-21 08:40:37.946253: step: 928/526, loss: 4.0356292724609375 2023-01-21 08:40:39.058506: step: 932/526, loss: 2.0324747562408447 2023-01-21 08:40:40.185693: step: 936/526, loss: 0.6113661527633667 2023-01-21 08:40:41.310766: step: 940/526, loss: 2.455641984939575 2023-01-21 08:40:42.431432: step: 944/526, loss: 0.4836846888065338 2023-01-21 08:40:43.568674: step: 948/526, loss: 0.7143409252166748 2023-01-21 08:40:44.759413: step: 952/526, loss: 0.5144330859184265 2023-01-21 08:40:45.884440: step: 956/526, loss: 0.7575192451477051 2023-01-21 08:40:47.029344: step: 960/526, loss: 1.4706535339355469 2023-01-21 08:40:48.175646: step: 964/526, loss: 3.752131700515747 2023-01-21 08:40:49.285790: step: 968/526, loss: 1.010324478149414 2023-01-21 08:40:50.404092: step: 972/526, loss: 1.2312567234039307 2023-01-21 08:40:51.528837: step: 976/526, loss: 2.3406429290771484 2023-01-21 08:40:52.666866: step: 980/526, loss: 0.9074604511260986 2023-01-21 08:40:53.783738: step: 984/526, loss: 0.6529759168624878 2023-01-21 08:40:54.921090: step: 988/526, loss: 0.2764093577861786 2023-01-21 08:40:56.047721: step: 992/526, loss: 0.3429710865020752 2023-01-21 08:40:57.160766: step: 996/526, loss: 0.3740905821323395 2023-01-21 08:40:58.277753: step: 1000/526, loss: 0.5420551896095276 2023-01-21 08:40:59.411506: step: 1004/526, loss: 0.45901262760162354 2023-01-21 08:41:00.550037: step: 1008/526, loss: 0.890532374382019 2023-01-21 08:41:01.657761: step: 1012/526, loss: 0.3833976089954376 2023-01-21 08:41:02.780231: step: 1016/526, loss: 0.4700547456741333 2023-01-21 08:41:03.898701: step: 1020/526, loss: 0.4177636206150055 2023-01-21 08:41:05.022887: step: 1024/526, loss: 1.8971961736679077 2023-01-21 08:41:06.153121: step: 1028/526, loss: 2.336444139480591 2023-01-21 08:41:07.270113: step: 1032/526, loss: 0.19568252563476562 2023-01-21 08:41:08.373500: step: 1036/526, loss: 0.4682191014289856 2023-01-21 08:41:09.503467: step: 1040/526, loss: 1.6938953399658203 2023-01-21 08:41:10.648228: step: 1044/526, loss: 0.4424464702606201 2023-01-21 08:41:11.755930: step: 1048/526, loss: 0.5344431400299072 2023-01-21 08:41:12.910025: step: 1052/526, loss: 1.361646056175232 2023-01-21 08:41:14.044078: step: 1056/526, loss: 0.36036643385887146 2023-01-21 08:41:15.172894: step: 1060/526, loss: 0.8183083534240723 2023-01-21 08:41:16.319288: step: 1064/526, loss: 0.3149799704551697 2023-01-21 08:41:17.459358: step: 1068/526, loss: 0.4732183516025543 2023-01-21 08:41:18.568215: step: 1072/526, loss: 0.19402772188186646 2023-01-21 08:41:19.689958: step: 1076/526, loss: 0.9440011978149414 2023-01-21 08:41:20.789731: step: 1080/526, loss: 1.884748101234436 2023-01-21 08:41:21.898170: step: 1084/526, loss: 1.3135898113250732 2023-01-21 08:41:23.010919: step: 1088/526, loss: 0.36445218324661255 2023-01-21 08:41:24.119362: step: 1092/526, loss: 0.5245578289031982 2023-01-21 08:41:25.278937: step: 1096/526, loss: 0.5936914682388306 2023-01-21 08:41:26.408650: step: 1100/526, loss: 0.8939400911331177 2023-01-21 08:41:27.527219: step: 1104/526, loss: 1.305603265762329 2023-01-21 08:41:28.662993: step: 1108/526, loss: 2.9500813484191895 2023-01-21 08:41:29.767800: step: 1112/526, loss: 0.7107867002487183 2023-01-21 08:41:30.887169: step: 1116/526, loss: 0.07613463699817657 2023-01-21 08:41:32.002670: step: 1120/526, loss: 0.30348050594329834 2023-01-21 08:41:33.128274: step: 1124/526, loss: 1.5272237062454224 2023-01-21 08:41:34.257114: step: 1128/526, loss: 1.7898383140563965 2023-01-21 08:41:35.396684: step: 1132/526, loss: 0.47858771681785583 2023-01-21 08:41:36.527833: step: 1136/526, loss: 0.3667069971561432 2023-01-21 08:41:37.661892: step: 1140/526, loss: 0.5482919812202454 2023-01-21 08:41:38.777806: step: 1144/526, loss: 0.37970858812332153 2023-01-21 08:41:39.901668: step: 1148/526, loss: 0.9039019346237183 2023-01-21 08:41:41.026247: step: 1152/526, loss: 4.6626691818237305 2023-01-21 08:41:42.168348: step: 1156/526, loss: 0.8359564542770386 2023-01-21 08:41:43.322424: step: 1160/526, loss: 4.585060119628906 2023-01-21 08:41:44.451098: step: 1164/526, loss: 1.1588298082351685 2023-01-21 08:41:45.568572: step: 1168/526, loss: 1.0112495422363281 2023-01-21 08:41:46.697289: step: 1172/526, loss: 0.4761594235897064 2023-01-21 08:41:47.805119: step: 1176/526, loss: 0.31148630380630493 2023-01-21 08:41:48.965622: step: 1180/526, loss: 0.5811941027641296 2023-01-21 08:41:50.130926: step: 1184/526, loss: 0.33053335547447205 2023-01-21 08:41:51.289966: step: 1188/526, loss: 0.4351780414581299 2023-01-21 08:41:52.406919: step: 1192/526, loss: 1.6941075325012207 2023-01-21 08:41:53.546233: step: 1196/526, loss: 0.2341902256011963 2023-01-21 08:41:54.698436: step: 1200/526, loss: 0.772669792175293 2023-01-21 08:41:55.843299: step: 1204/526, loss: 0.9370273947715759 2023-01-21 08:41:56.987392: step: 1208/526, loss: 0.5496181845664978 2023-01-21 08:41:58.097843: step: 1212/526, loss: 0.28019237518310547 2023-01-21 08:41:59.256492: step: 1216/526, loss: 0.5980945825576782 2023-01-21 08:42:00.407974: step: 1220/526, loss: 0.33800241351127625 2023-01-21 08:42:01.542599: step: 1224/526, loss: 0.5343923568725586 2023-01-21 08:42:02.694557: step: 1228/526, loss: 0.3986811339855194 2023-01-21 08:42:03.846660: step: 1232/526, loss: 0.6579622030258179 2023-01-21 08:42:04.982664: step: 1236/526, loss: 1.0317293405532837 2023-01-21 08:42:06.153158: step: 1240/526, loss: 1.3522977828979492 2023-01-21 08:42:07.262052: step: 1244/526, loss: 0.8583003282546997 2023-01-21 08:42:08.399721: step: 1248/526, loss: 0.3571805953979492 2023-01-21 08:42:09.531653: step: 1252/526, loss: 0.5468298196792603 2023-01-21 08:42:10.684586: step: 1256/526, loss: 0.5932650566101074 2023-01-21 08:42:11.804836: step: 1260/526, loss: 0.5215433835983276 2023-01-21 08:42:12.970786: step: 1264/526, loss: 1.1786279678344727 2023-01-21 08:42:14.120409: step: 1268/526, loss: 3.7608094215393066 2023-01-21 08:42:15.280603: step: 1272/526, loss: 1.6436172723770142 2023-01-21 08:42:16.405202: step: 1276/526, loss: 0.4854907989501953 2023-01-21 08:42:17.527766: step: 1280/526, loss: 0.728380024433136 2023-01-21 08:42:18.701011: step: 1284/526, loss: 0.3912392556667328 2023-01-21 08:42:19.817190: step: 1288/526, loss: 1.0520737171173096 2023-01-21 08:42:20.947218: step: 1292/526, loss: 0.3631967604160309 2023-01-21 08:42:22.068971: step: 1296/526, loss: 0.8719676733016968 2023-01-21 08:42:23.189447: step: 1300/526, loss: 0.7937418222427368 2023-01-21 08:42:24.330016: step: 1304/526, loss: 0.5097143650054932 2023-01-21 08:42:25.449249: step: 1308/526, loss: 1.2263057231903076 2023-01-21 08:42:26.571822: step: 1312/526, loss: 0.2847522795200348 2023-01-21 08:42:27.690402: step: 1316/526, loss: 0.31432899832725525 2023-01-21 08:42:28.833663: step: 1320/526, loss: 3.461671829223633 2023-01-21 08:42:29.973735: step: 1324/526, loss: 0.6790217161178589 2023-01-21 08:42:31.119996: step: 1328/526, loss: 0.9770418405532837 2023-01-21 08:42:32.239654: step: 1332/526, loss: 0.4239872097969055 2023-01-21 08:42:33.358190: step: 1336/526, loss: 0.6387863159179688 2023-01-21 08:42:34.480407: step: 1340/526, loss: 0.45950907468795776 2023-01-21 08:42:35.645756: step: 1344/526, loss: 0.6816931366920471 2023-01-21 08:42:36.762136: step: 1348/526, loss: 0.9068479537963867 2023-01-21 08:42:37.881900: step: 1352/526, loss: 0.4446907043457031 2023-01-21 08:42:39.018863: step: 1356/526, loss: 0.4947296977043152 2023-01-21 08:42:40.162891: step: 1360/526, loss: 1.7559940814971924 2023-01-21 08:42:41.302882: step: 1364/526, loss: 2.938032388687134 2023-01-21 08:42:42.468596: step: 1368/526, loss: 0.8173549175262451 2023-01-21 08:42:43.591563: step: 1372/526, loss: 0.510219395160675 2023-01-21 08:42:44.721679: step: 1376/526, loss: 1.6447769403457642 2023-01-21 08:42:45.903321: step: 1380/526, loss: 0.4797564446926117 2023-01-21 08:42:47.047567: step: 1384/526, loss: 0.532613217830658 2023-01-21 08:42:48.179613: step: 1388/526, loss: 0.7305388450622559 2023-01-21 08:42:49.336368: step: 1392/526, loss: 0.8144544363021851 2023-01-21 08:42:50.483244: step: 1396/526, loss: 1.5057934522628784 2023-01-21 08:42:51.590353: step: 1400/526, loss: 1.0043981075286865 2023-01-21 08:42:52.733085: step: 1404/526, loss: 4.387916564941406 2023-01-21 08:42:53.855346: step: 1408/526, loss: 0.29746735095977783 2023-01-21 08:42:54.984015: step: 1412/526, loss: 0.5198326110839844 2023-01-21 08:42:56.109552: step: 1416/526, loss: 0.27850157022476196 2023-01-21 08:42:57.257208: step: 1420/526, loss: 2.4872961044311523 2023-01-21 08:42:58.390517: step: 1424/526, loss: 2.2086496353149414 2023-01-21 08:42:59.511594: step: 1428/526, loss: 1.7859861850738525 2023-01-21 08:43:00.656933: step: 1432/526, loss: 0.4378030598163605 2023-01-21 08:43:01.779229: step: 1436/526, loss: 0.4930229187011719 2023-01-21 08:43:02.922762: step: 1440/526, loss: 0.6065353155136108 2023-01-21 08:43:04.051118: step: 1444/526, loss: 0.47361889481544495 2023-01-21 08:43:05.168318: step: 1448/526, loss: 0.37605801224708557 2023-01-21 08:43:06.318732: step: 1452/526, loss: 1.4961729049682617 2023-01-21 08:43:07.470445: step: 1456/526, loss: 1.583017349243164 2023-01-21 08:43:08.578139: step: 1460/526, loss: 0.2598903179168701 2023-01-21 08:43:09.711997: step: 1464/526, loss: 0.45981264114379883 2023-01-21 08:43:10.864785: step: 1468/526, loss: 0.44727057218551636 2023-01-21 08:43:11.991326: step: 1472/526, loss: 0.404015451669693 2023-01-21 08:43:13.122069: step: 1476/526, loss: 2.0474090576171875 2023-01-21 08:43:14.250137: step: 1480/526, loss: 1.4260640144348145 2023-01-21 08:43:15.408497: step: 1484/526, loss: 1.044939637184143 2023-01-21 08:43:16.531486: step: 1488/526, loss: 3.0949370861053467 2023-01-21 08:43:17.679082: step: 1492/526, loss: 0.4306080937385559 2023-01-21 08:43:18.856756: step: 1496/526, loss: 0.46039143204689026 2023-01-21 08:43:19.987220: step: 1500/526, loss: 0.3191283941268921 2023-01-21 08:43:21.098480: step: 1504/526, loss: 0.2647761404514313 2023-01-21 08:43:22.244012: step: 1508/526, loss: 2.128882646560669 2023-01-21 08:43:23.379245: step: 1512/526, loss: 0.4349292516708374 2023-01-21 08:43:24.516993: step: 1516/526, loss: 1.600499153137207 2023-01-21 08:43:25.666083: step: 1520/526, loss: 1.4844815731048584 2023-01-21 08:43:26.795547: step: 1524/526, loss: 2.144289970397949 2023-01-21 08:43:27.950809: step: 1528/526, loss: 0.36953020095825195 2023-01-21 08:43:29.077225: step: 1532/526, loss: 0.37135472893714905 2023-01-21 08:43:30.211165: step: 1536/526, loss: 0.39337289333343506 2023-01-21 08:43:31.335830: step: 1540/526, loss: 0.500944972038269 2023-01-21 08:43:32.469101: step: 1544/526, loss: 2.1210741996765137 2023-01-21 08:43:33.609749: step: 1548/526, loss: 0.3153996765613556 2023-01-21 08:43:34.759337: step: 1552/526, loss: 1.2591428756713867 2023-01-21 08:43:35.872909: step: 1556/526, loss: 2.7339954376220703 2023-01-21 08:43:36.986133: step: 1560/526, loss: 0.3999224603176117 2023-01-21 08:43:38.095185: step: 1564/526, loss: 2.504122257232666 2023-01-21 08:43:39.253507: step: 1568/526, loss: 1.0817643404006958 2023-01-21 08:43:40.399213: step: 1572/526, loss: 0.7084323763847351 2023-01-21 08:43:41.558297: step: 1576/526, loss: 0.23036609590053558 2023-01-21 08:43:42.689780: step: 1580/526, loss: 7.019631385803223 2023-01-21 08:43:43.850799: step: 1584/526, loss: 0.628699779510498 2023-01-21 08:43:44.962153: step: 1588/526, loss: 0.6608818769454956 2023-01-21 08:43:46.073163: step: 1592/526, loss: 0.3139715790748596 2023-01-21 08:43:47.238011: step: 1596/526, loss: 5.062445640563965 2023-01-21 08:43:48.394780: step: 1600/526, loss: 2.2490780353546143 2023-01-21 08:43:49.549400: step: 1604/526, loss: 0.4725354313850403 2023-01-21 08:43:50.700472: step: 1608/526, loss: 1.6625943183898926 2023-01-21 08:43:51.812832: step: 1612/526, loss: 7.107010364532471 2023-01-21 08:43:52.933741: step: 1616/526, loss: 1.544629454612732 2023-01-21 08:43:54.088290: step: 1620/526, loss: 0.4934345483779907 2023-01-21 08:43:55.241207: step: 1624/526, loss: 2.2757115364074707 2023-01-21 08:43:56.340411: step: 1628/526, loss: 2.29428768157959 2023-01-21 08:43:57.441034: step: 1632/526, loss: 2.0969605445861816 2023-01-21 08:43:58.595572: step: 1636/526, loss: 0.96044921875 2023-01-21 08:43:59.717736: step: 1640/526, loss: 0.9435324668884277 2023-01-21 08:44:00.835008: step: 1644/526, loss: 2.538663387298584 2023-01-21 08:44:01.988910: step: 1648/526, loss: 0.36689937114715576 2023-01-21 08:44:03.149357: step: 1652/526, loss: 0.7993355989456177 2023-01-21 08:44:04.264981: step: 1656/526, loss: 0.2025838941335678 2023-01-21 08:44:05.384949: step: 1660/526, loss: 1.9917449951171875 2023-01-21 08:44:06.536476: step: 1664/526, loss: 1.9251827001571655 2023-01-21 08:44:07.663548: step: 1668/526, loss: 0.31708812713623047 2023-01-21 08:44:08.806440: step: 1672/526, loss: 1.0962234735488892 2023-01-21 08:44:09.954895: step: 1676/526, loss: 0.4059239625930786 2023-01-21 08:44:11.100329: step: 1680/526, loss: 2.2722420692443848 2023-01-21 08:44:12.235426: step: 1684/526, loss: 0.312305748462677 2023-01-21 08:44:13.378841: step: 1688/526, loss: 0.444745808839798 2023-01-21 08:44:14.525528: step: 1692/526, loss: 0.24685117602348328 2023-01-21 08:44:15.657498: step: 1696/526, loss: 0.2933158874511719 2023-01-21 08:44:16.797275: step: 1700/526, loss: 1.0083562135696411 2023-01-21 08:44:17.925560: step: 1704/526, loss: 0.9939299821853638 2023-01-21 08:44:19.074318: step: 1708/526, loss: 2.3145675659179688 2023-01-21 08:44:20.185883: step: 1712/526, loss: 0.7160344123840332 2023-01-21 08:44:21.318178: step: 1716/526, loss: 0.3281521797180176 2023-01-21 08:44:22.501193: step: 1720/526, loss: 4.843984127044678 2023-01-21 08:44:23.629322: step: 1724/526, loss: 3.154914617538452 2023-01-21 08:44:24.749002: step: 1728/526, loss: 0.23179420828819275 2023-01-21 08:44:25.878050: step: 1732/526, loss: 0.7848429679870605 2023-01-21 08:44:26.989128: step: 1736/526, loss: 0.5189828276634216 2023-01-21 08:44:28.142896: step: 1740/526, loss: 0.5119626522064209 2023-01-21 08:44:29.244628: step: 1744/526, loss: 2.896693706512451 2023-01-21 08:44:30.401633: step: 1748/526, loss: 0.4422931671142578 2023-01-21 08:44:31.551536: step: 1752/526, loss: 0.3722724914550781 2023-01-21 08:44:32.713782: step: 1756/526, loss: 0.8425112962722778 2023-01-21 08:44:33.831352: step: 1760/526, loss: 2.4931066036224365 2023-01-21 08:44:34.956008: step: 1764/526, loss: 1.1446192264556885 2023-01-21 08:44:36.102831: step: 1768/526, loss: 0.49720528721809387 2023-01-21 08:44:37.223154: step: 1772/526, loss: 0.12635165452957153 2023-01-21 08:44:38.348833: step: 1776/526, loss: 2.4786667823791504 2023-01-21 08:44:39.497262: step: 1780/526, loss: 0.29175227880477905 2023-01-21 08:44:40.610165: step: 1784/526, loss: 1.8485543727874756 2023-01-21 08:44:41.756474: step: 1788/526, loss: 0.2906613051891327 2023-01-21 08:44:42.904939: step: 1792/526, loss: 0.3093242645263672 2023-01-21 08:44:44.001666: step: 1796/526, loss: 0.31108731031417847 2023-01-21 08:44:45.126232: step: 1800/526, loss: 0.3179604411125183 2023-01-21 08:44:46.262171: step: 1804/526, loss: 1.2491549253463745 2023-01-21 08:44:47.390741: step: 1808/526, loss: 1.4476888179779053 2023-01-21 08:44:48.504318: step: 1812/526, loss: 0.44461360573768616 2023-01-21 08:44:49.630892: step: 1816/526, loss: 1.6376667022705078 2023-01-21 08:44:50.754814: step: 1820/526, loss: 0.47399258613586426 2023-01-21 08:44:51.883384: step: 1824/526, loss: 0.8550583720207214 2023-01-21 08:44:53.005576: step: 1828/526, loss: 1.1353684663772583 2023-01-21 08:44:54.126187: step: 1832/526, loss: 0.15098491311073303 2023-01-21 08:44:55.221483: step: 1836/526, loss: 0.7288172245025635 2023-01-21 08:44:56.339746: step: 1840/526, loss: 1.0133200883865356 2023-01-21 08:44:57.521331: step: 1844/526, loss: 0.1850666105747223 2023-01-21 08:44:58.694704: step: 1848/526, loss: 0.9694145321846008 2023-01-21 08:44:59.822682: step: 1852/526, loss: 1.2353425025939941 2023-01-21 08:45:00.943569: step: 1856/526, loss: 0.2507871687412262 2023-01-21 08:45:02.046714: step: 1860/526, loss: 0.31660011410713196 2023-01-21 08:45:03.177553: step: 1864/526, loss: 0.23736849427223206 2023-01-21 08:45:04.308394: step: 1868/526, loss: 0.5693188309669495 2023-01-21 08:45:05.440181: step: 1872/526, loss: 0.8955358862876892 2023-01-21 08:45:06.576534: step: 1876/526, loss: 1.1898239850997925 2023-01-21 08:45:07.706998: step: 1880/526, loss: 0.28685086965560913 2023-01-21 08:45:08.847019: step: 1884/526, loss: 0.8071651458740234 2023-01-21 08:45:09.966365: step: 1888/526, loss: 3.286255359649658 2023-01-21 08:45:11.107878: step: 1892/526, loss: 0.22234492003917694 2023-01-21 08:45:12.241697: step: 1896/526, loss: 0.23985934257507324 2023-01-21 08:45:13.362222: step: 1900/526, loss: 2.318176031112671 2023-01-21 08:45:14.488168: step: 1904/526, loss: 1.1696834564208984 2023-01-21 08:45:15.601999: step: 1908/526, loss: 2.8154244422912598 2023-01-21 08:45:16.726091: step: 1912/526, loss: 0.9494587182998657 2023-01-21 08:45:17.928511: step: 1916/526, loss: 3.0047805309295654 2023-01-21 08:45:19.079192: step: 1920/526, loss: 2.5172293186187744 2023-01-21 08:45:20.233956: step: 1924/526, loss: 0.3164713978767395 2023-01-21 08:45:21.370043: step: 1928/526, loss: 1.2114191055297852 2023-01-21 08:45:22.509846: step: 1932/526, loss: 1.6604071855545044 2023-01-21 08:45:23.644357: step: 1936/526, loss: 2.9787697792053223 2023-01-21 08:45:24.777783: step: 1940/526, loss: 0.4908496141433716 2023-01-21 08:45:25.935100: step: 1944/526, loss: 1.097630262374878 2023-01-21 08:45:27.054293: step: 1948/526, loss: 2.469313859939575 2023-01-21 08:45:28.173349: step: 1952/526, loss: 2.3053290843963623 2023-01-21 08:45:29.295814: step: 1956/526, loss: 0.5947933197021484 2023-01-21 08:45:30.454714: step: 1960/526, loss: 0.5297248959541321 2023-01-21 08:45:31.562752: step: 1964/526, loss: 0.5264659523963928 2023-01-21 08:45:32.690662: step: 1968/526, loss: 0.4361874759197235 2023-01-21 08:45:33.798134: step: 1972/526, loss: 0.23376017808914185 2023-01-21 08:45:34.928231: step: 1976/526, loss: 1.3044383525848389 2023-01-21 08:45:36.073898: step: 1980/526, loss: 0.31719255447387695 2023-01-21 08:45:37.206747: step: 1984/526, loss: 0.5426143407821655 2023-01-21 08:45:38.341446: step: 1988/526, loss: 0.7885656356811523 2023-01-21 08:45:39.455846: step: 1992/526, loss: 2.0058581829071045 2023-01-21 08:45:40.575118: step: 1996/526, loss: 0.5770419836044312 2023-01-21 08:45:41.675243: step: 2000/526, loss: 1.8433730602264404 2023-01-21 08:45:42.837387: step: 2004/526, loss: 1.8296287059783936 2023-01-21 08:45:43.944967: step: 2008/526, loss: 0.6065354943275452 2023-01-21 08:45:45.080198: step: 2012/526, loss: 0.9866287708282471 2023-01-21 08:45:46.215293: step: 2016/526, loss: 0.9422327280044556 2023-01-21 08:45:47.340587: step: 2020/526, loss: 0.593647837638855 2023-01-21 08:45:48.477396: step: 2024/526, loss: 0.3329729437828064 2023-01-21 08:45:49.638988: step: 2028/526, loss: 0.49724704027175903 2023-01-21 08:45:50.774913: step: 2032/526, loss: 0.6353542804718018 2023-01-21 08:45:51.898503: step: 2036/526, loss: 0.20455598831176758 2023-01-21 08:45:53.032246: step: 2040/526, loss: 0.7911707162857056 2023-01-21 08:45:54.161035: step: 2044/526, loss: 0.6074792742729187 2023-01-21 08:45:55.287435: step: 2048/526, loss: 0.3341962695121765 2023-01-21 08:45:56.416521: step: 2052/526, loss: 0.5979087948799133 2023-01-21 08:45:57.542512: step: 2056/526, loss: 0.28880637884140015 2023-01-21 08:45:58.673512: step: 2060/526, loss: 0.8539525866508484 2023-01-21 08:45:59.821388: step: 2064/526, loss: 0.5040515661239624 2023-01-21 08:46:00.956248: step: 2068/526, loss: 0.37739840149879456 2023-01-21 08:46:02.060885: step: 2072/526, loss: 2.66009783744812 2023-01-21 08:46:03.202652: step: 2076/526, loss: 1.1328926086425781 2023-01-21 08:46:04.335880: step: 2080/526, loss: 0.32803452014923096 2023-01-21 08:46:05.492900: step: 2084/526, loss: 0.8675100803375244 2023-01-21 08:46:06.608046: step: 2088/526, loss: 1.6779357194900513 2023-01-21 08:46:07.752722: step: 2092/526, loss: 0.5400898456573486 2023-01-21 08:46:08.900484: step: 2096/526, loss: 0.5662949681282043 2023-01-21 08:46:10.041973: step: 2100/526, loss: 0.6584249138832092 2023-01-21 08:46:11.205265: step: 2104/526, loss: 1.1076157093048096 ================================================== Loss: 3.266 -------------------- Dev: {'event': {'p': 0.5421994884910486, 'r': 0.5645805592543276, 'f1': 0.553163731245923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Test: {'event': {'p': 0.4775, 'r': 0.3987473903966597, 'f1': 0.43458475540386804}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Chinese: {'event': {'p': 0.5319148936170213, 'r': 0.46296296296296297, 'f1': 0.495049504950495}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Korean: {'event': {'p': 0.8421052631578947, 'r': 0.25396825396825395, 'f1': 0.3902439024390244}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Russian: {'event': {'p': 0.5, 'r': 0.16666666666666666, 'f1': 0.25}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.5421994884910486, 'r': 0.5645805592543276, 'f1': 0.553163731245923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Chinese: {'event': {'p': 0.4775, 'r': 0.3987473903966597, 'f1': 0.43458475540386804}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Chinese: {'event': {'p': 0.5319148936170213, 'r': 0.46296296296296297, 'f1': 0.495049504950495}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Korean: {'event': {'p': 0.5421994884910486, 'r': 0.5645805592543276, 'f1': 0.553163731245923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Korean: {'event': {'p': 0.4775, 'r': 0.3987473903966597, 'f1': 0.43458475540386804}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Korean: {'event': {'p': 0.8421052631578947, 'r': 0.25396825396825395, 'f1': 0.3902439024390244}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} -------------------- Eng Dev for Russian: {'event': {'p': 0.5421994884910486, 'r': 0.5645805592543276, 'f1': 0.553163731245923}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Eng Test for Russian: {'event': {'p': 0.4775, 'r': 0.3987473903966597, 'f1': 0.43458475540386804}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} Sample Russian: {'event': {'p': 0.5, 'r': 0.16666666666666666, 'f1': 0.25}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 0} ****************************** Epoch: 1 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:47:09.639327: step: 4/526, loss: 2.083930015563965 2023-01-21 08:47:10.786236: step: 8/526, loss: 0.8274605870246887 2023-01-21 08:47:11.927907: step: 12/526, loss: 0.3782390356063843 2023-01-21 08:47:13.107536: step: 16/526, loss: 0.7598814964294434 2023-01-21 08:47:14.223909: step: 20/526, loss: 0.1971059888601303 2023-01-21 08:47:15.363319: step: 24/526, loss: 0.23529118299484253 2023-01-21 08:47:16.498113: step: 28/526, loss: 0.26415252685546875 2023-01-21 08:47:17.644987: step: 32/526, loss: 0.12952308356761932 2023-01-21 08:47:18.781802: step: 36/526, loss: 0.1944410502910614 2023-01-21 08:47:19.933027: step: 40/526, loss: 0.1660228818655014 2023-01-21 08:47:21.078724: step: 44/526, loss: 1.4282119274139404 2023-01-21 08:47:22.205133: step: 48/526, loss: 0.7807324528694153 2023-01-21 08:47:23.316377: step: 52/526, loss: 0.0864735096693039 2023-01-21 08:47:24.452149: step: 56/526, loss: 0.8017598986625671 2023-01-21 08:47:25.588801: step: 60/526, loss: 0.28034257888793945 2023-01-21 08:47:26.708398: step: 64/526, loss: 1.1089951992034912 2023-01-21 08:47:27.825688: step: 68/526, loss: 0.3048839569091797 2023-01-21 08:47:28.950040: step: 72/526, loss: 1.0833467245101929 2023-01-21 08:47:30.068281: step: 76/526, loss: 1.0029594898223877 2023-01-21 08:47:31.203176: step: 80/526, loss: 0.09686202555894852 2023-01-21 08:47:32.325825: step: 84/526, loss: 0.5943330526351929 2023-01-21 08:47:33.461875: step: 88/526, loss: 0.5998398065567017 2023-01-21 08:47:34.593442: step: 92/526, loss: 0.5607078671455383 2023-01-21 08:47:35.707184: step: 96/526, loss: 0.7707836627960205 2023-01-21 08:47:36.834130: step: 100/526, loss: 0.23311491310596466 2023-01-21 08:47:37.982114: step: 104/526, loss: 0.2576221227645874 2023-01-21 08:47:39.129560: step: 108/526, loss: 0.6494616866111755 2023-01-21 08:47:40.264525: step: 112/526, loss: 0.4253585934638977 2023-01-21 08:47:41.384499: step: 116/526, loss: 0.23099356889724731 2023-01-21 08:47:42.531834: step: 120/526, loss: 0.7637468576431274 2023-01-21 08:47:43.706252: step: 124/526, loss: 0.2622276842594147 2023-01-21 08:47:44.841071: step: 128/526, loss: 2.7345898151397705 2023-01-21 08:47:45.951863: step: 132/526, loss: 0.3178275227546692 2023-01-21 08:47:47.084750: step: 136/526, loss: 0.5001682043075562 2023-01-21 08:47:48.254436: step: 140/526, loss: 7.056186676025391 2023-01-21 08:47:49.388725: step: 144/526, loss: 0.9011992812156677 2023-01-21 08:47:50.525021: step: 148/526, loss: 1.5721018314361572 2023-01-21 08:47:51.656836: step: 152/526, loss: 0.16192373633384705 2023-01-21 08:47:52.841875: step: 156/526, loss: 0.26265040040016174 2023-01-21 08:47:53.976119: step: 160/526, loss: 0.46940135955810547 2023-01-21 08:47:55.109577: step: 164/526, loss: 0.4644995927810669 2023-01-21 08:47:56.229102: step: 168/526, loss: 0.3633946478366852 2023-01-21 08:47:57.354510: step: 172/526, loss: 0.9035346508026123 2023-01-21 08:47:58.489278: step: 176/526, loss: 1.9735026359558105 2023-01-21 08:47:59.607614: step: 180/526, loss: 2.5919289588928223 2023-01-21 08:48:00.728402: step: 184/526, loss: 0.6723855137825012 2023-01-21 08:48:01.873098: step: 188/526, loss: 0.24602484703063965 2023-01-21 08:48:02.985708: step: 192/526, loss: 1.0357261896133423 2023-01-21 08:48:04.105582: step: 196/526, loss: 0.9602370262145996 2023-01-21 08:48:05.236784: step: 200/526, loss: 0.9325742721557617 2023-01-21 08:48:06.411877: step: 204/526, loss: 3.030963182449341 2023-01-21 08:48:07.605387: step: 208/526, loss: 0.4387546181678772 2023-01-21 08:48:08.796349: step: 212/526, loss: 0.6036388278007507 2023-01-21 08:48:09.917585: step: 216/526, loss: 0.33836671710014343 2023-01-21 08:48:11.074209: step: 220/526, loss: 0.7502118945121765 2023-01-21 08:48:12.228314: step: 224/526, loss: 2.5615463256835938 2023-01-21 08:48:13.359960: step: 228/526, loss: 0.8718830943107605 2023-01-21 08:48:14.481803: step: 232/526, loss: 0.7705024480819702 2023-01-21 08:48:15.626381: step: 236/526, loss: 0.9986544251441956 2023-01-21 08:48:16.758124: step: 240/526, loss: 0.3948686718940735 2023-01-21 08:48:17.892311: step: 244/526, loss: 1.4745467901229858 2023-01-21 08:48:19.001456: step: 248/526, loss: 0.3927941620349884 2023-01-21 08:48:20.129079: step: 252/526, loss: 0.1897200047969818 2023-01-21 08:48:21.271798: step: 256/526, loss: 3.3282723426818848 2023-01-21 08:48:22.384497: step: 260/526, loss: 0.5444931983947754 2023-01-21 08:48:23.524367: step: 264/526, loss: 0.22589468955993652 2023-01-21 08:48:24.690824: step: 268/526, loss: 0.6399433016777039 2023-01-21 08:48:25.807548: step: 272/526, loss: 1.3531510829925537 2023-01-21 08:48:26.940849: step: 276/526, loss: 0.9859811067581177 2023-01-21 08:48:28.095944: step: 280/526, loss: 0.3848627209663391 2023-01-21 08:48:29.254604: step: 284/526, loss: 1.6057392358779907 2023-01-21 08:48:30.422908: step: 288/526, loss: 1.0108039379119873 2023-01-21 08:48:31.558271: step: 292/526, loss: 0.26069778203964233 2023-01-21 08:48:32.697075: step: 296/526, loss: 0.41763579845428467 2023-01-21 08:48:33.853130: step: 300/526, loss: 0.3531990051269531 2023-01-21 08:48:34.963503: step: 304/526, loss: 0.4743443727493286 2023-01-21 08:48:36.120345: step: 308/526, loss: 0.9863955974578857 2023-01-21 08:48:37.282975: step: 312/526, loss: 0.6807553768157959 2023-01-21 08:48:38.407258: step: 316/526, loss: 1.6670469045639038 2023-01-21 08:48:39.576694: step: 320/526, loss: 0.17421141266822815 2023-01-21 08:48:40.725605: step: 324/526, loss: 0.8518913388252258 2023-01-21 08:48:41.845586: step: 328/526, loss: 0.651017963886261 2023-01-21 08:48:43.000432: step: 332/526, loss: 0.48933982849121094 2023-01-21 08:48:44.152460: step: 336/526, loss: 0.18417192995548248 2023-01-21 08:48:45.312736: step: 340/526, loss: 0.6215408444404602 2023-01-21 08:48:46.440241: step: 344/526, loss: 0.36384862661361694 2023-01-21 08:48:47.583877: step: 348/526, loss: 0.5557636618614197 2023-01-21 08:48:48.716442: step: 352/526, loss: 2.640521287918091 2023-01-21 08:48:49.854723: step: 356/526, loss: 0.5376696586608887 2023-01-21 08:48:50.989817: step: 360/526, loss: 0.9676294922828674 2023-01-21 08:48:52.135168: step: 364/526, loss: 0.44783440232276917 2023-01-21 08:48:53.301548: step: 368/526, loss: 0.3965347111225128 2023-01-21 08:48:54.440349: step: 372/526, loss: 0.5311633944511414 2023-01-21 08:48:55.577930: step: 376/526, loss: 0.22307558357715607 2023-01-21 08:48:56.711289: step: 380/526, loss: 0.2216019332408905 2023-01-21 08:48:57.857588: step: 384/526, loss: 0.38986721634864807 2023-01-21 08:48:58.973263: step: 388/526, loss: 0.30713891983032227 2023-01-21 08:49:00.104560: step: 392/526, loss: 0.19910621643066406 2023-01-21 08:49:01.248532: step: 396/526, loss: 1.0961734056472778 2023-01-21 08:49:02.380117: step: 400/526, loss: 0.8917814493179321 2023-01-21 08:49:03.519163: step: 404/526, loss: 0.2355732023715973 2023-01-21 08:49:04.661863: step: 408/526, loss: 0.793698787689209 2023-01-21 08:49:05.781194: step: 412/526, loss: 1.4421148300170898 2023-01-21 08:49:06.912811: step: 416/526, loss: 0.25117313861846924 2023-01-21 08:49:08.045024: step: 420/526, loss: 0.2528945207595825 2023-01-21 08:49:09.187415: step: 424/526, loss: 2.6126599311828613 2023-01-21 08:49:10.308871: step: 428/526, loss: 0.25804147124290466 2023-01-21 08:49:11.432551: step: 432/526, loss: 0.22289638221263885 2023-01-21 08:49:12.551489: step: 436/526, loss: 1.0976557731628418 2023-01-21 08:49:13.728131: step: 440/526, loss: 0.22537703812122345 2023-01-21 08:49:14.859985: step: 444/526, loss: 0.5675263404846191 2023-01-21 08:49:16.024106: step: 448/526, loss: 0.8727865219116211 2023-01-21 08:49:17.168400: step: 452/526, loss: 0.32158422470092773 2023-01-21 08:49:18.322084: step: 456/526, loss: 0.526586651802063 2023-01-21 08:49:19.446127: step: 460/526, loss: 0.8260121941566467 2023-01-21 08:49:20.587672: step: 464/526, loss: 7.516510009765625 2023-01-21 08:49:21.752058: step: 468/526, loss: 1.013238549232483 2023-01-21 08:49:22.918838: step: 472/526, loss: 0.19077129662036896 2023-01-21 08:49:24.077975: step: 476/526, loss: 0.595874011516571 2023-01-21 08:49:25.208275: step: 480/526, loss: 0.7450437545776367 2023-01-21 08:49:26.378970: step: 484/526, loss: 1.3177281618118286 2023-01-21 08:49:27.535074: step: 488/526, loss: 0.6919177174568176 2023-01-21 08:49:28.658120: step: 492/526, loss: 0.8793939352035522 2023-01-21 08:49:29.784700: step: 496/526, loss: 0.770618200302124 2023-01-21 08:49:30.968987: step: 500/526, loss: 0.8184714913368225 2023-01-21 08:49:32.091047: step: 504/526, loss: 0.20872879028320312 2023-01-21 08:49:33.228967: step: 508/526, loss: 0.13976740837097168 2023-01-21 08:49:34.363174: step: 512/526, loss: 0.3190687298774719 2023-01-21 08:49:35.524923: step: 516/526, loss: 0.29576951265335083 2023-01-21 08:49:36.696281: step: 520/526, loss: 0.3784988820552826 2023-01-21 08:49:37.873502: step: 524/526, loss: 1.279805064201355 2023-01-21 08:49:38.994890: step: 528/526, loss: 0.514992892742157 2023-01-21 08:49:40.142087: step: 532/526, loss: 0.4659263789653778 2023-01-21 08:49:41.287563: step: 536/526, loss: 0.3378411531448364 2023-01-21 08:49:42.444964: step: 540/526, loss: 0.27676069736480713 2023-01-21 08:49:43.603491: step: 544/526, loss: 0.37308037281036377 2023-01-21 08:49:44.723464: step: 548/526, loss: 0.16481605172157288 2023-01-21 08:49:45.854662: step: 552/526, loss: 0.30347806215286255 2023-01-21 08:49:47.025887: step: 556/526, loss: 0.2569205164909363 2023-01-21 08:49:48.159096: step: 560/526, loss: 0.4747818112373352 2023-01-21 08:49:49.280875: step: 564/526, loss: 0.5102481245994568 2023-01-21 08:49:50.398892: step: 568/526, loss: 0.41791316866874695 2023-01-21 08:49:51.535625: step: 572/526, loss: 2.084381580352783 2023-01-21 08:49:52.662121: step: 576/526, loss: 0.9264667630195618 2023-01-21 08:49:53.780412: step: 580/526, loss: 0.6340084075927734 2023-01-21 08:49:54.908650: step: 584/526, loss: 0.9388248920440674 2023-01-21 08:49:56.029815: step: 588/526, loss: 0.4145932197570801 2023-01-21 08:49:57.166267: step: 592/526, loss: 0.5316857099533081 2023-01-21 08:49:58.300622: step: 596/526, loss: 1.4013551473617554 2023-01-21 08:49:59.448768: step: 600/526, loss: 0.5107519626617432 2023-01-21 08:50:00.571735: step: 604/526, loss: 0.26548969745635986 2023-01-21 08:50:01.730913: step: 608/526, loss: 0.8689706921577454 2023-01-21 08:50:02.895881: step: 612/526, loss: 0.7169291973114014 2023-01-21 08:50:04.063363: step: 616/526, loss: 1.7207502126693726 2023-01-21 08:50:05.213924: step: 620/526, loss: 0.5860875248908997 2023-01-21 08:50:06.336875: step: 624/526, loss: 1.6433438062667847 2023-01-21 08:50:07.491123: step: 628/526, loss: 5.409465789794922 2023-01-21 08:50:08.629906: step: 632/526, loss: 0.296293169260025 2023-01-21 08:50:09.787539: step: 636/526, loss: 0.6234157681465149 2023-01-21 08:50:10.909746: step: 640/526, loss: 0.7712377309799194 2023-01-21 08:50:12.092491: step: 644/526, loss: 0.36940136551856995 2023-01-21 08:50:13.217891: step: 648/526, loss: 0.26981449127197266 2023-01-21 08:50:14.331412: step: 652/526, loss: 0.4232860207557678 2023-01-21 08:50:15.457525: step: 656/526, loss: 0.6999709606170654 2023-01-21 08:50:16.602291: step: 660/526, loss: 0.19666405022144318 2023-01-21 08:50:17.712725: step: 664/526, loss: 0.8148563504219055 2023-01-21 08:50:18.820421: step: 668/526, loss: 0.47100716829299927 2023-01-21 08:50:19.947489: step: 672/526, loss: 0.23274736106395721 2023-01-21 08:50:21.084818: step: 676/526, loss: 2.2901201248168945 2023-01-21 08:50:22.203398: step: 680/526, loss: 0.46401405334472656 2023-01-21 08:50:23.337309: step: 684/526, loss: 2.6145966053009033 2023-01-21 08:50:24.482057: step: 688/526, loss: 0.777539849281311 2023-01-21 08:50:25.597248: step: 692/526, loss: 1.050235629081726 2023-01-21 08:50:26.728144: step: 696/526, loss: 0.1734866201877594 2023-01-21 08:50:27.840731: step: 700/526, loss: 0.07925238460302353 2023-01-21 08:50:28.978734: step: 704/526, loss: 0.6025959849357605 2023-01-21 08:50:30.132090: step: 708/526, loss: 7.720644474029541 2023-01-21 08:50:31.253932: step: 712/526, loss: 0.1756322681903839 2023-01-21 08:50:32.363458: step: 716/526, loss: 0.1689709722995758 2023-01-21 08:50:33.532211: step: 720/526, loss: 0.1403297483921051 2023-01-21 08:50:34.692937: step: 724/526, loss: 1.9489986896514893 2023-01-21 08:50:35.839026: step: 728/526, loss: 0.5420126914978027 2023-01-21 08:50:36.968284: step: 732/526, loss: 0.4235738515853882 2023-01-21 08:50:38.115793: step: 736/526, loss: 0.7603631019592285 2023-01-21 08:50:39.284409: step: 740/526, loss: 0.30566656589508057 2023-01-21 08:50:40.412693: step: 744/526, loss: 1.6442592144012451 2023-01-21 08:50:41.547420: step: 748/526, loss: 1.2697744369506836 2023-01-21 08:50:42.696249: step: 752/526, loss: 0.14380040764808655 2023-01-21 08:50:43.838536: step: 756/526, loss: 0.1881619393825531 2023-01-21 08:50:44.981579: step: 760/526, loss: 0.17431902885437012 2023-01-21 08:50:46.105650: step: 764/526, loss: 0.4401032328605652 2023-01-21 08:50:47.219768: step: 768/526, loss: 0.45035457611083984 2023-01-21 08:50:48.325780: step: 772/526, loss: 0.12683796882629395 2023-01-21 08:50:49.474015: step: 776/526, loss: 6.887025356292725 2023-01-21 08:50:50.568822: step: 780/526, loss: 1.1555850505828857 2023-01-21 08:50:51.693356: step: 784/526, loss: 1.0256932973861694 2023-01-21 08:50:52.839919: step: 788/526, loss: 1.7286767959594727 2023-01-21 08:50:53.986441: step: 792/526, loss: 0.2834101617336273 2023-01-21 08:50:55.131857: step: 796/526, loss: 0.6657474637031555 2023-01-21 08:50:56.276214: step: 800/526, loss: 0.39743462204933167 2023-01-21 08:50:57.393794: step: 804/526, loss: 1.0026803016662598 2023-01-21 08:50:58.531176: step: 808/526, loss: 0.8649067878723145 2023-01-21 08:50:59.661283: step: 812/526, loss: 0.756666362285614 2023-01-21 08:51:00.772781: step: 816/526, loss: 0.13735933601856232 2023-01-21 08:51:01.891101: step: 820/526, loss: 0.2243185043334961 2023-01-21 08:51:03.032505: step: 824/526, loss: 0.6144806146621704 2023-01-21 08:51:04.178752: step: 828/526, loss: 0.36953428387641907 2023-01-21 08:51:05.268697: step: 832/526, loss: 0.112358957529068 2023-01-21 08:51:06.374811: step: 836/526, loss: 7.910516262054443 2023-01-21 08:51:07.508596: step: 840/526, loss: 0.6389080286026001 2023-01-21 08:51:08.669068: step: 844/526, loss: 0.47347506880760193 2023-01-21 08:51:09.784711: step: 848/526, loss: 0.23369541764259338 2023-01-21 08:51:10.918010: step: 852/526, loss: 0.15017472207546234 2023-01-21 08:51:12.053678: step: 856/526, loss: 1.034529685974121 2023-01-21 08:51:13.163805: step: 860/526, loss: 0.768824577331543 2023-01-21 08:51:14.299335: step: 864/526, loss: 0.7620635628700256 2023-01-21 08:51:15.448578: step: 868/526, loss: 0.26703566312789917 2023-01-21 08:51:16.596978: step: 872/526, loss: 1.6306886672973633 2023-01-21 08:51:17.702522: step: 876/526, loss: 0.20424380898475647 2023-01-21 08:51:18.858626: step: 880/526, loss: 0.2717951834201813 2023-01-21 08:51:20.028039: step: 884/526, loss: 0.3805252015590668 2023-01-21 08:51:21.155463: step: 888/526, loss: 0.46414369344711304 2023-01-21 08:51:22.328923: step: 892/526, loss: 0.18890619277954102 2023-01-21 08:51:23.466144: step: 896/526, loss: 0.826716959476471 2023-01-21 08:51:24.631507: step: 900/526, loss: 0.6143976449966431 2023-01-21 08:51:25.773396: step: 904/526, loss: 0.17708587646484375 2023-01-21 08:51:26.892930: step: 908/526, loss: 1.067354440689087 2023-01-21 08:51:28.041697: step: 912/526, loss: 0.1992177963256836 2023-01-21 08:51:29.160509: step: 916/526, loss: 0.08626718819141388 2023-01-21 08:51:30.292919: step: 920/526, loss: 0.601593017578125 2023-01-21 08:51:31.419087: step: 924/526, loss: 0.3136301636695862 2023-01-21 08:51:32.526398: step: 928/526, loss: 0.3361783027648926 2023-01-21 08:51:33.712458: step: 932/526, loss: 1.8437751531600952 2023-01-21 08:51:34.862117: step: 936/526, loss: 0.5151111483573914 2023-01-21 08:51:36.022656: step: 940/526, loss: 0.49617502093315125 2023-01-21 08:51:37.193936: step: 944/526, loss: 1.0481302738189697 2023-01-21 08:51:38.323148: step: 948/526, loss: 0.24285149574279785 2023-01-21 08:51:39.433711: step: 952/526, loss: 0.16453714668750763 2023-01-21 08:51:40.575744: step: 956/526, loss: 0.5571976900100708 2023-01-21 08:51:41.700778: step: 960/526, loss: 1.0675063133239746 2023-01-21 08:51:42.859120: step: 964/526, loss: 0.31882303953170776 2023-01-21 08:51:43.978236: step: 968/526, loss: 0.16361112892627716 2023-01-21 08:51:45.135437: step: 972/526, loss: 0.3353150486946106 2023-01-21 08:51:46.293922: step: 976/526, loss: 0.37121570110321045 2023-01-21 08:51:47.431050: step: 980/526, loss: 0.18925437331199646 2023-01-21 08:51:48.537889: step: 984/526, loss: 1.0003020763397217 2023-01-21 08:51:49.655557: step: 988/526, loss: 2.0658602714538574 2023-01-21 08:51:50.807482: step: 992/526, loss: 0.2758404016494751 2023-01-21 08:51:51.968517: step: 996/526, loss: 0.784070611000061 2023-01-21 08:51:53.085419: step: 1000/526, loss: 0.3153071999549866 2023-01-21 08:51:54.188234: step: 1004/526, loss: 1.74001944065094 2023-01-21 08:51:55.311285: step: 1008/526, loss: 0.6710470914840698 2023-01-21 08:51:56.456670: step: 1012/526, loss: 0.17495279014110565 2023-01-21 08:51:57.608321: step: 1016/526, loss: 0.6081430315971375 2023-01-21 08:51:58.764906: step: 1020/526, loss: 0.3611958622932434 2023-01-21 08:51:59.900539: step: 1024/526, loss: 0.2290334403514862 2023-01-21 08:52:01.025255: step: 1028/526, loss: 0.16033358871936798 2023-01-21 08:52:02.153276: step: 1032/526, loss: 0.07119403034448624 2023-01-21 08:52:03.305049: step: 1036/526, loss: 0.3850610554218292 2023-01-21 08:52:04.444541: step: 1040/526, loss: 0.7330614924430847 2023-01-21 08:52:05.573602: step: 1044/526, loss: 0.9605895280838013 2023-01-21 08:52:06.688410: step: 1048/526, loss: 0.5487978458404541 2023-01-21 08:52:07.846387: step: 1052/526, loss: 0.20520305633544922 2023-01-21 08:52:08.965682: step: 1056/526, loss: 0.17443647980690002 2023-01-21 08:52:10.085639: step: 1060/526, loss: 0.32667216658592224 2023-01-21 08:52:11.201839: step: 1064/526, loss: 0.7762643694877625 2023-01-21 08:52:12.354651: step: 1068/526, loss: 0.45331984758377075 2023-01-21 08:52:13.497249: step: 1072/526, loss: 0.22069521248340607 2023-01-21 08:52:14.624257: step: 1076/526, loss: 0.29446014761924744 2023-01-21 08:52:15.780113: step: 1080/526, loss: 0.305077463388443 2023-01-21 08:52:16.907242: step: 1084/526, loss: 0.12300291657447815 2023-01-21 08:52:18.049899: step: 1088/526, loss: 0.8958039283752441 2023-01-21 08:52:19.191931: step: 1092/526, loss: 2.183286428451538 2023-01-21 08:52:20.339355: step: 1096/526, loss: 0.5702366232872009 2023-01-21 08:52:21.458114: step: 1100/526, loss: 0.36265257000923157 2023-01-21 08:52:22.568316: step: 1104/526, loss: 0.7502535581588745 2023-01-21 08:52:23.671519: step: 1108/526, loss: 1.2186193466186523 2023-01-21 08:52:24.801027: step: 1112/526, loss: 0.20356245338916779 2023-01-21 08:52:25.939293: step: 1116/526, loss: 0.33965975046157837 2023-01-21 08:52:27.079604: step: 1120/526, loss: 0.5156302452087402 2023-01-21 08:52:28.207438: step: 1124/526, loss: 0.14819243550300598 2023-01-21 08:52:29.374824: step: 1128/526, loss: 2.252439498901367 2023-01-21 08:52:30.508527: step: 1132/526, loss: 0.35062360763549805 2023-01-21 08:52:31.648373: step: 1136/526, loss: 1.8226397037506104 2023-01-21 08:52:32.788798: step: 1140/526, loss: 0.3080207407474518 2023-01-21 08:52:33.892130: step: 1144/526, loss: 1.0768003463745117 2023-01-21 08:52:35.022779: step: 1148/526, loss: 0.012936020269989967 2023-01-21 08:52:36.148136: step: 1152/526, loss: 1.0421228408813477 2023-01-21 08:52:37.285077: step: 1156/526, loss: 0.14714708924293518 2023-01-21 08:52:38.402097: step: 1160/526, loss: 1.5155606269836426 2023-01-21 08:52:39.502888: step: 1164/526, loss: 0.13709335029125214 2023-01-21 08:52:40.612346: step: 1168/526, loss: 0.20482495427131653 2023-01-21 08:52:41.717850: step: 1172/526, loss: 0.476881206035614 2023-01-21 08:52:42.857536: step: 1176/526, loss: 0.28138744831085205 2023-01-21 08:52:43.979894: step: 1180/526, loss: 0.6077210903167725 2023-01-21 08:52:45.128132: step: 1184/526, loss: 0.29387348890304565 2023-01-21 08:52:46.254493: step: 1188/526, loss: 0.12247458100318909 2023-01-21 08:52:47.370403: step: 1192/526, loss: 0.8134750127792358 2023-01-21 08:52:48.472954: step: 1196/526, loss: 3.0776665210723877 2023-01-21 08:52:49.642043: step: 1200/526, loss: 0.644314706325531 2023-01-21 08:52:50.771248: step: 1204/526, loss: 0.9402607679367065 2023-01-21 08:52:51.917718: step: 1208/526, loss: 0.9494227766990662 2023-01-21 08:52:53.066803: step: 1212/526, loss: 1.0180188417434692 2023-01-21 08:52:54.197220: step: 1216/526, loss: 0.3018779456615448 2023-01-21 08:52:55.311108: step: 1220/526, loss: 0.1744004786014557 2023-01-21 08:52:56.439479: step: 1224/526, loss: 1.2292245626449585 2023-01-21 08:52:57.546869: step: 1228/526, loss: 0.2391524314880371 2023-01-21 08:52:58.663109: step: 1232/526, loss: 0.6449756622314453 2023-01-21 08:52:59.760921: step: 1236/526, loss: 0.1707058846950531 2023-01-21 08:53:00.923148: step: 1240/526, loss: 0.2891367971897125 2023-01-21 08:53:02.034717: step: 1244/526, loss: 0.16197548806667328 2023-01-21 08:53:03.157813: step: 1248/526, loss: 0.08786487579345703 2023-01-21 08:53:04.307001: step: 1252/526, loss: 0.2576826214790344 2023-01-21 08:53:05.445122: step: 1256/526, loss: 1.7650480270385742 2023-01-21 08:53:06.623334: step: 1260/526, loss: 0.38966912031173706 2023-01-21 08:53:07.805664: step: 1264/526, loss: 1.028980016708374 2023-01-21 08:53:08.971292: step: 1268/526, loss: 0.2723108232021332 2023-01-21 08:53:10.081311: step: 1272/526, loss: 0.33362141251564026 2023-01-21 08:53:11.200970: step: 1276/526, loss: 0.28495320677757263 2023-01-21 08:53:12.308640: step: 1280/526, loss: 0.6047466397285461 2023-01-21 08:53:13.429446: step: 1284/526, loss: 0.24461022019386292 2023-01-21 08:53:14.547425: step: 1288/526, loss: 0.7112138867378235 2023-01-21 08:53:15.688978: step: 1292/526, loss: 0.9876381754875183 2023-01-21 08:53:16.846098: step: 1296/526, loss: 0.3674299120903015 2023-01-21 08:53:17.988330: step: 1300/526, loss: 0.2671385407447815 2023-01-21 08:53:19.113287: step: 1304/526, loss: 0.4772930145263672 2023-01-21 08:53:20.280322: step: 1308/526, loss: 0.40917739272117615 2023-01-21 08:53:21.394021: step: 1312/526, loss: 0.09050621837377548 2023-01-21 08:53:22.513531: step: 1316/526, loss: 0.2704673409461975 2023-01-21 08:53:23.618237: step: 1320/526, loss: 0.5377576351165771 2023-01-21 08:53:24.750333: step: 1324/526, loss: 1.5563396215438843 2023-01-21 08:53:25.886983: step: 1328/526, loss: 0.11209554970264435 2023-01-21 08:53:27.011266: step: 1332/526, loss: 0.15192189812660217 2023-01-21 08:53:28.147563: step: 1336/526, loss: 0.1536126732826233 2023-01-21 08:53:29.242033: step: 1340/526, loss: 1.1947139501571655 2023-01-21 08:53:30.384682: step: 1344/526, loss: 0.10649147629737854 2023-01-21 08:53:31.533432: step: 1348/526, loss: 0.24128800630569458 2023-01-21 08:53:32.662553: step: 1352/526, loss: 0.2277817726135254 2023-01-21 08:53:33.806188: step: 1356/526, loss: 2.7785356044769287 2023-01-21 08:53:34.925121: step: 1360/526, loss: 0.1620945930480957 2023-01-21 08:53:36.074378: step: 1364/526, loss: 0.8325731754302979 2023-01-21 08:53:37.213947: step: 1368/526, loss: 0.7937656044960022 2023-01-21 08:53:38.332976: step: 1372/526, loss: 1.5738050937652588 2023-01-21 08:53:39.489542: step: 1376/526, loss: 0.0945199504494667 2023-01-21 08:53:40.637667: step: 1380/526, loss: 1.0409876108169556 2023-01-21 08:53:41.762628: step: 1384/526, loss: 0.9487485289573669 2023-01-21 08:53:42.909470: step: 1388/526, loss: 0.26488518714904785 2023-01-21 08:53:44.041918: step: 1392/526, loss: 0.27072596549987793 2023-01-21 08:53:45.147558: step: 1396/526, loss: 0.3947053551673889 2023-01-21 08:53:46.242880: step: 1400/526, loss: 1.791111707687378 2023-01-21 08:53:47.401532: step: 1404/526, loss: 1.0617741346359253 2023-01-21 08:53:48.525698: step: 1408/526, loss: 0.37687110900878906 2023-01-21 08:53:49.678140: step: 1412/526, loss: 0.19886387884616852 2023-01-21 08:53:50.815012: step: 1416/526, loss: 0.5679737329483032 2023-01-21 08:53:51.955023: step: 1420/526, loss: 0.5390400290489197 2023-01-21 08:53:53.090460: step: 1424/526, loss: 0.8597105145454407 2023-01-21 08:53:54.213476: step: 1428/526, loss: 0.8764166831970215 2023-01-21 08:53:55.341809: step: 1432/526, loss: 0.39880692958831787 2023-01-21 08:53:56.470168: step: 1436/526, loss: 0.8203771710395813 2023-01-21 08:53:57.634885: step: 1440/526, loss: 2.040961265563965 2023-01-21 08:53:58.745166: step: 1444/526, loss: 0.22099782526493073 2023-01-21 08:53:59.894654: step: 1448/526, loss: 0.4093784689903259 2023-01-21 08:54:01.029822: step: 1452/526, loss: 0.3832346796989441 2023-01-21 08:54:02.178033: step: 1456/526, loss: 0.34543895721435547 2023-01-21 08:54:03.340409: step: 1460/526, loss: 0.4288085699081421 2023-01-21 08:54:04.488475: step: 1464/526, loss: 0.49686431884765625 2023-01-21 08:54:05.610337: step: 1468/526, loss: 0.8526090383529663 2023-01-21 08:54:06.733733: step: 1472/526, loss: 1.317407488822937 2023-01-21 08:54:07.872151: step: 1476/526, loss: 1.2262513637542725 2023-01-21 08:54:08.988217: step: 1480/526, loss: 1.0341687202453613 2023-01-21 08:54:10.116712: step: 1484/526, loss: 1.428326964378357 2023-01-21 08:54:11.222999: step: 1488/526, loss: 0.1320885717868805 2023-01-21 08:54:12.349593: step: 1492/526, loss: 0.16648346185684204 2023-01-21 08:54:13.455126: step: 1496/526, loss: 1.1512404680252075 2023-01-21 08:54:14.583444: step: 1500/526, loss: 0.13755732774734497 2023-01-21 08:54:15.692976: step: 1504/526, loss: 0.16974076628684998 2023-01-21 08:54:16.833383: step: 1508/526, loss: 0.45059338212013245 2023-01-21 08:54:18.022996: step: 1512/526, loss: 1.5823800563812256 2023-01-21 08:54:19.165615: step: 1516/526, loss: 0.364286333322525 2023-01-21 08:54:20.296513: step: 1520/526, loss: 1.3566417694091797 2023-01-21 08:54:21.453583: step: 1524/526, loss: 1.0856091976165771 2023-01-21 08:54:22.583227: step: 1528/526, loss: 0.5379573702812195 2023-01-21 08:54:23.691147: step: 1532/526, loss: 0.05152153968811035 2023-01-21 08:54:24.828038: step: 1536/526, loss: 0.2345394790172577 2023-01-21 08:54:25.953679: step: 1540/526, loss: 0.11498479545116425 2023-01-21 08:54:27.083510: step: 1544/526, loss: 1.08973228931427 2023-01-21 08:54:28.229659: step: 1548/526, loss: 0.12101593613624573 2023-01-21 08:54:29.390677: step: 1552/526, loss: 0.251290500164032 2023-01-21 08:54:30.529218: step: 1556/526, loss: 1.249951720237732 2023-01-21 08:54:31.674331: step: 1560/526, loss: 0.5401656031608582 2023-01-21 08:54:32.803301: step: 1564/526, loss: 0.5475764274597168 2023-01-21 08:54:33.919521: step: 1568/526, loss: 0.27133098244667053 2023-01-21 08:54:35.070343: step: 1572/526, loss: 1.6317224502563477 2023-01-21 08:54:36.204835: step: 1576/526, loss: 0.3935227394104004 2023-01-21 08:54:37.357506: step: 1580/526, loss: 0.5932140350341797 2023-01-21 08:54:38.522635: step: 1584/526, loss: 0.8116245865821838 2023-01-21 08:54:39.649796: step: 1588/526, loss: 0.17813152074813843 2023-01-21 08:54:40.790895: step: 1592/526, loss: 1.099025845527649 2023-01-21 08:54:41.915907: step: 1596/526, loss: 0.3425900638103485 2023-01-21 08:54:43.061495: step: 1600/526, loss: 1.39274263381958 2023-01-21 08:54:44.249321: step: 1604/526, loss: 0.9731160402297974 2023-01-21 08:54:45.380787: step: 1608/526, loss: 0.0604243278503418 2023-01-21 08:54:46.505217: step: 1612/526, loss: 0.6072297692298889 2023-01-21 08:54:47.674475: step: 1616/526, loss: 0.7625547647476196 2023-01-21 08:54:48.810266: step: 1620/526, loss: 2.1663389205932617 2023-01-21 08:54:49.943360: step: 1624/526, loss: 0.12372012436389923 2023-01-21 08:54:51.089934: step: 1628/526, loss: 0.10592193901538849 2023-01-21 08:54:52.215841: step: 1632/526, loss: 0.13152508437633514 2023-01-21 08:54:53.382741: step: 1636/526, loss: 0.44631558656692505 2023-01-21 08:54:54.534333: step: 1640/526, loss: 0.27061766386032104 2023-01-21 08:54:55.673821: step: 1644/526, loss: 8.072844505310059 2023-01-21 08:54:56.804026: step: 1648/526, loss: 0.2594289481639862 2023-01-21 08:54:57.945142: step: 1652/526, loss: 0.33607929944992065 2023-01-21 08:54:59.101445: step: 1656/526, loss: 0.4192092716693878 2023-01-21 08:55:00.260790: step: 1660/526, loss: 1.0942312479019165 2023-01-21 08:55:01.374750: step: 1664/526, loss: 0.3006388247013092 2023-01-21 08:55:02.530581: step: 1668/526, loss: 0.5258996486663818 2023-01-21 08:55:03.695523: step: 1672/526, loss: 0.2088792324066162 2023-01-21 08:55:04.857242: step: 1676/526, loss: 0.16801539063453674 2023-01-21 08:55:06.018746: step: 1680/526, loss: 0.8115229606628418 2023-01-21 08:55:07.170926: step: 1684/526, loss: 0.27260667085647583 2023-01-21 08:55:08.308440: step: 1688/526, loss: 2.6050796508789062 2023-01-21 08:55:09.476659: step: 1692/526, loss: 0.2026577889919281 2023-01-21 08:55:10.612789: step: 1696/526, loss: 0.3631228506565094 2023-01-21 08:55:11.734637: step: 1700/526, loss: 0.42658910155296326 2023-01-21 08:55:12.863424: step: 1704/526, loss: 0.20788437128067017 2023-01-21 08:55:13.998778: step: 1708/526, loss: 0.20939479768276215 2023-01-21 08:55:15.125115: step: 1712/526, loss: 0.23475365340709686 2023-01-21 08:55:16.248994: step: 1716/526, loss: 0.6525284647941589 2023-01-21 08:55:17.388689: step: 1720/526, loss: 0.8020996451377869 2023-01-21 08:55:18.522179: step: 1724/526, loss: 0.9109253883361816 2023-01-21 08:55:19.649015: step: 1728/526, loss: 0.13469500839710236 2023-01-21 08:55:20.800003: step: 1732/526, loss: 0.3011285364627838 2023-01-21 08:55:21.970300: step: 1736/526, loss: 0.4028759002685547 2023-01-21 08:55:23.081277: step: 1740/526, loss: 0.8307867050170898 2023-01-21 08:55:24.182606: step: 1744/526, loss: 0.7511939406394958 2023-01-21 08:55:25.332088: step: 1748/526, loss: 1.1502940654754639 2023-01-21 08:55:26.476839: step: 1752/526, loss: 0.7582845091819763 2023-01-21 08:55:27.599114: step: 1756/526, loss: 1.912783145904541 2023-01-21 08:55:28.741570: step: 1760/526, loss: 0.090985968708992 2023-01-21 08:55:29.859244: step: 1764/526, loss: 0.15521731972694397 2023-01-21 08:55:31.006539: step: 1768/526, loss: 0.4850344657897949 2023-01-21 08:55:32.135107: step: 1772/526, loss: 0.08559072017669678 2023-01-21 08:55:33.264891: step: 1776/526, loss: 0.17542386054992676 2023-01-21 08:55:34.422576: step: 1780/526, loss: 0.7627726793289185 2023-01-21 08:55:35.545822: step: 1784/526, loss: 0.09132376313209534 2023-01-21 08:55:36.674492: step: 1788/526, loss: 0.09789452701807022 2023-01-21 08:55:37.811820: step: 1792/526, loss: 0.15408021211624146 2023-01-21 08:55:38.961138: step: 1796/526, loss: 1.2216967344284058 2023-01-21 08:55:40.088404: step: 1800/526, loss: 0.15602655708789825 2023-01-21 08:55:41.217812: step: 1804/526, loss: 0.6172432899475098 2023-01-21 08:55:42.344736: step: 1808/526, loss: 0.48861628770828247 2023-01-21 08:55:43.509193: step: 1812/526, loss: 0.3700433075428009 2023-01-21 08:55:44.678818: step: 1816/526, loss: 0.5947478413581848 2023-01-21 08:55:45.781593: step: 1820/526, loss: 0.16661682724952698 2023-01-21 08:55:46.912451: step: 1824/526, loss: 0.7500246167182922 2023-01-21 08:55:48.048352: step: 1828/526, loss: 0.33961400389671326 2023-01-21 08:55:49.185563: step: 1832/526, loss: 0.2885696589946747 2023-01-21 08:55:50.307046: step: 1836/526, loss: 0.5720929503440857 2023-01-21 08:55:51.431429: step: 1840/526, loss: 0.4878579080104828 2023-01-21 08:55:52.570820: step: 1844/526, loss: 0.5794669985771179 2023-01-21 08:55:53.680200: step: 1848/526, loss: 0.26986151933670044 2023-01-21 08:55:54.793298: step: 1852/526, loss: 0.7763718962669373 2023-01-21 08:55:55.928758: step: 1856/526, loss: 1.6587055921554565 2023-01-21 08:55:57.034319: step: 1860/526, loss: 1.0288796424865723 2023-01-21 08:55:58.188965: step: 1864/526, loss: 0.21489034593105316 2023-01-21 08:55:59.322107: step: 1868/526, loss: 2.1826789379119873 2023-01-21 08:56:00.461577: step: 1872/526, loss: 1.1303766965866089 2023-01-21 08:56:01.621931: step: 1876/526, loss: 0.4038127362728119 2023-01-21 08:56:02.762924: step: 1880/526, loss: 0.40224477648735046 2023-01-21 08:56:03.906979: step: 1884/526, loss: 0.1496295928955078 2023-01-21 08:56:05.051044: step: 1888/526, loss: 0.12028808891773224 2023-01-21 08:56:06.214935: step: 1892/526, loss: 0.43746626377105713 2023-01-21 08:56:07.336403: step: 1896/526, loss: 0.1351669430732727 2023-01-21 08:56:08.513236: step: 1900/526, loss: 0.5046834945678711 2023-01-21 08:56:09.666372: step: 1904/526, loss: 0.6530091762542725 2023-01-21 08:56:10.796388: step: 1908/526, loss: 0.37235116958618164 2023-01-21 08:56:11.919795: step: 1912/526, loss: 0.07220964133739471 2023-01-21 08:56:13.051679: step: 1916/526, loss: 1.303880214691162 2023-01-21 08:56:14.177326: step: 1920/526, loss: 0.1236216127872467 2023-01-21 08:56:15.289568: step: 1924/526, loss: 0.597282886505127 2023-01-21 08:56:16.401010: step: 1928/526, loss: 0.9237990975379944 2023-01-21 08:56:17.540626: step: 1932/526, loss: 0.6608754396438599 2023-01-21 08:56:18.672109: step: 1936/526, loss: 0.45004063844680786 2023-01-21 08:56:19.822666: step: 1940/526, loss: 0.23730525374412537 2023-01-21 08:56:20.975254: step: 1944/526, loss: 0.2515888214111328 2023-01-21 08:56:22.087934: step: 1948/526, loss: 0.19622564315795898 2023-01-21 08:56:23.197809: step: 1952/526, loss: 0.8251916170120239 2023-01-21 08:56:24.331992: step: 1956/526, loss: 0.19301262497901917 2023-01-21 08:56:25.463662: step: 1960/526, loss: 0.11155568063259125 2023-01-21 08:56:26.616390: step: 1964/526, loss: 0.10877638310194016 2023-01-21 08:56:27.761738: step: 1968/526, loss: 1.6936662197113037 2023-01-21 08:56:28.907717: step: 1972/526, loss: 1.429900884628296 2023-01-21 08:56:30.047112: step: 1976/526, loss: 1.3313575983047485 2023-01-21 08:56:31.210025: step: 1980/526, loss: 0.16067099571228027 2023-01-21 08:56:32.318901: step: 1984/526, loss: 0.18011170625686646 2023-01-21 08:56:33.438661: step: 1988/526, loss: 0.5535411238670349 2023-01-21 08:56:34.558210: step: 1992/526, loss: 0.4083411991596222 2023-01-21 08:56:35.688121: step: 1996/526, loss: 1.4199025630950928 2023-01-21 08:56:36.815159: step: 2000/526, loss: 2.347923517227173 2023-01-21 08:56:37.917975: step: 2004/526, loss: 0.16935959458351135 2023-01-21 08:56:39.037006: step: 2008/526, loss: 0.6884486675262451 2023-01-21 08:56:40.173160: step: 2012/526, loss: 0.12982749938964844 2023-01-21 08:56:41.304185: step: 2016/526, loss: 6.484982967376709 2023-01-21 08:56:42.439483: step: 2020/526, loss: 0.7090138792991638 2023-01-21 08:56:43.643301: step: 2024/526, loss: 0.8112053871154785 2023-01-21 08:56:44.766293: step: 2028/526, loss: 0.2834510803222656 2023-01-21 08:56:45.905860: step: 2032/526, loss: 0.16633328795433044 2023-01-21 08:56:47.023652: step: 2036/526, loss: 0.35938215255737305 2023-01-21 08:56:48.147985: step: 2040/526, loss: 0.2283536046743393 2023-01-21 08:56:49.250372: step: 2044/526, loss: 0.07451625168323517 2023-01-21 08:56:50.368636: step: 2048/526, loss: 0.08642594516277313 2023-01-21 08:56:51.517417: step: 2052/526, loss: 0.7784126400947571 2023-01-21 08:56:52.636651: step: 2056/526, loss: 0.8092414140701294 2023-01-21 08:56:53.801863: step: 2060/526, loss: 0.40342777967453003 2023-01-21 08:56:54.911965: step: 2064/526, loss: 0.14191943407058716 2023-01-21 08:56:56.040773: step: 2068/526, loss: 0.278459370136261 2023-01-21 08:56:57.174441: step: 2072/526, loss: 0.5136095285415649 2023-01-21 08:56:58.296717: step: 2076/526, loss: 0.27368029952049255 2023-01-21 08:56:59.455073: step: 2080/526, loss: 1.7214633226394653 2023-01-21 08:57:00.545666: step: 2084/526, loss: 0.5242714881896973 2023-01-21 08:57:01.654083: step: 2088/526, loss: 0.3880229592323303 2023-01-21 08:57:02.810881: step: 2092/526, loss: 0.2507127821445465 2023-01-21 08:57:03.909020: step: 2096/526, loss: 0.2552480399608612 2023-01-21 08:57:05.000550: step: 2100/526, loss: 0.44141992926597595 2023-01-21 08:57:06.143626: step: 2104/526, loss: 6.530412673950195 ================================================== Loss: 0.751 -------------------- Dev: {'event': {'p': 0.6193853427895981, 'r': 0.6977363515312917, 'f1': 0.6562304320601127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Test: {'event': {'p': 0.5890142226581657, 'r': 0.6268267223382046, 'f1': 0.6073324905183313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Chinese: {'event': {'p': 0.6376811594202898, 'r': 0.8148148148148148, 'f1': 0.7154471544715448}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Korean: {'event': {'p': 0.7, 'r': 0.3333333333333333, 'f1': 0.45161290322580644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Russian: {'event': {'p': 0.5, 'r': 0.3611111111111111, 'f1': 0.4193548387096774}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} New best chinese model... New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6193853427895981, 'r': 0.6977363515312917, 'f1': 0.6562304320601127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Chinese: {'event': {'p': 0.5890142226581657, 'r': 0.6268267223382046, 'f1': 0.6073324905183313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Chinese: {'event': {'p': 0.6376811594202898, 'r': 0.8148148148148148, 'f1': 0.7154471544715448}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Korean: {'event': {'p': 0.6193853427895981, 'r': 0.6977363515312917, 'f1': 0.6562304320601127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Korean: {'event': {'p': 0.5890142226581657, 'r': 0.6268267223382046, 'f1': 0.6073324905183313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Korean: {'event': {'p': 0.7, 'r': 0.3333333333333333, 'f1': 0.45161290322580644}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} -------------------- Eng Dev for Russian: {'event': {'p': 0.6193853427895981, 'r': 0.6977363515312917, 'f1': 0.6562304320601127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5890142226581657, 'r': 0.6268267223382046, 'f1': 0.6073324905183313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.5, 'r': 0.3611111111111111, 'f1': 0.4193548387096774}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} ****************************** Epoch: 2 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 08:58:11.917687: step: 4/526, loss: 0.7371686697006226 2023-01-21 08:58:13.080565: step: 8/526, loss: 0.14277620613574982 2023-01-21 08:58:14.232651: step: 12/526, loss: 0.22620564699172974 2023-01-21 08:58:15.359607: step: 16/526, loss: 6.459168910980225 2023-01-21 08:58:16.496808: step: 20/526, loss: 2.7405848503112793 2023-01-21 08:58:17.609861: step: 24/526, loss: 0.4026455879211426 2023-01-21 08:58:18.735569: step: 28/526, loss: 0.19612395763397217 2023-01-21 08:58:19.827987: step: 32/526, loss: 0.0955914556980133 2023-01-21 08:58:20.934256: step: 36/526, loss: 0.849429190158844 2023-01-21 08:58:22.048020: step: 40/526, loss: 0.28417879343032837 2023-01-21 08:58:23.148907: step: 44/526, loss: 0.18784980475902557 2023-01-21 08:58:24.287193: step: 48/526, loss: 0.6011751890182495 2023-01-21 08:58:25.441879: step: 52/526, loss: 1.947126865386963 2023-01-21 08:58:26.538288: step: 56/526, loss: 0.05753608047962189 2023-01-21 08:58:27.652399: step: 60/526, loss: 0.8347592353820801 2023-01-21 08:58:28.779108: step: 64/526, loss: 0.8283432126045227 2023-01-21 08:58:29.919716: step: 68/526, loss: 0.192641019821167 2023-01-21 08:58:31.051790: step: 72/526, loss: 1.207589030265808 2023-01-21 08:58:32.175976: step: 76/526, loss: 0.28988510370254517 2023-01-21 08:58:33.321625: step: 80/526, loss: 0.2214467078447342 2023-01-21 08:58:34.446533: step: 84/526, loss: 0.054837897419929504 2023-01-21 08:58:35.573444: step: 88/526, loss: 0.17066727578639984 2023-01-21 08:58:36.716331: step: 92/526, loss: 0.26862430572509766 2023-01-21 08:58:37.839826: step: 96/526, loss: 0.1193912997841835 2023-01-21 08:58:38.974693: step: 100/526, loss: 0.1942417174577713 2023-01-21 08:58:40.111931: step: 104/526, loss: 0.7256829738616943 2023-01-21 08:58:41.238424: step: 108/526, loss: 0.34531232714653015 2023-01-21 08:58:42.348156: step: 112/526, loss: 0.3780747354030609 2023-01-21 08:58:43.477712: step: 116/526, loss: 0.14871588349342346 2023-01-21 08:58:44.629703: step: 120/526, loss: 1.8272624015808105 2023-01-21 08:58:45.756549: step: 124/526, loss: 2.5648231506347656 2023-01-21 08:58:46.874652: step: 128/526, loss: 0.2851431369781494 2023-01-21 08:58:47.991332: step: 132/526, loss: 0.4197579622268677 2023-01-21 08:58:49.102298: step: 136/526, loss: 0.3052535057067871 2023-01-21 08:58:50.205973: step: 140/526, loss: 0.5125947594642639 2023-01-21 08:58:51.346040: step: 144/526, loss: 0.7838599681854248 2023-01-21 08:58:52.491400: step: 148/526, loss: 0.3850439786911011 2023-01-21 08:58:53.625261: step: 152/526, loss: 0.8700082898139954 2023-01-21 08:58:54.737082: step: 156/526, loss: 0.1666758954524994 2023-01-21 08:58:55.884268: step: 160/526, loss: 0.18689113855361938 2023-01-21 08:58:57.008953: step: 164/526, loss: 0.14302244782447815 2023-01-21 08:58:58.131904: step: 168/526, loss: 0.158074289560318 2023-01-21 08:58:59.243886: step: 172/526, loss: 1.471127986907959 2023-01-21 08:59:00.360800: step: 176/526, loss: 0.5885858535766602 2023-01-21 08:59:01.479575: step: 180/526, loss: 0.30178844928741455 2023-01-21 08:59:02.591765: step: 184/526, loss: 0.1546308994293213 2023-01-21 08:59:03.711651: step: 188/526, loss: 0.5008574724197388 2023-01-21 08:59:04.832633: step: 192/526, loss: 0.12119532376527786 2023-01-21 08:59:05.938735: step: 196/526, loss: 0.14665204286575317 2023-01-21 08:59:07.081406: step: 200/526, loss: 0.2566226124763489 2023-01-21 08:59:08.197417: step: 204/526, loss: 6.856566905975342 2023-01-21 08:59:09.344676: step: 208/526, loss: 0.8183758854866028 2023-01-21 08:59:10.450133: step: 212/526, loss: 0.7002421021461487 2023-01-21 08:59:11.592205: step: 216/526, loss: 1.0828109979629517 2023-01-21 08:59:12.704160: step: 220/526, loss: 0.2938995957374573 2023-01-21 08:59:13.858032: step: 224/526, loss: 0.2463708519935608 2023-01-21 08:59:14.989685: step: 228/526, loss: 0.7030066251754761 2023-01-21 08:59:16.200399: step: 232/526, loss: 0.27724459767341614 2023-01-21 08:59:17.320390: step: 236/526, loss: 0.8027188777923584 2023-01-21 08:59:18.433436: step: 240/526, loss: 0.16752897202968597 2023-01-21 08:59:19.539928: step: 244/526, loss: 6.357608795166016 2023-01-21 08:59:20.682365: step: 248/526, loss: 0.22123274207115173 2023-01-21 08:59:21.823630: step: 252/526, loss: 0.13078276813030243 2023-01-21 08:59:23.012250: step: 256/526, loss: 0.4018796980381012 2023-01-21 08:59:24.146637: step: 260/526, loss: 0.5865241885185242 2023-01-21 08:59:25.306325: step: 264/526, loss: 0.8238250017166138 2023-01-21 08:59:26.467111: step: 268/526, loss: 1.88313889503479 2023-01-21 08:59:27.585887: step: 272/526, loss: 0.9527637362480164 2023-01-21 08:59:28.693792: step: 276/526, loss: 0.9735413789749146 2023-01-21 08:59:29.835568: step: 280/526, loss: 0.21730008721351624 2023-01-21 08:59:30.965664: step: 284/526, loss: 0.6474210619926453 2023-01-21 08:59:32.098109: step: 288/526, loss: 0.2266136258840561 2023-01-21 08:59:33.222494: step: 292/526, loss: 0.5801045894622803 2023-01-21 08:59:34.354493: step: 296/526, loss: 6.723257541656494 2023-01-21 08:59:35.473895: step: 300/526, loss: 0.19547128677368164 2023-01-21 08:59:36.627171: step: 304/526, loss: 0.2741946280002594 2023-01-21 08:59:37.732113: step: 308/526, loss: 0.25209468603134155 2023-01-21 08:59:38.875853: step: 312/526, loss: 0.7836558222770691 2023-01-21 08:59:39.982843: step: 316/526, loss: 0.08856067806482315 2023-01-21 08:59:41.087042: step: 320/526, loss: 0.13040843605995178 2023-01-21 08:59:42.245628: step: 324/526, loss: 0.6577426195144653 2023-01-21 08:59:43.350211: step: 328/526, loss: 0.3646667003631592 2023-01-21 08:59:44.493472: step: 332/526, loss: 0.7247281074523926 2023-01-21 08:59:45.638250: step: 336/526, loss: 0.17073917388916016 2023-01-21 08:59:46.752532: step: 340/526, loss: 0.21138477325439453 2023-01-21 08:59:47.876541: step: 344/526, loss: 0.095563605427742 2023-01-21 08:59:49.023225: step: 348/526, loss: 0.9368394017219543 2023-01-21 08:59:50.131843: step: 352/526, loss: 0.28148743510246277 2023-01-21 08:59:51.244347: step: 356/526, loss: 0.3404358923435211 2023-01-21 08:59:52.410736: step: 360/526, loss: 0.3894495964050293 2023-01-21 08:59:53.512241: step: 364/526, loss: 0.21750307083129883 2023-01-21 08:59:54.633368: step: 368/526, loss: 0.10416451096534729 2023-01-21 08:59:55.734827: step: 372/526, loss: 0.07759085297584534 2023-01-21 08:59:56.840667: step: 376/526, loss: 0.3454691469669342 2023-01-21 08:59:57.966346: step: 380/526, loss: 0.8341060876846313 2023-01-21 08:59:59.084017: step: 384/526, loss: 0.3944700360298157 2023-01-21 09:00:00.207149: step: 388/526, loss: 0.7250019311904907 2023-01-21 09:00:01.323052: step: 392/526, loss: 0.20458531379699707 2023-01-21 09:00:02.480858: step: 396/526, loss: 0.826136589050293 2023-01-21 09:00:03.619588: step: 400/526, loss: 0.685352623462677 2023-01-21 09:00:04.755796: step: 404/526, loss: 0.27412986755371094 2023-01-21 09:00:05.917200: step: 408/526, loss: 0.10080757737159729 2023-01-21 09:00:07.059811: step: 412/526, loss: 0.17872238159179688 2023-01-21 09:00:08.208147: step: 416/526, loss: 1.8121154308319092 2023-01-21 09:00:09.332155: step: 420/526, loss: 0.2770763337612152 2023-01-21 09:00:10.431456: step: 424/526, loss: 0.3830799460411072 2023-01-21 09:00:11.556202: step: 428/526, loss: 0.34010785818099976 2023-01-21 09:00:12.691745: step: 432/526, loss: 0.16602563858032227 2023-01-21 09:00:13.864208: step: 436/526, loss: 1.202040195465088 2023-01-21 09:00:14.996786: step: 440/526, loss: 0.21638837456703186 2023-01-21 09:00:16.124890: step: 444/526, loss: 0.08371257781982422 2023-01-21 09:00:17.272446: step: 448/526, loss: 1.234601378440857 2023-01-21 09:00:18.417889: step: 452/526, loss: 0.8448450565338135 2023-01-21 09:00:19.578857: step: 456/526, loss: 0.15782856941223145 2023-01-21 09:00:20.729423: step: 460/526, loss: 2.00057315826416 2023-01-21 09:00:21.867389: step: 464/526, loss: 0.1114925891160965 2023-01-21 09:00:22.988582: step: 468/526, loss: 0.2380194216966629 2023-01-21 09:00:24.121163: step: 472/526, loss: 0.8867142796516418 2023-01-21 09:00:25.284772: step: 476/526, loss: 3.263335704803467 2023-01-21 09:00:26.417355: step: 480/526, loss: 0.6447303891181946 2023-01-21 09:00:27.537760: step: 484/526, loss: 0.4122544825077057 2023-01-21 09:00:28.682488: step: 488/526, loss: 0.3632197380065918 2023-01-21 09:00:29.816232: step: 492/526, loss: 0.17861899733543396 2023-01-21 09:00:30.960802: step: 496/526, loss: 0.9973654747009277 2023-01-21 09:00:32.076810: step: 500/526, loss: 0.3834887742996216 2023-01-21 09:00:33.204398: step: 504/526, loss: 0.6213089227676392 2023-01-21 09:00:34.340911: step: 508/526, loss: 0.5560882091522217 2023-01-21 09:00:35.471592: step: 512/526, loss: 0.5832465291023254 2023-01-21 09:00:36.595106: step: 516/526, loss: 0.08488158881664276 2023-01-21 09:00:37.730137: step: 520/526, loss: 0.6259737610816956 2023-01-21 09:00:38.893851: step: 524/526, loss: 0.24635343253612518 2023-01-21 09:00:40.029145: step: 528/526, loss: 0.2052966207265854 2023-01-21 09:00:41.144310: step: 532/526, loss: 0.14800778031349182 2023-01-21 09:00:42.277596: step: 536/526, loss: 0.34591084718704224 2023-01-21 09:00:43.396380: step: 540/526, loss: 0.21279869973659515 2023-01-21 09:00:44.512465: step: 544/526, loss: 0.2472452074289322 2023-01-21 09:00:45.663491: step: 548/526, loss: 9.263084411621094 2023-01-21 09:00:46.783833: step: 552/526, loss: 0.1365702748298645 2023-01-21 09:00:47.922312: step: 556/526, loss: 1.2602787017822266 2023-01-21 09:00:49.089792: step: 560/526, loss: 0.9019102454185486 2023-01-21 09:00:50.250643: step: 564/526, loss: 0.05515509098768234 2023-01-21 09:00:51.384552: step: 568/526, loss: 1.2054473161697388 2023-01-21 09:00:52.503893: step: 572/526, loss: 0.20976220071315765 2023-01-21 09:00:53.647918: step: 576/526, loss: 1.3283308744430542 2023-01-21 09:00:54.771086: step: 580/526, loss: 0.9090974926948547 2023-01-21 09:00:55.894323: step: 584/526, loss: 1.6399478912353516 2023-01-21 09:00:56.992215: step: 588/526, loss: 0.17284272611141205 2023-01-21 09:00:58.121621: step: 592/526, loss: 0.17959660291671753 2023-01-21 09:00:59.272804: step: 596/526, loss: 0.0896090567111969 2023-01-21 09:01:00.425301: step: 600/526, loss: 0.13831454515457153 2023-01-21 09:01:01.585715: step: 604/526, loss: 0.9562665224075317 2023-01-21 09:01:02.700170: step: 608/526, loss: 0.17174744606018066 2023-01-21 09:01:03.905394: step: 612/526, loss: 0.10748949646949768 2023-01-21 09:01:05.014646: step: 616/526, loss: 0.6629589796066284 2023-01-21 09:01:06.163718: step: 620/526, loss: 0.21342550218105316 2023-01-21 09:01:07.361456: step: 624/526, loss: 0.204206183552742 2023-01-21 09:01:08.492645: step: 628/526, loss: 0.08018426597118378 2023-01-21 09:01:09.663655: step: 632/526, loss: 0.10665550082921982 2023-01-21 09:01:10.803443: step: 636/526, loss: 0.2275547981262207 2023-01-21 09:01:11.949911: step: 640/526, loss: 0.6749961376190186 2023-01-21 09:01:13.090308: step: 644/526, loss: 0.06434555351734161 2023-01-21 09:01:14.181258: step: 648/526, loss: 0.2734025716781616 2023-01-21 09:01:15.310968: step: 652/526, loss: 0.13301712274551392 2023-01-21 09:01:16.438164: step: 656/526, loss: 0.2600051164627075 2023-01-21 09:01:17.584014: step: 660/526, loss: 0.08476734161376953 2023-01-21 09:01:18.720034: step: 664/526, loss: 1.7141621112823486 2023-01-21 09:01:19.867002: step: 668/526, loss: 0.474673330783844 2023-01-21 09:01:20.979231: step: 672/526, loss: 0.4861367642879486 2023-01-21 09:01:22.091995: step: 676/526, loss: 0.6617634892463684 2023-01-21 09:01:23.266317: step: 680/526, loss: 0.08454294502735138 2023-01-21 09:01:24.425868: step: 684/526, loss: 0.13664832711219788 2023-01-21 09:01:25.529981: step: 688/526, loss: 0.7994793653488159 2023-01-21 09:01:26.710926: step: 692/526, loss: 0.7563218474388123 2023-01-21 09:01:27.864943: step: 696/526, loss: 0.12473689019680023 2023-01-21 09:01:28.983440: step: 700/526, loss: 0.1568729430437088 2023-01-21 09:01:30.118724: step: 704/526, loss: 0.520408034324646 2023-01-21 09:01:31.256818: step: 708/526, loss: 0.10370621830224991 2023-01-21 09:01:32.352436: step: 712/526, loss: 0.3513716757297516 2023-01-21 09:01:33.489837: step: 716/526, loss: 0.18958741426467896 2023-01-21 09:01:34.657566: step: 720/526, loss: 0.4684511125087738 2023-01-21 09:01:35.805167: step: 724/526, loss: 0.07368616759777069 2023-01-21 09:01:36.924485: step: 728/526, loss: 0.8527272939682007 2023-01-21 09:01:38.067235: step: 732/526, loss: 0.2130545675754547 2023-01-21 09:01:39.186930: step: 736/526, loss: 0.2087687999010086 2023-01-21 09:01:40.323250: step: 740/526, loss: 0.270187646150589 2023-01-21 09:01:41.441358: step: 744/526, loss: 0.36126840114593506 2023-01-21 09:01:42.576371: step: 748/526, loss: 0.19674549996852875 2023-01-21 09:01:43.707049: step: 752/526, loss: 0.8210179805755615 2023-01-21 09:01:44.852545: step: 756/526, loss: 1.8814308643341064 2023-01-21 09:01:46.002815: step: 760/526, loss: 0.1634916365146637 2023-01-21 09:01:47.117839: step: 764/526, loss: 0.08084812760353088 2023-01-21 09:01:48.258065: step: 768/526, loss: 0.0727061778306961 2023-01-21 09:01:49.411000: step: 772/526, loss: 0.16369609534740448 2023-01-21 09:01:50.550328: step: 776/526, loss: 1.20604407787323 2023-01-21 09:01:51.660908: step: 780/526, loss: 0.1588907241821289 2023-01-21 09:01:52.818761: step: 784/526, loss: 0.03581704944372177 2023-01-21 09:01:53.954805: step: 788/526, loss: 0.19073882699012756 2023-01-21 09:01:55.078991: step: 792/526, loss: 0.1830032467842102 2023-01-21 09:01:56.257101: step: 796/526, loss: 1.1516741514205933 2023-01-21 09:01:57.387205: step: 800/526, loss: 6.965559959411621 2023-01-21 09:01:58.515140: step: 804/526, loss: 0.18981438875198364 2023-01-21 09:01:59.654597: step: 808/526, loss: 0.21846535801887512 2023-01-21 09:02:00.799162: step: 812/526, loss: 0.09237129986286163 2023-01-21 09:02:01.965187: step: 816/526, loss: 0.27701517939567566 2023-01-21 09:02:03.073579: step: 820/526, loss: 0.7196527719497681 2023-01-21 09:02:04.203832: step: 824/526, loss: 0.19946356117725372 2023-01-21 09:02:05.325556: step: 828/526, loss: 0.15538763999938965 2023-01-21 09:02:06.461429: step: 832/526, loss: 0.15155087411403656 2023-01-21 09:02:07.630507: step: 836/526, loss: 1.4473398923873901 2023-01-21 09:02:08.784142: step: 840/526, loss: 0.7193412184715271 2023-01-21 09:02:09.891831: step: 844/526, loss: 0.14692479372024536 2023-01-21 09:02:11.024509: step: 848/526, loss: 0.19061346352100372 2023-01-21 09:02:12.126627: step: 852/526, loss: 0.33641982078552246 2023-01-21 09:02:13.290024: step: 856/526, loss: 0.34093427658081055 2023-01-21 09:02:14.425693: step: 860/526, loss: 0.7681768536567688 2023-01-21 09:02:15.566578: step: 864/526, loss: 0.18542137742042542 2023-01-21 09:02:16.694213: step: 868/526, loss: 0.037743210792541504 2023-01-21 09:02:17.810193: step: 872/526, loss: 0.9006453156471252 2023-01-21 09:02:18.949285: step: 876/526, loss: 0.11338487267494202 2023-01-21 09:02:20.096127: step: 880/526, loss: 0.4533006548881531 2023-01-21 09:02:21.224614: step: 884/526, loss: 0.8425458669662476 2023-01-21 09:02:22.346124: step: 888/526, loss: 0.7941291332244873 2023-01-21 09:02:23.481937: step: 892/526, loss: 1.0038068294525146 2023-01-21 09:02:24.603418: step: 896/526, loss: 0.20018891990184784 2023-01-21 09:02:25.715129: step: 900/526, loss: 0.6852834224700928 2023-01-21 09:02:26.830091: step: 904/526, loss: 1.060499906539917 2023-01-21 09:02:27.973707: step: 908/526, loss: 0.1644587516784668 2023-01-21 09:02:29.088262: step: 912/526, loss: 0.2475832998752594 2023-01-21 09:02:30.221391: step: 916/526, loss: 4.642247676849365 2023-01-21 09:02:31.348883: step: 920/526, loss: 1.515236735343933 2023-01-21 09:02:32.452786: step: 924/526, loss: 0.2769314646720886 2023-01-21 09:02:33.594413: step: 928/526, loss: 0.8623466491699219 2023-01-21 09:02:34.738571: step: 932/526, loss: 0.3963509798049927 2023-01-21 09:02:35.875457: step: 936/526, loss: 0.19093409180641174 2023-01-21 09:02:37.056928: step: 940/526, loss: 0.4518064558506012 2023-01-21 09:02:38.182661: step: 944/526, loss: 0.1292782723903656 2023-01-21 09:02:39.319313: step: 948/526, loss: 7.877197742462158 2023-01-21 09:02:40.452251: step: 952/526, loss: 0.12697263062000275 2023-01-21 09:02:41.595575: step: 956/526, loss: 0.17100726068019867 2023-01-21 09:02:42.733481: step: 960/526, loss: 0.21051812171936035 2023-01-21 09:02:43.884245: step: 964/526, loss: 0.9328678250312805 2023-01-21 09:02:45.038309: step: 968/526, loss: 1.9260833263397217 2023-01-21 09:02:46.194462: step: 972/526, loss: 0.11043057590723038 2023-01-21 09:02:47.361244: step: 976/526, loss: 0.3045993745326996 2023-01-21 09:02:48.487959: step: 980/526, loss: 0.15302905440330505 2023-01-21 09:02:49.635252: step: 984/526, loss: 0.5110117197036743 2023-01-21 09:02:50.779333: step: 988/526, loss: 0.06770721077919006 2023-01-21 09:02:51.935998: step: 992/526, loss: 0.13780584931373596 2023-01-21 09:02:53.073292: step: 996/526, loss: 0.2611147463321686 2023-01-21 09:02:54.214103: step: 1000/526, loss: 0.060150954872369766 2023-01-21 09:02:55.350574: step: 1004/526, loss: 1.1544982194900513 2023-01-21 09:02:56.484233: step: 1008/526, loss: 0.20008069276809692 2023-01-21 09:02:57.618535: step: 1012/526, loss: 0.7357773780822754 2023-01-21 09:02:58.767820: step: 1016/526, loss: 0.07235307991504669 2023-01-21 09:02:59.905036: step: 1020/526, loss: 0.3647525906562805 2023-01-21 09:03:01.069088: step: 1024/526, loss: 0.40268927812576294 2023-01-21 09:03:02.197152: step: 1028/526, loss: 0.7556501626968384 2023-01-21 09:03:03.346395: step: 1032/526, loss: 0.6995542645454407 2023-01-21 09:03:04.477226: step: 1036/526, loss: 0.124393992125988 2023-01-21 09:03:05.618493: step: 1040/526, loss: 0.35593196749687195 2023-01-21 09:03:06.755363: step: 1044/526, loss: 0.13892945647239685 2023-01-21 09:03:07.872516: step: 1048/526, loss: 0.989820122718811 2023-01-21 09:03:09.015342: step: 1052/526, loss: 0.09860043227672577 2023-01-21 09:03:10.153384: step: 1056/526, loss: 0.5358060598373413 2023-01-21 09:03:11.247369: step: 1060/526, loss: 0.21409329771995544 2023-01-21 09:03:12.371431: step: 1064/526, loss: 0.6925015449523926 2023-01-21 09:03:13.547027: step: 1068/526, loss: 0.4330235421657562 2023-01-21 09:03:14.701806: step: 1072/526, loss: 6.833107948303223 2023-01-21 09:03:15.844067: step: 1076/526, loss: 0.1685158610343933 2023-01-21 09:03:16.954169: step: 1080/526, loss: 0.17064112424850464 2023-01-21 09:03:18.102571: step: 1084/526, loss: 1.2955894470214844 2023-01-21 09:03:19.209255: step: 1088/526, loss: 0.029970981180667877 2023-01-21 09:03:20.312716: step: 1092/526, loss: 0.7043238282203674 2023-01-21 09:03:21.441854: step: 1096/526, loss: 0.09527397155761719 2023-01-21 09:03:22.569457: step: 1100/526, loss: 0.12092152237892151 2023-01-21 09:03:23.700937: step: 1104/526, loss: 0.06790585815906525 2023-01-21 09:03:24.833549: step: 1108/526, loss: 0.10850711166858673 2023-01-21 09:03:26.012800: step: 1112/526, loss: 0.20133551955223083 2023-01-21 09:03:27.132411: step: 1116/526, loss: 0.15107566118240356 2023-01-21 09:03:28.240473: step: 1120/526, loss: 0.21529769897460938 2023-01-21 09:03:29.383371: step: 1124/526, loss: 0.4878457188606262 2023-01-21 09:03:30.515950: step: 1128/526, loss: 0.241888627409935 2023-01-21 09:03:31.644162: step: 1132/526, loss: 1.7180202007293701 2023-01-21 09:03:32.769805: step: 1136/526, loss: 0.7908123135566711 2023-01-21 09:03:33.909831: step: 1140/526, loss: 0.2929447293281555 2023-01-21 09:03:35.059646: step: 1144/526, loss: 0.5712848901748657 2023-01-21 09:03:36.180617: step: 1148/526, loss: 1.495749831199646 2023-01-21 09:03:37.332186: step: 1152/526, loss: 0.6782424449920654 2023-01-21 09:03:38.465305: step: 1156/526, loss: 0.052136801183223724 2023-01-21 09:03:39.629832: step: 1160/526, loss: 0.1865929216146469 2023-01-21 09:03:40.751064: step: 1164/526, loss: 0.18351951241493225 2023-01-21 09:03:41.871246: step: 1168/526, loss: 0.7199469804763794 2023-01-21 09:03:43.021262: step: 1172/526, loss: 0.6309493780136108 2023-01-21 09:03:44.145090: step: 1176/526, loss: 0.179520845413208 2023-01-21 09:03:45.275500: step: 1180/526, loss: 0.4822305738925934 2023-01-21 09:03:46.382064: step: 1184/526, loss: 0.3642409145832062 2023-01-21 09:03:47.512231: step: 1188/526, loss: 0.8015809655189514 2023-01-21 09:03:48.682844: step: 1192/526, loss: 0.14055633544921875 2023-01-21 09:03:49.817915: step: 1196/526, loss: 0.13793951272964478 2023-01-21 09:03:50.950134: step: 1200/526, loss: 0.06664423644542694 2023-01-21 09:03:52.077244: step: 1204/526, loss: 0.07401466369628906 2023-01-21 09:03:53.204012: step: 1208/526, loss: 0.2749904692173004 2023-01-21 09:03:54.367289: step: 1212/526, loss: 0.5550336241722107 2023-01-21 09:03:55.512799: step: 1216/526, loss: 0.3880097270011902 2023-01-21 09:03:56.681909: step: 1220/526, loss: 0.49445676803588867 2023-01-21 09:03:57.814539: step: 1224/526, loss: 0.06755051761865616 2023-01-21 09:03:58.937868: step: 1228/526, loss: 1.6481280326843262 2023-01-21 09:04:00.091791: step: 1232/526, loss: 0.12867221236228943 2023-01-21 09:04:01.249130: step: 1236/526, loss: 0.13796454668045044 2023-01-21 09:04:02.383815: step: 1240/526, loss: 0.11804446578025818 2023-01-21 09:04:03.537967: step: 1244/526, loss: 0.6119052171707153 2023-01-21 09:04:04.687601: step: 1248/526, loss: 0.4132540822029114 2023-01-21 09:04:05.828807: step: 1252/526, loss: 0.1319485306739807 2023-01-21 09:04:06.952644: step: 1256/526, loss: 0.1947527825832367 2023-01-21 09:04:08.113096: step: 1260/526, loss: 0.13552546501159668 2023-01-21 09:04:09.223194: step: 1264/526, loss: 0.5897731781005859 2023-01-21 09:04:10.332423: step: 1268/526, loss: 1.033892035484314 2023-01-21 09:04:11.465863: step: 1272/526, loss: 0.7521183490753174 2023-01-21 09:04:12.633360: step: 1276/526, loss: 0.16722755134105682 2023-01-21 09:04:13.804509: step: 1280/526, loss: 0.15022316575050354 2023-01-21 09:04:14.929207: step: 1284/526, loss: 6.773693561553955 2023-01-21 09:04:16.075793: step: 1288/526, loss: 0.12554121017456055 2023-01-21 09:04:17.230442: step: 1292/526, loss: 0.8333118557929993 2023-01-21 09:04:18.374595: step: 1296/526, loss: 0.23240791261196136 2023-01-21 09:04:19.503824: step: 1300/526, loss: 0.13540272414684296 2023-01-21 09:04:20.659577: step: 1304/526, loss: 0.09288950264453888 2023-01-21 09:04:21.809858: step: 1308/526, loss: 0.5963178277015686 2023-01-21 09:04:22.937816: step: 1312/526, loss: 0.39611929655075073 2023-01-21 09:04:24.075560: step: 1316/526, loss: 0.09164667129516602 2023-01-21 09:04:25.233573: step: 1320/526, loss: 0.5031039118766785 2023-01-21 09:04:26.361023: step: 1324/526, loss: 0.20707297325134277 2023-01-21 09:04:27.491463: step: 1328/526, loss: 0.2976667284965515 2023-01-21 09:04:28.613147: step: 1332/526, loss: 0.13202795386314392 2023-01-21 09:04:29.736423: step: 1336/526, loss: 0.3469514846801758 2023-01-21 09:04:30.879701: step: 1340/526, loss: 2.144334554672241 2023-01-21 09:04:32.021044: step: 1344/526, loss: 0.26964569091796875 2023-01-21 09:04:33.133361: step: 1348/526, loss: 0.27661553025245667 2023-01-21 09:04:34.298291: step: 1352/526, loss: 0.4165247082710266 2023-01-21 09:04:35.428628: step: 1356/526, loss: 0.3352002501487732 2023-01-21 09:04:36.543094: step: 1360/526, loss: 0.25387871265411377 2023-01-21 09:04:37.668945: step: 1364/526, loss: 0.7760627865791321 2023-01-21 09:04:38.807667: step: 1368/526, loss: 0.9163665771484375 2023-01-21 09:04:39.937687: step: 1372/526, loss: 0.09707331657409668 2023-01-21 09:04:41.125035: step: 1376/526, loss: 0.0896996483206749 2023-01-21 09:04:42.248508: step: 1380/526, loss: 0.1081552505493164 2023-01-21 09:04:43.429073: step: 1384/526, loss: 0.40700292587280273 2023-01-21 09:04:44.572063: step: 1388/526, loss: 0.04897632449865341 2023-01-21 09:04:45.737964: step: 1392/526, loss: 1.671891450881958 2023-01-21 09:04:46.855334: step: 1396/526, loss: 0.25511035323143005 2023-01-21 09:04:48.025799: step: 1400/526, loss: 0.4179447293281555 2023-01-21 09:04:49.157361: step: 1404/526, loss: 0.22246304154396057 2023-01-21 09:04:50.335135: step: 1408/526, loss: 0.7529239654541016 2023-01-21 09:04:51.474356: step: 1412/526, loss: 0.19054308533668518 2023-01-21 09:04:52.624488: step: 1416/526, loss: 0.22854405641555786 2023-01-21 09:04:53.763392: step: 1420/526, loss: 0.3220301568508148 2023-01-21 09:04:54.898243: step: 1424/526, loss: 1.4963210821151733 2023-01-21 09:04:56.029186: step: 1428/526, loss: 0.0571308359503746 2023-01-21 09:04:57.144201: step: 1432/526, loss: 0.12312392890453339 2023-01-21 09:04:58.256289: step: 1436/526, loss: 0.08499126136302948 2023-01-21 09:04:59.380271: step: 1440/526, loss: 0.33305859565734863 2023-01-21 09:05:00.542351: step: 1444/526, loss: 0.7659327387809753 2023-01-21 09:05:01.686739: step: 1448/526, loss: 0.11860928684473038 2023-01-21 09:05:02.795586: step: 1452/526, loss: 0.09401998668909073 2023-01-21 09:05:03.881512: step: 1456/526, loss: 0.20078416168689728 2023-01-21 09:05:05.010543: step: 1460/526, loss: 0.7208659648895264 2023-01-21 09:05:06.197391: step: 1464/526, loss: 0.5318990349769592 2023-01-21 09:05:07.323246: step: 1468/526, loss: 1.0348161458969116 2023-01-21 09:05:08.461079: step: 1472/526, loss: 0.19070421159267426 2023-01-21 09:05:09.575681: step: 1476/526, loss: 0.8465617895126343 2023-01-21 09:05:10.729934: step: 1480/526, loss: 0.10943231731653214 2023-01-21 09:05:11.907924: step: 1484/526, loss: 0.7222205996513367 2023-01-21 09:05:13.058825: step: 1488/526, loss: 0.33950796723365784 2023-01-21 09:05:14.228454: step: 1492/526, loss: 0.14517097175121307 2023-01-21 09:05:15.339156: step: 1496/526, loss: 0.2597583830356598 2023-01-21 09:05:16.452866: step: 1500/526, loss: 0.13723182678222656 2023-01-21 09:05:17.597405: step: 1504/526, loss: 0.7158119678497314 2023-01-21 09:05:18.737439: step: 1508/526, loss: 0.9934319257736206 2023-01-21 09:05:19.879270: step: 1512/526, loss: 0.2308272421360016 2023-01-21 09:05:21.006185: step: 1516/526, loss: 0.1680740863084793 2023-01-21 09:05:22.139731: step: 1520/526, loss: 0.404039204120636 2023-01-21 09:05:23.253809: step: 1524/526, loss: 0.04582517221570015 2023-01-21 09:05:24.386139: step: 1528/526, loss: 0.14079482853412628 2023-01-21 09:05:25.493745: step: 1532/526, loss: 0.20451106131076813 2023-01-21 09:05:26.647744: step: 1536/526, loss: 0.23486891388893127 2023-01-21 09:05:27.804163: step: 1540/526, loss: 1.179813265800476 2023-01-21 09:05:28.925615: step: 1544/526, loss: 0.12898799777030945 2023-01-21 09:05:30.073963: step: 1548/526, loss: 1.0655649900436401 2023-01-21 09:05:31.223453: step: 1552/526, loss: 0.11519122123718262 2023-01-21 09:05:32.353406: step: 1556/526, loss: 0.961085319519043 2023-01-21 09:05:33.476364: step: 1560/526, loss: 0.8817575573921204 2023-01-21 09:05:34.631895: step: 1564/526, loss: 0.6503612399101257 2023-01-21 09:05:35.766243: step: 1568/526, loss: 0.7036305665969849 2023-01-21 09:05:36.885895: step: 1572/526, loss: 0.14672155678272247 2023-01-21 09:05:38.035529: step: 1576/526, loss: 0.05438194423913956 2023-01-21 09:05:39.178045: step: 1580/526, loss: 0.162797212600708 2023-01-21 09:05:40.306606: step: 1584/526, loss: 1.4852162599563599 2023-01-21 09:05:41.444906: step: 1588/526, loss: 0.9209295511245728 2023-01-21 09:05:42.564407: step: 1592/526, loss: 0.16470471024513245 2023-01-21 09:05:43.695667: step: 1596/526, loss: 0.1885172724723816 2023-01-21 09:05:44.809519: step: 1600/526, loss: 0.13555240631103516 2023-01-21 09:05:45.952791: step: 1604/526, loss: 1.318207025527954 2023-01-21 09:05:47.094784: step: 1608/526, loss: 0.604112982749939 2023-01-21 09:05:48.235724: step: 1612/526, loss: 1.2439247369766235 2023-01-21 09:05:49.367265: step: 1616/526, loss: 0.14844894409179688 2023-01-21 09:05:50.511108: step: 1620/526, loss: 0.114264577627182 2023-01-21 09:05:51.662370: step: 1624/526, loss: 0.7023523449897766 2023-01-21 09:05:52.815232: step: 1628/526, loss: 0.1391819268465042 2023-01-21 09:05:53.965481: step: 1632/526, loss: 0.41977816820144653 2023-01-21 09:05:55.102856: step: 1636/526, loss: 0.019954919815063477 2023-01-21 09:05:56.270937: step: 1640/526, loss: 0.17443951964378357 2023-01-21 09:05:57.402696: step: 1644/526, loss: 0.08527708053588867 2023-01-21 09:05:58.529507: step: 1648/526, loss: 0.16211938858032227 2023-01-21 09:05:59.676852: step: 1652/526, loss: 0.6825048923492432 2023-01-21 09:06:00.814206: step: 1656/526, loss: 0.2446775883436203 2023-01-21 09:06:01.950866: step: 1660/526, loss: 0.13465556502342224 2023-01-21 09:06:03.084154: step: 1664/526, loss: 0.2758150100708008 2023-01-21 09:06:04.232690: step: 1668/526, loss: 0.16842412948608398 2023-01-21 09:06:05.379245: step: 1672/526, loss: 0.5834553837776184 2023-01-21 09:06:06.523248: step: 1676/526, loss: 0.5031963586807251 2023-01-21 09:06:07.702916: step: 1680/526, loss: 0.7752524614334106 2023-01-21 09:06:08.853322: step: 1684/526, loss: 0.4572591781616211 2023-01-21 09:06:10.001018: step: 1688/526, loss: 0.20999452471733093 2023-01-21 09:06:11.133799: step: 1692/526, loss: 0.12950201332569122 2023-01-21 09:06:12.235296: step: 1696/526, loss: 0.12381734699010849 2023-01-21 09:06:13.370654: step: 1700/526, loss: 1.1109201908111572 2023-01-21 09:06:14.484064: step: 1704/526, loss: 0.23636063933372498 2023-01-21 09:06:15.645362: step: 1708/526, loss: 0.1920696347951889 2023-01-21 09:06:16.769082: step: 1712/526, loss: 0.2930752635002136 2023-01-21 09:06:17.890515: step: 1716/526, loss: 0.2131677269935608 2023-01-21 09:06:19.018484: step: 1720/526, loss: 0.2644510269165039 2023-01-21 09:06:20.165915: step: 1724/526, loss: 0.08906535804271698 2023-01-21 09:06:21.291161: step: 1728/526, loss: 1.166796326637268 2023-01-21 09:06:22.416561: step: 1732/526, loss: 1.7660728693008423 2023-01-21 09:06:23.559217: step: 1736/526, loss: 0.13750696182250977 2023-01-21 09:06:24.698826: step: 1740/526, loss: 1.9681421518325806 2023-01-21 09:06:25.826543: step: 1744/526, loss: 0.08952713012695312 2023-01-21 09:06:26.927239: step: 1748/526, loss: 0.7064719200134277 2023-01-21 09:06:28.085339: step: 1752/526, loss: 1.5497710704803467 2023-01-21 09:06:29.242369: step: 1756/526, loss: 0.2173960655927658 2023-01-21 09:06:30.401301: step: 1760/526, loss: 0.4258689880371094 2023-01-21 09:06:31.542550: step: 1764/526, loss: 8.201467514038086 2023-01-21 09:06:32.666062: step: 1768/526, loss: 0.2752320170402527 2023-01-21 09:06:33.818276: step: 1772/526, loss: 0.1689353585243225 2023-01-21 09:06:34.962023: step: 1776/526, loss: 0.09599890559911728 2023-01-21 09:06:36.092053: step: 1780/526, loss: 0.06656942516565323 2023-01-21 09:06:37.189872: step: 1784/526, loss: 0.37990808486938477 2023-01-21 09:06:38.338853: step: 1788/526, loss: 0.6800745129585266 2023-01-21 09:06:39.446083: step: 1792/526, loss: 0.12408037483692169 2023-01-21 09:06:40.564202: step: 1796/526, loss: 0.4488007128238678 2023-01-21 09:06:41.719691: step: 1800/526, loss: 0.1357475221157074 2023-01-21 09:06:42.857617: step: 1804/526, loss: 0.582709014415741 2023-01-21 09:06:43.970263: step: 1808/526, loss: 0.057540036737918854 2023-01-21 09:06:45.136087: step: 1812/526, loss: 1.0467431545257568 2023-01-21 09:06:46.265648: step: 1816/526, loss: 0.08513914048671722 2023-01-21 09:06:47.424589: step: 1820/526, loss: 1.3753395080566406 2023-01-21 09:06:48.535299: step: 1824/526, loss: 0.10260339081287384 2023-01-21 09:06:49.626928: step: 1828/526, loss: 0.21843966841697693 2023-01-21 09:06:50.761691: step: 1832/526, loss: 0.18705901503562927 2023-01-21 09:06:51.902160: step: 1836/526, loss: 0.0969705581665039 2023-01-21 09:06:53.035017: step: 1840/526, loss: 0.19723044335842133 2023-01-21 09:06:54.127477: step: 1844/526, loss: 0.6745023131370544 2023-01-21 09:06:55.248059: step: 1848/526, loss: 0.6798454523086548 2023-01-21 09:06:56.396966: step: 1852/526, loss: 0.33070430159568787 2023-01-21 09:06:57.532215: step: 1856/526, loss: 0.0542755126953125 2023-01-21 09:06:58.668398: step: 1860/526, loss: 0.12937946617603302 2023-01-21 09:06:59.813731: step: 1864/526, loss: 0.6542806029319763 2023-01-21 09:07:00.938335: step: 1868/526, loss: 0.11233168095350266 2023-01-21 09:07:02.067005: step: 1872/526, loss: 0.22717523574829102 2023-01-21 09:07:03.177540: step: 1876/526, loss: 0.8232075572013855 2023-01-21 09:07:04.298373: step: 1880/526, loss: 0.11775856465101242 2023-01-21 09:07:05.469441: step: 1884/526, loss: 7.02653169631958 2023-01-21 09:07:06.605691: step: 1888/526, loss: 0.0841517448425293 2023-01-21 09:07:07.846849: step: 1892/526, loss: 0.241241455078125 2023-01-21 09:07:08.953795: step: 1896/526, loss: 0.8380448818206787 2023-01-21 09:07:10.108264: step: 1900/526, loss: 0.5783689618110657 2023-01-21 09:07:11.228054: step: 1904/526, loss: 0.09035640209913254 2023-01-21 09:07:12.366715: step: 1908/526, loss: 0.26674729585647583 2023-01-21 09:07:13.528179: step: 1912/526, loss: 0.15535469353199005 2023-01-21 09:07:14.670944: step: 1916/526, loss: 0.90913987159729 2023-01-21 09:07:15.800693: step: 1920/526, loss: 0.24711419641971588 2023-01-21 09:07:16.929948: step: 1924/526, loss: 0.6851876378059387 2023-01-21 09:07:18.078232: step: 1928/526, loss: 0.7727624773979187 2023-01-21 09:07:19.224020: step: 1932/526, loss: 0.15901513397693634 2023-01-21 09:07:20.398743: step: 1936/526, loss: 0.2840121388435364 2023-01-21 09:07:21.547865: step: 1940/526, loss: 0.23126983642578125 2023-01-21 09:07:22.682521: step: 1944/526, loss: 1.0606579780578613 2023-01-21 09:07:23.814128: step: 1948/526, loss: 0.12540951371192932 2023-01-21 09:07:24.949074: step: 1952/526, loss: 0.12876950204372406 2023-01-21 09:07:26.082318: step: 1956/526, loss: 0.7682791948318481 2023-01-21 09:07:27.246695: step: 1960/526, loss: 0.4332424998283386 2023-01-21 09:07:28.371861: step: 1964/526, loss: 0.16579023003578186 2023-01-21 09:07:29.506422: step: 1968/526, loss: 0.32374557852745056 2023-01-21 09:07:30.652948: step: 1972/526, loss: 0.2833237648010254 2023-01-21 09:07:31.786203: step: 1976/526, loss: 0.14649434387683868 2023-01-21 09:07:32.934125: step: 1980/526, loss: 0.4473107159137726 2023-01-21 09:07:34.067255: step: 1984/526, loss: 0.04439125210046768 2023-01-21 09:07:35.185321: step: 1988/526, loss: 0.07865715026855469 2023-01-21 09:07:36.315255: step: 1992/526, loss: 0.1248352974653244 2023-01-21 09:07:37.443504: step: 1996/526, loss: 0.24855881929397583 2023-01-21 09:07:38.576755: step: 2000/526, loss: 0.20954833924770355 2023-01-21 09:07:39.705686: step: 2004/526, loss: 0.07233085483312607 2023-01-21 09:07:40.871679: step: 2008/526, loss: 0.12671375274658203 2023-01-21 09:07:41.995097: step: 2012/526, loss: 0.16479754447937012 2023-01-21 09:07:43.102008: step: 2016/526, loss: 0.0788472592830658 2023-01-21 09:07:44.218492: step: 2020/526, loss: 0.25870028138160706 2023-01-21 09:07:45.359143: step: 2024/526, loss: 0.15761509537696838 2023-01-21 09:07:46.484185: step: 2028/526, loss: 0.12140502780675888 2023-01-21 09:07:47.632447: step: 2032/526, loss: 0.1563182920217514 2023-01-21 09:07:48.792504: step: 2036/526, loss: 0.2727026045322418 2023-01-21 09:07:49.943186: step: 2040/526, loss: 0.48599594831466675 2023-01-21 09:07:51.080668: step: 2044/526, loss: 0.21796751022338867 2023-01-21 09:07:52.205731: step: 2048/526, loss: 0.17315025627613068 2023-01-21 09:07:53.376585: step: 2052/526, loss: 1.8426520824432373 2023-01-21 09:07:54.501766: step: 2056/526, loss: 0.175628662109375 2023-01-21 09:07:55.656430: step: 2060/526, loss: 0.27656039595603943 2023-01-21 09:07:56.799610: step: 2064/526, loss: 0.1536390781402588 2023-01-21 09:07:57.933222: step: 2068/526, loss: 0.10318709164857864 2023-01-21 09:07:59.075544: step: 2072/526, loss: 0.38626405596733093 2023-01-21 09:08:00.226167: step: 2076/526, loss: 0.14542122185230255 2023-01-21 09:08:01.380986: step: 2080/526, loss: 0.21770288050174713 2023-01-21 09:08:02.466346: step: 2084/526, loss: 0.6639150977134705 2023-01-21 09:08:03.613565: step: 2088/526, loss: 0.668742835521698 2023-01-21 09:08:04.746938: step: 2092/526, loss: 0.12401580810546875 2023-01-21 09:08:05.890063: step: 2096/526, loss: 0.14313897490501404 2023-01-21 09:08:07.035851: step: 2100/526, loss: 0.46609991788864136 2023-01-21 09:08:08.212076: step: 2104/526, loss: 1.8064841032028198 ================================================== Loss: 0.603 -------------------- Dev: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Test: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Korean: {'event': {'p': 0.7575757575757576, 'r': 0.3968253968253968, 'f1': 0.5208333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Russian: {'event': {'p': 0.44, 'r': 0.3055555555555556, 'f1': 0.36065573770491804}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} New best chinese model... New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.7575757575757576, 'r': 0.3968253968253968, 'f1': 0.5208333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.6193853427895981, 'r': 0.6977363515312917, 'f1': 0.6562304320601127}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Eng Test for Russian: {'event': {'p': 0.5890142226581657, 'r': 0.6268267223382046, 'f1': 0.6073324905183313}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} Sample Russian: {'event': {'p': 0.5, 'r': 0.3611111111111111, 'f1': 0.4193548387096774}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 1} ****************************** Epoch: 3 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:09:06.732629: step: 4/526, loss: 0.09815549850463867 2023-01-21 09:09:07.863813: step: 8/526, loss: 0.778786301612854 2023-01-21 09:09:08.994378: step: 12/526, loss: 0.26237422227859497 2023-01-21 09:09:10.146512: step: 16/526, loss: 1.166905164718628 2023-01-21 09:09:11.272900: step: 20/526, loss: 0.48532190918922424 2023-01-21 09:09:12.404671: step: 24/526, loss: 0.1599363535642624 2023-01-21 09:09:13.538895: step: 28/526, loss: 0.7683050036430359 2023-01-21 09:09:14.675113: step: 32/526, loss: 0.24998140335083008 2023-01-21 09:09:15.831834: step: 36/526, loss: 0.16066434979438782 2023-01-21 09:09:16.960003: step: 40/526, loss: 0.40726643800735474 2023-01-21 09:09:18.103174: step: 44/526, loss: 0.39912334084510803 2023-01-21 09:09:19.282813: step: 48/526, loss: 0.7967689037322998 2023-01-21 09:09:20.424832: step: 52/526, loss: 0.15700674057006836 2023-01-21 09:09:21.557964: step: 56/526, loss: 0.052364349365234375 2023-01-21 09:09:22.678438: step: 60/526, loss: 0.017160989344120026 2023-01-21 09:09:23.826611: step: 64/526, loss: 0.15097865462303162 2023-01-21 09:09:24.934676: step: 68/526, loss: 5.9862589836120605 2023-01-21 09:09:26.048705: step: 72/526, loss: 0.2158377766609192 2023-01-21 09:09:27.185755: step: 76/526, loss: 0.24023528397083282 2023-01-21 09:09:28.362546: step: 80/526, loss: 0.5475010871887207 2023-01-21 09:09:29.510536: step: 84/526, loss: 0.7246114015579224 2023-01-21 09:09:30.654312: step: 88/526, loss: 0.16104355454444885 2023-01-21 09:09:31.765655: step: 92/526, loss: 1.012978434562683 2023-01-21 09:09:32.890046: step: 96/526, loss: 0.12779411673545837 2023-01-21 09:09:34.034108: step: 100/526, loss: 0.055496931076049805 2023-01-21 09:09:35.198329: step: 104/526, loss: 0.7779055833816528 2023-01-21 09:09:36.322715: step: 108/526, loss: 0.5048141479492188 2023-01-21 09:09:37.492561: step: 112/526, loss: 0.9158059358596802 2023-01-21 09:09:38.594659: step: 116/526, loss: 0.39264994859695435 2023-01-21 09:09:39.720204: step: 120/526, loss: 0.11361436545848846 2023-01-21 09:09:40.831560: step: 124/526, loss: 0.7745078802108765 2023-01-21 09:09:41.942587: step: 128/526, loss: 0.05784139782190323 2023-01-21 09:09:43.084860: step: 132/526, loss: 0.1789465993642807 2023-01-21 09:09:44.237302: step: 136/526, loss: 0.466744989156723 2023-01-21 09:09:45.382484: step: 140/526, loss: 0.32892686128616333 2023-01-21 09:09:46.515361: step: 144/526, loss: 6.197875022888184 2023-01-21 09:09:47.670187: step: 148/526, loss: 0.18039646744728088 2023-01-21 09:09:48.840121: step: 152/526, loss: 1.033241629600525 2023-01-21 09:09:49.955434: step: 156/526, loss: 0.41054707765579224 2023-01-21 09:09:51.065797: step: 160/526, loss: 0.0938936248421669 2023-01-21 09:09:52.189714: step: 164/526, loss: 0.661960780620575 2023-01-21 09:09:53.357186: step: 168/526, loss: 0.5396683812141418 2023-01-21 09:09:54.479556: step: 172/526, loss: 0.23382607102394104 2023-01-21 09:09:55.604598: step: 176/526, loss: 0.7011641263961792 2023-01-21 09:09:56.743482: step: 180/526, loss: 0.6363601684570312 2023-01-21 09:09:57.866311: step: 184/526, loss: 0.10609379410743713 2023-01-21 09:09:59.010782: step: 188/526, loss: 0.7307679653167725 2023-01-21 09:10:00.146419: step: 192/526, loss: 0.873341977596283 2023-01-21 09:10:01.293921: step: 196/526, loss: 0.15034499764442444 2023-01-21 09:10:02.414503: step: 200/526, loss: 0.18926949799060822 2023-01-21 09:10:03.551996: step: 204/526, loss: 0.2140326052904129 2023-01-21 09:10:04.664927: step: 208/526, loss: 0.13158798217773438 2023-01-21 09:10:05.790461: step: 212/526, loss: 0.14059486985206604 2023-01-21 09:10:06.942572: step: 216/526, loss: 0.3620586395263672 2023-01-21 09:10:08.069403: step: 220/526, loss: 0.09814868122339249 2023-01-21 09:10:09.186049: step: 224/526, loss: 0.14739857614040375 2023-01-21 09:10:10.329037: step: 228/526, loss: 0.8174866437911987 2023-01-21 09:10:11.486915: step: 232/526, loss: 0.6810035705566406 2023-01-21 09:10:12.616308: step: 236/526, loss: 0.729864776134491 2023-01-21 09:10:13.742261: step: 240/526, loss: 0.05735397711396217 2023-01-21 09:10:14.912686: step: 244/526, loss: 0.6320679783821106 2023-01-21 09:10:16.023223: step: 248/526, loss: 0.026183797046542168 2023-01-21 09:10:17.176110: step: 252/526, loss: 0.18542452156543732 2023-01-21 09:10:18.349564: step: 256/526, loss: 0.15911169350147247 2023-01-21 09:10:19.538327: step: 260/526, loss: 0.16515932977199554 2023-01-21 09:10:20.707569: step: 264/526, loss: 0.10121984034776688 2023-01-21 09:10:21.909119: step: 268/526, loss: 0.15083780884742737 2023-01-21 09:10:23.040888: step: 272/526, loss: 0.149017333984375 2023-01-21 09:10:24.205336: step: 276/526, loss: 1.0262253284454346 2023-01-21 09:10:25.325575: step: 280/526, loss: 0.3056522309780121 2023-01-21 09:10:26.449074: step: 284/526, loss: 0.7167744040489197 2023-01-21 09:10:27.577270: step: 288/526, loss: 0.06786961853504181 2023-01-21 09:10:28.678776: step: 292/526, loss: 0.12627820670604706 2023-01-21 09:10:29.815065: step: 296/526, loss: 0.16959428787231445 2023-01-21 09:10:30.972774: step: 300/526, loss: 0.5408000946044922 2023-01-21 09:10:32.110372: step: 304/526, loss: 0.748309850692749 2023-01-21 09:10:33.238573: step: 308/526, loss: 0.1924075037240982 2023-01-21 09:10:34.374544: step: 312/526, loss: 0.18067866563796997 2023-01-21 09:10:35.537220: step: 316/526, loss: 0.06243553012609482 2023-01-21 09:10:36.699630: step: 320/526, loss: 0.12429027259349823 2023-01-21 09:10:37.839275: step: 324/526, loss: 0.10332375019788742 2023-01-21 09:10:38.988831: step: 328/526, loss: 0.2423768937587738 2023-01-21 09:10:40.120456: step: 332/526, loss: 0.6232358813285828 2023-01-21 09:10:41.242466: step: 336/526, loss: 0.2174820601940155 2023-01-21 09:10:42.377344: step: 340/526, loss: 1.1696345806121826 2023-01-21 09:10:43.501650: step: 344/526, loss: 0.6681364178657532 2023-01-21 09:10:44.637135: step: 348/526, loss: 0.761664867401123 2023-01-21 09:10:45.781714: step: 352/526, loss: 0.6626855731010437 2023-01-21 09:10:46.905460: step: 356/526, loss: 0.2573251724243164 2023-01-21 09:10:48.023976: step: 360/526, loss: 0.12913647294044495 2023-01-21 09:10:49.148938: step: 364/526, loss: 0.6908438205718994 2023-01-21 09:10:50.294787: step: 368/526, loss: 6.724062442779541 2023-01-21 09:10:51.453135: step: 372/526, loss: 0.1610884666442871 2023-01-21 09:10:52.582878: step: 376/526, loss: 0.13350383937358856 2023-01-21 09:10:53.711809: step: 380/526, loss: 0.315802663564682 2023-01-21 09:10:54.827149: step: 384/526, loss: 0.09661498665809631 2023-01-21 09:10:55.968514: step: 388/526, loss: 0.283730149269104 2023-01-21 09:10:57.083853: step: 392/526, loss: 0.2510172724723816 2023-01-21 09:10:58.212861: step: 396/526, loss: 0.21686744689941406 2023-01-21 09:10:59.378309: step: 400/526, loss: 0.8394155502319336 2023-01-21 09:11:00.551042: step: 404/526, loss: 0.19308948516845703 2023-01-21 09:11:01.681047: step: 408/526, loss: 0.5364732146263123 2023-01-21 09:11:02.794591: step: 412/526, loss: 0.13504552841186523 2023-01-21 09:11:03.924567: step: 416/526, loss: 0.05910434573888779 2023-01-21 09:11:05.040258: step: 420/526, loss: 0.09674377739429474 2023-01-21 09:11:06.164638: step: 424/526, loss: 0.2659017741680145 2023-01-21 09:11:07.303387: step: 428/526, loss: 0.10842514038085938 2023-01-21 09:11:08.429639: step: 432/526, loss: 0.6830092072486877 2023-01-21 09:11:09.573219: step: 436/526, loss: 0.11732621490955353 2023-01-21 09:11:10.696252: step: 440/526, loss: 0.06782722473144531 2023-01-21 09:11:11.837658: step: 444/526, loss: 1.1526572704315186 2023-01-21 09:11:12.969620: step: 448/526, loss: 0.8452734351158142 2023-01-21 09:11:14.109690: step: 452/526, loss: 0.0719631165266037 2023-01-21 09:11:15.250302: step: 456/526, loss: 0.11287655681371689 2023-01-21 09:11:16.420249: step: 460/526, loss: 0.08829379081726074 2023-01-21 09:11:17.525810: step: 464/526, loss: 0.24529676139354706 2023-01-21 09:11:18.708087: step: 468/526, loss: 0.2507755160331726 2023-01-21 09:11:19.862972: step: 472/526, loss: 0.1434486359357834 2023-01-21 09:11:21.021882: step: 476/526, loss: 1.9669455289840698 2023-01-21 09:11:22.181843: step: 480/526, loss: 0.03885490819811821 2023-01-21 09:11:23.331525: step: 484/526, loss: 0.333885133266449 2023-01-21 09:11:24.445239: step: 488/526, loss: 0.24597254395484924 2023-01-21 09:11:25.591917: step: 492/526, loss: 0.15722675621509552 2023-01-21 09:11:26.757900: step: 496/526, loss: 0.2573823630809784 2023-01-21 09:11:27.901660: step: 500/526, loss: 0.05392961576581001 2023-01-21 09:11:29.087706: step: 504/526, loss: 0.06517043709754944 2023-01-21 09:11:30.212209: step: 508/526, loss: 0.15791350603103638 2023-01-21 09:11:31.318496: step: 512/526, loss: 0.1807699203491211 2023-01-21 09:11:32.432864: step: 516/526, loss: 0.14367209374904633 2023-01-21 09:11:33.571882: step: 520/526, loss: 0.2890182435512543 2023-01-21 09:11:34.703872: step: 524/526, loss: 0.2728877663612366 2023-01-21 09:11:35.834607: step: 528/526, loss: 0.15734797716140747 2023-01-21 09:11:36.952290: step: 532/526, loss: 0.0902770459651947 2023-01-21 09:11:38.099374: step: 536/526, loss: 0.042649269104003906 2023-01-21 09:11:39.238844: step: 540/526, loss: 0.7358492612838745 2023-01-21 09:11:40.386020: step: 544/526, loss: 0.19873514771461487 2023-01-21 09:11:41.526535: step: 548/526, loss: 0.14755135774612427 2023-01-21 09:11:42.690479: step: 552/526, loss: 0.2229456901550293 2023-01-21 09:11:43.874779: step: 556/526, loss: 0.9254617691040039 2023-01-21 09:11:45.005352: step: 560/526, loss: 0.2923905551433563 2023-01-21 09:11:46.112522: step: 564/526, loss: 0.314846396446228 2023-01-21 09:11:47.232222: step: 568/526, loss: 0.3457811772823334 2023-01-21 09:11:48.375025: step: 572/526, loss: 0.15723343193531036 2023-01-21 09:11:49.556206: step: 576/526, loss: 1.830252766609192 2023-01-21 09:11:50.668223: step: 580/526, loss: 0.032427262514829636 2023-01-21 09:11:51.788813: step: 584/526, loss: 0.1188659742474556 2023-01-21 09:11:52.911040: step: 588/526, loss: 1.2814996242523193 2023-01-21 09:11:54.041781: step: 592/526, loss: 1.1448551416397095 2023-01-21 09:11:55.200615: step: 596/526, loss: 0.3831920623779297 2023-01-21 09:11:56.320172: step: 600/526, loss: 0.06013665348291397 2023-01-21 09:11:57.449966: step: 604/526, loss: 0.690144419670105 2023-01-21 09:11:58.580587: step: 608/526, loss: 0.20937128365039825 2023-01-21 09:11:59.733517: step: 612/526, loss: 0.06971664726734161 2023-01-21 09:12:00.862562: step: 616/526, loss: 0.19894886016845703 2023-01-21 09:12:02.008247: step: 620/526, loss: 0.07591447979211807 2023-01-21 09:12:03.112050: step: 624/526, loss: 5.988174915313721 2023-01-21 09:12:04.243711: step: 628/526, loss: 0.1725023239850998 2023-01-21 09:12:05.375663: step: 632/526, loss: 0.06016264110803604 2023-01-21 09:12:06.497253: step: 636/526, loss: 0.7884900569915771 2023-01-21 09:12:07.611061: step: 640/526, loss: 1.5777852535247803 2023-01-21 09:12:08.756077: step: 644/526, loss: 0.14996632933616638 2023-01-21 09:12:09.895380: step: 648/526, loss: 0.17883777618408203 2023-01-21 09:12:11.039491: step: 652/526, loss: 1.7639210224151611 2023-01-21 09:12:12.160541: step: 656/526, loss: 0.05362577736377716 2023-01-21 09:12:13.346124: step: 660/526, loss: 0.08198356628417969 2023-01-21 09:12:14.484253: step: 664/526, loss: 0.1766609251499176 2023-01-21 09:12:15.636376: step: 668/526, loss: 0.167727991938591 2023-01-21 09:12:16.762717: step: 672/526, loss: 0.17878219485282898 2023-01-21 09:12:17.889605: step: 676/526, loss: 0.051870059221982956 2023-01-21 09:12:19.066704: step: 680/526, loss: 0.13780784606933594 2023-01-21 09:12:20.192992: step: 684/526, loss: 0.1823573112487793 2023-01-21 09:12:21.299431: step: 688/526, loss: 0.10534921288490295 2023-01-21 09:12:22.458877: step: 692/526, loss: 0.3059524595737457 2023-01-21 09:12:23.588720: step: 696/526, loss: 0.30730992555618286 2023-01-21 09:12:24.735182: step: 700/526, loss: 0.33931687474250793 2023-01-21 09:12:25.848306: step: 704/526, loss: 0.9796249866485596 2023-01-21 09:12:26.992450: step: 708/526, loss: 0.31933584809303284 2023-01-21 09:12:28.129139: step: 712/526, loss: 0.4349270761013031 2023-01-21 09:12:29.256169: step: 716/526, loss: 0.9338370561599731 2023-01-21 09:12:30.396415: step: 720/526, loss: 0.0669802725315094 2023-01-21 09:12:31.525570: step: 724/526, loss: 0.1706235408782959 2023-01-21 09:12:32.661088: step: 728/526, loss: 0.7605264782905579 2023-01-21 09:12:33.788524: step: 732/526, loss: 0.7951940298080444 2023-01-21 09:12:34.938403: step: 736/526, loss: 0.0916227400302887 2023-01-21 09:12:36.056409: step: 740/526, loss: 0.5308011174201965 2023-01-21 09:12:37.175194: step: 744/526, loss: 0.12734170258045197 2023-01-21 09:12:38.308179: step: 748/526, loss: 0.4225795567035675 2023-01-21 09:12:39.450851: step: 752/526, loss: 0.9455537796020508 2023-01-21 09:12:40.624350: step: 756/526, loss: 0.11672095954418182 2023-01-21 09:12:41.763242: step: 760/526, loss: 0.4404076635837555 2023-01-21 09:12:42.910038: step: 764/526, loss: 0.36769819259643555 2023-01-21 09:12:44.038879: step: 768/526, loss: 0.1434079110622406 2023-01-21 09:12:45.159798: step: 772/526, loss: 0.23262205719947815 2023-01-21 09:12:46.258971: step: 776/526, loss: 0.0569973923265934 2023-01-21 09:12:47.381291: step: 780/526, loss: 0.09339595586061478 2023-01-21 09:12:48.512169: step: 784/526, loss: 0.12500867247581482 2023-01-21 09:12:49.661531: step: 788/526, loss: 2.8017613887786865 2023-01-21 09:12:50.833836: step: 792/526, loss: 0.2985422909259796 2023-01-21 09:12:51.956467: step: 796/526, loss: 0.6917413473129272 2023-01-21 09:12:53.070053: step: 800/526, loss: 1.242168664932251 2023-01-21 09:12:54.200410: step: 804/526, loss: 0.19024498760700226 2023-01-21 09:12:55.339133: step: 808/526, loss: 1.0680491924285889 2023-01-21 09:12:56.493460: step: 812/526, loss: 0.21203525364398956 2023-01-21 09:12:57.614584: step: 816/526, loss: 1.0118982791900635 2023-01-21 09:12:58.751621: step: 820/526, loss: 0.1602334976196289 2023-01-21 09:12:59.877415: step: 824/526, loss: 0.020087718963623047 2023-01-21 09:13:00.984501: step: 828/526, loss: 0.25196534395217896 2023-01-21 09:13:02.110807: step: 832/526, loss: 0.1407441645860672 2023-01-21 09:13:03.243511: step: 836/526, loss: 0.13415169715881348 2023-01-21 09:13:04.333422: step: 840/526, loss: 0.07295289635658264 2023-01-21 09:13:05.463902: step: 844/526, loss: 0.11090050637722015 2023-01-21 09:13:06.590376: step: 848/526, loss: 0.08863496780395508 2023-01-21 09:13:07.700894: step: 852/526, loss: 6.1598801612854 2023-01-21 09:13:08.867510: step: 856/526, loss: 0.21306943893432617 2023-01-21 09:13:10.031887: step: 860/526, loss: 0.3341161608695984 2023-01-21 09:13:11.168660: step: 864/526, loss: 0.8827815055847168 2023-01-21 09:13:12.302993: step: 868/526, loss: 0.24832621216773987 2023-01-21 09:13:13.436146: step: 872/526, loss: 0.748885452747345 2023-01-21 09:13:14.575057: step: 876/526, loss: 0.14316312968730927 2023-01-21 09:13:15.710235: step: 880/526, loss: 0.1617463082075119 2023-01-21 09:13:16.828934: step: 884/526, loss: 1.1235769987106323 2023-01-21 09:13:17.962673: step: 888/526, loss: 0.2687419354915619 2023-01-21 09:13:19.085917: step: 892/526, loss: 5.968334197998047 2023-01-21 09:13:20.212329: step: 896/526, loss: 0.1484842300415039 2023-01-21 09:13:21.363788: step: 900/526, loss: 0.15555091202259064 2023-01-21 09:13:22.468421: step: 904/526, loss: 0.184285968542099 2023-01-21 09:13:23.634118: step: 908/526, loss: 0.11641374230384827 2023-01-21 09:13:24.796682: step: 912/526, loss: 0.09140090644359589 2023-01-21 09:13:25.944182: step: 916/526, loss: 0.6824744343757629 2023-01-21 09:13:27.100132: step: 920/526, loss: 0.05462322384119034 2023-01-21 09:13:28.232591: step: 924/526, loss: 0.12663468718528748 2023-01-21 09:13:29.352561: step: 928/526, loss: 0.05434150993824005 2023-01-21 09:13:30.474508: step: 932/526, loss: 0.214741051197052 2023-01-21 09:13:31.574542: step: 936/526, loss: 0.07408924400806427 2023-01-21 09:13:32.716724: step: 940/526, loss: 0.5570886135101318 2023-01-21 09:13:33.858978: step: 944/526, loss: 1.0303354263305664 2023-01-21 09:13:34.980762: step: 948/526, loss: 0.1643667221069336 2023-01-21 09:13:36.087407: step: 952/526, loss: 0.23496118187904358 2023-01-21 09:13:37.232623: step: 956/526, loss: 0.19179022312164307 2023-01-21 09:13:38.370989: step: 960/526, loss: 1.0404634475708008 2023-01-21 09:13:39.499291: step: 964/526, loss: 0.05198941379785538 2023-01-21 09:13:40.651779: step: 968/526, loss: 0.5997173190116882 2023-01-21 09:13:41.781763: step: 972/526, loss: 6.237577438354492 2023-01-21 09:13:42.919837: step: 976/526, loss: 0.10701847076416016 2023-01-21 09:13:44.094415: step: 980/526, loss: 0.2570188641548157 2023-01-21 09:13:45.235803: step: 984/526, loss: 0.3070027232170105 2023-01-21 09:13:46.371651: step: 988/526, loss: 1.7618612051010132 2023-01-21 09:13:47.506064: step: 992/526, loss: 0.8996191024780273 2023-01-21 09:13:48.613324: step: 996/526, loss: 0.5811383128166199 2023-01-21 09:13:49.754258: step: 1000/526, loss: 0.664803147315979 2023-01-21 09:13:50.901103: step: 1004/526, loss: 0.3209201991558075 2023-01-21 09:13:52.035639: step: 1008/526, loss: 0.4834885597229004 2023-01-21 09:13:53.164698: step: 1012/526, loss: 0.2890813946723938 2023-01-21 09:13:54.272858: step: 1016/526, loss: 0.6176754236221313 2023-01-21 09:13:55.425904: step: 1020/526, loss: 0.08344468474388123 2023-01-21 09:13:56.539970: step: 1024/526, loss: 0.7532416582107544 2023-01-21 09:13:57.668789: step: 1028/526, loss: 0.4188307821750641 2023-01-21 09:13:58.801246: step: 1032/526, loss: 0.7252113223075867 2023-01-21 09:13:59.921185: step: 1036/526, loss: 0.17177867889404297 2023-01-21 09:14:01.036099: step: 1040/526, loss: 0.06400027126073837 2023-01-21 09:14:02.158119: step: 1044/526, loss: 0.10868249088525772 2023-01-21 09:14:03.341064: step: 1048/526, loss: 0.11496695876121521 2023-01-21 09:14:04.444428: step: 1052/526, loss: 0.1320648193359375 2023-01-21 09:14:05.564767: step: 1056/526, loss: 0.019594576209783554 2023-01-21 09:14:06.705970: step: 1060/526, loss: 0.04037170484662056 2023-01-21 09:14:07.849797: step: 1064/526, loss: 0.06519465148448944 2023-01-21 09:14:08.973914: step: 1068/526, loss: 0.19335699081420898 2023-01-21 09:14:10.096327: step: 1072/526, loss: 0.22089740633964539 2023-01-21 09:14:11.241247: step: 1076/526, loss: 0.02359180524945259 2023-01-21 09:14:12.370574: step: 1080/526, loss: 0.07745294272899628 2023-01-21 09:14:13.488117: step: 1084/526, loss: 0.10699639469385147 2023-01-21 09:14:14.624459: step: 1088/526, loss: 0.11986284703016281 2023-01-21 09:14:15.852231: step: 1092/526, loss: 0.09723816066980362 2023-01-21 09:14:16.981631: step: 1096/526, loss: 0.15060502290725708 2023-01-21 09:14:18.126917: step: 1100/526, loss: 0.25521835684776306 2023-01-21 09:14:19.252854: step: 1104/526, loss: 0.064361572265625 2023-01-21 09:14:20.374797: step: 1108/526, loss: 0.060060977935791016 2023-01-21 09:14:21.474832: step: 1112/526, loss: 0.04093647003173828 2023-01-21 09:14:22.591311: step: 1116/526, loss: 0.8150016069412231 2023-01-21 09:14:23.721967: step: 1120/526, loss: 1.2796907424926758 2023-01-21 09:14:24.849872: step: 1124/526, loss: 0.8485506772994995 2023-01-21 09:14:25.972810: step: 1128/526, loss: 0.745945155620575 2023-01-21 09:14:27.066016: step: 1132/526, loss: 0.30736014246940613 2023-01-21 09:14:28.199449: step: 1136/526, loss: 0.14030304551124573 2023-01-21 09:14:29.335318: step: 1140/526, loss: 0.06971073150634766 2023-01-21 09:14:30.474393: step: 1144/526, loss: 0.12525349855422974 2023-01-21 09:14:31.597908: step: 1148/526, loss: 6.075376033782959 2023-01-21 09:14:32.704508: step: 1152/526, loss: 0.09767132252454758 2023-01-21 09:14:33.837033: step: 1156/526, loss: 0.16151876747608185 2023-01-21 09:14:34.966438: step: 1160/526, loss: 0.03954639285802841 2023-01-21 09:14:36.078989: step: 1164/526, loss: 0.0148452278226614 2023-01-21 09:14:37.204833: step: 1168/526, loss: 0.46187302470207214 2023-01-21 09:14:38.339724: step: 1172/526, loss: 0.24328966438770294 2023-01-21 09:14:39.453214: step: 1176/526, loss: 0.05940687656402588 2023-01-21 09:14:40.627278: step: 1180/526, loss: 0.42158180475234985 2023-01-21 09:14:41.740825: step: 1184/526, loss: 0.29104599356651306 2023-01-21 09:14:42.857825: step: 1188/526, loss: 0.5036236047744751 2023-01-21 09:14:43.965648: step: 1192/526, loss: 0.04713907465338707 2023-01-21 09:14:45.115951: step: 1196/526, loss: 0.5283783078193665 2023-01-21 09:14:46.242230: step: 1200/526, loss: 0.10162925720214844 2023-01-21 09:14:47.387156: step: 1204/526, loss: 0.224803164601326 2023-01-21 09:14:48.501015: step: 1208/526, loss: 0.7440999150276184 2023-01-21 09:14:49.625759: step: 1212/526, loss: 0.08854293823242188 2023-01-21 09:14:50.750645: step: 1216/526, loss: 0.07764258235692978 2023-01-21 09:14:51.884876: step: 1220/526, loss: 1.003868579864502 2023-01-21 09:14:53.017074: step: 1224/526, loss: 0.14478349685668945 2023-01-21 09:14:54.140248: step: 1228/526, loss: 0.7403308153152466 2023-01-21 09:14:55.280273: step: 1232/526, loss: 0.08355551213026047 2023-01-21 09:14:56.420154: step: 1236/526, loss: 0.6245163083076477 2023-01-21 09:14:57.539517: step: 1240/526, loss: 0.580281138420105 2023-01-21 09:14:58.669397: step: 1244/526, loss: 0.7963274717330933 2023-01-21 09:14:59.791468: step: 1248/526, loss: 0.37675970792770386 2023-01-21 09:15:00.930425: step: 1252/526, loss: 0.8511053323745728 2023-01-21 09:15:02.071003: step: 1256/526, loss: 0.028861189261078835 2023-01-21 09:15:03.218924: step: 1260/526, loss: 0.05666828155517578 2023-01-21 09:15:04.346799: step: 1264/526, loss: 0.0780644416809082 2023-01-21 09:15:05.488212: step: 1268/526, loss: 1.047023057937622 2023-01-21 09:15:06.621953: step: 1272/526, loss: 0.09973640739917755 2023-01-21 09:15:07.760556: step: 1276/526, loss: 1.1144388914108276 2023-01-21 09:15:08.888869: step: 1280/526, loss: 0.9942584037780762 2023-01-21 09:15:10.010405: step: 1284/526, loss: 0.11726050823926926 2023-01-21 09:15:11.132825: step: 1288/526, loss: 0.24627934396266937 2023-01-21 09:15:12.265839: step: 1292/526, loss: 0.10007962584495544 2023-01-21 09:15:13.406744: step: 1296/526, loss: 0.04827446863055229 2023-01-21 09:15:14.519319: step: 1300/526, loss: 0.19467994570732117 2023-01-21 09:15:15.692243: step: 1304/526, loss: 0.25225162506103516 2023-01-21 09:15:16.840448: step: 1308/526, loss: 0.11150360107421875 2023-01-21 09:15:17.977760: step: 1312/526, loss: 1.4192262887954712 2023-01-21 09:15:19.106247: step: 1316/526, loss: 0.40786734223365784 2023-01-21 09:15:20.228713: step: 1320/526, loss: 0.06680698692798615 2023-01-21 09:15:21.360190: step: 1324/526, loss: 0.06681499630212784 2023-01-21 09:15:22.525497: step: 1328/526, loss: 0.22055187821388245 2023-01-21 09:15:23.636906: step: 1332/526, loss: 0.2694765031337738 2023-01-21 09:15:24.799701: step: 1336/526, loss: 0.9559891223907471 2023-01-21 09:15:25.985558: step: 1340/526, loss: 0.8912206292152405 2023-01-21 09:15:27.127722: step: 1344/526, loss: 0.7527163624763489 2023-01-21 09:15:28.279146: step: 1348/526, loss: 0.18553757667541504 2023-01-21 09:15:29.389833: step: 1352/526, loss: 0.23619085550308228 2023-01-21 09:15:30.539435: step: 1356/526, loss: 0.13935032486915588 2023-01-21 09:15:31.688745: step: 1360/526, loss: 0.06779885292053223 2023-01-21 09:15:32.815540: step: 1364/526, loss: 0.09334440529346466 2023-01-21 09:15:33.972921: step: 1368/526, loss: 0.20999857783317566 2023-01-21 09:15:35.104056: step: 1372/526, loss: 0.11324353516101837 2023-01-21 09:15:36.268575: step: 1376/526, loss: 0.13618770241737366 2023-01-21 09:15:37.394204: step: 1380/526, loss: 0.16702528297901154 2023-01-21 09:15:38.529275: step: 1384/526, loss: 0.5035191178321838 2023-01-21 09:15:39.660553: step: 1388/526, loss: 1.3180017471313477 2023-01-21 09:15:40.754357: step: 1392/526, loss: 0.2432619035243988 2023-01-21 09:15:41.852319: step: 1396/526, loss: 0.1816757172346115 2023-01-21 09:15:42.976040: step: 1400/526, loss: 0.2634263038635254 2023-01-21 09:15:44.076918: step: 1404/526, loss: 0.0674934983253479 2023-01-21 09:15:45.188912: step: 1408/526, loss: 0.25318947434425354 2023-01-21 09:15:46.321115: step: 1412/526, loss: 0.2854065001010895 2023-01-21 09:15:47.400067: step: 1416/526, loss: 0.3882119059562683 2023-01-21 09:15:48.526754: step: 1420/526, loss: 0.05792813375592232 2023-01-21 09:15:49.644247: step: 1424/526, loss: 0.38272157311439514 2023-01-21 09:15:50.751663: step: 1428/526, loss: 0.5331307649612427 2023-01-21 09:15:51.892957: step: 1432/526, loss: 0.13443392515182495 2023-01-21 09:15:53.026504: step: 1436/526, loss: 0.09153075516223907 2023-01-21 09:15:54.136894: step: 1440/526, loss: 0.07008647918701172 2023-01-21 09:15:55.303214: step: 1444/526, loss: 0.6439488530158997 2023-01-21 09:15:56.423649: step: 1448/526, loss: 1.077980875968933 2023-01-21 09:15:57.585539: step: 1452/526, loss: 0.44732579588890076 2023-01-21 09:15:58.735333: step: 1456/526, loss: 0.2313501387834549 2023-01-21 09:15:59.881450: step: 1460/526, loss: 0.11343111842870712 2023-01-21 09:16:01.029698: step: 1464/526, loss: 0.04825792461633682 2023-01-21 09:16:02.166640: step: 1468/526, loss: 0.3280300199985504 2023-01-21 09:16:03.301448: step: 1472/526, loss: 6.1980414390563965 2023-01-21 09:16:04.452161: step: 1476/526, loss: 0.7206302881240845 2023-01-21 09:16:05.583820: step: 1480/526, loss: 0.09934735298156738 2023-01-21 09:16:06.730721: step: 1484/526, loss: 0.3719167709350586 2023-01-21 09:16:07.867215: step: 1488/526, loss: 0.09125781804323196 2023-01-21 09:16:08.986731: step: 1492/526, loss: 0.10218754410743713 2023-01-21 09:16:10.129192: step: 1496/526, loss: 0.5845492482185364 2023-01-21 09:16:11.266237: step: 1500/526, loss: 0.7742104530334473 2023-01-21 09:16:12.427507: step: 1504/526, loss: 0.15945816040039062 2023-01-21 09:16:13.532538: step: 1508/526, loss: 0.11177407205104828 2023-01-21 09:16:14.649996: step: 1512/526, loss: 0.08908472210168839 2023-01-21 09:16:15.798490: step: 1516/526, loss: 0.20404338836669922 2023-01-21 09:16:16.953287: step: 1520/526, loss: 0.75209641456604 2023-01-21 09:16:18.091981: step: 1524/526, loss: 0.16547413170337677 2023-01-21 09:16:19.212414: step: 1528/526, loss: 0.10301017761230469 2023-01-21 09:16:20.325754: step: 1532/526, loss: 0.5217010378837585 2023-01-21 09:16:21.475400: step: 1536/526, loss: 0.12957553565502167 2023-01-21 09:16:22.589055: step: 1540/526, loss: 0.10902738571166992 2023-01-21 09:16:23.701699: step: 1544/526, loss: 0.05682273209095001 2023-01-21 09:16:24.837178: step: 1548/526, loss: 0.8489383459091187 2023-01-21 09:16:25.962943: step: 1552/526, loss: 0.8127043843269348 2023-01-21 09:16:27.085548: step: 1556/526, loss: 0.34159165620803833 2023-01-21 09:16:28.230358: step: 1560/526, loss: 0.1747705042362213 2023-01-21 09:16:29.371159: step: 1564/526, loss: 0.4296688139438629 2023-01-21 09:16:30.467488: step: 1568/526, loss: 0.1284904032945633 2023-01-21 09:16:31.626374: step: 1572/526, loss: 0.2206592559814453 2023-01-21 09:16:32.774447: step: 1576/526, loss: 0.12147083133459091 2023-01-21 09:16:33.898891: step: 1580/526, loss: 0.2198609858751297 2023-01-21 09:16:35.024426: step: 1584/526, loss: 0.2692309617996216 2023-01-21 09:16:36.199632: step: 1588/526, loss: 0.21258258819580078 2023-01-21 09:16:37.322535: step: 1592/526, loss: 0.07004199177026749 2023-01-21 09:16:38.482590: step: 1596/526, loss: 0.10684271156787872 2023-01-21 09:16:39.636451: step: 1600/526, loss: 0.08016996085643768 2023-01-21 09:16:40.778606: step: 1604/526, loss: 0.08632688969373703 2023-01-21 09:16:41.922586: step: 1608/526, loss: 0.28077125549316406 2023-01-21 09:16:43.114017: step: 1612/526, loss: 0.1809266209602356 2023-01-21 09:16:44.245434: step: 1616/526, loss: 0.1500811129808426 2023-01-21 09:16:45.399581: step: 1620/526, loss: 0.5783336758613586 2023-01-21 09:16:46.531483: step: 1624/526, loss: 0.46894150972366333 2023-01-21 09:16:47.664818: step: 1628/526, loss: 0.6204140186309814 2023-01-21 09:16:48.795167: step: 1632/526, loss: 0.17772865295410156 2023-01-21 09:16:49.919668: step: 1636/526, loss: 0.1586422324180603 2023-01-21 09:16:51.040465: step: 1640/526, loss: 0.048552703112363815 2023-01-21 09:16:52.179587: step: 1644/526, loss: 0.05477786436676979 2023-01-21 09:16:53.301258: step: 1648/526, loss: 0.15095233917236328 2023-01-21 09:16:54.429588: step: 1652/526, loss: 0.5051612854003906 2023-01-21 09:16:55.585224: step: 1656/526, loss: 0.1724916398525238 2023-01-21 09:16:56.724956: step: 1660/526, loss: 0.158155158162117 2023-01-21 09:16:57.882319: step: 1664/526, loss: 0.2265586405992508 2023-01-21 09:16:59.026635: step: 1668/526, loss: 0.4416227340698242 2023-01-21 09:17:00.147295: step: 1672/526, loss: 0.048676300793886185 2023-01-21 09:17:01.283860: step: 1676/526, loss: 0.18028397858142853 2023-01-21 09:17:02.435696: step: 1680/526, loss: 0.7623763680458069 2023-01-21 09:17:03.594158: step: 1684/526, loss: 0.3002418577671051 2023-01-21 09:17:04.741788: step: 1688/526, loss: 0.07221928238868713 2023-01-21 09:17:05.841922: step: 1692/526, loss: 0.1144382506608963 2023-01-21 09:17:06.969289: step: 1696/526, loss: 0.9417555928230286 2023-01-21 09:17:08.086976: step: 1700/526, loss: 0.7007885575294495 2023-01-21 09:17:09.230952: step: 1704/526, loss: 0.4379734694957733 2023-01-21 09:17:10.369430: step: 1708/526, loss: 0.27464112639427185 2023-01-21 09:17:11.503773: step: 1712/526, loss: 0.1818973571062088 2023-01-21 09:17:12.649295: step: 1716/526, loss: 0.15587759017944336 2023-01-21 09:17:13.783437: step: 1720/526, loss: 0.5782361030578613 2023-01-21 09:17:14.919172: step: 1724/526, loss: 0.3107401132583618 2023-01-21 09:17:16.030792: step: 1728/526, loss: 0.10782603919506073 2023-01-21 09:17:17.194381: step: 1732/526, loss: 0.18604522943496704 2023-01-21 09:17:18.313101: step: 1736/526, loss: 0.06759343296289444 2023-01-21 09:17:19.462313: step: 1740/526, loss: 0.13604554533958435 2023-01-21 09:17:20.582228: step: 1744/526, loss: 0.306617796421051 2023-01-21 09:17:21.716169: step: 1748/526, loss: 0.3607856035232544 2023-01-21 09:17:22.857065: step: 1752/526, loss: 0.1634874939918518 2023-01-21 09:17:24.000575: step: 1756/526, loss: 0.8868076801300049 2023-01-21 09:17:25.142667: step: 1760/526, loss: 0.3137521743774414 2023-01-21 09:17:26.314629: step: 1764/526, loss: 0.22938479483127594 2023-01-21 09:17:27.466441: step: 1768/526, loss: 0.6368406414985657 2023-01-21 09:17:28.596039: step: 1772/526, loss: 0.4434700012207031 2023-01-21 09:17:29.715023: step: 1776/526, loss: 0.20543384552001953 2023-01-21 09:17:30.834851: step: 1780/526, loss: 0.7326292991638184 2023-01-21 09:17:31.957357: step: 1784/526, loss: 0.13726906478405 2023-01-21 09:17:33.081948: step: 1788/526, loss: 0.14172545075416565 2023-01-21 09:17:34.217632: step: 1792/526, loss: 0.101201631128788 2023-01-21 09:17:35.341899: step: 1796/526, loss: 0.7137531042098999 2023-01-21 09:17:36.494111: step: 1800/526, loss: 0.5728021264076233 2023-01-21 09:17:37.624346: step: 1804/526, loss: 0.1522558331489563 2023-01-21 09:17:38.760473: step: 1808/526, loss: 0.3777618706226349 2023-01-21 09:17:39.899889: step: 1812/526, loss: 0.46837204694747925 2023-01-21 09:17:41.037668: step: 1816/526, loss: 0.40504521131515503 2023-01-21 09:17:42.155510: step: 1820/526, loss: 0.11311206221580505 2023-01-21 09:17:43.319214: step: 1824/526, loss: 0.8778169751167297 2023-01-21 09:17:44.447405: step: 1828/526, loss: 0.25798356533050537 2023-01-21 09:17:45.561687: step: 1832/526, loss: 0.20984306931495667 2023-01-21 09:17:46.704523: step: 1836/526, loss: 0.674612820148468 2023-01-21 09:17:47.799265: step: 1840/526, loss: 0.35536304116249084 2023-01-21 09:17:48.945255: step: 1844/526, loss: 0.6872149705886841 2023-01-21 09:17:50.095086: step: 1848/526, loss: 0.2538905143737793 2023-01-21 09:17:51.231615: step: 1852/526, loss: 0.7141143083572388 2023-01-21 09:17:52.359675: step: 1856/526, loss: 0.04651389271020889 2023-01-21 09:17:53.477443: step: 1860/526, loss: 0.13599276542663574 2023-01-21 09:17:54.585627: step: 1864/526, loss: 0.1452498883008957 2023-01-21 09:17:55.740543: step: 1868/526, loss: 0.17661356925964355 2023-01-21 09:17:56.868179: step: 1872/526, loss: 0.1657661348581314 2023-01-21 09:17:58.015580: step: 1876/526, loss: 0.10716857761144638 2023-01-21 09:17:59.183572: step: 1880/526, loss: 0.20729699730873108 2023-01-21 09:18:00.350135: step: 1884/526, loss: 1.8192706108093262 2023-01-21 09:18:01.502686: step: 1888/526, loss: 0.13287487626075745 2023-01-21 09:18:02.661746: step: 1892/526, loss: 0.1459612399339676 2023-01-21 09:18:03.784607: step: 1896/526, loss: 0.19155669212341309 2023-01-21 09:18:04.911192: step: 1900/526, loss: 0.03620195388793945 2023-01-21 09:18:06.033188: step: 1904/526, loss: 0.10412587970495224 2023-01-21 09:18:07.152223: step: 1908/526, loss: 0.12637433409690857 2023-01-21 09:18:08.291502: step: 1912/526, loss: 0.3685949444770813 2023-01-21 09:18:09.447365: step: 1916/526, loss: 0.43134477734565735 2023-01-21 09:18:10.568999: step: 1920/526, loss: 0.022488880902528763 2023-01-21 09:18:11.706176: step: 1924/526, loss: 0.6904934048652649 2023-01-21 09:18:12.907780: step: 1928/526, loss: 1.1879143714904785 2023-01-21 09:18:14.021495: step: 1932/526, loss: 0.13188600540161133 2023-01-21 09:18:15.161672: step: 1936/526, loss: 1.479654312133789 2023-01-21 09:18:16.285452: step: 1940/526, loss: 0.07299375534057617 2023-01-21 09:18:17.408302: step: 1944/526, loss: 0.23859329521656036 2023-01-21 09:18:18.536777: step: 1948/526, loss: 0.289031982421875 2023-01-21 09:18:19.696626: step: 1952/526, loss: 0.35942333936691284 2023-01-21 09:18:20.829793: step: 1956/526, loss: 0.5721483826637268 2023-01-21 09:18:21.937247: step: 1960/526, loss: 0.09510289132595062 2023-01-21 09:18:23.086471: step: 1964/526, loss: 0.19095487892627716 2023-01-21 09:18:24.229513: step: 1968/526, loss: 0.24901218712329865 2023-01-21 09:18:25.367496: step: 1972/526, loss: 0.6099697351455688 2023-01-21 09:18:26.503301: step: 1976/526, loss: 0.1544574797153473 2023-01-21 09:18:27.619298: step: 1980/526, loss: 0.25021684169769287 2023-01-21 09:18:28.765175: step: 1984/526, loss: 0.09940137714147568 2023-01-21 09:18:29.930097: step: 1988/526, loss: 0.11141081154346466 2023-01-21 09:18:31.063552: step: 1992/526, loss: 0.12098207324743271 2023-01-21 09:18:32.204302: step: 1996/526, loss: 0.037741970270872116 2023-01-21 09:18:33.361650: step: 2000/526, loss: 0.12642598152160645 2023-01-21 09:18:34.501819: step: 2004/526, loss: 0.14484553039073944 2023-01-21 09:18:35.615840: step: 2008/526, loss: 0.54514479637146 2023-01-21 09:18:36.717156: step: 2012/526, loss: 0.4738819897174835 2023-01-21 09:18:37.867562: step: 2016/526, loss: 0.10210972279310226 2023-01-21 09:18:39.009247: step: 2020/526, loss: 0.22374558448791504 2023-01-21 09:18:40.113395: step: 2024/526, loss: 0.13703079521656036 2023-01-21 09:18:41.262835: step: 2028/526, loss: 1.143784523010254 2023-01-21 09:18:42.376621: step: 2032/526, loss: 0.22043628990650177 2023-01-21 09:18:43.494822: step: 2036/526, loss: 0.7103164792060852 2023-01-21 09:18:44.655605: step: 2040/526, loss: 0.32267388701438904 2023-01-21 09:18:45.801334: step: 2044/526, loss: 2.1595511436462402 2023-01-21 09:18:46.926499: step: 2048/526, loss: 0.07906289398670197 2023-01-21 09:18:48.082289: step: 2052/526, loss: 0.4950440526008606 2023-01-21 09:18:49.202241: step: 2056/526, loss: 0.27676939964294434 2023-01-21 09:18:50.330618: step: 2060/526, loss: 0.8087915778160095 2023-01-21 09:18:51.464858: step: 2064/526, loss: 0.09626217186450958 2023-01-21 09:18:52.621393: step: 2068/526, loss: 0.15070191025733948 2023-01-21 09:18:53.773181: step: 2072/526, loss: 0.143851175904274 2023-01-21 09:18:54.895933: step: 2076/526, loss: 0.10410786420106888 2023-01-21 09:18:56.040965: step: 2080/526, loss: 0.02433171309530735 2023-01-21 09:18:57.213245: step: 2084/526, loss: 0.23612503707408905 2023-01-21 09:18:58.366004: step: 2088/526, loss: 0.188856840133667 2023-01-21 09:18:59.508412: step: 2092/526, loss: 0.18363535404205322 2023-01-21 09:19:00.649176: step: 2096/526, loss: 0.712185800075531 2023-01-21 09:19:01.785112: step: 2100/526, loss: 0.1364772915840149 2023-01-21 09:19:02.918821: step: 2104/526, loss: 0.19660745561122894 ================================================== Loss: 0.456 -------------------- Dev: {'event': {'p': 0.587251828631139, 'r': 0.748335552596538, 'f1': 0.6580796252927401}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Test: {'event': {'p': 0.5717155484130173, 'r': 0.7426931106471816, 'f1': 0.6460839954597049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Chinese: {'event': {'p': 0.5529411764705883, 'r': 0.8703703703703703, 'f1': 0.6762589928057553}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Korean: {'event': {'p': 0.6590909090909091, 'r': 0.4603174603174603, 'f1': 0.5420560747663551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Russian: {'event': {'p': 0.47368421052631576, 'r': 0.5, 'f1': 0.4864864864864865}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Korean: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Korean: {'event': {'p': 0.7575757575757576, 'r': 0.3968253968253968, 'f1': 0.5208333333333334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Russian: {'event': {'p': 0.587251828631139, 'r': 0.748335552596538, 'f1': 0.6580796252927401}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Eng Test for Russian: {'event': {'p': 0.5717155484130173, 'r': 0.7426931106471816, 'f1': 0.6460839954597049}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} Sample Russian: {'event': {'p': 0.47368421052631576, 'r': 0.5, 'f1': 0.4864864864864865}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 3} ****************************** Epoch: 4 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:19:54.874640: step: 4/526, loss: 0.4162517488002777 2023-01-21 09:19:56.028350: step: 8/526, loss: 0.16281470656394958 2023-01-21 09:19:57.140129: step: 12/526, loss: 0.09604306519031525 2023-01-21 09:19:58.271831: step: 16/526, loss: 0.12089300155639648 2023-01-21 09:19:59.413319: step: 20/526, loss: 0.7449560165405273 2023-01-21 09:20:00.524818: step: 24/526, loss: 0.07508730888366699 2023-01-21 09:20:01.657617: step: 28/526, loss: 0.38911762833595276 2023-01-21 09:20:02.796482: step: 32/526, loss: 0.12695255875587463 2023-01-21 09:20:03.938504: step: 36/526, loss: 0.27916955947875977 2023-01-21 09:20:05.084754: step: 40/526, loss: 0.07797583937644958 2023-01-21 09:20:06.205601: step: 44/526, loss: 0.1345663070678711 2023-01-21 09:20:07.335485: step: 48/526, loss: 0.8785447478294373 2023-01-21 09:20:08.468135: step: 52/526, loss: 0.023710155859589577 2023-01-21 09:20:09.623299: step: 56/526, loss: 0.18689842522144318 2023-01-21 09:20:10.750714: step: 60/526, loss: 0.19478636980056763 2023-01-21 09:20:11.880390: step: 64/526, loss: 0.06995487213134766 2023-01-21 09:20:12.998344: step: 68/526, loss: 0.9890401363372803 2023-01-21 09:20:14.142568: step: 72/526, loss: 0.09284988045692444 2023-01-21 09:20:15.263684: step: 76/526, loss: 0.2822338938713074 2023-01-21 09:20:16.371662: step: 80/526, loss: 0.01050415076315403 2023-01-21 09:20:17.485718: step: 84/526, loss: 0.18396396934986115 2023-01-21 09:20:18.619042: step: 88/526, loss: 0.8235489130020142 2023-01-21 09:20:19.785288: step: 92/526, loss: 0.0786537230014801 2023-01-21 09:20:20.944878: step: 96/526, loss: 0.12528057396411896 2023-01-21 09:20:22.071774: step: 100/526, loss: 0.22777271270751953 2023-01-21 09:20:23.235582: step: 104/526, loss: 0.3339591920375824 2023-01-21 09:20:24.386152: step: 108/526, loss: 0.13051939010620117 2023-01-21 09:20:25.514057: step: 112/526, loss: 0.13963966071605682 2023-01-21 09:20:26.643553: step: 116/526, loss: 0.13987894356250763 2023-01-21 09:20:27.775799: step: 120/526, loss: 0.25016745924949646 2023-01-21 09:20:28.923243: step: 124/526, loss: 0.7342115640640259 2023-01-21 09:20:30.030546: step: 128/526, loss: 0.020869970321655273 2023-01-21 09:20:31.155666: step: 132/526, loss: 0.1124420166015625 2023-01-21 09:20:32.292167: step: 136/526, loss: 0.3675425350666046 2023-01-21 09:20:33.422465: step: 140/526, loss: 0.8147329092025757 2023-01-21 09:20:34.539886: step: 144/526, loss: 0.1429820954799652 2023-01-21 09:20:35.689064: step: 148/526, loss: 0.09115403145551682 2023-01-21 09:20:36.805792: step: 152/526, loss: 0.708909809589386 2023-01-21 09:20:37.952782: step: 156/526, loss: 0.05897693708539009 2023-01-21 09:20:39.095744: step: 160/526, loss: 0.2730226516723633 2023-01-21 09:20:40.199864: step: 164/526, loss: 0.12081471085548401 2023-01-21 09:20:41.353943: step: 168/526, loss: 0.04553051292896271 2023-01-21 09:20:42.509138: step: 172/526, loss: 0.5359403491020203 2023-01-21 09:20:43.630738: step: 176/526, loss: 0.5294515490531921 2023-01-21 09:20:44.791957: step: 180/526, loss: 0.28645047545433044 2023-01-21 09:20:45.936157: step: 184/526, loss: 0.4109261631965637 2023-01-21 09:20:47.083077: step: 188/526, loss: 0.08603839576244354 2023-01-21 09:20:48.218262: step: 192/526, loss: 0.1760941594839096 2023-01-21 09:20:49.371625: step: 196/526, loss: 0.29432833194732666 2023-01-21 09:20:50.500618: step: 200/526, loss: 0.14655523002147675 2023-01-21 09:20:51.652043: step: 204/526, loss: 0.048816584050655365 2023-01-21 09:20:52.788418: step: 208/526, loss: 0.1075296401977539 2023-01-21 09:20:53.932604: step: 212/526, loss: 0.6211971640586853 2023-01-21 09:20:55.064993: step: 216/526, loss: 0.430537611246109 2023-01-21 09:20:56.205237: step: 220/526, loss: 0.16335105895996094 2023-01-21 09:20:57.356656: step: 224/526, loss: 0.13757658004760742 2023-01-21 09:20:58.521580: step: 228/526, loss: 0.10219473391771317 2023-01-21 09:20:59.674468: step: 232/526, loss: 0.47321754693984985 2023-01-21 09:21:00.830186: step: 236/526, loss: 0.09550885856151581 2023-01-21 09:21:02.004795: step: 240/526, loss: 0.1250135451555252 2023-01-21 09:21:03.108197: step: 244/526, loss: 0.14286527037620544 2023-01-21 09:21:04.260644: step: 248/526, loss: 0.0886419266462326 2023-01-21 09:21:05.393003: step: 252/526, loss: 0.4233490228652954 2023-01-21 09:21:06.514711: step: 256/526, loss: 0.08562731742858887 2023-01-21 09:21:07.617673: step: 260/526, loss: 0.032315947115421295 2023-01-21 09:21:08.712608: step: 264/526, loss: 0.7196832299232483 2023-01-21 09:21:09.830282: step: 268/526, loss: 0.3318479359149933 2023-01-21 09:21:10.951526: step: 272/526, loss: 0.016775943338871002 2023-01-21 09:21:12.093272: step: 276/526, loss: 0.43076688051223755 2023-01-21 09:21:13.252928: step: 280/526, loss: 0.15205231308937073 2023-01-21 09:21:14.364064: step: 284/526, loss: 0.2976920008659363 2023-01-21 09:21:15.510552: step: 288/526, loss: 0.12062001973390579 2023-01-21 09:21:16.638911: step: 292/526, loss: 0.6778050065040588 2023-01-21 09:21:17.789405: step: 296/526, loss: 0.24091148376464844 2023-01-21 09:21:18.924330: step: 300/526, loss: 0.8197460174560547 2023-01-21 09:21:20.046742: step: 304/526, loss: 0.14999523758888245 2023-01-21 09:21:21.215770: step: 308/526, loss: 0.1704537719488144 2023-01-21 09:21:22.362721: step: 312/526, loss: 1.2849609851837158 2023-01-21 09:21:23.539744: step: 316/526, loss: 0.09026508033275604 2023-01-21 09:21:24.664722: step: 320/526, loss: 0.05670461431145668 2023-01-21 09:21:25.804674: step: 324/526, loss: 0.01755533181130886 2023-01-21 09:21:26.945044: step: 328/526, loss: 0.44096651673316956 2023-01-21 09:21:28.071093: step: 332/526, loss: 0.24535000324249268 2023-01-21 09:21:29.223137: step: 336/526, loss: 0.10419130325317383 2023-01-21 09:21:30.381981: step: 340/526, loss: 0.3023788630962372 2023-01-21 09:21:31.531358: step: 344/526, loss: 0.18359366059303284 2023-01-21 09:21:32.678999: step: 348/526, loss: 0.09482508152723312 2023-01-21 09:21:33.806871: step: 352/526, loss: 0.4031192660331726 2023-01-21 09:21:34.944444: step: 356/526, loss: 0.19881489872932434 2023-01-21 09:21:36.106134: step: 360/526, loss: 0.6746804714202881 2023-01-21 09:21:37.231547: step: 364/526, loss: 0.43863701820373535 2023-01-21 09:21:38.396459: step: 368/526, loss: 0.11959227919578552 2023-01-21 09:21:39.533234: step: 372/526, loss: 0.10329818725585938 2023-01-21 09:21:40.649436: step: 376/526, loss: 0.08187904208898544 2023-01-21 09:21:41.803025: step: 380/526, loss: 0.2239125370979309 2023-01-21 09:21:42.917234: step: 384/526, loss: 0.41037869453430176 2023-01-21 09:21:44.054945: step: 388/526, loss: 0.11743126064538956 2023-01-21 09:21:45.223959: step: 392/526, loss: 0.24530839920043945 2023-01-21 09:21:46.354873: step: 396/526, loss: 0.5493795275688171 2023-01-21 09:21:47.483325: step: 400/526, loss: 0.07184629142284393 2023-01-21 09:21:48.606039: step: 404/526, loss: 0.3614213466644287 2023-01-21 09:21:49.766182: step: 408/526, loss: 0.23081770539283752 2023-01-21 09:21:50.900771: step: 412/526, loss: 0.02835226058959961 2023-01-21 09:21:52.041381: step: 416/526, loss: 0.37867841124534607 2023-01-21 09:21:53.188903: step: 420/526, loss: 0.2118498831987381 2023-01-21 09:21:54.346485: step: 424/526, loss: 0.5382680892944336 2023-01-21 09:21:55.479221: step: 428/526, loss: 0.8340715169906616 2023-01-21 09:21:56.619968: step: 432/526, loss: 0.1693919599056244 2023-01-21 09:21:57.768679: step: 436/526, loss: 0.22669143974781036 2023-01-21 09:21:58.905844: step: 440/526, loss: 0.17380136251449585 2023-01-21 09:22:00.062899: step: 444/526, loss: 0.09809380024671555 2023-01-21 09:22:01.210809: step: 448/526, loss: 0.5440462231636047 2023-01-21 09:22:02.328438: step: 452/526, loss: 0.06541042029857635 2023-01-21 09:22:03.469205: step: 456/526, loss: 0.2109546661376953 2023-01-21 09:22:04.576468: step: 460/526, loss: 0.09080924838781357 2023-01-21 09:22:05.713211: step: 464/526, loss: 0.07684745639562607 2023-01-21 09:22:06.848448: step: 468/526, loss: 0.042144108563661575 2023-01-21 09:22:07.961676: step: 472/526, loss: 0.11273002624511719 2023-01-21 09:22:09.134126: step: 476/526, loss: 0.7454760074615479 2023-01-21 09:22:10.269164: step: 480/526, loss: 0.08959593623876572 2023-01-21 09:22:11.408840: step: 484/526, loss: 0.14918716251850128 2023-01-21 09:22:12.559631: step: 488/526, loss: 0.3742007613182068 2023-01-21 09:22:13.693898: step: 492/526, loss: 0.037870120257139206 2023-01-21 09:22:14.835870: step: 496/526, loss: 0.08254127204418182 2023-01-21 09:22:15.964555: step: 500/526, loss: 0.1438557207584381 2023-01-21 09:22:17.126019: step: 504/526, loss: 0.1511041224002838 2023-01-21 09:22:18.297456: step: 508/526, loss: 0.06995344161987305 2023-01-21 09:22:19.405850: step: 512/526, loss: 0.06293439865112305 2023-01-21 09:22:20.559070: step: 516/526, loss: 0.1041233092546463 2023-01-21 09:22:21.719472: step: 520/526, loss: 0.3784911036491394 2023-01-21 09:22:22.860580: step: 524/526, loss: 0.03926382213830948 2023-01-21 09:22:23.984360: step: 528/526, loss: 0.019102763384580612 2023-01-21 09:22:25.122300: step: 532/526, loss: 0.09351787716150284 2023-01-21 09:22:26.255705: step: 536/526, loss: 0.5679179430007935 2023-01-21 09:22:27.367478: step: 540/526, loss: 0.14807163178920746 2023-01-21 09:22:28.492870: step: 544/526, loss: 0.24053727090358734 2023-01-21 09:22:29.587713: step: 548/526, loss: 0.05138111114501953 2023-01-21 09:22:30.706493: step: 552/526, loss: 0.23480865359306335 2023-01-21 09:22:31.850701: step: 556/526, loss: 0.12402839958667755 2023-01-21 09:22:32.982464: step: 560/526, loss: 0.07263145595788956 2023-01-21 09:22:34.115334: step: 564/526, loss: 0.13666324317455292 2023-01-21 09:22:35.245098: step: 568/526, loss: 0.2843009829521179 2023-01-21 09:22:36.405413: step: 572/526, loss: 0.18428076803684235 2023-01-21 09:22:37.539258: step: 576/526, loss: 0.09730225056409836 2023-01-21 09:22:38.703507: step: 580/526, loss: 0.04092860221862793 2023-01-21 09:22:39.849307: step: 584/526, loss: 0.11186066269874573 2023-01-21 09:22:41.020657: step: 588/526, loss: 0.9921743869781494 2023-01-21 09:22:42.184599: step: 592/526, loss: 0.056658077985048294 2023-01-21 09:22:43.348408: step: 596/526, loss: 0.17836570739746094 2023-01-21 09:22:44.484371: step: 600/526, loss: 0.033994387835264206 2023-01-21 09:22:45.598768: step: 604/526, loss: 1.318795084953308 2023-01-21 09:22:46.724335: step: 608/526, loss: 0.20577049255371094 2023-01-21 09:22:47.858816: step: 612/526, loss: 0.10561619699001312 2023-01-21 09:22:48.982152: step: 616/526, loss: 0.10459490120410919 2023-01-21 09:22:50.114985: step: 620/526, loss: 0.076439768075943 2023-01-21 09:22:51.268145: step: 624/526, loss: 0.10842490196228027 2023-01-21 09:22:52.409545: step: 628/526, loss: 0.10341091454029083 2023-01-21 09:22:53.545657: step: 632/526, loss: 0.1505311131477356 2023-01-21 09:22:54.685417: step: 636/526, loss: 0.08439254760742188 2023-01-21 09:22:55.823112: step: 640/526, loss: 0.1255919337272644 2023-01-21 09:22:56.949130: step: 644/526, loss: 0.4810018539428711 2023-01-21 09:22:58.071347: step: 648/526, loss: 0.09029150009155273 2023-01-21 09:22:59.195577: step: 652/526, loss: 1.07391357421875 2023-01-21 09:23:00.348530: step: 656/526, loss: 0.11303277313709259 2023-01-21 09:23:01.466757: step: 660/526, loss: 0.10102319717407227 2023-01-21 09:23:02.593916: step: 664/526, loss: 0.03169383853673935 2023-01-21 09:23:03.715321: step: 668/526, loss: 0.12495499104261398 2023-01-21 09:23:04.862548: step: 672/526, loss: 0.722877562046051 2023-01-21 09:23:05.983075: step: 676/526, loss: 0.057461075484752655 2023-01-21 09:23:07.127832: step: 680/526, loss: 0.6774193048477173 2023-01-21 09:23:08.285019: step: 684/526, loss: 0.130553737282753 2023-01-21 09:23:09.417970: step: 688/526, loss: 0.5497356057167053 2023-01-21 09:23:10.596328: step: 692/526, loss: 0.34561729431152344 2023-01-21 09:23:11.756912: step: 696/526, loss: 0.8197286128997803 2023-01-21 09:23:12.915392: step: 700/526, loss: 0.07111091911792755 2023-01-21 09:23:14.057692: step: 704/526, loss: 0.23404255509376526 2023-01-21 09:23:15.159661: step: 708/526, loss: 0.17759867012500763 2023-01-21 09:23:16.314380: step: 712/526, loss: 0.2189093679189682 2023-01-21 09:23:17.459455: step: 716/526, loss: 0.17411461472511292 2023-01-21 09:23:18.589325: step: 720/526, loss: 0.1080346405506134 2023-01-21 09:23:19.717627: step: 724/526, loss: 0.08979358524084091 2023-01-21 09:23:20.858689: step: 728/526, loss: 0.09780425578355789 2023-01-21 09:23:21.978590: step: 732/526, loss: 0.12218132615089417 2023-01-21 09:23:23.090253: step: 736/526, loss: 0.08891335129737854 2023-01-21 09:23:24.245717: step: 740/526, loss: 0.1156400665640831 2023-01-21 09:23:25.386987: step: 744/526, loss: 0.1883530616760254 2023-01-21 09:23:26.509352: step: 748/526, loss: 1.6280313730239868 2023-01-21 09:23:27.641122: step: 752/526, loss: 0.17897024750709534 2023-01-21 09:23:28.773806: step: 756/526, loss: 0.3709869384765625 2023-01-21 09:23:29.913776: step: 760/526, loss: 0.06960678100585938 2023-01-21 09:23:31.063343: step: 764/526, loss: 0.22618350386619568 2023-01-21 09:23:32.231946: step: 768/526, loss: 1.4186476469039917 2023-01-21 09:23:33.348997: step: 772/526, loss: 0.1943950653076172 2023-01-21 09:23:34.468940: step: 776/526, loss: 0.11608649045228958 2023-01-21 09:23:35.613935: step: 780/526, loss: 0.1909475326538086 2023-01-21 09:23:36.747666: step: 784/526, loss: 0.19988222420215607 2023-01-21 09:23:37.859700: step: 788/526, loss: 0.11071443557739258 2023-01-21 09:23:38.985250: step: 792/526, loss: 0.5383886694908142 2023-01-21 09:23:40.134213: step: 796/526, loss: 0.16698570549488068 2023-01-21 09:23:41.287510: step: 800/526, loss: 0.009955167770385742 2023-01-21 09:23:42.411385: step: 804/526, loss: 0.08415904641151428 2023-01-21 09:23:43.578358: step: 808/526, loss: 0.48933279514312744 2023-01-21 09:23:44.703703: step: 812/526, loss: 0.9976007342338562 2023-01-21 09:23:45.839459: step: 816/526, loss: 0.18563146889209747 2023-01-21 09:23:46.950105: step: 820/526, loss: 6.025500774383545 2023-01-21 09:23:48.103782: step: 824/526, loss: 0.28811654448509216 2023-01-21 09:23:49.231314: step: 828/526, loss: 0.11262817680835724 2023-01-21 09:23:50.356278: step: 832/526, loss: 0.08733835071325302 2023-01-21 09:23:51.480205: step: 836/526, loss: 0.500493586063385 2023-01-21 09:23:52.579843: step: 840/526, loss: 0.15617340803146362 2023-01-21 09:23:53.711465: step: 844/526, loss: 0.16936083137989044 2023-01-21 09:23:54.851597: step: 848/526, loss: 0.09375190734863281 2023-01-21 09:23:55.946032: step: 852/526, loss: 0.7823423743247986 2023-01-21 09:23:57.063284: step: 856/526, loss: 0.14134198427200317 2023-01-21 09:23:58.191715: step: 860/526, loss: 0.24290743470191956 2023-01-21 09:23:59.327840: step: 864/526, loss: 0.18269950151443481 2023-01-21 09:24:00.442494: step: 868/526, loss: 0.11488529294729233 2023-01-21 09:24:01.559536: step: 872/526, loss: 0.6648132801055908 2023-01-21 09:24:02.692064: step: 876/526, loss: 0.1677156388759613 2023-01-21 09:24:03.841817: step: 880/526, loss: 0.27670565247535706 2023-01-21 09:24:04.959036: step: 884/526, loss: 0.7141363620758057 2023-01-21 09:24:06.096811: step: 888/526, loss: 5.9438157081604 2023-01-21 09:24:07.231822: step: 892/526, loss: 0.2637576460838318 2023-01-21 09:24:08.385504: step: 896/526, loss: 0.15523844957351685 2023-01-21 09:24:09.544155: step: 900/526, loss: 0.6245636343955994 2023-01-21 09:24:10.664710: step: 904/526, loss: 0.12494106590747833 2023-01-21 09:24:11.797715: step: 908/526, loss: 0.21034583449363708 2023-01-21 09:24:12.950003: step: 912/526, loss: 0.9144895672798157 2023-01-21 09:24:14.101131: step: 916/526, loss: 0.6842329502105713 2023-01-21 09:24:15.218799: step: 920/526, loss: 0.18424977362155914 2023-01-21 09:24:16.373101: step: 924/526, loss: 2.065530300140381 2023-01-21 09:24:17.565144: step: 928/526, loss: 0.09701290726661682 2023-01-21 09:24:18.710105: step: 932/526, loss: 0.125346377491951 2023-01-21 09:24:19.809947: step: 936/526, loss: 0.11811709403991699 2023-01-21 09:24:20.932223: step: 940/526, loss: 0.26700520515441895 2023-01-21 09:24:22.058068: step: 944/526, loss: 0.14254026114940643 2023-01-21 09:24:23.183890: step: 948/526, loss: 0.06474772095680237 2023-01-21 09:24:24.321999: step: 952/526, loss: 0.18564969301223755 2023-01-21 09:24:25.430029: step: 956/526, loss: 0.15506859123706818 2023-01-21 09:24:26.576561: step: 960/526, loss: 0.22779254615306854 2023-01-21 09:24:27.712712: step: 964/526, loss: 0.26531124114990234 2023-01-21 09:24:28.821998: step: 968/526, loss: 0.22185811400413513 2023-01-21 09:24:29.971781: step: 972/526, loss: 0.05372581630945206 2023-01-21 09:24:31.074341: step: 976/526, loss: 0.22016258537769318 2023-01-21 09:24:32.199465: step: 980/526, loss: 0.7096502780914307 2023-01-21 09:24:33.350165: step: 984/526, loss: 0.9157988429069519 2023-01-21 09:24:34.531602: step: 988/526, loss: 0.27391892671585083 2023-01-21 09:24:35.665405: step: 992/526, loss: 0.6831054091453552 2023-01-21 09:24:36.794192: step: 996/526, loss: 0.22025933861732483 2023-01-21 09:24:37.942808: step: 1000/526, loss: 0.2376731038093567 2023-01-21 09:24:39.064587: step: 1004/526, loss: 0.21200771629810333 2023-01-21 09:24:40.190104: step: 1008/526, loss: 0.16856956481933594 2023-01-21 09:24:41.339476: step: 1012/526, loss: 0.15849237143993378 2023-01-21 09:24:42.482701: step: 1016/526, loss: 6.842772006988525 2023-01-21 09:24:43.615412: step: 1020/526, loss: 0.09412240982055664 2023-01-21 09:24:44.763543: step: 1024/526, loss: 0.20343151688575745 2023-01-21 09:24:45.921565: step: 1028/526, loss: 0.5002622604370117 2023-01-21 09:24:47.055102: step: 1032/526, loss: 0.18950414657592773 2023-01-21 09:24:48.185751: step: 1036/526, loss: 0.20053061842918396 2023-01-21 09:24:49.289247: step: 1040/526, loss: 0.13936929404735565 2023-01-21 09:24:50.414388: step: 1044/526, loss: 0.18831472098827362 2023-01-21 09:24:51.572628: step: 1048/526, loss: 0.19123177230358124 2023-01-21 09:24:52.682314: step: 1052/526, loss: 0.05880117416381836 2023-01-21 09:24:53.822107: step: 1056/526, loss: 0.6875283122062683 2023-01-21 09:24:54.951378: step: 1060/526, loss: 0.456093966960907 2023-01-21 09:24:56.068797: step: 1064/526, loss: 0.685244083404541 2023-01-21 09:24:57.213660: step: 1068/526, loss: 0.5029582381248474 2023-01-21 09:24:58.403014: step: 1072/526, loss: 0.0618559867143631 2023-01-21 09:24:59.547071: step: 1076/526, loss: 0.15304584801197052 2023-01-21 09:25:00.699426: step: 1080/526, loss: 0.0708322525024414 2023-01-21 09:25:01.847243: step: 1084/526, loss: 0.8185214996337891 2023-01-21 09:25:02.981534: step: 1088/526, loss: 0.0909125804901123 2023-01-21 09:25:04.122063: step: 1092/526, loss: 0.7081989645957947 2023-01-21 09:25:05.265575: step: 1096/526, loss: 0.32577913999557495 2023-01-21 09:25:06.365436: step: 1100/526, loss: 0.13986030220985413 2023-01-21 09:25:07.525132: step: 1104/526, loss: 0.39668768644332886 2023-01-21 09:25:08.653011: step: 1108/526, loss: 0.3519403338432312 2023-01-21 09:25:09.788631: step: 1112/526, loss: 0.5454993844032288 2023-01-21 09:25:10.919956: step: 1116/526, loss: 0.12598438560962677 2023-01-21 09:25:12.035591: step: 1120/526, loss: 0.08856096118688583 2023-01-21 09:25:13.169356: step: 1124/526, loss: 0.12020225822925568 2023-01-21 09:25:14.329285: step: 1128/526, loss: 0.13518276810646057 2023-01-21 09:25:15.451528: step: 1132/526, loss: 0.12709268927574158 2023-01-21 09:25:16.563259: step: 1136/526, loss: 0.02418675646185875 2023-01-21 09:25:17.683934: step: 1140/526, loss: 0.4214306175708771 2023-01-21 09:25:18.806667: step: 1144/526, loss: 0.3428903818130493 2023-01-21 09:25:19.960363: step: 1148/526, loss: 0.17736349999904633 2023-01-21 09:25:21.065629: step: 1152/526, loss: 0.32006072998046875 2023-01-21 09:25:22.198446: step: 1156/526, loss: 0.717534601688385 2023-01-21 09:25:23.340311: step: 1160/526, loss: 0.43570470809936523 2023-01-21 09:25:24.470223: step: 1164/526, loss: 1.1295288801193237 2023-01-21 09:25:25.592190: step: 1168/526, loss: 0.07961931824684143 2023-01-21 09:25:26.735692: step: 1172/526, loss: 0.43556660413742065 2023-01-21 09:25:27.876068: step: 1176/526, loss: 0.26903438568115234 2023-01-21 09:25:29.003079: step: 1180/526, loss: 0.05784440040588379 2023-01-21 09:25:30.135600: step: 1184/526, loss: 0.28725466132164 2023-01-21 09:25:31.272859: step: 1188/526, loss: 0.07860822975635529 2023-01-21 09:25:32.394795: step: 1192/526, loss: 1.069618582725525 2023-01-21 09:25:33.540641: step: 1196/526, loss: 0.09319524466991425 2023-01-21 09:25:34.692165: step: 1200/526, loss: 0.9479278922080994 2023-01-21 09:25:35.824496: step: 1204/526, loss: 0.7471533417701721 2023-01-21 09:25:36.973377: step: 1208/526, loss: 0.5247141122817993 2023-01-21 09:25:38.070199: step: 1212/526, loss: 0.0443631187081337 2023-01-21 09:25:39.212865: step: 1216/526, loss: 0.053660012781620026 2023-01-21 09:25:40.318346: step: 1220/526, loss: 0.7454779148101807 2023-01-21 09:25:41.473391: step: 1224/526, loss: 0.13610154390335083 2023-01-21 09:25:42.595885: step: 1228/526, loss: 0.2525690197944641 2023-01-21 09:25:43.712503: step: 1232/526, loss: 0.8536794185638428 2023-01-21 09:25:44.855529: step: 1236/526, loss: 0.06363983452320099 2023-01-21 09:25:45.993359: step: 1240/526, loss: 0.29542914032936096 2023-01-21 09:25:47.142937: step: 1244/526, loss: 0.4992457926273346 2023-01-21 09:25:48.274997: step: 1248/526, loss: 0.44941824674606323 2023-01-21 09:25:49.400994: step: 1252/526, loss: 0.20342311263084412 2023-01-21 09:25:50.535066: step: 1256/526, loss: 0.0764416754245758 2023-01-21 09:25:51.673237: step: 1260/526, loss: 0.03515281900763512 2023-01-21 09:25:52.793407: step: 1264/526, loss: 0.1457241028547287 2023-01-21 09:25:53.891830: step: 1268/526, loss: 0.07068701088428497 2023-01-21 09:25:55.015128: step: 1272/526, loss: 0.7477499842643738 2023-01-21 09:25:56.140540: step: 1276/526, loss: 0.23456035554409027 2023-01-21 09:25:57.232996: step: 1280/526, loss: 0.19361944496631622 2023-01-21 09:25:58.346392: step: 1284/526, loss: 0.11824379116296768 2023-01-21 09:25:59.467549: step: 1288/526, loss: 0.2297314703464508 2023-01-21 09:26:00.593180: step: 1292/526, loss: 0.6895031929016113 2023-01-21 09:26:01.707118: step: 1296/526, loss: 0.05504312366247177 2023-01-21 09:26:02.834779: step: 1300/526, loss: 0.6723909974098206 2023-01-21 09:26:03.942692: step: 1304/526, loss: 0.2251339852809906 2023-01-21 09:26:05.069041: step: 1308/526, loss: 0.2795290946960449 2023-01-21 09:26:06.214016: step: 1312/526, loss: 0.023305200040340424 2023-01-21 09:26:07.335924: step: 1316/526, loss: 0.032586004585027695 2023-01-21 09:26:08.420681: step: 1320/526, loss: 0.6801646947860718 2023-01-21 09:26:09.551209: step: 1324/526, loss: 0.05224008858203888 2023-01-21 09:26:10.708339: step: 1328/526, loss: 0.1410684585571289 2023-01-21 09:26:11.819370: step: 1332/526, loss: 0.19551940262317657 2023-01-21 09:26:12.951765: step: 1336/526, loss: 0.1493641436100006 2023-01-21 09:26:14.109490: step: 1340/526, loss: 0.1305527687072754 2023-01-21 09:26:15.222497: step: 1344/526, loss: 0.19192147254943848 2023-01-21 09:26:16.330421: step: 1348/526, loss: 0.056697942316532135 2023-01-21 09:26:17.466001: step: 1352/526, loss: 0.23513685166835785 2023-01-21 09:26:18.596737: step: 1356/526, loss: 0.15572190284729004 2023-01-21 09:26:19.726933: step: 1360/526, loss: 0.09732604026794434 2023-01-21 09:26:20.866318: step: 1364/526, loss: 0.30835264921188354 2023-01-21 09:26:21.996896: step: 1368/526, loss: 0.1821792721748352 2023-01-21 09:26:23.135541: step: 1372/526, loss: 0.15931901335716248 2023-01-21 09:26:24.280222: step: 1376/526, loss: 0.2861195504665375 2023-01-21 09:26:25.437491: step: 1380/526, loss: 0.9929283857345581 2023-01-21 09:26:26.565821: step: 1384/526, loss: 0.1610986739397049 2023-01-21 09:26:27.707740: step: 1388/526, loss: 0.47864967584609985 2023-01-21 09:26:28.833695: step: 1392/526, loss: 0.2566388249397278 2023-01-21 09:26:29.936017: step: 1396/526, loss: 0.07148456573486328 2023-01-21 09:26:31.078340: step: 1400/526, loss: 0.1619223654270172 2023-01-21 09:26:32.195524: step: 1404/526, loss: 0.06925344467163086 2023-01-21 09:26:33.311994: step: 1408/526, loss: 0.47106000781059265 2023-01-21 09:26:34.423855: step: 1412/526, loss: 1.3208248615264893 2023-01-21 09:26:35.577300: step: 1416/526, loss: 0.042852021753787994 2023-01-21 09:26:36.675723: step: 1420/526, loss: 1.056248664855957 2023-01-21 09:26:37.829715: step: 1424/526, loss: 0.7417402267456055 2023-01-21 09:26:38.967094: step: 1428/526, loss: 0.07049369812011719 2023-01-21 09:26:40.131411: step: 1432/526, loss: 0.3194744288921356 2023-01-21 09:26:41.256673: step: 1436/526, loss: 0.5398289561271667 2023-01-21 09:26:42.395461: step: 1440/526, loss: 0.2468746155500412 2023-01-21 09:26:43.557012: step: 1444/526, loss: 0.24635668098926544 2023-01-21 09:26:44.731422: step: 1448/526, loss: 2.023756265640259 2023-01-21 09:26:45.884656: step: 1452/526, loss: 0.4297069311141968 2023-01-21 09:26:47.003364: step: 1456/526, loss: 0.10008516162633896 2023-01-21 09:26:48.159277: step: 1460/526, loss: 0.35322466492652893 2023-01-21 09:26:49.275571: step: 1464/526, loss: 0.06808061897754669 2023-01-21 09:26:50.420351: step: 1468/526, loss: 0.05945263057947159 2023-01-21 09:26:51.563237: step: 1472/526, loss: 0.24721908569335938 2023-01-21 09:26:52.691433: step: 1476/526, loss: 0.7393164038658142 2023-01-21 09:26:53.821586: step: 1480/526, loss: 0.28931379318237305 2023-01-21 09:26:54.929198: step: 1484/526, loss: 0.2223377227783203 2023-01-21 09:26:56.098561: step: 1488/526, loss: 0.23443478345870972 2023-01-21 09:26:57.220046: step: 1492/526, loss: 0.15631341934204102 2023-01-21 09:26:58.344501: step: 1496/526, loss: 2.540395736694336 2023-01-21 09:26:59.442492: step: 1500/526, loss: 0.1442979872226715 2023-01-21 09:27:00.621036: step: 1504/526, loss: 0.12117157131433487 2023-01-21 09:27:01.750436: step: 1508/526, loss: 0.11972665786743164 2023-01-21 09:27:02.856668: step: 1512/526, loss: 0.10798332095146179 2023-01-21 09:27:03.973629: step: 1516/526, loss: 0.1232258751988411 2023-01-21 09:27:05.078081: step: 1520/526, loss: 0.10492291301488876 2023-01-21 09:27:06.229977: step: 1524/526, loss: 0.12372155487537384 2023-01-21 09:27:07.373390: step: 1528/526, loss: 0.7846032977104187 2023-01-21 09:27:08.518719: step: 1532/526, loss: 1.361919641494751 2023-01-21 09:27:09.669880: step: 1536/526, loss: 0.144123837351799 2023-01-21 09:27:10.795686: step: 1540/526, loss: 0.4321043789386749 2023-01-21 09:27:11.923844: step: 1544/526, loss: 0.062157824635505676 2023-01-21 09:27:13.015280: step: 1548/526, loss: 0.135029137134552 2023-01-21 09:27:14.158613: step: 1552/526, loss: 0.02161407470703125 2023-01-21 09:27:15.292634: step: 1556/526, loss: 0.16113395988941193 2023-01-21 09:27:16.453207: step: 1560/526, loss: 0.5720523595809937 2023-01-21 09:27:17.582329: step: 1564/526, loss: 0.5385809540748596 2023-01-21 09:27:18.740690: step: 1568/526, loss: 0.07579260319471359 2023-01-21 09:27:19.879247: step: 1572/526, loss: 0.09935131669044495 2023-01-21 09:27:21.001646: step: 1576/526, loss: 0.09231805801391602 2023-01-21 09:27:22.127823: step: 1580/526, loss: 0.11738023161888123 2023-01-21 09:27:23.249423: step: 1584/526, loss: 0.37796393036842346 2023-01-21 09:27:24.362855: step: 1588/526, loss: 0.5186008214950562 2023-01-21 09:27:25.475769: step: 1592/526, loss: 0.4013316035270691 2023-01-21 09:27:26.612274: step: 1596/526, loss: 0.09492354094982147 2023-01-21 09:27:27.759507: step: 1600/526, loss: 0.2895212173461914 2023-01-21 09:27:28.893444: step: 1604/526, loss: 0.11452098190784454 2023-01-21 09:27:30.057838: step: 1608/526, loss: 0.17952455580234528 2023-01-21 09:27:31.183067: step: 1612/526, loss: 0.020197726786136627 2023-01-21 09:27:32.314959: step: 1616/526, loss: 0.20291881263256073 2023-01-21 09:27:33.456147: step: 1620/526, loss: 0.19766837358474731 2023-01-21 09:27:34.556491: step: 1624/526, loss: 0.11001381278038025 2023-01-21 09:27:35.667449: step: 1628/526, loss: 0.5903832316398621 2023-01-21 09:27:36.799123: step: 1632/526, loss: 0.5519746541976929 2023-01-21 09:27:37.920477: step: 1636/526, loss: 0.34842386841773987 2023-01-21 09:27:39.067708: step: 1640/526, loss: 0.09857045114040375 2023-01-21 09:27:40.212171: step: 1644/526, loss: 0.8871277570724487 2023-01-21 09:27:41.328214: step: 1648/526, loss: 0.2080329954624176 2023-01-21 09:27:42.465127: step: 1652/526, loss: 0.3938787579536438 2023-01-21 09:27:43.618403: step: 1656/526, loss: 0.7692900896072388 2023-01-21 09:27:44.807738: step: 1660/526, loss: 0.0781501829624176 2023-01-21 09:27:45.981547: step: 1664/526, loss: 0.04093952476978302 2023-01-21 09:27:47.084908: step: 1668/526, loss: 0.045726679265499115 2023-01-21 09:27:48.212472: step: 1672/526, loss: 0.8742673397064209 2023-01-21 09:27:49.344086: step: 1676/526, loss: 1.3643262386322021 2023-01-21 09:27:50.502543: step: 1680/526, loss: 0.1408170759677887 2023-01-21 09:27:51.673507: step: 1684/526, loss: 0.4666202664375305 2023-01-21 09:27:52.792286: step: 1688/526, loss: 0.41231030225753784 2023-01-21 09:27:53.950627: step: 1692/526, loss: 0.8471347093582153 2023-01-21 09:27:55.103769: step: 1696/526, loss: 0.3264133930206299 2023-01-21 09:27:56.240901: step: 1700/526, loss: 0.47895461320877075 2023-01-21 09:27:57.345035: step: 1704/526, loss: 0.13776779174804688 2023-01-21 09:27:58.500930: step: 1708/526, loss: 0.3295432925224304 2023-01-21 09:27:59.648774: step: 1712/526, loss: 0.11990585178136826 2023-01-21 09:28:00.785162: step: 1716/526, loss: 0.6645368933677673 2023-01-21 09:28:01.894697: step: 1720/526, loss: 1.2137835025787354 2023-01-21 09:28:03.049259: step: 1724/526, loss: 0.28978538513183594 2023-01-21 09:28:04.173320: step: 1728/526, loss: 0.0360596664249897 2023-01-21 09:28:05.284071: step: 1732/526, loss: 0.05216164514422417 2023-01-21 09:28:06.398338: step: 1736/526, loss: 0.22510471940040588 2023-01-21 09:28:07.507578: step: 1740/526, loss: 0.5259994864463806 2023-01-21 09:28:08.640441: step: 1744/526, loss: 0.1807439923286438 2023-01-21 09:28:09.783628: step: 1748/526, loss: 0.08307859301567078 2023-01-21 09:28:10.911540: step: 1752/526, loss: 0.20417433977127075 2023-01-21 09:28:12.051264: step: 1756/526, loss: 0.7567815780639648 2023-01-21 09:28:13.198157: step: 1760/526, loss: 0.9521654844284058 2023-01-21 09:28:14.319266: step: 1764/526, loss: 0.17875318229198456 2023-01-21 09:28:15.442929: step: 1768/526, loss: 0.06773886829614639 2023-01-21 09:28:16.573245: step: 1772/526, loss: 0.33285924792289734 2023-01-21 09:28:17.719410: step: 1776/526, loss: 0.10147294402122498 2023-01-21 09:28:18.854005: step: 1780/526, loss: 0.960585355758667 2023-01-21 09:28:19.993560: step: 1784/526, loss: 0.10967588424682617 2023-01-21 09:28:21.123015: step: 1788/526, loss: 0.6423585414886475 2023-01-21 09:28:22.261310: step: 1792/526, loss: 0.6108390688896179 2023-01-21 09:28:23.396233: step: 1796/526, loss: 0.5428126454353333 2023-01-21 09:28:24.519977: step: 1800/526, loss: 0.015728093683719635 2023-01-21 09:28:25.638616: step: 1804/526, loss: 0.17113475501537323 2023-01-21 09:28:26.754046: step: 1808/526, loss: 0.23737744987010956 2023-01-21 09:28:27.880254: step: 1812/526, loss: 0.4769209027290344 2023-01-21 09:28:29.018922: step: 1816/526, loss: 0.08623366057872772 2023-01-21 09:28:30.169891: step: 1820/526, loss: 0.11496620625257492 2023-01-21 09:28:31.301900: step: 1824/526, loss: 0.30621927976608276 2023-01-21 09:28:32.443583: step: 1828/526, loss: 0.7393184900283813 2023-01-21 09:28:33.556188: step: 1832/526, loss: 0.0417332649230957 2023-01-21 09:28:34.679780: step: 1836/526, loss: 0.7059459686279297 2023-01-21 09:28:35.779484: step: 1840/526, loss: 0.6556960344314575 2023-01-21 09:28:36.959111: step: 1844/526, loss: 0.03392457962036133 2023-01-21 09:28:38.124577: step: 1848/526, loss: 0.06405811756849289 2023-01-21 09:28:39.246066: step: 1852/526, loss: 0.30063286423683167 2023-01-21 09:28:40.391732: step: 1856/526, loss: 0.31495171785354614 2023-01-21 09:28:41.509992: step: 1860/526, loss: 0.33719176054000854 2023-01-21 09:28:42.650297: step: 1864/526, loss: 0.6194790005683899 2023-01-21 09:28:43.785184: step: 1868/526, loss: 0.15346379578113556 2023-01-21 09:28:44.946866: step: 1872/526, loss: 0.3342325687408447 2023-01-21 09:28:46.088197: step: 1876/526, loss: 0.06054558977484703 2023-01-21 09:28:47.201973: step: 1880/526, loss: 0.1685403436422348 2023-01-21 09:28:48.343848: step: 1884/526, loss: 0.054334595799446106 2023-01-21 09:28:49.493228: step: 1888/526, loss: 0.06665010750293732 2023-01-21 09:28:50.641889: step: 1892/526, loss: 0.3820188641548157 2023-01-21 09:28:51.808318: step: 1896/526, loss: 0.14128609001636505 2023-01-21 09:28:52.934259: step: 1900/526, loss: 0.04584827274084091 2023-01-21 09:28:54.037870: step: 1904/526, loss: 0.08219747245311737 2023-01-21 09:28:55.172426: step: 1908/526, loss: 0.028188802301883698 2023-01-21 09:28:56.293554: step: 1912/526, loss: 0.04881878197193146 2023-01-21 09:28:57.414030: step: 1916/526, loss: 0.1517138034105301 2023-01-21 09:28:58.580510: step: 1920/526, loss: 1.1820714473724365 2023-01-21 09:28:59.707537: step: 1924/526, loss: 0.32395753264427185 2023-01-21 09:29:00.820817: step: 1928/526, loss: 1.1668269634246826 2023-01-21 09:29:01.948420: step: 1932/526, loss: 0.127878338098526 2023-01-21 09:29:03.058383: step: 1936/526, loss: 0.8874449729919434 2023-01-21 09:29:04.208426: step: 1940/526, loss: 0.744866132736206 2023-01-21 09:29:05.317590: step: 1944/526, loss: 0.08502760529518127 2023-01-21 09:29:06.464243: step: 1948/526, loss: 0.08729524910449982 2023-01-21 09:29:07.579036: step: 1952/526, loss: 0.08264384418725967 2023-01-21 09:29:08.720951: step: 1956/526, loss: 0.08795285224914551 2023-01-21 09:29:09.847188: step: 1960/526, loss: 0.19313374161720276 2023-01-21 09:29:10.982120: step: 1964/526, loss: 5.26082181930542 2023-01-21 09:29:12.136838: step: 1968/526, loss: 0.09294624626636505 2023-01-21 09:29:13.255780: step: 1972/526, loss: 0.10113152861595154 2023-01-21 09:29:14.376060: step: 1976/526, loss: 0.2049887776374817 2023-01-21 09:29:15.525441: step: 1980/526, loss: 0.19875651597976685 2023-01-21 09:29:16.664475: step: 1984/526, loss: 0.2498459815979004 2023-01-21 09:29:17.786115: step: 1988/526, loss: 0.10887555778026581 2023-01-21 09:29:18.921896: step: 1992/526, loss: 0.05493581295013428 2023-01-21 09:29:20.062867: step: 1996/526, loss: 0.15251369774341583 2023-01-21 09:29:21.183391: step: 2000/526, loss: 0.01222152728587389 2023-01-21 09:29:22.319241: step: 2004/526, loss: 0.7983397245407104 2023-01-21 09:29:23.431584: step: 2008/526, loss: 0.22303137183189392 2023-01-21 09:29:24.555389: step: 2012/526, loss: 0.6977473497390747 2023-01-21 09:29:25.730089: step: 2016/526, loss: 0.3306276202201843 2023-01-21 09:29:26.897720: step: 2020/526, loss: 0.40909910202026367 2023-01-21 09:29:28.070347: step: 2024/526, loss: 0.10456257313489914 2023-01-21 09:29:29.206609: step: 2028/526, loss: 0.23742122948169708 2023-01-21 09:29:30.369675: step: 2032/526, loss: 0.16777239739894867 2023-01-21 09:29:31.515065: step: 2036/526, loss: 0.02179795503616333 2023-01-21 09:29:32.684776: step: 2040/526, loss: 0.3406356871128082 2023-01-21 09:29:33.822241: step: 2044/526, loss: 0.7128070592880249 2023-01-21 09:29:34.987147: step: 2048/526, loss: 0.10555877536535263 2023-01-21 09:29:36.089005: step: 2052/526, loss: 0.06222992390394211 2023-01-21 09:29:37.223679: step: 2056/526, loss: 0.7264218330383301 2023-01-21 09:29:38.365565: step: 2060/526, loss: 0.04870662838220596 2023-01-21 09:29:39.513130: step: 2064/526, loss: 1.328598976135254 2023-01-21 09:29:40.632534: step: 2068/526, loss: 0.09405260533094406 2023-01-21 09:29:41.805856: step: 2072/526, loss: 0.12173452228307724 2023-01-21 09:29:42.958984: step: 2076/526, loss: 0.15678229928016663 2023-01-21 09:29:44.083395: step: 2080/526, loss: 1.41208815574646 2023-01-21 09:29:45.222333: step: 2084/526, loss: 0.6326181888580322 2023-01-21 09:29:46.394895: step: 2088/526, loss: 0.1623951941728592 2023-01-21 09:29:47.531346: step: 2092/526, loss: 0.15030059218406677 2023-01-21 09:29:48.685704: step: 2096/526, loss: 0.39091941714286804 2023-01-21 09:29:49.833257: step: 2100/526, loss: 0.8702443242073059 2023-01-21 09:29:50.980044: step: 2104/526, loss: 0.07611255347728729 ================================================== Loss: 0.353 -------------------- Dev: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Test: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Chinese: {'event': {'p': 0.5434782608695652, 'r': 0.9259259259259259, 'f1': 0.6849315068493151}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Korean: {'event': {'p': 0.5932203389830508, 'r': 0.5555555555555556, 'f1': 0.5737704918032787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Russian: {'event': {'p': 0.5116279069767442, 'r': 0.6111111111111112, 'f1': 0.5569620253164557}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} New best korean model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.5932203389830508, 'r': 0.5555555555555556, 'f1': 0.5737704918032787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5116279069767442, 'r': 0.6111111111111112, 'f1': 0.5569620253164557}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 5 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:30:47.785023: step: 4/526, loss: 0.07552170753479004 2023-01-21 09:30:48.905121: step: 8/526, loss: 0.1778736114501953 2023-01-21 09:30:50.072262: step: 12/526, loss: 0.47908085584640503 2023-01-21 09:30:51.194754: step: 16/526, loss: 0.41332149505615234 2023-01-21 09:30:52.329012: step: 20/526, loss: 0.22914238274097443 2023-01-21 09:30:53.452973: step: 24/526, loss: 0.49168384075164795 2023-01-21 09:30:54.587675: step: 28/526, loss: 0.7054325342178345 2023-01-21 09:30:55.716390: step: 32/526, loss: 0.22354355454444885 2023-01-21 09:30:56.881568: step: 36/526, loss: 0.2816998362541199 2023-01-21 09:30:57.997060: step: 40/526, loss: 0.15015316009521484 2023-01-21 09:30:59.148980: step: 44/526, loss: 0.3128385543823242 2023-01-21 09:31:00.288921: step: 48/526, loss: 0.1136753112077713 2023-01-21 09:31:01.441877: step: 52/526, loss: 0.3024848699569702 2023-01-21 09:31:02.581090: step: 56/526, loss: 0.2961660325527191 2023-01-21 09:31:03.707676: step: 60/526, loss: 0.16076450049877167 2023-01-21 09:31:04.867914: step: 64/526, loss: 0.70372474193573 2023-01-21 09:31:05.985785: step: 68/526, loss: 0.05580272898077965 2023-01-21 09:31:07.111349: step: 72/526, loss: 0.33491912484169006 2023-01-21 09:31:08.240945: step: 76/526, loss: 0.08502702414989471 2023-01-21 09:31:09.366104: step: 80/526, loss: 0.6640203595161438 2023-01-21 09:31:10.503277: step: 84/526, loss: 0.06171112135052681 2023-01-21 09:31:11.654022: step: 88/526, loss: 0.2955389618873596 2023-01-21 09:31:12.827485: step: 92/526, loss: 0.07607822120189667 2023-01-21 09:31:13.933267: step: 96/526, loss: 0.15505895018577576 2023-01-21 09:31:15.062475: step: 100/526, loss: 1.0233112573623657 2023-01-21 09:31:16.190549: step: 104/526, loss: 0.13770484924316406 2023-01-21 09:31:17.339892: step: 108/526, loss: 0.4573703706264496 2023-01-21 09:31:18.496429: step: 112/526, loss: 0.3016144931316376 2023-01-21 09:31:19.646484: step: 116/526, loss: 0.6969193816184998 2023-01-21 09:31:20.775280: step: 120/526, loss: 0.20527324080467224 2023-01-21 09:31:21.902007: step: 124/526, loss: 1.2636239528656006 2023-01-21 09:31:23.052012: step: 128/526, loss: 0.5671699047088623 2023-01-21 09:31:24.197542: step: 132/526, loss: 0.10811157524585724 2023-01-21 09:31:25.329133: step: 136/526, loss: 0.12215442955493927 2023-01-21 09:31:26.460524: step: 140/526, loss: 0.21227779984474182 2023-01-21 09:31:27.566396: step: 144/526, loss: 0.030367136001586914 2023-01-21 09:31:28.695921: step: 148/526, loss: 0.4781329333782196 2023-01-21 09:31:29.824561: step: 152/526, loss: 0.06072940677404404 2023-01-21 09:31:30.961826: step: 156/526, loss: 0.05557294189929962 2023-01-21 09:31:32.091057: step: 160/526, loss: 0.4160362482070923 2023-01-21 09:31:33.224147: step: 164/526, loss: 0.051721859723329544 2023-01-21 09:31:34.354657: step: 168/526, loss: 0.0803864449262619 2023-01-21 09:31:35.490770: step: 172/526, loss: 0.08706312626600266 2023-01-21 09:31:36.596189: step: 176/526, loss: 0.12174461781978607 2023-01-21 09:31:37.747326: step: 180/526, loss: 0.14703483879566193 2023-01-21 09:31:38.897084: step: 184/526, loss: 0.19677190482616425 2023-01-21 09:31:40.066011: step: 188/526, loss: 0.5028398036956787 2023-01-21 09:31:41.191196: step: 192/526, loss: 0.03458528593182564 2023-01-21 09:31:42.312593: step: 196/526, loss: 0.08830176293849945 2023-01-21 09:31:43.436856: step: 200/526, loss: 0.05585050582885742 2023-01-21 09:31:44.604404: step: 204/526, loss: 0.19362641870975494 2023-01-21 09:31:45.721536: step: 208/526, loss: 0.08621148765087128 2023-01-21 09:31:46.880640: step: 212/526, loss: 0.5853188633918762 2023-01-21 09:31:48.031193: step: 216/526, loss: 0.4329465925693512 2023-01-21 09:31:49.167058: step: 220/526, loss: 0.4039674699306488 2023-01-21 09:31:50.289395: step: 224/526, loss: 0.026792338117957115 2023-01-21 09:31:51.450924: step: 228/526, loss: 0.1944846212863922 2023-01-21 09:31:52.582266: step: 232/526, loss: 0.10644569993019104 2023-01-21 09:31:53.696739: step: 236/526, loss: 3.594597101211548 2023-01-21 09:31:54.833542: step: 240/526, loss: 0.23088054358959198 2023-01-21 09:31:55.968139: step: 244/526, loss: 0.060803867876529694 2023-01-21 09:31:57.122746: step: 248/526, loss: 0.07625265419483185 2023-01-21 09:31:58.254097: step: 252/526, loss: 0.3246402144432068 2023-01-21 09:31:59.363917: step: 256/526, loss: 0.222834974527359 2023-01-21 09:32:00.538368: step: 260/526, loss: 0.15671049058437347 2023-01-21 09:32:01.657757: step: 264/526, loss: 0.2793780565261841 2023-01-21 09:32:02.793468: step: 268/526, loss: 0.3518807291984558 2023-01-21 09:32:03.940485: step: 272/526, loss: 0.16502493619918823 2023-01-21 09:32:05.088078: step: 276/526, loss: 0.13538670539855957 2023-01-21 09:32:06.242607: step: 280/526, loss: 0.33921700716018677 2023-01-21 09:32:07.393598: step: 284/526, loss: 0.09293756633996964 2023-01-21 09:32:08.535688: step: 288/526, loss: 0.12042374908924103 2023-01-21 09:32:09.655787: step: 292/526, loss: 0.10213012993335724 2023-01-21 09:32:10.770418: step: 296/526, loss: 0.14765429496765137 2023-01-21 09:32:11.900355: step: 300/526, loss: 0.11585512012243271 2023-01-21 09:32:13.080345: step: 304/526, loss: 0.2017158567905426 2023-01-21 09:32:14.210837: step: 308/526, loss: 0.4023115336894989 2023-01-21 09:32:15.332824: step: 312/526, loss: 0.051944926381111145 2023-01-21 09:32:16.491221: step: 316/526, loss: 0.03483166918158531 2023-01-21 09:32:17.670005: step: 320/526, loss: 0.1493522673845291 2023-01-21 09:32:18.801878: step: 324/526, loss: 0.436555951833725 2023-01-21 09:32:19.926445: step: 328/526, loss: 0.09910402446985245 2023-01-21 09:32:21.063443: step: 332/526, loss: 0.10703545063734055 2023-01-21 09:32:22.180211: step: 336/526, loss: 0.01687011681497097 2023-01-21 09:32:23.333021: step: 340/526, loss: 0.6482803821563721 2023-01-21 09:32:24.507972: step: 344/526, loss: 0.3501931428909302 2023-01-21 09:32:25.646549: step: 348/526, loss: 0.5176986455917358 2023-01-21 09:32:26.782072: step: 352/526, loss: 0.41397643089294434 2023-01-21 09:32:27.926725: step: 356/526, loss: 0.11719612777233124 2023-01-21 09:32:29.087559: step: 360/526, loss: 0.14571914076805115 2023-01-21 09:32:30.220236: step: 364/526, loss: 0.1404467523097992 2023-01-21 09:32:31.373514: step: 368/526, loss: 0.04848051071166992 2023-01-21 09:32:32.517820: step: 372/526, loss: 0.029836654663085938 2023-01-21 09:32:33.631771: step: 376/526, loss: 0.049016378819942474 2023-01-21 09:32:34.799265: step: 380/526, loss: 0.4558943212032318 2023-01-21 09:32:35.948925: step: 384/526, loss: 0.07975959777832031 2023-01-21 09:32:37.119482: step: 388/526, loss: 0.24870933592319489 2023-01-21 09:32:38.273302: step: 392/526, loss: 0.07261504977941513 2023-01-21 09:32:39.402209: step: 396/526, loss: 0.2094138115644455 2023-01-21 09:32:40.534400: step: 400/526, loss: 0.06749296188354492 2023-01-21 09:32:41.664421: step: 404/526, loss: 0.6789333820343018 2023-01-21 09:32:42.840704: step: 408/526, loss: 0.14371486008167267 2023-01-21 09:32:43.944890: step: 412/526, loss: 0.11847849190235138 2023-01-21 09:32:45.062013: step: 416/526, loss: 0.14260311424732208 2023-01-21 09:32:46.221788: step: 420/526, loss: 0.6579089164733887 2023-01-21 09:32:47.360355: step: 424/526, loss: 0.13759641349315643 2023-01-21 09:32:48.522095: step: 428/526, loss: 0.1361497938632965 2023-01-21 09:32:49.648714: step: 432/526, loss: 0.11495156586170197 2023-01-21 09:32:50.787279: step: 436/526, loss: 0.10495725274085999 2023-01-21 09:32:51.901646: step: 440/526, loss: 0.2477591633796692 2023-01-21 09:32:53.077771: step: 444/526, loss: 0.3187906742095947 2023-01-21 09:32:54.206378: step: 448/526, loss: 0.11388960480690002 2023-01-21 09:32:55.321847: step: 452/526, loss: 0.1595078557729721 2023-01-21 09:32:56.470884: step: 456/526, loss: 0.4562973082065582 2023-01-21 09:32:57.627342: step: 460/526, loss: 0.2635514736175537 2023-01-21 09:32:58.752196: step: 464/526, loss: 0.1092710942029953 2023-01-21 09:32:59.881323: step: 468/526, loss: 0.030313827097415924 2023-01-21 09:33:00.982182: step: 472/526, loss: 0.4443342983722687 2023-01-21 09:33:02.118924: step: 476/526, loss: 0.0713747963309288 2023-01-21 09:33:03.279762: step: 480/526, loss: 0.084652379155159 2023-01-21 09:33:04.404858: step: 484/526, loss: 0.08752064406871796 2023-01-21 09:33:05.547725: step: 488/526, loss: 0.03136329725384712 2023-01-21 09:33:06.676738: step: 492/526, loss: 0.03622160106897354 2023-01-21 09:33:07.802055: step: 496/526, loss: 0.05147838592529297 2023-01-21 09:33:08.922542: step: 500/526, loss: 0.08350391685962677 2023-01-21 09:33:10.083046: step: 504/526, loss: 0.15375709533691406 2023-01-21 09:33:11.227348: step: 508/526, loss: 0.6762579083442688 2023-01-21 09:33:12.382635: step: 512/526, loss: 0.6304488182067871 2023-01-21 09:33:13.508828: step: 516/526, loss: 0.2314133644104004 2023-01-21 09:33:14.630556: step: 520/526, loss: 0.10436764359474182 2023-01-21 09:33:15.774186: step: 524/526, loss: 0.5557795763015747 2023-01-21 09:33:16.932102: step: 528/526, loss: 0.15001335740089417 2023-01-21 09:33:18.061613: step: 532/526, loss: 0.6348365545272827 2023-01-21 09:33:19.174409: step: 536/526, loss: 0.08023472130298615 2023-01-21 09:33:20.276841: step: 540/526, loss: 0.287660151720047 2023-01-21 09:33:21.430107: step: 544/526, loss: 0.08438985794782639 2023-01-21 09:33:22.573267: step: 548/526, loss: 0.14376278221607208 2023-01-21 09:33:23.731980: step: 552/526, loss: 0.057227421551942825 2023-01-21 09:33:24.853825: step: 556/526, loss: 0.06003871187567711 2023-01-21 09:33:25.989776: step: 560/526, loss: 0.08796177059412003 2023-01-21 09:33:27.127408: step: 564/526, loss: 0.44969606399536133 2023-01-21 09:33:28.274649: step: 568/526, loss: 0.5560224652290344 2023-01-21 09:33:29.426945: step: 572/526, loss: 0.19575929641723633 2023-01-21 09:33:30.569611: step: 576/526, loss: 0.10686217248439789 2023-01-21 09:33:31.731324: step: 580/526, loss: 0.27057409286499023 2023-01-21 09:33:32.859357: step: 584/526, loss: 0.6121435165405273 2023-01-21 09:33:33.998795: step: 588/526, loss: 0.0969381332397461 2023-01-21 09:33:35.112115: step: 592/526, loss: 0.09959135949611664 2023-01-21 09:33:36.269934: step: 596/526, loss: 0.0922972708940506 2023-01-21 09:33:37.393063: step: 600/526, loss: 0.10342487692832947 2023-01-21 09:33:38.535017: step: 604/526, loss: 0.019901657477021217 2023-01-21 09:33:39.689441: step: 608/526, loss: 0.19516621530056 2023-01-21 09:33:40.817230: step: 612/526, loss: 0.3263491094112396 2023-01-21 09:33:41.942913: step: 616/526, loss: 0.1799975484609604 2023-01-21 09:33:43.111304: step: 620/526, loss: 0.06523466110229492 2023-01-21 09:33:44.252694: step: 624/526, loss: 0.4737827181816101 2023-01-21 09:33:45.383544: step: 628/526, loss: 0.05707339942455292 2023-01-21 09:33:46.514044: step: 632/526, loss: 0.1440323442220688 2023-01-21 09:33:47.646093: step: 636/526, loss: 0.10384152084589005 2023-01-21 09:33:48.804589: step: 640/526, loss: 0.39088764786720276 2023-01-21 09:33:49.948766: step: 644/526, loss: 0.21252554655075073 2023-01-21 09:33:51.089892: step: 648/526, loss: 0.23186130821704865 2023-01-21 09:33:52.238833: step: 652/526, loss: 0.14491558074951172 2023-01-21 09:33:53.351962: step: 656/526, loss: 0.0809914618730545 2023-01-21 09:33:54.524026: step: 660/526, loss: 0.581156849861145 2023-01-21 09:33:55.654348: step: 664/526, loss: 0.11716137081384659 2023-01-21 09:33:56.763476: step: 668/526, loss: 0.42929807305336 2023-01-21 09:33:57.866535: step: 672/526, loss: 0.06738128513097763 2023-01-21 09:33:58.995398: step: 676/526, loss: 0.03126473352313042 2023-01-21 09:34:00.118553: step: 680/526, loss: 0.5546294450759888 2023-01-21 09:34:01.277570: step: 684/526, loss: 0.40436506271362305 2023-01-21 09:34:02.422519: step: 688/526, loss: 0.1566580832004547 2023-01-21 09:34:03.538842: step: 692/526, loss: 0.17345371842384338 2023-01-21 09:34:04.644006: step: 696/526, loss: 0.05924072116613388 2023-01-21 09:34:05.779072: step: 700/526, loss: 0.3746162950992584 2023-01-21 09:34:06.922046: step: 704/526, loss: 0.1387009620666504 2023-01-21 09:34:08.071743: step: 708/526, loss: 0.1399168074131012 2023-01-21 09:34:09.186833: step: 712/526, loss: 0.05567455291748047 2023-01-21 09:34:10.362912: step: 716/526, loss: 0.3600977659225464 2023-01-21 09:34:11.542885: step: 720/526, loss: 0.33110275864601135 2023-01-21 09:34:12.720907: step: 724/526, loss: 0.2367338240146637 2023-01-21 09:34:13.883404: step: 728/526, loss: 0.049689389765262604 2023-01-21 09:34:15.076630: step: 732/526, loss: 0.06639156490564346 2023-01-21 09:34:16.238565: step: 736/526, loss: 0.10460825264453888 2023-01-21 09:34:17.390475: step: 740/526, loss: 0.2921673059463501 2023-01-21 09:34:18.530616: step: 744/526, loss: 0.082593634724617 2023-01-21 09:34:19.677784: step: 748/526, loss: 0.22028112411499023 2023-01-21 09:34:20.810726: step: 752/526, loss: 0.20759665966033936 2023-01-21 09:34:21.943819: step: 756/526, loss: 0.04025144502520561 2023-01-21 09:34:23.074405: step: 760/526, loss: 0.06561832129955292 2023-01-21 09:34:24.199254: step: 764/526, loss: 0.10911436378955841 2023-01-21 09:34:25.346674: step: 768/526, loss: 0.07309374958276749 2023-01-21 09:34:26.480561: step: 772/526, loss: 0.13165739178657532 2023-01-21 09:34:27.606622: step: 776/526, loss: 0.3502717614173889 2023-01-21 09:34:28.738523: step: 780/526, loss: 0.08526067435741425 2023-01-21 09:34:29.880074: step: 784/526, loss: 0.30387717485427856 2023-01-21 09:34:31.013083: step: 788/526, loss: 0.4214094281196594 2023-01-21 09:34:32.186585: step: 792/526, loss: 0.677837610244751 2023-01-21 09:34:33.334108: step: 796/526, loss: 0.11714563518762589 2023-01-21 09:34:34.461421: step: 800/526, loss: 0.5716214179992676 2023-01-21 09:34:35.587580: step: 804/526, loss: 0.0568116195499897 2023-01-21 09:34:36.720277: step: 808/526, loss: 0.632706880569458 2023-01-21 09:34:37.813871: step: 812/526, loss: 0.31289201974868774 2023-01-21 09:34:38.931895: step: 816/526, loss: 0.5513056516647339 2023-01-21 09:34:40.090573: step: 820/526, loss: 0.12334317713975906 2023-01-21 09:34:41.244138: step: 824/526, loss: 0.056501008570194244 2023-01-21 09:34:42.400143: step: 828/526, loss: 0.16306714713573456 2023-01-21 09:34:43.547011: step: 832/526, loss: 0.12435455620288849 2023-01-21 09:34:44.689295: step: 836/526, loss: 0.07763805985450745 2023-01-21 09:34:45.816664: step: 840/526, loss: 0.2870665490627289 2023-01-21 09:34:46.974154: step: 844/526, loss: 0.4671330451965332 2023-01-21 09:34:48.109040: step: 848/526, loss: 0.016177939251065254 2023-01-21 09:34:49.257513: step: 852/526, loss: 0.025417424738407135 2023-01-21 09:34:50.385163: step: 856/526, loss: 0.043744660913944244 2023-01-21 09:34:51.525537: step: 860/526, loss: 0.581939697265625 2023-01-21 09:34:52.653980: step: 864/526, loss: 0.061492159962654114 2023-01-21 09:34:53.782239: step: 868/526, loss: 0.6203882098197937 2023-01-21 09:34:54.939157: step: 872/526, loss: 0.055752232670784 2023-01-21 09:34:56.057006: step: 876/526, loss: 0.024624967947602272 2023-01-21 09:34:57.183737: step: 880/526, loss: 0.16453096270561218 2023-01-21 09:34:58.323058: step: 884/526, loss: 0.10548434406518936 2023-01-21 09:34:59.425076: step: 888/526, loss: 0.20627892017364502 2023-01-21 09:35:00.577994: step: 892/526, loss: 0.07390408962965012 2023-01-21 09:35:01.735742: step: 896/526, loss: 0.12392453849315643 2023-01-21 09:35:02.867702: step: 900/526, loss: 0.622948944568634 2023-01-21 09:35:04.019161: step: 904/526, loss: 0.1790519654750824 2023-01-21 09:35:05.163940: step: 908/526, loss: 0.8041629195213318 2023-01-21 09:35:06.343542: step: 912/526, loss: 0.29022684693336487 2023-01-21 09:35:07.500429: step: 916/526, loss: 0.35412710905075073 2023-01-21 09:35:08.676672: step: 920/526, loss: 0.1469673216342926 2023-01-21 09:35:09.849816: step: 924/526, loss: 0.18353509902954102 2023-01-21 09:35:10.999650: step: 928/526, loss: 0.3620956242084503 2023-01-21 09:35:12.148934: step: 932/526, loss: 0.12959997355937958 2023-01-21 09:35:13.274903: step: 936/526, loss: 0.27869871258735657 2023-01-21 09:35:14.413064: step: 940/526, loss: 0.25083011388778687 2023-01-21 09:35:15.539697: step: 944/526, loss: 0.1387719213962555 2023-01-21 09:35:16.681848: step: 948/526, loss: 0.18806734681129456 2023-01-21 09:35:17.824966: step: 952/526, loss: 0.37217214703559875 2023-01-21 09:35:18.950568: step: 956/526, loss: 0.17451295256614685 2023-01-21 09:35:20.069891: step: 960/526, loss: 0.12138272076845169 2023-01-21 09:35:21.205356: step: 964/526, loss: 0.987384021282196 2023-01-21 09:35:22.368512: step: 968/526, loss: 0.08471079170703888 2023-01-21 09:35:23.500462: step: 972/526, loss: 0.16412298381328583 2023-01-21 09:35:24.667309: step: 976/526, loss: 0.09780779480934143 2023-01-21 09:35:25.798203: step: 980/526, loss: 0.07905187457799911 2023-01-21 09:35:26.942092: step: 984/526, loss: 0.030390549451112747 2023-01-21 09:35:28.081191: step: 988/526, loss: 0.0787535235285759 2023-01-21 09:35:29.197322: step: 992/526, loss: 0.17775791883468628 2023-01-21 09:35:30.314237: step: 996/526, loss: 0.059416815638542175 2023-01-21 09:35:31.448061: step: 1000/526, loss: 0.09653346240520477 2023-01-21 09:35:32.554437: step: 1004/526, loss: 0.13981446623802185 2023-01-21 09:35:33.708491: step: 1008/526, loss: 1.0894023180007935 2023-01-21 09:35:34.832343: step: 1012/526, loss: 0.0881490707397461 2023-01-21 09:35:35.984566: step: 1016/526, loss: 0.0698762908577919 2023-01-21 09:35:37.127399: step: 1020/526, loss: 0.16481313109397888 2023-01-21 09:35:38.287530: step: 1024/526, loss: 0.07063284516334534 2023-01-21 09:35:39.458171: step: 1028/526, loss: 0.09497375786304474 2023-01-21 09:35:40.591998: step: 1032/526, loss: 0.122956283390522 2023-01-21 09:35:41.712259: step: 1036/526, loss: 0.1150793582201004 2023-01-21 09:35:42.858441: step: 1040/526, loss: 0.10032553970813751 2023-01-21 09:35:43.992364: step: 1044/526, loss: 0.0470617301762104 2023-01-21 09:35:45.154865: step: 1048/526, loss: 0.3499331474304199 2023-01-21 09:35:46.332680: step: 1052/526, loss: 0.30687639117240906 2023-01-21 09:35:47.482906: step: 1056/526, loss: 0.37629234790802 2023-01-21 09:35:48.606848: step: 1060/526, loss: 0.06239809840917587 2023-01-21 09:35:49.734263: step: 1064/526, loss: 0.004215431399643421 2023-01-21 09:35:50.865964: step: 1068/526, loss: 0.05564947426319122 2023-01-21 09:35:52.019566: step: 1072/526, loss: 0.11828222870826721 2023-01-21 09:35:53.130239: step: 1076/526, loss: 0.048153355717659 2023-01-21 09:35:54.268740: step: 1080/526, loss: 0.07006406784057617 2023-01-21 09:35:55.423179: step: 1084/526, loss: 0.2897661328315735 2023-01-21 09:35:56.611892: step: 1088/526, loss: 0.02233278751373291 2023-01-21 09:35:57.754824: step: 1092/526, loss: 0.20013219118118286 2023-01-21 09:35:58.900558: step: 1096/526, loss: 0.07541098445653915 2023-01-21 09:36:00.032718: step: 1100/526, loss: 0.18152180314064026 2023-01-21 09:36:01.186318: step: 1104/526, loss: 0.09701547026634216 2023-01-21 09:36:02.303331: step: 1108/526, loss: 0.007830810733139515 2023-01-21 09:36:03.451690: step: 1112/526, loss: 0.20827673375606537 2023-01-21 09:36:04.566110: step: 1116/526, loss: 0.07904062420129776 2023-01-21 09:36:05.694683: step: 1120/526, loss: 0.06303844600915909 2023-01-21 09:36:06.897375: step: 1124/526, loss: 0.27868205308914185 2023-01-21 09:36:08.056047: step: 1128/526, loss: 0.4336077570915222 2023-01-21 09:36:09.217118: step: 1132/526, loss: 0.0917481929063797 2023-01-21 09:36:10.355618: step: 1136/526, loss: 0.5258356928825378 2023-01-21 09:36:11.490102: step: 1140/526, loss: 0.17753896117210388 2023-01-21 09:36:12.621898: step: 1144/526, loss: 0.23496904969215393 2023-01-21 09:36:13.748774: step: 1148/526, loss: 0.06634197384119034 2023-01-21 09:36:14.890025: step: 1152/526, loss: 0.26711511611938477 2023-01-21 09:36:16.035887: step: 1156/526, loss: 0.13119515776634216 2023-01-21 09:36:17.226134: step: 1160/526, loss: 0.07149533927440643 2023-01-21 09:36:18.371157: step: 1164/526, loss: 0.47180628776550293 2023-01-21 09:36:19.521978: step: 1168/526, loss: 0.13471731543540955 2023-01-21 09:36:20.660564: step: 1172/526, loss: 0.28347986936569214 2023-01-21 09:36:21.771985: step: 1176/526, loss: 0.01544198952615261 2023-01-21 09:36:22.908481: step: 1180/526, loss: 0.1351659893989563 2023-01-21 09:36:24.035662: step: 1184/526, loss: 0.16764900088310242 2023-01-21 09:36:25.157195: step: 1188/526, loss: 5.7529215812683105 2023-01-21 09:36:26.282808: step: 1192/526, loss: 0.6821755170822144 2023-01-21 09:36:27.413996: step: 1196/526, loss: 1.002524971961975 2023-01-21 09:36:28.543315: step: 1200/526, loss: 0.21645474433898926 2023-01-21 09:36:29.681484: step: 1204/526, loss: 0.15088805556297302 2023-01-21 09:36:30.798675: step: 1208/526, loss: 0.5905170440673828 2023-01-21 09:36:31.933481: step: 1212/526, loss: 0.046729087829589844 2023-01-21 09:36:33.089164: step: 1216/526, loss: 0.1753438413143158 2023-01-21 09:36:34.237835: step: 1220/526, loss: 0.08497485518455505 2023-01-21 09:36:35.366314: step: 1224/526, loss: 0.1213504746556282 2023-01-21 09:36:36.465019: step: 1228/526, loss: 0.4379407465457916 2023-01-21 09:36:37.602443: step: 1232/526, loss: 0.12683840095996857 2023-01-21 09:36:38.770364: step: 1236/526, loss: 0.4335874617099762 2023-01-21 09:36:39.911842: step: 1240/526, loss: 0.11594896018505096 2023-01-21 09:36:41.072140: step: 1244/526, loss: 0.09811754524707794 2023-01-21 09:36:42.215455: step: 1248/526, loss: 0.9354307651519775 2023-01-21 09:36:43.360906: step: 1252/526, loss: 0.136609748005867 2023-01-21 09:36:44.519942: step: 1256/526, loss: 0.14618225395679474 2023-01-21 09:36:45.677754: step: 1260/526, loss: 0.049700167030096054 2023-01-21 09:36:46.812270: step: 1264/526, loss: 0.026849079877138138 2023-01-21 09:36:47.939621: step: 1268/526, loss: 0.030932903289794922 2023-01-21 09:36:49.094058: step: 1272/526, loss: 0.12946195900440216 2023-01-21 09:36:50.223596: step: 1276/526, loss: 0.3413374125957489 2023-01-21 09:36:51.373003: step: 1280/526, loss: 0.5484434962272644 2023-01-21 09:36:52.522098: step: 1284/526, loss: 0.09699907153844833 2023-01-21 09:36:53.644601: step: 1288/526, loss: 0.10249491035938263 2023-01-21 09:36:54.782761: step: 1292/526, loss: 0.528791606426239 2023-01-21 09:36:55.929540: step: 1296/526, loss: 0.08172684162855148 2023-01-21 09:36:57.062362: step: 1300/526, loss: 0.15835723280906677 2023-01-21 09:36:58.232486: step: 1304/526, loss: 3.7234480381011963 2023-01-21 09:36:59.374400: step: 1308/526, loss: 0.15857048332691193 2023-01-21 09:37:00.509572: step: 1312/526, loss: 0.15546664595603943 2023-01-21 09:37:01.648877: step: 1316/526, loss: 0.013378048315644264 2023-01-21 09:37:02.824018: step: 1320/526, loss: 0.31611576676368713 2023-01-21 09:37:03.972309: step: 1324/526, loss: 0.056382276117801666 2023-01-21 09:37:05.083809: step: 1328/526, loss: 0.1151147335767746 2023-01-21 09:37:06.200195: step: 1332/526, loss: 0.19945715367794037 2023-01-21 09:37:07.331667: step: 1336/526, loss: 0.10670605301856995 2023-01-21 09:37:08.501481: step: 1340/526, loss: 0.06457419693470001 2023-01-21 09:37:09.674999: step: 1344/526, loss: 0.1427433043718338 2023-01-21 09:37:10.793393: step: 1348/526, loss: 0.07333040237426758 2023-01-21 09:37:11.946061: step: 1352/526, loss: 0.28943243622779846 2023-01-21 09:37:13.067437: step: 1356/526, loss: 0.04551959037780762 2023-01-21 09:37:14.225855: step: 1360/526, loss: 0.03165636211633682 2023-01-21 09:37:15.351625: step: 1364/526, loss: 0.110833540558815 2023-01-21 09:37:16.476962: step: 1368/526, loss: 0.18557500839233398 2023-01-21 09:37:17.637280: step: 1372/526, loss: 0.7316573858261108 2023-01-21 09:37:18.783965: step: 1376/526, loss: 0.6557825803756714 2023-01-21 09:37:19.932668: step: 1380/526, loss: 0.0473148375749588 2023-01-21 09:37:21.065643: step: 1384/526, loss: 0.19366998970508575 2023-01-21 09:37:22.214390: step: 1388/526, loss: 0.07655983418226242 2023-01-21 09:37:23.336888: step: 1392/526, loss: 0.08151821792125702 2023-01-21 09:37:24.462574: step: 1396/526, loss: 0.14024314284324646 2023-01-21 09:37:25.620854: step: 1400/526, loss: 0.12762165069580078 2023-01-21 09:37:26.747908: step: 1404/526, loss: 0.29477283358573914 2023-01-21 09:37:27.889475: step: 1408/526, loss: 0.1507505476474762 2023-01-21 09:37:29.014546: step: 1412/526, loss: 0.055031586438417435 2023-01-21 09:37:30.152380: step: 1416/526, loss: 0.19547760486602783 2023-01-21 09:37:31.311292: step: 1420/526, loss: 0.08987510204315186 2023-01-21 09:37:32.435633: step: 1424/526, loss: 0.17260035872459412 2023-01-21 09:37:33.574229: step: 1428/526, loss: 0.4401676654815674 2023-01-21 09:37:34.712197: step: 1432/526, loss: 0.10761260986328125 2023-01-21 09:37:35.853566: step: 1436/526, loss: 0.17115606367588043 2023-01-21 09:37:36.980920: step: 1440/526, loss: 0.12303884327411652 2023-01-21 09:37:38.153142: step: 1444/526, loss: 0.3938596844673157 2023-01-21 09:37:39.283169: step: 1448/526, loss: 0.13859958946704865 2023-01-21 09:37:40.442636: step: 1452/526, loss: 0.991059422492981 2023-01-21 09:37:41.584993: step: 1456/526, loss: 0.11598372459411621 2023-01-21 09:37:42.710913: step: 1460/526, loss: 0.06403837352991104 2023-01-21 09:37:43.856803: step: 1464/526, loss: 0.04212365299463272 2023-01-21 09:37:44.967036: step: 1468/526, loss: 0.045427776873111725 2023-01-21 09:37:46.111079: step: 1472/526, loss: 0.1310184895992279 2023-01-21 09:37:47.254969: step: 1476/526, loss: 0.20100879669189453 2023-01-21 09:37:48.400142: step: 1480/526, loss: 0.230610653758049 2023-01-21 09:37:49.537522: step: 1484/526, loss: 0.9139903783798218 2023-01-21 09:37:50.671798: step: 1488/526, loss: 0.2325393706560135 2023-01-21 09:37:51.820284: step: 1492/526, loss: 0.1776561737060547 2023-01-21 09:37:52.963562: step: 1496/526, loss: 0.09078922867774963 2023-01-21 09:37:54.118806: step: 1500/526, loss: 0.5354470014572144 2023-01-21 09:37:55.228522: step: 1504/526, loss: 0.040494538843631744 2023-01-21 09:37:56.373724: step: 1508/526, loss: 0.18460139632225037 2023-01-21 09:37:57.512161: step: 1512/526, loss: 0.21998730301856995 2023-01-21 09:37:58.684950: step: 1516/526, loss: 0.23641423881053925 2023-01-21 09:37:59.819336: step: 1520/526, loss: 0.05305318534374237 2023-01-21 09:38:00.958366: step: 1524/526, loss: 0.14219188690185547 2023-01-21 09:38:02.094818: step: 1528/526, loss: 0.192413330078125 2023-01-21 09:38:03.242003: step: 1532/526, loss: 0.19365087151527405 2023-01-21 09:38:04.370169: step: 1536/526, loss: 0.007358169183135033 2023-01-21 09:38:05.479424: step: 1540/526, loss: 0.06071644276380539 2023-01-21 09:38:06.624235: step: 1544/526, loss: 0.3244949281215668 2023-01-21 09:38:07.777856: step: 1548/526, loss: 0.9575450420379639 2023-01-21 09:38:08.951557: step: 1552/526, loss: 0.06787557899951935 2023-01-21 09:38:10.104277: step: 1556/526, loss: 0.04984130710363388 2023-01-21 09:38:11.222265: step: 1560/526, loss: 0.05042486637830734 2023-01-21 09:38:12.346244: step: 1564/526, loss: 0.7666254043579102 2023-01-21 09:38:13.495150: step: 1568/526, loss: 0.09522084891796112 2023-01-21 09:38:14.656044: step: 1572/526, loss: 0.21847449243068695 2023-01-21 09:38:15.823872: step: 1576/526, loss: 0.14648428559303284 2023-01-21 09:38:16.946100: step: 1580/526, loss: 0.2526691257953644 2023-01-21 09:38:18.086653: step: 1584/526, loss: 0.2150009274482727 2023-01-21 09:38:19.204742: step: 1588/526, loss: 0.10367265343666077 2023-01-21 09:38:20.338654: step: 1592/526, loss: 0.07769985496997833 2023-01-21 09:38:21.489247: step: 1596/526, loss: 0.08559217303991318 2023-01-21 09:38:22.612820: step: 1600/526, loss: 0.66338711977005 2023-01-21 09:38:23.732379: step: 1604/526, loss: 0.04394664987921715 2023-01-21 09:38:24.881789: step: 1608/526, loss: 0.2186059057712555 2023-01-21 09:38:26.015272: step: 1612/526, loss: 0.14132104814052582 2023-01-21 09:38:27.144715: step: 1616/526, loss: 0.013515759259462357 2023-01-21 09:38:28.296679: step: 1620/526, loss: 2.7745673656463623 2023-01-21 09:38:29.426530: step: 1624/526, loss: 0.35903164744377136 2023-01-21 09:38:30.565908: step: 1628/526, loss: 0.06225328892469406 2023-01-21 09:38:31.701766: step: 1632/526, loss: 0.09368248283863068 2023-01-21 09:38:32.846037: step: 1636/526, loss: 0.11353512108325958 2023-01-21 09:38:34.015314: step: 1640/526, loss: 0.4567698538303375 2023-01-21 09:38:35.186334: step: 1644/526, loss: 0.34618473052978516 2023-01-21 09:38:36.336676: step: 1648/526, loss: 0.5828515291213989 2023-01-21 09:38:37.475642: step: 1652/526, loss: 0.4022310972213745 2023-01-21 09:38:38.639917: step: 1656/526, loss: 0.18383732438087463 2023-01-21 09:38:39.761861: step: 1660/526, loss: 0.02237219735980034 2023-01-21 09:38:40.872109: step: 1664/526, loss: 0.028775788843631744 2023-01-21 09:38:42.012688: step: 1668/526, loss: 0.5737066268920898 2023-01-21 09:38:43.166325: step: 1672/526, loss: 5.0599565505981445 2023-01-21 09:38:44.283442: step: 1676/526, loss: 0.12092189490795135 2023-01-21 09:38:45.432938: step: 1680/526, loss: 0.0936523973941803 2023-01-21 09:38:46.573911: step: 1684/526, loss: 0.6087787747383118 2023-01-21 09:38:47.719837: step: 1688/526, loss: 0.5766590237617493 2023-01-21 09:38:48.838986: step: 1692/526, loss: 0.07612409442663193 2023-01-21 09:38:49.967380: step: 1696/526, loss: 0.1062956377863884 2023-01-21 09:38:51.107788: step: 1700/526, loss: 0.13765287399291992 2023-01-21 09:38:52.258529: step: 1704/526, loss: 2.2298364639282227 2023-01-21 09:38:53.403007: step: 1708/526, loss: 0.1256028264760971 2023-01-21 09:38:54.550623: step: 1712/526, loss: 0.12371645867824554 2023-01-21 09:38:55.674311: step: 1716/526, loss: 0.12302427738904953 2023-01-21 09:38:56.831010: step: 1720/526, loss: 0.13936057686805725 2023-01-21 09:38:57.962995: step: 1724/526, loss: 0.2171371579170227 2023-01-21 09:38:59.095657: step: 1728/526, loss: 0.07089905440807343 2023-01-21 09:39:00.231657: step: 1732/526, loss: 0.12172970920801163 2023-01-21 09:39:01.400179: step: 1736/526, loss: 0.1670808494091034 2023-01-21 09:39:02.534550: step: 1740/526, loss: 0.23992501199245453 2023-01-21 09:39:03.659688: step: 1744/526, loss: 0.1299276351928711 2023-01-21 09:39:04.851568: step: 1748/526, loss: 0.0321536548435688 2023-01-21 09:39:05.971007: step: 1752/526, loss: 0.0880836546421051 2023-01-21 09:39:07.106035: step: 1756/526, loss: 0.17066097259521484 2023-01-21 09:39:08.271599: step: 1760/526, loss: 0.21126055717468262 2023-01-21 09:39:09.407706: step: 1764/526, loss: 0.19855758547782898 2023-01-21 09:39:10.522802: step: 1768/526, loss: 0.4041768014431 2023-01-21 09:39:11.673355: step: 1772/526, loss: 0.08282260596752167 2023-01-21 09:39:12.793718: step: 1776/526, loss: 0.17709942162036896 2023-01-21 09:39:13.945013: step: 1780/526, loss: 0.9613229036331177 2023-01-21 09:39:15.084017: step: 1784/526, loss: 0.05794067680835724 2023-01-21 09:39:16.226209: step: 1788/526, loss: 0.08860006183385849 2023-01-21 09:39:17.355090: step: 1792/526, loss: 1.0366246700286865 2023-01-21 09:39:18.517476: step: 1796/526, loss: 0.26304614543914795 2023-01-21 09:39:19.678576: step: 1800/526, loss: 0.2124791443347931 2023-01-21 09:39:20.822150: step: 1804/526, loss: 0.23847246170043945 2023-01-21 09:39:21.969480: step: 1808/526, loss: 0.06172237545251846 2023-01-21 09:39:23.097790: step: 1812/526, loss: 0.3944447338581085 2023-01-21 09:39:24.213525: step: 1816/526, loss: 0.059876300394535065 2023-01-21 09:39:25.374758: step: 1820/526, loss: 0.08906774967908859 2023-01-21 09:39:26.500438: step: 1824/526, loss: 0.033185675740242004 2023-01-21 09:39:27.658642: step: 1828/526, loss: 0.15181049704551697 2023-01-21 09:39:28.814996: step: 1832/526, loss: 0.04933585971593857 2023-01-21 09:39:29.951054: step: 1836/526, loss: 0.14813023805618286 2023-01-21 09:39:31.080723: step: 1840/526, loss: 0.17138242721557617 2023-01-21 09:39:32.268601: step: 1844/526, loss: 0.23095454275608063 2023-01-21 09:39:33.409549: step: 1848/526, loss: 0.1301533728837967 2023-01-21 09:39:34.534734: step: 1852/526, loss: 0.10752621293067932 2023-01-21 09:39:35.664905: step: 1856/526, loss: 0.15956458449363708 2023-01-21 09:39:36.823319: step: 1860/526, loss: 0.20112381875514984 2023-01-21 09:39:37.959263: step: 1864/526, loss: 0.11384125053882599 2023-01-21 09:39:39.105713: step: 1868/526, loss: 0.18870334327220917 2023-01-21 09:39:40.246189: step: 1872/526, loss: 0.186344712972641 2023-01-21 09:39:41.397433: step: 1876/526, loss: 0.120759516954422 2023-01-21 09:39:42.537824: step: 1880/526, loss: 1.128904104232788 2023-01-21 09:39:43.699990: step: 1884/526, loss: 0.05447559431195259 2023-01-21 09:39:44.823828: step: 1888/526, loss: 0.2696797549724579 2023-01-21 09:39:45.954198: step: 1892/526, loss: 0.07769393920898438 2023-01-21 09:39:47.081581: step: 1896/526, loss: 0.1314132809638977 2023-01-21 09:39:48.216692: step: 1900/526, loss: 0.613822340965271 2023-01-21 09:39:49.362827: step: 1904/526, loss: 0.04486870765686035 2023-01-21 09:39:50.521942: step: 1908/526, loss: 1.2491724491119385 2023-01-21 09:39:51.663027: step: 1912/526, loss: 0.535128116607666 2023-01-21 09:39:52.824078: step: 1916/526, loss: 0.37151986360549927 2023-01-21 09:39:53.964022: step: 1920/526, loss: 0.7147814035415649 2023-01-21 09:39:55.121580: step: 1924/526, loss: 0.306810200214386 2023-01-21 09:39:56.267482: step: 1928/526, loss: 0.019554901868104935 2023-01-21 09:39:57.383475: step: 1932/526, loss: 0.32457178831100464 2023-01-21 09:39:58.501112: step: 1936/526, loss: 0.17129212617874146 2023-01-21 09:39:59.633714: step: 1940/526, loss: 0.08126652240753174 2023-01-21 09:40:00.780596: step: 1944/526, loss: 0.10410746186971664 2023-01-21 09:40:01.916797: step: 1948/526, loss: 0.18546819686889648 2023-01-21 09:40:03.050750: step: 1952/526, loss: 0.10252432525157928 2023-01-21 09:40:04.182225: step: 1956/526, loss: 0.501430094242096 2023-01-21 09:40:05.318096: step: 1960/526, loss: 0.17966899275779724 2023-01-21 09:40:06.465367: step: 1964/526, loss: 0.09827251732349396 2023-01-21 09:40:07.601614: step: 1968/526, loss: 0.06325512379407883 2023-01-21 09:40:08.759330: step: 1972/526, loss: 0.04881734773516655 2023-01-21 09:40:09.899985: step: 1976/526, loss: 0.30985507369041443 2023-01-21 09:40:11.062121: step: 1980/526, loss: 0.16137337684631348 2023-01-21 09:40:12.187409: step: 1984/526, loss: 0.6488516926765442 2023-01-21 09:40:13.355256: step: 1988/526, loss: 0.1319808065891266 2023-01-21 09:40:14.490184: step: 1992/526, loss: 0.09083929657936096 2023-01-21 09:40:15.647696: step: 1996/526, loss: 0.10225944966077805 2023-01-21 09:40:16.795019: step: 2000/526, loss: 0.041106320917606354 2023-01-21 09:40:17.924751: step: 2004/526, loss: 0.1926548033952713 2023-01-21 09:40:19.074952: step: 2008/526, loss: 0.1910114288330078 2023-01-21 09:40:20.206058: step: 2012/526, loss: 0.12288536876440048 2023-01-21 09:40:21.366662: step: 2016/526, loss: 0.02734251134097576 2023-01-21 09:40:22.516909: step: 2020/526, loss: 2.473437547683716 2023-01-21 09:40:23.685961: step: 2024/526, loss: 0.04134635999798775 2023-01-21 09:40:24.822018: step: 2028/526, loss: 0.2366454154253006 2023-01-21 09:40:25.965903: step: 2032/526, loss: 0.06661701202392578 2023-01-21 09:40:27.123919: step: 2036/526, loss: 0.148019939661026 2023-01-21 09:40:28.260200: step: 2040/526, loss: 0.030071640387177467 2023-01-21 09:40:29.409207: step: 2044/526, loss: 0.1777830272912979 2023-01-21 09:40:30.557740: step: 2048/526, loss: 0.21318073570728302 2023-01-21 09:40:31.678068: step: 2052/526, loss: 0.06619720160961151 2023-01-21 09:40:32.829014: step: 2056/526, loss: 0.07484984397888184 2023-01-21 09:40:33.966715: step: 2060/526, loss: 0.02998533472418785 2023-01-21 09:40:35.093650: step: 2064/526, loss: 0.03785824775695801 2023-01-21 09:40:36.241869: step: 2068/526, loss: 0.08734522014856339 2023-01-21 09:40:37.379750: step: 2072/526, loss: 0.13103142380714417 2023-01-21 09:40:38.523983: step: 2076/526, loss: 0.1434842050075531 2023-01-21 09:40:39.647344: step: 2080/526, loss: 0.03428039699792862 2023-01-21 09:40:40.805761: step: 2084/526, loss: 0.03596749156713486 2023-01-21 09:40:41.955905: step: 2088/526, loss: 1.0822758674621582 2023-01-21 09:40:43.091058: step: 2092/526, loss: 0.0823885053396225 2023-01-21 09:40:44.205350: step: 2096/526, loss: 0.06300631165504456 2023-01-21 09:40:45.384661: step: 2100/526, loss: 0.11963353306055069 2023-01-21 09:40:46.527217: step: 2104/526, loss: 0.05372738838195801 ================================================== Loss: 0.268 -------------------- Dev: {'event': {'p': 0.5482954545454546, 'r': 0.7709720372836218, 'f1': 0.6408411732152739}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Test: {'event': {'p': 0.584652862362972, 'r': 0.7515657620041754, 'f1': 0.6576844028316966}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Korean: {'event': {'p': 0.5689655172413793, 'r': 0.5238095238095238, 'f1': 0.5454545454545455}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} Russian: {'event': {'p': 0.3902439024390244, 'r': 0.4444444444444444, 'f1': 0.4155844155844156}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 5} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Korean: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Korean: {'event': {'p': 0.5932203389830508, 'r': 0.5555555555555556, 'f1': 0.5737704918032787}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} -------------------- Eng Dev for Russian: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5116279069767442, 'r': 0.6111111111111112, 'f1': 0.5569620253164557}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 6 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:41:30.030034: step: 4/526, loss: 0.1935199648141861 2023-01-21 09:41:31.144786: step: 8/526, loss: 0.053366661071777344 2023-01-21 09:41:32.284080: step: 12/526, loss: 0.09214058518409729 2023-01-21 09:41:33.426060: step: 16/526, loss: 0.05108308792114258 2023-01-21 09:41:34.559287: step: 20/526, loss: 0.16564340889453888 2023-01-21 09:41:35.693876: step: 24/526, loss: 0.12585657835006714 2023-01-21 09:41:36.851808: step: 28/526, loss: 0.0700259655714035 2023-01-21 09:41:37.975976: step: 32/526, loss: 0.04057908058166504 2023-01-21 09:41:39.137799: step: 36/526, loss: 0.27317723631858826 2023-01-21 09:41:40.321565: step: 40/526, loss: 0.10032758116722107 2023-01-21 09:41:41.465653: step: 44/526, loss: 0.1474703848361969 2023-01-21 09:41:42.595472: step: 48/526, loss: 0.10151688009500504 2023-01-21 09:41:43.731146: step: 52/526, loss: 0.06741821765899658 2023-01-21 09:41:44.878612: step: 56/526, loss: 0.08254604786634445 2023-01-21 09:41:46.009157: step: 60/526, loss: 0.09099225699901581 2023-01-21 09:41:47.123947: step: 64/526, loss: 0.13013902306556702 2023-01-21 09:41:48.251829: step: 68/526, loss: 0.1279090940952301 2023-01-21 09:41:49.406916: step: 72/526, loss: 0.3624299168586731 2023-01-21 09:41:50.527164: step: 76/526, loss: 1.7218503952026367 2023-01-21 09:41:51.690856: step: 80/526, loss: 0.17056140303611755 2023-01-21 09:41:52.890021: step: 84/526, loss: 0.10572975873947144 2023-01-21 09:41:54.027120: step: 88/526, loss: 0.10263185948133469 2023-01-21 09:41:55.177937: step: 92/526, loss: 0.09498701244592667 2023-01-21 09:41:56.294742: step: 96/526, loss: 0.14327764511108398 2023-01-21 09:41:57.436483: step: 100/526, loss: 0.2205335646867752 2023-01-21 09:41:58.565825: step: 104/526, loss: 0.43257367610931396 2023-01-21 09:41:59.696758: step: 108/526, loss: 0.23544475436210632 2023-01-21 09:42:00.863810: step: 112/526, loss: 0.06714353710412979 2023-01-21 09:42:02.062270: step: 116/526, loss: 0.20405682921409607 2023-01-21 09:42:03.190783: step: 120/526, loss: 0.12117824703454971 2023-01-21 09:42:04.347848: step: 124/526, loss: 0.23634929955005646 2023-01-21 09:42:05.490614: step: 128/526, loss: 0.11167869716882706 2023-01-21 09:42:06.637462: step: 132/526, loss: 0.7614037990570068 2023-01-21 09:42:07.789129: step: 136/526, loss: 0.0993494987487793 2023-01-21 09:42:08.953281: step: 140/526, loss: 0.04996833950281143 2023-01-21 09:42:10.102894: step: 144/526, loss: 0.14071054756641388 2023-01-21 09:42:11.217352: step: 148/526, loss: 0.04061935096979141 2023-01-21 09:42:12.370100: step: 152/526, loss: 0.4921053647994995 2023-01-21 09:42:13.494097: step: 156/526, loss: 0.3915863037109375 2023-01-21 09:42:14.657304: step: 160/526, loss: 0.047487977892160416 2023-01-21 09:42:15.790081: step: 164/526, loss: 0.12497777491807938 2023-01-21 09:42:16.920263: step: 168/526, loss: 0.031023599207401276 2023-01-21 09:42:18.077698: step: 172/526, loss: 0.061802055686712265 2023-01-21 09:42:19.255942: step: 176/526, loss: 0.12532693147659302 2023-01-21 09:42:20.407137: step: 180/526, loss: 0.08658051490783691 2023-01-21 09:42:21.560681: step: 184/526, loss: 0.08162536472082138 2023-01-21 09:42:22.671179: step: 188/526, loss: 0.12107090651988983 2023-01-21 09:42:23.860965: step: 192/526, loss: 0.33039236068725586 2023-01-21 09:42:24.993004: step: 196/526, loss: 0.4681624174118042 2023-01-21 09:42:26.145109: step: 200/526, loss: 0.19839076697826385 2023-01-21 09:42:27.268016: step: 204/526, loss: 0.13721294701099396 2023-01-21 09:42:28.413245: step: 208/526, loss: 0.1353965401649475 2023-01-21 09:42:29.558331: step: 212/526, loss: 0.05025773122906685 2023-01-21 09:42:30.709638: step: 216/526, loss: 0.8137813806533813 2023-01-21 09:42:31.831824: step: 220/526, loss: 0.05085926130414009 2023-01-21 09:42:32.964366: step: 224/526, loss: 0.08609800785779953 2023-01-21 09:42:34.118681: step: 228/526, loss: 0.041141461580991745 2023-01-21 09:42:35.245522: step: 232/526, loss: 0.13425445556640625 2023-01-21 09:42:36.402819: step: 236/526, loss: 0.17219281196594238 2023-01-21 09:42:37.523137: step: 240/526, loss: 0.28969812393188477 2023-01-21 09:42:38.654250: step: 244/526, loss: 0.12828321754932404 2023-01-21 09:42:39.765590: step: 248/526, loss: 0.21680259704589844 2023-01-21 09:42:40.884267: step: 252/526, loss: 0.5963222980499268 2023-01-21 09:42:42.011320: step: 256/526, loss: 0.14127665758132935 2023-01-21 09:42:43.164804: step: 260/526, loss: 0.40478143095970154 2023-01-21 09:42:44.275625: step: 264/526, loss: 0.022491455078125 2023-01-21 09:42:45.740140: step: 268/526, loss: 0.14795736968517303 2023-01-21 09:42:46.855074: step: 272/526, loss: 0.2272837609052658 2023-01-21 09:42:48.011407: step: 276/526, loss: 0.14946284890174866 2023-01-21 09:42:49.174483: step: 280/526, loss: 0.09964227676391602 2023-01-21 09:42:50.303922: step: 284/526, loss: 0.14771641790866852 2023-01-21 09:42:51.466699: step: 288/526, loss: 0.03897056728601456 2023-01-21 09:42:52.601369: step: 292/526, loss: 0.29770296812057495 2023-01-21 09:42:53.714612: step: 296/526, loss: 0.21242466568946838 2023-01-21 09:42:54.855720: step: 300/526, loss: 0.041317179799079895 2023-01-21 09:42:55.989077: step: 304/526, loss: 0.0701996311545372 2023-01-21 09:42:57.122451: step: 308/526, loss: 0.29508471488952637 2023-01-21 09:42:58.243312: step: 312/526, loss: 0.29923582077026367 2023-01-21 09:42:59.360012: step: 316/526, loss: 0.04200158268213272 2023-01-21 09:43:00.468604: step: 320/526, loss: 0.3863620162010193 2023-01-21 09:43:01.583722: step: 324/526, loss: 0.0825316458940506 2023-01-21 09:43:02.710447: step: 328/526, loss: 0.21170654892921448 2023-01-21 09:43:03.874222: step: 332/526, loss: 0.13917456567287445 2023-01-21 09:43:05.025843: step: 336/526, loss: 0.28148922324180603 2023-01-21 09:43:06.164112: step: 340/526, loss: 0.10612602531909943 2023-01-21 09:43:07.277039: step: 344/526, loss: 0.03621101379394531 2023-01-21 09:43:08.392918: step: 348/526, loss: 0.04159374162554741 2023-01-21 09:43:09.536979: step: 352/526, loss: 0.09239482879638672 2023-01-21 09:43:10.699881: step: 356/526, loss: 0.2786267399787903 2023-01-21 09:43:11.843093: step: 360/526, loss: 0.019766712561249733 2023-01-21 09:43:13.003182: step: 364/526, loss: 0.08411598205566406 2023-01-21 09:43:14.132940: step: 368/526, loss: 0.12203660607337952 2023-01-21 09:43:15.291241: step: 372/526, loss: 0.2746692895889282 2023-01-21 09:43:16.455708: step: 376/526, loss: 0.030750369653105736 2023-01-21 09:43:17.575553: step: 380/526, loss: 0.08497676998376846 2023-01-21 09:43:18.712238: step: 384/526, loss: 0.04459733888506889 2023-01-21 09:43:19.833128: step: 388/526, loss: 0.06859102100133896 2023-01-21 09:43:20.967565: step: 392/526, loss: 0.08576574921607971 2023-01-21 09:43:22.110296: step: 396/526, loss: 0.2812884449958801 2023-01-21 09:43:23.239402: step: 400/526, loss: 0.39120596647262573 2023-01-21 09:43:24.354116: step: 404/526, loss: 0.03319845348596573 2023-01-21 09:43:25.484930: step: 408/526, loss: 0.31789249181747437 2023-01-21 09:43:26.616656: step: 412/526, loss: 0.16425448656082153 2023-01-21 09:43:27.757169: step: 416/526, loss: 0.07633178681135178 2023-01-21 09:43:28.935710: step: 420/526, loss: 0.1328599900007248 2023-01-21 09:43:30.057282: step: 424/526, loss: 0.17225494980812073 2023-01-21 09:43:31.191013: step: 428/526, loss: 0.024799346923828125 2023-01-21 09:43:32.335593: step: 432/526, loss: 0.0743323341012001 2023-01-21 09:43:33.453736: step: 436/526, loss: 0.048112429678440094 2023-01-21 09:43:34.606321: step: 440/526, loss: 0.10075311362743378 2023-01-21 09:43:35.772039: step: 444/526, loss: 0.06957521289587021 2023-01-21 09:43:36.916076: step: 448/526, loss: 0.05378551781177521 2023-01-21 09:43:38.030443: step: 452/526, loss: 0.01987457275390625 2023-01-21 09:43:39.152257: step: 456/526, loss: 0.34529200196266174 2023-01-21 09:43:40.250065: step: 460/526, loss: 0.1810527890920639 2023-01-21 09:43:41.365012: step: 464/526, loss: 0.18962498009204865 2023-01-21 09:43:42.470758: step: 468/526, loss: 0.27324995398521423 2023-01-21 09:43:43.626982: step: 472/526, loss: 0.1481439620256424 2023-01-21 09:43:44.781147: step: 476/526, loss: 0.0918925330042839 2023-01-21 09:43:45.921235: step: 480/526, loss: 0.0877266451716423 2023-01-21 09:43:47.070958: step: 484/526, loss: 0.06493854522705078 2023-01-21 09:43:48.214658: step: 488/526, loss: 0.03394820913672447 2023-01-21 09:43:49.355632: step: 492/526, loss: 0.08866987377405167 2023-01-21 09:43:50.483925: step: 496/526, loss: 0.14991122484207153 2023-01-21 09:43:51.626232: step: 500/526, loss: 0.20004349946975708 2023-01-21 09:43:52.849761: step: 504/526, loss: 0.0679527297616005 2023-01-21 09:43:53.986290: step: 508/526, loss: 0.10013322532176971 2023-01-21 09:43:55.106577: step: 512/526, loss: 0.21924495697021484 2023-01-21 09:43:56.222106: step: 516/526, loss: 0.060643672943115234 2023-01-21 09:43:57.342601: step: 520/526, loss: 0.01569042168557644 2023-01-21 09:43:58.535812: step: 524/526, loss: 0.16761493682861328 2023-01-21 09:43:59.712341: step: 528/526, loss: 0.02835388109087944 2023-01-21 09:44:00.851919: step: 532/526, loss: 0.19586822390556335 2023-01-21 09:44:01.999299: step: 536/526, loss: 0.03471088409423828 2023-01-21 09:44:03.171638: step: 540/526, loss: 0.20807743072509766 2023-01-21 09:44:04.330521: step: 544/526, loss: 0.028271447867155075 2023-01-21 09:44:05.471035: step: 548/526, loss: 0.08852767944335938 2023-01-21 09:44:06.588338: step: 552/526, loss: 0.11253197491168976 2023-01-21 09:44:07.714188: step: 556/526, loss: 0.13039885461330414 2023-01-21 09:44:08.851632: step: 560/526, loss: 0.23515664041042328 2023-01-21 09:44:09.982459: step: 564/526, loss: 0.1011715903878212 2023-01-21 09:44:11.147009: step: 568/526, loss: 0.14560852944850922 2023-01-21 09:44:12.281129: step: 572/526, loss: 0.5282018184661865 2023-01-21 09:44:13.439154: step: 576/526, loss: 0.0902009904384613 2023-01-21 09:44:14.568373: step: 580/526, loss: 0.008965587243437767 2023-01-21 09:44:15.657296: step: 584/526, loss: 0.08957748860120773 2023-01-21 09:44:16.804561: step: 588/526, loss: 0.3128474950790405 2023-01-21 09:44:17.929589: step: 592/526, loss: 0.027715492993593216 2023-01-21 09:44:19.065561: step: 596/526, loss: 0.05699920654296875 2023-01-21 09:44:20.205149: step: 600/526, loss: 0.04225330427289009 2023-01-21 09:44:21.339928: step: 604/526, loss: 0.06309624016284943 2023-01-21 09:44:22.478257: step: 608/526, loss: 0.15198016166687012 2023-01-21 09:44:23.615696: step: 612/526, loss: 0.30850622057914734 2023-01-21 09:44:24.738606: step: 616/526, loss: 0.03823041915893555 2023-01-21 09:44:25.863552: step: 620/526, loss: 0.45894908905029297 2023-01-21 09:44:26.983083: step: 624/526, loss: 0.16934317350387573 2023-01-21 09:44:28.130439: step: 628/526, loss: 0.23666810989379883 2023-01-21 09:44:29.266656: step: 632/526, loss: 0.02895507961511612 2023-01-21 09:44:30.392910: step: 636/526, loss: 0.10448131710290909 2023-01-21 09:44:31.544969: step: 640/526, loss: 0.14572878181934357 2023-01-21 09:44:32.660794: step: 644/526, loss: 0.3547572195529938 2023-01-21 09:44:33.817273: step: 648/526, loss: 0.010396289639174938 2023-01-21 09:44:34.949594: step: 652/526, loss: 0.12235183268785477 2023-01-21 09:44:36.109028: step: 656/526, loss: 0.03808736801147461 2023-01-21 09:44:37.234899: step: 660/526, loss: 0.10837996006011963 2023-01-21 09:44:38.375079: step: 664/526, loss: 0.10509152710437775 2023-01-21 09:44:39.497015: step: 668/526, loss: 0.37106457352638245 2023-01-21 09:44:40.653590: step: 672/526, loss: 0.08305959403514862 2023-01-21 09:44:41.802320: step: 676/526, loss: 0.3925187289714813 2023-01-21 09:44:42.975657: step: 680/526, loss: 0.2616066336631775 2023-01-21 09:44:44.130152: step: 684/526, loss: 0.0772009864449501 2023-01-21 09:44:45.275222: step: 688/526, loss: 1.1067917346954346 2023-01-21 09:44:46.417482: step: 692/526, loss: 0.03790278360247612 2023-01-21 09:44:47.574443: step: 696/526, loss: 0.04826851189136505 2023-01-21 09:44:48.702934: step: 700/526, loss: 0.21131621301174164 2023-01-21 09:44:49.830979: step: 704/526, loss: 0.3306438624858856 2023-01-21 09:44:50.955165: step: 708/526, loss: 0.398441880941391 2023-01-21 09:44:52.123738: step: 712/526, loss: 0.12625399231910706 2023-01-21 09:44:53.245583: step: 716/526, loss: 0.06526875495910645 2023-01-21 09:44:54.376929: step: 720/526, loss: 0.09623154997825623 2023-01-21 09:44:55.511191: step: 724/526, loss: 0.05037336423993111 2023-01-21 09:44:56.661202: step: 728/526, loss: 0.0637001097202301 2023-01-21 09:44:57.800488: step: 732/526, loss: 1.1518621444702148 2023-01-21 09:44:58.944967: step: 736/526, loss: 0.1728355586528778 2023-01-21 09:45:00.142425: step: 740/526, loss: 0.039498139172792435 2023-01-21 09:45:01.276400: step: 744/526, loss: 0.14737872779369354 2023-01-21 09:45:02.401887: step: 748/526, loss: 0.3067273199558258 2023-01-21 09:45:03.520166: step: 752/526, loss: 0.17174334824085236 2023-01-21 09:45:04.678068: step: 756/526, loss: 0.06000871583819389 2023-01-21 09:45:05.803152: step: 760/526, loss: 0.22411632537841797 2023-01-21 09:45:06.930622: step: 764/526, loss: 0.08788815140724182 2023-01-21 09:45:08.052804: step: 768/526, loss: 0.15324965119361877 2023-01-21 09:45:09.183911: step: 772/526, loss: 0.08564749360084534 2023-01-21 09:45:10.325688: step: 776/526, loss: 0.08029551804065704 2023-01-21 09:45:11.495866: step: 780/526, loss: 0.22256916761398315 2023-01-21 09:45:12.625944: step: 784/526, loss: 0.04142565652728081 2023-01-21 09:45:13.757281: step: 788/526, loss: 0.013770533725619316 2023-01-21 09:45:14.855570: step: 792/526, loss: 0.1091160774230957 2023-01-21 09:45:15.968187: step: 796/526, loss: 0.17877742648124695 2023-01-21 09:45:17.116757: step: 800/526, loss: 0.23655076324939728 2023-01-21 09:45:18.247878: step: 804/526, loss: 0.151583731174469 2023-01-21 09:45:19.392019: step: 808/526, loss: 0.0842476338148117 2023-01-21 09:45:20.530539: step: 812/526, loss: 0.12497062981128693 2023-01-21 09:45:21.637835: step: 816/526, loss: 0.04380984231829643 2023-01-21 09:45:22.785607: step: 820/526, loss: 0.16521388292312622 2023-01-21 09:45:23.938917: step: 824/526, loss: 0.08629884570837021 2023-01-21 09:45:25.095474: step: 828/526, loss: 0.08899412304162979 2023-01-21 09:45:26.226715: step: 832/526, loss: 0.25268885493278503 2023-01-21 09:45:27.360427: step: 836/526, loss: 0.04107518121600151 2023-01-21 09:45:28.498206: step: 840/526, loss: 0.17141857743263245 2023-01-21 09:45:29.622285: step: 844/526, loss: 0.056225232779979706 2023-01-21 09:45:30.747051: step: 848/526, loss: 0.21195407211780548 2023-01-21 09:45:31.901439: step: 852/526, loss: 0.13661302626132965 2023-01-21 09:45:33.031768: step: 856/526, loss: 0.11779718101024628 2023-01-21 09:45:34.160131: step: 860/526, loss: 0.10724586993455887 2023-01-21 09:45:35.303155: step: 864/526, loss: 0.17789001762866974 2023-01-21 09:45:36.422982: step: 868/526, loss: 0.07723245769739151 2023-01-21 09:45:37.577562: step: 872/526, loss: 0.12877464294433594 2023-01-21 09:45:38.721609: step: 876/526, loss: 0.1516997516155243 2023-01-21 09:45:39.863082: step: 880/526, loss: 0.08661232143640518 2023-01-21 09:45:41.010089: step: 884/526, loss: 0.20451012253761292 2023-01-21 09:45:42.144170: step: 888/526, loss: 0.031790636479854584 2023-01-21 09:45:43.338488: step: 892/526, loss: 0.2826294004917145 2023-01-21 09:45:44.486227: step: 896/526, loss: 0.3116837739944458 2023-01-21 09:45:45.627506: step: 900/526, loss: 0.020956754684448242 2023-01-21 09:45:46.756564: step: 904/526, loss: 0.15053005516529083 2023-01-21 09:45:47.881912: step: 908/526, loss: 0.044760704040527344 2023-01-21 09:45:49.037603: step: 912/526, loss: 0.4009353518486023 2023-01-21 09:45:50.187755: step: 916/526, loss: 0.4486294686794281 2023-01-21 09:45:51.299481: step: 920/526, loss: 0.06704731285572052 2023-01-21 09:45:52.420185: step: 924/526, loss: 0.08138275146484375 2023-01-21 09:45:53.523701: step: 928/526, loss: 0.060701943933963776 2023-01-21 09:45:54.622700: step: 932/526, loss: 0.17838707566261292 2023-01-21 09:45:55.769233: step: 936/526, loss: 0.1150943711400032 2023-01-21 09:45:56.870712: step: 940/526, loss: 0.08999890834093094 2023-01-21 09:45:57.974745: step: 944/526, loss: 0.07424888759851456 2023-01-21 09:45:59.108585: step: 948/526, loss: 0.10326433181762695 2023-01-21 09:46:00.259335: step: 952/526, loss: 0.1869238018989563 2023-01-21 09:46:01.378031: step: 956/526, loss: 0.13016529381275177 2023-01-21 09:46:02.544395: step: 960/526, loss: 0.0988527312874794 2023-01-21 09:46:03.684910: step: 964/526, loss: 0.23519906401634216 2023-01-21 09:46:04.794004: step: 968/526, loss: 0.08537960052490234 2023-01-21 09:46:05.946868: step: 972/526, loss: 0.1011384055018425 2023-01-21 09:46:07.097306: step: 976/526, loss: 0.04553098976612091 2023-01-21 09:46:08.258328: step: 980/526, loss: 0.0717797800898552 2023-01-21 09:46:09.407063: step: 984/526, loss: 0.18373413383960724 2023-01-21 09:46:10.531978: step: 988/526, loss: 0.11869249492883682 2023-01-21 09:46:11.658754: step: 992/526, loss: 0.07695799320936203 2023-01-21 09:46:12.858298: step: 996/526, loss: 0.034386828541755676 2023-01-21 09:46:13.975486: step: 1000/526, loss: 0.17930403351783752 2023-01-21 09:46:15.139921: step: 1004/526, loss: 1.1596510410308838 2023-01-21 09:46:16.265860: step: 1008/526, loss: 0.06212444230914116 2023-01-21 09:46:17.385257: step: 1012/526, loss: 0.06146228685975075 2023-01-21 09:46:18.521583: step: 1016/526, loss: 0.029346086084842682 2023-01-21 09:46:19.653144: step: 1020/526, loss: 0.27891793847084045 2023-01-21 09:46:20.785975: step: 1024/526, loss: 0.8286097049713135 2023-01-21 09:46:21.929014: step: 1028/526, loss: 0.1077735424041748 2023-01-21 09:46:23.081744: step: 1032/526, loss: 0.3889457583427429 2023-01-21 09:46:24.227143: step: 1036/526, loss: 0.10716943442821503 2023-01-21 09:46:25.363318: step: 1040/526, loss: 0.05354709550738335 2023-01-21 09:46:26.516108: step: 1044/526, loss: 0.0801825076341629 2023-01-21 09:46:27.634827: step: 1048/526, loss: 0.24855437874794006 2023-01-21 09:46:28.746887: step: 1052/526, loss: 0.15052232146263123 2023-01-21 09:46:29.869527: step: 1056/526, loss: 0.26865842938423157 2023-01-21 09:46:30.982510: step: 1060/526, loss: 0.10129585862159729 2023-01-21 09:46:32.139621: step: 1064/526, loss: 0.2093355655670166 2023-01-21 09:46:33.298450: step: 1068/526, loss: 0.19879885017871857 2023-01-21 09:46:34.455023: step: 1072/526, loss: 0.2345268428325653 2023-01-21 09:46:35.582699: step: 1076/526, loss: 0.23189297318458557 2023-01-21 09:46:36.714810: step: 1080/526, loss: 0.2789713442325592 2023-01-21 09:46:37.863691: step: 1084/526, loss: 0.09879694133996964 2023-01-21 09:46:39.009228: step: 1088/526, loss: 0.1386704444885254 2023-01-21 09:46:40.138115: step: 1092/526, loss: 0.1418129950761795 2023-01-21 09:46:41.295596: step: 1096/526, loss: 0.19827881455421448 2023-01-21 09:46:42.424426: step: 1100/526, loss: 0.03734779357910156 2023-01-21 09:46:43.527718: step: 1104/526, loss: 0.16050490736961365 2023-01-21 09:46:44.675200: step: 1108/526, loss: 0.06221742555499077 2023-01-21 09:46:45.809158: step: 1112/526, loss: 0.11807861179113388 2023-01-21 09:46:46.961975: step: 1116/526, loss: 0.1265970766544342 2023-01-21 09:46:48.084830: step: 1120/526, loss: 0.05973844602704048 2023-01-21 09:46:49.207588: step: 1124/526, loss: 0.04374570772051811 2023-01-21 09:46:50.329052: step: 1128/526, loss: 0.40221673250198364 2023-01-21 09:46:51.469909: step: 1132/526, loss: 0.057587288320064545 2023-01-21 09:46:52.630937: step: 1136/526, loss: 0.07253055274486542 2023-01-21 09:46:53.795076: step: 1140/526, loss: 0.5135291814804077 2023-01-21 09:46:54.928378: step: 1144/526, loss: 0.0750647485256195 2023-01-21 09:46:56.050999: step: 1148/526, loss: 0.23290835320949554 2023-01-21 09:46:57.169191: step: 1152/526, loss: 0.030810022726655006 2023-01-21 09:46:58.290267: step: 1156/526, loss: 0.24512425065040588 2023-01-21 09:46:59.455693: step: 1160/526, loss: 0.09919628500938416 2023-01-21 09:47:00.567932: step: 1164/526, loss: 0.09027914702892303 2023-01-21 09:47:01.729827: step: 1168/526, loss: 0.04261169582605362 2023-01-21 09:47:02.922094: step: 1172/526, loss: 0.09574628621339798 2023-01-21 09:47:04.035871: step: 1176/526, loss: 0.13626104593276978 2023-01-21 09:47:05.209532: step: 1180/526, loss: 0.5341628193855286 2023-01-21 09:47:06.358877: step: 1184/526, loss: 0.7170593738555908 2023-01-21 09:47:07.484799: step: 1188/526, loss: 0.11014915257692337 2023-01-21 09:47:08.650441: step: 1192/526, loss: 0.2766742408275604 2023-01-21 09:47:09.835495: step: 1196/526, loss: 0.046143628656864166 2023-01-21 09:47:10.996086: step: 1200/526, loss: 0.07663488388061523 2023-01-21 09:47:12.133351: step: 1204/526, loss: 0.04668016731739044 2023-01-21 09:47:13.275934: step: 1208/526, loss: 0.07996115833520889 2023-01-21 09:47:14.416859: step: 1212/526, loss: 0.033505868166685104 2023-01-21 09:47:15.602503: step: 1216/526, loss: 0.3432506322860718 2023-01-21 09:47:16.754927: step: 1220/526, loss: 0.05837153270840645 2023-01-21 09:47:17.881903: step: 1224/526, loss: 0.04403328895568848 2023-01-21 09:47:19.015362: step: 1228/526, loss: 0.39112210273742676 2023-01-21 09:47:20.168697: step: 1232/526, loss: 0.13740229606628418 2023-01-21 09:47:21.328053: step: 1236/526, loss: 0.8640855550765991 2023-01-21 09:47:22.484717: step: 1240/526, loss: 0.11695460975170135 2023-01-21 09:47:23.683472: step: 1244/526, loss: 0.04489259794354439 2023-01-21 09:47:24.830460: step: 1248/526, loss: 0.09422874450683594 2023-01-21 09:47:25.947696: step: 1252/526, loss: 0.06230144202709198 2023-01-21 09:47:27.111628: step: 1256/526, loss: 1.684261679649353 2023-01-21 09:47:28.224053: step: 1260/526, loss: 0.05754508823156357 2023-01-21 09:47:29.341106: step: 1264/526, loss: 0.12799835205078125 2023-01-21 09:47:30.491001: step: 1268/526, loss: 0.04727745056152344 2023-01-21 09:47:31.638492: step: 1272/526, loss: 0.12582802772521973 2023-01-21 09:47:32.812741: step: 1276/526, loss: 0.23694530129432678 2023-01-21 09:47:33.946106: step: 1280/526, loss: 0.7765634059906006 2023-01-21 09:47:35.064000: step: 1284/526, loss: 0.07010293006896973 2023-01-21 09:47:36.207687: step: 1288/526, loss: 0.12948857247829437 2023-01-21 09:47:37.367481: step: 1292/526, loss: 1.0716722011566162 2023-01-21 09:47:38.491704: step: 1296/526, loss: 0.3529067933559418 2023-01-21 09:47:39.644954: step: 1300/526, loss: 0.08829350769519806 2023-01-21 09:47:40.744120: step: 1304/526, loss: 0.14155983924865723 2023-01-21 09:47:41.873282: step: 1308/526, loss: 0.05764905363321304 2023-01-21 09:47:43.047097: step: 1312/526, loss: 0.18648245930671692 2023-01-21 09:47:44.173572: step: 1316/526, loss: 0.16608795523643494 2023-01-21 09:47:45.296329: step: 1320/526, loss: 0.058429718017578125 2023-01-21 09:47:46.426707: step: 1324/526, loss: 1.1081345081329346 2023-01-21 09:47:47.588358: step: 1328/526, loss: 0.05601778253912926 2023-01-21 09:47:48.721996: step: 1332/526, loss: 0.1123056411743164 2023-01-21 09:47:49.900986: step: 1336/526, loss: 0.21761107444763184 2023-01-21 09:47:51.034796: step: 1340/526, loss: 0.12184090912342072 2023-01-21 09:47:52.167716: step: 1344/526, loss: 0.05934043228626251 2023-01-21 09:47:53.284537: step: 1348/526, loss: 0.3498401641845703 2023-01-21 09:47:54.429283: step: 1352/526, loss: 0.057191941887140274 2023-01-21 09:47:55.575694: step: 1356/526, loss: 0.1820422112941742 2023-01-21 09:47:56.718218: step: 1360/526, loss: 0.2146422266960144 2023-01-21 09:47:57.847465: step: 1364/526, loss: 0.06824059784412384 2023-01-21 09:47:58.957035: step: 1368/526, loss: 0.3996272087097168 2023-01-21 09:48:00.099207: step: 1372/526, loss: 0.13198700547218323 2023-01-21 09:48:01.258855: step: 1376/526, loss: 0.08759908378124237 2023-01-21 09:48:02.386465: step: 1380/526, loss: 0.09477319568395615 2023-01-21 09:48:03.503076: step: 1384/526, loss: 0.03702554851770401 2023-01-21 09:48:04.639654: step: 1388/526, loss: 0.040091514587402344 2023-01-21 09:48:05.784141: step: 1392/526, loss: 0.0537782683968544 2023-01-21 09:48:06.910091: step: 1396/526, loss: 0.30851611495018005 2023-01-21 09:48:08.049142: step: 1400/526, loss: 0.077253058552742 2023-01-21 09:48:09.153119: step: 1404/526, loss: 0.12993822991847992 2023-01-21 09:48:10.300132: step: 1408/526, loss: 0.027674105018377304 2023-01-21 09:48:11.435549: step: 1412/526, loss: 0.458151638507843 2023-01-21 09:48:12.584894: step: 1416/526, loss: 0.20200443267822266 2023-01-21 09:48:13.755939: step: 1420/526, loss: 0.08398628234863281 2023-01-21 09:48:14.901965: step: 1424/526, loss: 0.1619892120361328 2023-01-21 09:48:16.027045: step: 1428/526, loss: 0.18240490555763245 2023-01-21 09:48:17.149560: step: 1432/526, loss: 0.09315434098243713 2023-01-21 09:48:18.280564: step: 1436/526, loss: 0.14366689324378967 2023-01-21 09:48:19.454011: step: 1440/526, loss: 0.027073001489043236 2023-01-21 09:48:20.572833: step: 1444/526, loss: 0.03297267109155655 2023-01-21 09:48:21.684764: step: 1448/526, loss: 0.05766334757208824 2023-01-21 09:48:22.867749: step: 1452/526, loss: 0.11763410270214081 2023-01-21 09:48:24.024804: step: 1456/526, loss: 0.1841091513633728 2023-01-21 09:48:25.130802: step: 1460/526, loss: 0.09012174606323242 2023-01-21 09:48:26.254822: step: 1464/526, loss: 0.1229579895734787 2023-01-21 09:48:27.388956: step: 1468/526, loss: 0.22106781601905823 2023-01-21 09:48:28.549925: step: 1472/526, loss: 0.03912201151251793 2023-01-21 09:48:29.701387: step: 1476/526, loss: 0.08545112609863281 2023-01-21 09:48:30.822140: step: 1480/526, loss: 0.29759687185287476 2023-01-21 09:48:31.939834: step: 1484/526, loss: 0.22104902565479279 2023-01-21 09:48:33.074662: step: 1488/526, loss: 0.1259743571281433 2023-01-21 09:48:34.194127: step: 1492/526, loss: 0.01933879964053631 2023-01-21 09:48:35.318693: step: 1496/526, loss: 0.4715174436569214 2023-01-21 09:48:36.450461: step: 1500/526, loss: 0.7592087984085083 2023-01-21 09:48:37.578665: step: 1504/526, loss: 0.24064360558986664 2023-01-21 09:48:38.706165: step: 1508/526, loss: 0.22151318192481995 2023-01-21 09:48:39.818644: step: 1512/526, loss: 0.21759772300720215 2023-01-21 09:48:40.962024: step: 1516/526, loss: 0.689658522605896 2023-01-21 09:48:42.100999: step: 1520/526, loss: 0.1900985836982727 2023-01-21 09:48:43.284328: step: 1524/526, loss: 0.2347254753112793 2023-01-21 09:48:44.416632: step: 1528/526, loss: 0.027950286865234375 2023-01-21 09:48:45.562589: step: 1532/526, loss: 0.049568794667720795 2023-01-21 09:48:46.715648: step: 1536/526, loss: 0.20738163590431213 2023-01-21 09:48:47.867453: step: 1540/526, loss: 0.13126210868358612 2023-01-21 09:48:49.038832: step: 1544/526, loss: 0.04920470714569092 2023-01-21 09:48:50.205191: step: 1548/526, loss: 0.0811530128121376 2023-01-21 09:48:51.325510: step: 1552/526, loss: 0.13413038849830627 2023-01-21 09:48:52.489365: step: 1556/526, loss: 0.2211053967475891 2023-01-21 09:48:53.622503: step: 1560/526, loss: 0.11402948200702667 2023-01-21 09:48:54.727197: step: 1564/526, loss: 0.052236318588256836 2023-01-21 09:48:55.893129: step: 1568/526, loss: 0.18630047142505646 2023-01-21 09:48:57.027928: step: 1572/526, loss: 0.0687480941414833 2023-01-21 09:48:58.151792: step: 1576/526, loss: 0.03628816828131676 2023-01-21 09:48:59.266822: step: 1580/526, loss: 0.052163124084472656 2023-01-21 09:49:00.390155: step: 1584/526, loss: 0.015277600847184658 2023-01-21 09:49:01.525019: step: 1588/526, loss: 0.08992882072925568 2023-01-21 09:49:02.695282: step: 1592/526, loss: 0.7267881631851196 2023-01-21 09:49:03.840503: step: 1596/526, loss: 0.028608275577425957 2023-01-21 09:49:04.967580: step: 1600/526, loss: 0.07127423584461212 2023-01-21 09:49:06.124383: step: 1604/526, loss: 0.07827906310558319 2023-01-21 09:49:07.272001: step: 1608/526, loss: 0.09033241868019104 2023-01-21 09:49:08.423320: step: 1612/526, loss: 0.19561472535133362 2023-01-21 09:49:09.568954: step: 1616/526, loss: 0.10225506126880646 2023-01-21 09:49:10.690363: step: 1620/526, loss: 0.15876071155071259 2023-01-21 09:49:11.848923: step: 1624/526, loss: 0.1647598296403885 2023-01-21 09:49:13.082960: step: 1628/526, loss: 1.0362557172775269 2023-01-21 09:49:14.218691: step: 1632/526, loss: 0.08430089801549911 2023-01-21 09:49:15.339391: step: 1636/526, loss: 0.06773605942726135 2023-01-21 09:49:16.480165: step: 1640/526, loss: 0.12646789848804474 2023-01-21 09:49:17.617503: step: 1644/526, loss: 0.13007886707782745 2023-01-21 09:49:18.765683: step: 1648/526, loss: 0.08169527351856232 2023-01-21 09:49:19.899651: step: 1652/526, loss: 0.013237333856523037 2023-01-21 09:49:21.014009: step: 1656/526, loss: 0.044371794909238815 2023-01-21 09:49:22.209705: step: 1660/526, loss: 0.08450069278478622 2023-01-21 09:49:23.336546: step: 1664/526, loss: 0.16088305413722992 2023-01-21 09:49:24.456503: step: 1668/526, loss: 0.23504170775413513 2023-01-21 09:49:25.573481: step: 1672/526, loss: 0.059360697865486145 2023-01-21 09:49:26.699778: step: 1676/526, loss: 0.13032637536525726 2023-01-21 09:49:27.849482: step: 1680/526, loss: 0.12923890352249146 2023-01-21 09:49:28.989995: step: 1684/526, loss: 0.15137061476707458 2023-01-21 09:49:30.114900: step: 1688/526, loss: 0.10355682671070099 2023-01-21 09:49:31.248981: step: 1692/526, loss: 0.20478634536266327 2023-01-21 09:49:32.370196: step: 1696/526, loss: 0.026871968060731888 2023-01-21 09:49:33.493742: step: 1700/526, loss: 0.22027406096458435 2023-01-21 09:49:34.620607: step: 1704/526, loss: 0.20819568634033203 2023-01-21 09:49:35.772523: step: 1708/526, loss: 0.07477288693189621 2023-01-21 09:49:36.914933: step: 1712/526, loss: 0.1203809306025505 2023-01-21 09:49:38.069813: step: 1716/526, loss: 0.12176313996315002 2023-01-21 09:49:39.228138: step: 1720/526, loss: 0.11805567890405655 2023-01-21 09:49:40.360555: step: 1724/526, loss: 0.12193985283374786 2023-01-21 09:49:41.496349: step: 1728/526, loss: 0.13236084580421448 2023-01-21 09:49:42.645433: step: 1732/526, loss: 0.14280566573143005 2023-01-21 09:49:43.775065: step: 1736/526, loss: 0.13417492806911469 2023-01-21 09:49:44.913883: step: 1740/526, loss: 0.02441840060055256 2023-01-21 09:49:46.029597: step: 1744/526, loss: 0.0666445717215538 2023-01-21 09:49:47.154498: step: 1748/526, loss: 0.32476407289505005 2023-01-21 09:49:48.296901: step: 1752/526, loss: 0.18137098848819733 2023-01-21 09:49:49.428693: step: 1756/526, loss: 0.022738946601748466 2023-01-21 09:49:50.583574: step: 1760/526, loss: 0.16420021653175354 2023-01-21 09:49:51.728262: step: 1764/526, loss: 0.20444834232330322 2023-01-21 09:49:52.864790: step: 1768/526, loss: 0.175619974732399 2023-01-21 09:49:53.995388: step: 1772/526, loss: 0.10529603809118271 2023-01-21 09:49:55.123944: step: 1776/526, loss: 0.29934147000312805 2023-01-21 09:49:56.256410: step: 1780/526, loss: 0.03155786916613579 2023-01-21 09:49:57.375607: step: 1784/526, loss: 0.1482965052127838 2023-01-21 09:49:58.498762: step: 1788/526, loss: 0.05663476139307022 2023-01-21 09:49:59.649765: step: 1792/526, loss: 0.147626131772995 2023-01-21 09:50:00.770798: step: 1796/526, loss: 0.32595062255859375 2023-01-21 09:50:01.902985: step: 1800/526, loss: 0.1381494253873825 2023-01-21 09:50:03.056483: step: 1804/526, loss: 0.0920679122209549 2023-01-21 09:50:04.155953: step: 1808/526, loss: 0.02328047715127468 2023-01-21 09:50:05.315658: step: 1812/526, loss: 0.09933920204639435 2023-01-21 09:50:06.442467: step: 1816/526, loss: 0.1050751656293869 2023-01-21 09:50:07.568293: step: 1820/526, loss: 0.028495408594608307 2023-01-21 09:50:08.702140: step: 1824/526, loss: 0.20693102478981018 2023-01-21 09:50:09.828422: step: 1828/526, loss: 0.2184181958436966 2023-01-21 09:50:10.963713: step: 1832/526, loss: 0.13041381537914276 2023-01-21 09:50:12.092040: step: 1836/526, loss: 0.16703061759471893 2023-01-21 09:50:13.204504: step: 1840/526, loss: 1.2478207349777222 2023-01-21 09:50:14.327237: step: 1844/526, loss: 0.05235252529382706 2023-01-21 09:50:15.466885: step: 1848/526, loss: 0.08937931060791016 2023-01-21 09:50:16.598950: step: 1852/526, loss: 0.1811787486076355 2023-01-21 09:50:17.721487: step: 1856/526, loss: 0.23315058648586273 2023-01-21 09:50:18.852650: step: 1860/526, loss: 0.09772854298353195 2023-01-21 09:50:20.024000: step: 1864/526, loss: 0.24598237872123718 2023-01-21 09:50:21.207492: step: 1868/526, loss: 0.08545856177806854 2023-01-21 09:50:22.356991: step: 1872/526, loss: 0.01973877102136612 2023-01-21 09:50:23.517811: step: 1876/526, loss: 0.19826316833496094 2023-01-21 09:50:24.624013: step: 1880/526, loss: 0.03031148947775364 2023-01-21 09:50:25.747169: step: 1884/526, loss: 0.12070666253566742 2023-01-21 09:50:26.858050: step: 1888/526, loss: 0.05679841339588165 2023-01-21 09:50:27.988394: step: 1892/526, loss: 0.1682707816362381 2023-01-21 09:50:29.115961: step: 1896/526, loss: 0.04011411592364311 2023-01-21 09:50:30.271540: step: 1900/526, loss: 0.06990289688110352 2023-01-21 09:50:31.413185: step: 1904/526, loss: 0.2473526895046234 2023-01-21 09:50:32.568458: step: 1908/526, loss: 0.028064489364624023 2023-01-21 09:50:33.731947: step: 1912/526, loss: 0.04668784141540527 2023-01-21 09:50:34.888399: step: 1916/526, loss: 0.10080371052026749 2023-01-21 09:50:36.028440: step: 1920/526, loss: 0.07001761347055435 2023-01-21 09:50:37.169325: step: 1924/526, loss: 0.823540210723877 2023-01-21 09:50:38.297396: step: 1928/526, loss: 0.03893604502081871 2023-01-21 09:50:39.445032: step: 1932/526, loss: 0.11929807811975479 2023-01-21 09:50:40.597451: step: 1936/526, loss: 0.08397804200649261 2023-01-21 09:50:41.719305: step: 1940/526, loss: 0.034690700471401215 2023-01-21 09:50:42.868796: step: 1944/526, loss: 0.30808085203170776 2023-01-21 09:50:43.974093: step: 1948/526, loss: 0.06326466053724289 2023-01-21 09:50:45.123891: step: 1952/526, loss: 0.021103668957948685 2023-01-21 09:50:46.251784: step: 1956/526, loss: 0.06706027686595917 2023-01-21 09:50:47.400079: step: 1960/526, loss: 0.11220850795507431 2023-01-21 09:50:48.550368: step: 1964/526, loss: 0.14683294296264648 2023-01-21 09:50:49.702621: step: 1968/526, loss: 0.1673121452331543 2023-01-21 09:50:50.844320: step: 1972/526, loss: 0.017523478716611862 2023-01-21 09:50:51.969109: step: 1976/526, loss: 0.10771122574806213 2023-01-21 09:50:53.133252: step: 1980/526, loss: 0.11405238509178162 2023-01-21 09:50:54.275929: step: 1984/526, loss: 0.8534290194511414 2023-01-21 09:50:55.413738: step: 1988/526, loss: 0.16459941864013672 2023-01-21 09:50:56.555013: step: 1992/526, loss: 0.07454963028430939 2023-01-21 09:50:57.710530: step: 1996/526, loss: 0.19767847657203674 2023-01-21 09:50:58.831051: step: 2000/526, loss: 0.22813721001148224 2023-01-21 09:50:59.995226: step: 2004/526, loss: 0.05680818855762482 2023-01-21 09:51:01.127803: step: 2008/526, loss: 0.15537257492542267 2023-01-21 09:51:02.259362: step: 2012/526, loss: 0.15140561759471893 2023-01-21 09:51:03.371828: step: 2016/526, loss: 0.12234115600585938 2023-01-21 09:51:04.486199: step: 2020/526, loss: 0.05316825211048126 2023-01-21 09:51:05.613498: step: 2024/526, loss: 0.0906367301940918 2023-01-21 09:51:06.755773: step: 2028/526, loss: 0.26286783814430237 2023-01-21 09:51:07.898850: step: 2032/526, loss: 0.35145533084869385 2023-01-21 09:51:09.042040: step: 2036/526, loss: 0.2321152687072754 2023-01-21 09:51:10.198554: step: 2040/526, loss: 0.13255424797534943 2023-01-21 09:51:11.342273: step: 2044/526, loss: 0.0745459571480751 2023-01-21 09:51:12.475523: step: 2048/526, loss: 0.154963880777359 2023-01-21 09:51:13.594367: step: 2052/526, loss: 0.10433606803417206 2023-01-21 09:51:14.703949: step: 2056/526, loss: 0.0325261615216732 2023-01-21 09:51:15.851550: step: 2060/526, loss: 0.27381545305252075 2023-01-21 09:51:17.003623: step: 2064/526, loss: 0.16113099455833435 2023-01-21 09:51:18.130507: step: 2068/526, loss: 0.0280564334243536 2023-01-21 09:51:19.285676: step: 2072/526, loss: 0.08778181672096252 2023-01-21 09:51:20.451471: step: 2076/526, loss: 0.13478480279445648 2023-01-21 09:51:21.576609: step: 2080/526, loss: 0.1894073486328125 2023-01-21 09:51:22.706593: step: 2084/526, loss: 0.01869783364236355 2023-01-21 09:51:23.837648: step: 2088/526, loss: 0.11841030418872833 2023-01-21 09:51:24.971746: step: 2092/526, loss: 0.04905872046947479 2023-01-21 09:51:26.100968: step: 2096/526, loss: 0.12393054366111755 2023-01-21 09:51:27.268430: step: 2100/526, loss: 0.241658017039299 2023-01-21 09:51:28.407976: step: 2104/526, loss: 0.17300310730934143 ================================================== Loss: 0.172 -------------------- Dev: {'event': {'p': 0.5892494929006086, 'r': 0.7736351531291611, 'f1': 0.6689694876223374}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Test: {'event': {'p': 0.6005770816158286, 'r': 0.7604384133611691, 'f1': 0.6711192998618148}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Chinese: {'event': {'p': 0.5308641975308642, 'r': 0.7962962962962963, 'f1': 0.637037037037037}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Korean: {'event': {'p': 0.7073170731707317, 'r': 0.4603174603174603, 'f1': 0.5576923076923077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Russian: {'event': {'p': 0.4772727272727273, 'r': 0.5833333333333334, 'f1': 0.5250000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.5892494929006086, 'r': 0.7736351531291611, 'f1': 0.6689694876223374}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Eng Test for Korean: {'event': {'p': 0.6005770816158286, 'r': 0.7604384133611691, 'f1': 0.6711192998618148}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} Sample Korean: {'event': {'p': 0.7073170731707317, 'r': 0.4603174603174603, 'f1': 0.5576923076923077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 6} -------------------- Eng Dev for Russian: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5116279069767442, 'r': 0.6111111111111112, 'f1': 0.5569620253164557}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 7 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 09:52:16.543183: step: 4/526, loss: 0.08167391270399094 2023-01-21 09:52:17.673537: step: 8/526, loss: 0.07296915352344513 2023-01-21 09:52:18.857455: step: 12/526, loss: 0.10415878891944885 2023-01-21 09:52:20.022869: step: 16/526, loss: 0.339252769947052 2023-01-21 09:52:21.147982: step: 20/526, loss: 0.11001763492822647 2023-01-21 09:52:22.300797: step: 24/526, loss: 0.12424440681934357 2023-01-21 09:52:23.463396: step: 28/526, loss: 0.13503007590770721 2023-01-21 09:52:24.607282: step: 32/526, loss: 0.04702162742614746 2023-01-21 09:52:25.742090: step: 36/526, loss: 0.18683910369873047 2023-01-21 09:52:26.864876: step: 40/526, loss: 0.13439694046974182 2023-01-21 09:52:27.972133: step: 44/526, loss: 0.10474987328052521 2023-01-21 09:52:29.114468: step: 48/526, loss: 0.07858066260814667 2023-01-21 09:52:30.248223: step: 52/526, loss: 0.31191468238830566 2023-01-21 09:52:31.374365: step: 56/526, loss: 0.07207565009593964 2023-01-21 09:52:32.518496: step: 60/526, loss: 0.013375366106629372 2023-01-21 09:52:33.640125: step: 64/526, loss: 0.19718004763126373 2023-01-21 09:52:34.752136: step: 68/526, loss: 0.05647735670208931 2023-01-21 09:52:35.896616: step: 72/526, loss: 0.15307402610778809 2023-01-21 09:52:37.052366: step: 76/526, loss: 0.18623638153076172 2023-01-21 09:52:38.168522: step: 80/526, loss: 0.13776616752147675 2023-01-21 09:52:39.307671: step: 84/526, loss: 0.6230600476264954 2023-01-21 09:52:40.440856: step: 88/526, loss: 0.023225117474794388 2023-01-21 09:52:41.569857: step: 92/526, loss: 0.05773868411779404 2023-01-21 09:52:42.706199: step: 96/526, loss: 0.08946572244167328 2023-01-21 09:52:43.850442: step: 100/526, loss: 0.20988251268863678 2023-01-21 09:52:44.998769: step: 104/526, loss: 0.5690016150474548 2023-01-21 09:52:46.137264: step: 108/526, loss: 0.18187038600444794 2023-01-21 09:52:47.288243: step: 112/526, loss: 0.08731038868427277 2023-01-21 09:52:48.407920: step: 116/526, loss: 0.07598146796226501 2023-01-21 09:52:49.585436: step: 120/526, loss: 0.13496838510036469 2023-01-21 09:52:50.732913: step: 124/526, loss: 0.2077789306640625 2023-01-21 09:52:51.874019: step: 128/526, loss: 0.013856267556548119 2023-01-21 09:52:53.014032: step: 132/526, loss: 0.09541435539722443 2023-01-21 09:52:54.156669: step: 136/526, loss: 0.05809245631098747 2023-01-21 09:52:55.284429: step: 140/526, loss: 0.3651152551174164 2023-01-21 09:52:56.419340: step: 144/526, loss: 0.018068598583340645 2023-01-21 09:52:57.538246: step: 148/526, loss: 0.24460040032863617 2023-01-21 09:52:58.669914: step: 152/526, loss: 0.04399614408612251 2023-01-21 09:52:59.820241: step: 156/526, loss: 0.22436374425888062 2023-01-21 09:53:00.958796: step: 160/526, loss: 0.0024582864716649055 2023-01-21 09:53:02.102061: step: 164/526, loss: 0.021137237548828125 2023-01-21 09:53:03.213506: step: 168/526, loss: 0.06161389499902725 2023-01-21 09:53:04.353828: step: 172/526, loss: 0.15993185341358185 2023-01-21 09:53:05.511038: step: 176/526, loss: 0.9113231897354126 2023-01-21 09:53:06.636046: step: 180/526, loss: 0.060880571603775024 2023-01-21 09:53:07.746723: step: 184/526, loss: 0.050478171557188034 2023-01-21 09:53:08.854582: step: 188/526, loss: 0.015278911218047142 2023-01-21 09:53:10.023739: step: 192/526, loss: 0.07800436019897461 2023-01-21 09:53:11.170636: step: 196/526, loss: 0.5849026441574097 2023-01-21 09:53:12.303978: step: 200/526, loss: 0.042290449142456055 2023-01-21 09:53:13.429789: step: 204/526, loss: 0.0784408301115036 2023-01-21 09:53:14.567096: step: 208/526, loss: 0.02679157257080078 2023-01-21 09:53:15.734826: step: 212/526, loss: 0.09209270775318146 2023-01-21 09:53:16.858415: step: 216/526, loss: 0.047389887273311615 2023-01-21 09:53:18.000984: step: 220/526, loss: 0.5147913098335266 2023-01-21 09:53:19.146608: step: 224/526, loss: 0.05951347574591637 2023-01-21 09:53:20.285627: step: 228/526, loss: 0.06431560218334198 2023-01-21 09:53:21.378363: step: 232/526, loss: 0.06869430840015411 2023-01-21 09:53:22.541396: step: 236/526, loss: 0.07268104702234268 2023-01-21 09:53:23.660723: step: 240/526, loss: 0.18074187636375427 2023-01-21 09:53:24.798125: step: 244/526, loss: 0.038242436945438385 2023-01-21 09:53:25.943104: step: 248/526, loss: 0.037408068776130676 2023-01-21 09:53:27.091531: step: 252/526, loss: 0.015559673309326172 2023-01-21 09:53:28.248316: step: 256/526, loss: 0.07357215881347656 2023-01-21 09:53:29.356171: step: 260/526, loss: 0.06849727034568787 2023-01-21 09:53:30.506405: step: 264/526, loss: 0.46111923456192017 2023-01-21 09:53:31.676731: step: 268/526, loss: 0.20764578878879547 2023-01-21 09:53:32.800263: step: 272/526, loss: 0.1614522486925125 2023-01-21 09:53:33.952813: step: 276/526, loss: 0.22258442640304565 2023-01-21 09:53:35.107535: step: 280/526, loss: 0.08382568508386612 2023-01-21 09:53:36.277872: step: 284/526, loss: 0.06825542449951172 2023-01-21 09:53:37.397778: step: 288/526, loss: 0.13082657754421234 2023-01-21 09:53:38.540460: step: 292/526, loss: 0.48294275999069214 2023-01-21 09:53:39.699809: step: 296/526, loss: 0.03033742867410183 2023-01-21 09:53:40.841185: step: 300/526, loss: 0.42863520979881287 2023-01-21 09:53:41.979625: step: 304/526, loss: 0.05352230370044708 2023-01-21 09:53:43.100410: step: 308/526, loss: 0.10876922309398651 2023-01-21 09:53:44.255301: step: 312/526, loss: 0.028925180435180664 2023-01-21 09:53:45.390475: step: 316/526, loss: 0.0429142490029335 2023-01-21 09:53:46.488011: step: 320/526, loss: 0.005161619279533625 2023-01-21 09:53:47.586531: step: 324/526, loss: 0.016247272491455078 2023-01-21 09:53:48.756769: step: 328/526, loss: 0.05932874232530594 2023-01-21 09:53:49.898453: step: 332/526, loss: 0.03951101377606392 2023-01-21 09:53:51.049165: step: 336/526, loss: 0.04240245744585991 2023-01-21 09:53:52.208584: step: 340/526, loss: 0.02077507972717285 2023-01-21 09:53:53.366386: step: 344/526, loss: 0.0703771561384201 2023-01-21 09:53:54.499825: step: 348/526, loss: 0.041604042053222656 2023-01-21 09:53:55.644470: step: 352/526, loss: 0.1324130892753601 2023-01-21 09:53:56.794796: step: 356/526, loss: 0.10725397616624832 2023-01-21 09:53:57.938283: step: 360/526, loss: 0.038194939494132996 2023-01-21 09:53:59.076073: step: 364/526, loss: 0.7604475617408752 2023-01-21 09:54:00.224082: step: 368/526, loss: 0.10343198478221893 2023-01-21 09:54:01.388512: step: 372/526, loss: 0.30154791474342346 2023-01-21 09:54:02.501786: step: 376/526, loss: 0.7934285402297974 2023-01-21 09:54:03.623233: step: 380/526, loss: 0.10674865543842316 2023-01-21 09:54:04.781498: step: 384/526, loss: 0.07076716423034668 2023-01-21 09:54:05.918016: step: 388/526, loss: 0.22519856691360474 2023-01-21 09:54:07.037085: step: 392/526, loss: 0.03980549797415733 2023-01-21 09:54:08.191176: step: 396/526, loss: 0.03682880476117134 2023-01-21 09:54:09.324213: step: 400/526, loss: 0.005124473944306374 2023-01-21 09:54:10.444194: step: 404/526, loss: 0.19035053253173828 2023-01-21 09:54:11.574663: step: 408/526, loss: 0.04564166069030762 2023-01-21 09:54:12.736920: step: 412/526, loss: 0.16892023384571075 2023-01-21 09:54:13.923947: step: 416/526, loss: 0.0794258564710617 2023-01-21 09:54:15.066396: step: 420/526, loss: 0.5449943542480469 2023-01-21 09:54:16.189246: step: 424/526, loss: 0.09458093345165253 2023-01-21 09:54:17.315757: step: 428/526, loss: 0.04249515384435654 2023-01-21 09:54:18.430851: step: 432/526, loss: 0.14795684814453125 2023-01-21 09:54:19.587882: step: 436/526, loss: 0.11780376732349396 2023-01-21 09:54:20.716368: step: 440/526, loss: 0.11471491307020187 2023-01-21 09:54:21.857133: step: 444/526, loss: 0.04942111670970917 2023-01-21 09:54:22.995188: step: 448/526, loss: 0.055294036865234375 2023-01-21 09:54:24.159330: step: 452/526, loss: 0.07405147701501846 2023-01-21 09:54:25.312711: step: 456/526, loss: 0.07961221039295197 2023-01-21 09:54:26.447172: step: 460/526, loss: 0.026602745056152344 2023-01-21 09:54:27.574063: step: 464/526, loss: 0.292308509349823 2023-01-21 09:54:28.725960: step: 468/526, loss: 0.07435820251703262 2023-01-21 09:54:29.868909: step: 472/526, loss: 0.046309471130371094 2023-01-21 09:54:31.017747: step: 476/526, loss: 0.15707239508628845 2023-01-21 09:54:32.155871: step: 480/526, loss: 0.047875214368104935 2023-01-21 09:54:33.292291: step: 484/526, loss: 0.07487936317920685 2023-01-21 09:54:34.424541: step: 488/526, loss: 0.0793735533952713 2023-01-21 09:54:35.554612: step: 492/526, loss: 0.13200893998146057 2023-01-21 09:54:36.696097: step: 496/526, loss: 0.11926007270812988 2023-01-21 09:54:37.856538: step: 500/526, loss: 0.1535160094499588 2023-01-21 09:54:38.975404: step: 504/526, loss: 0.18116417527198792 2023-01-21 09:54:40.103181: step: 508/526, loss: 0.580575168132782 2023-01-21 09:54:41.223891: step: 512/526, loss: 0.04906139522790909 2023-01-21 09:54:42.337213: step: 516/526, loss: 0.0695432648062706 2023-01-21 09:54:43.484873: step: 520/526, loss: 0.17038078606128693 2023-01-21 09:54:44.608580: step: 524/526, loss: 0.10769033432006836 2023-01-21 09:54:45.734145: step: 528/526, loss: 0.49971145391464233 2023-01-21 09:54:46.872780: step: 532/526, loss: 0.25485116243362427 2023-01-21 09:54:48.003607: step: 536/526, loss: 0.05196123570203781 2023-01-21 09:54:49.138296: step: 540/526, loss: 0.005896282382309437 2023-01-21 09:54:50.271272: step: 544/526, loss: 0.09783181548118591 2023-01-21 09:54:51.399022: step: 548/526, loss: 0.10216712951660156 2023-01-21 09:54:52.526513: step: 552/526, loss: 0.03494014963507652 2023-01-21 09:54:53.646724: step: 556/526, loss: 0.020683670416474342 2023-01-21 09:54:54.792467: step: 560/526, loss: 0.03055420145392418 2023-01-21 09:54:55.930564: step: 564/526, loss: 0.12073612958192825 2023-01-21 09:54:57.085490: step: 568/526, loss: 0.2799626290798187 2023-01-21 09:54:58.232865: step: 572/526, loss: 0.08815017342567444 2023-01-21 09:54:59.358408: step: 576/526, loss: 0.05846752971410751 2023-01-21 09:55:00.501983: step: 580/526, loss: 0.12896637618541718 2023-01-21 09:55:01.640805: step: 584/526, loss: 0.43202272057533264 2023-01-21 09:55:02.791973: step: 588/526, loss: 0.034606173634529114 2023-01-21 09:55:03.932188: step: 592/526, loss: 0.31149905920028687 2023-01-21 09:55:05.083477: step: 596/526, loss: 0.11661787331104279 2023-01-21 09:55:06.232537: step: 600/526, loss: 0.6148589849472046 2023-01-21 09:55:07.358827: step: 604/526, loss: 1.1019624471664429 2023-01-21 09:55:08.482430: step: 608/526, loss: 0.0805574506521225 2023-01-21 09:55:09.615784: step: 612/526, loss: 0.05362563207745552 2023-01-21 09:55:10.776784: step: 616/526, loss: 0.08867712318897247 2023-01-21 09:55:11.908386: step: 620/526, loss: 0.07213282585144043 2023-01-21 09:55:13.023662: step: 624/526, loss: 0.03709716722369194 2023-01-21 09:55:14.172906: step: 628/526, loss: 0.1803010106086731 2023-01-21 09:55:15.328007: step: 632/526, loss: 0.0391879566013813 2023-01-21 09:55:16.464729: step: 636/526, loss: 0.10424366593360901 2023-01-21 09:55:17.597273: step: 640/526, loss: 0.009069728665053844 2023-01-21 09:55:18.734620: step: 644/526, loss: 0.5905860662460327 2023-01-21 09:55:19.841663: step: 648/526, loss: 0.023659229278564453 2023-01-21 09:55:20.956853: step: 652/526, loss: 0.0717647597193718 2023-01-21 09:55:22.128909: step: 656/526, loss: 0.01780710183084011 2023-01-21 09:55:23.271124: step: 660/526, loss: 0.10806316882371902 2023-01-21 09:55:24.432167: step: 664/526, loss: 0.11200366169214249 2023-01-21 09:55:25.561507: step: 668/526, loss: 0.0597173236310482 2023-01-21 09:55:26.715644: step: 672/526, loss: 0.07035350799560547 2023-01-21 09:55:27.893871: step: 676/526, loss: 0.274576872587204 2023-01-21 09:55:29.023936: step: 680/526, loss: 0.042966749519109726 2023-01-21 09:55:30.193803: step: 684/526, loss: 0.2340126931667328 2023-01-21 09:55:31.309147: step: 688/526, loss: 0.061487678438425064 2023-01-21 09:55:32.452426: step: 692/526, loss: 0.11537900567054749 2023-01-21 09:55:33.584901: step: 696/526, loss: 0.04064655303955078 2023-01-21 09:55:34.725856: step: 700/526, loss: 0.008150959387421608 2023-01-21 09:55:35.835012: step: 704/526, loss: 0.08518867194652557 2023-01-21 09:55:36.966547: step: 708/526, loss: 0.08306531608104706 2023-01-21 09:55:38.104247: step: 712/526, loss: 0.03696250915527344 2023-01-21 09:55:39.262499: step: 716/526, loss: 0.1326136291027069 2023-01-21 09:55:40.370047: step: 720/526, loss: 0.27834969758987427 2023-01-21 09:55:41.518686: step: 724/526, loss: 0.08183804154396057 2023-01-21 09:55:42.660546: step: 728/526, loss: 0.15969830751419067 2023-01-21 09:55:43.771494: step: 732/526, loss: 0.014976215548813343 2023-01-21 09:55:44.927966: step: 736/526, loss: 0.1051931381225586 2023-01-21 09:55:46.045220: step: 740/526, loss: 0.025025749579072 2023-01-21 09:55:47.195624: step: 744/526, loss: 0.014679431915283203 2023-01-21 09:55:48.353580: step: 748/526, loss: 0.3080042004585266 2023-01-21 09:55:49.448554: step: 752/526, loss: 0.15704306960105896 2023-01-21 09:55:50.583808: step: 756/526, loss: 0.06792859733104706 2023-01-21 09:55:51.707562: step: 760/526, loss: 0.05796823650598526 2023-01-21 09:55:52.852874: step: 764/526, loss: 0.37051573395729065 2023-01-21 09:55:53.983263: step: 768/526, loss: 0.0008899688837118447 2023-01-21 09:55:55.121987: step: 772/526, loss: 0.4569458067417145 2023-01-21 09:55:56.231350: step: 776/526, loss: 0.10436372458934784 2023-01-21 09:55:57.388731: step: 780/526, loss: 0.005532217212021351 2023-01-21 09:55:58.532134: step: 784/526, loss: 0.1366383582353592 2023-01-21 09:55:59.654067: step: 788/526, loss: 0.03198595345020294 2023-01-21 09:56:00.769084: step: 792/526, loss: 0.1335197389125824 2023-01-21 09:56:01.928965: step: 796/526, loss: 0.40154755115509033 2023-01-21 09:56:03.093077: step: 800/526, loss: 0.19560456275939941 2023-01-21 09:56:04.254997: step: 804/526, loss: 0.10589656978845596 2023-01-21 09:56:05.405313: step: 808/526, loss: 0.10243845731019974 2023-01-21 09:56:06.541231: step: 812/526, loss: 0.09133201092481613 2023-01-21 09:56:07.681818: step: 816/526, loss: 0.17204904556274414 2023-01-21 09:56:08.854421: step: 820/526, loss: 0.06681585311889648 2023-01-21 09:56:10.018262: step: 824/526, loss: 0.33932629227638245 2023-01-21 09:56:11.156193: step: 828/526, loss: 0.07811050862073898 2023-01-21 09:56:12.288987: step: 832/526, loss: 0.15597376227378845 2023-01-21 09:56:13.433506: step: 836/526, loss: 0.07745952904224396 2023-01-21 09:56:14.585829: step: 840/526, loss: 0.07306583225727081 2023-01-21 09:56:15.732509: step: 844/526, loss: 0.054602622985839844 2023-01-21 09:56:16.887370: step: 848/526, loss: 0.2932031750679016 2023-01-21 09:56:17.995652: step: 852/526, loss: 0.06582732498645782 2023-01-21 09:56:19.116993: step: 856/526, loss: 0.03746090084314346 2023-01-21 09:56:20.248694: step: 860/526, loss: 0.11692628264427185 2023-01-21 09:56:21.375078: step: 864/526, loss: 0.4945685565471649 2023-01-21 09:56:22.498436: step: 868/526, loss: 0.1825283020734787 2023-01-21 09:56:23.631714: step: 872/526, loss: 0.03533501550555229 2023-01-21 09:56:24.752861: step: 876/526, loss: 0.037752602249383926 2023-01-21 09:56:25.881104: step: 880/526, loss: 0.10571518540382385 2023-01-21 09:56:27.030997: step: 884/526, loss: 0.33125782012939453 2023-01-21 09:56:28.170882: step: 888/526, loss: 0.03770214319229126 2023-01-21 09:56:29.297020: step: 892/526, loss: 0.1424632966518402 2023-01-21 09:56:30.459222: step: 896/526, loss: 0.3080737292766571 2023-01-21 09:56:31.612747: step: 900/526, loss: 0.35783377289772034 2023-01-21 09:56:32.742793: step: 904/526, loss: 0.029872512444853783 2023-01-21 09:56:33.898255: step: 908/526, loss: 0.16637524962425232 2023-01-21 09:56:35.025426: step: 912/526, loss: 0.020266462117433548 2023-01-21 09:56:36.206386: step: 916/526, loss: 0.26609399914741516 2023-01-21 09:56:37.335288: step: 920/526, loss: 0.0461542122066021 2023-01-21 09:56:38.491959: step: 924/526, loss: 0.017359159886837006 2023-01-21 09:56:39.619537: step: 928/526, loss: 0.07158572971820831 2023-01-21 09:56:40.731732: step: 932/526, loss: 0.05541105195879936 2023-01-21 09:56:41.881405: step: 936/526, loss: 0.05066251754760742 2023-01-21 09:56:42.983835: step: 940/526, loss: 0.04320118948817253 2023-01-21 09:56:44.121037: step: 944/526, loss: 0.0726872906088829 2023-01-21 09:56:45.246905: step: 948/526, loss: 1.1193865537643433 2023-01-21 09:56:46.390420: step: 952/526, loss: 0.009647369384765625 2023-01-21 09:56:47.544764: step: 956/526, loss: 0.07161865383386612 2023-01-21 09:56:48.695335: step: 960/526, loss: 0.21820850670337677 2023-01-21 09:56:49.818743: step: 964/526, loss: 0.11262059211730957 2023-01-21 09:56:50.977369: step: 968/526, loss: 0.052465248852968216 2023-01-21 09:56:52.115346: step: 972/526, loss: 0.0840580016374588 2023-01-21 09:56:53.266663: step: 976/526, loss: 0.08422450721263885 2023-01-21 09:56:54.379446: step: 980/526, loss: 0.05547799915075302 2023-01-21 09:56:55.522805: step: 984/526, loss: 0.3026929497718811 2023-01-21 09:56:56.642534: step: 988/526, loss: 0.05694770812988281 2023-01-21 09:56:57.767902: step: 992/526, loss: 0.04771725833415985 2023-01-21 09:56:58.873406: step: 996/526, loss: 0.0326298251748085 2023-01-21 09:57:00.034080: step: 1000/526, loss: 0.1268974244594574 2023-01-21 09:57:01.158596: step: 1004/526, loss: 0.09495782852172852 2023-01-21 09:57:02.282575: step: 1008/526, loss: 0.05754394456744194 2023-01-21 09:57:03.414706: step: 1012/526, loss: 0.07991227507591248 2023-01-21 09:57:04.537464: step: 1016/526, loss: 0.06967845559120178 2023-01-21 09:57:05.662845: step: 1020/526, loss: 0.18973246216773987 2023-01-21 09:57:06.813751: step: 1024/526, loss: 0.07780519127845764 2023-01-21 09:57:07.980850: step: 1028/526, loss: 0.15296021103858948 2023-01-21 09:57:09.141845: step: 1032/526, loss: 0.10831265151500702 2023-01-21 09:57:10.296697: step: 1036/526, loss: 0.15890845656394958 2023-01-21 09:57:11.448001: step: 1040/526, loss: 0.06265049427747726 2023-01-21 09:57:12.596578: step: 1044/526, loss: 0.11585778743028641 2023-01-21 09:57:13.746066: step: 1048/526, loss: 0.024000167846679688 2023-01-21 09:57:14.888292: step: 1052/526, loss: 0.007708740420639515 2023-01-21 09:57:16.041524: step: 1056/526, loss: 0.17126807570457458 2023-01-21 09:57:17.192797: step: 1060/526, loss: 0.6998818516731262 2023-01-21 09:57:18.313889: step: 1064/526, loss: 0.0391172394156456 2023-01-21 09:57:19.438125: step: 1068/526, loss: 0.25107860565185547 2023-01-21 09:57:20.556489: step: 1072/526, loss: 0.10010509938001633 2023-01-21 09:57:21.679451: step: 1076/526, loss: 0.0942656546831131 2023-01-21 09:57:22.774201: step: 1080/526, loss: 0.05831451714038849 2023-01-21 09:57:23.930456: step: 1084/526, loss: 0.10765514522790909 2023-01-21 09:57:25.060671: step: 1088/526, loss: 0.04110097885131836 2023-01-21 09:57:26.187624: step: 1092/526, loss: 0.1538499891757965 2023-01-21 09:57:27.314279: step: 1096/526, loss: 0.09977855533361435 2023-01-21 09:57:28.452212: step: 1100/526, loss: 0.3947006165981293 2023-01-21 09:57:29.601205: step: 1104/526, loss: 0.021377945318818092 2023-01-21 09:57:30.719388: step: 1108/526, loss: 0.18397732079029083 2023-01-21 09:57:31.826796: step: 1112/526, loss: 0.18016815185546875 2023-01-21 09:57:32.993474: step: 1116/526, loss: 0.5049295425415039 2023-01-21 09:57:34.134231: step: 1120/526, loss: 0.46759310364723206 2023-01-21 09:57:35.288093: step: 1124/526, loss: 0.11646925657987595 2023-01-21 09:57:36.403523: step: 1128/526, loss: 0.10861194133758545 2023-01-21 09:57:37.540001: step: 1132/526, loss: 0.19022607803344727 2023-01-21 09:57:38.662671: step: 1136/526, loss: 0.1617298126220703 2023-01-21 09:57:39.823607: step: 1140/526, loss: 0.11830015480518341 2023-01-21 09:57:40.991825: step: 1144/526, loss: 0.8109237551689148 2023-01-21 09:57:42.133920: step: 1148/526, loss: 0.03343292325735092 2023-01-21 09:57:43.265890: step: 1152/526, loss: 0.09109906852245331 2023-01-21 09:57:44.416854: step: 1156/526, loss: 0.06347141414880753 2023-01-21 09:57:45.536445: step: 1160/526, loss: 0.024823855608701706 2023-01-21 09:57:46.690826: step: 1164/526, loss: 0.08118382096290588 2023-01-21 09:57:47.806394: step: 1168/526, loss: 0.1971965879201889 2023-01-21 09:57:48.960543: step: 1172/526, loss: 0.05900769308209419 2023-01-21 09:57:50.085369: step: 1176/526, loss: 0.043454360216856 2023-01-21 09:57:51.209694: step: 1180/526, loss: 0.019617462530732155 2023-01-21 09:57:52.330044: step: 1184/526, loss: 0.10672197490930557 2023-01-21 09:57:53.475374: step: 1188/526, loss: 0.23176135122776031 2023-01-21 09:57:54.596489: step: 1192/526, loss: 0.07296357303857803 2023-01-21 09:57:55.750560: step: 1196/526, loss: 0.09023533016443253 2023-01-21 09:57:56.856096: step: 1200/526, loss: 0.14960722625255585 2023-01-21 09:57:57.997802: step: 1204/526, loss: 0.708601176738739 2023-01-21 09:57:59.144831: step: 1208/526, loss: 0.07830296456813812 2023-01-21 09:58:00.280259: step: 1212/526, loss: 0.0588352233171463 2023-01-21 09:58:01.428749: step: 1216/526, loss: 0.09022102504968643 2023-01-21 09:58:02.556099: step: 1220/526, loss: 0.5049480199813843 2023-01-21 09:58:03.706332: step: 1224/526, loss: 0.10298814624547958 2023-01-21 09:58:04.876211: step: 1228/526, loss: 0.030216598883271217 2023-01-21 09:58:06.010145: step: 1232/526, loss: 0.02164621278643608 2023-01-21 09:58:07.177395: step: 1236/526, loss: 0.03098478354513645 2023-01-21 09:58:08.335469: step: 1240/526, loss: 0.07389402389526367 2023-01-21 09:58:09.488955: step: 1244/526, loss: 0.14863519370555878 2023-01-21 09:58:10.621131: step: 1248/526, loss: 0.032288551330566406 2023-01-21 09:58:11.756640: step: 1252/526, loss: 0.5651698112487793 2023-01-21 09:58:12.915117: step: 1256/526, loss: 0.0921720489859581 2023-01-21 09:58:14.038211: step: 1260/526, loss: 0.20742206275463104 2023-01-21 09:58:15.202490: step: 1264/526, loss: 0.1320362091064453 2023-01-21 09:58:16.331200: step: 1268/526, loss: 0.10012664645910263 2023-01-21 09:58:17.489176: step: 1272/526, loss: 0.09101162105798721 2023-01-21 09:58:18.645594: step: 1276/526, loss: 0.2577333450317383 2023-01-21 09:58:19.794661: step: 1280/526, loss: 0.7802594900131226 2023-01-21 09:58:20.923422: step: 1284/526, loss: 0.35978859663009644 2023-01-21 09:58:22.086606: step: 1288/526, loss: 0.05459742620587349 2023-01-21 09:58:23.239557: step: 1292/526, loss: 0.04782591015100479 2023-01-21 09:58:24.336146: step: 1296/526, loss: 0.06419219821691513 2023-01-21 09:58:25.490842: step: 1300/526, loss: 0.03440551832318306 2023-01-21 09:58:26.626869: step: 1304/526, loss: 0.15025176107883453 2023-01-21 09:58:27.804937: step: 1308/526, loss: 0.06674442440271378 2023-01-21 09:58:28.994647: step: 1312/526, loss: 0.3306504189968109 2023-01-21 09:58:30.145530: step: 1316/526, loss: 0.23197859525680542 2023-01-21 09:58:31.248278: step: 1320/526, loss: 0.22558268904685974 2023-01-21 09:58:32.397197: step: 1324/526, loss: 0.015024662017822266 2023-01-21 09:58:33.531268: step: 1328/526, loss: 0.2526015341281891 2023-01-21 09:58:34.700016: step: 1332/526, loss: 0.1268499344587326 2023-01-21 09:58:35.833313: step: 1336/526, loss: 0.21193313598632812 2023-01-21 09:58:36.967640: step: 1340/526, loss: 0.028816748410463333 2023-01-21 09:58:38.125264: step: 1344/526, loss: 0.10718526691198349 2023-01-21 09:58:39.265383: step: 1348/526, loss: 0.13887767493724823 2023-01-21 09:58:40.382358: step: 1352/526, loss: 0.07422390580177307 2023-01-21 09:58:41.516654: step: 1356/526, loss: 0.028196675702929497 2023-01-21 09:58:42.671481: step: 1360/526, loss: 0.03765721619129181 2023-01-21 09:58:43.802580: step: 1364/526, loss: 0.5351154804229736 2023-01-21 09:58:44.906313: step: 1368/526, loss: 0.28474217653274536 2023-01-21 09:58:46.022601: step: 1372/526, loss: 0.34652385115623474 2023-01-21 09:58:47.157822: step: 1376/526, loss: 0.03533210977911949 2023-01-21 09:58:48.312851: step: 1380/526, loss: 0.06955623626708984 2023-01-21 09:58:49.479208: step: 1384/526, loss: 0.222881019115448 2023-01-21 09:58:50.617005: step: 1388/526, loss: 0.06616034358739853 2023-01-21 09:58:51.775751: step: 1392/526, loss: 0.08102045208215714 2023-01-21 09:58:52.970325: step: 1396/526, loss: 0.04788055270910263 2023-01-21 09:58:54.114802: step: 1400/526, loss: 0.6647990942001343 2023-01-21 09:58:55.239335: step: 1404/526, loss: 0.06681656837463379 2023-01-21 09:58:56.367945: step: 1408/526, loss: 0.010775423608720303 2023-01-21 09:58:57.501097: step: 1412/526, loss: 0.12645049393177032 2023-01-21 09:58:58.640239: step: 1416/526, loss: 0.05009651184082031 2023-01-21 09:58:59.797822: step: 1420/526, loss: 0.0318637490272522 2023-01-21 09:59:00.924839: step: 1424/526, loss: 0.12748508155345917 2023-01-21 09:59:02.076637: step: 1428/526, loss: 0.028865624219179153 2023-01-21 09:59:03.247259: step: 1432/526, loss: 0.16792979836463928 2023-01-21 09:59:04.380657: step: 1436/526, loss: 0.06971435993909836 2023-01-21 09:59:05.513929: step: 1440/526, loss: 0.14157867431640625 2023-01-21 09:59:06.663627: step: 1444/526, loss: 0.06895770877599716 2023-01-21 09:59:07.803782: step: 1448/526, loss: 0.04891672357916832 2023-01-21 09:59:08.914863: step: 1452/526, loss: 0.26673200726509094 2023-01-21 09:59:10.047166: step: 1456/526, loss: 0.003355693770572543 2023-01-21 09:59:11.179559: step: 1460/526, loss: 0.06873917579650879 2023-01-21 09:59:12.351284: step: 1464/526, loss: 0.12451129406690598 2023-01-21 09:59:13.500718: step: 1468/526, loss: 0.079241082072258 2023-01-21 09:59:14.647615: step: 1472/526, loss: 0.05995222553610802 2023-01-21 09:59:15.789326: step: 1476/526, loss: 0.16911935806274414 2023-01-21 09:59:16.922056: step: 1480/526, loss: 0.13927793502807617 2023-01-21 09:59:18.033747: step: 1484/526, loss: 0.05743388831615448 2023-01-21 09:59:19.147862: step: 1488/526, loss: 0.07564163208007812 2023-01-21 09:59:20.252612: step: 1492/526, loss: 0.18532094359397888 2023-01-21 09:59:21.431900: step: 1496/526, loss: 0.028899766504764557 2023-01-21 09:59:22.533627: step: 1500/526, loss: 0.01598033867776394 2023-01-21 09:59:23.659126: step: 1504/526, loss: 0.04201669618487358 2023-01-21 09:59:24.802067: step: 1508/526, loss: 0.0795467421412468 2023-01-21 09:59:25.939428: step: 1512/526, loss: 0.09611830860376358 2023-01-21 09:59:27.103548: step: 1516/526, loss: 0.16045865416526794 2023-01-21 09:59:28.280531: step: 1520/526, loss: 0.8075896501541138 2023-01-21 09:59:29.425341: step: 1524/526, loss: 0.07167863845825195 2023-01-21 09:59:30.587206: step: 1528/526, loss: 0.052689362317323685 2023-01-21 09:59:31.718229: step: 1532/526, loss: 0.23393507301807404 2023-01-21 09:59:32.868156: step: 1536/526, loss: 0.2833879590034485 2023-01-21 09:59:33.995954: step: 1540/526, loss: 0.07227067649364471 2023-01-21 09:59:35.127435: step: 1544/526, loss: 0.002843809314072132 2023-01-21 09:59:36.266999: step: 1548/526, loss: 0.11440396308898926 2023-01-21 09:59:37.418602: step: 1552/526, loss: 0.09573821723461151 2023-01-21 09:59:38.577465: step: 1556/526, loss: 0.12741537392139435 2023-01-21 09:59:39.694219: step: 1560/526, loss: 0.09237485378980637 2023-01-21 09:59:40.855909: step: 1564/526, loss: 0.09336452931165695 2023-01-21 09:59:41.997261: step: 1568/526, loss: 0.07695751637220383 2023-01-21 09:59:43.170175: step: 1572/526, loss: 0.21299171447753906 2023-01-21 09:59:44.342121: step: 1576/526, loss: 0.05524163320660591 2023-01-21 09:59:45.484910: step: 1580/526, loss: 0.09359641373157501 2023-01-21 09:59:46.640864: step: 1584/526, loss: 0.2421269416809082 2023-01-21 09:59:47.765827: step: 1588/526, loss: 0.21581211686134338 2023-01-21 09:59:48.886481: step: 1592/526, loss: 0.21436166763305664 2023-01-21 09:59:50.012940: step: 1596/526, loss: 0.06375961750745773 2023-01-21 09:59:51.137043: step: 1600/526, loss: 0.2839374542236328 2023-01-21 09:59:52.269380: step: 1604/526, loss: 0.12226247787475586 2023-01-21 09:59:53.408495: step: 1608/526, loss: 0.10494418442249298 2023-01-21 09:59:54.533039: step: 1612/526, loss: 0.024761630222201347 2023-01-21 09:59:55.660499: step: 1616/526, loss: 0.17037610709667206 2023-01-21 09:59:56.825955: step: 1620/526, loss: 0.07723579555749893 2023-01-21 09:59:57.962113: step: 1624/526, loss: 0.2092466801404953 2023-01-21 09:59:59.114655: step: 1628/526, loss: 0.05044609308242798 2023-01-21 10:00:00.240818: step: 1632/526, loss: 0.029678424820303917 2023-01-21 10:00:01.398472: step: 1636/526, loss: 0.40183210372924805 2023-01-21 10:00:02.526983: step: 1640/526, loss: 0.08136948943138123 2023-01-21 10:00:03.657700: step: 1644/526, loss: 0.027004718780517578 2023-01-21 10:00:04.774791: step: 1648/526, loss: 0.01864652708172798 2023-01-21 10:00:05.942605: step: 1652/526, loss: 0.21829204261302948 2023-01-21 10:00:07.117982: step: 1656/526, loss: 0.2569066882133484 2023-01-21 10:00:08.266550: step: 1660/526, loss: 0.026708031073212624 2023-01-21 10:00:09.390306: step: 1664/526, loss: 0.06775055080652237 2023-01-21 10:00:10.532120: step: 1668/526, loss: 0.44076114892959595 2023-01-21 10:00:11.660105: step: 1672/526, loss: 0.054117538034915924 2023-01-21 10:00:12.800600: step: 1676/526, loss: 0.13281765580177307 2023-01-21 10:00:13.965075: step: 1680/526, loss: 0.365447998046875 2023-01-21 10:00:15.124267: step: 1684/526, loss: 0.12681007385253906 2023-01-21 10:00:16.270344: step: 1688/526, loss: 0.16925403475761414 2023-01-21 10:00:17.408548: step: 1692/526, loss: 0.06788118183612823 2023-01-21 10:00:18.545320: step: 1696/526, loss: 0.1189342513680458 2023-01-21 10:00:19.708922: step: 1700/526, loss: 1.2117369174957275 2023-01-21 10:00:20.832967: step: 1704/526, loss: 0.01751413382589817 2023-01-21 10:00:21.965991: step: 1708/526, loss: 0.11873197555541992 2023-01-21 10:00:23.083668: step: 1712/526, loss: 0.03187905251979828 2023-01-21 10:00:24.234290: step: 1716/526, loss: 0.12629947066307068 2023-01-21 10:00:25.371876: step: 1720/526, loss: 0.11023445427417755 2023-01-21 10:00:26.544453: step: 1724/526, loss: 0.11636095494031906 2023-01-21 10:00:27.677250: step: 1728/526, loss: 0.1528707593679428 2023-01-21 10:00:28.807744: step: 1732/526, loss: 0.1430281698703766 2023-01-21 10:00:29.945794: step: 1736/526, loss: 0.17953424155712128 2023-01-21 10:00:31.077547: step: 1740/526, loss: 0.04234180599451065 2023-01-21 10:00:32.222858: step: 1744/526, loss: 0.046218253672122955 2023-01-21 10:00:33.352640: step: 1748/526, loss: 0.29042285680770874 2023-01-21 10:00:34.478188: step: 1752/526, loss: 0.3707937002182007 2023-01-21 10:00:35.592863: step: 1756/526, loss: 0.419533908367157 2023-01-21 10:00:36.736428: step: 1760/526, loss: 0.6809808015823364 2023-01-21 10:00:37.864295: step: 1764/526, loss: 0.17096862196922302 2023-01-21 10:00:39.009556: step: 1768/526, loss: 0.1278887391090393 2023-01-21 10:00:40.174634: step: 1772/526, loss: 0.15587787330150604 2023-01-21 10:00:41.309855: step: 1776/526, loss: 0.06174030527472496 2023-01-21 10:00:42.457192: step: 1780/526, loss: 0.15425196290016174 2023-01-21 10:00:43.624985: step: 1784/526, loss: 0.12806811928749084 2023-01-21 10:00:44.774958: step: 1788/526, loss: 0.03488569334149361 2023-01-21 10:00:45.897068: step: 1792/526, loss: 0.15859737992286682 2023-01-21 10:00:47.006393: step: 1796/526, loss: 0.03506731986999512 2023-01-21 10:00:48.121169: step: 1800/526, loss: 0.11343927681446075 2023-01-21 10:00:49.258710: step: 1804/526, loss: 0.1599036157131195 2023-01-21 10:00:50.416398: step: 1808/526, loss: 0.07686062157154083 2023-01-21 10:00:51.567503: step: 1812/526, loss: 0.17101605236530304 2023-01-21 10:00:52.726969: step: 1816/526, loss: 0.10806016623973846 2023-01-21 10:00:53.868660: step: 1820/526, loss: 0.21551105380058289 2023-01-21 10:00:54.997791: step: 1824/526, loss: 0.1386154741048813 2023-01-21 10:00:56.124049: step: 1828/526, loss: 0.04928422346711159 2023-01-21 10:00:57.270807: step: 1832/526, loss: 0.02080964669585228 2023-01-21 10:00:58.412450: step: 1836/526, loss: 0.08711685985326767 2023-01-21 10:00:59.542562: step: 1840/526, loss: 0.1628192961215973 2023-01-21 10:01:00.673042: step: 1844/526, loss: 0.7590702772140503 2023-01-21 10:01:01.810995: step: 1848/526, loss: 0.09123792499303818 2023-01-21 10:01:02.940116: step: 1852/526, loss: 0.01990513876080513 2023-01-21 10:01:04.093145: step: 1856/526, loss: 0.14835429191589355 2023-01-21 10:01:05.216269: step: 1860/526, loss: 0.05281124264001846 2023-01-21 10:01:06.373035: step: 1864/526, loss: 0.05277001112699509 2023-01-21 10:01:07.489683: step: 1868/526, loss: 0.8418101072311401 2023-01-21 10:01:08.662499: step: 1872/526, loss: 0.055646900087594986 2023-01-21 10:01:09.810232: step: 1876/526, loss: 0.10546045005321503 2023-01-21 10:01:10.958999: step: 1880/526, loss: 0.11477122455835342 2023-01-21 10:01:12.077244: step: 1884/526, loss: 0.15601015090942383 2023-01-21 10:01:13.228754: step: 1888/526, loss: 0.022717859596014023 2023-01-21 10:01:14.363385: step: 1892/526, loss: 0.11235609650611877 2023-01-21 10:01:15.499200: step: 1896/526, loss: 0.22358913719654083 2023-01-21 10:01:16.633797: step: 1900/526, loss: 0.03359527513384819 2023-01-21 10:01:17.779677: step: 1904/526, loss: 0.0324854850769043 2023-01-21 10:01:18.908030: step: 1908/526, loss: 0.06537675857543945 2023-01-21 10:01:20.044277: step: 1912/526, loss: 0.20563125610351562 2023-01-21 10:01:21.197407: step: 1916/526, loss: 0.06448874622583389 2023-01-21 10:01:22.303647: step: 1920/526, loss: 0.37325984239578247 2023-01-21 10:01:23.413860: step: 1924/526, loss: 0.05528712272644043 2023-01-21 10:01:24.563445: step: 1928/526, loss: 0.0896279364824295 2023-01-21 10:01:25.728088: step: 1932/526, loss: 0.27288904786109924 2023-01-21 10:01:26.889719: step: 1936/526, loss: 0.14644156396389008 2023-01-21 10:01:28.050453: step: 1940/526, loss: 0.3932479918003082 2023-01-21 10:01:29.182018: step: 1944/526, loss: 0.0750247985124588 2023-01-21 10:01:30.334236: step: 1948/526, loss: 0.1488696038722992 2023-01-21 10:01:31.463798: step: 1952/526, loss: 0.01288747787475586 2023-01-21 10:01:32.600079: step: 1956/526, loss: 0.30777817964553833 2023-01-21 10:01:33.720868: step: 1960/526, loss: 0.013335800729691982 2023-01-21 10:01:34.847587: step: 1964/526, loss: 0.09965267032384872 2023-01-21 10:01:35.990558: step: 1968/526, loss: 0.033304404467344284 2023-01-21 10:01:37.135644: step: 1972/526, loss: 0.10899333655834198 2023-01-21 10:01:38.288337: step: 1976/526, loss: 1.0351192951202393 2023-01-21 10:01:39.419329: step: 1980/526, loss: 0.02187070995569229 2023-01-21 10:01:40.561764: step: 1984/526, loss: 0.11713247001171112 2023-01-21 10:01:41.698716: step: 1988/526, loss: 0.06531868129968643 2023-01-21 10:01:42.866611: step: 1992/526, loss: 0.3635888993740082 2023-01-21 10:01:44.017374: step: 1996/526, loss: 0.049531079828739166 2023-01-21 10:01:45.182796: step: 2000/526, loss: 0.022366192191839218 2023-01-21 10:01:46.345817: step: 2004/526, loss: 0.19946500658988953 2023-01-21 10:01:47.489418: step: 2008/526, loss: 0.36543503403663635 2023-01-21 10:01:48.665145: step: 2012/526, loss: 0.13704414665699005 2023-01-21 10:01:49.812699: step: 2016/526, loss: 0.1697574108839035 2023-01-21 10:01:50.946137: step: 2020/526, loss: 0.029215335845947266 2023-01-21 10:01:52.085397: step: 2024/526, loss: 0.02939128875732422 2023-01-21 10:01:53.198666: step: 2028/526, loss: 0.159742534160614 2023-01-21 10:01:54.364327: step: 2032/526, loss: 0.7922096252441406 2023-01-21 10:01:55.515853: step: 2036/526, loss: 0.04607725143432617 2023-01-21 10:01:56.643430: step: 2040/526, loss: 0.07328657805919647 2023-01-21 10:01:57.803015: step: 2044/526, loss: 0.050852395594120026 2023-01-21 10:01:58.979164: step: 2048/526, loss: 0.020374394953250885 2023-01-21 10:02:00.132188: step: 2052/526, loss: 0.06905481219291687 2023-01-21 10:02:01.284441: step: 2056/526, loss: 0.023433685302734375 2023-01-21 10:02:02.452316: step: 2060/526, loss: 0.07002801448106766 2023-01-21 10:02:03.633649: step: 2064/526, loss: 1.9444684982299805 2023-01-21 10:02:04.782141: step: 2068/526, loss: 0.061034154146909714 2023-01-21 10:02:05.918263: step: 2072/526, loss: 0.052844252437353134 2023-01-21 10:02:07.073769: step: 2076/526, loss: 0.0982128232717514 2023-01-21 10:02:08.236048: step: 2080/526, loss: 0.12395502626895905 2023-01-21 10:02:09.407857: step: 2084/526, loss: 0.08458928763866425 2023-01-21 10:02:10.538288: step: 2088/526, loss: 0.05322227627038956 2023-01-21 10:02:11.671886: step: 2092/526, loss: 0.021163761615753174 2023-01-21 10:02:12.841782: step: 2096/526, loss: 0.7529973983764648 2023-01-21 10:02:13.985573: step: 2100/526, loss: 0.06707258522510529 2023-01-21 10:02:15.125442: step: 2104/526, loss: 0.2566892206668854 ================================================== Loss: 0.155 -------------------- Dev: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Test: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Chinese: {'event': {'p': 0.5569620253164557, 'r': 0.8148148148148148, 'f1': 0.6616541353383458}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Russian: {'event': {'p': 0.37777777777777777, 'r': 0.4722222222222222, 'f1': 0.4197530864197531}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} New best korean model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5563041385948027, 'r': 0.7696404793608522, 'f1': 0.6458100558659218}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Eng Test for Russian: {'event': {'p': 0.5336806847785635, 'r': 0.7484342379958246, 'f1': 0.6230719096241583}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} Sample Russian: {'event': {'p': 0.5116279069767442, 'r': 0.6111111111111112, 'f1': 0.5569620253164557}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 4} ****************************** Epoch: 8 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:03:05.085723: step: 4/526, loss: 0.4509592652320862 2023-01-21 10:03:06.244436: step: 8/526, loss: 0.009682273492217064 2023-01-21 10:03:07.378610: step: 12/526, loss: 0.20073539018630981 2023-01-21 10:03:08.535474: step: 16/526, loss: 0.04364137724041939 2023-01-21 10:03:09.668916: step: 20/526, loss: 0.12309151142835617 2023-01-21 10:03:10.809670: step: 24/526, loss: 0.05186261981725693 2023-01-21 10:03:11.983356: step: 28/526, loss: 0.03564329445362091 2023-01-21 10:03:13.089598: step: 32/526, loss: 0.11937236785888672 2023-01-21 10:03:14.194448: step: 36/526, loss: 0.13100528717041016 2023-01-21 10:03:15.337380: step: 40/526, loss: 0.12632770836353302 2023-01-21 10:03:16.502281: step: 44/526, loss: 0.08592571318149567 2023-01-21 10:03:17.676347: step: 48/526, loss: 0.07140274345874786 2023-01-21 10:03:18.827223: step: 52/526, loss: 0.17488251626491547 2023-01-21 10:03:19.974169: step: 56/526, loss: 0.1383393406867981 2023-01-21 10:03:21.127752: step: 60/526, loss: 0.027023887261748314 2023-01-21 10:03:22.241493: step: 64/526, loss: 0.02628622017800808 2023-01-21 10:03:23.369340: step: 68/526, loss: 0.010891628451645374 2023-01-21 10:03:24.513158: step: 72/526, loss: 0.07957735657691956 2023-01-21 10:03:25.641583: step: 76/526, loss: 0.18381062150001526 2023-01-21 10:03:26.788656: step: 80/526, loss: 0.13099898397922516 2023-01-21 10:03:27.923894: step: 84/526, loss: 0.11448946595191956 2023-01-21 10:03:29.045751: step: 88/526, loss: 0.07045440375804901 2023-01-21 10:03:30.161677: step: 92/526, loss: 0.014455914497375488 2023-01-21 10:03:31.249806: step: 96/526, loss: 0.8065840601921082 2023-01-21 10:03:32.357840: step: 100/526, loss: 0.04395761340856552 2023-01-21 10:03:33.504546: step: 104/526, loss: 0.09363555908203125 2023-01-21 10:03:34.629579: step: 108/526, loss: 0.12660837173461914 2023-01-21 10:03:35.768538: step: 112/526, loss: 0.018051816150546074 2023-01-21 10:03:36.907417: step: 116/526, loss: 0.06137971952557564 2023-01-21 10:03:38.070199: step: 120/526, loss: 0.07128267735242844 2023-01-21 10:03:39.202333: step: 124/526, loss: 0.16976423561573029 2023-01-21 10:03:40.322846: step: 128/526, loss: 0.05111045762896538 2023-01-21 10:03:41.425159: step: 132/526, loss: 0.07206392288208008 2023-01-21 10:03:42.540642: step: 136/526, loss: 0.049035023897886276 2023-01-21 10:03:43.677452: step: 140/526, loss: 0.062110044062137604 2023-01-21 10:03:44.798352: step: 144/526, loss: 0.07096338272094727 2023-01-21 10:03:45.993974: step: 148/526, loss: 0.4366329312324524 2023-01-21 10:03:47.138866: step: 152/526, loss: 0.060074903070926666 2023-01-21 10:03:48.303751: step: 156/526, loss: 0.08557100594043732 2023-01-21 10:03:49.473047: step: 160/526, loss: 0.08230438083410263 2023-01-21 10:03:50.606052: step: 164/526, loss: 0.03566126897931099 2023-01-21 10:03:51.727271: step: 168/526, loss: 0.09695682674646378 2023-01-21 10:03:52.868186: step: 172/526, loss: 0.05592308193445206 2023-01-21 10:03:54.010280: step: 176/526, loss: 0.019489290192723274 2023-01-21 10:03:55.165606: step: 180/526, loss: 0.07788868248462677 2023-01-21 10:03:56.341528: step: 184/526, loss: 0.06774282455444336 2023-01-21 10:03:57.490410: step: 188/526, loss: 0.10261721909046173 2023-01-21 10:03:58.629097: step: 192/526, loss: 0.05845542252063751 2023-01-21 10:03:59.751251: step: 196/526, loss: 0.05750560760498047 2023-01-21 10:04:00.860049: step: 200/526, loss: 0.02927863597869873 2023-01-21 10:04:01.957750: step: 204/526, loss: 0.08807583153247833 2023-01-21 10:04:03.103862: step: 208/526, loss: 0.047608375549316406 2023-01-21 10:04:04.218777: step: 212/526, loss: 0.099675752222538 2023-01-21 10:04:05.361090: step: 216/526, loss: 0.11963607370853424 2023-01-21 10:04:06.486215: step: 220/526, loss: 0.11921291798353195 2023-01-21 10:04:07.616602: step: 224/526, loss: 0.2667018175125122 2023-01-21 10:04:08.772164: step: 228/526, loss: 0.1565166860818863 2023-01-21 10:04:09.900917: step: 232/526, loss: 0.459891140460968 2023-01-21 10:04:11.044284: step: 236/526, loss: 0.023046493530273438 2023-01-21 10:04:12.161963: step: 240/526, loss: 2.904085636138916 2023-01-21 10:04:13.265552: step: 244/526, loss: 0.021080588921904564 2023-01-21 10:04:14.400311: step: 248/526, loss: 0.014219284057617188 2023-01-21 10:04:15.531753: step: 252/526, loss: 0.025008201599121094 2023-01-21 10:04:16.651300: step: 256/526, loss: 0.05641651153564453 2023-01-21 10:04:17.804602: step: 260/526, loss: 0.0062920572236180305 2023-01-21 10:04:18.955198: step: 264/526, loss: 0.12470598518848419 2023-01-21 10:04:20.105330: step: 268/526, loss: 0.08369922637939453 2023-01-21 10:04:21.240673: step: 272/526, loss: 0.07556037604808807 2023-01-21 10:04:22.359867: step: 276/526, loss: 0.08341007679700851 2023-01-21 10:04:23.501150: step: 280/526, loss: 0.11428480595350266 2023-01-21 10:04:24.675085: step: 284/526, loss: 0.9325697422027588 2023-01-21 10:04:25.821830: step: 288/526, loss: 0.18574738502502441 2023-01-21 10:04:26.975022: step: 292/526, loss: 0.02388610877096653 2023-01-21 10:04:28.137011: step: 296/526, loss: 0.04833984375 2023-01-21 10:04:29.304907: step: 300/526, loss: 0.020642375573515892 2023-01-21 10:04:30.432076: step: 304/526, loss: 0.1163792610168457 2023-01-21 10:04:31.590573: step: 308/526, loss: 0.02852320671081543 2023-01-21 10:04:32.725880: step: 312/526, loss: 0.0456182025372982 2023-01-21 10:04:33.836841: step: 316/526, loss: 0.12594766914844513 2023-01-21 10:04:34.967487: step: 320/526, loss: 0.17843884229660034 2023-01-21 10:04:36.104431: step: 324/526, loss: 0.020578479394316673 2023-01-21 10:04:37.267555: step: 328/526, loss: 0.04206524044275284 2023-01-21 10:04:38.403944: step: 332/526, loss: 0.10274887084960938 2023-01-21 10:04:39.531846: step: 336/526, loss: 0.4566787779331207 2023-01-21 10:04:40.667515: step: 340/526, loss: 0.09298267215490341 2023-01-21 10:04:41.787354: step: 344/526, loss: 0.08521173894405365 2023-01-21 10:04:42.919501: step: 348/526, loss: 0.16558514535427094 2023-01-21 10:04:44.059003: step: 352/526, loss: 0.01702265813946724 2023-01-21 10:04:45.176405: step: 356/526, loss: 0.32536476850509644 2023-01-21 10:04:46.321815: step: 360/526, loss: 0.11419686675071716 2023-01-21 10:04:47.434445: step: 364/526, loss: 0.18552808463573456 2023-01-21 10:04:48.571750: step: 368/526, loss: 0.7736044526100159 2023-01-21 10:04:49.721185: step: 372/526, loss: 0.05100994184613228 2023-01-21 10:04:50.841472: step: 376/526, loss: 0.039560794830322266 2023-01-21 10:04:51.990951: step: 380/526, loss: 0.06244611740112305 2023-01-21 10:04:53.133316: step: 384/526, loss: 0.01614541932940483 2023-01-21 10:04:54.276926: step: 388/526, loss: 0.05539894104003906 2023-01-21 10:04:55.378874: step: 392/526, loss: 0.07028555870056152 2023-01-21 10:04:56.520820: step: 396/526, loss: 0.027847956866025925 2023-01-21 10:04:57.658733: step: 400/526, loss: 0.09425429999828339 2023-01-21 10:04:58.772095: step: 404/526, loss: 0.05379796028137207 2023-01-21 10:04:59.896871: step: 408/526, loss: 0.36586126685142517 2023-01-21 10:05:01.048235: step: 412/526, loss: 0.13255101442337036 2023-01-21 10:05:02.214259: step: 416/526, loss: 0.011668015271425247 2023-01-21 10:05:03.365388: step: 420/526, loss: 0.0749419704079628 2023-01-21 10:05:04.477303: step: 424/526, loss: 0.01673254929482937 2023-01-21 10:05:05.576600: step: 428/526, loss: 1.02268385887146 2023-01-21 10:05:06.725070: step: 432/526, loss: 0.21064051985740662 2023-01-21 10:05:07.861305: step: 436/526, loss: 0.25405243039131165 2023-01-21 10:05:09.013462: step: 440/526, loss: 0.0852547436952591 2023-01-21 10:05:10.156782: step: 444/526, loss: 0.006771135609596968 2023-01-21 10:05:11.277547: step: 448/526, loss: 0.02016305923461914 2023-01-21 10:05:12.426219: step: 452/526, loss: 0.01931324042379856 2023-01-21 10:05:13.581297: step: 456/526, loss: 0.046138957142829895 2023-01-21 10:05:14.703233: step: 460/526, loss: 0.01810484007000923 2023-01-21 10:05:15.838932: step: 464/526, loss: 0.2569112777709961 2023-01-21 10:05:17.004366: step: 468/526, loss: 0.0521366149187088 2023-01-21 10:05:18.131258: step: 472/526, loss: 0.21542473137378693 2023-01-21 10:05:19.247703: step: 476/526, loss: 0.10165233910083771 2023-01-21 10:05:20.395020: step: 480/526, loss: 0.1466052085161209 2023-01-21 10:05:21.563906: step: 484/526, loss: 0.06927242130041122 2023-01-21 10:05:22.725412: step: 488/526, loss: 0.1103677749633789 2023-01-21 10:05:23.834087: step: 492/526, loss: 0.02180485799908638 2023-01-21 10:05:24.959609: step: 496/526, loss: 0.13236021995544434 2023-01-21 10:05:26.107252: step: 500/526, loss: 0.008610916323959827 2023-01-21 10:05:27.257164: step: 504/526, loss: 0.05118846893310547 2023-01-21 10:05:28.384327: step: 508/526, loss: 0.02141270600259304 2023-01-21 10:05:29.571183: step: 512/526, loss: 0.0804070457816124 2023-01-21 10:05:30.697062: step: 516/526, loss: 0.11349506676197052 2023-01-21 10:05:31.829982: step: 520/526, loss: 0.36216410994529724 2023-01-21 10:05:32.965140: step: 524/526, loss: 0.033472299575805664 2023-01-21 10:05:34.139745: step: 528/526, loss: 0.024210095405578613 2023-01-21 10:05:35.271595: step: 532/526, loss: 0.0711517333984375 2023-01-21 10:05:36.381932: step: 536/526, loss: 0.1883375197649002 2023-01-21 10:05:37.507007: step: 540/526, loss: 0.03312330320477486 2023-01-21 10:05:38.659533: step: 544/526, loss: 0.16953782737255096 2023-01-21 10:05:39.774149: step: 548/526, loss: 0.07899179309606552 2023-01-21 10:05:40.904918: step: 552/526, loss: 0.0881553664803505 2023-01-21 10:05:42.101797: step: 556/526, loss: 0.05763111263513565 2023-01-21 10:05:43.266633: step: 560/526, loss: 0.10114964842796326 2023-01-21 10:05:44.399201: step: 564/526, loss: 0.07979154586791992 2023-01-21 10:05:45.567539: step: 568/526, loss: 0.21123342216014862 2023-01-21 10:05:46.702735: step: 572/526, loss: 0.049878671765327454 2023-01-21 10:05:47.835102: step: 576/526, loss: 0.057035066187381744 2023-01-21 10:05:48.982658: step: 580/526, loss: 0.13447979092597961 2023-01-21 10:05:50.082330: step: 584/526, loss: 0.16881036758422852 2023-01-21 10:05:51.231748: step: 588/526, loss: 0.08310695737600327 2023-01-21 10:05:52.377393: step: 592/526, loss: 0.14080849289894104 2023-01-21 10:05:53.510170: step: 596/526, loss: 0.12084797024726868 2023-01-21 10:05:54.651646: step: 600/526, loss: 0.03946685791015625 2023-01-21 10:05:55.796964: step: 604/526, loss: 0.11840762943029404 2023-01-21 10:05:56.924201: step: 608/526, loss: 0.38136404752731323 2023-01-21 10:05:58.023153: step: 612/526, loss: 0.0014248609077185392 2023-01-21 10:05:59.137804: step: 616/526, loss: 0.03033771552145481 2023-01-21 10:06:00.302430: step: 620/526, loss: 0.030315257608890533 2023-01-21 10:06:01.461283: step: 624/526, loss: 0.08245439827442169 2023-01-21 10:06:02.587465: step: 628/526, loss: 0.017532777041196823 2023-01-21 10:06:03.739375: step: 632/526, loss: 0.2039492428302765 2023-01-21 10:06:04.894026: step: 636/526, loss: 0.027325155213475227 2023-01-21 10:06:06.015659: step: 640/526, loss: 0.07786641269922256 2023-01-21 10:06:07.148986: step: 644/526, loss: 0.009795594029128551 2023-01-21 10:06:08.291695: step: 648/526, loss: 0.14057111740112305 2023-01-21 10:06:09.439148: step: 652/526, loss: 0.08568086475133896 2023-01-21 10:06:10.571478: step: 656/526, loss: 0.12078313529491425 2023-01-21 10:06:11.741287: step: 660/526, loss: 0.07834787666797638 2023-01-21 10:06:12.904966: step: 664/526, loss: 0.10383033752441406 2023-01-21 10:06:14.081948: step: 668/526, loss: 0.05333023518323898 2023-01-21 10:06:15.222378: step: 672/526, loss: 0.08539829403162003 2023-01-21 10:06:16.368759: step: 676/526, loss: 0.04530048370361328 2023-01-21 10:06:17.507288: step: 680/526, loss: 0.019302750006318092 2023-01-21 10:06:18.658144: step: 684/526, loss: 0.44927406311035156 2023-01-21 10:06:19.813015: step: 688/526, loss: 0.07670411467552185 2023-01-21 10:06:20.989606: step: 692/526, loss: 0.03258876875042915 2023-01-21 10:06:22.128587: step: 696/526, loss: 0.11091528087854385 2023-01-21 10:06:23.247472: step: 700/526, loss: 0.01544036902487278 2023-01-21 10:06:24.375061: step: 704/526, loss: 0.036835383623838425 2023-01-21 10:06:25.522540: step: 708/526, loss: 0.031983375549316406 2023-01-21 10:06:26.659009: step: 712/526, loss: 0.026570988819003105 2023-01-21 10:06:27.794722: step: 716/526, loss: 0.013112116605043411 2023-01-21 10:06:28.927857: step: 720/526, loss: 0.06713657081127167 2023-01-21 10:06:30.089538: step: 724/526, loss: 0.18842096626758575 2023-01-21 10:06:31.227604: step: 728/526, loss: 0.3138576149940491 2023-01-21 10:06:32.366713: step: 732/526, loss: 0.030364610254764557 2023-01-21 10:06:33.484044: step: 736/526, loss: 0.01774129830300808 2023-01-21 10:06:34.644941: step: 740/526, loss: 0.02115321159362793 2023-01-21 10:06:35.762337: step: 744/526, loss: 0.10952109843492508 2023-01-21 10:06:36.885884: step: 748/526, loss: 0.1439904272556305 2023-01-21 10:06:38.030985: step: 752/526, loss: 0.015142631717026234 2023-01-21 10:06:39.163300: step: 756/526, loss: 0.09855957329273224 2023-01-21 10:06:40.315465: step: 760/526, loss: 0.04211006313562393 2023-01-21 10:06:41.470860: step: 764/526, loss: 0.2005455195903778 2023-01-21 10:06:42.621970: step: 768/526, loss: 0.6141564846038818 2023-01-21 10:06:43.744653: step: 772/526, loss: 0.09879302978515625 2023-01-21 10:06:44.856961: step: 776/526, loss: 0.4876483976840973 2023-01-21 10:06:45.990694: step: 780/526, loss: 0.11870536208152771 2023-01-21 10:06:47.132492: step: 784/526, loss: 0.03438301011919975 2023-01-21 10:06:48.283185: step: 788/526, loss: 0.07214164733886719 2023-01-21 10:06:49.429626: step: 792/526, loss: 0.04700946807861328 2023-01-21 10:06:50.580127: step: 796/526, loss: 0.005422401241958141 2023-01-21 10:06:51.717433: step: 800/526, loss: 0.07056216895580292 2023-01-21 10:06:52.843547: step: 804/526, loss: 0.11131231486797333 2023-01-21 10:06:53.956804: step: 808/526, loss: 0.15480156242847443 2023-01-21 10:06:55.089719: step: 812/526, loss: 0.026541804894804955 2023-01-21 10:06:56.210689: step: 816/526, loss: 0.06710920482873917 2023-01-21 10:06:57.316560: step: 820/526, loss: 0.11994200199842453 2023-01-21 10:06:58.463207: step: 824/526, loss: 0.01672639697790146 2023-01-21 10:06:59.589668: step: 828/526, loss: 0.0629308670759201 2023-01-21 10:07:00.719907: step: 832/526, loss: 0.14385686814785004 2023-01-21 10:07:01.821841: step: 836/526, loss: 0.010058593936264515 2023-01-21 10:07:02.982347: step: 840/526, loss: 0.07296065986156464 2023-01-21 10:07:04.116955: step: 844/526, loss: 0.2944096624851227 2023-01-21 10:07:05.259481: step: 848/526, loss: 0.022469425573945045 2023-01-21 10:07:06.406342: step: 852/526, loss: 0.46562662720680237 2023-01-21 10:07:07.534585: step: 856/526, loss: 0.029773712158203125 2023-01-21 10:07:08.673158: step: 860/526, loss: 0.13548488914966583 2023-01-21 10:07:09.833388: step: 864/526, loss: 0.09199275821447372 2023-01-21 10:07:10.983335: step: 868/526, loss: 0.1318870633840561 2023-01-21 10:07:12.147155: step: 872/526, loss: 0.2812376022338867 2023-01-21 10:07:13.345114: step: 876/526, loss: 0.040105536580085754 2023-01-21 10:07:14.492747: step: 880/526, loss: 0.18516980111598969 2023-01-21 10:07:15.646772: step: 884/526, loss: 0.5230588316917419 2023-01-21 10:07:16.806367: step: 888/526, loss: 0.08030033111572266 2023-01-21 10:07:17.904076: step: 892/526, loss: 0.06456517428159714 2023-01-21 10:07:19.037934: step: 896/526, loss: 0.11795501410961151 2023-01-21 10:07:20.151433: step: 900/526, loss: 0.02095651626586914 2023-01-21 10:07:21.304591: step: 904/526, loss: 0.023101331666111946 2023-01-21 10:07:22.446039: step: 908/526, loss: 0.04510049894452095 2023-01-21 10:07:23.584078: step: 912/526, loss: 0.01420135423541069 2023-01-21 10:07:24.734292: step: 916/526, loss: 0.014659309759736061 2023-01-21 10:07:25.892566: step: 920/526, loss: 0.2086995393037796 2023-01-21 10:07:27.025654: step: 924/526, loss: 0.12090644985437393 2023-01-21 10:07:28.142527: step: 928/526, loss: 0.12550897896289825 2023-01-21 10:07:29.281251: step: 932/526, loss: 0.10923528671264648 2023-01-21 10:07:30.424766: step: 936/526, loss: 0.13390541076660156 2023-01-21 10:07:31.567904: step: 940/526, loss: 0.31669092178344727 2023-01-21 10:07:32.694587: step: 944/526, loss: 0.08669109642505646 2023-01-21 10:07:33.804797: step: 948/526, loss: 0.004893970675766468 2023-01-21 10:07:34.942935: step: 952/526, loss: 0.07659421116113663 2023-01-21 10:07:36.083466: step: 956/526, loss: 0.04791421815752983 2023-01-21 10:07:37.234332: step: 960/526, loss: 0.0898553803563118 2023-01-21 10:07:38.402526: step: 964/526, loss: 0.1253407597541809 2023-01-21 10:07:39.557096: step: 968/526, loss: 0.09322299808263779 2023-01-21 10:07:40.689679: step: 972/526, loss: 0.032363034784793854 2023-01-21 10:07:41.841418: step: 976/526, loss: 0.12266630679368973 2023-01-21 10:07:43.011786: step: 980/526, loss: 0.12242374569177628 2023-01-21 10:07:44.153680: step: 984/526, loss: 0.0991489440202713 2023-01-21 10:07:45.290621: step: 988/526, loss: 0.35024648904800415 2023-01-21 10:07:46.402503: step: 992/526, loss: 0.03630800172686577 2023-01-21 10:07:47.558427: step: 996/526, loss: 0.07523689419031143 2023-01-21 10:07:48.685488: step: 1000/526, loss: 0.03704657778143883 2023-01-21 10:07:49.835413: step: 1004/526, loss: 0.265733927488327 2023-01-21 10:07:50.971300: step: 1008/526, loss: 0.06460990756750107 2023-01-21 10:07:52.136157: step: 1012/526, loss: 0.08989782631397247 2023-01-21 10:07:53.265114: step: 1016/526, loss: 0.07344035804271698 2023-01-21 10:07:54.420514: step: 1020/526, loss: 0.23651272058486938 2023-01-21 10:07:55.546261: step: 1024/526, loss: 0.014575100503861904 2023-01-21 10:07:56.668766: step: 1028/526, loss: 1.0429644584655762 2023-01-21 10:07:57.824414: step: 1032/526, loss: 0.1339537650346756 2023-01-21 10:07:58.960640: step: 1036/526, loss: 0.12161803245544434 2023-01-21 10:08:00.081143: step: 1040/526, loss: 0.082611083984375 2023-01-21 10:08:01.183066: step: 1044/526, loss: 0.036517977714538574 2023-01-21 10:08:02.285220: step: 1048/526, loss: 0.05288155376911163 2023-01-21 10:08:03.437350: step: 1052/526, loss: 0.0292205810546875 2023-01-21 10:08:04.557970: step: 1056/526, loss: 0.09984450787305832 2023-01-21 10:08:05.676663: step: 1060/526, loss: 0.03174543380737305 2023-01-21 10:08:06.769997: step: 1064/526, loss: 0.14022216200828552 2023-01-21 10:08:07.905038: step: 1068/526, loss: 0.06268835067749023 2023-01-21 10:08:09.019471: step: 1072/526, loss: 0.26589494943618774 2023-01-21 10:08:10.193618: step: 1076/526, loss: 0.11476574093103409 2023-01-21 10:08:11.348248: step: 1080/526, loss: 0.3752923905849457 2023-01-21 10:08:12.464790: step: 1084/526, loss: 0.02142627164721489 2023-01-21 10:08:13.671921: step: 1088/526, loss: 0.5172773599624634 2023-01-21 10:08:14.807069: step: 1092/526, loss: 0.044009022414684296 2023-01-21 10:08:15.955056: step: 1096/526, loss: 0.08822555840015411 2023-01-21 10:08:17.090720: step: 1100/526, loss: 0.019056033343076706 2023-01-21 10:08:18.237988: step: 1104/526, loss: 0.03240513801574707 2023-01-21 10:08:19.378953: step: 1108/526, loss: 0.08362092822790146 2023-01-21 10:08:20.540198: step: 1112/526, loss: 0.02197418175637722 2023-01-21 10:08:21.687003: step: 1116/526, loss: 0.05853843688964844 2023-01-21 10:08:22.825703: step: 1120/526, loss: 0.030892515555024147 2023-01-21 10:08:23.984924: step: 1124/526, loss: 0.02810525894165039 2023-01-21 10:08:25.159661: step: 1128/526, loss: 0.19932176172733307 2023-01-21 10:08:26.331539: step: 1132/526, loss: 0.13839474320411682 2023-01-21 10:08:27.459407: step: 1136/526, loss: 0.05189061537384987 2023-01-21 10:08:28.571427: step: 1140/526, loss: 0.058396339416503906 2023-01-21 10:08:29.701272: step: 1144/526, loss: 0.05872382968664169 2023-01-21 10:08:30.826664: step: 1148/526, loss: 0.7279013395309448 2023-01-21 10:08:31.955932: step: 1152/526, loss: 0.03947272524237633 2023-01-21 10:08:33.100821: step: 1156/526, loss: 0.00186672224663198 2023-01-21 10:08:34.218390: step: 1160/526, loss: 0.06262989342212677 2023-01-21 10:08:35.354394: step: 1164/526, loss: 0.06388092786073685 2023-01-21 10:08:36.485071: step: 1168/526, loss: 0.018650412559509277 2023-01-21 10:08:37.652746: step: 1172/526, loss: 0.12528343498706818 2023-01-21 10:08:38.793176: step: 1176/526, loss: 0.1625482589006424 2023-01-21 10:08:39.900122: step: 1180/526, loss: 0.025317644700407982 2023-01-21 10:08:41.047077: step: 1184/526, loss: 0.03203906863927841 2023-01-21 10:08:42.219756: step: 1188/526, loss: 0.03545050695538521 2023-01-21 10:08:43.369298: step: 1192/526, loss: 0.04661398008465767 2023-01-21 10:08:44.510083: step: 1196/526, loss: 0.08548067510128021 2023-01-21 10:08:45.653325: step: 1200/526, loss: 0.3050301671028137 2023-01-21 10:08:46.782460: step: 1204/526, loss: 0.010405827313661575 2023-01-21 10:08:47.920710: step: 1208/526, loss: 0.09387321770191193 2023-01-21 10:08:49.028429: step: 1212/526, loss: 0.02991638332605362 2023-01-21 10:08:50.145450: step: 1216/526, loss: 0.20187081396579742 2023-01-21 10:08:51.271217: step: 1220/526, loss: 0.0730188861489296 2023-01-21 10:08:52.440616: step: 1224/526, loss: 0.05721969157457352 2023-01-21 10:08:53.579318: step: 1228/526, loss: 0.044106196612119675 2023-01-21 10:08:54.768892: step: 1232/526, loss: 0.035642530769109726 2023-01-21 10:08:55.945604: step: 1236/526, loss: 0.05703568831086159 2023-01-21 10:08:57.102603: step: 1240/526, loss: 0.016860295087099075 2023-01-21 10:08:58.300822: step: 1244/526, loss: 0.06117277592420578 2023-01-21 10:08:59.465436: step: 1248/526, loss: 0.027732277289032936 2023-01-21 10:09:00.602675: step: 1252/526, loss: 0.04078684002161026 2023-01-21 10:09:01.767740: step: 1256/526, loss: 0.03658314049243927 2023-01-21 10:09:02.928864: step: 1260/526, loss: 0.5496713519096375 2023-01-21 10:09:04.067184: step: 1264/526, loss: 0.05313711240887642 2023-01-21 10:09:05.237263: step: 1268/526, loss: 0.08549108356237411 2023-01-21 10:09:06.397126: step: 1272/526, loss: 0.06908149272203445 2023-01-21 10:09:07.513904: step: 1276/526, loss: 0.014507818035781384 2023-01-21 10:09:08.645554: step: 1280/526, loss: 0.10577650368213654 2023-01-21 10:09:09.784805: step: 1284/526, loss: 0.29365384578704834 2023-01-21 10:09:10.961951: step: 1288/526, loss: 0.04140627384185791 2023-01-21 10:09:12.132444: step: 1292/526, loss: 0.2889742851257324 2023-01-21 10:09:13.288106: step: 1296/526, loss: 0.5034885406494141 2023-01-21 10:09:14.445677: step: 1300/526, loss: 0.13891348242759705 2023-01-21 10:09:15.610814: step: 1304/526, loss: 0.111005499958992 2023-01-21 10:09:16.777881: step: 1308/526, loss: 0.11884107440710068 2023-01-21 10:09:17.945208: step: 1312/526, loss: 0.008743476122617722 2023-01-21 10:09:19.125020: step: 1316/526, loss: 0.10264606773853302 2023-01-21 10:09:20.269223: step: 1320/526, loss: 0.07501468807458878 2023-01-21 10:09:21.440089: step: 1324/526, loss: 0.16715145111083984 2023-01-21 10:09:22.609821: step: 1328/526, loss: 0.09068889915943146 2023-01-21 10:09:23.749983: step: 1332/526, loss: 0.9852508306503296 2023-01-21 10:09:24.883907: step: 1336/526, loss: 0.11147145926952362 2023-01-21 10:09:26.010397: step: 1340/526, loss: 0.08434734493494034 2023-01-21 10:09:27.127169: step: 1344/526, loss: 0.08265219628810883 2023-01-21 10:09:28.268275: step: 1348/526, loss: 0.01094741839915514 2023-01-21 10:09:29.424199: step: 1352/526, loss: 0.07162294536828995 2023-01-21 10:09:30.578336: step: 1356/526, loss: 0.02768879011273384 2023-01-21 10:09:31.692948: step: 1360/526, loss: 0.06671638786792755 2023-01-21 10:09:32.832898: step: 1364/526, loss: 0.46332302689552307 2023-01-21 10:09:33.972924: step: 1368/526, loss: 0.09154520183801651 2023-01-21 10:09:35.089762: step: 1372/526, loss: 0.09058380126953125 2023-01-21 10:09:36.222214: step: 1376/526, loss: 0.11177539825439453 2023-01-21 10:09:37.379833: step: 1380/526, loss: 0.405598908662796 2023-01-21 10:09:38.538517: step: 1384/526, loss: 0.0658111572265625 2023-01-21 10:09:39.688949: step: 1388/526, loss: 0.0830145925283432 2023-01-21 10:09:40.848285: step: 1392/526, loss: 0.025028228759765625 2023-01-21 10:09:41.984175: step: 1396/526, loss: 0.04585247114300728 2023-01-21 10:09:43.127052: step: 1400/526, loss: 0.046415187418460846 2023-01-21 10:09:44.254370: step: 1404/526, loss: 0.08061465620994568 2023-01-21 10:09:45.426497: step: 1408/526, loss: 0.1907079815864563 2023-01-21 10:09:46.559754: step: 1412/526, loss: 0.117953822016716 2023-01-21 10:09:47.700694: step: 1416/526, loss: 0.6954421997070312 2023-01-21 10:09:48.830140: step: 1420/526, loss: 0.06044788658618927 2023-01-21 10:09:49.974869: step: 1424/526, loss: 0.2675694525241852 2023-01-21 10:09:51.108662: step: 1428/526, loss: 0.042508531361818314 2023-01-21 10:09:52.226462: step: 1432/526, loss: 0.013435173779726028 2023-01-21 10:09:53.362097: step: 1436/526, loss: 0.0046634674072265625 2023-01-21 10:09:54.508489: step: 1440/526, loss: 0.14389826357364655 2023-01-21 10:09:55.627221: step: 1444/526, loss: 0.13436804711818695 2023-01-21 10:09:56.795123: step: 1448/526, loss: 1.344647765159607 2023-01-21 10:09:57.938775: step: 1452/526, loss: 0.20523148775100708 2023-01-21 10:09:59.066213: step: 1456/526, loss: 0.09275570511817932 2023-01-21 10:10:00.190663: step: 1460/526, loss: 0.058936458081007004 2023-01-21 10:10:01.319719: step: 1464/526, loss: 0.1774713695049286 2023-01-21 10:10:02.442406: step: 1468/526, loss: 0.18755581974983215 2023-01-21 10:10:03.571371: step: 1472/526, loss: 0.06118784099817276 2023-01-21 10:10:04.694136: step: 1476/526, loss: 0.0021563530899584293 2023-01-21 10:10:05.846607: step: 1480/526, loss: 0.21430516242980957 2023-01-21 10:10:06.982944: step: 1484/526, loss: 0.14232167601585388 2023-01-21 10:10:08.121187: step: 1488/526, loss: 0.09897498786449432 2023-01-21 10:10:09.255328: step: 1492/526, loss: 0.1261250078678131 2023-01-21 10:10:10.400940: step: 1496/526, loss: 0.13158316910266876 2023-01-21 10:10:11.541914: step: 1500/526, loss: 0.04002013057470322 2023-01-21 10:10:12.689960: step: 1504/526, loss: 0.14873509109020233 2023-01-21 10:10:13.866541: step: 1508/526, loss: 0.14446449279785156 2023-01-21 10:10:15.007156: step: 1512/526, loss: 0.08695831894874573 2023-01-21 10:10:16.137725: step: 1516/526, loss: 0.24661703407764435 2023-01-21 10:10:17.254367: step: 1520/526, loss: 0.14397135376930237 2023-01-21 10:10:18.410850: step: 1524/526, loss: 0.1595521867275238 2023-01-21 10:10:19.560667: step: 1528/526, loss: 0.04168272018432617 2023-01-21 10:10:20.692939: step: 1532/526, loss: 0.11637590080499649 2023-01-21 10:10:21.815498: step: 1536/526, loss: 0.20395943522453308 2023-01-21 10:10:22.971922: step: 1540/526, loss: 0.30232658982276917 2023-01-21 10:10:24.124828: step: 1544/526, loss: 0.11829891800880432 2023-01-21 10:10:25.286621: step: 1548/526, loss: 0.06865549087524414 2023-01-21 10:10:26.454909: step: 1552/526, loss: 0.2947154939174652 2023-01-21 10:10:27.585257: step: 1556/526, loss: 0.005326509475708008 2023-01-21 10:10:28.723395: step: 1560/526, loss: 0.04194007068872452 2023-01-21 10:10:29.865533: step: 1564/526, loss: 0.1535348892211914 2023-01-21 10:10:31.012991: step: 1568/526, loss: 0.1386076956987381 2023-01-21 10:10:32.161184: step: 1572/526, loss: 0.1179957389831543 2023-01-21 10:10:33.280787: step: 1576/526, loss: 0.09382247924804688 2023-01-21 10:10:34.395039: step: 1580/526, loss: 0.07488055527210236 2023-01-21 10:10:35.526593: step: 1584/526, loss: 0.03157339245080948 2023-01-21 10:10:36.670694: step: 1588/526, loss: 0.05361314117908478 2023-01-21 10:10:37.814602: step: 1592/526, loss: 0.06537432968616486 2023-01-21 10:10:38.945872: step: 1596/526, loss: 0.02526872232556343 2023-01-21 10:10:40.113316: step: 1600/526, loss: 0.13180875778198242 2023-01-21 10:10:41.251167: step: 1604/526, loss: 0.012719536200165749 2023-01-21 10:10:42.386589: step: 1608/526, loss: 0.015569686889648438 2023-01-21 10:10:43.515194: step: 1612/526, loss: 0.2748333513736725 2023-01-21 10:10:44.655424: step: 1616/526, loss: 0.11798276752233505 2023-01-21 10:10:45.770010: step: 1620/526, loss: 0.059235960245132446 2023-01-21 10:10:46.926423: step: 1624/526, loss: 0.054703906178474426 2023-01-21 10:10:48.042735: step: 1628/526, loss: 0.08646111190319061 2023-01-21 10:10:49.199619: step: 1632/526, loss: 0.13414783775806427 2023-01-21 10:10:50.342598: step: 1636/526, loss: 0.15013009309768677 2023-01-21 10:10:51.488144: step: 1640/526, loss: 0.06078920513391495 2023-01-21 10:10:52.631852: step: 1644/526, loss: 0.05789661407470703 2023-01-21 10:10:53.739769: step: 1648/526, loss: 0.009781932458281517 2023-01-21 10:10:54.907336: step: 1652/526, loss: 0.088352732360363 2023-01-21 10:10:56.070773: step: 1656/526, loss: 0.11192426830530167 2023-01-21 10:10:57.206878: step: 1660/526, loss: 0.1688070297241211 2023-01-21 10:10:58.335280: step: 1664/526, loss: 0.05373840406537056 2023-01-21 10:10:59.495172: step: 1668/526, loss: 0.15043525397777557 2023-01-21 10:11:00.635909: step: 1672/526, loss: 0.293840616941452 2023-01-21 10:11:01.767946: step: 1676/526, loss: 0.0830683708190918 2023-01-21 10:11:02.926132: step: 1680/526, loss: 0.21626701951026917 2023-01-21 10:11:04.055818: step: 1684/526, loss: 0.4208146631717682 2023-01-21 10:11:05.232786: step: 1688/526, loss: 0.09813422709703445 2023-01-21 10:11:06.386211: step: 1692/526, loss: 0.2739488184452057 2023-01-21 10:11:07.501166: step: 1696/526, loss: 0.018120765686035156 2023-01-21 10:11:08.642096: step: 1700/526, loss: 0.27012139558792114 2023-01-21 10:11:09.788625: step: 1704/526, loss: 0.3360114097595215 2023-01-21 10:11:10.899148: step: 1708/526, loss: 0.17406173050403595 2023-01-21 10:11:12.041406: step: 1712/526, loss: 0.015485573559999466 2023-01-21 10:11:13.185057: step: 1716/526, loss: 0.09435214847326279 2023-01-21 10:11:14.344066: step: 1720/526, loss: 0.7589371204376221 2023-01-21 10:11:15.512446: step: 1724/526, loss: 0.05659975856542587 2023-01-21 10:11:16.671384: step: 1728/526, loss: 0.1485244780778885 2023-01-21 10:11:17.853982: step: 1732/526, loss: 0.09022854268550873 2023-01-21 10:11:18.968191: step: 1736/526, loss: 0.02278771437704563 2023-01-21 10:11:20.088926: step: 1740/526, loss: 0.11888370662927628 2023-01-21 10:11:21.210475: step: 1744/526, loss: 0.1495838165283203 2023-01-21 10:11:22.350454: step: 1748/526, loss: 0.08892793953418732 2023-01-21 10:11:23.493657: step: 1752/526, loss: 0.2684028744697571 2023-01-21 10:11:24.639793: step: 1756/526, loss: 0.06688585132360458 2023-01-21 10:11:25.761895: step: 1760/526, loss: 0.015359879471361637 2023-01-21 10:11:26.881553: step: 1764/526, loss: 0.24693012237548828 2023-01-21 10:11:28.045785: step: 1768/526, loss: 0.08525620400905609 2023-01-21 10:11:29.185256: step: 1772/526, loss: 0.22598764300346375 2023-01-21 10:11:30.301460: step: 1776/526, loss: 0.06120362505316734 2023-01-21 10:11:31.453047: step: 1780/526, loss: 0.0734894722700119 2023-01-21 10:11:32.613307: step: 1784/526, loss: 0.011290360242128372 2023-01-21 10:11:33.743100: step: 1788/526, loss: 0.07968025654554367 2023-01-21 10:11:34.852844: step: 1792/526, loss: 0.05257873609662056 2023-01-21 10:11:35.979747: step: 1796/526, loss: 0.11113350093364716 2023-01-21 10:11:37.114593: step: 1800/526, loss: 0.07039546966552734 2023-01-21 10:11:38.232763: step: 1804/526, loss: 0.20111922919750214 2023-01-21 10:11:39.352310: step: 1808/526, loss: 0.017680644989013672 2023-01-21 10:11:40.468665: step: 1812/526, loss: 0.026076506823301315 2023-01-21 10:11:41.612161: step: 1816/526, loss: 0.08258219063282013 2023-01-21 10:11:42.770511: step: 1820/526, loss: 0.08950748294591904 2023-01-21 10:11:43.874653: step: 1824/526, loss: 0.031737376004457474 2023-01-21 10:11:45.035705: step: 1828/526, loss: 0.27701932191848755 2023-01-21 10:11:46.187861: step: 1832/526, loss: 0.013955689035356045 2023-01-21 10:11:47.340576: step: 1836/526, loss: 0.45228147506713867 2023-01-21 10:11:48.496165: step: 1840/526, loss: 0.02460331842303276 2023-01-21 10:11:49.617373: step: 1844/526, loss: 0.0606042854487896 2023-01-21 10:11:50.758707: step: 1848/526, loss: 0.3637872636318207 2023-01-21 10:11:51.883631: step: 1852/526, loss: 0.09964275360107422 2023-01-21 10:11:53.016968: step: 1856/526, loss: 0.032918740063905716 2023-01-21 10:11:54.145508: step: 1860/526, loss: 0.21822606027126312 2023-01-21 10:11:55.282843: step: 1864/526, loss: 0.029240990057587624 2023-01-21 10:11:56.426577: step: 1868/526, loss: 0.11280885338783264 2023-01-21 10:11:57.559532: step: 1872/526, loss: 0.08848877251148224 2023-01-21 10:11:58.689291: step: 1876/526, loss: 0.11782693862915039 2023-01-21 10:11:59.832409: step: 1880/526, loss: 0.10418033599853516 2023-01-21 10:12:00.980439: step: 1884/526, loss: 0.03960847854614258 2023-01-21 10:12:02.116908: step: 1888/526, loss: 0.029823780059814453 2023-01-21 10:12:03.256263: step: 1892/526, loss: 0.04397788271307945 2023-01-21 10:12:04.371429: step: 1896/526, loss: 0.11380930989980698 2023-01-21 10:12:05.486214: step: 1900/526, loss: 0.06160225719213486 2023-01-21 10:12:06.626843: step: 1904/526, loss: 0.2855139374732971 2023-01-21 10:12:07.748114: step: 1908/526, loss: 0.7506707906723022 2023-01-21 10:12:08.902191: step: 1912/526, loss: 0.0538908950984478 2023-01-21 10:12:10.023996: step: 1916/526, loss: 0.22454842925071716 2023-01-21 10:12:11.152802: step: 1920/526, loss: 0.003399467561393976 2023-01-21 10:12:12.279895: step: 1924/526, loss: 0.026694048196077347 2023-01-21 10:12:13.440022: step: 1928/526, loss: 0.5568329095840454 2023-01-21 10:12:14.612979: step: 1932/526, loss: 0.0950227826833725 2023-01-21 10:12:15.754638: step: 1936/526, loss: 0.08688707649707794 2023-01-21 10:12:16.893855: step: 1940/526, loss: 0.12432795017957687 2023-01-21 10:12:18.022772: step: 1944/526, loss: 0.09275256097316742 2023-01-21 10:12:19.156921: step: 1948/526, loss: 0.02868051640689373 2023-01-21 10:12:20.313478: step: 1952/526, loss: 0.0698302760720253 2023-01-21 10:12:21.464978: step: 1956/526, loss: 0.07321586459875107 2023-01-21 10:12:22.589099: step: 1960/526, loss: 0.040228843688964844 2023-01-21 10:12:23.744238: step: 1964/526, loss: 0.17586413025856018 2023-01-21 10:12:24.872186: step: 1968/526, loss: 0.07969598472118378 2023-01-21 10:12:25.985132: step: 1972/526, loss: 0.11393904685974121 2023-01-21 10:12:27.137137: step: 1976/526, loss: 0.036074426025152206 2023-01-21 10:12:28.262137: step: 1980/526, loss: 0.39023932814598083 2023-01-21 10:12:29.406701: step: 1984/526, loss: 0.02489762380719185 2023-01-21 10:12:30.555592: step: 1988/526, loss: 0.039016008377075195 2023-01-21 10:12:31.705803: step: 1992/526, loss: 0.21317899227142334 2023-01-21 10:12:32.842003: step: 1996/526, loss: 0.037840940058231354 2023-01-21 10:12:33.989245: step: 2000/526, loss: 0.17182107269763947 2023-01-21 10:12:35.133391: step: 2004/526, loss: 0.032541655004024506 2023-01-21 10:12:36.279761: step: 2008/526, loss: 0.060860924422740936 2023-01-21 10:12:37.413045: step: 2012/526, loss: 0.04920406639575958 2023-01-21 10:12:38.536829: step: 2016/526, loss: 0.08759719878435135 2023-01-21 10:12:39.666653: step: 2020/526, loss: 0.07130040973424911 2023-01-21 10:12:40.800885: step: 2024/526, loss: 0.06774139404296875 2023-01-21 10:12:41.937273: step: 2028/526, loss: 0.15149059891700745 2023-01-21 10:12:43.107826: step: 2032/526, loss: 0.03901328891515732 2023-01-21 10:12:44.316098: step: 2036/526, loss: 0.04768018424510956 2023-01-21 10:12:45.489512: step: 2040/526, loss: 0.13834437727928162 2023-01-21 10:12:46.629754: step: 2044/526, loss: 0.13579322397708893 2023-01-21 10:12:47.813833: step: 2048/526, loss: 0.11051268875598907 2023-01-21 10:12:48.956499: step: 2052/526, loss: 0.05337200313806534 2023-01-21 10:12:50.116027: step: 2056/526, loss: 0.02866397053003311 2023-01-21 10:12:51.292252: step: 2060/526, loss: 0.12159357219934464 2023-01-21 10:12:52.437914: step: 2064/526, loss: 0.046639252454042435 2023-01-21 10:12:53.572190: step: 2068/526, loss: 0.3356172740459442 2023-01-21 10:12:54.740185: step: 2072/526, loss: 0.1432534158229828 2023-01-21 10:12:55.884868: step: 2076/526, loss: 0.2417449951171875 2023-01-21 10:12:57.053940: step: 2080/526, loss: 0.07272524386644363 2023-01-21 10:12:58.201078: step: 2084/526, loss: 0.05484972149133682 2023-01-21 10:12:59.327278: step: 2088/526, loss: 0.12363357841968536 2023-01-21 10:13:00.461791: step: 2092/526, loss: 0.025763511657714844 2023-01-21 10:13:01.593582: step: 2096/526, loss: 0.0769733414053917 2023-01-21 10:13:02.728787: step: 2100/526, loss: 0.11375121772289276 2023-01-21 10:13:03.890472: step: 2104/526, loss: 0.049809932708740234 ================================================== Loss: 0.130 -------------------- Dev: {'event': {'p': 0.60687432867884, 'r': 0.7523302263648469, 'f1': 0.6718192627824019}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Test: {'event': {'p': 0.6179966044142614, 'r': 0.7599164926931107, 'f1': 0.6816479400749064}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Chinese: {'event': {'p': 0.5844155844155844, 'r': 0.8333333333333334, 'f1': 0.6870229007633588}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Korean: {'event': {'p': 0.7714285714285715, 'r': 0.42857142857142855, 'f1': 0.5510204081632654}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Russian: {'event': {'p': 0.5263157894736842, 'r': 0.5555555555555556, 'f1': 0.5405405405405405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.60687432867884, 'r': 0.7523302263648469, 'f1': 0.6718192627824019}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Russian: {'event': {'p': 0.6179966044142614, 'r': 0.7599164926931107, 'f1': 0.6816479400749064}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Russian: {'event': {'p': 0.5263157894736842, 'r': 0.5555555555555556, 'f1': 0.5405405405405405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} ****************************** Epoch: 9 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:13:53.115114: step: 4/526, loss: 0.06273460388183594 2023-01-21 10:13:54.229467: step: 8/526, loss: 0.04972662776708603 2023-01-21 10:13:55.363225: step: 12/526, loss: 0.02827734872698784 2023-01-21 10:13:56.502921: step: 16/526, loss: 0.052943114191293716 2023-01-21 10:13:57.647989: step: 20/526, loss: 0.008646011352539062 2023-01-21 10:13:58.774980: step: 24/526, loss: 0.06134873628616333 2023-01-21 10:13:59.879503: step: 28/526, loss: 0.2242942452430725 2023-01-21 10:14:00.984352: step: 32/526, loss: 0.29536008834838867 2023-01-21 10:14:02.127842: step: 36/526, loss: 0.055123090744018555 2023-01-21 10:14:03.251964: step: 40/526, loss: 0.05992274358868599 2023-01-21 10:14:04.422277: step: 44/526, loss: 0.024144649505615234 2023-01-21 10:14:05.530069: step: 48/526, loss: 0.08518228679895401 2023-01-21 10:14:06.661256: step: 52/526, loss: 0.018128395080566406 2023-01-21 10:14:07.835107: step: 56/526, loss: 0.12564639747142792 2023-01-21 10:14:09.055042: step: 60/526, loss: 0.07249632477760315 2023-01-21 10:14:10.217453: step: 64/526, loss: 0.2735300064086914 2023-01-21 10:14:11.329786: step: 68/526, loss: 0.024007225409150124 2023-01-21 10:14:12.473885: step: 72/526, loss: 0.05489225313067436 2023-01-21 10:14:13.623631: step: 76/526, loss: 0.045737460255622864 2023-01-21 10:14:14.758862: step: 80/526, loss: 0.13226623833179474 2023-01-21 10:14:15.905289: step: 84/526, loss: 0.1825447678565979 2023-01-21 10:14:17.110473: step: 88/526, loss: 0.06590290367603302 2023-01-21 10:14:18.273576: step: 92/526, loss: 0.16964063048362732 2023-01-21 10:14:19.412624: step: 96/526, loss: 0.2701391577720642 2023-01-21 10:14:20.514291: step: 100/526, loss: 0.002492809435352683 2023-01-21 10:14:21.657836: step: 104/526, loss: 0.00204639439471066 2023-01-21 10:14:22.805317: step: 108/526, loss: 0.226235494017601 2023-01-21 10:14:23.935633: step: 112/526, loss: 0.0693480521440506 2023-01-21 10:14:25.082423: step: 116/526, loss: 0.02584218978881836 2023-01-21 10:14:26.205960: step: 120/526, loss: 0.07365093380212784 2023-01-21 10:14:27.329734: step: 124/526, loss: 0.018468167632818222 2023-01-21 10:14:28.489466: step: 128/526, loss: 3.8758349418640137 2023-01-21 10:14:29.628827: step: 132/526, loss: 0.052092649042606354 2023-01-21 10:14:30.763535: step: 136/526, loss: 0.08069362491369247 2023-01-21 10:14:31.889541: step: 140/526, loss: 0.03643073886632919 2023-01-21 10:14:33.028898: step: 144/526, loss: 0.008904552087187767 2023-01-21 10:14:34.161652: step: 148/526, loss: 0.061319828033447266 2023-01-21 10:14:35.280876: step: 152/526, loss: 0.10991770774126053 2023-01-21 10:14:36.409292: step: 156/526, loss: 0.12624579668045044 2023-01-21 10:14:37.548746: step: 160/526, loss: 0.004436207003891468 2023-01-21 10:14:38.685427: step: 164/526, loss: 0.12727917730808258 2023-01-21 10:14:39.806654: step: 168/526, loss: 0.05051498860120773 2023-01-21 10:14:40.950571: step: 172/526, loss: 0.018753433600068092 2023-01-21 10:14:42.092710: step: 176/526, loss: 0.09445858001708984 2023-01-21 10:14:43.245014: step: 180/526, loss: 0.03955969959497452 2023-01-21 10:14:44.379758: step: 184/526, loss: 0.01619882509112358 2023-01-21 10:14:45.472190: step: 188/526, loss: 0.004410171881318092 2023-01-21 10:14:46.606831: step: 192/526, loss: 0.10134325921535492 2023-01-21 10:14:47.738258: step: 196/526, loss: 0.03623542934656143 2023-01-21 10:14:48.867936: step: 200/526, loss: 0.025229329243302345 2023-01-21 10:14:50.009764: step: 204/526, loss: 0.046385716646909714 2023-01-21 10:14:51.158361: step: 208/526, loss: 0.14255085587501526 2023-01-21 10:14:52.282142: step: 212/526, loss: 0.009249973110854626 2023-01-21 10:14:53.448756: step: 216/526, loss: 0.029299069195985794 2023-01-21 10:14:54.606056: step: 220/526, loss: 0.09028711169958115 2023-01-21 10:14:55.739935: step: 224/526, loss: 0.05342698097229004 2023-01-21 10:14:56.902035: step: 228/526, loss: 0.03964881971478462 2023-01-21 10:14:58.066003: step: 232/526, loss: 0.05981273576617241 2023-01-21 10:14:59.240305: step: 236/526, loss: 0.12419013679027557 2023-01-21 10:15:00.376166: step: 240/526, loss: 0.026497889310121536 2023-01-21 10:15:01.517423: step: 244/526, loss: 0.02921009063720703 2023-01-21 10:15:02.667344: step: 248/526, loss: 0.07892656326293945 2023-01-21 10:15:03.821135: step: 252/526, loss: 0.4995053708553314 2023-01-21 10:15:04.941943: step: 256/526, loss: 0.03964080661535263 2023-01-21 10:15:06.091074: step: 260/526, loss: 0.09100300818681717 2023-01-21 10:15:07.248245: step: 264/526, loss: 0.23894843459129333 2023-01-21 10:15:08.382513: step: 268/526, loss: 0.0653919205069542 2023-01-21 10:15:09.569711: step: 272/526, loss: 0.11007685959339142 2023-01-21 10:15:10.715147: step: 276/526, loss: 0.04790840297937393 2023-01-21 10:15:11.842019: step: 280/526, loss: 0.010727310553193092 2023-01-21 10:15:12.988812: step: 284/526, loss: 0.08917751908302307 2023-01-21 10:15:14.116282: step: 288/526, loss: 0.04526824876666069 2023-01-21 10:15:15.274553: step: 292/526, loss: 0.023631764575839043 2023-01-21 10:15:16.437596: step: 296/526, loss: 0.10619097948074341 2023-01-21 10:15:17.560878: step: 300/526, loss: 0.07016344368457794 2023-01-21 10:15:18.706432: step: 304/526, loss: 0.06495552510023117 2023-01-21 10:15:19.873479: step: 308/526, loss: 0.20252735912799835 2023-01-21 10:15:20.980005: step: 312/526, loss: 0.06283549964427948 2023-01-21 10:15:22.109103: step: 316/526, loss: 0.07670155167579651 2023-01-21 10:15:23.265065: step: 320/526, loss: 0.02313871495425701 2023-01-21 10:15:24.375739: step: 324/526, loss: 0.01654043234884739 2023-01-21 10:15:25.508686: step: 328/526, loss: 0.03600268438458443 2023-01-21 10:15:26.647359: step: 332/526, loss: 0.051339223980903625 2023-01-21 10:15:27.799613: step: 336/526, loss: 0.012384273111820221 2023-01-21 10:15:28.941069: step: 340/526, loss: 0.06428880989551544 2023-01-21 10:15:30.097645: step: 344/526, loss: 0.2045794576406479 2023-01-21 10:15:31.217675: step: 348/526, loss: 0.01100549753755331 2023-01-21 10:15:32.388386: step: 352/526, loss: 0.09947919845581055 2023-01-21 10:15:33.538945: step: 356/526, loss: 0.5032371282577515 2023-01-21 10:15:34.662157: step: 360/526, loss: 0.03166599199175835 2023-01-21 10:15:35.804322: step: 364/526, loss: 0.02063298225402832 2023-01-21 10:15:36.918512: step: 368/526, loss: 0.04896404966711998 2023-01-21 10:15:38.063180: step: 372/526, loss: 0.06352724879980087 2023-01-21 10:15:39.233514: step: 376/526, loss: 0.09708347916603088 2023-01-21 10:15:40.377424: step: 380/526, loss: 0.09912853688001633 2023-01-21 10:15:41.488045: step: 384/526, loss: 0.02930302731692791 2023-01-21 10:15:42.653979: step: 388/526, loss: 0.23564453423023224 2023-01-21 10:15:43.791140: step: 392/526, loss: 0.10602133721113205 2023-01-21 10:15:44.959850: step: 396/526, loss: 0.1418243646621704 2023-01-21 10:15:46.091268: step: 400/526, loss: 0.06861257553100586 2023-01-21 10:15:47.230650: step: 404/526, loss: 0.3773897886276245 2023-01-21 10:15:48.362265: step: 408/526, loss: 0.008510899730026722 2023-01-21 10:15:49.489695: step: 412/526, loss: 0.012688159942626953 2023-01-21 10:15:50.623011: step: 416/526, loss: 0.0464058443903923 2023-01-21 10:15:51.779445: step: 420/526, loss: 0.07371998578310013 2023-01-21 10:15:52.940934: step: 424/526, loss: 0.7404410243034363 2023-01-21 10:15:54.105212: step: 428/526, loss: 0.01785125769674778 2023-01-21 10:15:55.236459: step: 432/526, loss: 0.12167225033044815 2023-01-21 10:15:56.387620: step: 436/526, loss: 0.04297056049108505 2023-01-21 10:15:57.533712: step: 440/526, loss: 0.10440711677074432 2023-01-21 10:15:58.693841: step: 444/526, loss: 0.06877918541431427 2023-01-21 10:15:59.867569: step: 448/526, loss: 0.029997684061527252 2023-01-21 10:16:01.038126: step: 452/526, loss: 0.15336699783802032 2023-01-21 10:16:02.204875: step: 456/526, loss: 0.03872213512659073 2023-01-21 10:16:03.348833: step: 460/526, loss: 0.10762529075145721 2023-01-21 10:16:04.489513: step: 464/526, loss: 0.05059080198407173 2023-01-21 10:16:05.628920: step: 468/526, loss: 0.04152107238769531 2023-01-21 10:16:06.779784: step: 472/526, loss: 0.03788471221923828 2023-01-21 10:16:07.932072: step: 476/526, loss: 0.0924578607082367 2023-01-21 10:16:09.070023: step: 480/526, loss: 0.04869556799530983 2023-01-21 10:16:10.230949: step: 484/526, loss: 0.03446640446782112 2023-01-21 10:16:11.359647: step: 488/526, loss: 0.030467605218291283 2023-01-21 10:16:12.500304: step: 492/526, loss: 0.04888105392456055 2023-01-21 10:16:13.653204: step: 496/526, loss: 0.10127200931310654 2023-01-21 10:16:14.790112: step: 500/526, loss: 0.49693623185157776 2023-01-21 10:16:15.900708: step: 504/526, loss: 0.06561170518398285 2023-01-21 10:16:17.030163: step: 508/526, loss: 0.03338127210736275 2023-01-21 10:16:18.196570: step: 512/526, loss: 0.053220320492982864 2023-01-21 10:16:19.340224: step: 516/526, loss: 0.26733750104904175 2023-01-21 10:16:20.477518: step: 520/526, loss: 0.022270098328590393 2023-01-21 10:16:21.594211: step: 524/526, loss: 0.017182255163788795 2023-01-21 10:16:22.724290: step: 528/526, loss: 0.08973188698291779 2023-01-21 10:16:23.898432: step: 532/526, loss: 0.16665706038475037 2023-01-21 10:16:25.080908: step: 536/526, loss: 0.0938805565237999 2023-01-21 10:16:26.245874: step: 540/526, loss: 0.01415166910737753 2023-01-21 10:16:27.407360: step: 544/526, loss: 0.22182941436767578 2023-01-21 10:16:28.540424: step: 548/526, loss: 0.018256567418575287 2023-01-21 10:16:29.660789: step: 552/526, loss: 0.013985920697450638 2023-01-21 10:16:30.796286: step: 556/526, loss: 0.05294542759656906 2023-01-21 10:16:31.905284: step: 560/526, loss: 0.026349354535341263 2023-01-21 10:16:33.049596: step: 564/526, loss: 0.027312468737363815 2023-01-21 10:16:34.195482: step: 568/526, loss: 0.05296573415398598 2023-01-21 10:16:35.359188: step: 572/526, loss: 0.029798317700624466 2023-01-21 10:16:36.486487: step: 576/526, loss: 0.03664436563849449 2023-01-21 10:16:37.634059: step: 580/526, loss: 0.11082687973976135 2023-01-21 10:16:38.776284: step: 584/526, loss: 0.0678778663277626 2023-01-21 10:16:39.909239: step: 588/526, loss: 0.056672096252441406 2023-01-21 10:16:41.026524: step: 592/526, loss: 0.14632777869701385 2023-01-21 10:16:42.176130: step: 596/526, loss: 0.14618340134620667 2023-01-21 10:16:43.323237: step: 600/526, loss: 0.1745828539133072 2023-01-21 10:16:44.454548: step: 604/526, loss: 0.0998816043138504 2023-01-21 10:16:45.596037: step: 608/526, loss: 0.08128509670495987 2023-01-21 10:16:46.747969: step: 612/526, loss: 0.034986067563295364 2023-01-21 10:16:47.877613: step: 616/526, loss: 0.06140480190515518 2023-01-21 10:16:49.012162: step: 620/526, loss: 0.10063949972391129 2023-01-21 10:16:50.141225: step: 624/526, loss: 0.08500318229198456 2023-01-21 10:16:51.286375: step: 628/526, loss: 0.04294443503022194 2023-01-21 10:16:52.426695: step: 632/526, loss: 0.115880087018013 2023-01-21 10:16:53.594487: step: 636/526, loss: 0.007815957069396973 2023-01-21 10:16:54.718811: step: 640/526, loss: 0.007673311047255993 2023-01-21 10:16:55.879806: step: 644/526, loss: 0.07724981009960175 2023-01-21 10:16:57.005800: step: 648/526, loss: 0.09962892532348633 2023-01-21 10:16:58.147094: step: 652/526, loss: 0.06111174076795578 2023-01-21 10:16:59.283274: step: 656/526, loss: 0.0499509796500206 2023-01-21 10:17:00.394337: step: 660/526, loss: 0.019877910614013672 2023-01-21 10:17:01.553471: step: 664/526, loss: 0.1742444932460785 2023-01-21 10:17:02.701077: step: 668/526, loss: 0.07345151901245117 2023-01-21 10:17:03.839443: step: 672/526, loss: 0.05171380192041397 2023-01-21 10:17:05.001303: step: 676/526, loss: 0.06844054162502289 2023-01-21 10:17:06.127107: step: 680/526, loss: 0.19001130759716034 2023-01-21 10:17:07.264138: step: 684/526, loss: 0.0820087417960167 2023-01-21 10:17:08.392726: step: 688/526, loss: 0.04524059221148491 2023-01-21 10:17:09.532366: step: 692/526, loss: 0.02237701416015625 2023-01-21 10:17:10.672037: step: 696/526, loss: 0.013925934210419655 2023-01-21 10:17:11.793717: step: 700/526, loss: 0.13059782981872559 2023-01-21 10:17:12.940898: step: 704/526, loss: 0.005258369259536266 2023-01-21 10:17:14.090498: step: 708/526, loss: 0.1428043395280838 2023-01-21 10:17:15.216487: step: 712/526, loss: 0.13605232536792755 2023-01-21 10:17:16.362198: step: 716/526, loss: 0.05469207838177681 2023-01-21 10:17:17.523602: step: 720/526, loss: 0.029755020514130592 2023-01-21 10:17:18.644392: step: 724/526, loss: 0.019161224365234375 2023-01-21 10:17:19.802764: step: 728/526, loss: 0.15146532654762268 2023-01-21 10:17:20.946459: step: 732/526, loss: 0.12099442631006241 2023-01-21 10:17:22.108882: step: 736/526, loss: 0.11903706192970276 2023-01-21 10:17:23.254828: step: 740/526, loss: 0.035573769360780716 2023-01-21 10:17:24.378886: step: 744/526, loss: 0.023453522473573685 2023-01-21 10:17:25.513909: step: 748/526, loss: 0.06212405860424042 2023-01-21 10:17:26.683758: step: 752/526, loss: 0.041171833872795105 2023-01-21 10:17:27.822982: step: 756/526, loss: 0.0913642942905426 2023-01-21 10:17:28.966723: step: 760/526, loss: 0.060819149017333984 2023-01-21 10:17:30.100156: step: 764/526, loss: 0.005241966340690851 2023-01-21 10:17:31.220278: step: 768/526, loss: 0.04731016606092453 2023-01-21 10:17:32.350258: step: 772/526, loss: 0.0874660536646843 2023-01-21 10:17:33.490762: step: 776/526, loss: 0.0473664253950119 2023-01-21 10:17:34.608594: step: 780/526, loss: 0.04849643632769585 2023-01-21 10:17:35.735775: step: 784/526, loss: 0.021467208862304688 2023-01-21 10:17:36.901166: step: 788/526, loss: 0.02276330068707466 2023-01-21 10:17:38.033987: step: 792/526, loss: 0.7945021390914917 2023-01-21 10:17:39.163189: step: 796/526, loss: 0.07966355979442596 2023-01-21 10:17:40.285608: step: 800/526, loss: 0.10666169971227646 2023-01-21 10:17:41.437700: step: 804/526, loss: 0.12571974098682404 2023-01-21 10:17:42.567827: step: 808/526, loss: 0.030166054144501686 2023-01-21 10:17:43.730579: step: 812/526, loss: 0.0077298167161643505 2023-01-21 10:17:44.856929: step: 816/526, loss: 0.09210185706615448 2023-01-21 10:17:45.996559: step: 820/526, loss: 0.04695863649249077 2023-01-21 10:17:47.143124: step: 824/526, loss: 0.11744165420532227 2023-01-21 10:17:48.265740: step: 828/526, loss: 0.038575079292058945 2023-01-21 10:17:49.428725: step: 832/526, loss: 0.027442552149295807 2023-01-21 10:17:50.582275: step: 836/526, loss: 0.04608777165412903 2023-01-21 10:17:51.719757: step: 840/526, loss: 0.06978344917297363 2023-01-21 10:17:52.867353: step: 844/526, loss: 0.01463708933442831 2023-01-21 10:17:54.006446: step: 848/526, loss: 0.05460663139820099 2023-01-21 10:17:55.149203: step: 852/526, loss: 0.03759412840008736 2023-01-21 10:17:56.319459: step: 856/526, loss: 0.049581434577703476 2023-01-21 10:17:57.463446: step: 860/526, loss: 0.028264619410037994 2023-01-21 10:17:58.599004: step: 864/526, loss: 0.06847391277551651 2023-01-21 10:17:59.737537: step: 868/526, loss: 0.05833626165986061 2023-01-21 10:18:00.886849: step: 872/526, loss: 0.04056265950202942 2023-01-21 10:18:01.982483: step: 876/526, loss: 0.06393267959356308 2023-01-21 10:18:03.142654: step: 880/526, loss: 0.07144995033740997 2023-01-21 10:18:04.270129: step: 884/526, loss: 0.1611798256635666 2023-01-21 10:18:05.395633: step: 888/526, loss: 0.061147596687078476 2023-01-21 10:18:06.550773: step: 892/526, loss: 0.04013872146606445 2023-01-21 10:18:07.709383: step: 896/526, loss: 0.07072458416223526 2023-01-21 10:18:08.841882: step: 900/526, loss: 0.4843043386936188 2023-01-21 10:18:09.968455: step: 904/526, loss: 0.13908196985721588 2023-01-21 10:18:11.101025: step: 908/526, loss: 0.04387928545475006 2023-01-21 10:18:12.236167: step: 912/526, loss: 0.07040281593799591 2023-01-21 10:18:13.379981: step: 916/526, loss: 0.04282999038696289 2023-01-21 10:18:14.517050: step: 920/526, loss: 0.07336115837097168 2023-01-21 10:18:15.651634: step: 924/526, loss: 0.10548292100429535 2023-01-21 10:18:16.803151: step: 928/526, loss: 0.15576358139514923 2023-01-21 10:18:17.971496: step: 932/526, loss: 0.10637989640235901 2023-01-21 10:18:19.101637: step: 936/526, loss: 0.09210458397865295 2023-01-21 10:18:20.254565: step: 940/526, loss: 0.046960845589637756 2023-01-21 10:18:21.377958: step: 944/526, loss: 0.030419539660215378 2023-01-21 10:18:22.490081: step: 948/526, loss: 0.06506214290857315 2023-01-21 10:18:23.609950: step: 952/526, loss: 0.09902973473072052 2023-01-21 10:18:24.733654: step: 956/526, loss: 0.06627140194177628 2023-01-21 10:18:25.872250: step: 960/526, loss: 0.10831494629383087 2023-01-21 10:18:26.994276: step: 964/526, loss: 0.13443775475025177 2023-01-21 10:18:28.151220: step: 968/526, loss: 0.012888241559267044 2023-01-21 10:18:29.285188: step: 972/526, loss: 0.18446750938892365 2023-01-21 10:18:30.423265: step: 976/526, loss: 0.339281290769577 2023-01-21 10:18:31.597568: step: 980/526, loss: 0.05691647529602051 2023-01-21 10:18:32.739518: step: 984/526, loss: 0.07747035473585129 2023-01-21 10:18:33.858153: step: 988/526, loss: 0.08023052662611008 2023-01-21 10:18:35.027860: step: 992/526, loss: 0.7032496333122253 2023-01-21 10:18:36.183280: step: 996/526, loss: 0.06381640583276749 2023-01-21 10:18:37.310510: step: 1000/526, loss: 0.056778714060783386 2023-01-21 10:18:38.415129: step: 1004/526, loss: 0.33411872386932373 2023-01-21 10:18:39.549515: step: 1008/526, loss: 0.1968100517988205 2023-01-21 10:18:40.683418: step: 1012/526, loss: 0.06786146014928818 2023-01-21 10:18:41.820061: step: 1016/526, loss: 0.034000590443611145 2023-01-21 10:18:42.936592: step: 1020/526, loss: 0.004477500915527344 2023-01-21 10:18:44.073953: step: 1024/526, loss: 0.19066371023654938 2023-01-21 10:18:45.193826: step: 1028/526, loss: 0.10681009292602539 2023-01-21 10:18:46.328623: step: 1032/526, loss: 0.03215327113866806 2023-01-21 10:18:47.466779: step: 1036/526, loss: 0.1189613863825798 2023-01-21 10:18:48.633939: step: 1040/526, loss: 0.016580868512392044 2023-01-21 10:18:49.749709: step: 1044/526, loss: 0.040842246264219284 2023-01-21 10:18:50.911167: step: 1048/526, loss: 0.07049208134412766 2023-01-21 10:18:52.069829: step: 1052/526, loss: 0.2909330427646637 2023-01-21 10:18:53.188284: step: 1056/526, loss: 0.07620034366846085 2023-01-21 10:18:54.334036: step: 1060/526, loss: 0.0831860601902008 2023-01-21 10:18:55.462416: step: 1064/526, loss: 0.08039799332618713 2023-01-21 10:18:56.587345: step: 1068/526, loss: 0.0762978121638298 2023-01-21 10:18:57.705507: step: 1072/526, loss: 0.005699443630874157 2023-01-21 10:18:58.836973: step: 1076/526, loss: 0.06466083228588104 2023-01-21 10:18:59.962928: step: 1080/526, loss: 0.20280568301677704 2023-01-21 10:19:01.078132: step: 1084/526, loss: 0.08602876216173172 2023-01-21 10:19:02.187901: step: 1088/526, loss: 0.03445424884557724 2023-01-21 10:19:03.321387: step: 1092/526, loss: 0.03289613872766495 2023-01-21 10:19:04.441145: step: 1096/526, loss: 0.03306560590863228 2023-01-21 10:19:05.590174: step: 1100/526, loss: 0.12068119645118713 2023-01-21 10:19:06.709440: step: 1104/526, loss: 0.05690737068653107 2023-01-21 10:19:07.850570: step: 1108/526, loss: 0.0273881908506155 2023-01-21 10:19:08.984815: step: 1112/526, loss: 0.23444966971874237 2023-01-21 10:19:10.140850: step: 1116/526, loss: 0.04446830600500107 2023-01-21 10:19:11.318894: step: 1120/526, loss: 0.025336647406220436 2023-01-21 10:19:12.438834: step: 1124/526, loss: 0.05155191570520401 2023-01-21 10:19:13.620357: step: 1128/526, loss: 0.10651607811450958 2023-01-21 10:19:14.755359: step: 1132/526, loss: 0.114251047372818 2023-01-21 10:19:15.880369: step: 1136/526, loss: 0.5184873342514038 2023-01-21 10:19:16.996150: step: 1140/526, loss: 0.09276838600635529 2023-01-21 10:19:18.158833: step: 1144/526, loss: 0.024493694305419922 2023-01-21 10:19:19.271151: step: 1148/526, loss: 0.842566192150116 2023-01-21 10:19:20.396508: step: 1152/526, loss: 0.008257674984633923 2023-01-21 10:19:21.541324: step: 1156/526, loss: 0.08722105622291565 2023-01-21 10:19:22.656467: step: 1160/526, loss: 0.09716796875 2023-01-21 10:19:23.766398: step: 1164/526, loss: 0.007197761442512274 2023-01-21 10:19:24.881509: step: 1168/526, loss: 0.08151836693286896 2023-01-21 10:19:26.041079: step: 1172/526, loss: 0.0497395284473896 2023-01-21 10:19:27.166279: step: 1176/526, loss: 0.06084489822387695 2023-01-21 10:19:28.321158: step: 1180/526, loss: 0.07828722149133682 2023-01-21 10:19:29.446112: step: 1184/526, loss: 0.042025040835142136 2023-01-21 10:19:30.602152: step: 1188/526, loss: 0.09930925071239471 2023-01-21 10:19:31.800647: step: 1192/526, loss: 0.07812199741601944 2023-01-21 10:19:32.919101: step: 1196/526, loss: 0.041902828961610794 2023-01-21 10:19:34.049905: step: 1200/526, loss: 0.22845545411109924 2023-01-21 10:19:35.179974: step: 1204/526, loss: 0.05532092973589897 2023-01-21 10:19:36.340045: step: 1208/526, loss: 0.06245307996869087 2023-01-21 10:19:37.471518: step: 1212/526, loss: 0.02949414774775505 2023-01-21 10:19:38.624657: step: 1216/526, loss: 0.02378559112548828 2023-01-21 10:19:39.755254: step: 1220/526, loss: 0.006373787298798561 2023-01-21 10:19:40.897174: step: 1224/526, loss: 0.08481474220752716 2023-01-21 10:19:42.022004: step: 1228/526, loss: 0.03761797025799751 2023-01-21 10:19:43.152125: step: 1232/526, loss: 0.0032770633697509766 2023-01-21 10:19:44.289882: step: 1236/526, loss: 0.05032806470990181 2023-01-21 10:19:45.417402: step: 1240/526, loss: 0.07737484574317932 2023-01-21 10:19:46.540647: step: 1244/526, loss: 0.026522256433963776 2023-01-21 10:19:47.672804: step: 1248/526, loss: 0.027126504108309746 2023-01-21 10:19:48.812833: step: 1252/526, loss: 0.07468719780445099 2023-01-21 10:19:49.963684: step: 1256/526, loss: 0.28271007537841797 2023-01-21 10:19:51.081496: step: 1260/526, loss: 0.28559890389442444 2023-01-21 10:19:52.267451: step: 1264/526, loss: 0.022942161187529564 2023-01-21 10:19:53.380452: step: 1268/526, loss: 0.018129730597138405 2023-01-21 10:19:54.487506: step: 1272/526, loss: 0.3420391082763672 2023-01-21 10:19:55.628567: step: 1276/526, loss: 0.0206514373421669 2023-01-21 10:19:56.747279: step: 1280/526, loss: 0.6973414421081543 2023-01-21 10:19:57.850878: step: 1284/526, loss: 0.09004630893468857 2023-01-21 10:19:58.960529: step: 1288/526, loss: 0.012910605408251286 2023-01-21 10:20:00.095023: step: 1292/526, loss: 0.052889443933963776 2023-01-21 10:20:01.251198: step: 1296/526, loss: 0.00937356986105442 2023-01-21 10:20:02.374674: step: 1300/526, loss: 0.009701442904770374 2023-01-21 10:20:03.507030: step: 1304/526, loss: 0.08429671078920364 2023-01-21 10:20:04.626802: step: 1308/526, loss: 0.05223708599805832 2023-01-21 10:20:05.734162: step: 1312/526, loss: 0.1049903929233551 2023-01-21 10:20:06.862842: step: 1316/526, loss: 0.15975895524024963 2023-01-21 10:20:07.990632: step: 1320/526, loss: 0.06687908619642258 2023-01-21 10:20:09.121762: step: 1324/526, loss: 0.0996524840593338 2023-01-21 10:20:10.226918: step: 1328/526, loss: 0.007422924041748047 2023-01-21 10:20:11.357134: step: 1332/526, loss: 0.1732240617275238 2023-01-21 10:20:12.495060: step: 1336/526, loss: 0.03923340141773224 2023-01-21 10:20:13.655837: step: 1340/526, loss: 0.007705879397690296 2023-01-21 10:20:14.819640: step: 1344/526, loss: 1.9873721599578857 2023-01-21 10:20:15.950262: step: 1348/526, loss: 0.06160922348499298 2023-01-21 10:20:17.075551: step: 1352/526, loss: 0.05101480334997177 2023-01-21 10:20:18.203816: step: 1356/526, loss: 0.04123516008257866 2023-01-21 10:20:19.322198: step: 1360/526, loss: 0.12192918360233307 2023-01-21 10:20:20.460423: step: 1364/526, loss: 0.14946289360523224 2023-01-21 10:20:21.592619: step: 1368/526, loss: 0.04394255951046944 2023-01-21 10:20:22.692558: step: 1372/526, loss: 0.09714050590991974 2023-01-21 10:20:23.847838: step: 1376/526, loss: 0.04991569742560387 2023-01-21 10:20:24.995893: step: 1380/526, loss: 0.07330255955457687 2023-01-21 10:20:26.129234: step: 1384/526, loss: 0.10196466743946075 2023-01-21 10:20:27.300463: step: 1388/526, loss: 0.11431503295898438 2023-01-21 10:20:28.446514: step: 1392/526, loss: 0.24093666672706604 2023-01-21 10:20:29.572189: step: 1396/526, loss: 0.016542624682188034 2023-01-21 10:20:30.706647: step: 1400/526, loss: 0.6701074838638306 2023-01-21 10:20:31.868012: step: 1404/526, loss: 0.0032260892912745476 2023-01-21 10:20:32.989442: step: 1408/526, loss: 0.03952236473560333 2023-01-21 10:20:34.126310: step: 1412/526, loss: 0.053475573658943176 2023-01-21 10:20:35.245107: step: 1416/526, loss: 0.10574407875537872 2023-01-21 10:20:36.386799: step: 1420/526, loss: 0.02206139639019966 2023-01-21 10:20:37.528215: step: 1424/526, loss: 0.23898696899414062 2023-01-21 10:20:38.635801: step: 1428/526, loss: 0.0604923740029335 2023-01-21 10:20:39.778104: step: 1432/526, loss: 0.22128941118717194 2023-01-21 10:20:40.917696: step: 1436/526, loss: 0.03140106052160263 2023-01-21 10:20:42.058376: step: 1440/526, loss: 0.07476844638586044 2023-01-21 10:20:43.161085: step: 1444/526, loss: 0.05649605020880699 2023-01-21 10:20:44.280811: step: 1448/526, loss: 0.04432068020105362 2023-01-21 10:20:45.381378: step: 1452/526, loss: 0.01875171810388565 2023-01-21 10:20:46.503364: step: 1456/526, loss: 0.10838525742292404 2023-01-21 10:20:47.630152: step: 1460/526, loss: 0.014918709173798561 2023-01-21 10:20:48.803238: step: 1464/526, loss: 0.05639658123254776 2023-01-21 10:20:49.936990: step: 1468/526, loss: 0.04339020326733589 2023-01-21 10:20:51.048303: step: 1472/526, loss: 0.024857616052031517 2023-01-21 10:20:52.207786: step: 1476/526, loss: 0.007811450865119696 2023-01-21 10:20:53.330679: step: 1480/526, loss: 0.01029291097074747 2023-01-21 10:20:54.469717: step: 1484/526, loss: 0.11069220304489136 2023-01-21 10:20:55.577165: step: 1488/526, loss: 0.009436321444809437 2023-01-21 10:20:56.713750: step: 1492/526, loss: 0.005244255065917969 2023-01-21 10:20:57.834744: step: 1496/526, loss: 0.05494390055537224 2023-01-21 10:20:58.976789: step: 1500/526, loss: 0.01738009601831436 2023-01-21 10:21:00.110038: step: 1504/526, loss: 0.041535090655088425 2023-01-21 10:21:01.238481: step: 1508/526, loss: 0.07702608406543732 2023-01-21 10:21:02.362431: step: 1512/526, loss: 0.08296613395214081 2023-01-21 10:21:03.494488: step: 1516/526, loss: 0.1479577124118805 2023-01-21 10:21:04.615406: step: 1520/526, loss: 0.06086378172039986 2023-01-21 10:21:05.785389: step: 1524/526, loss: 0.1092875525355339 2023-01-21 10:21:06.935120: step: 1528/526, loss: 0.0897495299577713 2023-01-21 10:21:08.090051: step: 1532/526, loss: 0.07425608485937119 2023-01-21 10:21:09.232862: step: 1536/526, loss: 0.027304459363222122 2023-01-21 10:21:10.372399: step: 1540/526, loss: 0.09604702144861221 2023-01-21 10:21:11.510663: step: 1544/526, loss: 0.10754013061523438 2023-01-21 10:21:12.633982: step: 1548/526, loss: 0.10374736785888672 2023-01-21 10:21:13.759734: step: 1552/526, loss: 0.037551261484622955 2023-01-21 10:21:14.885826: step: 1556/526, loss: 0.31077396869659424 2023-01-21 10:21:16.032525: step: 1560/526, loss: 0.08327203243970871 2023-01-21 10:21:17.182797: step: 1564/526, loss: 0.020313359797000885 2023-01-21 10:21:18.326157: step: 1568/526, loss: 0.08904695510864258 2023-01-21 10:21:19.474490: step: 1572/526, loss: 0.034650661051273346 2023-01-21 10:21:20.574658: step: 1576/526, loss: 0.04883251339197159 2023-01-21 10:21:21.704912: step: 1580/526, loss: 0.08138800412416458 2023-01-21 10:21:22.826662: step: 1584/526, loss: 0.03640337288379669 2023-01-21 10:21:23.947857: step: 1588/526, loss: 0.09982023388147354 2023-01-21 10:21:25.104719: step: 1592/526, loss: 0.10655536502599716 2023-01-21 10:21:26.239993: step: 1596/526, loss: 0.09950906783342361 2023-01-21 10:21:27.351570: step: 1600/526, loss: 0.0488312765955925 2023-01-21 10:21:28.469039: step: 1604/526, loss: 0.13597726821899414 2023-01-21 10:21:29.598721: step: 1608/526, loss: 0.03615992143750191 2023-01-21 10:21:30.733449: step: 1612/526, loss: 0.24791431427001953 2023-01-21 10:21:31.874236: step: 1616/526, loss: 0.1749028116464615 2023-01-21 10:21:33.025446: step: 1620/526, loss: 0.08051913231611252 2023-01-21 10:21:34.156306: step: 1624/526, loss: 0.151081845164299 2023-01-21 10:21:35.308910: step: 1628/526, loss: 0.09550247341394424 2023-01-21 10:21:36.457775: step: 1632/526, loss: 0.005854606628417969 2023-01-21 10:21:37.619616: step: 1636/526, loss: 0.08622942119836807 2023-01-21 10:21:38.738463: step: 1640/526, loss: 0.026195144280791283 2023-01-21 10:21:39.866673: step: 1644/526, loss: 0.006679725367575884 2023-01-21 10:21:41.049026: step: 1648/526, loss: 0.14648056030273438 2023-01-21 10:21:42.205091: step: 1652/526, loss: 0.12668178975582123 2023-01-21 10:21:43.374173: step: 1656/526, loss: 0.06289100646972656 2023-01-21 10:21:44.504186: step: 1660/526, loss: 0.1046912893652916 2023-01-21 10:21:45.651553: step: 1664/526, loss: 0.016279984265565872 2023-01-21 10:21:46.798646: step: 1668/526, loss: 0.09579434245824814 2023-01-21 10:21:47.924846: step: 1672/526, loss: 0.04399166256189346 2023-01-21 10:21:49.075228: step: 1676/526, loss: 0.024792686104774475 2023-01-21 10:21:50.240431: step: 1680/526, loss: 0.08130040764808655 2023-01-21 10:21:51.380822: step: 1684/526, loss: 0.025147533044219017 2023-01-21 10:21:52.538630: step: 1688/526, loss: 0.0655585303902626 2023-01-21 10:21:53.679125: step: 1692/526, loss: 0.0028779031708836555 2023-01-21 10:21:54.834709: step: 1696/526, loss: 0.03996109962463379 2023-01-21 10:21:55.978076: step: 1700/526, loss: 0.12408313155174255 2023-01-21 10:21:57.144530: step: 1704/526, loss: 0.011883596889674664 2023-01-21 10:21:58.278571: step: 1708/526, loss: 0.10415688157081604 2023-01-21 10:21:59.436463: step: 1712/526, loss: 0.01332016009837389 2023-01-21 10:22:00.571998: step: 1716/526, loss: 0.07270117104053497 2023-01-21 10:22:01.753597: step: 1720/526, loss: 0.08891163021326065 2023-01-21 10:22:02.889382: step: 1724/526, loss: 0.01740737073123455 2023-01-21 10:22:04.036740: step: 1728/526, loss: 0.023345567286014557 2023-01-21 10:22:05.207131: step: 1732/526, loss: 0.012402153573930264 2023-01-21 10:22:06.402789: step: 1736/526, loss: 0.3003942370414734 2023-01-21 10:22:07.526798: step: 1740/526, loss: 0.0820653960108757 2023-01-21 10:22:08.654289: step: 1744/526, loss: 0.03697824478149414 2023-01-21 10:22:09.780246: step: 1748/526, loss: 0.1366405040025711 2023-01-21 10:22:10.898754: step: 1752/526, loss: 0.13540410995483398 2023-01-21 10:22:12.066287: step: 1756/526, loss: 0.06986103951931 2023-01-21 10:22:13.221004: step: 1760/526, loss: 0.04328117519617081 2023-01-21 10:22:14.332105: step: 1764/526, loss: 0.005517005920410156 2023-01-21 10:22:15.470096: step: 1768/526, loss: 0.9555491805076599 2023-01-21 10:22:16.618609: step: 1772/526, loss: 0.38103771209716797 2023-01-21 10:22:17.765191: step: 1776/526, loss: 0.08931179344654083 2023-01-21 10:22:18.927323: step: 1780/526, loss: 0.14278793334960938 2023-01-21 10:22:20.058656: step: 1784/526, loss: 0.05463924631476402 2023-01-21 10:22:21.224557: step: 1788/526, loss: 0.17228588461875916 2023-01-21 10:22:22.374107: step: 1792/526, loss: 0.0715034008026123 2023-01-21 10:22:23.547139: step: 1796/526, loss: 0.08065047860145569 2023-01-21 10:22:24.688700: step: 1800/526, loss: 0.14247922599315643 2023-01-21 10:22:25.836916: step: 1804/526, loss: 0.2524127960205078 2023-01-21 10:22:27.002259: step: 1808/526, loss: 0.08272905647754669 2023-01-21 10:22:28.136834: step: 1812/526, loss: 0.014359379187226295 2023-01-21 10:22:29.270695: step: 1816/526, loss: 0.04249287024140358 2023-01-21 10:22:30.409159: step: 1820/526, loss: 0.6068402528762817 2023-01-21 10:22:31.569085: step: 1824/526, loss: 0.04011697694659233 2023-01-21 10:22:32.744394: step: 1828/526, loss: 0.04723620414733887 2023-01-21 10:22:33.927971: step: 1832/526, loss: 0.17213377356529236 2023-01-21 10:22:35.051568: step: 1836/526, loss: 0.4181341528892517 2023-01-21 10:22:36.245097: step: 1840/526, loss: 0.156813383102417 2023-01-21 10:22:37.413319: step: 1844/526, loss: 0.3373374044895172 2023-01-21 10:22:38.590488: step: 1848/526, loss: 0.1378398984670639 2023-01-21 10:22:39.756267: step: 1852/526, loss: 0.022688819095492363 2023-01-21 10:22:40.866882: step: 1856/526, loss: 0.0207995418459177 2023-01-21 10:22:42.056288: step: 1860/526, loss: 0.01760849915444851 2023-01-21 10:22:43.250096: step: 1864/526, loss: 0.06021890416741371 2023-01-21 10:22:44.406438: step: 1868/526, loss: 0.03155479580163956 2023-01-21 10:22:45.545666: step: 1872/526, loss: 0.03675680235028267 2023-01-21 10:22:46.677044: step: 1876/526, loss: 0.06314067542552948 2023-01-21 10:22:47.846669: step: 1880/526, loss: 0.03274441137909889 2023-01-21 10:22:49.001048: step: 1884/526, loss: 0.0290724765509367 2023-01-21 10:22:50.143299: step: 1888/526, loss: 0.0348079688847065 2023-01-21 10:22:51.320893: step: 1892/526, loss: 0.25051993131637573 2023-01-21 10:22:52.508075: step: 1896/526, loss: 0.08215999603271484 2023-01-21 10:22:53.658686: step: 1900/526, loss: 0.10269089043140411 2023-01-21 10:22:54.830445: step: 1904/526, loss: 0.01501312293112278 2023-01-21 10:22:56.001782: step: 1908/526, loss: 0.01247868500649929 2023-01-21 10:22:57.176535: step: 1912/526, loss: 0.05762819945812225 2023-01-21 10:22:58.326482: step: 1916/526, loss: 0.059395406395196915 2023-01-21 10:22:59.493956: step: 1920/526, loss: 0.04089965671300888 2023-01-21 10:23:00.646485: step: 1924/526, loss: 0.08226414024829865 2023-01-21 10:23:01.829619: step: 1928/526, loss: 0.027237892150878906 2023-01-21 10:23:02.970446: step: 1932/526, loss: 0.029527664184570312 2023-01-21 10:23:04.105542: step: 1936/526, loss: 0.22997260093688965 2023-01-21 10:23:05.238249: step: 1940/526, loss: 0.030162526294589043 2023-01-21 10:23:06.392867: step: 1944/526, loss: 0.3136647343635559 2023-01-21 10:23:07.537309: step: 1948/526, loss: 0.49713554978370667 2023-01-21 10:23:08.720910: step: 1952/526, loss: 0.004202508833259344 2023-01-21 10:23:09.895263: step: 1956/526, loss: 0.1447422057390213 2023-01-21 10:23:11.037374: step: 1960/526, loss: 0.017315387725830078 2023-01-21 10:23:12.178885: step: 1964/526, loss: 0.04121146351099014 2023-01-21 10:23:13.317250: step: 1968/526, loss: 0.053319741040468216 2023-01-21 10:23:14.451339: step: 1972/526, loss: 0.09213316440582275 2023-01-21 10:23:15.606992: step: 1976/526, loss: 0.09824752807617188 2023-01-21 10:23:16.761692: step: 1980/526, loss: 0.09497824311256409 2023-01-21 10:23:17.897922: step: 1984/526, loss: 0.14240819215774536 2023-01-21 10:23:19.004404: step: 1988/526, loss: 0.042424917221069336 2023-01-21 10:23:20.146003: step: 1992/526, loss: 0.048325732350349426 2023-01-21 10:23:21.287663: step: 1996/526, loss: 0.07114067673683167 2023-01-21 10:23:22.462321: step: 2000/526, loss: 0.06646886467933655 2023-01-21 10:23:23.589378: step: 2004/526, loss: 0.2667744755744934 2023-01-21 10:23:24.778114: step: 2008/526, loss: 0.011460686102509499 2023-01-21 10:23:25.927392: step: 2012/526, loss: 0.0881836861371994 2023-01-21 10:23:27.063711: step: 2016/526, loss: 0.03329572454094887 2023-01-21 10:23:28.238875: step: 2020/526, loss: 0.5911588072776794 2023-01-21 10:23:29.387055: step: 2024/526, loss: 0.05148658901453018 2023-01-21 10:23:30.529492: step: 2028/526, loss: 0.0896640345454216 2023-01-21 10:23:31.671371: step: 2032/526, loss: 0.08779249340295792 2023-01-21 10:23:32.810683: step: 2036/526, loss: 0.04977541044354439 2023-01-21 10:23:33.964484: step: 2040/526, loss: 0.06893427670001984 2023-01-21 10:23:35.106250: step: 2044/526, loss: 0.5643278956413269 2023-01-21 10:23:36.271117: step: 2048/526, loss: 0.18750472366809845 2023-01-21 10:23:37.427069: step: 2052/526, loss: 0.058579444885253906 2023-01-21 10:23:38.590099: step: 2056/526, loss: 0.06687898933887482 2023-01-21 10:23:39.767597: step: 2060/526, loss: 0.05959515646100044 2023-01-21 10:23:40.937306: step: 2064/526, loss: 0.04958000034093857 2023-01-21 10:23:42.081919: step: 2068/526, loss: 0.04323253780603409 2023-01-21 10:23:43.266619: step: 2072/526, loss: 0.05554013326764107 2023-01-21 10:23:44.433142: step: 2076/526, loss: 0.08727961033582687 2023-01-21 10:23:45.575291: step: 2080/526, loss: 1.0955184698104858 2023-01-21 10:23:46.764591: step: 2084/526, loss: 0.11226553469896317 2023-01-21 10:23:47.910420: step: 2088/526, loss: 0.06319661438465118 2023-01-21 10:23:49.091908: step: 2092/526, loss: 0.08107885718345642 2023-01-21 10:23:50.248426: step: 2096/526, loss: 0.04995689168572426 2023-01-21 10:23:51.417872: step: 2100/526, loss: 0.14095425605773926 2023-01-21 10:23:52.532042: step: 2104/526, loss: 0.03281044960021973 ================================================== Loss: 0.108 -------------------- Dev: {'event': {'p': 0.5503802281368821, 'r': 0.7709720372836218, 'f1': 0.6422628951747089}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Test: {'event': {'p': 0.5799212598425196, 'r': 0.7687891440501043, 'f1': 0.6611310592459605}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Chinese: {'event': {'p': 0.6, 'r': 0.9444444444444444, 'f1': 0.7338129496402879}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Korean: {'event': {'p': 0.717391304347826, 'r': 0.5238095238095238, 'f1': 0.6055045871559633}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} Russian: {'event': {'p': 0.4583333333333333, 'r': 0.6111111111111112, 'f1': 0.5238095238095238}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 9} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.60687432867884, 'r': 0.7523302263648469, 'f1': 0.6718192627824019}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Eng Test for Russian: {'event': {'p': 0.6179966044142614, 'r': 0.7599164926931107, 'f1': 0.6816479400749064}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} Sample Russian: {'event': {'p': 0.5263157894736842, 'r': 0.5555555555555556, 'f1': 0.5405405405405405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 8} ****************************** Epoch: 10 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:24:38.283568: step: 4/526, loss: 0.028320984914898872 2023-01-21 10:24:39.445862: step: 8/526, loss: 0.028383541852235794 2023-01-21 10:24:40.611765: step: 12/526, loss: 0.06290411949157715 2023-01-21 10:24:41.775400: step: 16/526, loss: 0.04937257617712021 2023-01-21 10:24:42.929838: step: 20/526, loss: 0.16861124336719513 2023-01-21 10:24:44.056983: step: 24/526, loss: 0.06576375663280487 2023-01-21 10:24:45.244838: step: 28/526, loss: 0.024859048426151276 2023-01-21 10:24:46.408600: step: 32/526, loss: 0.003791332244873047 2023-01-21 10:24:47.523718: step: 36/526, loss: 0.009054708294570446 2023-01-21 10:24:48.665366: step: 40/526, loss: 0.06260533630847931 2023-01-21 10:24:49.807713: step: 44/526, loss: 0.004513883963227272 2023-01-21 10:24:50.988809: step: 48/526, loss: 0.04992947727441788 2023-01-21 10:24:52.147970: step: 52/526, loss: 0.12240581214427948 2023-01-21 10:24:53.285497: step: 56/526, loss: 0.12667998671531677 2023-01-21 10:24:54.417328: step: 60/526, loss: 0.12655992805957794 2023-01-21 10:24:55.579830: step: 64/526, loss: 0.023884296417236328 2023-01-21 10:24:56.700889: step: 68/526, loss: 0.10118045657873154 2023-01-21 10:24:57.847065: step: 72/526, loss: 0.032205771654844284 2023-01-21 10:24:58.970894: step: 76/526, loss: 0.06212451681494713 2023-01-21 10:25:00.134369: step: 80/526, loss: 0.09330043941736221 2023-01-21 10:25:01.282593: step: 84/526, loss: 0.019300175830721855 2023-01-21 10:25:02.409072: step: 88/526, loss: 0.0037672044709324837 2023-01-21 10:25:03.543883: step: 92/526, loss: 0.0006354331853799522 2023-01-21 10:25:04.675568: step: 96/526, loss: 0.03278150409460068 2023-01-21 10:25:05.855592: step: 100/526, loss: 0.013471794314682484 2023-01-21 10:25:07.004760: step: 104/526, loss: 0.010409379377961159 2023-01-21 10:25:08.104110: step: 108/526, loss: 0.003136539366096258 2023-01-21 10:25:09.251755: step: 112/526, loss: 0.2667730450630188 2023-01-21 10:25:10.409165: step: 116/526, loss: 0.02319650538265705 2023-01-21 10:25:11.529148: step: 120/526, loss: 0.038399938493967056 2023-01-21 10:25:12.671079: step: 124/526, loss: 0.04065141826868057 2023-01-21 10:25:13.803050: step: 128/526, loss: 0.07226391136646271 2023-01-21 10:25:14.942033: step: 132/526, loss: 0.0392330177128315 2023-01-21 10:25:16.092804: step: 136/526, loss: 0.08125477284193039 2023-01-21 10:25:17.234127: step: 140/526, loss: 0.01510238740593195 2023-01-21 10:25:18.393138: step: 144/526, loss: 0.22804327309131622 2023-01-21 10:25:19.568828: step: 148/526, loss: 0.05395030975341797 2023-01-21 10:25:20.720944: step: 152/526, loss: 0.011246491223573685 2023-01-21 10:25:21.891425: step: 156/526, loss: 0.06688646972179413 2023-01-21 10:25:23.054577: step: 160/526, loss: 0.15941017866134644 2023-01-21 10:25:24.228809: step: 164/526, loss: 0.03614625707268715 2023-01-21 10:25:25.358811: step: 168/526, loss: 0.04990587383508682 2023-01-21 10:25:26.517703: step: 172/526, loss: 0.010376430116593838 2023-01-21 10:25:27.667211: step: 176/526, loss: 0.047675516456365585 2023-01-21 10:25:28.814780: step: 180/526, loss: 0.34093916416168213 2023-01-21 10:25:29.953749: step: 184/526, loss: 0.014830781146883965 2023-01-21 10:25:31.094173: step: 188/526, loss: 0.04991798475384712 2023-01-21 10:25:32.280928: step: 192/526, loss: 0.023826980963349342 2023-01-21 10:25:33.457146: step: 196/526, loss: 0.0010565996635705233 2023-01-21 10:25:34.628927: step: 200/526, loss: 0.04137573391199112 2023-01-21 10:25:35.801847: step: 204/526, loss: 0.07580319046974182 2023-01-21 10:25:36.965641: step: 208/526, loss: 0.030399512499570847 2023-01-21 10:25:38.106102: step: 212/526, loss: 0.06335706263780594 2023-01-21 10:25:39.267477: step: 216/526, loss: 0.0020377160981297493 2023-01-21 10:25:40.436167: step: 220/526, loss: 0.023698806762695312 2023-01-21 10:25:41.631187: step: 224/526, loss: 0.032497502863407135 2023-01-21 10:25:42.793109: step: 228/526, loss: 0.01242523267865181 2023-01-21 10:25:44.005995: step: 232/526, loss: 0.04625234752893448 2023-01-21 10:25:45.166666: step: 236/526, loss: 0.07313938438892365 2023-01-21 10:25:46.333274: step: 240/526, loss: 0.08535385131835938 2023-01-21 10:25:47.506445: step: 244/526, loss: 0.4234287738800049 2023-01-21 10:25:48.679360: step: 248/526, loss: 0.06997833400964737 2023-01-21 10:25:49.844403: step: 252/526, loss: 0.16430386900901794 2023-01-21 10:25:50.998801: step: 256/526, loss: 0.023958731442689896 2023-01-21 10:25:52.148363: step: 260/526, loss: 0.022513866424560547 2023-01-21 10:25:53.264178: step: 264/526, loss: 0.030522825196385384 2023-01-21 10:25:54.409074: step: 268/526, loss: 0.10798521339893341 2023-01-21 10:25:55.547650: step: 272/526, loss: 0.01524047926068306 2023-01-21 10:25:56.683616: step: 276/526, loss: 0.037169747054576874 2023-01-21 10:25:57.858017: step: 280/526, loss: 0.20213204622268677 2023-01-21 10:25:58.993904: step: 284/526, loss: 0.10551920533180237 2023-01-21 10:26:00.126909: step: 288/526, loss: 0.026165615767240524 2023-01-21 10:26:01.258098: step: 292/526, loss: 0.34799259901046753 2023-01-21 10:26:02.403893: step: 296/526, loss: 0.03233833611011505 2023-01-21 10:26:03.552119: step: 300/526, loss: 0.02612781524658203 2023-01-21 10:26:04.680267: step: 304/526, loss: 0.0483766570687294 2023-01-21 10:26:05.863595: step: 308/526, loss: 0.010968636721372604 2023-01-21 10:26:07.063064: step: 312/526, loss: 0.11256222426891327 2023-01-21 10:26:08.243134: step: 316/526, loss: 0.026598358526825905 2023-01-21 10:26:09.386648: step: 320/526, loss: 0.25557535886764526 2023-01-21 10:26:10.568271: step: 324/526, loss: 0.04780564457178116 2023-01-21 10:26:11.748222: step: 328/526, loss: 0.0025281906127929688 2023-01-21 10:26:12.913060: step: 332/526, loss: 0.0003452778037171811 2023-01-21 10:26:14.099427: step: 336/526, loss: 0.07560577243566513 2023-01-21 10:26:15.249663: step: 340/526, loss: 0.020679473876953125 2023-01-21 10:26:16.387984: step: 344/526, loss: 0.08357725292444229 2023-01-21 10:26:17.507225: step: 348/526, loss: 0.022170402109622955 2023-01-21 10:26:18.647631: step: 352/526, loss: 0.0702851265668869 2023-01-21 10:26:19.792425: step: 356/526, loss: 0.1083984375 2023-01-21 10:26:20.972513: step: 360/526, loss: 0.15933114290237427 2023-01-21 10:26:22.134200: step: 364/526, loss: 0.08095331490039825 2023-01-21 10:26:23.271588: step: 368/526, loss: 0.07179813832044601 2023-01-21 10:26:24.421770: step: 372/526, loss: 0.06184377521276474 2023-01-21 10:26:25.570175: step: 376/526, loss: 0.011195373721420765 2023-01-21 10:26:26.732002: step: 380/526, loss: 0.01475071907043457 2023-01-21 10:26:27.871336: step: 384/526, loss: 0.03557606041431427 2023-01-21 10:26:29.025771: step: 388/526, loss: 0.03391990810632706 2023-01-21 10:26:30.156906: step: 392/526, loss: 0.04469270631670952 2023-01-21 10:26:31.303325: step: 396/526, loss: 0.23098793625831604 2023-01-21 10:26:32.436707: step: 400/526, loss: 0.02031850814819336 2023-01-21 10:26:33.621601: step: 404/526, loss: 0.09688778221607208 2023-01-21 10:26:34.801006: step: 408/526, loss: 0.06264915317296982 2023-01-21 10:26:35.942673: step: 412/526, loss: 0.07998549938201904 2023-01-21 10:26:37.107219: step: 416/526, loss: 0.115503691136837 2023-01-21 10:26:38.256681: step: 420/526, loss: 0.025243043899536133 2023-01-21 10:26:39.388657: step: 424/526, loss: 0.017613792791962624 2023-01-21 10:26:40.547532: step: 428/526, loss: 0.1176474541425705 2023-01-21 10:26:41.731259: step: 432/526, loss: 0.043384552001953125 2023-01-21 10:26:42.913208: step: 436/526, loss: 0.007562064565718174 2023-01-21 10:26:44.108357: step: 440/526, loss: 0.09554615616798401 2023-01-21 10:26:45.251815: step: 444/526, loss: 0.0011378289200365543 2023-01-21 10:26:46.448242: step: 448/526, loss: 0.08882226794958115 2023-01-21 10:26:47.612313: step: 452/526, loss: 0.03608202934265137 2023-01-21 10:26:48.756340: step: 456/526, loss: 0.011974668130278587 2023-01-21 10:26:49.901673: step: 460/526, loss: 0.0049231527373194695 2023-01-21 10:26:51.044467: step: 464/526, loss: 0.14429016411304474 2023-01-21 10:26:52.210463: step: 468/526, loss: 0.09686315059661865 2023-01-21 10:26:53.389775: step: 472/526, loss: 0.02040109783411026 2023-01-21 10:26:54.545771: step: 476/526, loss: 0.01570301130414009 2023-01-21 10:26:55.712406: step: 480/526, loss: 0.024383926764130592 2023-01-21 10:26:56.866022: step: 484/526, loss: 0.18245373666286469 2023-01-21 10:26:58.003411: step: 488/526, loss: 0.052815936505794525 2023-01-21 10:26:59.138805: step: 492/526, loss: 0.03601484000682831 2023-01-21 10:27:00.292599: step: 496/526, loss: 0.02328806184232235 2023-01-21 10:27:01.450894: step: 500/526, loss: 0.06231861189007759 2023-01-21 10:27:02.597918: step: 504/526, loss: 0.038298606872558594 2023-01-21 10:27:03.762818: step: 508/526, loss: 0.09702196717262268 2023-01-21 10:27:04.937805: step: 512/526, loss: 0.04379863664507866 2023-01-21 10:27:06.072342: step: 516/526, loss: 0.055422186851501465 2023-01-21 10:27:07.203950: step: 520/526, loss: 0.017874527722597122 2023-01-21 10:27:08.350905: step: 524/526, loss: 0.02707233466207981 2023-01-21 10:27:09.513967: step: 528/526, loss: 0.053722333163022995 2023-01-21 10:27:10.679282: step: 532/526, loss: 0.11697397381067276 2023-01-21 10:27:11.850615: step: 536/526, loss: 0.06402167677879333 2023-01-21 10:27:13.018902: step: 540/526, loss: 0.04290056228637695 2023-01-21 10:27:14.162745: step: 544/526, loss: 0.029560375958681107 2023-01-21 10:27:15.333542: step: 548/526, loss: 0.0044727325439453125 2023-01-21 10:27:16.480970: step: 552/526, loss: 0.02180004119873047 2023-01-21 10:27:17.691679: step: 556/526, loss: 0.03216296434402466 2023-01-21 10:27:18.898247: step: 560/526, loss: 0.07602892071008682 2023-01-21 10:27:20.080533: step: 564/526, loss: 0.01582932472229004 2023-01-21 10:27:21.205345: step: 568/526, loss: 0.03550272062420845 2023-01-21 10:27:22.352741: step: 572/526, loss: 0.06587543338537216 2023-01-21 10:27:23.503289: step: 576/526, loss: 0.09434433281421661 2023-01-21 10:27:24.658371: step: 580/526, loss: 0.05430154874920845 2023-01-21 10:27:25.800278: step: 584/526, loss: 0.29151880741119385 2023-01-21 10:27:26.955023: step: 588/526, loss: 0.014581013470888138 2023-01-21 10:27:28.108178: step: 592/526, loss: 0.03231649473309517 2023-01-21 10:27:29.285607: step: 596/526, loss: 0.1652783304452896 2023-01-21 10:27:30.419498: step: 600/526, loss: 0.0006856441032141447 2023-01-21 10:27:31.603516: step: 604/526, loss: 0.10957708209753036 2023-01-21 10:27:32.738021: step: 608/526, loss: 0.0373167023062706 2023-01-21 10:27:33.884763: step: 612/526, loss: 0.0268572811037302 2023-01-21 10:27:35.043606: step: 616/526, loss: 0.043706513941287994 2023-01-21 10:27:36.193410: step: 620/526, loss: 0.1443004608154297 2023-01-21 10:27:37.351611: step: 624/526, loss: 0.032207682728767395 2023-01-21 10:27:38.497004: step: 628/526, loss: 0.14608363807201385 2023-01-21 10:27:39.659817: step: 632/526, loss: 1.1984728574752808 2023-01-21 10:27:40.816173: step: 636/526, loss: 0.04691801220178604 2023-01-21 10:27:41.969765: step: 640/526, loss: 0.13143710792064667 2023-01-21 10:27:43.145804: step: 644/526, loss: 0.04209737852215767 2023-01-21 10:27:44.281294: step: 648/526, loss: 0.07995434105396271 2023-01-21 10:27:45.457404: step: 652/526, loss: 0.005539512727409601 2023-01-21 10:27:46.631135: step: 656/526, loss: 0.032628726214170456 2023-01-21 10:27:47.762417: step: 660/526, loss: 0.030490685254335403 2023-01-21 10:27:48.930222: step: 664/526, loss: 0.073316290974617 2023-01-21 10:27:50.084532: step: 668/526, loss: 0.496309757232666 2023-01-21 10:27:51.248272: step: 672/526, loss: 0.1407543122768402 2023-01-21 10:27:52.426493: step: 676/526, loss: 0.004931735806167126 2023-01-21 10:27:53.598761: step: 680/526, loss: 0.011069202795624733 2023-01-21 10:27:54.762367: step: 684/526, loss: 0.011652231216430664 2023-01-21 10:27:55.919283: step: 688/526, loss: 0.05863799899816513 2023-01-21 10:27:57.050585: step: 692/526, loss: 0.12837858498096466 2023-01-21 10:27:58.223291: step: 696/526, loss: 0.010602951049804688 2023-01-21 10:27:59.352165: step: 700/526, loss: 0.0979333445429802 2023-01-21 10:28:00.494052: step: 704/526, loss: 0.014994812197983265 2023-01-21 10:28:01.627943: step: 708/526, loss: 0.00142078404314816 2023-01-21 10:28:02.784075: step: 712/526, loss: 0.01856342703104019 2023-01-21 10:28:03.922082: step: 716/526, loss: 0.06520041823387146 2023-01-21 10:28:05.060857: step: 720/526, loss: 0.01949338987469673 2023-01-21 10:28:06.196866: step: 724/526, loss: 0.011514759622514248 2023-01-21 10:28:07.355646: step: 728/526, loss: 0.22837772965431213 2023-01-21 10:28:08.478550: step: 732/526, loss: 0.08065585792064667 2023-01-21 10:28:09.622230: step: 736/526, loss: 0.10996055603027344 2023-01-21 10:28:10.791861: step: 740/526, loss: 0.042223989963531494 2023-01-21 10:28:11.935621: step: 744/526, loss: 0.0754416435956955 2023-01-21 10:28:13.098480: step: 748/526, loss: 0.05606689676642418 2023-01-21 10:28:14.249075: step: 752/526, loss: 0.00275592808611691 2023-01-21 10:28:15.398994: step: 756/526, loss: 0.10385185480117798 2023-01-21 10:28:16.605778: step: 760/526, loss: 0.09406185150146484 2023-01-21 10:28:17.763540: step: 764/526, loss: 0.04433479160070419 2023-01-21 10:28:18.956302: step: 768/526, loss: 0.03898601606488228 2023-01-21 10:28:20.076864: step: 772/526, loss: 0.047724343836307526 2023-01-21 10:28:21.213114: step: 776/526, loss: 0.014749574474990368 2023-01-21 10:28:22.374671: step: 780/526, loss: 0.009395027533173561 2023-01-21 10:28:23.539427: step: 784/526, loss: 0.132147878408432 2023-01-21 10:28:24.682940: step: 788/526, loss: 0.014745360240340233 2023-01-21 10:28:25.832412: step: 792/526, loss: 0.0352899543941021 2023-01-21 10:28:26.989412: step: 796/526, loss: 0.05112161859869957 2023-01-21 10:28:28.156093: step: 800/526, loss: 0.05016679689288139 2023-01-21 10:28:29.319498: step: 804/526, loss: 0.04091458395123482 2023-01-21 10:28:30.478547: step: 808/526, loss: 0.03102698363363743 2023-01-21 10:28:31.652359: step: 812/526, loss: 0.07510252296924591 2023-01-21 10:28:32.785508: step: 816/526, loss: 0.009077358059585094 2023-01-21 10:28:33.945288: step: 820/526, loss: 0.0262330062687397 2023-01-21 10:28:35.063956: step: 824/526, loss: 0.02296009100973606 2023-01-21 10:28:36.208483: step: 828/526, loss: 0.047461796551942825 2023-01-21 10:28:37.371416: step: 832/526, loss: 0.021651649847626686 2023-01-21 10:28:38.533802: step: 836/526, loss: 0.041078757494688034 2023-01-21 10:28:39.672121: step: 840/526, loss: 0.37376919388771057 2023-01-21 10:28:40.844121: step: 844/526, loss: 0.1277790069580078 2023-01-21 10:28:42.019200: step: 848/526, loss: 0.08828449249267578 2023-01-21 10:28:43.218784: step: 852/526, loss: 0.15048810839653015 2023-01-21 10:28:44.366264: step: 856/526, loss: 0.20188234746456146 2023-01-21 10:28:45.512905: step: 860/526, loss: 0.0610136054456234 2023-01-21 10:28:46.667689: step: 864/526, loss: 0.1468985676765442 2023-01-21 10:28:47.806569: step: 868/526, loss: 0.13093385100364685 2023-01-21 10:28:48.974045: step: 872/526, loss: 0.34043940901756287 2023-01-21 10:28:50.148502: step: 876/526, loss: 0.046022605150938034 2023-01-21 10:28:51.335166: step: 880/526, loss: 0.057181455194950104 2023-01-21 10:28:52.513426: step: 884/526, loss: 0.06816444545984268 2023-01-21 10:28:53.699639: step: 888/526, loss: 0.06227097287774086 2023-01-21 10:28:54.882016: step: 892/526, loss: 0.030421830713748932 2023-01-21 10:28:56.043898: step: 896/526, loss: 0.19506484270095825 2023-01-21 10:28:57.191849: step: 900/526, loss: 0.020766019821166992 2023-01-21 10:28:58.336156: step: 904/526, loss: 0.14779242873191833 2023-01-21 10:28:59.499924: step: 908/526, loss: 0.007510280702263117 2023-01-21 10:29:00.688394: step: 912/526, loss: 0.09319205582141876 2023-01-21 10:29:01.857902: step: 916/526, loss: 0.06066341698169708 2023-01-21 10:29:03.027325: step: 920/526, loss: 0.032367803156375885 2023-01-21 10:29:04.188738: step: 924/526, loss: 0.19828377664089203 2023-01-21 10:29:05.310575: step: 928/526, loss: 0.0059191701002418995 2023-01-21 10:29:06.447250: step: 932/526, loss: 0.00969848595559597 2023-01-21 10:29:07.616240: step: 936/526, loss: 0.019677067175507545 2023-01-21 10:29:08.746859: step: 940/526, loss: 0.06122016906738281 2023-01-21 10:29:09.879509: step: 944/526, loss: 0.011370975524187088 2023-01-21 10:29:11.010920: step: 948/526, loss: 0.013803817331790924 2023-01-21 10:29:12.133793: step: 952/526, loss: 0.02003936655819416 2023-01-21 10:29:13.272337: step: 956/526, loss: 0.02453165128827095 2023-01-21 10:29:14.418320: step: 960/526, loss: 0.007159424014389515 2023-01-21 10:29:15.585821: step: 964/526, loss: 0.027132224291563034 2023-01-21 10:29:16.736232: step: 968/526, loss: 0.06755886226892471 2023-01-21 10:29:17.881037: step: 972/526, loss: 0.08049201965332031 2023-01-21 10:29:19.061560: step: 976/526, loss: 0.16731682419776917 2023-01-21 10:29:20.229618: step: 980/526, loss: 0.08363030105829239 2023-01-21 10:29:21.385651: step: 984/526, loss: 0.01293573435395956 2023-01-21 10:29:22.531445: step: 988/526, loss: 0.044126223772764206 2023-01-21 10:29:23.690595: step: 992/526, loss: 0.23999693989753723 2023-01-21 10:29:24.838111: step: 996/526, loss: 0.03117094188928604 2023-01-21 10:29:25.979827: step: 1000/526, loss: 0.04149474948644638 2023-01-21 10:29:27.132372: step: 1004/526, loss: 0.030040835961699486 2023-01-21 10:29:28.255650: step: 1008/526, loss: 0.05211935192346573 2023-01-21 10:29:29.404031: step: 1012/526, loss: 0.07298517227172852 2023-01-21 10:29:30.595405: step: 1016/526, loss: 0.004798555746674538 2023-01-21 10:29:31.765774: step: 1020/526, loss: 0.07155527919530869 2023-01-21 10:29:32.899114: step: 1024/526, loss: 0.009870529174804688 2023-01-21 10:29:34.086512: step: 1028/526, loss: 0.06878840923309326 2023-01-21 10:29:35.260611: step: 1032/526, loss: 0.054642342031002045 2023-01-21 10:29:36.409013: step: 1036/526, loss: 0.059169963002204895 2023-01-21 10:29:37.551090: step: 1040/526, loss: 0.12754793465137482 2023-01-21 10:29:38.688223: step: 1044/526, loss: 0.030283259227871895 2023-01-21 10:29:39.817267: step: 1048/526, loss: 0.025376440957188606 2023-01-21 10:29:40.961785: step: 1052/526, loss: 0.01849508285522461 2023-01-21 10:29:42.091713: step: 1056/526, loss: 0.007206773851066828 2023-01-21 10:29:43.287364: step: 1060/526, loss: 0.022374916821718216 2023-01-21 10:29:44.447324: step: 1064/526, loss: 0.05935974046587944 2023-01-21 10:29:45.596967: step: 1068/526, loss: 0.02503185346722603 2023-01-21 10:29:46.793743: step: 1072/526, loss: 0.05175476148724556 2023-01-21 10:29:47.960535: step: 1076/526, loss: 0.05761241912841797 2023-01-21 10:29:49.148760: step: 1080/526, loss: 0.03208618238568306 2023-01-21 10:29:50.285546: step: 1084/526, loss: 0.013869166374206543 2023-01-21 10:29:51.424016: step: 1088/526, loss: 0.3287279009819031 2023-01-21 10:29:52.547065: step: 1092/526, loss: 0.12407165765762329 2023-01-21 10:29:53.671717: step: 1096/526, loss: 0.07733113318681717 2023-01-21 10:29:54.872582: step: 1100/526, loss: 0.02474966086447239 2023-01-21 10:29:56.040699: step: 1104/526, loss: 0.05765619874000549 2023-01-21 10:29:57.212044: step: 1108/526, loss: 0.055417634546756744 2023-01-21 10:29:58.372717: step: 1112/526, loss: 0.07246103137731552 2023-01-21 10:29:59.512854: step: 1116/526, loss: 0.05764341354370117 2023-01-21 10:30:00.653070: step: 1120/526, loss: 0.08796558529138565 2023-01-21 10:30:01.819096: step: 1124/526, loss: 0.11782994121313095 2023-01-21 10:30:02.981483: step: 1128/526, loss: 0.6517559289932251 2023-01-21 10:30:04.154320: step: 1132/526, loss: 0.03792142868041992 2023-01-21 10:30:05.299818: step: 1136/526, loss: 0.012488747015595436 2023-01-21 10:30:06.459503: step: 1140/526, loss: 0.003528594970703125 2023-01-21 10:30:07.628011: step: 1144/526, loss: 0.09680843353271484 2023-01-21 10:30:08.766774: step: 1148/526, loss: 0.023140262812376022 2023-01-21 10:30:09.950410: step: 1152/526, loss: 0.025685502216219902 2023-01-21 10:30:11.105530: step: 1156/526, loss: 0.00809102039784193 2023-01-21 10:30:12.283110: step: 1160/526, loss: 0.11041207611560822 2023-01-21 10:30:13.449570: step: 1164/526, loss: 0.06896954029798508 2023-01-21 10:30:14.590794: step: 1168/526, loss: 0.1656995713710785 2023-01-21 10:30:15.760213: step: 1172/526, loss: 0.08468323200941086 2023-01-21 10:30:16.911537: step: 1176/526, loss: 0.06340312957763672 2023-01-21 10:30:18.045666: step: 1180/526, loss: 0.054660990834236145 2023-01-21 10:30:19.191303: step: 1184/526, loss: 0.009999657049775124 2023-01-21 10:30:20.375623: step: 1188/526, loss: 0.19355127215385437 2023-01-21 10:30:21.503857: step: 1192/526, loss: 0.08167479187250137 2023-01-21 10:30:22.663330: step: 1196/526, loss: 0.04205026850104332 2023-01-21 10:30:23.844390: step: 1200/526, loss: 0.10201139748096466 2023-01-21 10:30:24.997739: step: 1204/526, loss: 0.08033923804759979 2023-01-21 10:30:26.136882: step: 1208/526, loss: 0.10385928303003311 2023-01-21 10:30:27.325342: step: 1212/526, loss: 0.08103971183300018 2023-01-21 10:30:28.519712: step: 1216/526, loss: 0.0748133659362793 2023-01-21 10:30:29.680241: step: 1220/526, loss: 0.022324658930301666 2023-01-21 10:30:30.840249: step: 1224/526, loss: 0.04760704189538956 2023-01-21 10:30:32.006113: step: 1228/526, loss: 0.014558887109160423 2023-01-21 10:30:33.175678: step: 1232/526, loss: 0.031389132142066956 2023-01-21 10:30:34.343197: step: 1236/526, loss: 0.02243518829345703 2023-01-21 10:30:35.481366: step: 1240/526, loss: 0.02817354165017605 2023-01-21 10:30:36.634038: step: 1244/526, loss: 0.04374952241778374 2023-01-21 10:30:37.784431: step: 1248/526, loss: 0.050734326243400574 2023-01-21 10:30:38.953900: step: 1252/526, loss: 0.09195242077112198 2023-01-21 10:30:40.117372: step: 1256/526, loss: 0.03470778465270996 2023-01-21 10:30:41.288751: step: 1260/526, loss: 0.0790136381983757 2023-01-21 10:30:42.473686: step: 1264/526, loss: 0.1336391419172287 2023-01-21 10:30:43.634014: step: 1268/526, loss: 0.03554688021540642 2023-01-21 10:30:44.774318: step: 1272/526, loss: 0.02658400498330593 2023-01-21 10:30:45.957876: step: 1276/526, loss: 0.06641463935375214 2023-01-21 10:30:47.106046: step: 1280/526, loss: 0.0794897973537445 2023-01-21 10:30:48.289430: step: 1284/526, loss: 0.05838785320520401 2023-01-21 10:30:49.436951: step: 1288/526, loss: 0.01993265189230442 2023-01-21 10:30:50.585307: step: 1292/526, loss: 0.2914840579032898 2023-01-21 10:30:51.745691: step: 1296/526, loss: 0.026393413543701172 2023-01-21 10:30:52.897463: step: 1300/526, loss: 0.051714133471250534 2023-01-21 10:30:54.047056: step: 1304/526, loss: 0.033529095351696014 2023-01-21 10:30:55.197733: step: 1308/526, loss: 0.15825581550598145 2023-01-21 10:30:56.349544: step: 1312/526, loss: 0.010504817590117455 2023-01-21 10:30:57.500798: step: 1316/526, loss: 0.1150369644165039 2023-01-21 10:30:58.692850: step: 1320/526, loss: 0.10432469844818115 2023-01-21 10:30:59.841585: step: 1324/526, loss: 0.023850345984101295 2023-01-21 10:31:00.999951: step: 1328/526, loss: 0.3994367718696594 2023-01-21 10:31:02.124421: step: 1332/526, loss: 0.003780448343604803 2023-01-21 10:31:03.303294: step: 1336/526, loss: 0.05550122261047363 2023-01-21 10:31:04.485611: step: 1340/526, loss: 0.05399675294756889 2023-01-21 10:31:05.607054: step: 1344/526, loss: 0.008647250942885876 2023-01-21 10:31:06.723217: step: 1348/526, loss: 0.03377723693847656 2023-01-21 10:31:07.867340: step: 1352/526, loss: 0.08250083774328232 2023-01-21 10:31:09.017616: step: 1356/526, loss: 0.09580240398645401 2023-01-21 10:31:10.172676: step: 1360/526, loss: 0.017348527908325195 2023-01-21 10:31:11.331266: step: 1364/526, loss: 0.0042724609375 2023-01-21 10:31:12.494585: step: 1368/526, loss: 0.017301440238952637 2023-01-21 10:31:13.682941: step: 1372/526, loss: 0.004537248983979225 2023-01-21 10:31:14.898933: step: 1376/526, loss: 0.14740857481956482 2023-01-21 10:31:16.081448: step: 1380/526, loss: 0.03909559175372124 2023-01-21 10:31:17.247806: step: 1384/526, loss: 0.16381597518920898 2023-01-21 10:31:18.397678: step: 1388/526, loss: 0.03183326870203018 2023-01-21 10:31:19.525742: step: 1392/526, loss: 0.0280427448451519 2023-01-21 10:31:20.698201: step: 1396/526, loss: 0.1238887831568718 2023-01-21 10:31:21.878483: step: 1400/526, loss: 0.07164116203784943 2023-01-21 10:31:23.058661: step: 1404/526, loss: 0.03279304504394531 2023-01-21 10:31:24.216508: step: 1408/526, loss: 0.05606594309210777 2023-01-21 10:31:25.384069: step: 1412/526, loss: 0.014059877023100853 2023-01-21 10:31:26.523821: step: 1416/526, loss: 0.004166269209235907 2023-01-21 10:31:27.670302: step: 1420/526, loss: 0.2109052687883377 2023-01-21 10:31:28.809886: step: 1424/526, loss: 0.0957794189453125 2023-01-21 10:31:29.986229: step: 1428/526, loss: 0.028122520074248314 2023-01-21 10:31:31.119306: step: 1432/526, loss: 0.0031442642211914062 2023-01-21 10:31:32.260251: step: 1436/526, loss: 0.2259502410888672 2023-01-21 10:31:33.415532: step: 1440/526, loss: 0.05520162731409073 2023-01-21 10:31:34.574059: step: 1444/526, loss: 0.09885750710964203 2023-01-21 10:31:35.719106: step: 1448/526, loss: 0.017478371039032936 2023-01-21 10:31:36.877291: step: 1452/526, loss: 0.055128954350948334 2023-01-21 10:31:38.023389: step: 1456/526, loss: 0.0517701655626297 2023-01-21 10:31:39.182071: step: 1460/526, loss: 0.011632347479462624 2023-01-21 10:31:40.350816: step: 1464/526, loss: 0.03318033367395401 2023-01-21 10:31:41.532740: step: 1468/526, loss: 0.11411857604980469 2023-01-21 10:31:42.672577: step: 1472/526, loss: 0.19330093264579773 2023-01-21 10:31:43.845733: step: 1476/526, loss: 0.06672678142786026 2023-01-21 10:31:45.028672: step: 1480/526, loss: 0.08597736805677414 2023-01-21 10:31:46.147707: step: 1484/526, loss: 0.10448607057332993 2023-01-21 10:31:47.334128: step: 1488/526, loss: 0.01931598410010338 2023-01-21 10:31:48.514984: step: 1492/526, loss: 0.013746977783739567 2023-01-21 10:31:49.651505: step: 1496/526, loss: 0.004908561706542969 2023-01-21 10:31:50.859253: step: 1500/526, loss: 0.03815726935863495 2023-01-21 10:31:51.996785: step: 1504/526, loss: 0.028313064947724342 2023-01-21 10:31:53.156828: step: 1508/526, loss: 0.06979440897703171 2023-01-21 10:31:54.354755: step: 1512/526, loss: 0.07203473895788193 2023-01-21 10:31:55.476198: step: 1516/526, loss: 0.05343141779303551 2023-01-21 10:31:56.663754: step: 1520/526, loss: 0.027323532849550247 2023-01-21 10:31:57.823221: step: 1524/526, loss: 0.07754402607679367 2023-01-21 10:31:58.974342: step: 1528/526, loss: 0.06882844120264053 2023-01-21 10:32:00.118519: step: 1532/526, loss: 0.01407613791525364 2023-01-21 10:32:01.296322: step: 1536/526, loss: 0.02982468530535698 2023-01-21 10:32:02.459667: step: 1540/526, loss: 0.19618263840675354 2023-01-21 10:32:03.592236: step: 1544/526, loss: 0.30988186597824097 2023-01-21 10:32:04.743525: step: 1548/526, loss: 0.03654642030596733 2023-01-21 10:32:05.868289: step: 1552/526, loss: 0.007747936062514782 2023-01-21 10:32:07.032126: step: 1556/526, loss: 0.10610080510377884 2023-01-21 10:32:08.191677: step: 1560/526, loss: 0.10457658767700195 2023-01-21 10:32:09.341344: step: 1564/526, loss: 0.35605552792549133 2023-01-21 10:32:10.521549: step: 1568/526, loss: 0.0793820396065712 2023-01-21 10:32:11.639708: step: 1572/526, loss: 0.13206414878368378 2023-01-21 10:32:12.800181: step: 1576/526, loss: 0.12931156158447266 2023-01-21 10:32:13.992089: step: 1580/526, loss: 0.052256204187870026 2023-01-21 10:32:15.149541: step: 1584/526, loss: 0.07471618801355362 2023-01-21 10:32:16.286492: step: 1588/526, loss: 0.14208492636680603 2023-01-21 10:32:17.428638: step: 1592/526, loss: 0.45212239027023315 2023-01-21 10:32:18.576785: step: 1596/526, loss: 0.019907426089048386 2023-01-21 10:32:19.754376: step: 1600/526, loss: 0.03246612846851349 2023-01-21 10:32:20.901682: step: 1604/526, loss: 0.05747561901807785 2023-01-21 10:32:22.041782: step: 1608/526, loss: 0.02671222761273384 2023-01-21 10:32:23.184233: step: 1612/526, loss: 0.03648632764816284 2023-01-21 10:32:24.329882: step: 1616/526, loss: 0.03674755245447159 2023-01-21 10:32:25.504168: step: 1620/526, loss: 0.008162951096892357 2023-01-21 10:32:26.665228: step: 1624/526, loss: 0.03517938032746315 2023-01-21 10:32:27.785583: step: 1628/526, loss: 0.08168144524097443 2023-01-21 10:32:28.915771: step: 1632/526, loss: 0.16347496211528778 2023-01-21 10:32:30.050219: step: 1636/526, loss: 0.04364819452166557 2023-01-21 10:32:31.198023: step: 1640/526, loss: 0.014599466696381569 2023-01-21 10:32:32.328341: step: 1644/526, loss: 0.026294518262147903 2023-01-21 10:32:33.525386: step: 1648/526, loss: 0.047118380665779114 2023-01-21 10:32:34.702603: step: 1652/526, loss: 0.07855091243982315 2023-01-21 10:32:35.893377: step: 1656/526, loss: 0.21931058168411255 2023-01-21 10:32:37.060819: step: 1660/526, loss: 0.08486151695251465 2023-01-21 10:32:38.192480: step: 1664/526, loss: 0.027772333472967148 2023-01-21 10:32:39.337726: step: 1668/526, loss: 0.04638214409351349 2023-01-21 10:32:40.497512: step: 1672/526, loss: 0.020376110449433327 2023-01-21 10:32:41.684565: step: 1676/526, loss: 0.12195605784654617 2023-01-21 10:32:42.859883: step: 1680/526, loss: 0.11245956271886826 2023-01-21 10:32:43.990432: step: 1684/526, loss: 0.18732263147830963 2023-01-21 10:32:45.139487: step: 1688/526, loss: 0.42811423540115356 2023-01-21 10:32:46.291056: step: 1692/526, loss: 0.033809103071689606 2023-01-21 10:32:47.435933: step: 1696/526, loss: 0.02467980608344078 2023-01-21 10:32:48.624121: step: 1700/526, loss: 0.07565413415431976 2023-01-21 10:32:49.764564: step: 1704/526, loss: 0.06971245259046555 2023-01-21 10:32:50.898956: step: 1708/526, loss: 0.02219533920288086 2023-01-21 10:32:52.036615: step: 1712/526, loss: 0.035179756581783295 2023-01-21 10:32:53.192617: step: 1716/526, loss: 0.024142790585756302 2023-01-21 10:32:54.347484: step: 1720/526, loss: 0.045146238058805466 2023-01-21 10:32:55.489492: step: 1724/526, loss: 0.008876467123627663 2023-01-21 10:32:56.665188: step: 1728/526, loss: 0.07120724022388458 2023-01-21 10:32:57.788371: step: 1732/526, loss: 0.004645633976906538 2023-01-21 10:32:58.944732: step: 1736/526, loss: 0.04314884915947914 2023-01-21 10:33:00.047314: step: 1740/526, loss: 0.12078976631164551 2023-01-21 10:33:01.200815: step: 1744/526, loss: 0.023958493024110794 2023-01-21 10:33:02.357201: step: 1748/526, loss: 0.08768635243177414 2023-01-21 10:33:03.507383: step: 1752/526, loss: 0.01675243303179741 2023-01-21 10:33:04.678707: step: 1756/526, loss: 0.028470231220126152 2023-01-21 10:33:05.819512: step: 1760/526, loss: 0.004552984144538641 2023-01-21 10:33:06.985186: step: 1764/526, loss: 0.10632573068141937 2023-01-21 10:33:08.173036: step: 1768/526, loss: 0.11652755737304688 2023-01-21 10:33:09.307003: step: 1772/526, loss: 0.13604259490966797 2023-01-21 10:33:10.446284: step: 1776/526, loss: 0.28422442078590393 2023-01-21 10:33:11.592833: step: 1780/526, loss: 0.048311520367860794 2023-01-21 10:33:12.723100: step: 1784/526, loss: 0.03549493849277496 2023-01-21 10:33:13.897943: step: 1788/526, loss: 0.01968097686767578 2023-01-21 10:33:15.054967: step: 1792/526, loss: 0.02958536148071289 2023-01-21 10:33:16.204935: step: 1796/526, loss: 0.007223033811897039 2023-01-21 10:33:17.376053: step: 1800/526, loss: 0.17660921812057495 2023-01-21 10:33:18.537494: step: 1804/526, loss: 0.010737895965576172 2023-01-21 10:33:19.713379: step: 1808/526, loss: 0.03359851986169815 2023-01-21 10:33:20.883336: step: 1812/526, loss: 0.0781548023223877 2023-01-21 10:33:22.053494: step: 1816/526, loss: 0.1672716736793518 2023-01-21 10:33:23.187764: step: 1820/526, loss: 0.009849358350038528 2023-01-21 10:33:24.345826: step: 1824/526, loss: 0.11711722612380981 2023-01-21 10:33:25.491004: step: 1828/526, loss: 0.019664956256747246 2023-01-21 10:33:26.630204: step: 1832/526, loss: 0.016196250915527344 2023-01-21 10:33:27.797629: step: 1836/526, loss: 0.12970009446144104 2023-01-21 10:33:28.937868: step: 1840/526, loss: 0.041359614580869675 2023-01-21 10:33:30.090927: step: 1844/526, loss: 0.02796030230820179 2023-01-21 10:33:31.227577: step: 1848/526, loss: 0.10098437964916229 2023-01-21 10:33:32.425549: step: 1852/526, loss: 0.0474674254655838 2023-01-21 10:33:33.595427: step: 1856/526, loss: 0.02993020974099636 2023-01-21 10:33:34.756616: step: 1860/526, loss: 0.08916177600622177 2023-01-21 10:33:35.923383: step: 1864/526, loss: 0.05043373256921768 2023-01-21 10:33:37.061914: step: 1868/526, loss: 0.13478927314281464 2023-01-21 10:33:38.221199: step: 1872/526, loss: 0.003322315402328968 2023-01-21 10:33:39.339343: step: 1876/526, loss: 0.048246145248413086 2023-01-21 10:33:40.486266: step: 1880/526, loss: 0.0334840789437294 2023-01-21 10:33:41.640447: step: 1884/526, loss: 0.0033965588081628084 2023-01-21 10:33:42.787884: step: 1888/526, loss: 0.018155859783291817 2023-01-21 10:33:43.935928: step: 1892/526, loss: 0.005624675657600164 2023-01-21 10:33:45.088387: step: 1896/526, loss: 0.027116205543279648 2023-01-21 10:33:46.214389: step: 1900/526, loss: 0.08187694847583771 2023-01-21 10:33:47.330921: step: 1904/526, loss: 0.0037211417220532894 2023-01-21 10:33:48.505757: step: 1908/526, loss: 0.04520740360021591 2023-01-21 10:33:49.645851: step: 1912/526, loss: 0.515699565410614 2023-01-21 10:33:50.770396: step: 1916/526, loss: 0.039029501378536224 2023-01-21 10:33:51.916097: step: 1920/526, loss: 0.06831901520490646 2023-01-21 10:33:53.047525: step: 1924/526, loss: 0.035895347595214844 2023-01-21 10:33:54.225916: step: 1928/526, loss: 0.024035751819610596 2023-01-21 10:33:55.377425: step: 1932/526, loss: 0.043511394411325455 2023-01-21 10:33:56.519617: step: 1936/526, loss: 0.023209620267152786 2023-01-21 10:33:57.697702: step: 1940/526, loss: 0.08647099137306213 2023-01-21 10:33:58.827798: step: 1944/526, loss: 0.08339639008045197 2023-01-21 10:34:00.015102: step: 1948/526, loss: 0.08672972023487091 2023-01-21 10:34:01.160676: step: 1952/526, loss: 0.1704721450805664 2023-01-21 10:34:02.314488: step: 1956/526, loss: 0.13986149430274963 2023-01-21 10:34:03.488349: step: 1960/526, loss: 0.10825100541114807 2023-01-21 10:34:04.642624: step: 1964/526, loss: 1.1889848709106445 2023-01-21 10:34:05.814089: step: 1968/526, loss: 0.027541637420654297 2023-01-21 10:34:06.970073: step: 1972/526, loss: 0.0064182281494140625 2023-01-21 10:34:08.141735: step: 1976/526, loss: 0.03887768089771271 2023-01-21 10:34:09.299027: step: 1980/526, loss: 0.07199759781360626 2023-01-21 10:34:10.477072: step: 1984/526, loss: 0.03553476184606552 2023-01-21 10:34:11.611590: step: 1988/526, loss: 0.04219427332282066 2023-01-21 10:34:12.805112: step: 1992/526, loss: 0.22363309562206268 2023-01-21 10:34:14.018085: step: 1996/526, loss: 0.04667210578918457 2023-01-21 10:34:15.130420: step: 2000/526, loss: 0.0815645232796669 2023-01-21 10:34:16.312542: step: 2004/526, loss: 0.0011970996856689453 2023-01-21 10:34:17.453832: step: 2008/526, loss: 0.05406847223639488 2023-01-21 10:34:18.586936: step: 2012/526, loss: 0.15167349576950073 2023-01-21 10:34:19.767331: step: 2016/526, loss: 0.08149395138025284 2023-01-21 10:34:20.922559: step: 2020/526, loss: 0.1271902173757553 2023-01-21 10:34:22.177348: step: 2024/526, loss: 0.035814858973026276 2023-01-21 10:34:23.373365: step: 2028/526, loss: 0.2741064131259918 2023-01-21 10:34:24.528030: step: 2032/526, loss: 0.19340907037258148 2023-01-21 10:34:25.669864: step: 2036/526, loss: 0.05843010172247887 2023-01-21 10:34:26.826545: step: 2040/526, loss: 0.0958930030465126 2023-01-21 10:34:27.980923: step: 2044/526, loss: 0.07399062812328339 2023-01-21 10:34:29.131090: step: 2048/526, loss: 0.08161049336194992 2023-01-21 10:34:30.314579: step: 2052/526, loss: 0.18087102472782135 2023-01-21 10:34:31.480910: step: 2056/526, loss: 0.09563814103603363 2023-01-21 10:34:32.620470: step: 2060/526, loss: 0.050290297716856 2023-01-21 10:34:33.776333: step: 2064/526, loss: 0.10206671059131622 2023-01-21 10:34:34.970904: step: 2068/526, loss: 0.07828088104724884 2023-01-21 10:34:36.113300: step: 2072/526, loss: 0.03274235874414444 2023-01-21 10:34:37.283893: step: 2076/526, loss: 0.012637329287827015 2023-01-21 10:34:38.450904: step: 2080/526, loss: 0.09160013496875763 2023-01-21 10:34:39.649170: step: 2084/526, loss: 0.02294163778424263 2023-01-21 10:34:40.790092: step: 2088/526, loss: 0.006971931550651789 2023-01-21 10:34:41.969616: step: 2092/526, loss: 0.0292356014251709 2023-01-21 10:34:43.126393: step: 2096/526, loss: 0.14679566025733948 2023-01-21 10:34:44.267668: step: 2100/526, loss: 0.5074905157089233 2023-01-21 10:34:45.393027: step: 2104/526, loss: 0.5970926284790039 ================================================== Loss: 0.078 -------------------- Dev: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Test: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Chinese: {'event': {'p': 0.5595238095238095, 'r': 0.8703703703703703, 'f1': 0.6811594202898551}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Korean: {'event': {'p': 0.6530612244897959, 'r': 0.5079365079365079, 'f1': 0.5714285714285714}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6604010025062657, 'r': 0.7017310252996005, 'f1': 0.6804389928986443}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Eng Test for Chinese: {'event': {'p': 0.6480836236933798, 'r': 0.6795407098121086, 'f1': 0.6634394904458599}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} Sample Chinese: {'event': {'p': 0.6323529411764706, 'r': 0.7962962962962963, 'f1': 0.7049180327868853}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 2} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 11 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:35:40.318557: step: 4/526, loss: 0.7875440716743469 2023-01-21 10:35:41.463125: step: 8/526, loss: 0.02249932289123535 2023-01-21 10:35:42.646122: step: 12/526, loss: 0.004925346467643976 2023-01-21 10:35:43.766581: step: 16/526, loss: 0.02322368696331978 2023-01-21 10:35:44.902029: step: 20/526, loss: 0.05778513103723526 2023-01-21 10:35:46.070943: step: 24/526, loss: 0.027384568005800247 2023-01-21 10:35:47.234391: step: 28/526, loss: 0.04867757484316826 2023-01-21 10:35:48.351168: step: 32/526, loss: 0.03923201560974121 2023-01-21 10:35:49.513458: step: 36/526, loss: 0.05661306530237198 2023-01-21 10:35:50.664594: step: 40/526, loss: 0.04953975975513458 2023-01-21 10:35:51.813595: step: 44/526, loss: 0.03000359609723091 2023-01-21 10:35:52.968861: step: 48/526, loss: 0.03891410678625107 2023-01-21 10:35:54.172075: step: 52/526, loss: 0.033743537962436676 2023-01-21 10:35:55.311183: step: 56/526, loss: 0.007989120669662952 2023-01-21 10:35:56.479525: step: 60/526, loss: 0.010918332263827324 2023-01-21 10:35:57.614068: step: 64/526, loss: 0.09751138836145401 2023-01-21 10:35:58.754856: step: 68/526, loss: 0.02198515087366104 2023-01-21 10:35:59.918372: step: 72/526, loss: 0.01547012384980917 2023-01-21 10:36:01.057252: step: 76/526, loss: 0.08279826492071152 2023-01-21 10:36:02.206265: step: 80/526, loss: 0.16883696615695953 2023-01-21 10:36:03.341115: step: 84/526, loss: 0.0665230080485344 2023-01-21 10:36:04.493113: step: 88/526, loss: 0.023172665387392044 2023-01-21 10:36:05.683462: step: 92/526, loss: 0.003418541047722101 2023-01-21 10:36:06.824423: step: 96/526, loss: 0.03295326232910156 2023-01-21 10:36:07.963413: step: 100/526, loss: 0.15467782318592072 2023-01-21 10:36:09.118452: step: 104/526, loss: 0.04071483761072159 2023-01-21 10:36:10.262463: step: 108/526, loss: 0.023386195302009583 2023-01-21 10:36:11.413401: step: 112/526, loss: 0.07622185349464417 2023-01-21 10:36:12.538522: step: 116/526, loss: 0.04514904320240021 2023-01-21 10:36:13.700662: step: 120/526, loss: 0.013074207119643688 2023-01-21 10:36:14.856014: step: 124/526, loss: 0.0023229599464684725 2023-01-21 10:36:16.060351: step: 128/526, loss: 0.015420341864228249 2023-01-21 10:36:17.220408: step: 132/526, loss: 0.041366007179021835 2023-01-21 10:36:18.370837: step: 136/526, loss: 0.06983594596385956 2023-01-21 10:36:19.509204: step: 140/526, loss: 0.06665734946727753 2023-01-21 10:36:20.651354: step: 144/526, loss: 0.0364743210375309 2023-01-21 10:36:21.807274: step: 148/526, loss: 0.07817334681749344 2023-01-21 10:36:22.955493: step: 152/526, loss: 0.06048621982336044 2023-01-21 10:36:24.112958: step: 156/526, loss: 0.008032036013901234 2023-01-21 10:36:25.266214: step: 160/526, loss: 0.10645437985658646 2023-01-21 10:36:26.442923: step: 164/526, loss: 0.017642784863710403 2023-01-21 10:36:27.570809: step: 168/526, loss: 0.010627923533320427 2023-01-21 10:36:28.731544: step: 172/526, loss: 0.015745734795928 2023-01-21 10:36:29.881232: step: 176/526, loss: 0.4386845529079437 2023-01-21 10:36:31.038766: step: 180/526, loss: 0.02267928048968315 2023-01-21 10:36:32.194675: step: 184/526, loss: 0.05037736892700195 2023-01-21 10:36:33.327471: step: 188/526, loss: 0.01201620139181614 2023-01-21 10:36:34.504363: step: 192/526, loss: 0.06117381900548935 2023-01-21 10:36:35.662555: step: 196/526, loss: 0.022325992584228516 2023-01-21 10:36:36.837437: step: 200/526, loss: 0.07804547250270844 2023-01-21 10:36:37.998388: step: 204/526, loss: 0.05018806457519531 2023-01-21 10:36:39.142019: step: 208/526, loss: 0.0376712828874588 2023-01-21 10:36:40.336264: step: 212/526, loss: 0.08962497860193253 2023-01-21 10:36:41.464650: step: 216/526, loss: 0.0306883342564106 2023-01-21 10:36:42.623807: step: 220/526, loss: 0.13745450973510742 2023-01-21 10:36:43.778365: step: 224/526, loss: 0.016959190368652344 2023-01-21 10:36:44.911601: step: 228/526, loss: 0.04527008533477783 2023-01-21 10:36:46.132838: step: 232/526, loss: 0.1183050125837326 2023-01-21 10:36:47.330184: step: 236/526, loss: 0.10537528991699219 2023-01-21 10:36:48.515605: step: 240/526, loss: 0.023282814770936966 2023-01-21 10:36:49.673664: step: 244/526, loss: 0.0017177105182781816 2023-01-21 10:36:50.818770: step: 248/526, loss: 0.07080788165330887 2023-01-21 10:36:51.959921: step: 252/526, loss: 0.006410121917724609 2023-01-21 10:36:53.110217: step: 256/526, loss: 0.07890644669532776 2023-01-21 10:36:54.227919: step: 260/526, loss: 0.0656648650765419 2023-01-21 10:36:55.398002: step: 264/526, loss: 0.014365863986313343 2023-01-21 10:36:56.621061: step: 268/526, loss: 0.4121030867099762 2023-01-21 10:36:57.773844: step: 272/526, loss: 0.011065102182328701 2023-01-21 10:36:58.949143: step: 276/526, loss: 0.07069685310125351 2023-01-21 10:37:00.099540: step: 280/526, loss: 2.1570065021514893 2023-01-21 10:37:01.246446: step: 284/526, loss: 0.08994360268115997 2023-01-21 10:37:02.386541: step: 288/526, loss: 0.05215740203857422 2023-01-21 10:37:03.495085: step: 292/526, loss: 0.00024046898761298507 2023-01-21 10:37:04.649813: step: 296/526, loss: 0.09027834236621857 2023-01-21 10:37:05.847669: step: 300/526, loss: 0.08672408759593964 2023-01-21 10:37:06.992145: step: 304/526, loss: 0.09557781368494034 2023-01-21 10:37:08.157516: step: 308/526, loss: 0.03354682773351669 2023-01-21 10:37:09.303693: step: 312/526, loss: 0.006915283389389515 2023-01-21 10:37:10.473099: step: 316/526, loss: 0.050173282623291016 2023-01-21 10:37:11.619128: step: 320/526, loss: 0.022202301770448685 2023-01-21 10:37:12.771920: step: 324/526, loss: 0.027315808460116386 2023-01-21 10:37:13.911805: step: 328/526, loss: 0.04655585438013077 2023-01-21 10:37:15.054488: step: 332/526, loss: 0.003204727079719305 2023-01-21 10:37:16.238838: step: 336/526, loss: 0.05240964889526367 2023-01-21 10:37:17.387146: step: 340/526, loss: 0.027200888842344284 2023-01-21 10:37:18.552512: step: 344/526, loss: 0.04323005676269531 2023-01-21 10:37:19.683123: step: 348/526, loss: 0.011305427178740501 2023-01-21 10:37:20.821625: step: 352/526, loss: 0.0011354446178302169 2023-01-21 10:37:21.999098: step: 356/526, loss: 0.03729061037302017 2023-01-21 10:37:23.151781: step: 360/526, loss: 0.0022614479530602694 2023-01-21 10:37:24.293025: step: 364/526, loss: 0.04951276630163193 2023-01-21 10:37:25.468016: step: 368/526, loss: 0.005857181269675493 2023-01-21 10:37:26.630073: step: 372/526, loss: 0.028737831860780716 2023-01-21 10:37:27.763169: step: 376/526, loss: 0.12956352531909943 2023-01-21 10:37:28.931297: step: 380/526, loss: 0.2627403140068054 2023-01-21 10:37:30.093311: step: 384/526, loss: 0.05890006944537163 2023-01-21 10:37:31.244214: step: 388/526, loss: 0.001965141389518976 2023-01-21 10:37:32.379353: step: 392/526, loss: 0.14423684775829315 2023-01-21 10:37:33.508468: step: 396/526, loss: 0.038520436733961105 2023-01-21 10:37:34.688122: step: 400/526, loss: 0.14634761214256287 2023-01-21 10:37:35.846443: step: 404/526, loss: 0.060933783650398254 2023-01-21 10:37:37.047870: step: 408/526, loss: 0.12302875518798828 2023-01-21 10:37:38.191700: step: 412/526, loss: 0.01497945748269558 2023-01-21 10:37:39.407680: step: 416/526, loss: 0.06960439682006836 2023-01-21 10:37:40.550667: step: 420/526, loss: 0.16288337111473083 2023-01-21 10:37:41.707356: step: 424/526, loss: 0.0877465307712555 2023-01-21 10:37:42.913182: step: 428/526, loss: 0.0016601562965661287 2023-01-21 10:37:44.093582: step: 432/526, loss: 0.02856769599020481 2023-01-21 10:37:45.259507: step: 436/526, loss: 0.45875853300094604 2023-01-21 10:37:46.408889: step: 440/526, loss: 0.04623117670416832 2023-01-21 10:37:47.577864: step: 444/526, loss: 0.08636818081140518 2023-01-21 10:37:48.714827: step: 448/526, loss: 0.17690840363502502 2023-01-21 10:37:49.868684: step: 452/526, loss: 0.04018135368824005 2023-01-21 10:37:51.014057: step: 456/526, loss: 0.0006117820739746094 2023-01-21 10:37:52.201550: step: 460/526, loss: 0.044963836669921875 2023-01-21 10:37:53.373369: step: 464/526, loss: 0.04150800779461861 2023-01-21 10:37:54.508134: step: 468/526, loss: 0.029713820666074753 2023-01-21 10:37:55.671397: step: 472/526, loss: 0.08301468193531036 2023-01-21 10:37:56.781000: step: 476/526, loss: 0.06878957897424698 2023-01-21 10:37:57.937438: step: 480/526, loss: 0.002543926239013672 2023-01-21 10:37:59.075343: step: 484/526, loss: 0.020890522748231888 2023-01-21 10:38:00.243839: step: 488/526, loss: 0.006577491760253906 2023-01-21 10:38:01.483412: step: 492/526, loss: 0.028504230082035065 2023-01-21 10:38:02.656986: step: 496/526, loss: 0.11988388001918793 2023-01-21 10:38:03.826003: step: 500/526, loss: 0.04588508605957031 2023-01-21 10:38:04.982901: step: 504/526, loss: 0.06446514278650284 2023-01-21 10:38:06.148410: step: 508/526, loss: 0.20085124671459198 2023-01-21 10:38:07.317780: step: 512/526, loss: 0.14871521294116974 2023-01-21 10:38:08.451542: step: 516/526, loss: 0.016603756695985794 2023-01-21 10:38:09.628042: step: 520/526, loss: 0.017780590802431107 2023-01-21 10:38:10.749643: step: 524/526, loss: 0.029329968616366386 2023-01-21 10:38:11.875884: step: 528/526, loss: 0.06928665935993195 2023-01-21 10:38:13.011222: step: 532/526, loss: 0.1715715527534485 2023-01-21 10:38:14.147465: step: 536/526, loss: 0.012115097604691982 2023-01-21 10:38:15.286186: step: 540/526, loss: 0.021280813962221146 2023-01-21 10:38:16.410509: step: 544/526, loss: 0.4144185185432434 2023-01-21 10:38:17.547731: step: 548/526, loss: 0.1649942547082901 2023-01-21 10:38:18.701457: step: 552/526, loss: 0.040679000318050385 2023-01-21 10:38:19.874615: step: 556/526, loss: 0.01966571807861328 2023-01-21 10:38:21.017190: step: 560/526, loss: 0.0019033432472497225 2023-01-21 10:38:22.190602: step: 564/526, loss: 0.2586946487426758 2023-01-21 10:38:23.365182: step: 568/526, loss: 0.07293100655078888 2023-01-21 10:38:24.516429: step: 572/526, loss: 0.09041118621826172 2023-01-21 10:38:25.687886: step: 576/526, loss: 0.0097503662109375 2023-01-21 10:38:26.839923: step: 580/526, loss: 0.05258836969733238 2023-01-21 10:38:28.022758: step: 584/526, loss: 0.02617006190121174 2023-01-21 10:38:29.186422: step: 588/526, loss: 0.0029286383651196957 2023-01-21 10:38:30.359686: step: 592/526, loss: 0.1657407581806183 2023-01-21 10:38:31.495568: step: 596/526, loss: 0.022541143000125885 2023-01-21 10:38:32.656105: step: 600/526, loss: 0.16706332564353943 2023-01-21 10:38:33.772252: step: 604/526, loss: 0.004824066534638405 2023-01-21 10:38:34.893509: step: 608/526, loss: 0.9176253080368042 2023-01-21 10:38:36.093204: step: 612/526, loss: 0.008787775412201881 2023-01-21 10:38:37.245441: step: 616/526, loss: 0.02135448530316353 2023-01-21 10:38:38.407705: step: 620/526, loss: 0.037813473492860794 2023-01-21 10:38:39.559275: step: 624/526, loss: 0.019387150183320045 2023-01-21 10:38:40.714468: step: 628/526, loss: 0.011272287927567959 2023-01-21 10:38:41.897144: step: 632/526, loss: 0.026995087042450905 2023-01-21 10:38:43.084339: step: 636/526, loss: 0.014538098126649857 2023-01-21 10:38:44.264319: step: 640/526, loss: 0.07078027725219727 2023-01-21 10:38:45.414343: step: 644/526, loss: 0.09060049057006836 2023-01-21 10:38:46.590498: step: 648/526, loss: 0.0013401031028479338 2023-01-21 10:38:47.760591: step: 652/526, loss: 0.06468410789966583 2023-01-21 10:38:48.932832: step: 656/526, loss: 0.09370870888233185 2023-01-21 10:38:50.099193: step: 660/526, loss: 0.09635104984045029 2023-01-21 10:38:51.241680: step: 664/526, loss: 0.15643177926540375 2023-01-21 10:38:52.402486: step: 668/526, loss: 0.07567481696605682 2023-01-21 10:38:53.567838: step: 672/526, loss: 0.0414765365421772 2023-01-21 10:38:54.711245: step: 676/526, loss: 0.06897668540477753 2023-01-21 10:38:55.837091: step: 680/526, loss: 0.13982850313186646 2023-01-21 10:38:56.992516: step: 684/526, loss: 0.0678325816988945 2023-01-21 10:38:58.145802: step: 688/526, loss: 0.1256331503391266 2023-01-21 10:38:59.298120: step: 692/526, loss: 0.09280691295862198 2023-01-21 10:39:00.462312: step: 696/526, loss: 0.08952607959508896 2023-01-21 10:39:01.648700: step: 700/526, loss: 0.04745883867144585 2023-01-21 10:39:02.834032: step: 704/526, loss: 0.02794818766415119 2023-01-21 10:39:03.991946: step: 708/526, loss: 0.01044159010052681 2023-01-21 10:39:05.156505: step: 712/526, loss: 0.024486351758241653 2023-01-21 10:39:06.323710: step: 716/526, loss: 0.04224910959601402 2023-01-21 10:39:07.493303: step: 720/526, loss: 0.01577301137149334 2023-01-21 10:39:08.680807: step: 724/526, loss: 0.030944060534238815 2023-01-21 10:39:09.786738: step: 728/526, loss: 0.04078331217169762 2023-01-21 10:39:10.967990: step: 732/526, loss: 0.06733899563550949 2023-01-21 10:39:12.130719: step: 736/526, loss: 0.03284921869635582 2023-01-21 10:39:13.316469: step: 740/526, loss: 0.011911321431398392 2023-01-21 10:39:14.463522: step: 744/526, loss: 0.03274650499224663 2023-01-21 10:39:15.646735: step: 748/526, loss: 0.05940263345837593 2023-01-21 10:39:16.811984: step: 752/526, loss: 0.12280355393886566 2023-01-21 10:39:17.971469: step: 756/526, loss: 0.0732354149222374 2023-01-21 10:39:19.136364: step: 760/526, loss: 0.022115040570497513 2023-01-21 10:39:20.292877: step: 764/526, loss: 0.014854812994599342 2023-01-21 10:39:21.433543: step: 768/526, loss: 0.006291211117058992 2023-01-21 10:39:22.575156: step: 772/526, loss: 0.033112525939941406 2023-01-21 10:39:23.713932: step: 776/526, loss: 0.06358539313077927 2023-01-21 10:39:24.875934: step: 780/526, loss: 0.0706118568778038 2023-01-21 10:39:26.055006: step: 784/526, loss: 0.06204243004322052 2023-01-21 10:39:27.207425: step: 788/526, loss: 0.054195500910282135 2023-01-21 10:39:28.347228: step: 792/526, loss: 0.01781935803592205 2023-01-21 10:39:29.468162: step: 796/526, loss: 0.007649135310202837 2023-01-21 10:39:30.653072: step: 800/526, loss: 0.02671995386481285 2023-01-21 10:39:31.798179: step: 804/526, loss: 0.06302700191736221 2023-01-21 10:39:32.936350: step: 808/526, loss: 0.05171575769782066 2023-01-21 10:39:34.105571: step: 812/526, loss: 0.07220802456140518 2023-01-21 10:39:35.280244: step: 816/526, loss: 0.058527376502752304 2023-01-21 10:39:36.439448: step: 820/526, loss: 0.021082116290926933 2023-01-21 10:39:37.608200: step: 824/526, loss: 0.1577732115983963 2023-01-21 10:39:38.788290: step: 828/526, loss: 0.04468078911304474 2023-01-21 10:39:39.921825: step: 832/526, loss: 0.10990381240844727 2023-01-21 10:39:41.047283: step: 836/526, loss: 0.01732635498046875 2023-01-21 10:39:42.237230: step: 840/526, loss: 0.032190799713134766 2023-01-21 10:39:43.404819: step: 844/526, loss: 0.03010735660791397 2023-01-21 10:39:44.540909: step: 848/526, loss: 0.061919644474983215 2023-01-21 10:39:45.693190: step: 852/526, loss: 0.009528684429824352 2023-01-21 10:39:46.866292: step: 856/526, loss: 0.013329267501831055 2023-01-21 10:39:48.007433: step: 860/526, loss: 0.08001852035522461 2023-01-21 10:39:49.155125: step: 864/526, loss: 0.0638311430811882 2023-01-21 10:39:50.318466: step: 868/526, loss: 0.018859058618545532 2023-01-21 10:39:51.486498: step: 872/526, loss: 0.08367881178855896 2023-01-21 10:39:52.646663: step: 876/526, loss: 0.0007014751899987459 2023-01-21 10:39:53.840150: step: 880/526, loss: 0.06270580738782883 2023-01-21 10:39:54.988718: step: 884/526, loss: 0.028011512011289597 2023-01-21 10:39:56.165693: step: 888/526, loss: 0.0805811882019043 2023-01-21 10:39:57.312036: step: 892/526, loss: 0.12155047059059143 2023-01-21 10:39:58.452709: step: 896/526, loss: 0.14108867943286896 2023-01-21 10:39:59.649832: step: 900/526, loss: 0.01929459534585476 2023-01-21 10:40:00.812031: step: 904/526, loss: 0.01830129697918892 2023-01-21 10:40:02.010387: step: 908/526, loss: 0.05306129530072212 2023-01-21 10:40:03.168892: step: 912/526, loss: 0.09303779155015945 2023-01-21 10:40:04.342612: step: 916/526, loss: 0.10501112788915634 2023-01-21 10:40:05.517504: step: 920/526, loss: 0.04940962791442871 2023-01-21 10:40:06.712633: step: 924/526, loss: 0.3828137218952179 2023-01-21 10:40:07.864961: step: 928/526, loss: 0.0023636340629309416 2023-01-21 10:40:09.006795: step: 932/526, loss: 0.02647123485803604 2023-01-21 10:40:10.184942: step: 936/526, loss: 0.09867896884679794 2023-01-21 10:40:11.319640: step: 940/526, loss: 0.046065427362918854 2023-01-21 10:40:12.570327: step: 944/526, loss: 0.03982894495129585 2023-01-21 10:40:13.734743: step: 948/526, loss: 0.045298196375370026 2023-01-21 10:40:14.872524: step: 952/526, loss: 0.03884153068065643 2023-01-21 10:40:16.032448: step: 956/526, loss: 0.03743448108434677 2023-01-21 10:40:17.154410: step: 960/526, loss: 0.046800754964351654 2023-01-21 10:40:18.306913: step: 964/526, loss: 0.005704021546989679 2023-01-21 10:40:19.448509: step: 968/526, loss: 0.01610422134399414 2023-01-21 10:40:20.598253: step: 972/526, loss: 0.06591711193323135 2023-01-21 10:40:21.770809: step: 976/526, loss: 0.026584720239043236 2023-01-21 10:40:22.949198: step: 980/526, loss: 0.103411965072155 2023-01-21 10:40:24.101663: step: 984/526, loss: 0.1741001158952713 2023-01-21 10:40:25.277048: step: 988/526, loss: 0.0450650230050087 2023-01-21 10:40:26.463521: step: 992/526, loss: 0.005143642891198397 2023-01-21 10:40:27.621812: step: 996/526, loss: 0.015406466089189053 2023-01-21 10:40:28.776348: step: 1000/526, loss: 0.012519837357103825 2023-01-21 10:40:29.906793: step: 1004/526, loss: 0.035891368985176086 2023-01-21 10:40:31.068444: step: 1008/526, loss: 0.08953256905078888 2023-01-21 10:40:32.243766: step: 1012/526, loss: 0.03525133058428764 2023-01-21 10:40:33.402057: step: 1016/526, loss: 0.0743812620639801 2023-01-21 10:40:34.584924: step: 1020/526, loss: 0.26910096406936646 2023-01-21 10:40:35.738152: step: 1024/526, loss: 0.0531248115003109 2023-01-21 10:40:36.907557: step: 1028/526, loss: 0.042627908289432526 2023-01-21 10:40:38.060665: step: 1032/526, loss: 0.08183574676513672 2023-01-21 10:40:39.234663: step: 1036/526, loss: 0.01733860932290554 2023-01-21 10:40:40.416996: step: 1040/526, loss: 0.05517435073852539 2023-01-21 10:40:41.588442: step: 1044/526, loss: 0.12523861229419708 2023-01-21 10:40:42.787660: step: 1048/526, loss: 0.16399379074573517 2023-01-21 10:40:43.913666: step: 1052/526, loss: 0.01727323606610298 2023-01-21 10:40:45.057586: step: 1056/526, loss: 0.40482062101364136 2023-01-21 10:40:46.216996: step: 1060/526, loss: 0.03465118259191513 2023-01-21 10:40:47.391057: step: 1064/526, loss: 0.09335631877183914 2023-01-21 10:40:48.537401: step: 1068/526, loss: 0.009600735269486904 2023-01-21 10:40:49.719436: step: 1072/526, loss: 0.03583049774169922 2023-01-21 10:40:50.917380: step: 1076/526, loss: 0.02284879796206951 2023-01-21 10:40:52.066049: step: 1080/526, loss: 0.01699678972363472 2023-01-21 10:40:53.232554: step: 1084/526, loss: 0.07446169853210449 2023-01-21 10:40:54.393216: step: 1088/526, loss: 0.04209384694695473 2023-01-21 10:40:55.528315: step: 1092/526, loss: 0.026517678052186966 2023-01-21 10:40:56.663107: step: 1096/526, loss: 0.006968879606574774 2023-01-21 10:40:57.798363: step: 1100/526, loss: 0.04068265110254288 2023-01-21 10:40:58.957850: step: 1104/526, loss: 0.22661571204662323 2023-01-21 10:41:00.121303: step: 1108/526, loss: 0.025042343884706497 2023-01-21 10:41:01.295578: step: 1112/526, loss: 0.011298942379653454 2023-01-21 10:41:02.479708: step: 1116/526, loss: 0.0077888015657663345 2023-01-21 10:41:03.666808: step: 1120/526, loss: 0.05329904705286026 2023-01-21 10:41:04.811676: step: 1124/526, loss: 0.029503442347049713 2023-01-21 10:41:05.972040: step: 1128/526, loss: 0.022494984790682793 2023-01-21 10:41:07.141400: step: 1132/526, loss: 0.11945438385009766 2023-01-21 10:41:08.335450: step: 1136/526, loss: 0.027576161548495293 2023-01-21 10:41:09.471757: step: 1140/526, loss: 0.012827587313950062 2023-01-21 10:41:10.610473: step: 1144/526, loss: 0.03182496875524521 2023-01-21 10:41:11.763711: step: 1148/526, loss: 0.05577068775892258 2023-01-21 10:41:12.937234: step: 1152/526, loss: 0.010259103029966354 2023-01-21 10:41:14.118970: step: 1156/526, loss: 0.05718441307544708 2023-01-21 10:41:15.293310: step: 1160/526, loss: 0.14575956761837006 2023-01-21 10:41:16.431960: step: 1164/526, loss: 0.014760971069335938 2023-01-21 10:41:17.587198: step: 1168/526, loss: 0.1333659142255783 2023-01-21 10:41:18.762100: step: 1172/526, loss: 0.006091308780014515 2023-01-21 10:41:19.922124: step: 1176/526, loss: 0.07075687497854233 2023-01-21 10:41:21.065258: step: 1180/526, loss: 0.029253197833895683 2023-01-21 10:41:22.230675: step: 1184/526, loss: 0.02695169672369957 2023-01-21 10:41:23.373161: step: 1188/526, loss: 0.023044109344482422 2023-01-21 10:41:24.546160: step: 1192/526, loss: 0.08442975580692291 2023-01-21 10:41:25.683672: step: 1196/526, loss: 0.02315702475607395 2023-01-21 10:41:26.884642: step: 1200/526, loss: 0.3992482125759125 2023-01-21 10:41:28.052506: step: 1204/526, loss: 0.06844520568847656 2023-01-21 10:41:29.204103: step: 1208/526, loss: 0.017757605761289597 2023-01-21 10:41:30.379991: step: 1212/526, loss: 0.016434574499726295 2023-01-21 10:41:31.544096: step: 1216/526, loss: 0.05191526561975479 2023-01-21 10:41:32.697936: step: 1220/526, loss: 0.040917206555604935 2023-01-21 10:41:33.867601: step: 1224/526, loss: 0.04365057870745659 2023-01-21 10:41:35.016173: step: 1228/526, loss: 0.03348054736852646 2023-01-21 10:41:36.174820: step: 1232/526, loss: 0.003781127743422985 2023-01-21 10:41:37.329010: step: 1236/526, loss: 0.022966481745243073 2023-01-21 10:41:38.495950: step: 1240/526, loss: 0.23435230553150177 2023-01-21 10:41:39.663929: step: 1244/526, loss: 0.07387170940637589 2023-01-21 10:41:40.802237: step: 1248/526, loss: 0.04220600426197052 2023-01-21 10:41:41.964707: step: 1252/526, loss: 0.22918277978897095 2023-01-21 10:41:43.125831: step: 1256/526, loss: 0.2728847563266754 2023-01-21 10:41:44.283859: step: 1260/526, loss: 0.1085447371006012 2023-01-21 10:41:45.487672: step: 1264/526, loss: 0.27389073371887207 2023-01-21 10:41:46.669917: step: 1268/526, loss: 0.0417238250374794 2023-01-21 10:41:47.858949: step: 1272/526, loss: 0.05303382873535156 2023-01-21 10:41:49.011077: step: 1276/526, loss: 0.030916860327124596 2023-01-21 10:41:50.142423: step: 1280/526, loss: 0.029001297429203987 2023-01-21 10:41:51.298356: step: 1284/526, loss: 0.07766552269458771 2023-01-21 10:41:52.491088: step: 1288/526, loss: 0.09346818923950195 2023-01-21 10:41:53.655383: step: 1292/526, loss: 0.0011556625831872225 2023-01-21 10:41:54.805349: step: 1296/526, loss: 0.08032999187707901 2023-01-21 10:41:55.940863: step: 1300/526, loss: 0.03483548015356064 2023-01-21 10:41:57.142894: step: 1304/526, loss: 0.04926605522632599 2023-01-21 10:41:58.295812: step: 1308/526, loss: 0.04237718507647514 2023-01-21 10:41:59.410498: step: 1312/526, loss: 0.015924930572509766 2023-01-21 10:42:00.552437: step: 1316/526, loss: 0.0515320785343647 2023-01-21 10:42:01.708112: step: 1320/526, loss: 0.017804985865950584 2023-01-21 10:42:02.855681: step: 1324/526, loss: 0.0012434959644451737 2023-01-21 10:42:04.007796: step: 1328/526, loss: 0.02672438696026802 2023-01-21 10:42:05.163365: step: 1332/526, loss: 0.05814165994524956 2023-01-21 10:42:06.289481: step: 1336/526, loss: 0.001913738320581615 2023-01-21 10:42:07.441310: step: 1340/526, loss: 0.008213615976274014 2023-01-21 10:42:08.579224: step: 1344/526, loss: 0.042535971850156784 2023-01-21 10:42:09.736839: step: 1348/526, loss: 0.3972126841545105 2023-01-21 10:42:10.892679: step: 1352/526, loss: 0.007982779294252396 2023-01-21 10:42:12.038125: step: 1356/526, loss: 0.023943137377500534 2023-01-21 10:42:13.191744: step: 1360/526, loss: 0.04347553476691246 2023-01-21 10:42:14.352858: step: 1364/526, loss: 0.0876404345035553 2023-01-21 10:42:15.546111: step: 1368/526, loss: 0.09897013008594513 2023-01-21 10:42:16.693658: step: 1372/526, loss: 0.03507370874285698 2023-01-21 10:42:17.853412: step: 1376/526, loss: 0.021694118157029152 2023-01-21 10:42:19.059767: step: 1380/526, loss: 0.008884811773896217 2023-01-21 10:42:20.202730: step: 1384/526, loss: 0.009718275628983974 2023-01-21 10:42:21.347066: step: 1388/526, loss: 0.033117104321718216 2023-01-21 10:42:22.521535: step: 1392/526, loss: 0.23833851516246796 2023-01-21 10:42:23.661610: step: 1396/526, loss: 0.022200584411621094 2023-01-21 10:42:24.868388: step: 1400/526, loss: 0.07133479416370392 2023-01-21 10:42:26.032376: step: 1404/526, loss: 0.021805191412568092 2023-01-21 10:42:27.163867: step: 1408/526, loss: 0.10773412883281708 2023-01-21 10:42:28.309959: step: 1412/526, loss: 0.020034123212099075 2023-01-21 10:42:29.462823: step: 1416/526, loss: 0.06970882415771484 2023-01-21 10:42:30.634552: step: 1420/526, loss: 0.012594223953783512 2023-01-21 10:42:31.768670: step: 1424/526, loss: 0.07655420899391174 2023-01-21 10:42:32.937164: step: 1428/526, loss: 0.01114511489868164 2023-01-21 10:42:34.090615: step: 1432/526, loss: 0.1356087177991867 2023-01-21 10:42:35.228407: step: 1436/526, loss: 0.004479885101318359 2023-01-21 10:42:36.385392: step: 1440/526, loss: 0.1724068820476532 2023-01-21 10:42:37.534975: step: 1444/526, loss: 0.061400797218084335 2023-01-21 10:42:38.676227: step: 1448/526, loss: 0.05605662986636162 2023-01-21 10:42:39.840319: step: 1452/526, loss: 0.003729534102603793 2023-01-21 10:42:41.031599: step: 1456/526, loss: 0.16875191032886505 2023-01-21 10:42:42.207802: step: 1460/526, loss: 0.04922027885913849 2023-01-21 10:42:43.375281: step: 1464/526, loss: 0.07462406903505325 2023-01-21 10:42:44.540956: step: 1468/526, loss: 0.026093482971191406 2023-01-21 10:42:45.696407: step: 1472/526, loss: 0.04679737240076065 2023-01-21 10:42:46.888737: step: 1476/526, loss: 0.05691566318273544 2023-01-21 10:42:48.045189: step: 1480/526, loss: 0.04170060157775879 2023-01-21 10:42:49.196805: step: 1484/526, loss: 0.03397498279809952 2023-01-21 10:42:50.311545: step: 1488/526, loss: 0.031627655029296875 2023-01-21 10:42:51.446208: step: 1492/526, loss: 0.020652294158935547 2023-01-21 10:42:52.591869: step: 1496/526, loss: 0.04655003547668457 2023-01-21 10:42:53.766108: step: 1500/526, loss: 0.09661178290843964 2023-01-21 10:42:54.913504: step: 1504/526, loss: 0.7788635492324829 2023-01-21 10:42:56.080702: step: 1508/526, loss: 0.054302215576171875 2023-01-21 10:42:57.257618: step: 1512/526, loss: 0.03610391542315483 2023-01-21 10:42:58.414306: step: 1516/526, loss: 0.04607558250427246 2023-01-21 10:42:59.594955: step: 1520/526, loss: 0.016023922711610794 2023-01-21 10:43:00.741261: step: 1524/526, loss: 0.005801391787827015 2023-01-21 10:43:01.896834: step: 1528/526, loss: 0.06605930626392365 2023-01-21 10:43:03.060450: step: 1532/526, loss: 0.015055418945848942 2023-01-21 10:43:04.223427: step: 1536/526, loss: 0.023659851402044296 2023-01-21 10:43:05.396957: step: 1540/526, loss: 0.023528099060058594 2023-01-21 10:43:06.536273: step: 1544/526, loss: 0.03215980902314186 2023-01-21 10:43:07.671955: step: 1548/526, loss: 0.02749185636639595 2023-01-21 10:43:08.827138: step: 1552/526, loss: 0.13465410470962524 2023-01-21 10:43:09.969508: step: 1556/526, loss: 0.05170431360602379 2023-01-21 10:43:11.159782: step: 1560/526, loss: 0.1625603288412094 2023-01-21 10:43:12.314113: step: 1564/526, loss: 0.045168355107307434 2023-01-21 10:43:13.474438: step: 1568/526, loss: 0.008575725369155407 2023-01-21 10:43:14.639068: step: 1572/526, loss: 0.0715770274400711 2023-01-21 10:43:15.777073: step: 1576/526, loss: 0.1396162509918213 2023-01-21 10:43:16.937567: step: 1580/526, loss: 0.05045466870069504 2023-01-21 10:43:18.065396: step: 1584/526, loss: 0.03575272485613823 2023-01-21 10:43:19.208412: step: 1588/526, loss: 0.01935606077313423 2023-01-21 10:43:20.346956: step: 1592/526, loss: 0.019881058484315872 2023-01-21 10:43:21.520971: step: 1596/526, loss: 0.042699433863162994 2023-01-21 10:43:22.677745: step: 1600/526, loss: 0.02517561987042427 2023-01-21 10:43:23.811540: step: 1604/526, loss: 0.010599327273666859 2023-01-21 10:43:25.000580: step: 1608/526, loss: 0.05540146678686142 2023-01-21 10:43:26.170510: step: 1612/526, loss: 0.09423694014549255 2023-01-21 10:43:27.305303: step: 1616/526, loss: 0.46620291471481323 2023-01-21 10:43:28.462833: step: 1620/526, loss: 0.035639550536870956 2023-01-21 10:43:29.616178: step: 1624/526, loss: 0.07630367577075958 2023-01-21 10:43:30.783053: step: 1628/526, loss: 0.14377766847610474 2023-01-21 10:43:31.921681: step: 1632/526, loss: 0.01673140563070774 2023-01-21 10:43:33.090114: step: 1636/526, loss: 0.029838846996426582 2023-01-21 10:43:34.251592: step: 1640/526, loss: 0.01385793648660183 2023-01-21 10:43:35.386992: step: 1644/526, loss: 0.027337361127138138 2023-01-21 10:43:36.561803: step: 1648/526, loss: 0.008957100100815296 2023-01-21 10:43:37.699335: step: 1652/526, loss: 0.027769137173891068 2023-01-21 10:43:38.880732: step: 1656/526, loss: 0.020801641047000885 2023-01-21 10:43:40.020852: step: 1660/526, loss: 0.0037563801743090153 2023-01-21 10:43:41.177709: step: 1664/526, loss: 0.10962390899658203 2023-01-21 10:43:42.362250: step: 1668/526, loss: 0.018858812749385834 2023-01-21 10:43:43.505067: step: 1672/526, loss: 0.04797947406768799 2023-01-21 10:43:44.663920: step: 1676/526, loss: 0.10421662032604218 2023-01-21 10:43:45.808025: step: 1680/526, loss: 0.02319207228720188 2023-01-21 10:43:46.942924: step: 1684/526, loss: 0.0046520233154296875 2023-01-21 10:43:48.100526: step: 1688/526, loss: 0.015748977661132812 2023-01-21 10:43:49.267774: step: 1692/526, loss: 0.0009660244104452431 2023-01-21 10:43:50.438267: step: 1696/526, loss: 0.46494781970977783 2023-01-21 10:43:51.592186: step: 1700/526, loss: 0.038410186767578125 2023-01-21 10:43:52.695930: step: 1704/526, loss: 0.0030962468590587378 2023-01-21 10:43:53.829763: step: 1708/526, loss: 0.13460254669189453 2023-01-21 10:43:54.950827: step: 1712/526, loss: 0.29013940691947937 2023-01-21 10:43:56.083148: step: 1716/526, loss: 0.0002465248107910156 2023-01-21 10:43:57.191089: step: 1720/526, loss: 0.00477290153503418 2023-01-21 10:43:58.389681: step: 1724/526, loss: 0.11452846229076385 2023-01-21 10:43:59.529658: step: 1728/526, loss: 0.18054704368114471 2023-01-21 10:44:00.698141: step: 1732/526, loss: 0.028684616088867188 2023-01-21 10:44:01.833358: step: 1736/526, loss: 0.08340243995189667 2023-01-21 10:44:02.993763: step: 1740/526, loss: 0.0382143035531044 2023-01-21 10:44:04.175918: step: 1744/526, loss: 0.03238735347986221 2023-01-21 10:44:05.349444: step: 1748/526, loss: 0.06665096431970596 2023-01-21 10:44:06.506220: step: 1752/526, loss: 0.08878107368946075 2023-01-21 10:44:07.676647: step: 1756/526, loss: 0.07616205513477325 2023-01-21 10:44:08.809418: step: 1760/526, loss: 0.07322315871715546 2023-01-21 10:44:09.952764: step: 1764/526, loss: 0.33517712354660034 2023-01-21 10:44:11.129805: step: 1768/526, loss: 0.018207263201475143 2023-01-21 10:44:12.264131: step: 1772/526, loss: 0.048223115503787994 2023-01-21 10:44:13.451458: step: 1776/526, loss: 0.007552933879196644 2023-01-21 10:44:14.622395: step: 1780/526, loss: 0.03580122068524361 2023-01-21 10:44:15.769660: step: 1784/526, loss: 0.08817722648382187 2023-01-21 10:44:16.888336: step: 1788/526, loss: 0.007825946435332298 2023-01-21 10:44:18.040453: step: 1792/526, loss: 0.6622940897941589 2023-01-21 10:44:19.207839: step: 1796/526, loss: 0.11396083980798721 2023-01-21 10:44:20.385133: step: 1800/526, loss: 0.0024312972091138363 2023-01-21 10:44:21.531538: step: 1804/526, loss: 0.03350124508142471 2023-01-21 10:44:22.674081: step: 1808/526, loss: 0.013551807031035423 2023-01-21 10:44:23.830712: step: 1812/526, loss: 0.058525752276182175 2023-01-21 10:44:24.965259: step: 1816/526, loss: 0.0364532470703125 2023-01-21 10:44:26.180462: step: 1820/526, loss: 0.013156699016690254 2023-01-21 10:44:27.319993: step: 1824/526, loss: 0.039473772048950195 2023-01-21 10:44:28.482433: step: 1828/526, loss: 0.030416011810302734 2023-01-21 10:44:29.626234: step: 1832/526, loss: 0.011693526059389114 2023-01-21 10:44:30.780735: step: 1836/526, loss: 0.057546332478523254 2023-01-21 10:44:31.957391: step: 1840/526, loss: 0.09516707062721252 2023-01-21 10:44:33.102063: step: 1844/526, loss: 0.04046466201543808 2023-01-21 10:44:34.249938: step: 1848/526, loss: 0.02912139892578125 2023-01-21 10:44:35.387654: step: 1852/526, loss: 0.0180098544806242 2023-01-21 10:44:36.555584: step: 1856/526, loss: 0.029911089688539505 2023-01-21 10:44:37.724959: step: 1860/526, loss: 0.0234573595225811 2023-01-21 10:44:38.868227: step: 1864/526, loss: 0.008993816561996937 2023-01-21 10:44:40.009861: step: 1868/526, loss: 0.027631284669041634 2023-01-21 10:44:41.168165: step: 1872/526, loss: 0.023629046976566315 2023-01-21 10:44:42.395580: step: 1876/526, loss: 0.02846803516149521 2023-01-21 10:44:43.557602: step: 1880/526, loss: 0.04903569445014 2023-01-21 10:44:44.712092: step: 1884/526, loss: 0.06552944332361221 2023-01-21 10:44:45.869629: step: 1888/526, loss: 0.013401031494140625 2023-01-21 10:44:47.017584: step: 1892/526, loss: 0.05478954315185547 2023-01-21 10:44:48.153241: step: 1896/526, loss: 0.11679296940565109 2023-01-21 10:44:49.330771: step: 1900/526, loss: 0.04600093513727188 2023-01-21 10:44:50.519142: step: 1904/526, loss: 0.007589483633637428 2023-01-21 10:44:51.676548: step: 1908/526, loss: 0.41737622022628784 2023-01-21 10:44:52.826930: step: 1912/526, loss: 0.031861305236816406 2023-01-21 10:44:53.977410: step: 1916/526, loss: 0.05798816680908203 2023-01-21 10:44:55.153942: step: 1920/526, loss: 0.07336101680994034 2023-01-21 10:44:56.343256: step: 1924/526, loss: 0.15314950048923492 2023-01-21 10:44:57.508128: step: 1928/526, loss: 0.002681732177734375 2023-01-21 10:44:58.685557: step: 1932/526, loss: 0.4707014262676239 2023-01-21 10:44:59.836649: step: 1936/526, loss: 0.03202028200030327 2023-01-21 10:45:00.965256: step: 1940/526, loss: 0.004622364416718483 2023-01-21 10:45:02.109102: step: 1944/526, loss: 0.0360906608402729 2023-01-21 10:45:03.263015: step: 1948/526, loss: 0.017039014026522636 2023-01-21 10:45:04.397094: step: 1952/526, loss: 0.003299760865047574 2023-01-21 10:45:05.540595: step: 1956/526, loss: 0.0649019256234169 2023-01-21 10:45:06.680827: step: 1960/526, loss: 0.11012105643749237 2023-01-21 10:45:07.815460: step: 1964/526, loss: 0.11162881553173065 2023-01-21 10:45:08.968685: step: 1968/526, loss: 0.0005999326822347939 2023-01-21 10:45:10.125680: step: 1972/526, loss: 0.009122943505644798 2023-01-21 10:45:11.258084: step: 1976/526, loss: 0.11528149247169495 2023-01-21 10:45:12.392023: step: 1980/526, loss: 0.09085922688245773 2023-01-21 10:45:13.582503: step: 1984/526, loss: 0.006344890221953392 2023-01-21 10:45:14.739289: step: 1988/526, loss: 0.04867715761065483 2023-01-21 10:45:15.898264: step: 1992/526, loss: 0.0987405776977539 2023-01-21 10:45:17.068166: step: 1996/526, loss: 0.05556058883666992 2023-01-21 10:45:18.207940: step: 2000/526, loss: 0.01620776578783989 2023-01-21 10:45:19.366659: step: 2004/526, loss: 0.03907070308923721 2023-01-21 10:45:20.551971: step: 2008/526, loss: 0.014750957489013672 2023-01-21 10:45:21.700069: step: 2012/526, loss: 0.07001190632581711 2023-01-21 10:45:22.824406: step: 2016/526, loss: 0.5143219232559204 2023-01-21 10:45:23.953115: step: 2020/526, loss: 0.010564994998276234 2023-01-21 10:45:25.091455: step: 2024/526, loss: 0.10760851204395294 2023-01-21 10:45:26.222930: step: 2028/526, loss: 0.07874422520399094 2023-01-21 10:45:27.378149: step: 2032/526, loss: 0.06146583706140518 2023-01-21 10:45:28.521054: step: 2036/526, loss: 0.05511474609375 2023-01-21 10:45:29.663083: step: 2040/526, loss: 0.09309883415699005 2023-01-21 10:45:30.804297: step: 2044/526, loss: 0.043385982513427734 2023-01-21 10:45:32.019333: step: 2048/526, loss: 0.06844620406627655 2023-01-21 10:45:33.194304: step: 2052/526, loss: 0.046548936516046524 2023-01-21 10:45:34.315421: step: 2056/526, loss: 0.016779804602265358 2023-01-21 10:45:35.437163: step: 2060/526, loss: 0.07168009132146835 2023-01-21 10:45:36.570270: step: 2064/526, loss: 0.029972266405820847 2023-01-21 10:45:37.698574: step: 2068/526, loss: 0.04488839954137802 2023-01-21 10:45:38.862888: step: 2072/526, loss: 0.16236476600170135 2023-01-21 10:45:40.014004: step: 2076/526, loss: 0.2640199661254883 2023-01-21 10:45:41.151140: step: 2080/526, loss: 0.014705944806337357 2023-01-21 10:45:42.319459: step: 2084/526, loss: 0.13029327988624573 2023-01-21 10:45:43.471379: step: 2088/526, loss: 0.41745373606681824 2023-01-21 10:45:44.610800: step: 2092/526, loss: 0.04197096824645996 2023-01-21 10:45:45.755534: step: 2096/526, loss: 0.007121086586266756 2023-01-21 10:45:46.916386: step: 2100/526, loss: 0.03950660303235054 2023-01-21 10:45:48.070970: step: 2104/526, loss: 0.27300509810447693 ================================================== Loss: 0.075 -------------------- Dev: {'event': {'p': 0.6263498920086393, 'r': 0.7723035952063915, 'f1': 0.6917113893858079}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Test: {'event': {'p': 0.6342208642514185, 'r': 0.7583507306889353, 'f1': 0.6907535060613264}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Chinese: {'event': {'p': 0.5714285714285714, 'r': 0.8888888888888888, 'f1': 0.6956521739130435}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Korean: {'event': {'p': 0.7380952380952381, 'r': 0.49206349206349204, 'f1': 0.5904761904761904}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Russian: {'event': {'p': 0.5, 'r': 0.5277777777777778, 'f1': 0.5135135135135136}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6263498920086393, 'r': 0.7723035952063915, 'f1': 0.6917113893858079}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Eng Test for Chinese: {'event': {'p': 0.6342208642514185, 'r': 0.7583507306889353, 'f1': 0.6907535060613264}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Sample Chinese: {'event': {'p': 0.5714285714285714, 'r': 0.8888888888888888, 'f1': 0.6956521739130435}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 12 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:46:41.323962: step: 4/526, loss: 0.005014276597648859 2023-01-21 10:46:42.516094: step: 8/526, loss: 0.15837565064430237 2023-01-21 10:46:43.701684: step: 12/526, loss: 0.02331562340259552 2023-01-21 10:46:44.846937: step: 16/526, loss: 0.05583696439862251 2023-01-21 10:46:46.024985: step: 20/526, loss: 0.02319946326315403 2023-01-21 10:46:47.170348: step: 24/526, loss: 0.07766366004943848 2023-01-21 10:46:48.291787: step: 28/526, loss: 0.007442283909767866 2023-01-21 10:46:49.434547: step: 32/526, loss: 0.11003002524375916 2023-01-21 10:46:50.556409: step: 36/526, loss: 0.0012899399735033512 2023-01-21 10:46:51.709322: step: 40/526, loss: 0.05518393591046333 2023-01-21 10:46:52.857192: step: 44/526, loss: 0.2991431951522827 2023-01-21 10:46:54.015866: step: 48/526, loss: 0.006704807281494141 2023-01-21 10:46:55.174898: step: 52/526, loss: 0.035034749656915665 2023-01-21 10:46:56.313684: step: 56/526, loss: 0.05994434654712677 2023-01-21 10:46:57.474528: step: 60/526, loss: 0.33890390396118164 2023-01-21 10:46:58.694126: step: 64/526, loss: 0.03724546357989311 2023-01-21 10:46:59.839252: step: 68/526, loss: 0.005110550206154585 2023-01-21 10:47:01.022097: step: 72/526, loss: 0.0189864169806242 2023-01-21 10:47:02.228734: step: 76/526, loss: 0.051998041570186615 2023-01-21 10:47:03.396663: step: 80/526, loss: 0.010298728942871094 2023-01-21 10:47:04.539700: step: 84/526, loss: 0.04920339584350586 2023-01-21 10:47:05.676899: step: 88/526, loss: 0.09373245388269424 2023-01-21 10:47:06.814545: step: 92/526, loss: 0.197038933634758 2023-01-21 10:47:07.980456: step: 96/526, loss: 0.07460059970617294 2023-01-21 10:47:09.140256: step: 100/526, loss: 0.018487263470888138 2023-01-21 10:47:10.320644: step: 104/526, loss: 0.19398756325244904 2023-01-21 10:47:11.505380: step: 108/526, loss: 0.0467960387468338 2023-01-21 10:47:12.661995: step: 112/526, loss: 0.03556842729449272 2023-01-21 10:47:13.897354: step: 116/526, loss: 0.07704783976078033 2023-01-21 10:47:15.030833: step: 120/526, loss: 0.033930111676454544 2023-01-21 10:47:16.154600: step: 124/526, loss: 0.09686551988124847 2023-01-21 10:47:17.316899: step: 128/526, loss: 0.08251826465129852 2023-01-21 10:47:18.463293: step: 132/526, loss: 0.22340141236782074 2023-01-21 10:47:19.633947: step: 136/526, loss: 0.23547811806201935 2023-01-21 10:47:20.792827: step: 140/526, loss: 0.16128763556480408 2023-01-21 10:47:21.944631: step: 144/526, loss: 0.01851787604391575 2023-01-21 10:47:23.090940: step: 148/526, loss: 0.0385284423828125 2023-01-21 10:47:24.261012: step: 152/526, loss: 0.06302586197853088 2023-01-21 10:47:25.401440: step: 156/526, loss: 0.007805442903190851 2023-01-21 10:47:26.547229: step: 160/526, loss: 0.0001601219264557585 2023-01-21 10:47:27.706787: step: 164/526, loss: 0.004538869950920343 2023-01-21 10:47:28.845771: step: 168/526, loss: 0.006539535708725452 2023-01-21 10:47:30.002457: step: 172/526, loss: 0.013204623013734818 2023-01-21 10:47:31.196795: step: 176/526, loss: 0.07103576511144638 2023-01-21 10:47:32.356951: step: 180/526, loss: 0.026652049273252487 2023-01-21 10:47:33.527779: step: 184/526, loss: 0.02602977864444256 2023-01-21 10:47:34.664361: step: 188/526, loss: 0.0026180269196629524 2023-01-21 10:47:35.862627: step: 192/526, loss: 0.05306531488895416 2023-01-21 10:47:37.006317: step: 196/526, loss: 0.0069921016693115234 2023-01-21 10:47:38.186958: step: 200/526, loss: 0.011813545599579811 2023-01-21 10:47:39.315243: step: 204/526, loss: 0.024253463372588158 2023-01-21 10:47:40.513414: step: 208/526, loss: 0.008948802947998047 2023-01-21 10:47:41.658330: step: 212/526, loss: 0.015288067981600761 2023-01-21 10:47:42.788822: step: 216/526, loss: 0.002563786692917347 2023-01-21 10:47:43.955513: step: 220/526, loss: 0.04061717912554741 2023-01-21 10:47:45.094633: step: 224/526, loss: 0.11216068267822266 2023-01-21 10:47:46.277754: step: 228/526, loss: 0.04761169105768204 2023-01-21 10:47:47.421889: step: 232/526, loss: 0.01874055713415146 2023-01-21 10:47:48.558286: step: 236/526, loss: 0.009713363833725452 2023-01-21 10:47:49.760864: step: 240/526, loss: 0.017380524426698685 2023-01-21 10:47:50.940205: step: 244/526, loss: 0.050794124603271484 2023-01-21 10:47:52.106275: step: 248/526, loss: 0.013465499505400658 2023-01-21 10:47:53.257225: step: 252/526, loss: 0.02010054513812065 2023-01-21 10:47:54.420330: step: 256/526, loss: 0.013569450005888939 2023-01-21 10:47:55.627654: step: 260/526, loss: 0.006478500552475452 2023-01-21 10:47:56.809018: step: 264/526, loss: 0.04996204376220703 2023-01-21 10:47:57.963816: step: 268/526, loss: 0.035019125789403915 2023-01-21 10:47:59.110572: step: 272/526, loss: 0.01232242677360773 2023-01-21 10:48:00.267402: step: 276/526, loss: 0.038022611290216446 2023-01-21 10:48:01.419835: step: 280/526, loss: 0.0251601692289114 2023-01-21 10:48:02.608036: step: 284/526, loss: 0.03603691980242729 2023-01-21 10:48:03.732252: step: 288/526, loss: 0.027198027819395065 2023-01-21 10:48:04.938610: step: 292/526, loss: 0.030568504706025124 2023-01-21 10:48:06.081709: step: 296/526, loss: 0.004569435492157936 2023-01-21 10:48:07.242948: step: 300/526, loss: 0.19565612077713013 2023-01-21 10:48:08.382916: step: 304/526, loss: 0.07910709083080292 2023-01-21 10:48:09.537469: step: 308/526, loss: 0.001977908657863736 2023-01-21 10:48:10.695864: step: 312/526, loss: 0.013912105932831764 2023-01-21 10:48:11.888227: step: 316/526, loss: 0.004699897952377796 2023-01-21 10:48:13.090540: step: 320/526, loss: 0.75470370054245 2023-01-21 10:48:14.243118: step: 324/526, loss: 0.06167927011847496 2023-01-21 10:48:15.422360: step: 328/526, loss: 0.06235914304852486 2023-01-21 10:48:16.574724: step: 332/526, loss: 0.08310098946094513 2023-01-21 10:48:17.702001: step: 336/526, loss: 0.01947193220257759 2023-01-21 10:48:18.835209: step: 340/526, loss: 0.04777107387781143 2023-01-21 10:48:20.019550: step: 344/526, loss: 0.053322505205869675 2023-01-21 10:48:21.205468: step: 348/526, loss: 0.06441116333007812 2023-01-21 10:48:22.344555: step: 352/526, loss: 0.0547216422855854 2023-01-21 10:48:23.541974: step: 356/526, loss: 0.09208846837282181 2023-01-21 10:48:24.704536: step: 360/526, loss: 0.03144874423742294 2023-01-21 10:48:25.863322: step: 364/526, loss: 0.09040851145982742 2023-01-21 10:48:27.009746: step: 368/526, loss: 0.3013055920600891 2023-01-21 10:48:28.160181: step: 372/526, loss: 0.049841977655887604 2023-01-21 10:48:29.277260: step: 376/526, loss: 0.04145689308643341 2023-01-21 10:48:30.420948: step: 380/526, loss: 0.055289652198553085 2023-01-21 10:48:31.565550: step: 384/526, loss: 0.05458955839276314 2023-01-21 10:48:32.730871: step: 388/526, loss: 0.006760788150131702 2023-01-21 10:48:33.876318: step: 392/526, loss: 0.015175485983490944 2023-01-21 10:48:35.028262: step: 396/526, loss: 0.043103788048028946 2023-01-21 10:48:36.162646: step: 400/526, loss: 0.05019550397992134 2023-01-21 10:48:37.307347: step: 404/526, loss: 0.023662948980927467 2023-01-21 10:48:38.451018: step: 408/526, loss: 0.02124500274658203 2023-01-21 10:48:39.605896: step: 412/526, loss: 0.06670932471752167 2023-01-21 10:48:40.748103: step: 416/526, loss: 0.05777731165289879 2023-01-21 10:48:41.932050: step: 420/526, loss: 0.03680391237139702 2023-01-21 10:48:43.118126: step: 424/526, loss: 0.047613002359867096 2023-01-21 10:48:44.273996: step: 428/526, loss: 0.7896060943603516 2023-01-21 10:48:45.419605: step: 432/526, loss: 0.023007284849882126 2023-01-21 10:48:46.552692: step: 436/526, loss: 0.11981189250946045 2023-01-21 10:48:47.710701: step: 440/526, loss: 0.03748941794037819 2023-01-21 10:48:48.891455: step: 444/526, loss: 0.011210059747099876 2023-01-21 10:48:50.047358: step: 448/526, loss: 0.0694499984383583 2023-01-21 10:48:51.176730: step: 452/526, loss: 0.08625636249780655 2023-01-21 10:48:52.352401: step: 456/526, loss: 0.1802116334438324 2023-01-21 10:48:53.523574: step: 460/526, loss: 0.035994913429021835 2023-01-21 10:48:54.671384: step: 464/526, loss: 0.07664966583251953 2023-01-21 10:48:55.828089: step: 468/526, loss: 0.06726408004760742 2023-01-21 10:48:56.969680: step: 472/526, loss: 0.10308542102575302 2023-01-21 10:48:58.107356: step: 476/526, loss: 0.02859029732644558 2023-01-21 10:48:59.271975: step: 480/526, loss: 0.013994693756103516 2023-01-21 10:49:00.477345: step: 484/526, loss: 0.03370704501867294 2023-01-21 10:49:01.627098: step: 488/526, loss: 0.03453044965863228 2023-01-21 10:49:02.772116: step: 492/526, loss: 0.0007882118225097656 2023-01-21 10:49:03.915222: step: 496/526, loss: 0.3698081970214844 2023-01-21 10:49:05.053467: step: 500/526, loss: 0.8205415606498718 2023-01-21 10:49:06.199501: step: 504/526, loss: 0.09431672096252441 2023-01-21 10:49:07.352099: step: 508/526, loss: 0.039953041821718216 2023-01-21 10:49:08.498457: step: 512/526, loss: 0.03703467920422554 2023-01-21 10:49:09.665387: step: 516/526, loss: 0.1137368232011795 2023-01-21 10:49:10.801697: step: 520/526, loss: 0.005227852147072554 2023-01-21 10:49:11.959234: step: 524/526, loss: 0.025608062744140625 2023-01-21 10:49:13.139895: step: 528/526, loss: 0.10578260570764542 2023-01-21 10:49:14.290246: step: 532/526, loss: 0.020121192559599876 2023-01-21 10:49:15.455315: step: 536/526, loss: 0.0334138385951519 2023-01-21 10:49:16.576952: step: 540/526, loss: 0.021785831078886986 2023-01-21 10:49:17.762071: step: 544/526, loss: 0.05236692726612091 2023-01-21 10:49:18.914901: step: 548/526, loss: 0.1314988136291504 2023-01-21 10:49:20.080032: step: 552/526, loss: 0.02799406088888645 2023-01-21 10:49:21.257191: step: 556/526, loss: 0.1003204807639122 2023-01-21 10:49:22.374918: step: 560/526, loss: 0.012031269259750843 2023-01-21 10:49:23.535102: step: 564/526, loss: 0.021151352673768997 2023-01-21 10:49:24.686512: step: 568/526, loss: 0.0068832398392260075 2023-01-21 10:49:25.828532: step: 572/526, loss: 0.016821861267089844 2023-01-21 10:49:26.996559: step: 576/526, loss: 0.018536090850830078 2023-01-21 10:49:28.169955: step: 580/526, loss: 0.01825409010052681 2023-01-21 10:49:29.357011: step: 584/526, loss: 0.003875351045280695 2023-01-21 10:49:30.543992: step: 588/526, loss: 0.00866842269897461 2023-01-21 10:49:31.728387: step: 592/526, loss: 0.01886777952313423 2023-01-21 10:49:32.867230: step: 596/526, loss: 0.009160804562270641 2023-01-21 10:49:34.043508: step: 600/526, loss: 0.006759452633559704 2023-01-21 10:49:35.224697: step: 604/526, loss: 0.006973934359848499 2023-01-21 10:49:36.420802: step: 608/526, loss: 0.05599198490381241 2023-01-21 10:49:37.568283: step: 612/526, loss: 0.031445980072021484 2023-01-21 10:49:38.711150: step: 616/526, loss: 0.013846492394804955 2023-01-21 10:49:39.828023: step: 620/526, loss: 0.08624868094921112 2023-01-21 10:49:40.969665: step: 624/526, loss: 0.01663370244204998 2023-01-21 10:49:42.135424: step: 628/526, loss: 0.027376916259527206 2023-01-21 10:49:43.303915: step: 632/526, loss: 0.004254770465195179 2023-01-21 10:49:44.493201: step: 636/526, loss: 0.03201127052307129 2023-01-21 10:49:45.624868: step: 640/526, loss: 0.013113665394484997 2023-01-21 10:49:46.763432: step: 644/526, loss: 0.16804885864257812 2023-01-21 10:49:47.979915: step: 648/526, loss: 0.0326998233795166 2023-01-21 10:49:49.146443: step: 652/526, loss: 0.020496368408203125 2023-01-21 10:49:50.299737: step: 656/526, loss: 0.008899211883544922 2023-01-21 10:49:51.464602: step: 660/526, loss: 0.013555800542235374 2023-01-21 10:49:52.674218: step: 664/526, loss: 0.06982650607824326 2023-01-21 10:49:53.826059: step: 668/526, loss: 0.14360694587230682 2023-01-21 10:49:54.999020: step: 672/526, loss: 0.020160485059022903 2023-01-21 10:49:56.170352: step: 676/526, loss: 0.03500986471772194 2023-01-21 10:49:57.306188: step: 680/526, loss: 0.18179956078529358 2023-01-21 10:49:58.506348: step: 684/526, loss: 0.015084696002304554 2023-01-21 10:49:59.692355: step: 688/526, loss: 0.018367290496826172 2023-01-21 10:50:00.859905: step: 692/526, loss: 0.021160507574677467 2023-01-21 10:50:02.032272: step: 696/526, loss: 0.020832061767578125 2023-01-21 10:50:03.215273: step: 700/526, loss: 0.02121915854513645 2023-01-21 10:50:04.363068: step: 704/526, loss: 0.03353748470544815 2023-01-21 10:50:05.499558: step: 708/526, loss: 0.012795353308320045 2023-01-21 10:50:06.626550: step: 712/526, loss: 0.05167360603809357 2023-01-21 10:50:07.754089: step: 716/526, loss: 0.0785159170627594 2023-01-21 10:50:08.953438: step: 720/526, loss: 0.14270934462547302 2023-01-21 10:50:10.103453: step: 724/526, loss: 0.017456531524658203 2023-01-21 10:50:11.238936: step: 728/526, loss: 0.04156322404742241 2023-01-21 10:50:12.397410: step: 732/526, loss: 0.019110679626464844 2023-01-21 10:50:13.537562: step: 736/526, loss: 0.13657379150390625 2023-01-21 10:50:14.687623: step: 740/526, loss: 0.028533553704619408 2023-01-21 10:50:15.837086: step: 744/526, loss: 0.04280510172247887 2023-01-21 10:50:16.984441: step: 748/526, loss: 0.014842415228486061 2023-01-21 10:50:18.136357: step: 752/526, loss: 0.0020779133774340153 2023-01-21 10:50:19.265899: step: 756/526, loss: 0.007045364938676357 2023-01-21 10:50:20.468179: step: 760/526, loss: 0.04721980169415474 2023-01-21 10:50:21.656475: step: 764/526, loss: 0.02477102167904377 2023-01-21 10:50:22.807476: step: 768/526, loss: 0.0012703895336017013 2023-01-21 10:50:23.944060: step: 772/526, loss: 0.031102323904633522 2023-01-21 10:50:25.103847: step: 776/526, loss: 0.04388155788183212 2023-01-21 10:50:26.265867: step: 780/526, loss: 0.009329509921371937 2023-01-21 10:50:27.400580: step: 784/526, loss: 0.022286999970674515 2023-01-21 10:50:28.514817: step: 788/526, loss: 0.022002220153808594 2023-01-21 10:50:29.654279: step: 792/526, loss: 0.021439289674162865 2023-01-21 10:50:30.823063: step: 796/526, loss: 0.09964265674352646 2023-01-21 10:50:31.989788: step: 800/526, loss: 0.00927190762013197 2023-01-21 10:50:33.138072: step: 804/526, loss: 0.026840783655643463 2023-01-21 10:50:34.306489: step: 808/526, loss: 0.05136584863066673 2023-01-21 10:50:35.463312: step: 812/526, loss: 0.02593545988202095 2023-01-21 10:50:36.669683: step: 816/526, loss: 0.024433420971035957 2023-01-21 10:50:37.817421: step: 820/526, loss: 0.02726602554321289 2023-01-21 10:50:38.957406: step: 824/526, loss: 0.006706523708999157 2023-01-21 10:50:40.123463: step: 828/526, loss: 0.00850820541381836 2023-01-21 10:50:41.260083: step: 832/526, loss: 0.01353550050407648 2023-01-21 10:50:42.402547: step: 836/526, loss: 0.0828624740242958 2023-01-21 10:50:43.544687: step: 840/526, loss: 0.0027029276825487614 2023-01-21 10:50:44.709640: step: 844/526, loss: 0.02274780347943306 2023-01-21 10:50:45.895148: step: 848/526, loss: 0.03883986547589302 2023-01-21 10:50:47.032232: step: 852/526, loss: 0.020419906824827194 2023-01-21 10:50:48.181713: step: 856/526, loss: 0.007386303041130304 2023-01-21 10:50:49.339121: step: 860/526, loss: 0.3408277630805969 2023-01-21 10:50:50.451952: step: 864/526, loss: 0.0542634017765522 2023-01-21 10:50:51.594401: step: 868/526, loss: 0.0017309188842773438 2023-01-21 10:50:52.737315: step: 872/526, loss: 0.0013473511207848787 2023-01-21 10:50:53.909335: step: 876/526, loss: 0.054708484560251236 2023-01-21 10:50:55.048076: step: 880/526, loss: 0.04383111000061035 2023-01-21 10:50:56.209659: step: 884/526, loss: 0.03033285029232502 2023-01-21 10:50:57.379389: step: 888/526, loss: 0.0433807373046875 2023-01-21 10:50:58.562112: step: 892/526, loss: 0.034485530108213425 2023-01-21 10:50:59.714753: step: 896/526, loss: 0.043548014014959335 2023-01-21 10:51:00.890172: step: 900/526, loss: 0.06304372847080231 2023-01-21 10:51:02.051359: step: 904/526, loss: 0.04075632244348526 2023-01-21 10:51:03.224402: step: 908/526, loss: 0.16977018117904663 2023-01-21 10:51:04.373280: step: 912/526, loss: 0.011780452914536 2023-01-21 10:51:05.529612: step: 916/526, loss: 0.07580890506505966 2023-01-21 10:51:06.687321: step: 920/526, loss: 0.03230275958776474 2023-01-21 10:51:07.836804: step: 924/526, loss: 0.11890307068824768 2023-01-21 10:51:08.990270: step: 928/526, loss: 0.06739797443151474 2023-01-21 10:51:10.144642: step: 932/526, loss: 0.017836762592196465 2023-01-21 10:51:11.321218: step: 936/526, loss: 0.00805444736033678 2023-01-21 10:51:12.477261: step: 940/526, loss: 0.01569375954568386 2023-01-21 10:51:13.672221: step: 944/526, loss: 0.0011993409134447575 2023-01-21 10:51:14.826763: step: 948/526, loss: 0.019652558490633965 2023-01-21 10:51:15.993305: step: 952/526, loss: 0.06442413479089737 2023-01-21 10:51:17.157244: step: 956/526, loss: 0.18526040017604828 2023-01-21 10:51:18.315703: step: 960/526, loss: 0.009407520294189453 2023-01-21 10:51:19.459561: step: 964/526, loss: 0.025765608996152878 2023-01-21 10:51:20.583603: step: 968/526, loss: 0.06046156957745552 2023-01-21 10:51:21.714738: step: 972/526, loss: 0.04392719268798828 2023-01-21 10:51:22.853794: step: 976/526, loss: 0.015723418444395065 2023-01-21 10:51:24.020734: step: 980/526, loss: 0.06410686671733856 2023-01-21 10:51:25.184348: step: 984/526, loss: 0.07546329498291016 2023-01-21 10:51:26.306423: step: 988/526, loss: 0.0746615082025528 2023-01-21 10:51:27.462733: step: 992/526, loss: 0.04820432886481285 2023-01-21 10:51:28.646436: step: 996/526, loss: 0.00711593683809042 2023-01-21 10:51:29.795132: step: 1000/526, loss: 0.045021869242191315 2023-01-21 10:51:30.965405: step: 1004/526, loss: 0.032410621643066406 2023-01-21 10:51:32.108398: step: 1008/526, loss: 0.03292074054479599 2023-01-21 10:51:33.306243: step: 1012/526, loss: 0.009679603390395641 2023-01-21 10:51:34.466946: step: 1016/526, loss: 0.02282867208123207 2023-01-21 10:51:35.643985: step: 1020/526, loss: 0.06127443164587021 2023-01-21 10:51:36.826226: step: 1024/526, loss: 0.02827281877398491 2023-01-21 10:51:37.965612: step: 1028/526, loss: 0.003234005067497492 2023-01-21 10:51:39.138773: step: 1032/526, loss: 0.031190872192382812 2023-01-21 10:51:40.296535: step: 1036/526, loss: 0.04781933128833771 2023-01-21 10:51:41.439976: step: 1040/526, loss: 0.012742806226015091 2023-01-21 10:51:42.573610: step: 1044/526, loss: 0.0018214226001873612 2023-01-21 10:51:43.763642: step: 1048/526, loss: 0.136883944272995 2023-01-21 10:51:44.915421: step: 1052/526, loss: 0.027689361944794655 2023-01-21 10:51:46.045525: step: 1056/526, loss: 0.4107941687107086 2023-01-21 10:51:47.218682: step: 1060/526, loss: 0.03291730955243111 2023-01-21 10:51:48.399080: step: 1064/526, loss: 0.01733236387372017 2023-01-21 10:51:49.561909: step: 1068/526, loss: 0.006503724958747625 2023-01-21 10:51:50.723230: step: 1072/526, loss: 0.050051119178533554 2023-01-21 10:51:51.885224: step: 1076/526, loss: 0.021442700177431107 2023-01-21 10:51:53.009470: step: 1080/526, loss: 0.032674409449100494 2023-01-21 10:51:54.164336: step: 1084/526, loss: 0.03428492695093155 2023-01-21 10:51:55.311820: step: 1088/526, loss: 0.0006778120878152549 2023-01-21 10:51:56.456486: step: 1092/526, loss: 0.04425087198615074 2023-01-21 10:51:57.590492: step: 1096/526, loss: 0.012255668640136719 2023-01-21 10:51:58.732601: step: 1100/526, loss: 0.05667443573474884 2023-01-21 10:51:59.886620: step: 1104/526, loss: 0.010410880669951439 2023-01-21 10:52:01.036385: step: 1108/526, loss: 0.0020513534545898438 2023-01-21 10:52:02.202724: step: 1112/526, loss: 0.180999755859375 2023-01-21 10:52:03.357701: step: 1116/526, loss: 0.023761559277772903 2023-01-21 10:52:04.506271: step: 1120/526, loss: 0.0240662582218647 2023-01-21 10:52:05.648567: step: 1124/526, loss: 0.01329665258526802 2023-01-21 10:52:06.788582: step: 1128/526, loss: 0.006125736050307751 2023-01-21 10:52:07.939286: step: 1132/526, loss: 0.03783588483929634 2023-01-21 10:52:09.070601: step: 1136/526, loss: 0.04968605190515518 2023-01-21 10:52:10.228968: step: 1140/526, loss: 0.1593664288520813 2023-01-21 10:52:11.383444: step: 1144/526, loss: 0.09417963027954102 2023-01-21 10:52:12.571153: step: 1148/526, loss: 0.026500225067138672 2023-01-21 10:52:13.750254: step: 1152/526, loss: 0.0037673949263989925 2023-01-21 10:52:14.906506: step: 1156/526, loss: 0.0488039031624794 2023-01-21 10:52:16.069176: step: 1160/526, loss: 0.013555717654526234 2023-01-21 10:52:17.245784: step: 1164/526, loss: 0.002695274306461215 2023-01-21 10:52:18.409828: step: 1168/526, loss: 0.016222668811678886 2023-01-21 10:52:19.553596: step: 1172/526, loss: 0.026630688458681107 2023-01-21 10:52:20.736878: step: 1176/526, loss: 0.1773972362279892 2023-01-21 10:52:21.883551: step: 1180/526, loss: 0.021808242425322533 2023-01-21 10:52:23.052690: step: 1184/526, loss: 0.2964572012424469 2023-01-21 10:52:24.223368: step: 1188/526, loss: 0.04487719386816025 2023-01-21 10:52:25.420092: step: 1192/526, loss: 0.00749549875035882 2023-01-21 10:52:26.554375: step: 1196/526, loss: 0.021151162683963776 2023-01-21 10:52:27.730554: step: 1200/526, loss: 0.0334961861371994 2023-01-21 10:52:28.874761: step: 1204/526, loss: 0.03635196387767792 2023-01-21 10:52:29.990469: step: 1208/526, loss: 0.006293904967606068 2023-01-21 10:52:31.156150: step: 1212/526, loss: 0.0027557373978197575 2023-01-21 10:52:32.312332: step: 1216/526, loss: 0.04963350296020508 2023-01-21 10:52:33.452113: step: 1220/526, loss: 0.2429819107055664 2023-01-21 10:52:34.662644: step: 1224/526, loss: 0.2772899866104126 2023-01-21 10:52:35.825078: step: 1228/526, loss: 0.03223705291748047 2023-01-21 10:52:37.005185: step: 1232/526, loss: 0.06376257538795471 2023-01-21 10:52:38.164998: step: 1236/526, loss: 0.04116266965866089 2023-01-21 10:52:39.323588: step: 1240/526, loss: 0.014807652682065964 2023-01-21 10:52:40.481844: step: 1244/526, loss: 0.02443566359579563 2023-01-21 10:52:41.635798: step: 1248/526, loss: 0.005798244383186102 2023-01-21 10:52:42.772279: step: 1252/526, loss: 0.024842072278261185 2023-01-21 10:52:43.908304: step: 1256/526, loss: 0.03334350511431694 2023-01-21 10:52:45.117383: step: 1260/526, loss: 0.0090491296723485 2023-01-21 10:52:46.307443: step: 1264/526, loss: 0.014495945535600185 2023-01-21 10:52:47.437952: step: 1268/526, loss: 0.004066848661750555 2023-01-21 10:52:48.663964: step: 1272/526, loss: 0.08167324215173721 2023-01-21 10:52:49.855899: step: 1276/526, loss: 0.03427610546350479 2023-01-21 10:52:50.997724: step: 1280/526, loss: 0.027289772406220436 2023-01-21 10:52:52.159403: step: 1284/526, loss: 0.003236436750739813 2023-01-21 10:52:53.294511: step: 1288/526, loss: 0.038630008697509766 2023-01-21 10:52:54.465696: step: 1292/526, loss: 0.012416648678481579 2023-01-21 10:52:55.669629: step: 1296/526, loss: 0.07008008658885956 2023-01-21 10:52:56.820919: step: 1300/526, loss: 0.05007324367761612 2023-01-21 10:52:57.980388: step: 1304/526, loss: 0.0747290700674057 2023-01-21 10:52:59.132990: step: 1308/526, loss: 0.03546304628252983 2023-01-21 10:53:00.320908: step: 1312/526, loss: 0.044303130358457565 2023-01-21 10:53:01.492365: step: 1316/526, loss: 0.1232721358537674 2023-01-21 10:53:02.681754: step: 1320/526, loss: 0.027375031262636185 2023-01-21 10:53:03.818260: step: 1324/526, loss: 0.009673118591308594 2023-01-21 10:53:04.966284: step: 1328/526, loss: 0.02348911762237549 2023-01-21 10:53:06.126662: step: 1332/526, loss: 0.07153091579675674 2023-01-21 10:53:07.292862: step: 1336/526, loss: 0.030302954837679863 2023-01-21 10:53:08.464409: step: 1340/526, loss: 0.041434336453676224 2023-01-21 10:53:09.625884: step: 1344/526, loss: 0.020363997668027878 2023-01-21 10:53:10.804136: step: 1348/526, loss: 0.07390232384204865 2023-01-21 10:53:11.979291: step: 1352/526, loss: 0.07209186255931854 2023-01-21 10:53:13.124298: step: 1356/526, loss: 0.004238558001816273 2023-01-21 10:53:14.274670: step: 1360/526, loss: 0.025953199714422226 2023-01-21 10:53:15.414620: step: 1364/526, loss: 0.02716651000082493 2023-01-21 10:53:16.547229: step: 1368/526, loss: 0.05875494331121445 2023-01-21 10:53:17.670302: step: 1372/526, loss: 0.031439781188964844 2023-01-21 10:53:18.821829: step: 1376/526, loss: 0.017386246472597122 2023-01-21 10:53:19.965446: step: 1380/526, loss: 0.008271598257124424 2023-01-21 10:53:21.132095: step: 1384/526, loss: 0.06272315979003906 2023-01-21 10:53:22.292479: step: 1388/526, loss: 0.011740971356630325 2023-01-21 10:53:23.438664: step: 1392/526, loss: 0.034822940826416016 2023-01-21 10:53:24.609796: step: 1396/526, loss: 0.0436285026371479 2023-01-21 10:53:25.759975: step: 1400/526, loss: 0.004963874816894531 2023-01-21 10:53:26.895013: step: 1404/526, loss: 0.053969480097293854 2023-01-21 10:53:28.062200: step: 1408/526, loss: 0.03966865688562393 2023-01-21 10:53:29.190969: step: 1412/526, loss: 0.05612468719482422 2023-01-21 10:53:30.332294: step: 1416/526, loss: 0.011247635819017887 2023-01-21 10:53:31.491073: step: 1420/526, loss: 0.025371933355927467 2023-01-21 10:53:32.626101: step: 1424/526, loss: 0.0461546890437603 2023-01-21 10:53:33.781713: step: 1428/526, loss: 0.003895664354786277 2023-01-21 10:53:34.949593: step: 1432/526, loss: 0.8182815313339233 2023-01-21 10:53:36.077604: step: 1436/526, loss: 0.006576252169907093 2023-01-21 10:53:37.296225: step: 1440/526, loss: 0.023845817893743515 2023-01-21 10:53:38.464851: step: 1444/526, loss: 0.03608150780200958 2023-01-21 10:53:39.615098: step: 1448/526, loss: 0.016951465979218483 2023-01-21 10:53:40.775015: step: 1452/526, loss: 0.0018060803413391113 2023-01-21 10:53:41.938207: step: 1456/526, loss: 0.08336801081895828 2023-01-21 10:53:43.085290: step: 1460/526, loss: 0.02992877922952175 2023-01-21 10:53:44.236045: step: 1464/526, loss: 0.6871873140335083 2023-01-21 10:53:45.398793: step: 1468/526, loss: 0.03498063236474991 2023-01-21 10:53:46.544347: step: 1472/526, loss: 0.06262655556201935 2023-01-21 10:53:47.679246: step: 1476/526, loss: 0.02201104164123535 2023-01-21 10:53:48.801096: step: 1480/526, loss: 0.005126094911247492 2023-01-21 10:53:49.948957: step: 1484/526, loss: 0.02070169523358345 2023-01-21 10:53:51.088305: step: 1488/526, loss: 0.0006317138904705644 2023-01-21 10:53:52.211377: step: 1492/526, loss: 0.00042033198405988514 2023-01-21 10:53:53.363805: step: 1496/526, loss: 0.013733958825469017 2023-01-21 10:53:54.492143: step: 1500/526, loss: 0.01101963222026825 2023-01-21 10:53:55.656834: step: 1504/526, loss: 0.10884075611829758 2023-01-21 10:53:56.798898: step: 1508/526, loss: 0.0911344513297081 2023-01-21 10:53:57.944167: step: 1512/526, loss: 0.14299726486206055 2023-01-21 10:53:59.092844: step: 1516/526, loss: 0.010817051865160465 2023-01-21 10:54:00.244768: step: 1520/526, loss: 0.023146916180849075 2023-01-21 10:54:01.415522: step: 1524/526, loss: 0.02089667320251465 2023-01-21 10:54:02.544259: step: 1528/526, loss: 0.14809398353099823 2023-01-21 10:54:03.667396: step: 1532/526, loss: 0.026880264282226562 2023-01-21 10:54:04.812215: step: 1536/526, loss: 0.05155186727643013 2023-01-21 10:54:06.002818: step: 1540/526, loss: 0.000682830810546875 2023-01-21 10:54:07.137220: step: 1544/526, loss: 0.022702883929014206 2023-01-21 10:54:08.287031: step: 1548/526, loss: 0.047187235206365585 2023-01-21 10:54:09.452137: step: 1552/526, loss: 0.17317704856395721 2023-01-21 10:54:10.607465: step: 1556/526, loss: 0.02258272096514702 2023-01-21 10:54:11.816063: step: 1560/526, loss: 0.026255272328853607 2023-01-21 10:54:12.944519: step: 1564/526, loss: 0.00017632247181609273 2023-01-21 10:54:14.093818: step: 1568/526, loss: 0.23204460740089417 2023-01-21 10:54:15.276830: step: 1572/526, loss: 0.05755005031824112 2023-01-21 10:54:16.415795: step: 1576/526, loss: 0.045473624020814896 2023-01-21 10:54:17.549106: step: 1580/526, loss: 0.002887773560360074 2023-01-21 10:54:18.678380: step: 1584/526, loss: 0.07780475914478302 2023-01-21 10:54:19.810862: step: 1588/526, loss: 0.0057201385498046875 2023-01-21 10:54:20.944774: step: 1592/526, loss: 0.0072193145751953125 2023-01-21 10:54:22.079314: step: 1596/526, loss: 0.02798943594098091 2023-01-21 10:54:23.240613: step: 1600/526, loss: 0.01578812673687935 2023-01-21 10:54:24.412065: step: 1604/526, loss: 0.4075937271118164 2023-01-21 10:54:25.573303: step: 1608/526, loss: 0.04425077512860298 2023-01-21 10:54:26.696467: step: 1612/526, loss: 0.023885346949100494 2023-01-21 10:54:27.845245: step: 1616/526, loss: 0.023654939606785774 2023-01-21 10:54:28.988067: step: 1620/526, loss: 0.08473014831542969 2023-01-21 10:54:30.113083: step: 1624/526, loss: 0.07060708105564117 2023-01-21 10:54:31.278403: step: 1628/526, loss: 0.054797910153865814 2023-01-21 10:54:32.437018: step: 1632/526, loss: 0.2670868933200836 2023-01-21 10:54:33.606701: step: 1636/526, loss: 0.0017000199295580387 2023-01-21 10:54:34.758768: step: 1640/526, loss: 0.038254451006650925 2023-01-21 10:54:35.896249: step: 1644/526, loss: 0.012462044134736061 2023-01-21 10:54:37.045327: step: 1648/526, loss: 0.07951248437166214 2023-01-21 10:54:38.183584: step: 1652/526, loss: 0.5755947828292847 2023-01-21 10:54:39.325985: step: 1656/526, loss: 0.019350241869688034 2023-01-21 10:54:40.457140: step: 1660/526, loss: 0.01771240308880806 2023-01-21 10:54:41.590143: step: 1664/526, loss: 0.10567188262939453 2023-01-21 10:54:42.717749: step: 1668/526, loss: 0.07023916393518448 2023-01-21 10:54:43.879887: step: 1672/526, loss: 0.05703587830066681 2023-01-21 10:54:45.048110: step: 1676/526, loss: 0.027640677988529205 2023-01-21 10:54:46.207645: step: 1680/526, loss: 0.025768471881747246 2023-01-21 10:54:47.350791: step: 1684/526, loss: 0.04692802205681801 2023-01-21 10:54:48.537037: step: 1688/526, loss: 0.05409812927246094 2023-01-21 10:54:49.727640: step: 1692/526, loss: 0.4012279510498047 2023-01-21 10:54:50.874056: step: 1696/526, loss: 0.1030871644616127 2023-01-21 10:54:52.020288: step: 1700/526, loss: 0.04265942797064781 2023-01-21 10:54:53.151743: step: 1704/526, loss: 0.15849293768405914 2023-01-21 10:54:54.305055: step: 1708/526, loss: 0.03170964866876602 2023-01-21 10:54:55.456605: step: 1712/526, loss: 0.02606491930782795 2023-01-21 10:54:56.598298: step: 1716/526, loss: 0.053172290325164795 2023-01-21 10:54:57.746245: step: 1720/526, loss: 0.03164386749267578 2023-01-21 10:54:58.910803: step: 1724/526, loss: 0.04469861835241318 2023-01-21 10:55:00.051388: step: 1728/526, loss: 0.005818176083266735 2023-01-21 10:55:01.199056: step: 1732/526, loss: 0.04537258297204971 2023-01-21 10:55:02.376212: step: 1736/526, loss: 0.04201517254114151 2023-01-21 10:55:03.527161: step: 1740/526, loss: 0.029452133923768997 2023-01-21 10:55:04.668923: step: 1744/526, loss: 0.0060163503512740135 2023-01-21 10:55:05.764162: step: 1748/526, loss: 0.28751277923583984 2023-01-21 10:55:06.933213: step: 1752/526, loss: 0.01087264996021986 2023-01-21 10:55:08.077910: step: 1756/526, loss: 0.048621464520692825 2023-01-21 10:55:09.255132: step: 1760/526, loss: 0.04401111602783203 2023-01-21 10:55:10.390879: step: 1764/526, loss: 0.022840406745672226 2023-01-21 10:55:11.534941: step: 1768/526, loss: 0.01822199858725071 2023-01-21 10:55:12.691631: step: 1772/526, loss: 0.07354164123535156 2023-01-21 10:55:13.850229: step: 1776/526, loss: 0.02441110834479332 2023-01-21 10:55:15.020185: step: 1780/526, loss: 0.019890213385224342 2023-01-21 10:55:16.184984: step: 1784/526, loss: 0.06719360500574112 2023-01-21 10:55:17.321755: step: 1788/526, loss: 0.05928993225097656 2023-01-21 10:55:18.462643: step: 1792/526, loss: 0.4545758366584778 2023-01-21 10:55:19.632432: step: 1796/526, loss: 0.17160768806934357 2023-01-21 10:55:20.799443: step: 1800/526, loss: 0.05804738774895668 2023-01-21 10:55:21.958353: step: 1804/526, loss: 0.04509871453046799 2023-01-21 10:55:23.111579: step: 1808/526, loss: 0.04761238396167755 2023-01-21 10:55:24.235192: step: 1812/526, loss: 0.01605534553527832 2023-01-21 10:55:25.399506: step: 1816/526, loss: 0.01584453694522381 2023-01-21 10:55:26.562098: step: 1820/526, loss: 0.09592743217945099 2023-01-21 10:55:27.692949: step: 1824/526, loss: 0.193315327167511 2023-01-21 10:55:28.847471: step: 1828/526, loss: 0.01300439890474081 2023-01-21 10:55:29.991625: step: 1832/526, loss: 0.00048437720397487283 2023-01-21 10:55:31.155070: step: 1836/526, loss: 0.01170816458761692 2023-01-21 10:55:32.358892: step: 1840/526, loss: 0.028622150421142578 2023-01-21 10:55:33.520780: step: 1844/526, loss: 0.09107961505651474 2023-01-21 10:55:34.659702: step: 1848/526, loss: 0.0006039619911462069 2023-01-21 10:55:35.811289: step: 1852/526, loss: 0.0005326271057128906 2023-01-21 10:55:36.955157: step: 1856/526, loss: 0.0033239363692700863 2023-01-21 10:55:38.138071: step: 1860/526, loss: 0.059676554054021835 2023-01-21 10:55:39.293512: step: 1864/526, loss: 0.0428556427359581 2023-01-21 10:55:40.461474: step: 1868/526, loss: 0.018061447888612747 2023-01-21 10:55:41.603846: step: 1872/526, loss: 0.03526115417480469 2023-01-21 10:55:42.738002: step: 1876/526, loss: 0.06230955570936203 2023-01-21 10:55:43.881567: step: 1880/526, loss: 0.07876358181238174 2023-01-21 10:55:45.024193: step: 1884/526, loss: 0.019382977858185768 2023-01-21 10:55:46.199528: step: 1888/526, loss: 0.0564669631421566 2023-01-21 10:55:47.336254: step: 1892/526, loss: 0.038663100451231 2023-01-21 10:55:48.531039: step: 1896/526, loss: 0.0011962890857830644 2023-01-21 10:55:49.676233: step: 1900/526, loss: 0.005804920569062233 2023-01-21 10:55:50.837764: step: 1904/526, loss: 0.0636088103055954 2023-01-21 10:55:51.985451: step: 1908/526, loss: 0.0152749540284276 2023-01-21 10:55:53.126319: step: 1912/526, loss: 0.007731437683105469 2023-01-21 10:55:54.262383: step: 1916/526, loss: 0.026102924719452858 2023-01-21 10:55:55.422396: step: 1920/526, loss: 0.42654716968536377 2023-01-21 10:55:56.583450: step: 1924/526, loss: 0.041433051228523254 2023-01-21 10:55:57.734424: step: 1928/526, loss: 0.038018226623535156 2023-01-21 10:55:58.871072: step: 1932/526, loss: 0.031040668487548828 2023-01-21 10:56:00.025157: step: 1936/526, loss: 0.011744118295609951 2023-01-21 10:56:01.158081: step: 1940/526, loss: 0.026145363226532936 2023-01-21 10:56:02.309953: step: 1944/526, loss: 0.04430332034826279 2023-01-21 10:56:03.508247: step: 1948/526, loss: 0.03115851990878582 2023-01-21 10:56:04.665557: step: 1952/526, loss: 0.00018043517775367945 2023-01-21 10:56:05.811882: step: 1956/526, loss: 0.018588924780488014 2023-01-21 10:56:06.977643: step: 1960/526, loss: 0.11597137153148651 2023-01-21 10:56:08.120935: step: 1964/526, loss: 0.04968447610735893 2023-01-21 10:56:09.249209: step: 1968/526, loss: 0.01731748692691326 2023-01-21 10:56:10.408528: step: 1972/526, loss: 0.02411813661456108 2023-01-21 10:56:11.554993: step: 1976/526, loss: 0.027556326240301132 2023-01-21 10:56:12.721875: step: 1980/526, loss: 0.04701070860028267 2023-01-21 10:56:13.892797: step: 1984/526, loss: 0.08717431873083115 2023-01-21 10:56:15.069002: step: 1988/526, loss: 0.039710044860839844 2023-01-21 10:56:16.226159: step: 1992/526, loss: 0.08350543677806854 2023-01-21 10:56:17.348924: step: 1996/526, loss: 0.04892873764038086 2023-01-21 10:56:18.518482: step: 2000/526, loss: 0.45091360807418823 2023-01-21 10:56:19.656311: step: 2004/526, loss: 0.029063750058412552 2023-01-21 10:56:20.800394: step: 2008/526, loss: 0.024001501500606537 2023-01-21 10:56:21.973957: step: 2012/526, loss: 0.013182925991714 2023-01-21 10:56:23.114491: step: 2016/526, loss: 0.08830604702234268 2023-01-21 10:56:24.271947: step: 2020/526, loss: 0.052048493176698685 2023-01-21 10:56:25.471411: step: 2024/526, loss: 0.03321409225463867 2023-01-21 10:56:26.638929: step: 2028/526, loss: 0.11997724324464798 2023-01-21 10:56:27.804309: step: 2032/526, loss: 0.15167662501335144 2023-01-21 10:56:28.979965: step: 2036/526, loss: 0.12710018455982208 2023-01-21 10:56:30.141897: step: 2040/526, loss: 0.5929369926452637 2023-01-21 10:56:31.334164: step: 2044/526, loss: 0.04465284198522568 2023-01-21 10:56:32.511057: step: 2048/526, loss: 0.06522063910961151 2023-01-21 10:56:33.654271: step: 2052/526, loss: 0.06809630990028381 2023-01-21 10:56:34.823796: step: 2056/526, loss: 0.0362459197640419 2023-01-21 10:56:35.973945: step: 2060/526, loss: 0.02742924727499485 2023-01-21 10:56:37.130526: step: 2064/526, loss: 0.03204236179590225 2023-01-21 10:56:38.265582: step: 2068/526, loss: 0.11362352222204208 2023-01-21 10:56:39.408081: step: 2072/526, loss: 0.08075514435768127 2023-01-21 10:56:40.548915: step: 2076/526, loss: 0.12044267356395721 2023-01-21 10:56:41.712146: step: 2080/526, loss: 0.03281516954302788 2023-01-21 10:56:42.859855: step: 2084/526, loss: 0.06396923214197159 2023-01-21 10:56:43.993198: step: 2088/526, loss: 0.03920001909136772 2023-01-21 10:56:45.180935: step: 2092/526, loss: 0.14201021194458008 2023-01-21 10:56:46.341824: step: 2096/526, loss: 0.031968019902706146 2023-01-21 10:56:47.493854: step: 2100/526, loss: 0.23675537109375 2023-01-21 10:56:48.658029: step: 2104/526, loss: 0.13140617311000824 ================================================== Loss: 0.062 -------------------- Dev: {'event': {'p': 0.5905349794238683, 'r': 0.7643142476697736, 'f1': 0.6662797446314569}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Test: {'event': {'p': 0.6230870135548754, 'r': 0.7437369519832986, 'f1': 0.6780870806566739}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Chinese: {'event': {'p': 0.5853658536585366, 'r': 0.8888888888888888, 'f1': 0.7058823529411764}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Korean: {'event': {'p': 0.6136363636363636, 'r': 0.42857142857142855, 'f1': 0.5046728971962617}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} Russian: {'event': {'p': 0.4418604651162791, 'r': 0.5277777777777778, 'f1': 0.4810126582278481}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 12} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6263498920086393, 'r': 0.7723035952063915, 'f1': 0.6917113893858079}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Eng Test for Chinese: {'event': {'p': 0.6342208642514185, 'r': 0.7583507306889353, 'f1': 0.6907535060613264}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} Sample Chinese: {'event': {'p': 0.5714285714285714, 'r': 0.8888888888888888, 'f1': 0.6956521739130435}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 11} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 13 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 10:57:33.631146: step: 4/526, loss: 0.003343868302181363 2023-01-21 10:57:34.785604: step: 8/526, loss: 0.04936399683356285 2023-01-21 10:57:35.940263: step: 12/526, loss: 0.007564449217170477 2023-01-21 10:57:37.125036: step: 16/526, loss: 0.01380014419555664 2023-01-21 10:57:38.260521: step: 20/526, loss: 0.052228882908821106 2023-01-21 10:57:39.437637: step: 24/526, loss: 0.049744606018066406 2023-01-21 10:57:40.618786: step: 28/526, loss: 0.06352625042200089 2023-01-21 10:57:41.745648: step: 32/526, loss: 0.031080342829227448 2023-01-21 10:57:42.860174: step: 36/526, loss: 0.0637393444776535 2023-01-21 10:57:44.004203: step: 40/526, loss: 0.03955040127038956 2023-01-21 10:57:45.158231: step: 44/526, loss: 0.018660355359315872 2023-01-21 10:57:46.324004: step: 48/526, loss: 0.03791084140539169 2023-01-21 10:57:47.484678: step: 52/526, loss: 0.4659709930419922 2023-01-21 10:57:48.669041: step: 56/526, loss: 0.02841362915933132 2023-01-21 10:57:49.842141: step: 60/526, loss: 0.02750368043780327 2023-01-21 10:57:50.987304: step: 64/526, loss: 0.004151105880737305 2023-01-21 10:57:52.139987: step: 68/526, loss: 0.019626140594482422 2023-01-21 10:57:53.292015: step: 72/526, loss: 0.026484966278076172 2023-01-21 10:57:54.462406: step: 76/526, loss: 0.09473247826099396 2023-01-21 10:57:55.583860: step: 80/526, loss: 0.11876068264245987 2023-01-21 10:57:56.732979: step: 84/526, loss: 0.05898723751306534 2023-01-21 10:57:57.860183: step: 88/526, loss: 0.02245493046939373 2023-01-21 10:57:59.017124: step: 92/526, loss: 0.0012227654224261642 2023-01-21 10:58:00.159581: step: 96/526, loss: 0.030055999755859375 2023-01-21 10:58:01.308618: step: 100/526, loss: 0.53765869140625 2023-01-21 10:58:02.474008: step: 104/526, loss: 0.1193990707397461 2023-01-21 10:58:03.638299: step: 108/526, loss: 0.6213350296020508 2023-01-21 10:58:04.805357: step: 112/526, loss: 0.03964824602007866 2023-01-21 10:58:05.937979: step: 116/526, loss: 0.04657583311200142 2023-01-21 10:58:07.079511: step: 120/526, loss: 0.017995595932006836 2023-01-21 10:58:08.240779: step: 124/526, loss: 0.1778421401977539 2023-01-21 10:58:09.380096: step: 128/526, loss: 0.12723799049854279 2023-01-21 10:58:10.535293: step: 132/526, loss: 0.004783248994499445 2023-01-21 10:58:11.672616: step: 136/526, loss: 0.07159395515918732 2023-01-21 10:58:12.908742: step: 140/526, loss: 0.04679832607507706 2023-01-21 10:58:14.071697: step: 144/526, loss: 0.043010879307985306 2023-01-21 10:58:15.182647: step: 148/526, loss: 0.04448862001299858 2023-01-21 10:58:16.328792: step: 152/526, loss: 0.11926116794347763 2023-01-21 10:58:17.517907: step: 156/526, loss: 0.4576537609100342 2023-01-21 10:58:18.710223: step: 160/526, loss: 0.0801454484462738 2023-01-21 10:58:19.890460: step: 164/526, loss: 0.08440323173999786 2023-01-21 10:58:21.053945: step: 168/526, loss: 0.04520683363080025 2023-01-21 10:58:22.208316: step: 172/526, loss: 0.010570526123046875 2023-01-21 10:58:23.347924: step: 176/526, loss: 0.005318975541740656 2023-01-21 10:58:24.487065: step: 180/526, loss: 0.012668800540268421 2023-01-21 10:58:25.631576: step: 184/526, loss: 0.040294550359249115 2023-01-21 10:58:26.774443: step: 188/526, loss: 0.08683434128761292 2023-01-21 10:58:27.924518: step: 192/526, loss: 0.012177658267319202 2023-01-21 10:58:29.064460: step: 196/526, loss: 0.00578994769603014 2023-01-21 10:58:30.199999: step: 200/526, loss: 0.013469576835632324 2023-01-21 10:58:31.362908: step: 204/526, loss: 0.00961923599243164 2023-01-21 10:58:32.510649: step: 208/526, loss: 0.0031724930740892887 2023-01-21 10:58:33.654054: step: 212/526, loss: 0.008566808886826038 2023-01-21 10:58:34.823713: step: 216/526, loss: 0.05053577199578285 2023-01-21 10:58:35.952417: step: 220/526, loss: 0.019422434270381927 2023-01-21 10:58:37.122282: step: 224/526, loss: 0.11163187026977539 2023-01-21 10:58:38.265173: step: 228/526, loss: 0.6892479658126831 2023-01-21 10:58:39.397117: step: 232/526, loss: 0.004634285345673561 2023-01-21 10:58:40.539356: step: 236/526, loss: 0.028853796422481537 2023-01-21 10:58:41.678652: step: 240/526, loss: 0.030071640387177467 2023-01-21 10:58:42.858744: step: 244/526, loss: 0.10433006286621094 2023-01-21 10:58:44.004922: step: 248/526, loss: 0.06990686058998108 2023-01-21 10:58:45.190993: step: 252/526, loss: 0.10982856154441833 2023-01-21 10:58:46.366091: step: 256/526, loss: 0.0402885302901268 2023-01-21 10:58:47.502094: step: 260/526, loss: 0.03999929502606392 2023-01-21 10:58:48.648357: step: 264/526, loss: 0.013419246301054955 2023-01-21 10:58:49.791562: step: 268/526, loss: 0.11001081764698029 2023-01-21 10:58:50.966917: step: 272/526, loss: 0.015373516827821732 2023-01-21 10:58:52.108148: step: 276/526, loss: 0.03369751200079918 2023-01-21 10:58:53.227786: step: 280/526, loss: 0.00595436105504632 2023-01-21 10:58:54.369005: step: 284/526, loss: 0.05754728242754936 2023-01-21 10:58:55.523857: step: 288/526, loss: 0.05466146767139435 2023-01-21 10:58:56.701899: step: 292/526, loss: 0.00962753314524889 2023-01-21 10:58:57.847859: step: 296/526, loss: 0.028749655932188034 2023-01-21 10:58:58.977468: step: 300/526, loss: 0.017904235050082207 2023-01-21 10:59:00.121155: step: 304/526, loss: 0.01133575476706028 2023-01-21 10:59:01.280419: step: 308/526, loss: 0.0036680935882031918 2023-01-21 10:59:02.482284: step: 312/526, loss: 0.0059254649095237255 2023-01-21 10:59:03.661206: step: 316/526, loss: 0.0036606788635253906 2023-01-21 10:59:04.803076: step: 320/526, loss: 0.009058046154677868 2023-01-21 10:59:05.934264: step: 324/526, loss: 0.00353584298864007 2023-01-21 10:59:07.082330: step: 328/526, loss: 0.0011196136474609375 2023-01-21 10:59:08.215136: step: 332/526, loss: 0.00653076171875 2023-01-21 10:59:09.343468: step: 336/526, loss: 0.015506076626479626 2023-01-21 10:59:10.511305: step: 340/526, loss: 0.0452730655670166 2023-01-21 10:59:11.637038: step: 344/526, loss: 0.027920866385102272 2023-01-21 10:59:12.791084: step: 348/526, loss: 0.0018477441044524312 2023-01-21 10:59:13.960829: step: 352/526, loss: 0.005360317416489124 2023-01-21 10:59:15.123466: step: 356/526, loss: 0.018458843231201172 2023-01-21 10:59:16.281640: step: 360/526, loss: 0.099718376994133 2023-01-21 10:59:17.436090: step: 364/526, loss: 0.025664901360869408 2023-01-21 10:59:18.593286: step: 368/526, loss: 0.04473366588354111 2023-01-21 10:59:19.780639: step: 372/526, loss: 0.056041717529296875 2023-01-21 10:59:20.958973: step: 376/526, loss: 0.007688617799431086 2023-01-21 10:59:22.130615: step: 380/526, loss: 0.020863723009824753 2023-01-21 10:59:23.271567: step: 384/526, loss: 0.0014172076480463147 2023-01-21 10:59:24.422890: step: 388/526, loss: 0.015442085452377796 2023-01-21 10:59:25.618596: step: 392/526, loss: 0.018088817596435547 2023-01-21 10:59:26.750164: step: 396/526, loss: 0.02909700945019722 2023-01-21 10:59:27.902042: step: 400/526, loss: 0.011687946505844593 2023-01-21 10:59:29.084297: step: 404/526, loss: 0.00018882752920035273 2023-01-21 10:59:30.263654: step: 408/526, loss: 0.00488700857385993 2023-01-21 10:59:31.438506: step: 412/526, loss: 0.057968903332948685 2023-01-21 10:59:32.578601: step: 416/526, loss: 0.005549335852265358 2023-01-21 10:59:33.775014: step: 420/526, loss: 0.012666416354477406 2023-01-21 10:59:34.903696: step: 424/526, loss: 0.0160387996584177 2023-01-21 10:59:36.053929: step: 428/526, loss: 0.04160728678107262 2023-01-21 10:59:37.213446: step: 432/526, loss: 0.08094330132007599 2023-01-21 10:59:38.402115: step: 436/526, loss: 0.027771759778261185 2023-01-21 10:59:39.563168: step: 440/526, loss: 0.07526607811450958 2023-01-21 10:59:40.755616: step: 444/526, loss: 0.05303172767162323 2023-01-21 10:59:41.901672: step: 448/526, loss: 0.13842131197452545 2023-01-21 10:59:43.083356: step: 452/526, loss: 0.0013718606205657125 2023-01-21 10:59:44.222844: step: 456/526, loss: 0.0044023036025464535 2023-01-21 10:59:45.355953: step: 460/526, loss: 0.01291036605834961 2023-01-21 10:59:46.540014: step: 464/526, loss: 0.03337745741009712 2023-01-21 10:59:47.670187: step: 468/526, loss: 0.023488139733672142 2023-01-21 10:59:48.830848: step: 472/526, loss: 0.0369526743888855 2023-01-21 10:59:49.985188: step: 476/526, loss: 0.08896227180957794 2023-01-21 10:59:51.118679: step: 480/526, loss: 0.022980643436312675 2023-01-21 10:59:52.235922: step: 484/526, loss: 0.0011325955856591463 2023-01-21 10:59:53.382015: step: 488/526, loss: 0.24746552109718323 2023-01-21 10:59:54.546231: step: 492/526, loss: 0.03898783028125763 2023-01-21 10:59:55.674150: step: 496/526, loss: 0.04852492734789848 2023-01-21 10:59:56.830651: step: 500/526, loss: 0.027479028329253197 2023-01-21 10:59:57.976109: step: 504/526, loss: 0.02696561999619007 2023-01-21 10:59:59.121706: step: 508/526, loss: 0.05223055183887482 2023-01-21 11:00:00.295451: step: 512/526, loss: 0.08334140479564667 2023-01-21 11:00:01.462035: step: 516/526, loss: 0.05268239974975586 2023-01-21 11:00:02.604245: step: 520/526, loss: 0.11760444194078445 2023-01-21 11:00:03.760970: step: 524/526, loss: 0.13242359459400177 2023-01-21 11:00:04.913081: step: 528/526, loss: 0.008893584832549095 2023-01-21 11:00:06.053627: step: 532/526, loss: 0.027087021619081497 2023-01-21 11:00:07.212586: step: 536/526, loss: 0.02698230743408203 2023-01-21 11:00:08.351836: step: 540/526, loss: 0.04895520210266113 2023-01-21 11:00:09.529825: step: 544/526, loss: 0.09065943211317062 2023-01-21 11:00:10.721161: step: 548/526, loss: 0.03667950630187988 2023-01-21 11:00:11.904809: step: 552/526, loss: 0.4843130111694336 2023-01-21 11:00:13.061221: step: 556/526, loss: 0.010173416696488857 2023-01-21 11:00:14.233632: step: 560/526, loss: 0.009769439697265625 2023-01-21 11:00:15.381160: step: 564/526, loss: 0.009202862158417702 2023-01-21 11:00:16.505548: step: 568/526, loss: 0.06249351426959038 2023-01-21 11:00:17.640600: step: 572/526, loss: 0.049231961369514465 2023-01-21 11:00:18.793620: step: 576/526, loss: 0.08227720856666565 2023-01-21 11:00:19.937970: step: 580/526, loss: 0.022121144458651543 2023-01-21 11:00:21.097206: step: 584/526, loss: 0.07328915596008301 2023-01-21 11:00:22.220424: step: 588/526, loss: 0.02001342736184597 2023-01-21 11:00:23.372636: step: 592/526, loss: 0.040125273168087006 2023-01-21 11:00:24.504227: step: 596/526, loss: 0.0020106793381273746 2023-01-21 11:00:25.625937: step: 600/526, loss: 0.01310644205659628 2023-01-21 11:00:26.773005: step: 604/526, loss: 0.0038715361151844263 2023-01-21 11:00:27.956361: step: 608/526, loss: 0.04984588548541069 2023-01-21 11:00:29.143475: step: 612/526, loss: 0.033686257898807526 2023-01-21 11:00:30.326967: step: 616/526, loss: 0.7480446696281433 2023-01-21 11:00:31.464161: step: 620/526, loss: 0.04864707216620445 2023-01-21 11:00:32.613865: step: 624/526, loss: 0.009557723999023438 2023-01-21 11:00:33.778447: step: 628/526, loss: 0.0551937110722065 2023-01-21 11:00:34.907189: step: 632/526, loss: 0.03425464779138565 2023-01-21 11:00:36.064865: step: 636/526, loss: 0.025500871241092682 2023-01-21 11:00:37.219056: step: 640/526, loss: 0.03884430229663849 2023-01-21 11:00:38.391743: step: 644/526, loss: 0.8188632130622864 2023-01-21 11:00:39.516064: step: 648/526, loss: 0.06335659325122833 2023-01-21 11:00:40.627623: step: 652/526, loss: 0.03251013904809952 2023-01-21 11:00:41.786154: step: 656/526, loss: 0.01196746900677681 2023-01-21 11:00:42.951128: step: 660/526, loss: 0.05213518068194389 2023-01-21 11:00:44.091400: step: 664/526, loss: 0.009797478094696999 2023-01-21 11:00:45.259810: step: 668/526, loss: 0.04624948650598526 2023-01-21 11:00:46.402176: step: 672/526, loss: 0.0673513412475586 2023-01-21 11:00:47.536857: step: 676/526, loss: 0.012476349249482155 2023-01-21 11:00:48.665645: step: 680/526, loss: 0.025736570358276367 2023-01-21 11:00:49.844559: step: 684/526, loss: 0.013802719302475452 2023-01-21 11:00:51.035273: step: 688/526, loss: 0.01801929622888565 2023-01-21 11:00:52.180708: step: 692/526, loss: 0.04194164276123047 2023-01-21 11:00:53.433138: step: 696/526, loss: 0.06269092857837677 2023-01-21 11:00:54.635434: step: 700/526, loss: 0.042037203907966614 2023-01-21 11:00:55.817834: step: 704/526, loss: 0.019763469696044922 2023-01-21 11:00:56.999709: step: 708/526, loss: 0.049951933324337006 2023-01-21 11:00:58.145392: step: 712/526, loss: 0.2472720742225647 2023-01-21 11:00:59.297034: step: 716/526, loss: 0.04955577850341797 2023-01-21 11:01:00.438897: step: 720/526, loss: 0.019998932257294655 2023-01-21 11:01:01.587488: step: 724/526, loss: 0.3687778413295746 2023-01-21 11:01:02.754381: step: 728/526, loss: 0.01746220700442791 2023-01-21 11:01:03.912169: step: 732/526, loss: 0.08555164933204651 2023-01-21 11:01:05.065474: step: 736/526, loss: 0.028433572500944138 2023-01-21 11:01:06.210393: step: 740/526, loss: 0.012831354513764381 2023-01-21 11:01:07.381795: step: 744/526, loss: 0.022426225244998932 2023-01-21 11:01:08.548938: step: 748/526, loss: 0.003670215606689453 2023-01-21 11:01:09.702554: step: 752/526, loss: 0.001497554825618863 2023-01-21 11:01:10.864926: step: 756/526, loss: 0.01020956039428711 2023-01-21 11:01:12.019373: step: 760/526, loss: 0.10498128831386566 2023-01-21 11:01:13.173068: step: 764/526, loss: 0.04622707515954971 2023-01-21 11:01:14.352423: step: 768/526, loss: 0.02273707464337349 2023-01-21 11:01:15.521269: step: 772/526, loss: 0.1283605992794037 2023-01-21 11:01:16.686872: step: 776/526, loss: 0.02354917675256729 2023-01-21 11:01:17.857340: step: 780/526, loss: 0.08163337409496307 2023-01-21 11:01:19.002866: step: 784/526, loss: 0.01126704178750515 2023-01-21 11:01:20.173652: step: 788/526, loss: 0.01570143736898899 2023-01-21 11:01:21.317654: step: 792/526, loss: 0.010365486145019531 2023-01-21 11:01:22.465318: step: 796/526, loss: 0.28114375472068787 2023-01-21 11:01:23.622305: step: 800/526, loss: 0.13259848952293396 2023-01-21 11:01:24.766463: step: 804/526, loss: 0.015461349859833717 2023-01-21 11:01:25.999562: step: 808/526, loss: 0.010441113263368607 2023-01-21 11:01:27.136416: step: 812/526, loss: 0.04111804813146591 2023-01-21 11:01:28.299194: step: 816/526, loss: 0.09982052445411682 2023-01-21 11:01:29.431704: step: 820/526, loss: 0.03638429567217827 2023-01-21 11:01:30.574084: step: 824/526, loss: 0.021789170801639557 2023-01-21 11:01:31.728800: step: 828/526, loss: 0.0339839942753315 2023-01-21 11:01:32.890581: step: 832/526, loss: 0.03462924808263779 2023-01-21 11:01:34.021295: step: 836/526, loss: 0.0086548812687397 2023-01-21 11:01:35.163347: step: 840/526, loss: 0.0022373199462890625 2023-01-21 11:01:36.315756: step: 844/526, loss: 0.0487859770655632 2023-01-21 11:01:37.465390: step: 848/526, loss: 0.0534062385559082 2023-01-21 11:01:38.609420: step: 852/526, loss: 0.0008760690689086914 2023-01-21 11:01:39.749350: step: 856/526, loss: 0.017857860773801804 2023-01-21 11:01:40.893094: step: 860/526, loss: 0.01971343904733658 2023-01-21 11:01:42.048246: step: 864/526, loss: 0.026155853644013405 2023-01-21 11:01:43.194000: step: 868/526, loss: 0.056824494153261185 2023-01-21 11:01:44.341085: step: 872/526, loss: 0.04470262676477432 2023-01-21 11:01:45.477809: step: 876/526, loss: 0.05740748345851898 2023-01-21 11:01:46.619413: step: 880/526, loss: 0.10411418229341507 2023-01-21 11:01:47.781539: step: 884/526, loss: 0.08634739369153976 2023-01-21 11:01:48.934862: step: 888/526, loss: 1.0524760484695435 2023-01-21 11:01:50.095611: step: 892/526, loss: 0.09735670685768127 2023-01-21 11:01:51.239511: step: 896/526, loss: 0.026569750159978867 2023-01-21 11:01:52.379681: step: 900/526, loss: 0.0115820886567235 2023-01-21 11:01:53.542987: step: 904/526, loss: 0.013141918927431107 2023-01-21 11:01:54.692878: step: 908/526, loss: 0.14877520501613617 2023-01-21 11:01:55.828503: step: 912/526, loss: 0.0027168274391442537 2023-01-21 11:01:56.972594: step: 916/526, loss: 0.02921476401388645 2023-01-21 11:01:58.150501: step: 920/526, loss: 0.01851959154009819 2023-01-21 11:01:59.309126: step: 924/526, loss: 0.027705764397978783 2023-01-21 11:02:00.445649: step: 928/526, loss: 0.08063450455665588 2023-01-21 11:02:01.608943: step: 932/526, loss: 0.0413387306034565 2023-01-21 11:02:02.761158: step: 936/526, loss: 0.11855144798755646 2023-01-21 11:02:03.913377: step: 940/526, loss: 0.019108010455965996 2023-01-21 11:02:05.060042: step: 944/526, loss: 0.06180715560913086 2023-01-21 11:02:06.197327: step: 948/526, loss: 0.01549997366964817 2023-01-21 11:02:07.313765: step: 952/526, loss: 0.01909341849386692 2023-01-21 11:02:08.430056: step: 956/526, loss: 0.08234482258558273 2023-01-21 11:02:09.582698: step: 960/526, loss: 0.12263260036706924 2023-01-21 11:02:10.699194: step: 964/526, loss: 0.027236558496952057 2023-01-21 11:02:11.850667: step: 968/526, loss: 0.034651611000299454 2023-01-21 11:02:13.059898: step: 972/526, loss: 0.039281558245420456 2023-01-21 11:02:14.247946: step: 976/526, loss: 0.018955709412693977 2023-01-21 11:02:15.393153: step: 980/526, loss: 0.03333435207605362 2023-01-21 11:02:16.548803: step: 984/526, loss: 0.03172741085290909 2023-01-21 11:02:17.724422: step: 988/526, loss: 0.03255309909582138 2023-01-21 11:02:18.857595: step: 992/526, loss: 0.00015573501877952367 2023-01-21 11:02:19.998023: step: 996/526, loss: 0.021399879828095436 2023-01-21 11:02:21.162746: step: 1000/526, loss: 0.02144594117999077 2023-01-21 11:02:22.320033: step: 1004/526, loss: 0.056670188903808594 2023-01-21 11:02:23.474899: step: 1008/526, loss: 0.022966861724853516 2023-01-21 11:02:24.621274: step: 1012/526, loss: 0.12093563377857208 2023-01-21 11:02:25.763773: step: 1016/526, loss: 0.0705813392996788 2023-01-21 11:02:26.915271: step: 1020/526, loss: 0.0035948753356933594 2023-01-21 11:02:28.047359: step: 1024/526, loss: 0.09655246883630753 2023-01-21 11:02:29.185031: step: 1028/526, loss: 0.04067058861255646 2023-01-21 11:02:30.352551: step: 1032/526, loss: 0.0689840316772461 2023-01-21 11:02:31.444873: step: 1036/526, loss: 0.03283090516924858 2023-01-21 11:02:32.600123: step: 1040/526, loss: 0.004947471432387829 2023-01-21 11:02:33.750613: step: 1044/526, loss: 0.0627809539437294 2023-01-21 11:02:34.915686: step: 1048/526, loss: 0.03735370934009552 2023-01-21 11:02:36.084518: step: 1052/526, loss: 0.034029770642519 2023-01-21 11:02:37.263616: step: 1056/526, loss: 0.013742304407060146 2023-01-21 11:02:38.453563: step: 1060/526, loss: 0.08030929416418076 2023-01-21 11:02:39.635197: step: 1064/526, loss: 0.007369565777480602 2023-01-21 11:02:40.783588: step: 1068/526, loss: 0.04058799520134926 2023-01-21 11:02:41.901004: step: 1072/526, loss: 0.04046201705932617 2023-01-21 11:02:43.094687: step: 1076/526, loss: 0.016430091112852097 2023-01-21 11:02:44.233946: step: 1080/526, loss: 0.13665838539600372 2023-01-21 11:02:45.398005: step: 1084/526, loss: 0.007668304722756147 2023-01-21 11:02:46.579948: step: 1088/526, loss: 0.03161673620343208 2023-01-21 11:02:47.724973: step: 1092/526, loss: 0.028898097574710846 2023-01-21 11:02:48.884042: step: 1096/526, loss: 0.05564174801111221 2023-01-21 11:02:50.042256: step: 1100/526, loss: 0.008345984853804111 2023-01-21 11:02:51.169123: step: 1104/526, loss: 0.030109405517578125 2023-01-21 11:02:52.304312: step: 1108/526, loss: 0.005655097775161266 2023-01-21 11:02:53.502757: step: 1112/526, loss: 0.12387847900390625 2023-01-21 11:02:54.651512: step: 1116/526, loss: 0.042987726628780365 2023-01-21 11:02:55.807045: step: 1120/526, loss: 0.02583017572760582 2023-01-21 11:02:56.969262: step: 1124/526, loss: 0.0813068374991417 2023-01-21 11:02:58.139530: step: 1128/526, loss: 0.034679412841796875 2023-01-21 11:02:59.318787: step: 1132/526, loss: 0.003002262208610773 2023-01-21 11:03:00.460105: step: 1136/526, loss: 0.01450738962739706 2023-01-21 11:03:01.577664: step: 1140/526, loss: 0.01184234581887722 2023-01-21 11:03:02.717949: step: 1144/526, loss: 0.1052946075797081 2023-01-21 11:03:03.925435: step: 1148/526, loss: 0.025901008397340775 2023-01-21 11:03:05.104064: step: 1152/526, loss: 0.03181643784046173 2023-01-21 11:03:06.233687: step: 1156/526, loss: 0.005623912904411554 2023-01-21 11:03:07.411858: step: 1160/526, loss: 0.04248478263616562 2023-01-21 11:03:08.564251: step: 1164/526, loss: 0.003130245255306363 2023-01-21 11:03:09.705742: step: 1168/526, loss: 0.030846262350678444 2023-01-21 11:03:10.864526: step: 1172/526, loss: 0.0061657908372581005 2023-01-21 11:03:12.052379: step: 1176/526, loss: 0.038248300552368164 2023-01-21 11:03:13.213868: step: 1180/526, loss: 0.04946904256939888 2023-01-21 11:03:14.353801: step: 1184/526, loss: 0.0031760213896632195 2023-01-21 11:03:15.498885: step: 1188/526, loss: 0.016347885131835938 2023-01-21 11:03:16.640493: step: 1192/526, loss: 0.0021251202560961246 2023-01-21 11:03:17.783998: step: 1196/526, loss: 0.015226269140839577 2023-01-21 11:03:18.954473: step: 1200/526, loss: 0.03696164861321449 2023-01-21 11:03:20.132513: step: 1204/526, loss: 0.0946749672293663 2023-01-21 11:03:21.290910: step: 1208/526, loss: 0.09247412532567978 2023-01-21 11:03:22.461825: step: 1212/526, loss: 0.02320464886724949 2023-01-21 11:03:23.643542: step: 1216/526, loss: 0.003293800400570035 2023-01-21 11:03:24.797741: step: 1220/526, loss: 0.00538978585973382 2023-01-21 11:03:25.981815: step: 1224/526, loss: 0.02916708029806614 2023-01-21 11:03:27.138338: step: 1228/526, loss: 0.029478073120117188 2023-01-21 11:03:28.285698: step: 1232/526, loss: 0.00650176964700222 2023-01-21 11:03:29.403211: step: 1236/526, loss: 0.0022499561309814453 2023-01-21 11:03:30.562947: step: 1240/526, loss: 0.007954406552016735 2023-01-21 11:03:31.724967: step: 1244/526, loss: 0.0009212493896484375 2023-01-21 11:03:32.888410: step: 1248/526, loss: 0.09075680375099182 2023-01-21 11:03:34.028627: step: 1252/526, loss: 0.008729934692382812 2023-01-21 11:03:35.177011: step: 1256/526, loss: 0.0055634500458836555 2023-01-21 11:03:36.333971: step: 1260/526, loss: 0.03166494518518448 2023-01-21 11:03:37.499320: step: 1264/526, loss: 0.0009611129644326866 2023-01-21 11:03:38.626209: step: 1268/526, loss: 0.011462783440947533 2023-01-21 11:03:39.762577: step: 1272/526, loss: 0.050531625747680664 2023-01-21 11:03:40.901746: step: 1276/526, loss: 0.057877495884895325 2023-01-21 11:03:42.060355: step: 1280/526, loss: 0.02187213860452175 2023-01-21 11:03:43.236555: step: 1284/526, loss: 0.02775602415204048 2023-01-21 11:03:44.422285: step: 1288/526, loss: 0.033225297927856445 2023-01-21 11:03:45.562876: step: 1292/526, loss: 0.009862899780273438 2023-01-21 11:03:46.732148: step: 1296/526, loss: 0.039669036865234375 2023-01-21 11:03:47.911796: step: 1300/526, loss: 0.0007548332796432078 2023-01-21 11:03:49.071971: step: 1304/526, loss: 0.05173492431640625 2023-01-21 11:03:50.186685: step: 1308/526, loss: 0.08933629840612411 2023-01-21 11:03:51.309070: step: 1312/526, loss: 0.05989236757159233 2023-01-21 11:03:52.435267: step: 1316/526, loss: 0.02345418930053711 2023-01-21 11:03:53.585617: step: 1320/526, loss: 0.03348731994628906 2023-01-21 11:03:54.726177: step: 1324/526, loss: 0.01136484183371067 2023-01-21 11:03:55.858049: step: 1328/526, loss: 0.0005401134840212762 2023-01-21 11:03:57.030390: step: 1332/526, loss: 0.09696026146411896 2023-01-21 11:03:58.196434: step: 1336/526, loss: 0.08942870795726776 2023-01-21 11:03:59.358723: step: 1340/526, loss: 0.017301559448242188 2023-01-21 11:04:00.528914: step: 1344/526, loss: 0.04023704677820206 2023-01-21 11:04:01.667372: step: 1348/526, loss: 0.006279659457504749 2023-01-21 11:04:02.817988: step: 1352/526, loss: 0.012971878051757812 2023-01-21 11:04:03.983648: step: 1356/526, loss: 0.012519550509750843 2023-01-21 11:04:05.137031: step: 1360/526, loss: 0.022723007947206497 2023-01-21 11:04:06.288801: step: 1364/526, loss: 0.03470955044031143 2023-01-21 11:04:07.422228: step: 1368/526, loss: 0.07795833796262741 2023-01-21 11:04:08.585940: step: 1372/526, loss: 0.016532326117157936 2023-01-21 11:04:09.722609: step: 1376/526, loss: 0.007169055752456188 2023-01-21 11:04:10.877361: step: 1380/526, loss: 0.04544110223650932 2023-01-21 11:04:12.059720: step: 1384/526, loss: 0.01293106097728014 2023-01-21 11:04:13.212332: step: 1388/526, loss: 0.011092161759734154 2023-01-21 11:04:14.355691: step: 1392/526, loss: 0.0023432732559740543 2023-01-21 11:04:15.476980: step: 1396/526, loss: 0.004767036531120539 2023-01-21 11:04:16.643123: step: 1400/526, loss: 0.01958489418029785 2023-01-21 11:04:17.784499: step: 1404/526, loss: 0.07528066635131836 2023-01-21 11:04:18.929366: step: 1408/526, loss: 0.2621818482875824 2023-01-21 11:04:20.075420: step: 1412/526, loss: 0.03466815873980522 2023-01-21 11:04:21.265455: step: 1416/526, loss: 0.04376044496893883 2023-01-21 11:04:22.420004: step: 1420/526, loss: 0.048659421503543854 2023-01-21 11:04:23.583196: step: 1424/526, loss: 0.0004706382751464844 2023-01-21 11:04:24.694398: step: 1428/526, loss: 0.010799694806337357 2023-01-21 11:04:25.861230: step: 1432/526, loss: 0.23829756677150726 2023-01-21 11:04:26.995104: step: 1436/526, loss: 0.09141998738050461 2023-01-21 11:04:28.121281: step: 1440/526, loss: 0.027706623077392578 2023-01-21 11:04:29.308463: step: 1444/526, loss: 0.06473293155431747 2023-01-21 11:04:30.459853: step: 1448/526, loss: 0.12289729714393616 2023-01-21 11:04:31.595341: step: 1452/526, loss: 0.021455192938447 2023-01-21 11:04:32.735184: step: 1456/526, loss: 0.04598226770758629 2023-01-21 11:04:33.906452: step: 1460/526, loss: 0.022632120177149773 2023-01-21 11:04:35.040350: step: 1464/526, loss: 0.07233905792236328 2023-01-21 11:04:36.188964: step: 1468/526, loss: 0.04462037235498428 2023-01-21 11:04:37.310484: step: 1472/526, loss: 0.04205169901251793 2023-01-21 11:04:38.458408: step: 1476/526, loss: 0.06706027686595917 2023-01-21 11:04:39.596029: step: 1480/526, loss: 0.03812894970178604 2023-01-21 11:04:40.732562: step: 1484/526, loss: 0.36283913254737854 2023-01-21 11:04:41.865016: step: 1488/526, loss: 0.0345952995121479 2023-01-21 11:04:43.009366: step: 1492/526, loss: 0.00749053992331028 2023-01-21 11:04:44.172792: step: 1496/526, loss: 0.026373863220214844 2023-01-21 11:04:45.329748: step: 1500/526, loss: 0.13010425865650177 2023-01-21 11:04:46.470489: step: 1504/526, loss: 0.008827782236039639 2023-01-21 11:04:47.625102: step: 1508/526, loss: 0.13581402599811554 2023-01-21 11:04:48.747380: step: 1512/526, loss: 0.0032296180725097656 2023-01-21 11:04:49.890094: step: 1516/526, loss: 0.018725205212831497 2023-01-21 11:04:51.066307: step: 1520/526, loss: 0.032004546374082565 2023-01-21 11:04:52.241986: step: 1524/526, loss: 0.04018688201904297 2023-01-21 11:04:53.388620: step: 1528/526, loss: 0.03525538370013237 2023-01-21 11:04:54.552574: step: 1532/526, loss: 0.011114586144685745 2023-01-21 11:04:55.712270: step: 1536/526, loss: 0.026351165026426315 2023-01-21 11:04:56.854672: step: 1540/526, loss: 0.04839515686035156 2023-01-21 11:04:57.977637: step: 1544/526, loss: 0.053110361099243164 2023-01-21 11:04:59.173968: step: 1548/526, loss: 0.010938739404082298 2023-01-21 11:05:00.334030: step: 1552/526, loss: 0.04854869842529297 2023-01-21 11:05:01.510059: step: 1556/526, loss: 0.7395030856132507 2023-01-21 11:05:02.638632: step: 1560/526, loss: 0.17734451591968536 2023-01-21 11:05:03.805555: step: 1564/526, loss: 0.015507126227021217 2023-01-21 11:05:04.966623: step: 1568/526, loss: 0.012645245529711246 2023-01-21 11:05:06.106499: step: 1572/526, loss: 0.0001027107282425277 2023-01-21 11:05:07.236044: step: 1576/526, loss: 0.018884658813476562 2023-01-21 11:05:08.394107: step: 1580/526, loss: 0.10373573005199432 2023-01-21 11:05:09.553844: step: 1584/526, loss: 0.009451961144804955 2023-01-21 11:05:10.711029: step: 1588/526, loss: 0.009971809573471546 2023-01-21 11:05:11.851062: step: 1592/526, loss: 0.012992668896913528 2023-01-21 11:05:13.005238: step: 1596/526, loss: 0.03197155147790909 2023-01-21 11:05:14.182757: step: 1600/526, loss: 0.012263203039765358 2023-01-21 11:05:15.367631: step: 1604/526, loss: 0.02445230446755886 2023-01-21 11:05:16.563777: step: 1608/526, loss: 0.016923904418945312 2023-01-21 11:05:17.712161: step: 1612/526, loss: 0.07473859935998917 2023-01-21 11:05:18.837750: step: 1616/526, loss: 0.02385730855166912 2023-01-21 11:05:19.976817: step: 1620/526, loss: 0.034781455993652344 2023-01-21 11:05:21.154593: step: 1624/526, loss: 0.07011270523071289 2023-01-21 11:05:22.328773: step: 1628/526, loss: 0.042014218866825104 2023-01-21 11:05:23.482959: step: 1632/526, loss: 0.04486117139458656 2023-01-21 11:05:24.654385: step: 1636/526, loss: 0.008313750848174095 2023-01-21 11:05:25.805363: step: 1640/526, loss: 0.02460041083395481 2023-01-21 11:05:26.941460: step: 1644/526, loss: 0.003979104105383158 2023-01-21 11:05:28.082350: step: 1648/526, loss: 0.01498270034790039 2023-01-21 11:05:29.237117: step: 1652/526, loss: 0.04777412489056587 2023-01-21 11:05:30.378186: step: 1656/526, loss: 0.11847671866416931 2023-01-21 11:05:31.528725: step: 1660/526, loss: 0.22739095985889435 2023-01-21 11:05:32.710293: step: 1664/526, loss: 0.00045385363046079874 2023-01-21 11:05:33.861251: step: 1668/526, loss: 0.318939208984375 2023-01-21 11:05:34.984116: step: 1672/526, loss: 0.0011371612781658769 2023-01-21 11:05:36.149287: step: 1676/526, loss: 0.01677551306784153 2023-01-21 11:05:37.276593: step: 1680/526, loss: 0.005530738737434149 2023-01-21 11:05:38.428953: step: 1684/526, loss: 0.04484749212861061 2023-01-21 11:05:39.576909: step: 1688/526, loss: 0.034762192517519 2023-01-21 11:05:40.743718: step: 1692/526, loss: 0.03612661361694336 2023-01-21 11:05:41.919709: step: 1696/526, loss: 0.24682140350341797 2023-01-21 11:05:43.119286: step: 1700/526, loss: 0.0073286062106490135 2023-01-21 11:05:44.239218: step: 1704/526, loss: 0.024684906005859375 2023-01-21 11:05:45.386150: step: 1708/526, loss: 0.011906147934496403 2023-01-21 11:05:46.525677: step: 1712/526, loss: 0.04727668687701225 2023-01-21 11:05:47.674122: step: 1716/526, loss: 0.021604634821414948 2023-01-21 11:05:48.804992: step: 1720/526, loss: 0.010314512066543102 2023-01-21 11:05:49.946334: step: 1724/526, loss: 0.04415760189294815 2023-01-21 11:05:51.101334: step: 1728/526, loss: 0.0003491402021609247 2023-01-21 11:05:52.267577: step: 1732/526, loss: 0.070029117166996 2023-01-21 11:05:53.406020: step: 1736/526, loss: 0.039321184158325195 2023-01-21 11:05:54.536213: step: 1740/526, loss: 0.04083128273487091 2023-01-21 11:05:55.701056: step: 1744/526, loss: 0.030246615409851074 2023-01-21 11:05:56.872218: step: 1748/526, loss: 0.0046062469482421875 2023-01-21 11:05:57.983975: step: 1752/526, loss: 0.004038667771965265 2023-01-21 11:05:59.136069: step: 1756/526, loss: 0.0022607804276049137 2023-01-21 11:06:00.272372: step: 1760/526, loss: 0.03269386291503906 2023-01-21 11:06:01.414626: step: 1764/526, loss: 0.08118186146020889 2023-01-21 11:06:02.542523: step: 1768/526, loss: 0.03135528787970543 2023-01-21 11:06:03.681658: step: 1772/526, loss: 0.04818706586956978 2023-01-21 11:06:04.807848: step: 1776/526, loss: 0.009947061538696289 2023-01-21 11:06:05.965233: step: 1780/526, loss: 0.045124150812625885 2023-01-21 11:06:07.120214: step: 1784/526, loss: 0.014792424626648426 2023-01-21 11:06:08.247794: step: 1788/526, loss: 0.01794910430908203 2023-01-21 11:06:09.426824: step: 1792/526, loss: 0.05271501839160919 2023-01-21 11:06:10.599216: step: 1796/526, loss: 0.015408897772431374 2023-01-21 11:06:11.744840: step: 1800/526, loss: 0.0028742789290845394 2023-01-21 11:06:12.917366: step: 1804/526, loss: 0.002128505613654852 2023-01-21 11:06:14.047413: step: 1808/526, loss: 0.132737934589386 2023-01-21 11:06:15.216611: step: 1812/526, loss: 0.10737285763025284 2023-01-21 11:06:16.345334: step: 1816/526, loss: 0.06412134319543839 2023-01-21 11:06:17.463489: step: 1820/526, loss: 0.11031875014305115 2023-01-21 11:06:18.623773: step: 1824/526, loss: 0.09334290772676468 2023-01-21 11:06:19.777287: step: 1828/526, loss: 0.024251842871308327 2023-01-21 11:06:20.900089: step: 1832/526, loss: 0.03070240095257759 2023-01-21 11:06:22.055144: step: 1836/526, loss: 0.06307859718799591 2023-01-21 11:06:23.188971: step: 1840/526, loss: 0.0629362091422081 2023-01-21 11:06:24.330437: step: 1844/526, loss: 0.004518937785178423 2023-01-21 11:06:25.518271: step: 1848/526, loss: 0.07525482028722763 2023-01-21 11:06:26.651592: step: 1852/526, loss: 0.035201456397771835 2023-01-21 11:06:27.817369: step: 1856/526, loss: 0.07313995063304901 2023-01-21 11:06:29.003553: step: 1860/526, loss: 0.06632176041603088 2023-01-21 11:06:30.160117: step: 1864/526, loss: 0.034888364374637604 2023-01-21 11:06:31.297046: step: 1868/526, loss: 0.04519472271203995 2023-01-21 11:06:32.418687: step: 1872/526, loss: 3.2901764370762976e-06 2023-01-21 11:06:33.594175: step: 1876/526, loss: 0.05560455098748207 2023-01-21 11:06:34.717489: step: 1880/526, loss: 0.0690622329711914 2023-01-21 11:06:35.847702: step: 1884/526, loss: 0.002453422639518976 2023-01-21 11:06:37.009694: step: 1888/526, loss: 0.0332578644156456 2023-01-21 11:06:38.171541: step: 1892/526, loss: 0.01622753217816353 2023-01-21 11:06:39.349259: step: 1896/526, loss: 0.08401737362146378 2023-01-21 11:06:40.478864: step: 1900/526, loss: 0.013556289486587048 2023-01-21 11:06:41.622070: step: 1904/526, loss: 0.016473103314638138 2023-01-21 11:06:42.772024: step: 1908/526, loss: 0.02500324323773384 2023-01-21 11:06:43.937231: step: 1912/526, loss: 0.03747615963220596 2023-01-21 11:06:45.125498: step: 1916/526, loss: 0.01623210869729519 2023-01-21 11:06:46.275940: step: 1920/526, loss: 0.020865153521299362 2023-01-21 11:06:47.436972: step: 1924/526, loss: 0.09426537156105042 2023-01-21 11:06:48.601304: step: 1928/526, loss: 0.03553667291998863 2023-01-21 11:06:49.767241: step: 1932/526, loss: 0.0386173278093338 2023-01-21 11:06:50.907766: step: 1936/526, loss: 0.04942674562335014 2023-01-21 11:06:52.078277: step: 1940/526, loss: 0.034838005900382996 2023-01-21 11:06:53.314258: step: 1944/526, loss: 0.11963681876659393 2023-01-21 11:06:54.441047: step: 1948/526, loss: 0.009502792730927467 2023-01-21 11:06:55.577711: step: 1952/526, loss: 0.05932999029755592 2023-01-21 11:06:56.774046: step: 1956/526, loss: 0.013558006845414639 2023-01-21 11:06:57.928373: step: 1960/526, loss: 0.3790111541748047 2023-01-21 11:06:59.100720: step: 1964/526, loss: 0.00024013521033339202 2023-01-21 11:07:00.293281: step: 1968/526, loss: 0.002020931337028742 2023-01-21 11:07:01.447652: step: 1972/526, loss: 0.008262872695922852 2023-01-21 11:07:02.609877: step: 1976/526, loss: 0.045568086206912994 2023-01-21 11:07:03.773019: step: 1980/526, loss: 0.03980877622961998 2023-01-21 11:07:04.900915: step: 1984/526, loss: 0.010819340124726295 2023-01-21 11:07:06.023471: step: 1988/526, loss: 0.08175916224718094 2023-01-21 11:07:07.180703: step: 1992/526, loss: 0.04726286232471466 2023-01-21 11:07:08.335002: step: 1996/526, loss: 0.013749408535659313 2023-01-21 11:07:09.509413: step: 2000/526, loss: 0.09512300789356232 2023-01-21 11:07:10.650275: step: 2004/526, loss: 0.003344666911289096 2023-01-21 11:07:11.789171: step: 2008/526, loss: 0.002407073974609375 2023-01-21 11:07:12.969210: step: 2012/526, loss: 0.0848415419459343 2023-01-21 11:07:14.129578: step: 2016/526, loss: 0.01933269575238228 2023-01-21 11:07:15.298387: step: 2020/526, loss: 0.003736400743946433 2023-01-21 11:07:16.450933: step: 2024/526, loss: 0.03413262590765953 2023-01-21 11:07:17.597411: step: 2028/526, loss: 0.024927139282226562 2023-01-21 11:07:18.732507: step: 2032/526, loss: 0.006679153069853783 2023-01-21 11:07:19.891718: step: 2036/526, loss: 0.024360084906220436 2023-01-21 11:07:21.081482: step: 2040/526, loss: 0.059807490557432175 2023-01-21 11:07:22.246610: step: 2044/526, loss: 0.11016955226659775 2023-01-21 11:07:23.378656: step: 2048/526, loss: 0.06936921924352646 2023-01-21 11:07:24.558832: step: 2052/526, loss: 0.004474448971450329 2023-01-21 11:07:25.688536: step: 2056/526, loss: 0.01922025717794895 2023-01-21 11:07:26.866633: step: 2060/526, loss: 0.04763145372271538 2023-01-21 11:07:27.990832: step: 2064/526, loss: 0.03901495784521103 2023-01-21 11:07:29.134067: step: 2068/526, loss: 0.05246410891413689 2023-01-21 11:07:30.292480: step: 2072/526, loss: 0.017983341589570045 2023-01-21 11:07:31.483179: step: 2076/526, loss: 0.024172592908143997 2023-01-21 11:07:32.603686: step: 2080/526, loss: 0.015199471265077591 2023-01-21 11:07:33.746901: step: 2084/526, loss: 0.01585559919476509 2023-01-21 11:07:34.925411: step: 2088/526, loss: 0.03293590620160103 2023-01-21 11:07:36.064882: step: 2092/526, loss: 0.03557705879211426 2023-01-21 11:07:37.192093: step: 2096/526, loss: 0.022069169208407402 2023-01-21 11:07:38.368276: step: 2100/526, loss: 0.0372493751347065 2023-01-21 11:07:39.513766: step: 2104/526, loss: 0.036190200597047806 ================================================== Loss: 0.055 -------------------- Dev: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Test: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Korean: {'event': {'p': 0.7045454545454546, 'r': 0.49206349206349204, 'f1': 0.5794392523364486}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Russian: {'event': {'p': 0.4634146341463415, 'r': 0.5277777777777778, 'f1': 0.4935064935064935}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} New best chinese model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 14 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:08:28.862227: step: 4/526, loss: 0.028769290074706078 2023-01-21 11:08:30.027257: step: 8/526, loss: 0.010029559955000877 2023-01-21 11:08:31.176298: step: 12/526, loss: 0.052431218326091766 2023-01-21 11:08:32.320000: step: 16/526, loss: 0.0002471923944540322 2023-01-21 11:08:33.458455: step: 20/526, loss: 0.020116042345762253 2023-01-21 11:08:34.658627: step: 24/526, loss: 0.04445667192339897 2023-01-21 11:08:35.832304: step: 28/526, loss: 0.0448613166809082 2023-01-21 11:08:36.967720: step: 32/526, loss: 0.0025829316582530737 2023-01-21 11:08:38.119821: step: 36/526, loss: 0.011233711615204811 2023-01-21 11:08:39.292882: step: 40/526, loss: 0.0006347656017169356 2023-01-21 11:08:40.435442: step: 44/526, loss: 0.004898023791611195 2023-01-21 11:08:41.563094: step: 48/526, loss: 0.0010742663871496916 2023-01-21 11:08:42.718282: step: 52/526, loss: 0.056396484375 2023-01-21 11:08:43.887886: step: 56/526, loss: 0.006851291749626398 2023-01-21 11:08:45.080638: step: 60/526, loss: 0.005477714352309704 2023-01-21 11:08:46.209062: step: 64/526, loss: 0.00089178082998842 2023-01-21 11:08:47.333874: step: 68/526, loss: 0.026175975799560547 2023-01-21 11:08:48.485919: step: 72/526, loss: 0.026777077466249466 2023-01-21 11:08:49.623450: step: 76/526, loss: 0.01096954382956028 2023-01-21 11:08:50.776551: step: 80/526, loss: 0.04775390774011612 2023-01-21 11:08:51.923466: step: 84/526, loss: 0.018092917278409004 2023-01-21 11:08:53.109931: step: 88/526, loss: 0.05182266607880592 2023-01-21 11:08:54.263286: step: 92/526, loss: 0.009153938852250576 2023-01-21 11:08:55.413745: step: 96/526, loss: 0.012828255072236061 2023-01-21 11:08:56.571644: step: 100/526, loss: 0.023183632642030716 2023-01-21 11:08:57.743954: step: 104/526, loss: 0.02047748677432537 2023-01-21 11:08:58.899931: step: 108/526, loss: 0.006088924594223499 2023-01-21 11:09:00.046147: step: 112/526, loss: 0.010431480593979359 2023-01-21 11:09:01.186273: step: 116/526, loss: 0.003316784044727683 2023-01-21 11:09:02.340109: step: 120/526, loss: 0.022585012018680573 2023-01-21 11:09:03.513989: step: 124/526, loss: 0.0050903321243822575 2023-01-21 11:09:04.651083: step: 128/526, loss: 0.039612531661987305 2023-01-21 11:09:05.816801: step: 132/526, loss: 0.013351202011108398 2023-01-21 11:09:06.956402: step: 136/526, loss: 0.026549531146883965 2023-01-21 11:09:08.156845: step: 140/526, loss: 0.04821624606847763 2023-01-21 11:09:09.302983: step: 144/526, loss: 0.006395912263542414 2023-01-21 11:09:10.488942: step: 148/526, loss: 0.010717582888901234 2023-01-21 11:09:11.648582: step: 152/526, loss: 0.04908733442425728 2023-01-21 11:09:12.763796: step: 156/526, loss: 0.0012494564289227128 2023-01-21 11:09:13.931496: step: 160/526, loss: 0.025116920471191406 2023-01-21 11:09:15.072056: step: 164/526, loss: 0.05233040079474449 2023-01-21 11:09:16.233381: step: 168/526, loss: 0.004216575529426336 2023-01-21 11:09:17.393238: step: 172/526, loss: 0.04400043562054634 2023-01-21 11:09:18.544196: step: 176/526, loss: 0.018488312140107155 2023-01-21 11:09:19.721362: step: 180/526, loss: 0.06219344213604927 2023-01-21 11:09:20.884743: step: 184/526, loss: 0.0006612777942791581 2023-01-21 11:09:22.031544: step: 188/526, loss: 1.0482269525527954 2023-01-21 11:09:23.172777: step: 192/526, loss: 0.012943553738296032 2023-01-21 11:09:24.325310: step: 196/526, loss: 0.007782650180160999 2023-01-21 11:09:25.479590: step: 200/526, loss: 0.00029478073702193797 2023-01-21 11:09:26.629074: step: 204/526, loss: 0.23826980590820312 2023-01-21 11:09:27.780662: step: 208/526, loss: 0.03896646574139595 2023-01-21 11:09:28.937487: step: 212/526, loss: 0.025578143075108528 2023-01-21 11:09:30.077497: step: 216/526, loss: 0.05037498474121094 2023-01-21 11:09:31.231856: step: 220/526, loss: 0.037569332867860794 2023-01-21 11:09:32.404655: step: 224/526, loss: 0.10282401740550995 2023-01-21 11:09:33.545674: step: 228/526, loss: 0.08367414772510529 2023-01-21 11:09:34.689443: step: 232/526, loss: 0.07944760471582413 2023-01-21 11:09:35.855992: step: 236/526, loss: 0.03297453001141548 2023-01-21 11:09:37.000899: step: 240/526, loss: 0.057761382311582565 2023-01-21 11:09:38.172094: step: 244/526, loss: 0.03840923309326172 2023-01-21 11:09:39.362074: step: 248/526, loss: 0.01747150346636772 2023-01-21 11:09:40.497481: step: 252/526, loss: 0.03914625570178032 2023-01-21 11:09:41.671898: step: 256/526, loss: 0.029239464551210403 2023-01-21 11:09:42.841100: step: 260/526, loss: 0.006405830383300781 2023-01-21 11:09:43.973185: step: 264/526, loss: 0.019260313361883163 2023-01-21 11:09:45.156181: step: 268/526, loss: 0.030454635620117188 2023-01-21 11:09:46.293160: step: 272/526, loss: 0.006913661956787109 2023-01-21 11:09:47.452970: step: 276/526, loss: 0.3178260624408722 2023-01-21 11:09:48.617825: step: 280/526, loss: 0.022599412128329277 2023-01-21 11:09:49.795302: step: 284/526, loss: 0.005079174414277077 2023-01-21 11:09:50.978109: step: 288/526, loss: 6.18457852397114e-05 2023-01-21 11:09:52.135328: step: 292/526, loss: 0.009868239983916283 2023-01-21 11:09:53.294707: step: 296/526, loss: 0.014399338513612747 2023-01-21 11:09:54.471798: step: 300/526, loss: 0.009044838137924671 2023-01-21 11:09:55.620890: step: 304/526, loss: 0.11165733635425568 2023-01-21 11:09:56.772933: step: 308/526, loss: 0.021715546026825905 2023-01-21 11:09:57.937639: step: 312/526, loss: 0.018271446228027344 2023-01-21 11:09:59.106018: step: 316/526, loss: 0.048165224492549896 2023-01-21 11:10:00.298927: step: 320/526, loss: 0.002834796905517578 2023-01-21 11:10:01.440642: step: 324/526, loss: 0.08821564167737961 2023-01-21 11:10:02.609829: step: 328/526, loss: 0.003925800323486328 2023-01-21 11:10:03.752098: step: 332/526, loss: 0.010478115640580654 2023-01-21 11:10:04.898339: step: 336/526, loss: 0.0032283782493323088 2023-01-21 11:10:06.057717: step: 340/526, loss: 0.1905813217163086 2023-01-21 11:10:07.260777: step: 344/526, loss: 0.06588344275951385 2023-01-21 11:10:08.424045: step: 348/526, loss: 0.005924320314079523 2023-01-21 11:10:09.581385: step: 352/526, loss: 0.06634500622749329 2023-01-21 11:10:10.741827: step: 356/526, loss: 0.006025886163115501 2023-01-21 11:10:11.888001: step: 360/526, loss: 0.008804893121123314 2023-01-21 11:10:13.081118: step: 364/526, loss: 0.03967122733592987 2023-01-21 11:10:14.213079: step: 368/526, loss: 0.001993370009586215 2023-01-21 11:10:15.374446: step: 372/526, loss: 0.07171478122472763 2023-01-21 11:10:16.521238: step: 376/526, loss: 0.013125038705766201 2023-01-21 11:10:17.709004: step: 380/526, loss: 0.09669800102710724 2023-01-21 11:10:18.875507: step: 384/526, loss: 0.010965729132294655 2023-01-21 11:10:20.030544: step: 388/526, loss: 0.03508472442626953 2023-01-21 11:10:21.205111: step: 392/526, loss: 0.010452366434037685 2023-01-21 11:10:22.371677: step: 396/526, loss: 0.007734966464340687 2023-01-21 11:10:23.544381: step: 400/526, loss: 0.0469091460108757 2023-01-21 11:10:24.673203: step: 404/526, loss: 0.008955668658018112 2023-01-21 11:10:25.839793: step: 408/526, loss: 0.0075807576067745686 2023-01-21 11:10:26.984783: step: 412/526, loss: 0.0035459520295262337 2023-01-21 11:10:28.140030: step: 416/526, loss: 0.056842658668756485 2023-01-21 11:10:29.304381: step: 420/526, loss: 0.07088947296142578 2023-01-21 11:10:30.452463: step: 424/526, loss: 0.006724882405251265 2023-01-21 11:10:31.600672: step: 428/526, loss: 0.0016445160144940019 2023-01-21 11:10:32.757899: step: 432/526, loss: 0.016928577795624733 2023-01-21 11:10:33.899224: step: 436/526, loss: 0.035657405853271484 2023-01-21 11:10:35.054121: step: 440/526, loss: 0.0058716777712106705 2023-01-21 11:10:36.202937: step: 444/526, loss: 0.0009237289777956903 2023-01-21 11:10:37.362581: step: 448/526, loss: 0.010971833020448685 2023-01-21 11:10:38.524188: step: 452/526, loss: 0.007061720360070467 2023-01-21 11:10:39.674932: step: 456/526, loss: 0.06919269263744354 2023-01-21 11:10:40.814317: step: 460/526, loss: 0.012153435498476028 2023-01-21 11:10:41.965215: step: 464/526, loss: 0.02987213246524334 2023-01-21 11:10:43.098401: step: 468/526, loss: 0.2166258692741394 2023-01-21 11:10:44.256292: step: 472/526, loss: 0.05221042409539223 2023-01-21 11:10:45.404479: step: 476/526, loss: 4.706382969743572e-05 2023-01-21 11:10:46.601925: step: 480/526, loss: 0.1522476226091385 2023-01-21 11:10:47.798446: step: 484/526, loss: 0.027977945283055305 2023-01-21 11:10:48.918964: step: 488/526, loss: 0.011879539117217064 2023-01-21 11:10:50.030297: step: 492/526, loss: 0.004175520036369562 2023-01-21 11:10:51.170792: step: 496/526, loss: 0.03925075754523277 2023-01-21 11:10:52.323942: step: 500/526, loss: 0.0074522970244288445 2023-01-21 11:10:53.474034: step: 504/526, loss: 0.018669797107577324 2023-01-21 11:10:54.620863: step: 508/526, loss: 0.0609624870121479 2023-01-21 11:10:55.791056: step: 512/526, loss: 0.00041131972102448344 2023-01-21 11:10:56.947113: step: 516/526, loss: 0.004340362269431353 2023-01-21 11:10:58.083906: step: 520/526, loss: 0.024741938337683678 2023-01-21 11:10:59.222285: step: 524/526, loss: 0.013489819131791592 2023-01-21 11:11:00.347475: step: 528/526, loss: 0.017365837469697 2023-01-21 11:11:01.552570: step: 532/526, loss: 0.06297626346349716 2023-01-21 11:11:02.688382: step: 536/526, loss: 0.011751556769013405 2023-01-21 11:11:03.796175: step: 540/526, loss: 0.024974536150693893 2023-01-21 11:11:04.944340: step: 544/526, loss: 0.03037581592798233 2023-01-21 11:11:06.076783: step: 548/526, loss: 0.1036502867937088 2023-01-21 11:11:07.239869: step: 552/526, loss: 0.001616668771021068 2023-01-21 11:11:08.378315: step: 556/526, loss: 0.10586948692798615 2023-01-21 11:11:09.532339: step: 560/526, loss: 0.020032214000821114 2023-01-21 11:11:10.664346: step: 564/526, loss: 0.01800668239593506 2023-01-21 11:11:11.852040: step: 568/526, loss: 0.05415081977844238 2023-01-21 11:11:12.989941: step: 572/526, loss: 0.02035980299115181 2023-01-21 11:11:14.145107: step: 576/526, loss: 0.07460957020521164 2023-01-21 11:11:15.314819: step: 580/526, loss: 0.23481902480125427 2023-01-21 11:11:16.463130: step: 584/526, loss: 0.004990768618881702 2023-01-21 11:11:17.616078: step: 588/526, loss: 0.005188083276152611 2023-01-21 11:11:18.758866: step: 592/526, loss: 0.035907745361328125 2023-01-21 11:11:19.927403: step: 596/526, loss: 0.11188049614429474 2023-01-21 11:11:21.081118: step: 600/526, loss: 0.011778736487030983 2023-01-21 11:11:22.228564: step: 604/526, loss: 0.03871631622314453 2023-01-21 11:11:23.386396: step: 608/526, loss: 0.06158256530761719 2023-01-21 11:11:24.545682: step: 612/526, loss: 0.029084108769893646 2023-01-21 11:11:25.733217: step: 616/526, loss: 0.0006958007579669356 2023-01-21 11:11:26.922035: step: 620/526, loss: 0.030344391241669655 2023-01-21 11:11:28.070127: step: 624/526, loss: 0.01116247195750475 2023-01-21 11:11:29.199115: step: 628/526, loss: 0.1720743328332901 2023-01-21 11:11:30.360122: step: 632/526, loss: 0.07003584504127502 2023-01-21 11:11:31.503741: step: 636/526, loss: 0.0007466792594641447 2023-01-21 11:11:32.668713: step: 640/526, loss: 0.0038945735432207584 2023-01-21 11:11:33.855429: step: 644/526, loss: 0.010624694637954235 2023-01-21 11:11:35.041007: step: 648/526, loss: 0.03835335001349449 2023-01-21 11:11:36.207742: step: 652/526, loss: 0.0007458686595782638 2023-01-21 11:11:37.347185: step: 656/526, loss: 0.022769834846258163 2023-01-21 11:11:38.479166: step: 660/526, loss: 0.022292137145996094 2023-01-21 11:11:39.641817: step: 664/526, loss: 0.03672933578491211 2023-01-21 11:11:40.793293: step: 668/526, loss: 0.010675812140107155 2023-01-21 11:11:41.930461: step: 672/526, loss: 0.08597974479198456 2023-01-21 11:11:43.076034: step: 676/526, loss: 0.0008501053089275956 2023-01-21 11:11:44.212187: step: 680/526, loss: 0.013707256875932217 2023-01-21 11:11:45.375376: step: 684/526, loss: 0.02264699898660183 2023-01-21 11:11:46.531861: step: 688/526, loss: 0.00027980803861282766 2023-01-21 11:11:47.674188: step: 692/526, loss: 0.04479851573705673 2023-01-21 11:11:48.816815: step: 696/526, loss: 0.017988301813602448 2023-01-21 11:11:49.975341: step: 700/526, loss: 0.08115081489086151 2023-01-21 11:11:51.141065: step: 704/526, loss: 0.1008453369140625 2023-01-21 11:11:52.316098: step: 708/526, loss: 0.03654451295733452 2023-01-21 11:11:53.457897: step: 712/526, loss: 0.022855425253510475 2023-01-21 11:11:54.597864: step: 716/526, loss: 0.009132814593613148 2023-01-21 11:11:55.758376: step: 720/526, loss: 0.05775337666273117 2023-01-21 11:11:56.947241: step: 724/526, loss: 0.05679187923669815 2023-01-21 11:11:58.084872: step: 728/526, loss: 0.000560760498046875 2023-01-21 11:11:59.237822: step: 732/526, loss: 0.026170063763856888 2023-01-21 11:12:00.414970: step: 736/526, loss: 0.031717684119939804 2023-01-21 11:12:01.552700: step: 740/526, loss: 0.04638538509607315 2023-01-21 11:12:02.697026: step: 744/526, loss: 0.007654215674847364 2023-01-21 11:12:03.847445: step: 748/526, loss: 0.016956090927124023 2023-01-21 11:12:05.029734: step: 752/526, loss: 0.04389238357543945 2023-01-21 11:12:06.198336: step: 756/526, loss: 0.00855178851634264 2023-01-21 11:12:07.340045: step: 760/526, loss: 0.034254275262355804 2023-01-21 11:12:08.501221: step: 764/526, loss: 0.14534887671470642 2023-01-21 11:12:09.682987: step: 768/526, loss: 0.008438730612397194 2023-01-21 11:12:10.866966: step: 772/526, loss: 0.00020227432833053172 2023-01-21 11:12:12.016889: step: 776/526, loss: 0.025569917634129524 2023-01-21 11:12:13.163535: step: 780/526, loss: 0.06152515113353729 2023-01-21 11:12:14.271071: step: 784/526, loss: 0.0010889529949054122 2023-01-21 11:12:15.437710: step: 788/526, loss: 0.01145782507956028 2023-01-21 11:12:16.605764: step: 792/526, loss: 0.00215320591814816 2023-01-21 11:12:17.740515: step: 796/526, loss: 0.019379330798983574 2023-01-21 11:12:18.883527: step: 800/526, loss: 0.009414101019501686 2023-01-21 11:12:20.045732: step: 804/526, loss: 0.0415528304874897 2023-01-21 11:12:21.207062: step: 808/526, loss: 0.005959892645478249 2023-01-21 11:12:22.446235: step: 812/526, loss: 0.015486336313188076 2023-01-21 11:12:23.586831: step: 816/526, loss: 0.3148905634880066 2023-01-21 11:12:24.769362: step: 820/526, loss: 0.006985187530517578 2023-01-21 11:12:25.923441: step: 824/526, loss: 0.0011864661937579513 2023-01-21 11:12:27.068627: step: 828/526, loss: 0.007863283157348633 2023-01-21 11:12:28.233623: step: 832/526, loss: 0.0014211655361577868 2023-01-21 11:12:29.370453: step: 836/526, loss: 0.031378746032714844 2023-01-21 11:12:30.601416: step: 840/526, loss: 0.0496830940246582 2023-01-21 11:12:31.763308: step: 844/526, loss: 0.026904677972197533 2023-01-21 11:12:32.911993: step: 848/526, loss: 0.006833505816757679 2023-01-21 11:12:34.052538: step: 852/526, loss: 0.005734109785407782 2023-01-21 11:12:35.231544: step: 856/526, loss: 0.017026234418153763 2023-01-21 11:12:36.395810: step: 860/526, loss: 0.008176708593964577 2023-01-21 11:12:37.543177: step: 864/526, loss: 0.005719518288969994 2023-01-21 11:12:38.687436: step: 868/526, loss: 0.014420128427445889 2023-01-21 11:12:39.811646: step: 872/526, loss: 0.052733805030584335 2023-01-21 11:12:40.961266: step: 876/526, loss: 0.015298079699277878 2023-01-21 11:12:42.103355: step: 880/526, loss: 0.05374488979578018 2023-01-21 11:12:43.294235: step: 884/526, loss: 0.6681743860244751 2023-01-21 11:12:44.449111: step: 888/526, loss: 0.00038733481778763235 2023-01-21 11:12:45.586866: step: 892/526, loss: 0.011528587900102139 2023-01-21 11:12:46.718400: step: 896/526, loss: 0.006939506158232689 2023-01-21 11:12:47.865848: step: 900/526, loss: 0.06101560965180397 2023-01-21 11:12:49.005926: step: 904/526, loss: 0.0038442614022642374 2023-01-21 11:12:50.137236: step: 908/526, loss: 0.08767890930175781 2023-01-21 11:12:51.290376: step: 912/526, loss: 0.07067527621984482 2023-01-21 11:12:52.421062: step: 916/526, loss: 0.06301408261060715 2023-01-21 11:12:53.605064: step: 920/526, loss: 0.05525340884923935 2023-01-21 11:12:54.744320: step: 924/526, loss: 0.03001587465405464 2023-01-21 11:12:55.873482: step: 928/526, loss: 0.05066210404038429 2023-01-21 11:12:56.992934: step: 932/526, loss: 0.09693536907434464 2023-01-21 11:12:58.123852: step: 936/526, loss: 0.01844949647784233 2023-01-21 11:12:59.251380: step: 940/526, loss: 0.001082134316675365 2023-01-21 11:13:00.389275: step: 944/526, loss: 0.0005603790050372481 2023-01-21 11:13:01.554309: step: 948/526, loss: 0.03271665796637535 2023-01-21 11:13:02.687447: step: 952/526, loss: 0.07516960054636002 2023-01-21 11:13:03.858784: step: 956/526, loss: 0.02696523629128933 2023-01-21 11:13:05.004608: step: 960/526, loss: 0.07556271553039551 2023-01-21 11:13:06.169342: step: 964/526, loss: 0.05851192772388458 2023-01-21 11:13:07.321497: step: 968/526, loss: 0.08257904648780823 2023-01-21 11:13:08.501674: step: 972/526, loss: 0.0008014679187908769 2023-01-21 11:13:09.646514: step: 976/526, loss: 0.47696352005004883 2023-01-21 11:13:10.815950: step: 980/526, loss: 0.0007006645319052041 2023-01-21 11:13:11.978719: step: 984/526, loss: 0.011383056640625 2023-01-21 11:13:13.111430: step: 988/526, loss: 0.00690307654440403 2023-01-21 11:13:14.244738: step: 992/526, loss: 0.031540773808956146 2023-01-21 11:13:15.394700: step: 996/526, loss: 0.02640690840780735 2023-01-21 11:13:16.567486: step: 1000/526, loss: 0.015194892883300781 2023-01-21 11:13:17.728552: step: 1004/526, loss: 0.024719620123505592 2023-01-21 11:13:18.872107: step: 1008/526, loss: 0.005755424965173006 2023-01-21 11:13:20.023351: step: 1012/526, loss: 0.02340087853372097 2023-01-21 11:13:21.179463: step: 1016/526, loss: 0.0287641528993845 2023-01-21 11:13:22.361630: step: 1020/526, loss: 0.0001539230433991179 2023-01-21 11:13:23.494470: step: 1024/526, loss: 0.00240249652415514 2023-01-21 11:13:24.636871: step: 1028/526, loss: 0.009704399853944778 2023-01-21 11:13:25.791163: step: 1032/526, loss: 0.008310128003358841 2023-01-21 11:13:26.938359: step: 1036/526, loss: 0.016400909051299095 2023-01-21 11:13:28.087326: step: 1040/526, loss: 0.01180114783346653 2023-01-21 11:13:29.230428: step: 1044/526, loss: 0.08272266387939453 2023-01-21 11:13:30.381983: step: 1048/526, loss: 0.021078776568174362 2023-01-21 11:13:31.574403: step: 1052/526, loss: 0.009411525912582874 2023-01-21 11:13:32.718240: step: 1056/526, loss: 0.11165113747119904 2023-01-21 11:13:33.867094: step: 1060/526, loss: 0.05851578712463379 2023-01-21 11:13:35.016586: step: 1064/526, loss: 0.06337108463048935 2023-01-21 11:13:36.194177: step: 1068/526, loss: 0.05727062374353409 2023-01-21 11:13:37.311206: step: 1072/526, loss: 0.01905827596783638 2023-01-21 11:13:38.445082: step: 1076/526, loss: 0.09094247221946716 2023-01-21 11:13:39.572523: step: 1080/526, loss: 0.007142829708755016 2023-01-21 11:13:40.732536: step: 1084/526, loss: 0.012323188595473766 2023-01-21 11:13:41.899903: step: 1088/526, loss: 0.027137089520692825 2023-01-21 11:13:43.063089: step: 1092/526, loss: 0.04914169758558273 2023-01-21 11:13:44.197559: step: 1096/526, loss: 0.00014047622971702367 2023-01-21 11:13:45.368666: step: 1100/526, loss: 0.020443536341190338 2023-01-21 11:13:46.553335: step: 1104/526, loss: 0.0012463569873943925 2023-01-21 11:13:47.692894: step: 1108/526, loss: 0.11151676625013351 2023-01-21 11:13:48.837497: step: 1112/526, loss: 0.03851685672998428 2023-01-21 11:13:50.010476: step: 1116/526, loss: 0.05852337181568146 2023-01-21 11:13:51.164410: step: 1120/526, loss: 0.017716217786073685 2023-01-21 11:13:52.340585: step: 1124/526, loss: 0.002372789429500699 2023-01-21 11:13:53.479230: step: 1128/526, loss: 0.020529747009277344 2023-01-21 11:13:54.635361: step: 1132/526, loss: 0.00903086643666029 2023-01-21 11:13:55.805438: step: 1136/526, loss: 0.01313171349465847 2023-01-21 11:13:56.990545: step: 1140/526, loss: 0.010787487961351871 2023-01-21 11:13:58.196682: step: 1144/526, loss: 0.003053951309993863 2023-01-21 11:13:59.360397: step: 1148/526, loss: 0.017096709460020065 2023-01-21 11:14:00.510805: step: 1152/526, loss: 0.00044465065002441406 2023-01-21 11:14:01.696750: step: 1156/526, loss: 0.008677482604980469 2023-01-21 11:14:02.864467: step: 1160/526, loss: 0.012208365835249424 2023-01-21 11:14:03.977830: step: 1164/526, loss: 0.021317673847079277 2023-01-21 11:14:05.147450: step: 1168/526, loss: 0.009976488538086414 2023-01-21 11:14:06.307048: step: 1172/526, loss: 0.005944728851318359 2023-01-21 11:14:07.478377: step: 1176/526, loss: 0.008188724517822266 2023-01-21 11:14:08.643083: step: 1180/526, loss: 0.0025188445579260588 2023-01-21 11:14:09.765836: step: 1184/526, loss: 0.011735345236957073 2023-01-21 11:14:10.933490: step: 1188/526, loss: 0.022648334503173828 2023-01-21 11:14:12.063609: step: 1192/526, loss: 0.00027484894962981343 2023-01-21 11:14:13.205541: step: 1196/526, loss: 0.09759931266307831 2023-01-21 11:14:14.368726: step: 1200/526, loss: 0.00784306600689888 2023-01-21 11:14:15.544321: step: 1204/526, loss: 0.0034612659364938736 2023-01-21 11:14:16.711980: step: 1208/526, loss: 0.007220268249511719 2023-01-21 11:14:17.889650: step: 1212/526, loss: 0.0019939423073083162 2023-01-21 11:14:19.007610: step: 1216/526, loss: 0.053132928907871246 2023-01-21 11:14:20.161273: step: 1220/526, loss: 0.01638355292379856 2023-01-21 11:14:21.315604: step: 1224/526, loss: 0.13880720734596252 2023-01-21 11:14:22.460394: step: 1228/526, loss: 0.00272884382866323 2023-01-21 11:14:23.591856: step: 1232/526, loss: 0.07232251763343811 2023-01-21 11:14:24.767571: step: 1236/526, loss: 0.028873920440673828 2023-01-21 11:14:25.922908: step: 1240/526, loss: 0.0020033835899084806 2023-01-21 11:14:27.060330: step: 1244/526, loss: 0.03896808624267578 2023-01-21 11:14:28.192487: step: 1248/526, loss: 0.07329349964857101 2023-01-21 11:14:29.345450: step: 1252/526, loss: 0.016579627990722656 2023-01-21 11:14:30.515103: step: 1256/526, loss: 0.05246315151453018 2023-01-21 11:14:31.648668: step: 1260/526, loss: 0.05313308537006378 2023-01-21 11:14:32.792714: step: 1264/526, loss: 0.14537039399147034 2023-01-21 11:14:33.936163: step: 1268/526, loss: 0.011049938388168812 2023-01-21 11:14:35.117349: step: 1272/526, loss: 0.03732319176197052 2023-01-21 11:14:36.279304: step: 1276/526, loss: 0.06474075466394424 2023-01-21 11:14:37.447960: step: 1280/526, loss: 0.13269072771072388 2023-01-21 11:14:38.599824: step: 1284/526, loss: 0.07875613868236542 2023-01-21 11:14:39.772063: step: 1288/526, loss: 0.025364112108945847 2023-01-21 11:14:40.921458: step: 1292/526, loss: 0.012532235123217106 2023-01-21 11:14:42.062455: step: 1296/526, loss: 0.03448314592242241 2023-01-21 11:14:43.209962: step: 1300/526, loss: 0.01663036271929741 2023-01-21 11:14:44.347147: step: 1304/526, loss: 0.015280676074326038 2023-01-21 11:14:45.495528: step: 1308/526, loss: 0.001991367433220148 2023-01-21 11:14:46.611531: step: 1312/526, loss: 0.0353623628616333 2023-01-21 11:14:47.754457: step: 1316/526, loss: 0.0299238208681345 2023-01-21 11:14:48.902256: step: 1320/526, loss: 0.011005115695297718 2023-01-21 11:14:50.035462: step: 1324/526, loss: 0.07054109871387482 2023-01-21 11:14:51.197124: step: 1328/526, loss: 0.06421032547950745 2023-01-21 11:14:52.331612: step: 1332/526, loss: 0.012379264459013939 2023-01-21 11:14:53.491947: step: 1336/526, loss: 0.014616680331528187 2023-01-21 11:14:54.638366: step: 1340/526, loss: 0.02322368696331978 2023-01-21 11:14:55.827212: step: 1344/526, loss: 0.016550827771425247 2023-01-21 11:14:56.999623: step: 1348/526, loss: 0.017902422696352005 2023-01-21 11:14:58.117401: step: 1352/526, loss: 0.005248356144875288 2023-01-21 11:14:59.245110: step: 1356/526, loss: 0.013132667168974876 2023-01-21 11:15:00.368919: step: 1360/526, loss: 0.00671000499278307 2023-01-21 11:15:01.531880: step: 1364/526, loss: 0.023415088653564453 2023-01-21 11:15:02.682523: step: 1368/526, loss: 0.0005354405147954822 2023-01-21 11:15:03.802037: step: 1372/526, loss: 0.00042765142279677093 2023-01-21 11:15:04.937983: step: 1376/526, loss: 0.015307044610381126 2023-01-21 11:15:06.060486: step: 1380/526, loss: 0.18836478888988495 2023-01-21 11:15:07.204788: step: 1384/526, loss: 0.001861572265625 2023-01-21 11:15:08.384066: step: 1388/526, loss: 0.028600502759218216 2023-01-21 11:15:09.521598: step: 1392/526, loss: 0.004447364713996649 2023-01-21 11:15:10.660788: step: 1396/526, loss: 0.07432775944471359 2023-01-21 11:15:11.775169: step: 1400/526, loss: 0.01283111609518528 2023-01-21 11:15:12.923520: step: 1404/526, loss: 0.03626260906457901 2023-01-21 11:15:14.036212: step: 1408/526, loss: 0.03151760250329971 2023-01-21 11:15:15.184809: step: 1412/526, loss: 0.002002716064453125 2023-01-21 11:15:16.303153: step: 1416/526, loss: 0.09061814099550247 2023-01-21 11:15:17.454772: step: 1420/526, loss: 0.03536491468548775 2023-01-21 11:15:18.582715: step: 1424/526, loss: 0.01896655559539795 2023-01-21 11:15:19.716733: step: 1428/526, loss: 0.14170297980308533 2023-01-21 11:15:20.847426: step: 1432/526, loss: 0.0007655143854208291 2023-01-21 11:15:21.994990: step: 1436/526, loss: 0.06198524683713913 2023-01-21 11:15:23.131778: step: 1440/526, loss: 0.008537674322724342 2023-01-21 11:15:24.275663: step: 1444/526, loss: 0.022623158991336823 2023-01-21 11:15:25.414104: step: 1448/526, loss: 0.05109348148107529 2023-01-21 11:15:26.584244: step: 1452/526, loss: 0.012701607309281826 2023-01-21 11:15:27.708034: step: 1456/526, loss: 0.003014135407283902 2023-01-21 11:15:28.856036: step: 1460/526, loss: 0.0011701583862304688 2023-01-21 11:15:29.995458: step: 1464/526, loss: 0.0027844428550451994 2023-01-21 11:15:31.133754: step: 1468/526, loss: 0.0018871307838708162 2023-01-21 11:15:32.275866: step: 1472/526, loss: 0.0028310774359852076 2023-01-21 11:15:33.413546: step: 1476/526, loss: 0.0021018029656261206 2023-01-21 11:15:34.550827: step: 1480/526, loss: 0.002915668534114957 2023-01-21 11:15:35.715560: step: 1484/526, loss: 0.09278788417577744 2023-01-21 11:15:36.904493: step: 1488/526, loss: 0.03566303104162216 2023-01-21 11:15:38.045054: step: 1492/526, loss: 0.007505846209824085 2023-01-21 11:15:39.204052: step: 1496/526, loss: 0.0024461746215820312 2023-01-21 11:15:40.357306: step: 1500/526, loss: 0.044089317321777344 2023-01-21 11:15:41.504367: step: 1504/526, loss: 0.000953292881604284 2023-01-21 11:15:42.662756: step: 1508/526, loss: 0.0310229305177927 2023-01-21 11:15:43.799966: step: 1512/526, loss: 0.0328710600733757 2023-01-21 11:15:44.955592: step: 1516/526, loss: 0.0010725974570959806 2023-01-21 11:15:46.117230: step: 1520/526, loss: 0.5288132429122925 2023-01-21 11:15:47.275850: step: 1524/526, loss: 0.07509537041187286 2023-01-21 11:15:48.435337: step: 1528/526, loss: 0.023479079827666283 2023-01-21 11:15:49.580643: step: 1532/526, loss: 0.0011591911315917969 2023-01-21 11:15:50.750584: step: 1536/526, loss: 0.053659629076719284 2023-01-21 11:15:51.921975: step: 1540/526, loss: 0.05722217634320259 2023-01-21 11:15:53.067368: step: 1544/526, loss: 0.02186732180416584 2023-01-21 11:15:54.246236: step: 1548/526, loss: 0.07524510473012924 2023-01-21 11:15:55.401874: step: 1552/526, loss: 0.032151415944099426 2023-01-21 11:15:56.584135: step: 1556/526, loss: 0.012040138244628906 2023-01-21 11:15:57.769188: step: 1560/526, loss: 0.018299389630556107 2023-01-21 11:15:58.937151: step: 1564/526, loss: 0.01961848884820938 2023-01-21 11:16:00.100506: step: 1568/526, loss: 0.03396425396203995 2023-01-21 11:16:01.267177: step: 1572/526, loss: 0.016263389959931374 2023-01-21 11:16:02.431837: step: 1576/526, loss: 0.012240070849657059 2023-01-21 11:16:03.564202: step: 1580/526, loss: 0.006938362028449774 2023-01-21 11:16:04.697953: step: 1584/526, loss: 0.012159060686826706 2023-01-21 11:16:05.842717: step: 1588/526, loss: 0.0010875702137127519 2023-01-21 11:16:06.981831: step: 1592/526, loss: 0.00943145714700222 2023-01-21 11:16:08.140286: step: 1596/526, loss: 0.021221160888671875 2023-01-21 11:16:09.322135: step: 1600/526, loss: 0.02083129808306694 2023-01-21 11:16:10.453666: step: 1604/526, loss: 0.0012420654529705644 2023-01-21 11:16:11.620286: step: 1608/526, loss: 0.025774193927645683 2023-01-21 11:16:12.757170: step: 1612/526, loss: 0.00968628004193306 2023-01-21 11:16:13.885639: step: 1616/526, loss: 0.07935958355665207 2023-01-21 11:16:15.057181: step: 1620/526, loss: 0.07276764512062073 2023-01-21 11:16:16.221723: step: 1624/526, loss: 0.031004145741462708 2023-01-21 11:16:17.413766: step: 1628/526, loss: 0.04893860965967178 2023-01-21 11:16:18.566260: step: 1632/526, loss: 0.02863292768597603 2023-01-21 11:16:19.735115: step: 1636/526, loss: 0.012957191094756126 2023-01-21 11:16:20.891674: step: 1640/526, loss: 0.012598990462720394 2023-01-21 11:16:22.064348: step: 1644/526, loss: 0.011762047186493874 2023-01-21 11:16:23.187533: step: 1648/526, loss: 0.003974628169089556 2023-01-21 11:16:24.316574: step: 1652/526, loss: 0.1054234504699707 2023-01-21 11:16:25.506177: step: 1656/526, loss: 0.03522396460175514 2023-01-21 11:16:26.655023: step: 1660/526, loss: 0.004057598300278187 2023-01-21 11:16:27.809156: step: 1664/526, loss: 0.04009886085987091 2023-01-21 11:16:28.937583: step: 1668/526, loss: 0.029004957526922226 2023-01-21 11:16:30.106764: step: 1672/526, loss: 0.029472827911376953 2023-01-21 11:16:31.288778: step: 1676/526, loss: 0.021584415808320045 2023-01-21 11:16:32.449905: step: 1680/526, loss: 0.06130675971508026 2023-01-21 11:16:33.591630: step: 1684/526, loss: 0.014807415194809437 2023-01-21 11:16:34.753681: step: 1688/526, loss: 0.07389964908361435 2023-01-21 11:16:35.930825: step: 1692/526, loss: 0.045704230666160583 2023-01-21 11:16:37.067820: step: 1696/526, loss: 0.07203865051269531 2023-01-21 11:16:38.250032: step: 1700/526, loss: 0.027575302869081497 2023-01-21 11:16:39.415338: step: 1704/526, loss: 0.007754468824714422 2023-01-21 11:16:40.591823: step: 1708/526, loss: 0.32650527358055115 2023-01-21 11:16:41.745462: step: 1712/526, loss: 0.05932359769940376 2023-01-21 11:16:42.913137: step: 1716/526, loss: 0.11338011920452118 2023-01-21 11:16:44.047208: step: 1720/526, loss: 0.03401365503668785 2023-01-21 11:16:45.204012: step: 1724/526, loss: 0.026148034259676933 2023-01-21 11:16:46.402141: step: 1728/526, loss: 0.018466949462890625 2023-01-21 11:16:47.585081: step: 1732/526, loss: 0.04690499231219292 2023-01-21 11:16:48.736953: step: 1736/526, loss: 0.0029813051223754883 2023-01-21 11:16:49.878747: step: 1740/526, loss: 0.11627130210399628 2023-01-21 11:16:51.049483: step: 1744/526, loss: 0.6963399648666382 2023-01-21 11:16:52.208185: step: 1748/526, loss: 0.02696824073791504 2023-01-21 11:16:53.359558: step: 1752/526, loss: 0.04826998710632324 2023-01-21 11:16:54.492499: step: 1756/526, loss: 0.10515594482421875 2023-01-21 11:16:55.666508: step: 1760/526, loss: 0.015928268432617188 2023-01-21 11:16:56.812440: step: 1764/526, loss: 0.007394728250801563 2023-01-21 11:16:57.954357: step: 1768/526, loss: 0.02510528452694416 2023-01-21 11:16:59.112539: step: 1772/526, loss: 0.006548500154167414 2023-01-21 11:17:00.269035: step: 1776/526, loss: 0.03842916339635849 2023-01-21 11:17:01.437928: step: 1780/526, loss: 0.011109876446425915 2023-01-21 11:17:02.625361: step: 1784/526, loss: 0.05668430030345917 2023-01-21 11:17:03.750796: step: 1788/526, loss: 0.14935742318630219 2023-01-21 11:17:04.903020: step: 1792/526, loss: 0.05277585983276367 2023-01-21 11:17:06.062650: step: 1796/526, loss: 0.0009180069318972528 2023-01-21 11:17:07.187119: step: 1800/526, loss: 0.0048984051682055 2023-01-21 11:17:08.332942: step: 1804/526, loss: 0.022557832300662994 2023-01-21 11:17:09.503240: step: 1808/526, loss: 0.024224378168582916 2023-01-21 11:17:10.626440: step: 1812/526, loss: 0.0031709671020507812 2023-01-21 11:17:11.749429: step: 1816/526, loss: 0.006091499235481024 2023-01-21 11:17:12.918908: step: 1820/526, loss: 0.3853173553943634 2023-01-21 11:17:14.066562: step: 1824/526, loss: 0.0038299560546875 2023-01-21 11:17:15.237245: step: 1828/526, loss: 0.029605364426970482 2023-01-21 11:17:16.381999: step: 1832/526, loss: 0.017946625128388405 2023-01-21 11:17:17.587017: step: 1836/526, loss: 0.0021326064597815275 2023-01-21 11:17:18.718813: step: 1840/526, loss: 0.004783248528838158 2023-01-21 11:17:19.881222: step: 1844/526, loss: 0.004403305239975452 2023-01-21 11:17:21.044746: step: 1848/526, loss: 0.769024670124054 2023-01-21 11:17:22.199429: step: 1852/526, loss: 0.24156780540943146 2023-01-21 11:17:23.348914: step: 1856/526, loss: 0.006679534912109375 2023-01-21 11:17:24.521868: step: 1860/526, loss: 0.3477468490600586 2023-01-21 11:17:25.707064: step: 1864/526, loss: 0.0831918716430664 2023-01-21 11:17:26.862965: step: 1868/526, loss: 0.0034605979453772306 2023-01-21 11:17:28.034031: step: 1872/526, loss: 0.05076313391327858 2023-01-21 11:17:29.163394: step: 1876/526, loss: 0.011259890161454678 2023-01-21 11:17:30.336029: step: 1880/526, loss: 0.016239548102021217 2023-01-21 11:17:31.463258: step: 1884/526, loss: 0.07064676284790039 2023-01-21 11:17:32.597987: step: 1888/526, loss: 0.0006339550018310547 2023-01-21 11:17:33.775257: step: 1892/526, loss: 0.031766317784786224 2023-01-21 11:17:34.928689: step: 1896/526, loss: 0.0023015975020825863 2023-01-21 11:17:36.125286: step: 1900/526, loss: 0.0020677566062659025 2023-01-21 11:17:37.265410: step: 1904/526, loss: 0.01207809429615736 2023-01-21 11:17:38.427493: step: 1908/526, loss: 0.1251041442155838 2023-01-21 11:17:39.575417: step: 1912/526, loss: 0.03783874586224556 2023-01-21 11:17:40.720009: step: 1916/526, loss: 0.011644363403320312 2023-01-21 11:17:41.857169: step: 1920/526, loss: 0.08902013301849365 2023-01-21 11:17:43.026626: step: 1924/526, loss: 0.0004311561642680317 2023-01-21 11:17:44.245745: step: 1928/526, loss: 0.10454969108104706 2023-01-21 11:17:45.391345: step: 1932/526, loss: 0.011892891488969326 2023-01-21 11:17:46.548539: step: 1936/526, loss: 0.029772471636533737 2023-01-21 11:17:47.698085: step: 1940/526, loss: 0.06390085071325302 2023-01-21 11:17:48.827327: step: 1944/526, loss: 0.015054702758789062 2023-01-21 11:17:49.982703: step: 1948/526, loss: 0.012243938632309437 2023-01-21 11:17:51.174165: step: 1952/526, loss: 0.4398822784423828 2023-01-21 11:17:52.321036: step: 1956/526, loss: 0.08239660412073135 2023-01-21 11:17:53.487902: step: 1960/526, loss: 0.0768928974866867 2023-01-21 11:17:54.639124: step: 1964/526, loss: 0.0029317857697606087 2023-01-21 11:17:55.772289: step: 1968/526, loss: 0.0447535514831543 2023-01-21 11:17:56.942681: step: 1972/526, loss: 0.043679095804691315 2023-01-21 11:17:58.082340: step: 1976/526, loss: 0.05214405059814453 2023-01-21 11:17:59.248950: step: 1980/526, loss: 0.058764077723026276 2023-01-21 11:18:00.409127: step: 1984/526, loss: 0.00429534912109375 2023-01-21 11:18:01.565917: step: 1988/526, loss: 0.033416081219911575 2023-01-21 11:18:02.738419: step: 1992/526, loss: 0.005925274454057217 2023-01-21 11:18:03.870572: step: 1996/526, loss: 0.041114043444395065 2023-01-21 11:18:05.052579: step: 2000/526, loss: 0.11051063239574432 2023-01-21 11:18:06.238689: step: 2004/526, loss: 0.031624604016542435 2023-01-21 11:18:07.391090: step: 2008/526, loss: 0.037841081619262695 2023-01-21 11:18:08.549449: step: 2012/526, loss: 0.008747626096010208 2023-01-21 11:18:09.728515: step: 2016/526, loss: 0.019548749551177025 2023-01-21 11:18:10.889322: step: 2020/526, loss: 0.020754002034664154 2023-01-21 11:18:12.042685: step: 2024/526, loss: 0.02051960863173008 2023-01-21 11:18:13.212170: step: 2028/526, loss: 0.13297158479690552 2023-01-21 11:18:14.345099: step: 2032/526, loss: 0.0016053318977355957 2023-01-21 11:18:15.501694: step: 2036/526, loss: 0.011286449618637562 2023-01-21 11:18:16.642979: step: 2040/526, loss: 0.03196869045495987 2023-01-21 11:18:17.817061: step: 2044/526, loss: 0.0447598472237587 2023-01-21 11:18:18.962834: step: 2048/526, loss: 0.0063200946897268295 2023-01-21 11:18:20.121958: step: 2052/526, loss: 0.0020450593437999487 2023-01-21 11:18:21.270662: step: 2056/526, loss: 0.1256854087114334 2023-01-21 11:18:22.435852: step: 2060/526, loss: 0.0715525671839714 2023-01-21 11:18:23.559491: step: 2064/526, loss: 0.001021099160425365 2023-01-21 11:18:24.695480: step: 2068/526, loss: 0.04251861572265625 2023-01-21 11:18:25.828008: step: 2072/526, loss: 0.0840839371085167 2023-01-21 11:18:26.951921: step: 2076/526, loss: 0.1330556869506836 2023-01-21 11:18:28.105456: step: 2080/526, loss: 0.09395141899585724 2023-01-21 11:18:29.250121: step: 2084/526, loss: 0.025026513263583183 2023-01-21 11:18:30.385506: step: 2088/526, loss: 0.07811050117015839 2023-01-21 11:18:31.571898: step: 2092/526, loss: 0.015866661444306374 2023-01-21 11:18:32.706290: step: 2096/526, loss: 0.014677238650619984 2023-01-21 11:18:33.818989: step: 2100/526, loss: 0.05850258469581604 2023-01-21 11:18:34.991253: step: 2104/526, loss: 0.012905311770737171 ================================================== Loss: 0.044 -------------------- Dev: {'event': {'p': 0.596757852077001, 'r': 0.7842876165113183, 'f1': 0.6777905638665133}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Test: {'event': {'p': 0.6237995824634656, 'r': 0.7797494780793319, 'f1': 0.6931106471816285}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Chinese: {'event': {'p': 0.5903614457831325, 'r': 0.9074074074074074, 'f1': 0.7153284671532847}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Korean: {'event': {'p': 0.7073170731707317, 'r': 0.4603174603174603, 'f1': 0.5576923076923077}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} Russian: {'event': {'p': 0.4146341463414634, 'r': 0.4722222222222222, 'f1': 0.4415584415584415}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 14} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 15 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:19:18.832459: step: 4/526, loss: 0.015715409070253372 2023-01-21 11:19:19.954156: step: 8/526, loss: 0.03385467454791069 2023-01-21 11:19:21.083964: step: 12/526, loss: 0.09915709495544434 2023-01-21 11:19:22.213815: step: 16/526, loss: 0.011584281921386719 2023-01-21 11:19:23.383186: step: 20/526, loss: 8.7738037109375e-05 2023-01-21 11:19:24.501699: step: 24/526, loss: 0.0006498337024822831 2023-01-21 11:19:25.650192: step: 28/526, loss: 0.013985633850097656 2023-01-21 11:19:26.773031: step: 32/526, loss: 0.020456504076719284 2023-01-21 11:19:27.914787: step: 36/526, loss: 0.008350754156708717 2023-01-21 11:19:29.042763: step: 40/526, loss: 0.026631547138094902 2023-01-21 11:19:30.164160: step: 44/526, loss: 0.058612849563360214 2023-01-21 11:19:31.292957: step: 48/526, loss: 0.018627453595399857 2023-01-21 11:19:32.457415: step: 52/526, loss: 0.07881756126880646 2023-01-21 11:19:33.668816: step: 56/526, loss: 0.011064434424042702 2023-01-21 11:19:34.813886: step: 60/526, loss: 0.027622507885098457 2023-01-21 11:19:36.002964: step: 64/526, loss: 0.02246999740600586 2023-01-21 11:19:37.153533: step: 68/526, loss: 0.007797289174050093 2023-01-21 11:19:38.319282: step: 72/526, loss: 0.000351142865838483 2023-01-21 11:19:39.456595: step: 76/526, loss: 0.07748118042945862 2023-01-21 11:19:40.624282: step: 80/526, loss: 0.006552505772560835 2023-01-21 11:19:41.778443: step: 84/526, loss: 0.032444000244140625 2023-01-21 11:19:42.926638: step: 88/526, loss: 0.002372264862060547 2023-01-21 11:19:44.065260: step: 92/526, loss: 0.06604146957397461 2023-01-21 11:19:45.218138: step: 96/526, loss: 0.006347847171127796 2023-01-21 11:19:46.372690: step: 100/526, loss: 0.03308863937854767 2023-01-21 11:19:47.491683: step: 104/526, loss: 0.011268901638686657 2023-01-21 11:19:48.657303: step: 108/526, loss: 0.049483153969049454 2023-01-21 11:19:49.803476: step: 112/526, loss: 0.012888908386230469 2023-01-21 11:19:50.958389: step: 116/526, loss: 0.06288042664527893 2023-01-21 11:19:52.143361: step: 120/526, loss: 0.006939411163330078 2023-01-21 11:19:53.316122: step: 124/526, loss: 0.017635727301239967 2023-01-21 11:19:54.486534: step: 128/526, loss: 0.0018420219421386719 2023-01-21 11:19:55.627585: step: 132/526, loss: 0.009213065728545189 2023-01-21 11:19:56.767020: step: 136/526, loss: 0.035201214253902435 2023-01-21 11:19:57.966476: step: 140/526, loss: 0.025753404945135117 2023-01-21 11:19:59.116282: step: 144/526, loss: 0.0016251562628895044 2023-01-21 11:20:00.286511: step: 148/526, loss: 0.12415160983800888 2023-01-21 11:20:01.436195: step: 152/526, loss: 0.007700157351791859 2023-01-21 11:20:02.578245: step: 156/526, loss: 0.00010375976853538305 2023-01-21 11:20:03.720513: step: 160/526, loss: 0.019460201263427734 2023-01-21 11:20:04.851222: step: 164/526, loss: 0.0004245758173055947 2023-01-21 11:20:06.021368: step: 168/526, loss: 0.02017951011657715 2023-01-21 11:20:07.160448: step: 172/526, loss: 0.07349959015846252 2023-01-21 11:20:08.320843: step: 176/526, loss: 0.01844015158712864 2023-01-21 11:20:09.495544: step: 180/526, loss: 0.032985687255859375 2023-01-21 11:20:10.668760: step: 184/526, loss: 0.008691406808793545 2023-01-21 11:20:11.848058: step: 188/526, loss: 0.0008455276256427169 2023-01-21 11:20:13.007175: step: 192/526, loss: 0.013279153034090996 2023-01-21 11:20:14.183887: step: 196/526, loss: 0.011546134948730469 2023-01-21 11:20:15.340931: step: 200/526, loss: 0.0751718059182167 2023-01-21 11:20:16.484554: step: 204/526, loss: 0.01450204849243164 2023-01-21 11:20:17.621237: step: 208/526, loss: 0.037872314453125 2023-01-21 11:20:18.758418: step: 212/526, loss: 0.009087909013032913 2023-01-21 11:20:19.887912: step: 216/526, loss: 0.010727119632065296 2023-01-21 11:20:21.032632: step: 220/526, loss: 0.0007208824390545487 2023-01-21 11:20:22.182353: step: 224/526, loss: 0.0003113746643066406 2023-01-21 11:20:23.350911: step: 228/526, loss: 0.001144456909969449 2023-01-21 11:20:24.520495: step: 232/526, loss: 0.004671192262321711 2023-01-21 11:20:25.707766: step: 236/526, loss: 0.007922370918095112 2023-01-21 11:20:26.852559: step: 240/526, loss: 0.015404893085360527 2023-01-21 11:20:28.004508: step: 244/526, loss: 0.001066494034603238 2023-01-21 11:20:29.138320: step: 248/526, loss: 0.04539012908935547 2023-01-21 11:20:30.291771: step: 252/526, loss: 0.021883105859160423 2023-01-21 11:20:31.452931: step: 256/526, loss: 0.0025157928466796875 2023-01-21 11:20:32.611315: step: 260/526, loss: 0.008914184756577015 2023-01-21 11:20:33.751000: step: 264/526, loss: 0.014430428855121136 2023-01-21 11:20:34.919116: step: 268/526, loss: 0.008715915493667126 2023-01-21 11:20:36.064477: step: 272/526, loss: 0.03676197677850723 2023-01-21 11:20:37.191614: step: 276/526, loss: 0.0006800651317462325 2023-01-21 11:20:38.336629: step: 280/526, loss: 0.025012493133544922 2023-01-21 11:20:39.482374: step: 284/526, loss: 0.035610102117061615 2023-01-21 11:20:40.659531: step: 288/526, loss: 0.008062934502959251 2023-01-21 11:20:41.798231: step: 292/526, loss: 0.0018789292080327868 2023-01-21 11:20:42.942161: step: 296/526, loss: 0.011466551572084427 2023-01-21 11:20:44.102848: step: 300/526, loss: 0.13480329513549805 2023-01-21 11:20:45.247934: step: 304/526, loss: 0.01765470579266548 2023-01-21 11:20:46.374118: step: 308/526, loss: 0.059966277331113815 2023-01-21 11:20:47.509596: step: 312/526, loss: 0.012594032101333141 2023-01-21 11:20:48.658112: step: 316/526, loss: 0.030475234612822533 2023-01-21 11:20:49.815605: step: 320/526, loss: 0.004731560125946999 2023-01-21 11:20:50.972502: step: 324/526, loss: 0.002152061555534601 2023-01-21 11:20:52.157490: step: 328/526, loss: 0.012895298190414906 2023-01-21 11:20:53.324513: step: 332/526, loss: 0.6520692706108093 2023-01-21 11:20:54.454258: step: 336/526, loss: 0.04438934475183487 2023-01-21 11:20:55.615388: step: 340/526, loss: 0.03441943973302841 2023-01-21 11:20:56.807877: step: 344/526, loss: 0.041034795343875885 2023-01-21 11:20:57.964589: step: 348/526, loss: 0.014559936709702015 2023-01-21 11:20:59.110978: step: 352/526, loss: 0.06233549118041992 2023-01-21 11:21:00.290150: step: 356/526, loss: 0.016116047278046608 2023-01-21 11:21:01.422855: step: 360/526, loss: 0.04400377348065376 2023-01-21 11:21:02.621596: step: 364/526, loss: 0.003484535263851285 2023-01-21 11:21:03.740326: step: 368/526, loss: 0.0056514739990234375 2023-01-21 11:21:04.904025: step: 372/526, loss: 0.011733055114746094 2023-01-21 11:21:06.056382: step: 376/526, loss: 0.0021009445190429688 2023-01-21 11:21:07.214645: step: 380/526, loss: 0.028389930725097656 2023-01-21 11:21:08.404818: step: 384/526, loss: 0.029287436977028847 2023-01-21 11:21:09.570562: step: 388/526, loss: 0.006999206729233265 2023-01-21 11:21:10.725826: step: 392/526, loss: 0.039727017283439636 2023-01-21 11:21:11.871185: step: 396/526, loss: 0.09039411693811417 2023-01-21 11:21:13.001533: step: 400/526, loss: 0.00019559860811568797 2023-01-21 11:21:14.142700: step: 404/526, loss: 0.02323608286678791 2023-01-21 11:21:15.271849: step: 408/526, loss: 0.006137060932815075 2023-01-21 11:21:16.470667: step: 412/526, loss: 0.03086681477725506 2023-01-21 11:21:17.594275: step: 416/526, loss: 0.07261276245117188 2023-01-21 11:21:18.743060: step: 420/526, loss: 0.0013107298873364925 2023-01-21 11:21:19.891047: step: 424/526, loss: 0.000828647636808455 2023-01-21 11:21:21.031469: step: 428/526, loss: 0.011659479700028896 2023-01-21 11:21:22.171892: step: 432/526, loss: 0.06942576915025711 2023-01-21 11:21:23.335239: step: 436/526, loss: 0.00394439697265625 2023-01-21 11:21:24.489772: step: 440/526, loss: 0.0008063316345214844 2023-01-21 11:21:25.670633: step: 444/526, loss: 0.009787273593246937 2023-01-21 11:21:26.811597: step: 448/526, loss: 0.1708083152770996 2023-01-21 11:21:27.956619: step: 452/526, loss: 0.00014190674119163305 2023-01-21 11:21:29.127677: step: 456/526, loss: 0.01170959509909153 2023-01-21 11:21:30.299438: step: 460/526, loss: 0.10126066207885742 2023-01-21 11:21:31.465634: step: 464/526, loss: 0.014204979874193668 2023-01-21 11:21:32.640086: step: 468/526, loss: 0.14807739853858948 2023-01-21 11:21:33.777567: step: 472/526, loss: 0.0011466980213299394 2023-01-21 11:21:34.912516: step: 476/526, loss: 0.013512516394257545 2023-01-21 11:21:36.055494: step: 480/526, loss: 0.007953834719955921 2023-01-21 11:21:37.182680: step: 484/526, loss: 0.006524467375129461 2023-01-21 11:21:38.354275: step: 488/526, loss: 0.0019288540352135897 2023-01-21 11:21:39.513925: step: 492/526, loss: 0.0006891250959597528 2023-01-21 11:21:40.651611: step: 496/526, loss: 0.04015376791357994 2023-01-21 11:21:41.779083: step: 500/526, loss: 0.6664199829101562 2023-01-21 11:21:42.927953: step: 504/526, loss: 0.040895987302064896 2023-01-21 11:21:44.092785: step: 508/526, loss: 0.050680164247751236 2023-01-21 11:21:45.247134: step: 512/526, loss: 0.014007186517119408 2023-01-21 11:21:46.410118: step: 516/526, loss: 0.00402679480612278 2023-01-21 11:21:47.553920: step: 520/526, loss: 0.04378996044397354 2023-01-21 11:21:48.698859: step: 524/526, loss: 0.0006673813331872225 2023-01-21 11:21:49.845795: step: 528/526, loss: 0.0032178880646824837 2023-01-21 11:21:51.011140: step: 532/526, loss: 0.008038091473281384 2023-01-21 11:21:52.164346: step: 536/526, loss: 0.009963417425751686 2023-01-21 11:21:53.286312: step: 540/526, loss: 0.0005750656127929688 2023-01-21 11:21:54.458471: step: 544/526, loss: 0.1334766298532486 2023-01-21 11:21:55.611611: step: 548/526, loss: 0.00023899078951217234 2023-01-21 11:21:56.765509: step: 552/526, loss: 0.09313793480396271 2023-01-21 11:21:57.917148: step: 556/526, loss: 0.07736358791589737 2023-01-21 11:21:59.037228: step: 560/526, loss: 0.013270950876176357 2023-01-21 11:22:00.159089: step: 564/526, loss: 0.08632683753967285 2023-01-21 11:22:01.302225: step: 568/526, loss: 0.055107977241277695 2023-01-21 11:22:02.446343: step: 572/526, loss: 0.0077197072096168995 2023-01-21 11:22:03.613107: step: 576/526, loss: 0.05015640705823898 2023-01-21 11:22:04.765890: step: 580/526, loss: 0.00459213275462389 2023-01-21 11:22:05.896394: step: 584/526, loss: 0.0058259014040231705 2023-01-21 11:22:07.027553: step: 588/526, loss: 0.019400596618652344 2023-01-21 11:22:08.209613: step: 592/526, loss: 0.0005949020269326866 2023-01-21 11:22:09.342600: step: 596/526, loss: 0.07880210876464844 2023-01-21 11:22:10.487397: step: 600/526, loss: 0.05722007900476456 2023-01-21 11:22:11.636760: step: 604/526, loss: 0.01571188122034073 2023-01-21 11:22:12.793495: step: 608/526, loss: 0.012703800573945045 2023-01-21 11:22:13.925751: step: 612/526, loss: 0.15576894581317902 2023-01-21 11:22:15.085993: step: 616/526, loss: 0.024607468396425247 2023-01-21 11:22:16.211272: step: 620/526, loss: 0.02085094526410103 2023-01-21 11:22:17.350091: step: 624/526, loss: 0.005806922912597656 2023-01-21 11:22:18.471487: step: 628/526, loss: 0.024089815095067024 2023-01-21 11:22:19.610022: step: 632/526, loss: 0.05824146419763565 2023-01-21 11:22:20.785969: step: 636/526, loss: 0.016150474548339844 2023-01-21 11:22:21.932180: step: 640/526, loss: 0.01067895907908678 2023-01-21 11:22:23.075744: step: 644/526, loss: 0.06559982150793076 2023-01-21 11:22:24.215799: step: 648/526, loss: 0.0009349823230877519 2023-01-21 11:22:25.361707: step: 652/526, loss: 0.0025187493301928043 2023-01-21 11:22:26.536463: step: 656/526, loss: 0.02749490737915039 2023-01-21 11:22:27.694260: step: 660/526, loss: 0.010251045227050781 2023-01-21 11:22:28.821473: step: 664/526, loss: 0.03259697183966637 2023-01-21 11:22:29.964722: step: 668/526, loss: 0.19118356704711914 2023-01-21 11:22:31.115955: step: 672/526, loss: 0.00396041851490736 2023-01-21 11:22:32.264371: step: 676/526, loss: 0.034720804542303085 2023-01-21 11:22:33.424330: step: 680/526, loss: 0.01591358333826065 2023-01-21 11:22:34.552714: step: 684/526, loss: 0.4573030471801758 2023-01-21 11:22:35.692754: step: 688/526, loss: 0.017461776733398438 2023-01-21 11:22:36.901300: step: 692/526, loss: 0.030918169766664505 2023-01-21 11:22:38.055669: step: 696/526, loss: 0.003731679869815707 2023-01-21 11:22:39.192664: step: 700/526, loss: 0.017079640179872513 2023-01-21 11:22:40.367921: step: 704/526, loss: 0.03582973778247833 2023-01-21 11:22:41.500578: step: 708/526, loss: 0.003549480577930808 2023-01-21 11:22:42.687195: step: 712/526, loss: 0.04098358005285263 2023-01-21 11:22:43.861427: step: 716/526, loss: 0.02720518223941326 2023-01-21 11:22:45.009583: step: 720/526, loss: 0.010844326578080654 2023-01-21 11:22:46.183061: step: 724/526, loss: 0.06526069343090057 2023-01-21 11:22:47.320622: step: 728/526, loss: 0.0010103702079504728 2023-01-21 11:22:48.495421: step: 732/526, loss: 0.013023662380874157 2023-01-21 11:22:49.632402: step: 736/526, loss: 0.10436000674962997 2023-01-21 11:22:50.776118: step: 740/526, loss: 0.018746376037597656 2023-01-21 11:22:51.929208: step: 744/526, loss: 0.010208320803940296 2023-01-21 11:22:53.108362: step: 748/526, loss: 0.03430977091193199 2023-01-21 11:22:54.272244: step: 752/526, loss: 0.011287307366728783 2023-01-21 11:22:55.419500: step: 756/526, loss: 0.0008378028869628906 2023-01-21 11:22:56.557032: step: 760/526, loss: 0.15765905380249023 2023-01-21 11:22:57.732370: step: 764/526, loss: 0.07822628319263458 2023-01-21 11:22:58.862638: step: 768/526, loss: 0.0011303424835205078 2023-01-21 11:23:00.035414: step: 772/526, loss: 0.009031486697494984 2023-01-21 11:23:01.179755: step: 776/526, loss: 0.01705322228372097 2023-01-21 11:23:02.348273: step: 780/526, loss: 0.7453729510307312 2023-01-21 11:23:03.474919: step: 784/526, loss: 0.03326597064733505 2023-01-21 11:23:04.595852: step: 788/526, loss: 0.018526364117860794 2023-01-21 11:23:05.724230: step: 792/526, loss: 0.01178207341581583 2023-01-21 11:23:06.899884: step: 796/526, loss: 0.0231457706540823 2023-01-21 11:23:08.016570: step: 800/526, loss: 9.384155418956652e-05 2023-01-21 11:23:09.166748: step: 804/526, loss: 0.009880256839096546 2023-01-21 11:23:10.312013: step: 808/526, loss: 0.021640969440340996 2023-01-21 11:23:11.467315: step: 812/526, loss: 0.010782241821289062 2023-01-21 11:23:12.607685: step: 816/526, loss: 0.003671550890430808 2023-01-21 11:23:13.725243: step: 820/526, loss: 0.006351756863296032 2023-01-21 11:23:14.879972: step: 824/526, loss: 0.0023437500931322575 2023-01-21 11:23:16.036035: step: 828/526, loss: 0.005060291383415461 2023-01-21 11:23:17.202698: step: 832/526, loss: 0.014135456643998623 2023-01-21 11:23:18.359450: step: 836/526, loss: 0.04877214506268501 2023-01-21 11:23:19.495600: step: 840/526, loss: 0.009589672088623047 2023-01-21 11:23:20.670406: step: 844/526, loss: 0.012925434857606888 2023-01-21 11:23:21.796382: step: 848/526, loss: 0.006515598390251398 2023-01-21 11:23:22.973688: step: 852/526, loss: 0.06548157334327698 2023-01-21 11:23:24.140276: step: 856/526, loss: 0.023426055908203125 2023-01-21 11:23:25.278179: step: 860/526, loss: 0.006204700563102961 2023-01-21 11:23:26.479051: step: 864/526, loss: 0.0011611938243731856 2023-01-21 11:23:27.629428: step: 868/526, loss: 0.012956619262695312 2023-01-21 11:23:28.777008: step: 872/526, loss: 0.026856709271669388 2023-01-21 11:23:29.904658: step: 876/526, loss: 0.0063949585892260075 2023-01-21 11:23:31.049187: step: 880/526, loss: 9.298325312556699e-05 2023-01-21 11:23:32.184269: step: 884/526, loss: 0.046695709228515625 2023-01-21 11:23:33.323358: step: 888/526, loss: 0.003896331647410989 2023-01-21 11:23:34.467401: step: 892/526, loss: 0.02193247154355049 2023-01-21 11:23:35.667560: step: 896/526, loss: 0.04974813759326935 2023-01-21 11:23:36.847561: step: 900/526, loss: 0.026309967041015625 2023-01-21 11:23:37.989051: step: 904/526, loss: 0.021569253876805305 2023-01-21 11:23:39.143556: step: 908/526, loss: 0.017328644171357155 2023-01-21 11:23:40.306369: step: 912/526, loss: 0.05048827826976776 2023-01-21 11:23:41.436215: step: 916/526, loss: 0.03735008090734482 2023-01-21 11:23:42.597354: step: 920/526, loss: 0.020472144708037376 2023-01-21 11:23:43.784903: step: 924/526, loss: 0.0218382366001606 2023-01-21 11:23:44.921423: step: 928/526, loss: 0.07334079593420029 2023-01-21 11:23:46.079462: step: 932/526, loss: 0.07878141105175018 2023-01-21 11:23:47.238330: step: 936/526, loss: 0.0017305375076830387 2023-01-21 11:23:48.429021: step: 940/526, loss: 0.002079582307487726 2023-01-21 11:23:49.569935: step: 944/526, loss: 0.0004038810729980469 2023-01-21 11:23:50.705056: step: 948/526, loss: 0.025563621893525124 2023-01-21 11:23:51.877055: step: 952/526, loss: 0.0032917975913733244 2023-01-21 11:23:53.074009: step: 956/526, loss: 0.005711269564926624 2023-01-21 11:23:54.265524: step: 960/526, loss: 0.042267415672540665 2023-01-21 11:23:55.412951: step: 964/526, loss: 0.06126260757446289 2023-01-21 11:23:56.560269: step: 968/526, loss: 0.11003799736499786 2023-01-21 11:23:57.719544: step: 972/526, loss: 0.054376937448978424 2023-01-21 11:23:58.894417: step: 976/526, loss: 0.324615478515625 2023-01-21 11:24:00.063807: step: 980/526, loss: 0.011299324221909046 2023-01-21 11:24:01.227847: step: 984/526, loss: 0.04791727289557457 2023-01-21 11:24:02.345801: step: 988/526, loss: 0.03502492979168892 2023-01-21 11:24:03.520219: step: 992/526, loss: 0.05111370235681534 2023-01-21 11:24:04.695792: step: 996/526, loss: 0.012689972296357155 2023-01-21 11:24:05.831840: step: 1000/526, loss: 0.01204681396484375 2023-01-21 11:24:06.969495: step: 1004/526, loss: 0.017117274925112724 2023-01-21 11:24:08.132422: step: 1008/526, loss: 0.055908337235450745 2023-01-21 11:24:09.284572: step: 1012/526, loss: 0.007498741149902344 2023-01-21 11:24:10.444089: step: 1016/526, loss: 0.009439468383789062 2023-01-21 11:24:11.568765: step: 1020/526, loss: 0.00013599396334029734 2023-01-21 11:24:12.738834: step: 1024/526, loss: 0.0611303336918354 2023-01-21 11:24:13.936832: step: 1028/526, loss: 0.08583197742700577 2023-01-21 11:24:15.125160: step: 1032/526, loss: 0.004662561230361462 2023-01-21 11:24:16.267998: step: 1036/526, loss: 0.017775440588593483 2023-01-21 11:24:17.400883: step: 1040/526, loss: 0.02116527408361435 2023-01-21 11:24:18.554820: step: 1044/526, loss: 0.02670269086956978 2023-01-21 11:24:19.699963: step: 1048/526, loss: 0.00222358712926507 2023-01-21 11:24:20.855462: step: 1052/526, loss: 0.03948249667882919 2023-01-21 11:24:22.016843: step: 1056/526, loss: 0.0013480663765221834 2023-01-21 11:24:23.165812: step: 1060/526, loss: 0.019669342786073685 2023-01-21 11:24:24.311910: step: 1064/526, loss: 0.013860797509551048 2023-01-21 11:24:25.505108: step: 1068/526, loss: 0.006151676177978516 2023-01-21 11:24:26.676881: step: 1072/526, loss: 0.14107413589954376 2023-01-21 11:24:27.829316: step: 1076/526, loss: 0.04501352459192276 2023-01-21 11:24:28.988286: step: 1080/526, loss: 0.0010957717895507812 2023-01-21 11:24:30.138432: step: 1084/526, loss: 0.040661975741386414 2023-01-21 11:24:31.289633: step: 1088/526, loss: 0.004996109288185835 2023-01-21 11:24:32.446646: step: 1092/526, loss: 0.017347240820527077 2023-01-21 11:24:33.627045: step: 1096/526, loss: 0.0011172294616699219 2023-01-21 11:24:34.756397: step: 1100/526, loss: 0.04806967079639435 2023-01-21 11:24:35.904286: step: 1104/526, loss: 0.11985526978969574 2023-01-21 11:24:37.066270: step: 1108/526, loss: 0.010937238112092018 2023-01-21 11:24:38.181587: step: 1112/526, loss: 0.07239770889282227 2023-01-21 11:24:39.329235: step: 1116/526, loss: 0.03562898933887482 2023-01-21 11:24:40.449571: step: 1120/526, loss: 0.04157219082117081 2023-01-21 11:24:41.632467: step: 1124/526, loss: 0.010804176330566406 2023-01-21 11:24:42.764075: step: 1128/526, loss: 0.020627882331609726 2023-01-21 11:24:43.918171: step: 1132/526, loss: 0.005881595890969038 2023-01-21 11:24:45.057566: step: 1136/526, loss: 0.015665389597415924 2023-01-21 11:24:46.200575: step: 1140/526, loss: 0.001398372696712613 2023-01-21 11:24:47.429969: step: 1144/526, loss: 0.11602059006690979 2023-01-21 11:24:48.594013: step: 1148/526, loss: 0.019445039331912994 2023-01-21 11:24:49.715719: step: 1152/526, loss: 0.0007688045734539628 2023-01-21 11:24:50.869080: step: 1156/526, loss: 0.03306932747364044 2023-01-21 11:24:52.020616: step: 1160/526, loss: 0.003934717271476984 2023-01-21 11:24:53.184766: step: 1164/526, loss: 0.009060859680175781 2023-01-21 11:24:54.363133: step: 1168/526, loss: 0.015883494168519974 2023-01-21 11:24:55.500098: step: 1172/526, loss: 0.06717921048402786 2023-01-21 11:24:56.650314: step: 1176/526, loss: 0.0018405914306640625 2023-01-21 11:24:57.791460: step: 1180/526, loss: 0.0010327368509024382 2023-01-21 11:24:58.932347: step: 1184/526, loss: 0.022007416933774948 2023-01-21 11:25:00.101854: step: 1188/526, loss: 0.020312214270234108 2023-01-21 11:25:01.297720: step: 1192/526, loss: 0.08988456428050995 2023-01-21 11:25:02.425491: step: 1196/526, loss: 0.015948869287967682 2023-01-21 11:25:03.559963: step: 1200/526, loss: 0.03337812423706055 2023-01-21 11:25:04.715969: step: 1204/526, loss: 0.007315540686249733 2023-01-21 11:25:05.885937: step: 1208/526, loss: 0.011577225290238857 2023-01-21 11:25:07.052352: step: 1212/526, loss: 0.008961677551269531 2023-01-21 11:25:08.209774: step: 1216/526, loss: 0.007487487979233265 2023-01-21 11:25:09.339652: step: 1220/526, loss: 0.006799888797104359 2023-01-21 11:25:10.494811: step: 1224/526, loss: 0.0004146576102357358 2023-01-21 11:25:11.641849: step: 1228/526, loss: 0.06662826985120773 2023-01-21 11:25:12.776275: step: 1232/526, loss: 0.013881397433578968 2023-01-21 11:25:13.950286: step: 1236/526, loss: 0.07209930568933487 2023-01-21 11:25:15.080883: step: 1240/526, loss: 0.0018220484489575028 2023-01-21 11:25:16.239042: step: 1244/526, loss: 0.01940889284014702 2023-01-21 11:25:17.399267: step: 1248/526, loss: 0.025511646643280983 2023-01-21 11:25:18.566003: step: 1252/526, loss: 0.00012807846360374242 2023-01-21 11:25:19.705963: step: 1256/526, loss: 0.024041080847382545 2023-01-21 11:25:20.833433: step: 1260/526, loss: 0.007365561090409756 2023-01-21 11:25:21.963986: step: 1264/526, loss: 0.36412307620048523 2023-01-21 11:25:23.119901: step: 1268/526, loss: 0.034429360181093216 2023-01-21 11:25:24.275481: step: 1272/526, loss: 0.010556411929428577 2023-01-21 11:25:25.439958: step: 1276/526, loss: 0.020966148003935814 2023-01-21 11:25:26.587281: step: 1280/526, loss: 0.0007737636333331466 2023-01-21 11:25:27.755827: step: 1284/526, loss: 0.0020576000679284334 2023-01-21 11:25:28.912331: step: 1288/526, loss: 0.05202013999223709 2023-01-21 11:25:30.097732: step: 1292/526, loss: 0.04629163816571236 2023-01-21 11:25:31.267938: step: 1296/526, loss: 0.04239015281200409 2023-01-21 11:25:32.447529: step: 1300/526, loss: 0.027730464935302734 2023-01-21 11:25:33.617417: step: 1304/526, loss: 0.0045299530029296875 2023-01-21 11:25:34.778731: step: 1308/526, loss: 0.05006713792681694 2023-01-21 11:25:35.959329: step: 1312/526, loss: 0.009278487414121628 2023-01-21 11:25:37.096933: step: 1316/526, loss: 0.01439080573618412 2023-01-21 11:25:38.262154: step: 1320/526, loss: 0.010241151787340641 2023-01-21 11:25:39.428331: step: 1324/526, loss: 0.017360782250761986 2023-01-21 11:25:40.569140: step: 1328/526, loss: 0.02648625522851944 2023-01-21 11:25:41.711125: step: 1332/526, loss: 0.014238834381103516 2023-01-21 11:25:42.892030: step: 1336/526, loss: 0.048409465700387955 2023-01-21 11:25:44.077965: step: 1340/526, loss: 0.02427687868475914 2023-01-21 11:25:45.228408: step: 1344/526, loss: 0.02034778520464897 2023-01-21 11:25:46.389455: step: 1348/526, loss: 0.02927885204553604 2023-01-21 11:25:47.570702: step: 1352/526, loss: 0.059866905212402344 2023-01-21 11:25:48.717360: step: 1356/526, loss: 0.03551807254552841 2023-01-21 11:25:49.863421: step: 1360/526, loss: 0.006972646806389093 2023-01-21 11:25:51.017470: step: 1364/526, loss: 0.01871929131448269 2023-01-21 11:25:52.158015: step: 1368/526, loss: 0.014097404666244984 2023-01-21 11:25:53.309633: step: 1372/526, loss: 0.017656372860074043 2023-01-21 11:25:54.461577: step: 1376/526, loss: 0.0036301612854003906 2023-01-21 11:25:55.604302: step: 1380/526, loss: 0.004176235292106867 2023-01-21 11:25:56.740915: step: 1384/526, loss: 0.004494952969253063 2023-01-21 11:25:57.882036: step: 1388/526, loss: 0.008702421560883522 2023-01-21 11:25:59.038098: step: 1392/526, loss: 0.007414627354592085 2023-01-21 11:26:00.231458: step: 1396/526, loss: 0.0048385621048510075 2023-01-21 11:26:01.390476: step: 1400/526, loss: 0.011631202884018421 2023-01-21 11:26:02.519514: step: 1404/526, loss: 0.0345911979675293 2023-01-21 11:26:03.665115: step: 1408/526, loss: 0.004528164863586426 2023-01-21 11:26:04.837244: step: 1412/526, loss: 0.005847168155014515 2023-01-21 11:26:05.981076: step: 1416/526, loss: 0.005957103334367275 2023-01-21 11:26:07.105553: step: 1420/526, loss: 0.03772792965173721 2023-01-21 11:26:08.221290: step: 1424/526, loss: 0.0217269416898489 2023-01-21 11:26:09.400767: step: 1428/526, loss: 0.051311589777469635 2023-01-21 11:26:10.555576: step: 1432/526, loss: 0.08636932075023651 2023-01-21 11:26:11.684976: step: 1436/526, loss: 0.029451416805386543 2023-01-21 11:26:12.848895: step: 1440/526, loss: 0.028063582256436348 2023-01-21 11:26:14.001012: step: 1444/526, loss: 0.009215736761689186 2023-01-21 11:26:15.143459: step: 1448/526, loss: 0.027518033981323242 2023-01-21 11:26:16.304975: step: 1452/526, loss: 0.007077121641486883 2023-01-21 11:26:17.459135: step: 1456/526, loss: 0.025943756103515625 2023-01-21 11:26:18.642768: step: 1460/526, loss: 0.06484813988208771 2023-01-21 11:26:19.795247: step: 1464/526, loss: 0.023143388330936432 2023-01-21 11:26:20.965043: step: 1468/526, loss: 0.021740246564149857 2023-01-21 11:26:22.128657: step: 1472/526, loss: 0.006299733649939299 2023-01-21 11:26:23.301973: step: 1476/526, loss: 0.1377815157175064 2023-01-21 11:26:24.441741: step: 1480/526, loss: 0.06675568222999573 2023-01-21 11:26:25.588246: step: 1484/526, loss: 0.02924194373190403 2023-01-21 11:26:26.720858: step: 1488/526, loss: 0.005580043885856867 2023-01-21 11:26:27.862773: step: 1492/526, loss: 0.02570180967450142 2023-01-21 11:26:29.004856: step: 1496/526, loss: 0.07973155379295349 2023-01-21 11:26:30.141084: step: 1500/526, loss: 0.016829872503876686 2023-01-21 11:26:31.303652: step: 1504/526, loss: 0.03310227394104004 2023-01-21 11:26:32.436096: step: 1508/526, loss: 0.011465931311249733 2023-01-21 11:26:33.576654: step: 1512/526, loss: 0.03176116943359375 2023-01-21 11:26:34.729757: step: 1516/526, loss: 0.06437158584594727 2023-01-21 11:26:35.865111: step: 1520/526, loss: 0.033135414123535156 2023-01-21 11:26:37.028351: step: 1524/526, loss: 0.7013838887214661 2023-01-21 11:26:38.219582: step: 1528/526, loss: 0.06686048954725266 2023-01-21 11:26:39.361905: step: 1532/526, loss: 0.04190211370587349 2023-01-21 11:26:40.489167: step: 1536/526, loss: 0.03224353864789009 2023-01-21 11:26:41.629840: step: 1540/526, loss: 0.15404310822486877 2023-01-21 11:26:42.830403: step: 1544/526, loss: 0.004715061280876398 2023-01-21 11:26:43.990221: step: 1548/526, loss: 0.055117275565862656 2023-01-21 11:26:45.149224: step: 1552/526, loss: 0.1209026351571083 2023-01-21 11:26:46.289768: step: 1556/526, loss: 0.031272079795598984 2023-01-21 11:26:47.484116: step: 1560/526, loss: 0.017296981066465378 2023-01-21 11:26:48.636159: step: 1564/526, loss: 0.558657169342041 2023-01-21 11:26:49.771881: step: 1568/526, loss: 0.003933954518288374 2023-01-21 11:26:50.964939: step: 1572/526, loss: 0.013286590576171875 2023-01-21 11:26:52.118129: step: 1576/526, loss: 0.009356856346130371 2023-01-21 11:26:53.288829: step: 1580/526, loss: 0.026461409404873848 2023-01-21 11:26:54.464472: step: 1584/526, loss: 0.0025741576682776213 2023-01-21 11:26:55.602948: step: 1588/526, loss: 0.0011539459228515625 2023-01-21 11:26:56.724446: step: 1592/526, loss: 0.0022188187576830387 2023-01-21 11:26:57.863989: step: 1596/526, loss: 0.002547240350395441 2023-01-21 11:26:59.024125: step: 1600/526, loss: 0.002344417618587613 2023-01-21 11:27:00.163657: step: 1604/526, loss: 0.051453590393066406 2023-01-21 11:27:01.328770: step: 1608/526, loss: 0.024554062634706497 2023-01-21 11:27:02.477814: step: 1612/526, loss: 0.00838475301861763 2023-01-21 11:27:03.644864: step: 1616/526, loss: 0.02063126675784588 2023-01-21 11:27:04.771411: step: 1620/526, loss: 0.009009934030473232 2023-01-21 11:27:05.933577: step: 1624/526, loss: 0.026799678802490234 2023-01-21 11:27:07.065171: step: 1628/526, loss: 0.05049638822674751 2023-01-21 11:27:08.242269: step: 1632/526, loss: 0.0020748137030750513 2023-01-21 11:27:09.416521: step: 1636/526, loss: -3.004073732881807e-06 2023-01-21 11:27:10.568232: step: 1640/526, loss: 0.07907982170581818 2023-01-21 11:27:11.715102: step: 1644/526, loss: 0.07097997516393661 2023-01-21 11:27:12.869438: step: 1648/526, loss: 0.062343597412109375 2023-01-21 11:27:14.002102: step: 1652/526, loss: 0.09758994728326797 2023-01-21 11:27:15.151672: step: 1656/526, loss: 0.11636219918727875 2023-01-21 11:27:16.318855: step: 1660/526, loss: 0.009195208549499512 2023-01-21 11:27:17.459735: step: 1664/526, loss: 0.0028181076049804688 2023-01-21 11:27:18.626454: step: 1668/526, loss: 0.0558408759534359 2023-01-21 11:27:19.747424: step: 1672/526, loss: 0.0030951977241784334 2023-01-21 11:27:20.879517: step: 1676/526, loss: 0.0024912357330322266 2023-01-21 11:27:22.015513: step: 1680/526, loss: 0.03528556600213051 2023-01-21 11:27:23.141133: step: 1684/526, loss: 0.04412698745727539 2023-01-21 11:27:24.297466: step: 1688/526, loss: 0.010465907864272594 2023-01-21 11:27:25.455437: step: 1692/526, loss: 0.012494850903749466 2023-01-21 11:27:26.620033: step: 1696/526, loss: 0.03230543062090874 2023-01-21 11:27:27.767066: step: 1700/526, loss: 0.005598688032478094 2023-01-21 11:27:28.939605: step: 1704/526, loss: 0.5719591379165649 2023-01-21 11:27:30.127855: step: 1708/526, loss: 0.00456085242331028 2023-01-21 11:27:31.297226: step: 1712/526, loss: 0.002419757889583707 2023-01-21 11:27:32.470321: step: 1716/526, loss: 0.003182315733283758 2023-01-21 11:27:33.574295: step: 1720/526, loss: 0.004273462574928999 2023-01-21 11:27:34.721254: step: 1724/526, loss: 0.08217382431030273 2023-01-21 11:27:35.882155: step: 1728/526, loss: 0.027550602331757545 2023-01-21 11:27:37.030471: step: 1732/526, loss: 0.04347586631774902 2023-01-21 11:27:38.177671: step: 1736/526, loss: 0.07771559059619904 2023-01-21 11:27:39.305061: step: 1740/526, loss: 0.013586331158876419 2023-01-21 11:27:40.457809: step: 1744/526, loss: 0.0018961429595947266 2023-01-21 11:27:41.598089: step: 1748/526, loss: 0.005118942353874445 2023-01-21 11:27:42.738689: step: 1752/526, loss: 0.026560688391327858 2023-01-21 11:27:43.872519: step: 1756/526, loss: 0.0005363464588299394 2023-01-21 11:27:45.021599: step: 1760/526, loss: 0.022364521399140358 2023-01-21 11:27:46.201045: step: 1764/526, loss: 0.018175601959228516 2023-01-21 11:27:47.362589: step: 1768/526, loss: 0.005443144124001265 2023-01-21 11:27:48.491559: step: 1772/526, loss: 0.01248016394674778 2023-01-21 11:27:49.656310: step: 1776/526, loss: 0.044884875416755676 2023-01-21 11:27:50.811684: step: 1780/526, loss: 0.01636476442217827 2023-01-21 11:27:52.006183: step: 1784/526, loss: 0.028263092041015625 2023-01-21 11:27:53.180305: step: 1788/526, loss: 0.03206014633178711 2023-01-21 11:27:54.330113: step: 1792/526, loss: 0.04886217415332794 2023-01-21 11:27:55.492560: step: 1796/526, loss: 0.0009314537164755166 2023-01-21 11:27:56.661353: step: 1800/526, loss: 0.047337912023067474 2023-01-21 11:27:57.817894: step: 1804/526, loss: 0.00031490327091887593 2023-01-21 11:27:58.988297: step: 1808/526, loss: 0.030181502923369408 2023-01-21 11:28:00.155383: step: 1812/526, loss: 0.018903352320194244 2023-01-21 11:28:01.304471: step: 1816/526, loss: 0.062339916825294495 2023-01-21 11:28:02.443211: step: 1820/526, loss: 0.0068701268173754215 2023-01-21 11:28:03.620038: step: 1824/526, loss: 0.03403320536017418 2023-01-21 11:28:04.743312: step: 1828/526, loss: 0.010880088433623314 2023-01-21 11:28:05.909977: step: 1832/526, loss: 0.017440224066376686 2023-01-21 11:28:07.043321: step: 1836/526, loss: 0.09154100716114044 2023-01-21 11:28:08.186719: step: 1840/526, loss: 0.07327929139137268 2023-01-21 11:28:09.329440: step: 1844/526, loss: 0.2678384780883789 2023-01-21 11:28:10.487632: step: 1848/526, loss: 0.005862140562385321 2023-01-21 11:28:11.639188: step: 1852/526, loss: 0.06640391796827316 2023-01-21 11:28:12.855153: step: 1856/526, loss: 0.14631310105323792 2023-01-21 11:28:14.025003: step: 1860/526, loss: 0.006291770841926336 2023-01-21 11:28:15.192030: step: 1864/526, loss: 0.02634124644100666 2023-01-21 11:28:16.334407: step: 1868/526, loss: 0.006655740551650524 2023-01-21 11:28:17.497130: step: 1872/526, loss: 0.035736940801143646 2023-01-21 11:28:18.639833: step: 1876/526, loss: 0.3259812295436859 2023-01-21 11:28:19.782480: step: 1880/526, loss: 0.045960187911987305 2023-01-21 11:28:20.974107: step: 1884/526, loss: 0.04650431126356125 2023-01-21 11:28:22.186172: step: 1888/526, loss: 0.006530856713652611 2023-01-21 11:28:23.315884: step: 1892/526, loss: 0.06782494485378265 2023-01-21 11:28:24.492827: step: 1896/526, loss: 0.005308437626808882 2023-01-21 11:28:25.685577: step: 1900/526, loss: 0.02348356321454048 2023-01-21 11:28:26.817604: step: 1904/526, loss: 0.014668083749711514 2023-01-21 11:28:27.950309: step: 1908/526, loss: 0.01696472242474556 2023-01-21 11:28:29.117251: step: 1912/526, loss: 0.020378874614834785 2023-01-21 11:28:30.267052: step: 1916/526, loss: 0.0008379936334677041 2023-01-21 11:28:31.409572: step: 1920/526, loss: 0.026038646697998047 2023-01-21 11:28:32.555185: step: 1924/526, loss: 0.026744937524199486 2023-01-21 11:28:33.721532: step: 1928/526, loss: 0.013733197003602982 2023-01-21 11:28:34.865302: step: 1932/526, loss: 0.0059073446318507195 2023-01-21 11:28:36.010053: step: 1936/526, loss: 0.16801396012306213 2023-01-21 11:28:37.161533: step: 1940/526, loss: 0.04417915269732475 2023-01-21 11:28:38.314991: step: 1944/526, loss: 0.04421444237232208 2023-01-21 11:28:39.453251: step: 1948/526, loss: 0.004235363099724054 2023-01-21 11:28:40.595611: step: 1952/526, loss: 0.003881836077198386 2023-01-21 11:28:41.725882: step: 1956/526, loss: 0.0014799118507653475 2023-01-21 11:28:42.894197: step: 1960/526, loss: 0.01616344414651394 2023-01-21 11:28:44.051809: step: 1964/526, loss: 0.027036476880311966 2023-01-21 11:28:45.214301: step: 1968/526, loss: 0.023020554333925247 2023-01-21 11:28:46.384968: step: 1972/526, loss: 0.027962446212768555 2023-01-21 11:28:47.534917: step: 1976/526, loss: 0.007340335752815008 2023-01-21 11:28:48.778403: step: 1980/526, loss: 0.007399177644401789 2023-01-21 11:28:49.924302: step: 1984/526, loss: 0.03774986416101456 2023-01-21 11:28:51.073204: step: 1988/526, loss: 0.027124499902129173 2023-01-21 11:28:52.218552: step: 1992/526, loss: 0.022027302533388138 2023-01-21 11:28:53.427482: step: 1996/526, loss: 0.009863234125077724 2023-01-21 11:28:54.582151: step: 2000/526, loss: 0.013858986087143421 2023-01-21 11:28:55.729661: step: 2004/526, loss: 0.008997059427201748 2023-01-21 11:28:56.860286: step: 2008/526, loss: 0.02874666266143322 2023-01-21 11:28:58.035521: step: 2012/526, loss: 0.01140441931784153 2023-01-21 11:28:59.187008: step: 2016/526, loss: 0.0008622646564617753 2023-01-21 11:29:00.357974: step: 2020/526, loss: 0.0729396864771843 2023-01-21 11:29:01.522954: step: 2024/526, loss: 0.016897058114409447 2023-01-21 11:29:02.701172: step: 2028/526, loss: 0.0374792106449604 2023-01-21 11:29:03.844906: step: 2032/526, loss: 0.010707378387451172 2023-01-21 11:29:05.002708: step: 2036/526, loss: 0.003799057099968195 2023-01-21 11:29:06.138846: step: 2040/526, loss: 0.11505670845508575 2023-01-21 11:29:07.279332: step: 2044/526, loss: 0.025504684075713158 2023-01-21 11:29:08.415400: step: 2048/526, loss: 0.00025835036649368703 2023-01-21 11:29:09.577597: step: 2052/526, loss: 0.019388772547245026 2023-01-21 11:29:10.745251: step: 2056/526, loss: 0.35503169894218445 2023-01-21 11:29:11.930377: step: 2060/526, loss: 0.027684496715664864 2023-01-21 11:29:13.091587: step: 2064/526, loss: 0.0491364486515522 2023-01-21 11:29:14.236051: step: 2068/526, loss: 0.006453514099121094 2023-01-21 11:29:15.404734: step: 2072/526, loss: 0.00449900608509779 2023-01-21 11:29:16.537424: step: 2076/526, loss: 0.031456850469112396 2023-01-21 11:29:17.711983: step: 2080/526, loss: 0.0054968358017504215 2023-01-21 11:29:18.864440: step: 2084/526, loss: 0.01752195507287979 2023-01-21 11:29:20.025675: step: 2088/526, loss: 0.016249418258666992 2023-01-21 11:29:21.190162: step: 2092/526, loss: 0.056441307067871094 2023-01-21 11:29:22.346918: step: 2096/526, loss: 0.020601511001586914 2023-01-21 11:29:23.471830: step: 2100/526, loss: 0.018584443256258965 2023-01-21 11:29:24.630760: step: 2104/526, loss: 0.002734279725700617 ================================================== Loss: 0.039 -------------------- Dev: {'event': {'p': 0.5861386138613861, 'r': 0.7882822902796272, 'f1': 0.6723452583759226}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Test: {'event': {'p': 0.6248436848686952, 'r': 0.7823590814196242, 'f1': 0.6947856315179607}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Chinese: {'event': {'p': 0.5813953488372093, 'r': 0.9259259259259259, 'f1': 0.7142857142857142}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Korean: {'event': {'p': 0.5818181818181818, 'r': 0.5079365079365079, 'f1': 0.5423728813559321}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} Russian: {'event': {'p': 0.4888888888888889, 'r': 0.6111111111111112, 'f1': 0.5432098765432098}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 15} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6217672413793104, 'r': 0.7683089214380826, 'f1': 0.6873138773079214}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Eng Test for Russian: {'event': {'p': 0.6300302899177845, 'r': 0.7599164926931107, 'f1': 0.6889046605157322}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} Sample Russian: {'event': {'p': 0.5, 'r': 0.5833333333333334, 'f1': 0.5384615384615384}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 10} ****************************** Epoch: 16 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:30:09.208183: step: 4/526, loss: 0.00011005401756847277 2023-01-21 11:30:10.331975: step: 8/526, loss: 0.000926971435546875 2023-01-21 11:30:11.497268: step: 12/526, loss: 0.029949188232421875 2023-01-21 11:30:12.669340: step: 16/526, loss: 0.006628227420151234 2023-01-21 11:30:13.777801: step: 20/526, loss: 0.037622738629579544 2023-01-21 11:30:14.897712: step: 24/526, loss: 0.009700726717710495 2023-01-21 11:30:16.029194: step: 28/526, loss: 0.003586006350815296 2023-01-21 11:30:17.200156: step: 32/526, loss: 0.0005657196161337197 2023-01-21 11:30:18.360377: step: 36/526, loss: 0.0002120971621479839 2023-01-21 11:30:19.510593: step: 40/526, loss: 0.0004032134893350303 2023-01-21 11:30:20.662140: step: 44/526, loss: 0.05403842777013779 2023-01-21 11:30:21.824070: step: 48/526, loss: 0.004210758488625288 2023-01-21 11:30:22.956420: step: 52/526, loss: 0.00917806662619114 2023-01-21 11:30:24.115637: step: 56/526, loss: 0.00089263916015625 2023-01-21 11:30:25.292594: step: 60/526, loss: 0.008420038037002087 2023-01-21 11:30:26.418492: step: 64/526, loss: 0.039496805518865585 2023-01-21 11:30:27.547679: step: 68/526, loss: 0.030106164515018463 2023-01-21 11:30:28.726060: step: 72/526, loss: 0.024308206513524055 2023-01-21 11:30:29.873721: step: 76/526, loss: 6.545782525790855e-05 2023-01-21 11:30:30.992529: step: 80/526, loss: 0.014850807376205921 2023-01-21 11:30:32.136207: step: 84/526, loss: 0.031487371772527695 2023-01-21 11:30:33.360564: step: 88/526, loss: 0.04479942470788956 2023-01-21 11:30:34.520129: step: 92/526, loss: 0.02416362799704075 2023-01-21 11:30:35.657258: step: 96/526, loss: 0.00467948941513896 2023-01-21 11:30:36.820446: step: 100/526, loss: 0.038279153406620026 2023-01-21 11:30:37.998163: step: 104/526, loss: 0.00205402378924191 2023-01-21 11:30:39.158695: step: 108/526, loss: 0.04345588758587837 2023-01-21 11:30:40.288970: step: 112/526, loss: 0.002930164337158203 2023-01-21 11:30:41.436667: step: 116/526, loss: 0.03169460594654083 2023-01-21 11:30:42.581917: step: 120/526, loss: 0.0056645870208740234 2023-01-21 11:30:43.740179: step: 124/526, loss: 0.007916831411421299 2023-01-21 11:30:44.897306: step: 128/526, loss: 0.003505706787109375 2023-01-21 11:30:46.083435: step: 132/526, loss: 0.03725452721118927 2023-01-21 11:30:47.247176: step: 136/526, loss: 0.0007162094116210938 2023-01-21 11:30:48.414556: step: 140/526, loss: 0.021389389410614967 2023-01-21 11:30:49.544836: step: 144/526, loss: 0.01701831817626953 2023-01-21 11:30:50.724614: step: 148/526, loss: 0.0033901215065270662 2023-01-21 11:30:51.877146: step: 152/526, loss: 0.0017707825172692537 2023-01-21 11:30:53.022198: step: 156/526, loss: 0.5684857368469238 2023-01-21 11:30:54.159466: step: 160/526, loss: 0.010584640316665173 2023-01-21 11:30:55.279605: step: 164/526, loss: 0.001024436904117465 2023-01-21 11:30:56.393694: step: 168/526, loss: 0.009119797497987747 2023-01-21 11:30:57.572439: step: 172/526, loss: 0.012661266140639782 2023-01-21 11:30:58.736501: step: 176/526, loss: 0.023590087890625 2023-01-21 11:30:59.858394: step: 180/526, loss: 0.01925811730325222 2023-01-21 11:31:00.994757: step: 184/526, loss: 0.011992835439741611 2023-01-21 11:31:02.150805: step: 188/526, loss: 0.06738948822021484 2023-01-21 11:31:03.319145: step: 192/526, loss: 0.007896996103227139 2023-01-21 11:31:04.469506: step: 196/526, loss: 0.0003678321954794228 2023-01-21 11:31:05.603084: step: 200/526, loss: 0.002997016767039895 2023-01-21 11:31:06.721990: step: 204/526, loss: 0.008821535855531693 2023-01-21 11:31:07.882421: step: 208/526, loss: 0.004204940982162952 2023-01-21 11:31:09.035018: step: 212/526, loss: 0.0031444551423192024 2023-01-21 11:31:10.172969: step: 216/526, loss: 0.045699309557676315 2023-01-21 11:31:11.348292: step: 220/526, loss: 0.009506797417998314 2023-01-21 11:31:12.526178: step: 224/526, loss: 0.07151356339454651 2023-01-21 11:31:13.710124: step: 228/526, loss: 0.03923730552196503 2023-01-21 11:31:14.878863: step: 232/526, loss: 0.005748939700424671 2023-01-21 11:31:16.090147: step: 236/526, loss: 0.08857040852308273 2023-01-21 11:31:17.278572: step: 240/526, loss: 0.006740999408066273 2023-01-21 11:31:18.400530: step: 244/526, loss: 0.001222896622493863 2023-01-21 11:31:19.550735: step: 248/526, loss: 0.00033245087251998484 2023-01-21 11:31:20.681477: step: 252/526, loss: 0.022715091705322266 2023-01-21 11:31:21.843075: step: 256/526, loss: 0.4950489103794098 2023-01-21 11:31:23.002023: step: 260/526, loss: 0.03233318403363228 2023-01-21 11:31:24.169504: step: 264/526, loss: 0.014944505877792835 2023-01-21 11:31:25.317835: step: 268/526, loss: 0.018395423889160156 2023-01-21 11:31:26.439110: step: 272/526, loss: 0.020806027576327324 2023-01-21 11:31:27.582958: step: 276/526, loss: 0.00037660598172806203 2023-01-21 11:31:28.710100: step: 280/526, loss: 0.0014060974353924394 2023-01-21 11:31:29.895991: step: 284/526, loss: 0.02727203443646431 2023-01-21 11:31:31.060358: step: 288/526, loss: 0.008841228671371937 2023-01-21 11:31:32.198105: step: 292/526, loss: 0.00018597840971779078 2023-01-21 11:31:33.350679: step: 296/526, loss: 0.00043563847430050373 2023-01-21 11:31:34.545378: step: 300/526, loss: 0.034189797937870026 2023-01-21 11:31:35.693684: step: 304/526, loss: 3.5953522456111386e-05 2023-01-21 11:31:36.846325: step: 308/526, loss: 0.04229774326086044 2023-01-21 11:31:38.009543: step: 312/526, loss: 0.05191965028643608 2023-01-21 11:31:39.189086: step: 316/526, loss: 0.036585234105587006 2023-01-21 11:31:40.320011: step: 320/526, loss: 0.0011564254527911544 2023-01-21 11:31:41.447018: step: 324/526, loss: 0.0020558834075927734 2023-01-21 11:31:42.610408: step: 328/526, loss: 0.0005407333374023438 2023-01-21 11:31:43.814622: step: 332/526, loss: 0.0032994269859045744 2023-01-21 11:31:44.965540: step: 336/526, loss: 0.010889149270951748 2023-01-21 11:31:46.124367: step: 340/526, loss: 0.00048341750516556203 2023-01-21 11:31:47.266973: step: 344/526, loss: 0.0035694122780114412 2023-01-21 11:31:48.414321: step: 348/526, loss: 2.3841867005103268e-06 2023-01-21 11:31:49.561303: step: 352/526, loss: 0.01700468175113201 2023-01-21 11:31:50.740053: step: 356/526, loss: 0.07163476943969727 2023-01-21 11:31:51.869851: step: 360/526, loss: 0.0033693313598632812 2023-01-21 11:31:53.022087: step: 364/526, loss: 0.005605125334113836 2023-01-21 11:31:54.178019: step: 368/526, loss: 0.010840988717973232 2023-01-21 11:31:55.325564: step: 372/526, loss: 0.0015970231033861637 2023-01-21 11:31:56.468759: step: 376/526, loss: 0.07096214592456818 2023-01-21 11:31:57.654411: step: 380/526, loss: 0.012497996911406517 2023-01-21 11:31:58.784883: step: 384/526, loss: 0.0019136428600177169 2023-01-21 11:31:59.933321: step: 388/526, loss: 0.019266463816165924 2023-01-21 11:32:01.121995: step: 392/526, loss: 0.0027828216552734375 2023-01-21 11:32:02.269581: step: 396/526, loss: 0.018727969378232956 2023-01-21 11:32:03.417592: step: 400/526, loss: 0.002200603485107422 2023-01-21 11:32:04.570768: step: 404/526, loss: 0.0007186889415606856 2023-01-21 11:32:05.757199: step: 408/526, loss: 0.016614913940429688 2023-01-21 11:32:06.892924: step: 412/526, loss: 0.10731811821460724 2023-01-21 11:32:08.054864: step: 416/526, loss: 0.007856511510908604 2023-01-21 11:32:09.229516: step: 420/526, loss: 0.07938919216394424 2023-01-21 11:32:10.391659: step: 424/526, loss: 0.00168781285174191 2023-01-21 11:32:11.557624: step: 428/526, loss: 0.017522001639008522 2023-01-21 11:32:12.719216: step: 432/526, loss: 0.028425121679902077 2023-01-21 11:32:13.904936: step: 436/526, loss: 0.1173354908823967 2023-01-21 11:32:15.037054: step: 440/526, loss: 0.0013078213669359684 2023-01-21 11:32:16.180699: step: 444/526, loss: 0.03006439283490181 2023-01-21 11:32:17.319272: step: 448/526, loss: 0.025170136243104935 2023-01-21 11:32:18.510968: step: 452/526, loss: 0.0012928962241858244 2023-01-21 11:32:19.648248: step: 456/526, loss: 0.012340545654296875 2023-01-21 11:32:20.793953: step: 460/526, loss: 5.1212311518611386e-05 2023-01-21 11:32:21.952791: step: 464/526, loss: 0.0627140998840332 2023-01-21 11:32:23.114317: step: 468/526, loss: 0.06278286129236221 2023-01-21 11:32:24.307014: step: 472/526, loss: 0.009520339779555798 2023-01-21 11:32:25.459054: step: 476/526, loss: 0.020020723342895508 2023-01-21 11:32:26.630439: step: 480/526, loss: 0.004179859068244696 2023-01-21 11:32:27.807547: step: 484/526, loss: 0.0035198212135583162 2023-01-21 11:32:28.953231: step: 488/526, loss: 0.005193233489990234 2023-01-21 11:32:30.123712: step: 492/526, loss: 0.0011620878940448165 2023-01-21 11:32:31.273490: step: 496/526, loss: 0.013108158484101295 2023-01-21 11:32:32.432527: step: 500/526, loss: 0.010790920816361904 2023-01-21 11:32:33.579160: step: 504/526, loss: 0.04725589603185654 2023-01-21 11:32:34.696651: step: 508/526, loss: 0.005801201332360506 2023-01-21 11:32:35.850316: step: 512/526, loss: 0.0016223908169195056 2023-01-21 11:32:36.997992: step: 516/526, loss: 0.033487893640995026 2023-01-21 11:32:38.133490: step: 520/526, loss: 0.0010188103187829256 2023-01-21 11:32:39.278055: step: 524/526, loss: 0.002458471106365323 2023-01-21 11:32:40.471501: step: 528/526, loss: 0.003955078311264515 2023-01-21 11:32:41.620056: step: 532/526, loss: 0.0006930351373739541 2023-01-21 11:32:42.771869: step: 536/526, loss: 0.00612220773473382 2023-01-21 11:32:43.928265: step: 540/526, loss: 0.029469633474946022 2023-01-21 11:32:45.077497: step: 544/526, loss: 0.07070689648389816 2023-01-21 11:32:46.239949: step: 548/526, loss: 0.03894519805908203 2023-01-21 11:32:47.398300: step: 552/526, loss: 0.006038665771484375 2023-01-21 11:32:48.556177: step: 556/526, loss: 0.01715373992919922 2023-01-21 11:32:49.755441: step: 560/526, loss: 0.02691631391644478 2023-01-21 11:32:50.920614: step: 564/526, loss: 0.03173065185546875 2023-01-21 11:32:52.083289: step: 568/526, loss: 0.05192279815673828 2023-01-21 11:32:53.232503: step: 572/526, loss: 0.7218326926231384 2023-01-21 11:32:54.373163: step: 576/526, loss: 0.007844829931855202 2023-01-21 11:32:55.533145: step: 580/526, loss: 0.014657092280685902 2023-01-21 11:32:56.686203: step: 584/526, loss: 0.0025338411796838045 2023-01-21 11:32:57.829761: step: 588/526, loss: 0.012466048821806908 2023-01-21 11:32:58.963916: step: 592/526, loss: 0.003747367998585105 2023-01-21 11:33:00.095162: step: 596/526, loss: 0.00015802383131813258 2023-01-21 11:33:01.265255: step: 600/526, loss: 0.0013956070179119706 2023-01-21 11:33:02.373909: step: 604/526, loss: 0.00013432501873467118 2023-01-21 11:33:03.528951: step: 608/526, loss: 0.0366269126534462 2023-01-21 11:33:04.688316: step: 612/526, loss: 0.0004051685391459614 2023-01-21 11:33:05.844197: step: 616/526, loss: 0.0031521799974143505 2023-01-21 11:33:07.020477: step: 620/526, loss: 0.011005687527358532 2023-01-21 11:33:08.168079: step: 624/526, loss: 0.004446124657988548 2023-01-21 11:33:09.290381: step: 628/526, loss: 0.0975554883480072 2023-01-21 11:33:10.442523: step: 632/526, loss: 0.03302812948822975 2023-01-21 11:33:11.577152: step: 636/526, loss: 0.008095837198197842 2023-01-21 11:33:12.717635: step: 640/526, loss: 0.06758633255958557 2023-01-21 11:33:13.887267: step: 644/526, loss: 0.04158306121826172 2023-01-21 11:33:15.042757: step: 648/526, loss: 0.009734058752655983 2023-01-21 11:33:16.180371: step: 652/526, loss: 0.05300493165850639 2023-01-21 11:33:17.347132: step: 656/526, loss: 0.0772184357047081 2023-01-21 11:33:18.509889: step: 660/526, loss: 0.0387025885283947 2023-01-21 11:33:19.666452: step: 664/526, loss: 0.056638289242982864 2023-01-21 11:33:20.838732: step: 668/526, loss: 0.021870804950594902 2023-01-21 11:33:21.970806: step: 672/526, loss: 0.002477932022884488 2023-01-21 11:33:23.113623: step: 676/526, loss: 0.0038053514435887337 2023-01-21 11:33:24.257039: step: 680/526, loss: 0.01116104144603014 2023-01-21 11:33:25.426944: step: 684/526, loss: 0.05805826559662819 2023-01-21 11:33:26.551560: step: 688/526, loss: 9.663105447543785e-05 2023-01-21 11:33:27.684236: step: 692/526, loss: 0.006488132290542126 2023-01-21 11:33:28.805611: step: 696/526, loss: 0.009644603356719017 2023-01-21 11:33:29.931765: step: 700/526, loss: 0.1351669430732727 2023-01-21 11:33:31.066392: step: 704/526, loss: 0.015242195688188076 2023-01-21 11:33:32.205733: step: 708/526, loss: 0.010523796081542969 2023-01-21 11:33:33.350976: step: 712/526, loss: 0.0018431664211675525 2023-01-21 11:33:34.501396: step: 716/526, loss: 0.03212776035070419 2023-01-21 11:33:35.675766: step: 720/526, loss: 0.029506683349609375 2023-01-21 11:33:36.822231: step: 724/526, loss: 0.019783878698945045 2023-01-21 11:33:37.977105: step: 728/526, loss: 0.03422107547521591 2023-01-21 11:33:39.122789: step: 732/526, loss: 0.00010271072096657008 2023-01-21 11:33:40.257828: step: 736/526, loss: 0.0025561333168298006 2023-01-21 11:33:41.418454: step: 740/526, loss: 0.007940388284623623 2023-01-21 11:33:42.583011: step: 744/526, loss: 0.018820475786924362 2023-01-21 11:33:43.722070: step: 748/526, loss: 0.007153606973588467 2023-01-21 11:33:44.881375: step: 752/526, loss: 0.013507843017578125 2023-01-21 11:33:46.028576: step: 756/526, loss: 0.008286476135253906 2023-01-21 11:33:47.195685: step: 760/526, loss: 0.0026101351249963045 2023-01-21 11:33:48.332626: step: 764/526, loss: 0.003011125372722745 2023-01-21 11:33:49.469709: step: 768/526, loss: 0.10370655357837677 2023-01-21 11:33:50.643431: step: 772/526, loss: 0.006145477294921875 2023-01-21 11:33:51.768745: step: 776/526, loss: 0.011046123690903187 2023-01-21 11:33:52.938869: step: 780/526, loss: 0.011315441690385342 2023-01-21 11:33:54.081878: step: 784/526, loss: 0.005061340518295765 2023-01-21 11:33:55.206834: step: 788/526, loss: 0.012497376650571823 2023-01-21 11:33:56.354177: step: 792/526, loss: 0.0424153208732605 2023-01-21 11:33:57.483710: step: 796/526, loss: 0.03855876624584198 2023-01-21 11:33:58.617287: step: 800/526, loss: 0.013813686557114124 2023-01-21 11:33:59.758270: step: 804/526, loss: 0.007894134148955345 2023-01-21 11:34:00.852955: step: 808/526, loss: 0.01721782609820366 2023-01-21 11:34:01.969381: step: 812/526, loss: 0.016414929181337357 2023-01-21 11:34:03.106491: step: 816/526, loss: 0.0011794089805334806 2023-01-21 11:34:04.248275: step: 820/526, loss: 0.018193436786532402 2023-01-21 11:34:05.357787: step: 824/526, loss: 0.0004116058407817036 2023-01-21 11:34:06.508039: step: 828/526, loss: 0.0032097341027110815 2023-01-21 11:34:07.678406: step: 832/526, loss: 8.18252592580393e-05 2023-01-21 11:34:08.793605: step: 836/526, loss: 0.0003183365042787045 2023-01-21 11:34:09.941501: step: 840/526, loss: 0.015307235531508923 2023-01-21 11:34:11.097837: step: 844/526, loss: 0.004409885033965111 2023-01-21 11:34:12.262011: step: 848/526, loss: 0.003408813616260886 2023-01-21 11:34:13.403091: step: 852/526, loss: 0.03038616105914116 2023-01-21 11:34:14.544404: step: 856/526, loss: 0.0010211020708084106 2023-01-21 11:34:15.714424: step: 860/526, loss: 0.030455972999334335 2023-01-21 11:34:16.847278: step: 864/526, loss: 0.000154399880557321 2023-01-21 11:34:17.986885: step: 868/526, loss: 0.008781814947724342 2023-01-21 11:34:19.162959: step: 872/526, loss: 0.00012149810936534777 2023-01-21 11:34:20.343276: step: 876/526, loss: 4.024505687993951e-05 2023-01-21 11:34:21.520654: step: 880/526, loss: 0.03061084821820259 2023-01-21 11:34:22.693907: step: 884/526, loss: 0.05472087860107422 2023-01-21 11:34:23.883082: step: 888/526, loss: 0.025283336639404297 2023-01-21 11:34:25.025870: step: 892/526, loss: 0.023142816498875618 2023-01-21 11:34:26.150747: step: 896/526, loss: 0.004337119869887829 2023-01-21 11:34:27.278115: step: 900/526, loss: 0.022684242576360703 2023-01-21 11:34:28.402282: step: 904/526, loss: 0.013498211279511452 2023-01-21 11:34:29.581844: step: 908/526, loss: 0.0207901019603014 2023-01-21 11:34:30.753334: step: 912/526, loss: 0.008356666192412376 2023-01-21 11:34:31.903349: step: 916/526, loss: 0.0105431554839015 2023-01-21 11:34:33.092099: step: 920/526, loss: 0.5433765649795532 2023-01-21 11:34:34.268199: step: 924/526, loss: 0.03345680236816406 2023-01-21 11:34:35.444750: step: 928/526, loss: 0.026952171698212624 2023-01-21 11:34:36.621680: step: 932/526, loss: 0.014483070932328701 2023-01-21 11:34:37.766400: step: 936/526, loss: 0.019674967974424362 2023-01-21 11:34:38.907688: step: 940/526, loss: 0.0019038678146898746 2023-01-21 11:34:40.055416: step: 944/526, loss: 0.008465766906738281 2023-01-21 11:34:41.198325: step: 948/526, loss: 0.015292310155928135 2023-01-21 11:34:42.357735: step: 952/526, loss: 0.0023895264603197575 2023-01-21 11:34:43.521423: step: 956/526, loss: 0.022774221375584602 2023-01-21 11:34:44.668679: step: 960/526, loss: 0.010180902667343616 2023-01-21 11:34:45.786961: step: 964/526, loss: 0.014191102236509323 2023-01-21 11:34:46.913496: step: 968/526, loss: 0.005507171154022217 2023-01-21 11:34:48.100548: step: 972/526, loss: 0.1057150810956955 2023-01-21 11:34:49.208248: step: 976/526, loss: 0.0006694793701171875 2023-01-21 11:34:50.387894: step: 980/526, loss: 0.017969131469726562 2023-01-21 11:34:51.585868: step: 984/526, loss: 0.034426022320985794 2023-01-21 11:34:52.751193: step: 988/526, loss: 0.06725058704614639 2023-01-21 11:34:53.890452: step: 992/526, loss: 0.09980545192956924 2023-01-21 11:34:55.047151: step: 996/526, loss: 0.006281280890107155 2023-01-21 11:34:56.256832: step: 1000/526, loss: 0.028983497992157936 2023-01-21 11:34:57.390599: step: 1004/526, loss: 0.01869940757751465 2023-01-21 11:34:58.587425: step: 1008/526, loss: 0.005263710394501686 2023-01-21 11:34:59.739739: step: 1012/526, loss: 0.029657935723662376 2023-01-21 11:35:00.902218: step: 1016/526, loss: 0.006319808773696423 2023-01-21 11:35:02.054302: step: 1020/526, loss: 0.01023800391703844 2023-01-21 11:35:03.209715: step: 1024/526, loss: 0.6852072477340698 2023-01-21 11:35:04.345164: step: 1028/526, loss: 0.011850547976791859 2023-01-21 11:35:05.509273: step: 1032/526, loss: 0.06599750369787216 2023-01-21 11:35:06.665024: step: 1036/526, loss: 0.003675317857414484 2023-01-21 11:35:07.829770: step: 1040/526, loss: 0.2622377574443817 2023-01-21 11:35:08.968922: step: 1044/526, loss: 0.019731713458895683 2023-01-21 11:35:10.146089: step: 1048/526, loss: 0.03058652952313423 2023-01-21 11:35:11.304627: step: 1052/526, loss: 0.04766717180609703 2023-01-21 11:35:12.443860: step: 1056/526, loss: 0.001121902372688055 2023-01-21 11:35:13.594135: step: 1060/526, loss: 0.014496040530502796 2023-01-21 11:35:14.759206: step: 1064/526, loss: 0.19050703942775726 2023-01-21 11:35:15.910225: step: 1068/526, loss: 0.016804981976747513 2023-01-21 11:35:17.078169: step: 1072/526, loss: 0.05126895755529404 2023-01-21 11:35:18.216187: step: 1076/526, loss: 0.029932117089629173 2023-01-21 11:35:19.381295: step: 1080/526, loss: 0.03689174726605415 2023-01-21 11:35:20.529958: step: 1084/526, loss: 0.018379880115389824 2023-01-21 11:35:21.682688: step: 1088/526, loss: 0.005470371339470148 2023-01-21 11:35:22.813498: step: 1092/526, loss: 0.017129898071289062 2023-01-21 11:35:23.992965: step: 1096/526, loss: 0.0006162643549032509 2023-01-21 11:35:25.140498: step: 1100/526, loss: 0.3256654739379883 2023-01-21 11:35:26.313832: step: 1104/526, loss: 0.001452636788599193 2023-01-21 11:35:27.483919: step: 1108/526, loss: 0.12877941131591797 2023-01-21 11:35:28.650752: step: 1112/526, loss: 0.0022145749535411596 2023-01-21 11:35:29.822793: step: 1116/526, loss: 0.017909908667206764 2023-01-21 11:35:30.980517: step: 1120/526, loss: 0.0266602523624897 2023-01-21 11:35:32.122626: step: 1124/526, loss: 0.011217355728149414 2023-01-21 11:35:33.264162: step: 1128/526, loss: 0.05366382747888565 2023-01-21 11:35:34.508988: step: 1132/526, loss: 0.004132366273552179 2023-01-21 11:35:35.670975: step: 1136/526, loss: 0.05129261314868927 2023-01-21 11:35:36.793252: step: 1140/526, loss: 0.012623215094208717 2023-01-21 11:35:37.917941: step: 1144/526, loss: 0.01526489295065403 2023-01-21 11:35:39.080742: step: 1148/526, loss: 0.0033623697236180305 2023-01-21 11:35:40.240264: step: 1152/526, loss: 0.0010648727184161544 2023-01-21 11:35:41.411416: step: 1156/526, loss: 0.04280386120080948 2023-01-21 11:35:42.568525: step: 1160/526, loss: 0.06125011295080185 2023-01-21 11:35:43.730173: step: 1164/526, loss: 0.0029735565185546875 2023-01-21 11:35:44.875693: step: 1168/526, loss: 0.10555687546730042 2023-01-21 11:35:46.055413: step: 1172/526, loss: 0.01919260062277317 2023-01-21 11:35:47.230256: step: 1176/526, loss: 0.0023391246795654297 2023-01-21 11:35:48.400503: step: 1180/526, loss: 0.03665027767419815 2023-01-21 11:35:49.547888: step: 1184/526, loss: 0.001665306044742465 2023-01-21 11:35:50.665505: step: 1188/526, loss: 0.00519218435510993 2023-01-21 11:35:51.833104: step: 1192/526, loss: 0.024559592828154564 2023-01-21 11:35:52.973276: step: 1196/526, loss: 0.0011234283447265625 2023-01-21 11:35:54.133187: step: 1200/526, loss: 0.003875160124152899 2023-01-21 11:35:55.302430: step: 1204/526, loss: 0.026830770075321198 2023-01-21 11:35:56.417965: step: 1208/526, loss: 7.731914229225367e-05 2023-01-21 11:35:57.605062: step: 1212/526, loss: 0.11283569037914276 2023-01-21 11:35:58.789099: step: 1216/526, loss: 0.03677120432257652 2023-01-21 11:35:59.971102: step: 1220/526, loss: 0.0005367517587728798 2023-01-21 11:36:01.125274: step: 1224/526, loss: 4.792213439941406e-05 2023-01-21 11:36:02.260796: step: 1228/526, loss: 0.042072393000125885 2023-01-21 11:36:03.396806: step: 1232/526, loss: 0.010038233362138271 2023-01-21 11:36:04.579223: step: 1236/526, loss: 0.047733500599861145 2023-01-21 11:36:05.717147: step: 1240/526, loss: 0.04888410493731499 2023-01-21 11:36:06.885480: step: 1244/526, loss: 0.0031155110336840153 2023-01-21 11:36:08.012460: step: 1248/526, loss: 0.039557456970214844 2023-01-21 11:36:09.147848: step: 1252/526, loss: 0.008772374130785465 2023-01-21 11:36:10.304344: step: 1256/526, loss: 0.039145469665527344 2023-01-21 11:36:11.459891: step: 1260/526, loss: 0.013201141729950905 2023-01-21 11:36:12.612979: step: 1264/526, loss: 0.00016717911057639867 2023-01-21 11:36:13.772765: step: 1268/526, loss: 0.06774864345788956 2023-01-21 11:36:14.915872: step: 1272/526, loss: 0.0020217897836118937 2023-01-21 11:36:16.080921: step: 1276/526, loss: 0.06763801723718643 2023-01-21 11:36:17.254884: step: 1280/526, loss: 0.1894427388906479 2023-01-21 11:36:18.395577: step: 1284/526, loss: 0.09113812446594238 2023-01-21 11:36:19.547294: step: 1288/526, loss: 0.07216491550207138 2023-01-21 11:36:20.680314: step: 1292/526, loss: 0.0015925407642498612 2023-01-21 11:36:21.827353: step: 1296/526, loss: 0.044083595275878906 2023-01-21 11:36:23.017770: step: 1300/526, loss: 0.03427238389849663 2023-01-21 11:36:24.160938: step: 1304/526, loss: 6.246566772460938e-05 2023-01-21 11:36:25.328010: step: 1308/526, loss: 0.030949022620916367 2023-01-21 11:36:26.444203: step: 1312/526, loss: 0.004623508546501398 2023-01-21 11:36:27.596906: step: 1316/526, loss: 0.01839613914489746 2023-01-21 11:36:28.764454: step: 1320/526, loss: 0.01687178574502468 2023-01-21 11:36:29.946835: step: 1324/526, loss: 0.00014228821964934468 2023-01-21 11:36:31.099157: step: 1328/526, loss: 0.08320808410644531 2023-01-21 11:36:32.233119: step: 1332/526, loss: 0.0011567592155188322 2023-01-21 11:36:33.395997: step: 1336/526, loss: 0.035341642796993256 2023-01-21 11:36:34.534630: step: 1340/526, loss: 0.007504272274672985 2023-01-21 11:36:35.681105: step: 1344/526, loss: 0.0002117156982421875 2023-01-21 11:36:36.819637: step: 1348/526, loss: 0.011392975226044655 2023-01-21 11:36:37.982554: step: 1352/526, loss: 0.0030735016334801912 2023-01-21 11:36:39.196085: step: 1356/526, loss: 0.022859174758195877 2023-01-21 11:36:40.326859: step: 1360/526, loss: 0.0028609277214854956 2023-01-21 11:36:41.477395: step: 1364/526, loss: 0.005170727148652077 2023-01-21 11:36:42.599380: step: 1368/526, loss: 0.06670016795396805 2023-01-21 11:36:43.748503: step: 1372/526, loss: 0.03216209635138512 2023-01-21 11:36:44.915847: step: 1376/526, loss: 0.002028340008109808 2023-01-21 11:36:46.024618: step: 1380/526, loss: 0.0033762454986572266 2023-01-21 11:36:47.163075: step: 1384/526, loss: 0.021107101812958717 2023-01-21 11:36:48.295428: step: 1388/526, loss: 0.07684288173913956 2023-01-21 11:36:49.438007: step: 1392/526, loss: 0.012826060876250267 2023-01-21 11:36:50.580329: step: 1396/526, loss: 0.027877237647771835 2023-01-21 11:36:51.713763: step: 1400/526, loss: 0.011816036887466908 2023-01-21 11:36:52.827181: step: 1404/526, loss: 2.6607514882925898e-05 2023-01-21 11:36:53.981736: step: 1408/526, loss: 0.00016078949556685984 2023-01-21 11:36:55.118749: step: 1412/526, loss: 0.03845162317156792 2023-01-21 11:36:56.246039: step: 1416/526, loss: 0.0684087723493576 2023-01-21 11:36:57.400023: step: 1420/526, loss: 0.04565334692597389 2023-01-21 11:36:58.563783: step: 1424/526, loss: 0.004640102386474609 2023-01-21 11:36:59.727303: step: 1428/526, loss: 0.015440654009580612 2023-01-21 11:37:00.912836: step: 1432/526, loss: 0.006703376770019531 2023-01-21 11:37:02.070858: step: 1436/526, loss: 0.0009208678966388106 2023-01-21 11:37:03.205232: step: 1440/526, loss: 0.001810264540836215 2023-01-21 11:37:04.346705: step: 1444/526, loss: 0.030171968042850494 2023-01-21 11:37:05.485860: step: 1448/526, loss: 0.013826275244355202 2023-01-21 11:37:06.656945: step: 1452/526, loss: 0.03273200988769531 2023-01-21 11:37:07.789797: step: 1456/526, loss: 0.020988276228308678 2023-01-21 11:37:08.958259: step: 1460/526, loss: 0.013797379098832607 2023-01-21 11:37:10.097602: step: 1464/526, loss: 0.022766876965761185 2023-01-21 11:37:11.232920: step: 1468/526, loss: 0.013074303045868874 2023-01-21 11:37:12.394455: step: 1472/526, loss: 0.009706258773803711 2023-01-21 11:37:13.517907: step: 1476/526, loss: 0.025360107421875 2023-01-21 11:37:14.680255: step: 1480/526, loss: 0.051818035542964935 2023-01-21 11:37:15.827047: step: 1484/526, loss: 0.023619651794433594 2023-01-21 11:37:16.997142: step: 1488/526, loss: 0.09215135872364044 2023-01-21 11:37:18.143684: step: 1492/526, loss: 0.059362031519412994 2023-01-21 11:37:19.309676: step: 1496/526, loss: 0.02387847937643528 2023-01-21 11:37:20.491239: step: 1500/526, loss: 0.02298736572265625 2023-01-21 11:37:21.652781: step: 1504/526, loss: 0.0031350136268883944 2023-01-21 11:37:22.799125: step: 1508/526, loss: 0.01287221908569336 2023-01-21 11:37:23.967582: step: 1512/526, loss: 0.007496643345803022 2023-01-21 11:37:25.132944: step: 1516/526, loss: 0.04456625133752823 2023-01-21 11:37:26.275009: step: 1520/526, loss: 0.036905575543642044 2023-01-21 11:37:27.420501: step: 1524/526, loss: 0.03574182838201523 2023-01-21 11:37:28.568768: step: 1528/526, loss: 0.020223427563905716 2023-01-21 11:37:29.725456: step: 1532/526, loss: 0.005044556222856045 2023-01-21 11:37:30.835021: step: 1536/526, loss: 0.01800546608865261 2023-01-21 11:37:31.976053: step: 1540/526, loss: 0.016041278839111328 2023-01-21 11:37:33.158584: step: 1544/526, loss: 0.009444808587431908 2023-01-21 11:37:34.366571: step: 1548/526, loss: 0.07086839526891708 2023-01-21 11:37:35.504510: step: 1552/526, loss: 0.0850076675415039 2023-01-21 11:37:36.621790: step: 1556/526, loss: 0.014645768329501152 2023-01-21 11:37:37.836958: step: 1560/526, loss: 0.00337562570348382 2023-01-21 11:37:38.976490: step: 1564/526, loss: 0.04385490342974663 2023-01-21 11:37:40.102862: step: 1568/526, loss: 0.03648347780108452 2023-01-21 11:37:41.255136: step: 1572/526, loss: 0.003805875778198242 2023-01-21 11:37:42.402696: step: 1576/526, loss: 0.01501169241964817 2023-01-21 11:37:43.566290: step: 1580/526, loss: 0.036768436431884766 2023-01-21 11:37:44.723039: step: 1584/526, loss: 0.005214214324951172 2023-01-21 11:37:45.838418: step: 1588/526, loss: 0.06704378128051758 2023-01-21 11:37:47.016246: step: 1592/526, loss: 0.0134324561804533 2023-01-21 11:37:48.169468: step: 1596/526, loss: 0.006333780474960804 2023-01-21 11:37:49.273482: step: 1600/526, loss: 0.003576993942260742 2023-01-21 11:37:50.426191: step: 1604/526, loss: 0.007436180487275124 2023-01-21 11:37:51.531206: step: 1608/526, loss: 0.10442066192626953 2023-01-21 11:37:52.679181: step: 1612/526, loss: 0.02715625800192356 2023-01-21 11:37:53.817062: step: 1616/526, loss: 0.0002761840878520161 2023-01-21 11:37:54.955044: step: 1620/526, loss: 0.02376708947122097 2023-01-21 11:37:56.120030: step: 1624/526, loss: 0.0425909049808979 2023-01-21 11:37:57.300535: step: 1628/526, loss: 0.007823002524673939 2023-01-21 11:37:58.441307: step: 1632/526, loss: 0.0069630625657737255 2023-01-21 11:37:59.570501: step: 1636/526, loss: 0.009379577822983265 2023-01-21 11:38:00.717616: step: 1640/526, loss: 0.04682111740112305 2023-01-21 11:38:01.860065: step: 1644/526, loss: 0.02999715879559517 2023-01-21 11:38:03.021220: step: 1648/526, loss: 5.0115584599552676e-05 2023-01-21 11:38:04.176093: step: 1652/526, loss: 0.1424911469221115 2023-01-21 11:38:05.295238: step: 1656/526, loss: 0.015407372266054153 2023-01-21 11:38:06.439028: step: 1660/526, loss: 0.013187408447265625 2023-01-21 11:38:07.571127: step: 1664/526, loss: 0.1278003752231598 2023-01-21 11:38:08.752191: step: 1668/526, loss: 0.04957018047571182 2023-01-21 11:38:09.915212: step: 1672/526, loss: 0.48830169439315796 2023-01-21 11:38:11.047597: step: 1676/526, loss: 0.04101238399744034 2023-01-21 11:38:12.171114: step: 1680/526, loss: 0.008026408962905407 2023-01-21 11:38:13.330575: step: 1684/526, loss: 0.06244621425867081 2023-01-21 11:38:14.486222: step: 1688/526, loss: 0.005667400546371937 2023-01-21 11:38:15.625113: step: 1692/526, loss: 0.029083536937832832 2023-01-21 11:38:16.802024: step: 1696/526, loss: 0.004749107174575329 2023-01-21 11:38:17.945906: step: 1700/526, loss: 0.008632849901914597 2023-01-21 11:38:19.101497: step: 1704/526, loss: 0.0051434519700706005 2023-01-21 11:38:20.265819: step: 1708/526, loss: 0.03107757493853569 2023-01-21 11:38:21.397340: step: 1712/526, loss: 0.06791611015796661 2023-01-21 11:38:22.572263: step: 1716/526, loss: 0.008446312509477139 2023-01-21 11:38:23.742460: step: 1720/526, loss: 0.0007907867548055947 2023-01-21 11:38:24.902233: step: 1724/526, loss: 0.05016594007611275 2023-01-21 11:38:26.057837: step: 1728/526, loss: 0.006982707884162664 2023-01-21 11:38:27.197801: step: 1732/526, loss: 0.012863541021943092 2023-01-21 11:38:28.358361: step: 1736/526, loss: 0.04176607355475426 2023-01-21 11:38:29.493060: step: 1740/526, loss: 0.023456193506717682 2023-01-21 11:38:30.618071: step: 1744/526, loss: 0.0015418052207678556 2023-01-21 11:38:31.765601: step: 1748/526, loss: 0.06011927127838135 2023-01-21 11:38:32.911129: step: 1752/526, loss: 0.012541437521576881 2023-01-21 11:38:34.097602: step: 1756/526, loss: 0.02001648023724556 2023-01-21 11:38:35.272764: step: 1760/526, loss: 0.0074253082275390625 2023-01-21 11:38:36.446866: step: 1764/526, loss: 0.05597114562988281 2023-01-21 11:38:37.600601: step: 1768/526, loss: 0.049950502812862396 2023-01-21 11:38:38.758077: step: 1772/526, loss: 0.06593179702758789 2023-01-21 11:38:39.913584: step: 1776/526, loss: 0.0014391898876056075 2023-01-21 11:38:41.102731: step: 1780/526, loss: 0.039015673100948334 2023-01-21 11:38:42.264402: step: 1784/526, loss: 0.02459716796875 2023-01-21 11:38:43.424946: step: 1788/526, loss: 0.020014189183712006 2023-01-21 11:38:44.563983: step: 1792/526, loss: 0.031155109405517578 2023-01-21 11:38:45.714129: step: 1796/526, loss: 0.006568908225744963 2023-01-21 11:38:46.870210: step: 1800/526, loss: 0.0071121216751635075 2023-01-21 11:38:48.089193: step: 1804/526, loss: 0.0051132203079760075 2023-01-21 11:38:49.262097: step: 1808/526, loss: 0.002435970352962613 2023-01-21 11:38:50.412684: step: 1812/526, loss: 0.005366229917854071 2023-01-21 11:38:51.558527: step: 1816/526, loss: 0.0003819465928245336 2023-01-21 11:38:52.752745: step: 1820/526, loss: 0.033179473131895065 2023-01-21 11:38:53.931604: step: 1824/526, loss: 0.0011067867744714022 2023-01-21 11:38:55.122924: step: 1828/526, loss: 0.004168510437011719 2023-01-21 11:38:56.335337: step: 1832/526, loss: 0.003395223757252097 2023-01-21 11:38:57.489266: step: 1836/526, loss: 0.003672695253044367 2023-01-21 11:38:58.630087: step: 1840/526, loss: 0.00021986961655784398 2023-01-21 11:38:59.795825: step: 1844/526, loss: 0.04323158040642738 2023-01-21 11:39:00.939518: step: 1848/526, loss: 0.03374312072992325 2023-01-21 11:39:02.143749: step: 1852/526, loss: 0.028297901153564453 2023-01-21 11:39:03.305477: step: 1856/526, loss: 0.01671133004128933 2023-01-21 11:39:04.435513: step: 1860/526, loss: 0.009083127602934837 2023-01-21 11:39:05.624237: step: 1864/526, loss: 0.006672191899269819 2023-01-21 11:39:06.765799: step: 1868/526, loss: 0.012367821298539639 2023-01-21 11:39:07.910735: step: 1872/526, loss: 0.04822073131799698 2023-01-21 11:39:09.052031: step: 1876/526, loss: 0.004676866345107555 2023-01-21 11:39:10.175704: step: 1880/526, loss: 0.000821018242277205 2023-01-21 11:39:11.368636: step: 1884/526, loss: 0.17603187263011932 2023-01-21 11:39:12.540799: step: 1888/526, loss: 0.04731273651123047 2023-01-21 11:39:13.706049: step: 1892/526, loss: 0.012217044830322266 2023-01-21 11:39:14.873393: step: 1896/526, loss: 0.13915623724460602 2023-01-21 11:39:16.032652: step: 1900/526, loss: 0.0026071546599268913 2023-01-21 11:39:17.188809: step: 1904/526, loss: 0.00045003893319517374 2023-01-21 11:39:18.329150: step: 1908/526, loss: 0.006118870340287685 2023-01-21 11:39:19.484290: step: 1912/526, loss: 0.04653315618634224 2023-01-21 11:39:20.672794: step: 1916/526, loss: 0.03326454386115074 2023-01-21 11:39:21.813789: step: 1920/526, loss: 0.0023725510109215975 2023-01-21 11:39:22.997951: step: 1924/526, loss: 0.0069987536408007145 2023-01-21 11:39:24.141692: step: 1928/526, loss: 0.0011455535423010588 2023-01-21 11:39:25.304107: step: 1932/526, loss: 0.01862793043255806 2023-01-21 11:39:26.429802: step: 1936/526, loss: 0.11043939739465714 2023-01-21 11:39:27.587683: step: 1940/526, loss: 0.007619666866958141 2023-01-21 11:39:28.727609: step: 1944/526, loss: 0.078722283244133 2023-01-21 11:39:29.892285: step: 1948/526, loss: 0.052454378455877304 2023-01-21 11:39:31.036887: step: 1952/526, loss: 0.022197771817445755 2023-01-21 11:39:32.192188: step: 1956/526, loss: 5.53131121705519e-06 2023-01-21 11:39:33.321926: step: 1960/526, loss: 0.06036224216222763 2023-01-21 11:39:34.487690: step: 1964/526, loss: 0.06706800311803818 2023-01-21 11:39:35.629894: step: 1968/526, loss: 0.02777853049337864 2023-01-21 11:39:36.802032: step: 1972/526, loss: 0.28656381368637085 2023-01-21 11:39:37.959236: step: 1976/526, loss: 0.07560410350561142 2023-01-21 11:39:39.106583: step: 1980/526, loss: 0.0016263008583337069 2023-01-21 11:39:40.254193: step: 1984/526, loss: 0.01938037946820259 2023-01-21 11:39:41.408150: step: 1988/526, loss: 0.01508321799337864 2023-01-21 11:39:42.593657: step: 1992/526, loss: 0.03717451170086861 2023-01-21 11:39:43.742395: step: 1996/526, loss: 0.03589196130633354 2023-01-21 11:39:44.883124: step: 2000/526, loss: 0.052363112568855286 2023-01-21 11:39:46.026833: step: 2004/526, loss: 0.01964740827679634 2023-01-21 11:39:47.181078: step: 2008/526, loss: 0.035630084574222565 2023-01-21 11:39:48.340556: step: 2012/526, loss: 0.035118866711854935 2023-01-21 11:39:49.464911: step: 2016/526, loss: 0.0008286953088827431 2023-01-21 11:39:50.610096: step: 2020/526, loss: 0.044460296630859375 2023-01-21 11:39:51.764547: step: 2024/526, loss: 0.01126632746309042 2023-01-21 11:39:52.927397: step: 2028/526, loss: 0.016485977917909622 2023-01-21 11:39:54.059092: step: 2032/526, loss: 0.007075023837387562 2023-01-21 11:39:55.223588: step: 2036/526, loss: 0.03506527096033096 2023-01-21 11:39:56.386695: step: 2040/526, loss: 0.005649232771247625 2023-01-21 11:39:57.516656: step: 2044/526, loss: 0.003901576856151223 2023-01-21 11:39:58.663317: step: 2048/526, loss: 0.00407066335901618 2023-01-21 11:39:59.814152: step: 2052/526, loss: 0.059488486498594284 2023-01-21 11:40:00.979973: step: 2056/526, loss: 0.011468411423265934 2023-01-21 11:40:02.127764: step: 2060/526, loss: 0.001214742660522461 2023-01-21 11:40:03.284831: step: 2064/526, loss: 0.0040495870634913445 2023-01-21 11:40:04.428460: step: 2068/526, loss: 0.002097702119499445 2023-01-21 11:40:05.606948: step: 2072/526, loss: 0.020907068625092506 2023-01-21 11:40:06.741231: step: 2076/526, loss: 0.0009271144517697394 2023-01-21 11:40:07.880247: step: 2080/526, loss: 0.05711951479315758 2023-01-21 11:40:09.024684: step: 2084/526, loss: 0.06822071224451065 2023-01-21 11:40:10.177205: step: 2088/526, loss: 0.0037601469084620476 2023-01-21 11:40:11.353456: step: 2092/526, loss: 0.00111980433575809 2023-01-21 11:40:12.525940: step: 2096/526, loss: 1.564025842526462e-05 2023-01-21 11:40:13.683419: step: 2100/526, loss: 0.02524433098733425 2023-01-21 11:40:14.834596: step: 2104/526, loss: 0.0006954193231649697 ================================================== Loss: 0.032 -------------------- Dev: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Test: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Chinese: {'event': {'p': 0.5747126436781609, 'r': 0.9259259259259259, 'f1': 0.7092198581560283}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Korean: {'event': {'p': 0.6166666666666667, 'r': 0.5873015873015873, 'f1': 0.6016260162601625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 17 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:41:09.144164: step: 4/526, loss: 0.02712268754839897 2023-01-21 11:41:10.270678: step: 8/526, loss: 0.001408039010129869 2023-01-21 11:41:11.450652: step: 12/526, loss: 0.012801457196474075 2023-01-21 11:41:12.619079: step: 16/526, loss: 0.0024095536209642887 2023-01-21 11:41:13.738615: step: 20/526, loss: 0.09601977467536926 2023-01-21 11:41:14.916272: step: 24/526, loss: 0.0018514633411541581 2023-01-21 11:41:16.059628: step: 28/526, loss: 0.025252724066376686 2023-01-21 11:41:17.235701: step: 32/526, loss: 0.0019448280800133944 2023-01-21 11:41:18.450300: step: 36/526, loss: 0.08267202973365784 2023-01-21 11:41:19.591474: step: 40/526, loss: 0.05796356499195099 2023-01-21 11:41:20.739253: step: 44/526, loss: 0.004552364349365234 2023-01-21 11:41:21.900927: step: 48/526, loss: 0.03374652937054634 2023-01-21 11:41:23.046299: step: 52/526, loss: 0.005305671598762274 2023-01-21 11:41:24.188240: step: 56/526, loss: 0.07305049896240234 2023-01-21 11:41:25.317024: step: 60/526, loss: 0.06041393429040909 2023-01-21 11:41:26.435535: step: 64/526, loss: 0.004543781280517578 2023-01-21 11:41:27.591860: step: 68/526, loss: 0.040546514093875885 2023-01-21 11:41:28.747528: step: 72/526, loss: 0.0002662658516783267 2023-01-21 11:41:29.886153: step: 76/526, loss: 0.0124657629057765 2023-01-21 11:41:31.046212: step: 80/526, loss: 0.08264666050672531 2023-01-21 11:41:32.203275: step: 84/526, loss: 0.005223465152084827 2023-01-21 11:41:33.360205: step: 88/526, loss: 0.015918731689453125 2023-01-21 11:41:34.505301: step: 92/526, loss: 0.003290843917056918 2023-01-21 11:41:35.641882: step: 96/526, loss: 0.01964450068771839 2023-01-21 11:41:36.776936: step: 100/526, loss: 0.029907705262303352 2023-01-21 11:41:37.930765: step: 104/526, loss: 0.0026391029823571444 2023-01-21 11:41:39.090338: step: 108/526, loss: 0.01197052001953125 2023-01-21 11:41:40.228179: step: 112/526, loss: 0.0004809380043298006 2023-01-21 11:41:41.338074: step: 116/526, loss: 0.02562703564763069 2023-01-21 11:41:42.495847: step: 120/526, loss: 1.3805336952209473 2023-01-21 11:41:43.638016: step: 124/526, loss: 0.030428458005189896 2023-01-21 11:41:44.777811: step: 128/526, loss: 0.0010852813720703125 2023-01-21 11:41:45.942371: step: 132/526, loss: 0.029212377965450287 2023-01-21 11:41:47.105798: step: 136/526, loss: 0.010221099480986595 2023-01-21 11:41:48.291771: step: 140/526, loss: 0.0008207320934161544 2023-01-21 11:41:49.409999: step: 144/526, loss: 0.03578338772058487 2023-01-21 11:41:50.572711: step: 148/526, loss: 0.028288459405303 2023-01-21 11:41:51.721253: step: 152/526, loss: 0.028358841314911842 2023-01-21 11:41:52.853211: step: 156/526, loss: -7.152557373046875e-06 2023-01-21 11:41:54.007299: step: 160/526, loss: 0.038114070892333984 2023-01-21 11:41:55.145774: step: 164/526, loss: 0.0076299188658595085 2023-01-21 11:41:56.285763: step: 168/526, loss: 0.02585907094180584 2023-01-21 11:41:57.432936: step: 172/526, loss: 0.0006128311506472528 2023-01-21 11:41:58.596541: step: 176/526, loss: 0.020081330090761185 2023-01-21 11:41:59.769068: step: 180/526, loss: 0.0035316466819494963 2023-01-21 11:42:00.878411: step: 184/526, loss: 0.0006196022150106728 2023-01-21 11:42:02.054884: step: 188/526, loss: 0.04908294603228569 2023-01-21 11:42:03.190915: step: 192/526, loss: 0.0001585006684763357 2023-01-21 11:42:04.359860: step: 196/526, loss: 0.03778076171875 2023-01-21 11:42:05.510855: step: 200/526, loss: 0.0007956505287438631 2023-01-21 11:42:06.696440: step: 204/526, loss: 0.0016698837280273438 2023-01-21 11:42:07.814543: step: 208/526, loss: 0.01373300515115261 2023-01-21 11:42:08.969471: step: 212/526, loss: 0.006088066380470991 2023-01-21 11:42:10.146962: step: 216/526, loss: 0.050337694585323334 2023-01-21 11:42:11.296673: step: 220/526, loss: 0.005057477857917547 2023-01-21 11:42:12.424367: step: 224/526, loss: 0.0005259513854980469 2023-01-21 11:42:13.603534: step: 228/526, loss: 0.018445681780576706 2023-01-21 11:42:14.750961: step: 232/526, loss: 0.001512980554252863 2023-01-21 11:42:15.910316: step: 236/526, loss: 0.000976181065198034 2023-01-21 11:42:17.045208: step: 240/526, loss: 0.008683395572006702 2023-01-21 11:42:18.197155: step: 244/526, loss: 0.03286619111895561 2023-01-21 11:42:19.349658: step: 248/526, loss: 0.0032364847138524055 2023-01-21 11:42:20.501837: step: 252/526, loss: 0.07784061133861542 2023-01-21 11:42:21.673918: step: 256/526, loss: 0.046559907495975494 2023-01-21 11:42:22.784665: step: 260/526, loss: 0.004661845974624157 2023-01-21 11:42:23.934860: step: 264/526, loss: 0.001417827676050365 2023-01-21 11:42:25.064603: step: 268/526, loss: 1.7642974853515625e-05 2023-01-21 11:42:26.187304: step: 272/526, loss: 0.007066536229103804 2023-01-21 11:42:27.320911: step: 276/526, loss: 0.00266265869140625 2023-01-21 11:42:28.505452: step: 280/526, loss: 0.00022516549506690353 2023-01-21 11:42:29.650016: step: 284/526, loss: 6.246566772460938e-05 2023-01-21 11:42:30.791689: step: 288/526, loss: 0.0021077156998217106 2023-01-21 11:42:31.934760: step: 292/526, loss: 0.023946000263094902 2023-01-21 11:42:33.043882: step: 296/526, loss: 0.004250526428222656 2023-01-21 11:42:34.167072: step: 300/526, loss: 0.023140527307987213 2023-01-21 11:42:35.328611: step: 304/526, loss: 0.008922828361392021 2023-01-21 11:42:36.482526: step: 308/526, loss: 0.008628463372588158 2023-01-21 11:42:37.639315: step: 312/526, loss: 0.013683510944247246 2023-01-21 11:42:38.788543: step: 316/526, loss: 0.038454435765743256 2023-01-21 11:42:39.939640: step: 320/526, loss: 0.0037343979347497225 2023-01-21 11:42:41.067956: step: 324/526, loss: 4.76837158203125e-06 2023-01-21 11:42:42.234418: step: 328/526, loss: 0.006391811650246382 2023-01-21 11:42:43.368679: step: 332/526, loss: 0.0008426666026934981 2023-01-21 11:42:44.510767: step: 336/526, loss: 0.01620674319565296 2023-01-21 11:42:45.620815: step: 340/526, loss: 0.0063485149294137955 2023-01-21 11:42:46.767878: step: 344/526, loss: 0.05706596374511719 2023-01-21 11:42:47.906963: step: 348/526, loss: 0.014854812994599342 2023-01-21 11:42:49.075860: step: 352/526, loss: 0.018103981390595436 2023-01-21 11:42:50.210815: step: 356/526, loss: 0.013170433230698109 2023-01-21 11:42:51.344610: step: 360/526, loss: 0.00872888509184122 2023-01-21 11:42:52.486690: step: 364/526, loss: 0.0010667801834642887 2023-01-21 11:42:53.619681: step: 368/526, loss: 0.010018778033554554 2023-01-21 11:42:54.772032: step: 372/526, loss: 0.003550910856574774 2023-01-21 11:42:55.906987: step: 376/526, loss: 0.0032329559326171875 2023-01-21 11:42:57.044430: step: 380/526, loss: 0.036969948559999466 2023-01-21 11:42:58.170186: step: 384/526, loss: 0.01533060148358345 2023-01-21 11:42:59.299324: step: 388/526, loss: 0.0010803223121911287 2023-01-21 11:43:00.422210: step: 392/526, loss: 0.03152661398053169 2023-01-21 11:43:01.556095: step: 396/526, loss: 0.03798218071460724 2023-01-21 11:43:02.720008: step: 400/526, loss: 0.022547628730535507 2023-01-21 11:43:03.879377: step: 404/526, loss: 0.0395687110722065 2023-01-21 11:43:05.014829: step: 408/526, loss: 0.018436836078763008 2023-01-21 11:43:06.139540: step: 412/526, loss: 0.0046918392181396484 2023-01-21 11:43:07.286476: step: 416/526, loss: 0.002600288251414895 2023-01-21 11:43:08.421644: step: 420/526, loss: 0.00018777848163153976 2023-01-21 11:43:09.526320: step: 424/526, loss: 0.0015184402000159025 2023-01-21 11:43:10.645257: step: 428/526, loss: 0.008309555239975452 2023-01-21 11:43:11.764296: step: 432/526, loss: 0.014017963781952858 2023-01-21 11:43:12.910943: step: 436/526, loss: 0.017172621563076973 2023-01-21 11:43:14.061948: step: 440/526, loss: 0.008749008178710938 2023-01-21 11:43:15.189657: step: 444/526, loss: 0.029176950454711914 2023-01-21 11:43:16.354351: step: 448/526, loss: 0.004161929711699486 2023-01-21 11:43:17.511880: step: 452/526, loss: 0.13465367257595062 2023-01-21 11:43:18.648473: step: 456/526, loss: 0.00015053153038024902 2023-01-21 11:43:19.790134: step: 460/526, loss: 0.015687370672822 2023-01-21 11:43:20.912190: step: 464/526, loss: 0.0012549401726573706 2023-01-21 11:43:22.039534: step: 468/526, loss: 0.0904931128025055 2023-01-21 11:43:23.187131: step: 472/526, loss: 0.0018959998851642013 2023-01-21 11:43:24.314893: step: 476/526, loss: 0.007970333099365234 2023-01-21 11:43:25.425037: step: 480/526, loss: 0.0007155895000323653 2023-01-21 11:43:26.590024: step: 484/526, loss: 0.019067956134676933 2023-01-21 11:43:27.728976: step: 488/526, loss: 0.039622265845537186 2023-01-21 11:43:28.884495: step: 492/526, loss: 0.009263277053833008 2023-01-21 11:43:30.010410: step: 496/526, loss: 0.0007642745622433722 2023-01-21 11:43:31.176764: step: 500/526, loss: 0.02133016474545002 2023-01-21 11:43:32.319081: step: 504/526, loss: 0.001066494034603238 2023-01-21 11:43:33.433427: step: 508/526, loss: 0.018526649102568626 2023-01-21 11:43:34.579010: step: 512/526, loss: 0.020804978907108307 2023-01-21 11:43:35.746004: step: 516/526, loss: 0.00030651091947220266 2023-01-21 11:43:36.919780: step: 520/526, loss: 0.004455185029655695 2023-01-21 11:43:38.079685: step: 524/526, loss: 0.017485808581113815 2023-01-21 11:43:39.241083: step: 528/526, loss: 0.07302455604076385 2023-01-21 11:43:40.366732: step: 532/526, loss: 0.013042259030044079 2023-01-21 11:43:41.497914: step: 536/526, loss: 0.10113011300563812 2023-01-21 11:43:42.645056: step: 540/526, loss: 0.09660720825195312 2023-01-21 11:43:43.789187: step: 544/526, loss: 0.1760028898715973 2023-01-21 11:43:44.927026: step: 548/526, loss: 0.03047924116253853 2023-01-21 11:43:46.042673: step: 552/526, loss: 0.00045604706974700093 2023-01-21 11:43:47.186010: step: 556/526, loss: 0.019284822046756744 2023-01-21 11:43:48.349598: step: 560/526, loss: 0.1884039044380188 2023-01-21 11:43:49.474012: step: 564/526, loss: 0.002670383546501398 2023-01-21 11:43:50.621916: step: 568/526, loss: 0.001455974648706615 2023-01-21 11:43:51.795882: step: 572/526, loss: 0.0010818482842296362 2023-01-21 11:43:52.914076: step: 576/526, loss: 0.027611160650849342 2023-01-21 11:43:54.038046: step: 580/526, loss: 0.01098489761352539 2023-01-21 11:43:55.186995: step: 584/526, loss: 0.00024318695068359375 2023-01-21 11:43:56.330035: step: 588/526, loss: 0.0003849029599223286 2023-01-21 11:43:57.476775: step: 592/526, loss: 0.0005496024969033897 2023-01-21 11:43:58.621793: step: 596/526, loss: 0.023192215710878372 2023-01-21 11:43:59.752225: step: 600/526, loss: 0.00019583702669478953 2023-01-21 11:44:00.900272: step: 604/526, loss: 0.03303699195384979 2023-01-21 11:44:02.022834: step: 608/526, loss: 0.005811119452118874 2023-01-21 11:44:03.178594: step: 612/526, loss: 0.019014358520507812 2023-01-21 11:44:04.329343: step: 616/526, loss: 0.016980934888124466 2023-01-21 11:44:05.463217: step: 620/526, loss: 0.020771218463778496 2023-01-21 11:44:06.614108: step: 624/526, loss: 0.007435798645019531 2023-01-21 11:44:07.773098: step: 628/526, loss: 0.009545231238007545 2023-01-21 11:44:08.914095: step: 632/526, loss: 0.0004153251647949219 2023-01-21 11:44:10.045417: step: 636/526, loss: 0.0009140014881268144 2023-01-21 11:44:11.188918: step: 640/526, loss: 0.0005583763122558594 2023-01-21 11:44:12.337834: step: 644/526, loss: 0.014668464660644531 2023-01-21 11:44:13.474124: step: 648/526, loss: 0.0015415191883221269 2023-01-21 11:44:14.619114: step: 652/526, loss: 0.0037648677825927734 2023-01-21 11:44:15.744563: step: 656/526, loss: 0.0073036192916333675 2023-01-21 11:44:16.878501: step: 660/526, loss: 0.04812183603644371 2023-01-21 11:44:17.993312: step: 664/526, loss: 0.010242628864943981 2023-01-21 11:44:19.129936: step: 668/526, loss: 0.010446930304169655 2023-01-21 11:44:20.261267: step: 672/526, loss: 0.0010035515297204256 2023-01-21 11:44:21.382571: step: 676/526, loss: 0.0029952051118016243 2023-01-21 11:44:22.513951: step: 680/526, loss: 0.013922125101089478 2023-01-21 11:44:23.624477: step: 684/526, loss: 0.007066535763442516 2023-01-21 11:44:24.772658: step: 688/526, loss: 0.00043640134390443563 2023-01-21 11:44:25.883224: step: 692/526, loss: 6.103515261202119e-06 2023-01-21 11:44:27.030635: step: 696/526, loss: 0.0038742662873119116 2023-01-21 11:44:28.150148: step: 700/526, loss: 0.00345191964879632 2023-01-21 11:44:29.305587: step: 704/526, loss: 0.052428338676691055 2023-01-21 11:44:30.451547: step: 708/526, loss: 0.0028247355949133635 2023-01-21 11:44:31.659843: step: 712/526, loss: 0.0007513046730309725 2023-01-21 11:44:32.809664: step: 716/526, loss: 0.011724472045898438 2023-01-21 11:44:33.937616: step: 720/526, loss: 7.221699343062937e-05 2023-01-21 11:44:35.066805: step: 724/526, loss: 0.018499374389648438 2023-01-21 11:44:36.188604: step: 728/526, loss: 0.024643326178193092 2023-01-21 11:44:37.307287: step: 732/526, loss: 0.010480117984116077 2023-01-21 11:44:38.451247: step: 736/526, loss: 0.0007181167602539062 2023-01-21 11:44:39.591617: step: 740/526, loss: 0.008913851343095303 2023-01-21 11:44:40.738878: step: 744/526, loss: 5.4168704082258046e-05 2023-01-21 11:44:41.895558: step: 748/526, loss: 0.009173410944640636 2023-01-21 11:44:43.087568: step: 752/526, loss: 0.0004017830069642514 2023-01-21 11:44:44.213085: step: 756/526, loss: 0.13221673667430878 2023-01-21 11:44:45.375545: step: 760/526, loss: 0.1255381554365158 2023-01-21 11:44:46.531763: step: 764/526, loss: 0.005768394563347101 2023-01-21 11:44:47.640533: step: 768/526, loss: 0.033943939954042435 2023-01-21 11:44:48.790573: step: 772/526, loss: 0.037000179290771484 2023-01-21 11:44:49.931969: step: 776/526, loss: 0.0602506622672081 2023-01-21 11:44:51.082928: step: 780/526, loss: 0.03467750549316406 2023-01-21 11:44:52.193767: step: 784/526, loss: 0.00912933424115181 2023-01-21 11:44:53.342182: step: 788/526, loss: 0.0054340362548828125 2023-01-21 11:44:54.482904: step: 792/526, loss: 0.0002578258572611958 2023-01-21 11:44:55.634294: step: 796/526, loss: 0.3241601884365082 2023-01-21 11:44:56.793022: step: 800/526, loss: 0.035428620874881744 2023-01-21 11:44:57.934770: step: 804/526, loss: 0.006259727291762829 2023-01-21 11:44:59.096899: step: 808/526, loss: 0.01959390565752983 2023-01-21 11:45:00.257794: step: 812/526, loss: 0.0006286621210165322 2023-01-21 11:45:01.361307: step: 816/526, loss: 0.0002814769686665386 2023-01-21 11:45:02.504575: step: 820/526, loss: 0.02241516299545765 2023-01-21 11:45:03.642422: step: 824/526, loss: 0.00877304095774889 2023-01-21 11:45:04.801288: step: 828/526, loss: 0.03837409242987633 2023-01-21 11:45:05.930575: step: 832/526, loss: 0.007070064544677734 2023-01-21 11:45:07.068354: step: 836/526, loss: 0.00011606216139625758 2023-01-21 11:45:08.224079: step: 840/526, loss: 0.0033384801354259253 2023-01-21 11:45:09.368490: step: 844/526, loss: 0.008058643899857998 2023-01-21 11:45:10.558803: step: 848/526, loss: 0.14827671647071838 2023-01-21 11:45:11.668108: step: 852/526, loss: 0.020801736041903496 2023-01-21 11:45:12.786438: step: 856/526, loss: 0.022509479895234108 2023-01-21 11:45:13.920574: step: 860/526, loss: 0.22603388130664825 2023-01-21 11:45:15.048206: step: 864/526, loss: 0.035506583750247955 2023-01-21 11:45:16.180231: step: 868/526, loss: 0.0006921768072061241 2023-01-21 11:45:17.321245: step: 872/526, loss: 0.0003002643643412739 2023-01-21 11:45:18.454464: step: 876/526, loss: 0.01151504460722208 2023-01-21 11:45:19.600663: step: 880/526, loss: 0.07331258058547974 2023-01-21 11:45:20.741606: step: 884/526, loss: 0.0001354217529296875 2023-01-21 11:45:21.876585: step: 888/526, loss: 0.4261000454425812 2023-01-21 11:45:23.008188: step: 892/526, loss: 0.0052032470703125 2023-01-21 11:45:24.132628: step: 896/526, loss: 0.10094952583312988 2023-01-21 11:45:25.274008: step: 900/526, loss: 0.01489410363137722 2023-01-21 11:45:26.388090: step: 904/526, loss: 0.014415360055863857 2023-01-21 11:45:27.510196: step: 908/526, loss: 0.004832458682358265 2023-01-21 11:45:28.629239: step: 912/526, loss: 0.010731220245361328 2023-01-21 11:45:29.814111: step: 916/526, loss: 0.031651876866817474 2023-01-21 11:45:30.961866: step: 920/526, loss: 0.00012969970703125 2023-01-21 11:45:32.093059: step: 924/526, loss: 0.003452110104262829 2023-01-21 11:45:33.211561: step: 928/526, loss: 0.0018125057686120272 2023-01-21 11:45:34.360090: step: 932/526, loss: 0.0011042595142498612 2023-01-21 11:45:35.522285: step: 936/526, loss: 0.0008923053974285722 2023-01-21 11:45:36.638480: step: 940/526, loss: 0.062303829938173294 2023-01-21 11:45:37.784956: step: 944/526, loss: 0.0046176910400390625 2023-01-21 11:45:38.951538: step: 948/526, loss: 0.013587666675448418 2023-01-21 11:45:40.076137: step: 952/526, loss: 0.027117349207401276 2023-01-21 11:45:41.235453: step: 956/526, loss: 0.01620473898947239 2023-01-21 11:45:42.386674: step: 960/526, loss: 0.024223709478974342 2023-01-21 11:45:43.521508: step: 964/526, loss: 0.0005677223089151084 2023-01-21 11:45:44.677688: step: 968/526, loss: 0.049275971949100494 2023-01-21 11:45:45.808036: step: 972/526, loss: 0.013196706771850586 2023-01-21 11:45:46.957750: step: 976/526, loss: 0.004477214999496937 2023-01-21 11:45:48.091540: step: 980/526, loss: 0.0018205642700195312 2023-01-21 11:45:49.222931: step: 984/526, loss: 0.0017698288429528475 2023-01-21 11:45:50.362671: step: 988/526, loss: 0.08145418763160706 2023-01-21 11:45:51.501757: step: 992/526, loss: 0.03809700161218643 2023-01-21 11:45:52.683570: step: 996/526, loss: 0.0009351730113849044 2023-01-21 11:45:53.826714: step: 1000/526, loss: 0.009448624216020107 2023-01-21 11:45:54.965331: step: 1004/526, loss: 0.005946922115981579 2023-01-21 11:45:56.092606: step: 1008/526, loss: 0.0005452156183309853 2023-01-21 11:45:57.231165: step: 1012/526, loss: 0.0021003722213208675 2023-01-21 11:45:58.361233: step: 1016/526, loss: 1.2731552487821318e-05 2023-01-21 11:45:59.523153: step: 1020/526, loss: 0.01615314558148384 2023-01-21 11:46:00.651413: step: 1024/526, loss: 0.04951968044042587 2023-01-21 11:46:01.791942: step: 1028/526, loss: 0.0113379480317235 2023-01-21 11:46:02.914573: step: 1032/526, loss: 0.003451442578807473 2023-01-21 11:46:04.047004: step: 1036/526, loss: 0.002733421279117465 2023-01-21 11:46:05.228021: step: 1040/526, loss: 0.059688374400138855 2023-01-21 11:46:06.366961: step: 1044/526, loss: 0.05275745689868927 2023-01-21 11:46:07.522803: step: 1048/526, loss: 0.0033521652221679688 2023-01-21 11:46:08.667264: step: 1052/526, loss: 8.0108642578125e-05 2023-01-21 11:46:09.771420: step: 1056/526, loss: 0.0004809856473002583 2023-01-21 11:46:10.901821: step: 1060/526, loss: 0.02368602715432644 2023-01-21 11:46:12.054001: step: 1064/526, loss: 0.025569915771484375 2023-01-21 11:46:13.184801: step: 1068/526, loss: 0.0009590148692950606 2023-01-21 11:46:14.316737: step: 1072/526, loss: 0.006788397207856178 2023-01-21 11:46:15.469762: step: 1076/526, loss: 0.007541084196418524 2023-01-21 11:46:16.591188: step: 1080/526, loss: 0.02177600748836994 2023-01-21 11:46:17.753559: step: 1084/526, loss: 0.010368348099291325 2023-01-21 11:46:18.863257: step: 1088/526, loss: 0.02678089216351509 2023-01-21 11:46:20.005258: step: 1092/526, loss: 0.00021057129197288305 2023-01-21 11:46:21.160503: step: 1096/526, loss: 0.06157388910651207 2023-01-21 11:46:22.302371: step: 1100/526, loss: 0.006807089317589998 2023-01-21 11:46:23.451359: step: 1104/526, loss: 0.0009675025357864797 2023-01-21 11:46:24.589144: step: 1108/526, loss: 0.0021232604049146175 2023-01-21 11:46:25.733430: step: 1112/526, loss: 0.013833428733050823 2023-01-21 11:46:26.856147: step: 1116/526, loss: 0.021033724769949913 2023-01-21 11:46:27.982742: step: 1120/526, loss: 0.03125705569982529 2023-01-21 11:46:29.094146: step: 1124/526, loss: 0.0031599043868482113 2023-01-21 11:46:30.235553: step: 1128/526, loss: 0.00025959016056731343 2023-01-21 11:46:31.391677: step: 1132/526, loss: 0.01864500157535076 2023-01-21 11:46:32.522229: step: 1136/526, loss: 0.03527240827679634 2023-01-21 11:46:33.643880: step: 1140/526, loss: 0.02262091636657715 2023-01-21 11:46:34.811240: step: 1144/526, loss: 0.008797263726592064 2023-01-21 11:46:35.943036: step: 1148/526, loss: 0.02583150938153267 2023-01-21 11:46:37.088011: step: 1152/526, loss: 0.05012664943933487 2023-01-21 11:46:38.263249: step: 1156/526, loss: -7.4386593951203395e-06 2023-01-21 11:46:39.405277: step: 1160/526, loss: 0.0020987512543797493 2023-01-21 11:46:40.549794: step: 1164/526, loss: 0.0019715309608727694 2023-01-21 11:46:41.664191: step: 1168/526, loss: 0.0058629512786865234 2023-01-21 11:46:42.778464: step: 1172/526, loss: 0.004943275358527899 2023-01-21 11:46:43.916210: step: 1176/526, loss: 0.053217317909002304 2023-01-21 11:46:45.040120: step: 1180/526, loss: 0.016385745257139206 2023-01-21 11:46:46.170353: step: 1184/526, loss: 0.017311764881014824 2023-01-21 11:46:47.325576: step: 1188/526, loss: 0.027155686169862747 2023-01-21 11:46:48.462535: step: 1192/526, loss: 0.09188447892665863 2023-01-21 11:46:49.602184: step: 1196/526, loss: 0.024518754333257675 2023-01-21 11:46:50.717407: step: 1200/526, loss: 0.0013476371532306075 2023-01-21 11:46:51.855378: step: 1204/526, loss: 0.028818178921937943 2023-01-21 11:46:52.967772: step: 1208/526, loss: 0.07680292427539825 2023-01-21 11:46:54.069493: step: 1212/526, loss: 0.005534553434699774 2023-01-21 11:46:55.193644: step: 1216/526, loss: 0.15463100373744965 2023-01-21 11:46:56.325396: step: 1220/526, loss: 0.019785745069384575 2023-01-21 11:46:57.438461: step: 1224/526, loss: 0.0016452790005132556 2023-01-21 11:46:58.641156: step: 1228/526, loss: 0.19144554436206818 2023-01-21 11:46:59.790562: step: 1232/526, loss: 0.0007670402410440147 2023-01-21 11:47:00.919019: step: 1236/526, loss: 0.024678993970155716 2023-01-21 11:47:02.060085: step: 1240/526, loss: 0.007084846496582031 2023-01-21 11:47:03.214325: step: 1244/526, loss: 0.02675933949649334 2023-01-21 11:47:04.385326: step: 1248/526, loss: 0.0006745815626345575 2023-01-21 11:47:05.528669: step: 1252/526, loss: 0.027397824451327324 2023-01-21 11:47:06.640724: step: 1256/526, loss: 0.001489925431087613 2023-01-21 11:47:07.758141: step: 1260/526, loss: 0.003269004635512829 2023-01-21 11:47:08.901262: step: 1264/526, loss: 0.01984553411602974 2023-01-21 11:47:10.037988: step: 1268/526, loss: 0.1360720694065094 2023-01-21 11:47:11.210234: step: 1272/526, loss: 0.2584635615348816 2023-01-21 11:47:12.343296: step: 1276/526, loss: 0.005172157660126686 2023-01-21 11:47:13.483947: step: 1280/526, loss: 0.04760890081524849 2023-01-21 11:47:14.625345: step: 1284/526, loss: 0.03189058229327202 2023-01-21 11:47:15.739354: step: 1288/526, loss: 0.0071957590989768505 2023-01-21 11:47:16.861801: step: 1292/526, loss: 0.0004414558643475175 2023-01-21 11:47:18.016655: step: 1296/526, loss: 0.005295705981552601 2023-01-21 11:47:19.163344: step: 1300/526, loss: 0.0015960694290697575 2023-01-21 11:47:20.295559: step: 1304/526, loss: 0.020749665796756744 2023-01-21 11:47:21.437317: step: 1308/526, loss: 0.016248416155576706 2023-01-21 11:47:22.613791: step: 1312/526, loss: 0.07596893608570099 2023-01-21 11:47:23.727295: step: 1316/526, loss: 0.00019531250291038305 2023-01-21 11:47:24.896791: step: 1320/526, loss: 0.009435654617846012 2023-01-21 11:47:26.035526: step: 1324/526, loss: 0.00395622244104743 2023-01-21 11:47:27.186221: step: 1328/526, loss: 0.14249287545681 2023-01-21 11:47:28.311249: step: 1332/526, loss: 0.028884029015898705 2023-01-21 11:47:29.444431: step: 1336/526, loss: 0.0010064125526696444 2023-01-21 11:47:30.589041: step: 1340/526, loss: 0.0003430366632528603 2023-01-21 11:47:31.743538: step: 1344/526, loss: 0.024555303156375885 2023-01-21 11:47:32.858850: step: 1348/526, loss: 0.05074882507324219 2023-01-21 11:47:34.008045: step: 1352/526, loss: 0.07233371585607529 2023-01-21 11:47:35.156932: step: 1356/526, loss: 0.022238444536924362 2023-01-21 11:47:36.283803: step: 1360/526, loss: 0.011472320184111595 2023-01-21 11:47:37.414364: step: 1364/526, loss: 0.011249256320297718 2023-01-21 11:47:38.520534: step: 1368/526, loss: 0.010108756832778454 2023-01-21 11:47:39.657443: step: 1372/526, loss: 0.026175880804657936 2023-01-21 11:47:40.776653: step: 1376/526, loss: 0.07446374744176865 2023-01-21 11:47:41.924411: step: 1380/526, loss: 0.007753944024443626 2023-01-21 11:47:43.085980: step: 1384/526, loss: 0.017905618995428085 2023-01-21 11:47:44.220144: step: 1388/526, loss: 0.07091198116540909 2023-01-21 11:47:45.349106: step: 1392/526, loss: 0.00263385777361691 2023-01-21 11:47:46.519925: step: 1396/526, loss: 0.0222946647554636 2023-01-21 11:47:47.664178: step: 1400/526, loss: 0.0017691315151751041 2023-01-21 11:47:48.802752: step: 1404/526, loss: 0.030976485460996628 2023-01-21 11:47:49.941204: step: 1408/526, loss: 0.002493190811946988 2023-01-21 11:47:51.112916: step: 1412/526, loss: 0.0020665167830884457 2023-01-21 11:47:52.240884: step: 1416/526, loss: 0.007055854890495539 2023-01-21 11:47:53.406781: step: 1420/526, loss: 0.00730972271412611 2023-01-21 11:47:54.509757: step: 1424/526, loss: 0.0026132583152502775 2023-01-21 11:47:55.635660: step: 1428/526, loss: 0.01791706122457981 2023-01-21 11:47:56.791309: step: 1432/526, loss: 0.0013868332607671618 2023-01-21 11:47:57.963569: step: 1436/526, loss: 0.021071529015898705 2023-01-21 11:47:59.111295: step: 1440/526, loss: 0.057860471308231354 2023-01-21 11:48:00.244562: step: 1444/526, loss: 0.012936783023178577 2023-01-21 11:48:01.392472: step: 1448/526, loss: 0.0010332107776775956 2023-01-21 11:48:02.518594: step: 1452/526, loss: 0.005069923587143421 2023-01-21 11:48:03.638286: step: 1456/526, loss: 0.0034369470085948706 2023-01-21 11:48:04.778145: step: 1460/526, loss: 3.409385681152344e-05 2023-01-21 11:48:05.919317: step: 1464/526, loss: 0.00012722014798782766 2023-01-21 11:48:07.078374: step: 1468/526, loss: 0.0047471048310399055 2023-01-21 11:48:08.208045: step: 1472/526, loss: 0.006742286495864391 2023-01-21 11:48:09.337394: step: 1476/526, loss: 0.00024127960205078125 2023-01-21 11:48:10.447363: step: 1480/526, loss: 0.0060630799271166325 2023-01-21 11:48:11.638324: step: 1484/526, loss: 0.0009752750629559159 2023-01-21 11:48:12.777683: step: 1488/526, loss: 0.0025264739524573088 2023-01-21 11:48:13.924519: step: 1492/526, loss: 0.019185258075594902 2023-01-21 11:48:15.062784: step: 1496/526, loss: 0.03000679239630699 2023-01-21 11:48:16.207218: step: 1500/526, loss: 0.0030374526977539062 2023-01-21 11:48:17.335024: step: 1504/526, loss: 0.00015316010103560984 2023-01-21 11:48:18.476346: step: 1508/526, loss: 0.03774048015475273 2023-01-21 11:48:19.623923: step: 1512/526, loss: 0.0017588615883141756 2023-01-21 11:48:20.737055: step: 1516/526, loss: 0.0760623961687088 2023-01-21 11:48:21.872690: step: 1520/526, loss: 0.026489639654755592 2023-01-21 11:48:23.011016: step: 1524/526, loss: 0.03952217102050781 2023-01-21 11:48:24.157262: step: 1528/526, loss: 0.03390655294060707 2023-01-21 11:48:25.297171: step: 1532/526, loss: 0.004974079318344593 2023-01-21 11:48:26.446639: step: 1536/526, loss: 0.001953601837158203 2023-01-21 11:48:27.584675: step: 1540/526, loss: 0.0025721550919115543 2023-01-21 11:48:28.721331: step: 1544/526, loss: 0.019340038299560547 2023-01-21 11:48:29.847653: step: 1548/526, loss: 0.0048046112060546875 2023-01-21 11:48:31.032962: step: 1552/526, loss: 0.04182777553796768 2023-01-21 11:48:32.173067: step: 1556/526, loss: 0.0015128612285479903 2023-01-21 11:48:33.307989: step: 1560/526, loss: 0.0004330635129008442 2023-01-21 11:48:34.463086: step: 1564/526, loss: 0.005412387661635876 2023-01-21 11:48:35.619199: step: 1568/526, loss: 0.07815251499414444 2023-01-21 11:48:36.757715: step: 1572/526, loss: 0.08765692263841629 2023-01-21 11:48:37.896907: step: 1576/526, loss: 0.1259164810180664 2023-01-21 11:48:39.011989: step: 1580/526, loss: 0.00020141601271461695 2023-01-21 11:48:40.150825: step: 1584/526, loss: 0.0005550385103560984 2023-01-21 11:48:41.309032: step: 1588/526, loss: 0.1288854479789734 2023-01-21 11:48:42.443415: step: 1592/526, loss: 0.009970474988222122 2023-01-21 11:48:43.595129: step: 1596/526, loss: 0.003493356751278043 2023-01-21 11:48:44.727000: step: 1600/526, loss: 0.03868560865521431 2023-01-21 11:48:45.839885: step: 1604/526, loss: 0.004835701081901789 2023-01-21 11:48:46.977918: step: 1608/526, loss: 0.012922096997499466 2023-01-21 11:48:48.092083: step: 1612/526, loss: 0.03222351148724556 2023-01-21 11:48:49.227304: step: 1616/526, loss: 0.02324523963034153 2023-01-21 11:48:50.371457: step: 1620/526, loss: 0.0004547596035990864 2023-01-21 11:48:51.523288: step: 1624/526, loss: 0.01389312744140625 2023-01-21 11:48:52.677064: step: 1628/526, loss: 0.016480065882205963 2023-01-21 11:48:53.818918: step: 1632/526, loss: 0.0018655776511877775 2023-01-21 11:48:54.955002: step: 1636/526, loss: 0.00044288637582212687 2023-01-21 11:48:56.085756: step: 1640/526, loss: 0.01916208304464817 2023-01-21 11:48:57.267933: step: 1644/526, loss: 0.0024417878594249487 2023-01-21 11:48:58.401199: step: 1648/526, loss: 0.003782844403758645 2023-01-21 11:48:59.519900: step: 1652/526, loss: 0.09213085472583771 2023-01-21 11:49:00.633835: step: 1656/526, loss: 0.0001305341866100207 2023-01-21 11:49:01.781037: step: 1660/526, loss: 0.023404503241181374 2023-01-21 11:49:02.910733: step: 1664/526, loss: 0.01796851120889187 2023-01-21 11:49:04.042268: step: 1668/526, loss: 0.07622671127319336 2023-01-21 11:49:05.271539: step: 1672/526, loss: 0.00355281843803823 2023-01-21 11:49:06.396211: step: 1676/526, loss: 0.011435317806899548 2023-01-21 11:49:07.513554: step: 1680/526, loss: 0.062442779541015625 2023-01-21 11:49:08.641690: step: 1684/526, loss: 0.003697299864143133 2023-01-21 11:49:09.744324: step: 1688/526, loss: 0.0004693031369242817 2023-01-21 11:49:10.883945: step: 1692/526, loss: 0.003188514616340399 2023-01-21 11:49:12.017674: step: 1696/526, loss: 0.0014663697220385075 2023-01-21 11:49:13.165007: step: 1700/526, loss: 0.008653735741972923 2023-01-21 11:49:14.347824: step: 1704/526, loss: 0.05044383928179741 2023-01-21 11:49:15.434940: step: 1708/526, loss: 0.0019284725422039628 2023-01-21 11:49:16.584808: step: 1712/526, loss: 0.0005409240839071572 2023-01-21 11:49:17.719202: step: 1716/526, loss: 0.00618248013779521 2023-01-21 11:49:18.859968: step: 1720/526, loss: 0.004835701081901789 2023-01-21 11:49:20.001263: step: 1724/526, loss: 0.0015562057960778475 2023-01-21 11:49:21.140021: step: 1728/526, loss: 0.03382463380694389 2023-01-21 11:49:22.271023: step: 1732/526, loss: 0.006190681364387274 2023-01-21 11:49:23.371842: step: 1736/526, loss: 0.012252426706254482 2023-01-21 11:49:24.536764: step: 1740/526, loss: 0.02145414426922798 2023-01-21 11:49:25.680822: step: 1744/526, loss: 0.010240936651825905 2023-01-21 11:49:26.835764: step: 1748/526, loss: 0.013096618466079235 2023-01-21 11:49:27.962428: step: 1752/526, loss: 0.07340545952320099 2023-01-21 11:49:29.116425: step: 1756/526, loss: 0.045792482793331146 2023-01-21 11:49:30.270655: step: 1760/526, loss: 0.041458990424871445 2023-01-21 11:49:31.433891: step: 1764/526, loss: 0.012446975335478783 2023-01-21 11:49:32.607328: step: 1768/526, loss: 0.015450382605195045 2023-01-21 11:49:33.774139: step: 1772/526, loss: 0.014376641251146793 2023-01-21 11:49:34.933428: step: 1776/526, loss: 0.00141487130895257 2023-01-21 11:49:36.087852: step: 1780/526, loss: 0.0013548850547522306 2023-01-21 11:49:37.240740: step: 1784/526, loss: 0.015245819464325905 2023-01-21 11:49:38.383347: step: 1788/526, loss: 0.002990627195686102 2023-01-21 11:49:39.535223: step: 1792/526, loss: 0.0029386519454419613 2023-01-21 11:49:40.671097: step: 1796/526, loss: 0.015460300259292126 2023-01-21 11:49:41.793293: step: 1800/526, loss: 0.0013916492462158203 2023-01-21 11:49:42.920758: step: 1804/526, loss: 0.0035120011307299137 2023-01-21 11:49:44.044476: step: 1808/526, loss: 0.024268342182040215 2023-01-21 11:49:45.169592: step: 1812/526, loss: 0.006486320868134499 2023-01-21 11:49:46.296167: step: 1816/526, loss: 0.008128166198730469 2023-01-21 11:49:47.397083: step: 1820/526, loss: 0.0009676933404989541 2023-01-21 11:49:48.532154: step: 1824/526, loss: 0.15685276687145233 2023-01-21 11:49:49.658853: step: 1828/526, loss: 0.00397916417568922 2023-01-21 11:49:50.811318: step: 1832/526, loss: 0.985771119594574 2023-01-21 11:49:51.940028: step: 1836/526, loss: 0.02320842817425728 2023-01-21 11:49:53.098266: step: 1840/526, loss: 0.30618131160736084 2023-01-21 11:49:54.232238: step: 1844/526, loss: 7.877349707996473e-05 2023-01-21 11:49:55.392156: step: 1848/526, loss: 0.012806988321244717 2023-01-21 11:49:56.520026: step: 1852/526, loss: 0.002792453859001398 2023-01-21 11:49:57.738099: step: 1856/526, loss: 0.01624937169253826 2023-01-21 11:49:58.884940: step: 1860/526, loss: 0.029878806322813034 2023-01-21 11:50:00.037670: step: 1864/526, loss: 0.01139225997030735 2023-01-21 11:50:01.154510: step: 1868/526, loss: 0.009100723080337048 2023-01-21 11:50:02.285014: step: 1872/526, loss: 0.004316282458603382 2023-01-21 11:50:03.439765: step: 1876/526, loss: 0.0011700630420818925 2023-01-21 11:50:04.571472: step: 1880/526, loss: 0.0056847091764211655 2023-01-21 11:50:05.704241: step: 1884/526, loss: 0.01797514036297798 2023-01-21 11:50:06.849050: step: 1888/526, loss: 0.013667869381606579 2023-01-21 11:50:07.974262: step: 1892/526, loss: 0.005320072639733553 2023-01-21 11:50:09.119396: step: 1896/526, loss: 0.0018579482566565275 2023-01-21 11:50:10.262479: step: 1900/526, loss: 0.0007995605119504035 2023-01-21 11:50:11.408558: step: 1904/526, loss: 0.01890571229159832 2023-01-21 11:50:12.539555: step: 1908/526, loss: 0.00027914048405364156 2023-01-21 11:50:13.725687: step: 1912/526, loss: 0.0026869773864746094 2023-01-21 11:50:14.864199: step: 1916/526, loss: 0.0664713904261589 2023-01-21 11:50:16.017110: step: 1920/526, loss: 0.0016141415107995272 2023-01-21 11:50:17.194207: step: 1924/526, loss: 0.05968360975384712 2023-01-21 11:50:18.345962: step: 1928/526, loss: 0.03923375532031059 2023-01-21 11:50:19.501293: step: 1932/526, loss: 0.0008770108688622713 2023-01-21 11:50:20.627339: step: 1936/526, loss: 0.0027625083457678556 2023-01-21 11:50:21.786783: step: 1940/526, loss: 0.0037410736549645662 2023-01-21 11:50:22.932911: step: 1944/526, loss: 0.010704612359404564 2023-01-21 11:50:24.055431: step: 1948/526, loss: 0.04006776958703995 2023-01-21 11:50:25.197657: step: 1952/526, loss: 0.0019824982155114412 2023-01-21 11:50:26.345151: step: 1956/526, loss: 0.0740472823381424 2023-01-21 11:50:27.460715: step: 1960/526, loss: 0.0035758339799940586 2023-01-21 11:50:28.583033: step: 1964/526, loss: 0.07524452358484268 2023-01-21 11:50:29.700264: step: 1968/526, loss: 0.00036373137845657766 2023-01-21 11:50:30.819053: step: 1972/526, loss: 0.0012372017372399569 2023-01-21 11:50:31.994129: step: 1976/526, loss: 0.06256141513586044 2023-01-21 11:50:33.131589: step: 1980/526, loss: 0.030743980780243874 2023-01-21 11:50:34.276974: step: 1984/526, loss: 0.00283985142596066 2023-01-21 11:50:35.407358: step: 1988/526, loss: 0.008736992254853249 2023-01-21 11:50:36.520459: step: 1992/526, loss: 0.0036020278930664062 2023-01-21 11:50:37.683516: step: 1996/526, loss: 0.035840604454278946 2023-01-21 11:50:38.861129: step: 2000/526, loss: 0.0004417419258970767 2023-01-21 11:50:40.000204: step: 2004/526, loss: 0.047985076904296875 2023-01-21 11:50:41.141268: step: 2008/526, loss: 0.0020084381103515625 2023-01-21 11:50:42.263787: step: 2012/526, loss: 0.0019193650223314762 2023-01-21 11:50:43.409332: step: 2016/526, loss: 0.027818774804472923 2023-01-21 11:50:44.591187: step: 2020/526, loss: 0.003223610110580921 2023-01-21 11:50:45.720292: step: 2024/526, loss: 0.05363426357507706 2023-01-21 11:50:46.847257: step: 2028/526, loss: 0.0018838882679119706 2023-01-21 11:50:47.981998: step: 2032/526, loss: 0.05442104488611221 2023-01-21 11:50:49.122369: step: 2036/526, loss: 0.005970573518425226 2023-01-21 11:50:50.265296: step: 2040/526, loss: 0.0636780709028244 2023-01-21 11:50:51.405315: step: 2044/526, loss: 0.007358265109360218 2023-01-21 11:50:52.562685: step: 2048/526, loss: 0.003989601042121649 2023-01-21 11:50:53.722658: step: 2052/526, loss: 0.05000562593340874 2023-01-21 11:50:54.875648: step: 2056/526, loss: 0.08075566589832306 2023-01-21 11:50:56.020544: step: 2060/526, loss: 0.006933736614882946 2023-01-21 11:50:57.171975: step: 2064/526, loss: 0.03279101848602295 2023-01-21 11:50:58.280471: step: 2068/526, loss: 0.0010629177559167147 2023-01-21 11:50:59.448029: step: 2072/526, loss: 0.001238059950992465 2023-01-21 11:51:00.602630: step: 2076/526, loss: 0.10594367980957031 2023-01-21 11:51:01.747906: step: 2080/526, loss: 0.002029800321906805 2023-01-21 11:51:02.896742: step: 2084/526, loss: 0.14116555452346802 2023-01-21 11:51:04.005410: step: 2088/526, loss: 0.0003570556582417339 2023-01-21 11:51:05.140224: step: 2092/526, loss: 0.016153432428836823 2023-01-21 11:51:06.262508: step: 2096/526, loss: 0.0487184040248394 2023-01-21 11:51:07.447782: step: 2100/526, loss: 0.2527340054512024 2023-01-21 11:51:08.571002: step: 2104/526, loss: 0.05380944907665253 ================================================== Loss: 0.029 -------------------- Dev: {'event': {'p': 0.5993657505285412, 'r': 0.7549933422103862, 'f1': 0.6682380671773719}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Test: {'event': {'p': 0.6199000832639467, 'r': 0.7771398747390397, 'f1': 0.6896711440481704}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Chinese: {'event': {'p': 0.5802469135802469, 'r': 0.8703703703703703, 'f1': 0.6962962962962963}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Korean: {'event': {'p': 0.6341463414634146, 'r': 0.4126984126984127, 'f1': 0.5}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} Russian: {'event': {'p': 0.5, 'r': 0.5555555555555556, 'f1': 0.5263157894736842}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 17} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 18 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 11:51:51.786485: step: 4/526, loss: 0.006751632783561945 2023-01-21 11:51:52.985996: step: 8/526, loss: 0.08588533848524094 2023-01-21 11:51:54.134529: step: 12/526, loss: 0.018348027020692825 2023-01-21 11:51:55.263171: step: 16/526, loss: 0.0015828133327886462 2023-01-21 11:51:56.392647: step: 20/526, loss: 0.011808491311967373 2023-01-21 11:51:57.486424: step: 24/526, loss: 0.0005998611450195312 2023-01-21 11:51:58.607815: step: 28/526, loss: 0.07798881828784943 2023-01-21 11:51:59.763502: step: 32/526, loss: 0.02732066996395588 2023-01-21 11:52:00.911102: step: 36/526, loss: 0.004439449403434992 2023-01-21 11:52:02.032328: step: 40/526, loss: 0.006948280613869429 2023-01-21 11:52:03.152242: step: 44/526, loss: 0.0681842640042305 2023-01-21 11:52:04.294805: step: 48/526, loss: 0.08184299618005753 2023-01-21 11:52:05.415465: step: 52/526, loss: 0.002940845675766468 2023-01-21 11:52:06.555348: step: 56/526, loss: 0.003729534102603793 2023-01-21 11:52:07.674886: step: 60/526, loss: 0.0034748080652207136 2023-01-21 11:52:08.854187: step: 64/526, loss: 0.001116943429224193 2023-01-21 11:52:09.989682: step: 68/526, loss: 0.044062234461307526 2023-01-21 11:52:11.160144: step: 72/526, loss: 0.0335724838078022 2023-01-21 11:52:12.284838: step: 76/526, loss: 0.03290729597210884 2023-01-21 11:52:13.440062: step: 80/526, loss: 0.0008903503767214715 2023-01-21 11:52:14.600919: step: 84/526, loss: 0.00013465881056617945 2023-01-21 11:52:15.727396: step: 88/526, loss: 0.0010342597961425781 2023-01-21 11:52:16.843142: step: 92/526, loss: 0.000347995781339705 2023-01-21 11:52:17.950146: step: 96/526, loss: 0.0003503561019897461 2023-01-21 11:52:19.117711: step: 100/526, loss: 0.09323463588953018 2023-01-21 11:52:20.275940: step: 104/526, loss: 0.013824081979691982 2023-01-21 11:52:21.415428: step: 108/526, loss: 0.0370057113468647 2023-01-21 11:52:22.533963: step: 112/526, loss: 0.012207889929413795 2023-01-21 11:52:23.660375: step: 116/526, loss: 0.03219299390912056 2023-01-21 11:52:24.821212: step: 120/526, loss: 0.011827278882265091 2023-01-21 11:52:25.951414: step: 124/526, loss: 0.007548856548964977 2023-01-21 11:52:27.093395: step: 128/526, loss: 0.025111723691225052 2023-01-21 11:52:28.221592: step: 132/526, loss: 0.0021045685280114412 2023-01-21 11:52:29.333918: step: 136/526, loss: 0.0032861712388694286 2023-01-21 11:52:30.460337: step: 140/526, loss: 0.047944262623786926 2023-01-21 11:52:31.578657: step: 144/526, loss: 0.012397385202348232 2023-01-21 11:52:32.694063: step: 148/526, loss: 0.008004856295883656 2023-01-21 11:52:33.833645: step: 152/526, loss: 0.051562026143074036 2023-01-21 11:52:34.944167: step: 156/526, loss: 0.02565789222717285 2023-01-21 11:52:36.077260: step: 160/526, loss: 0.020117856562137604 2023-01-21 11:52:37.184087: step: 164/526, loss: 0.04840121418237686 2023-01-21 11:52:38.317017: step: 168/526, loss: 0.0011371612781658769 2023-01-21 11:52:39.452806: step: 172/526, loss: 0.008522129617631435 2023-01-21 11:52:40.577010: step: 176/526, loss: 0.003469085553660989 2023-01-21 11:52:41.739213: step: 180/526, loss: 0.004070281982421875 2023-01-21 11:52:42.915898: step: 184/526, loss: 0.010434150695800781 2023-01-21 11:52:44.032215: step: 188/526, loss: 0.011482620611786842 2023-01-21 11:52:45.163486: step: 192/526, loss: 0.0036216736771166325 2023-01-21 11:52:46.280502: step: 196/526, loss: 0.06496667861938477 2023-01-21 11:52:47.414694: step: 200/526, loss: 0.00014371871657203883 2023-01-21 11:52:48.548800: step: 204/526, loss: 0.0011555672390386462 2023-01-21 11:52:49.682674: step: 208/526, loss: 0.003719329833984375 2023-01-21 11:52:50.811930: step: 212/526, loss: 0.0025718689430505037 2023-01-21 11:52:51.927604: step: 216/526, loss: 0.0031719207763671875 2023-01-21 11:52:53.036751: step: 220/526, loss: 0.004272651858627796 2023-01-21 11:52:54.202286: step: 224/526, loss: 0.0006236553308553994 2023-01-21 11:52:55.322802: step: 228/526, loss: 0.011204147711396217 2023-01-21 11:52:56.467533: step: 232/526, loss: 0.004658222198486328 2023-01-21 11:52:57.637045: step: 236/526, loss: 0.08171777427196503 2023-01-21 11:52:58.785003: step: 240/526, loss: 0.0040607452392578125 2023-01-21 11:52:59.902036: step: 244/526, loss: 0.018023205921053886 2023-01-21 11:53:01.068539: step: 248/526, loss: 0.0017507553566247225 2023-01-21 11:53:02.191821: step: 252/526, loss: 0.004785728175193071 2023-01-21 11:53:03.320878: step: 256/526, loss: 0.0007618427043780684 2023-01-21 11:53:04.445540: step: 260/526, loss: 0.00016336441331077367 2023-01-21 11:53:05.555630: step: 264/526, loss: 0.018542669713497162 2023-01-21 11:53:06.674994: step: 268/526, loss: 0.0008250236860476434 2023-01-21 11:53:07.831842: step: 272/526, loss: 0.10582657158374786 2023-01-21 11:53:08.982898: step: 276/526, loss: 0.022313691675662994 2023-01-21 11:53:10.115369: step: 280/526, loss: 0.02405899204313755 2023-01-21 11:53:11.245352: step: 284/526, loss: 0.0092652328312397 2023-01-21 11:53:12.368938: step: 288/526, loss: 0.025114823132753372 2023-01-21 11:53:13.511215: step: 292/526, loss: 0.0009074211120605469 2023-01-21 11:53:14.628538: step: 296/526, loss: 0.03528418764472008 2023-01-21 11:53:15.760139: step: 300/526, loss: 0.0016310691135004163 2023-01-21 11:53:16.872829: step: 304/526, loss: 0.0011116027599200606 2023-01-21 11:53:18.024570: step: 308/526, loss: 0.0032452105078846216 2023-01-21 11:53:19.159249: step: 312/526, loss: 0.007528114598244429 2023-01-21 11:53:20.305175: step: 316/526, loss: 0.003603696823120117 2023-01-21 11:53:21.424243: step: 320/526, loss: 0.04394540935754776 2023-01-21 11:53:22.575210: step: 324/526, loss: 0.01797657087445259 2023-01-21 11:53:23.718625: step: 328/526, loss: 0.01444101333618164 2023-01-21 11:53:24.825378: step: 332/526, loss: 0.01026687677949667 2023-01-21 11:53:25.958905: step: 336/526, loss: 0.03797459602355957 2023-01-21 11:53:27.141105: step: 340/526, loss: 0.00041179655818268657 2023-01-21 11:53:28.289060: step: 344/526, loss: 0.0007867813110351562 2023-01-21 11:53:29.416914: step: 348/526, loss: 0.00028057099552825093 2023-01-21 11:53:30.551024: step: 352/526, loss: 0.004292118828743696 2023-01-21 11:53:31.692103: step: 356/526, loss: 0.0612298958003521 2023-01-21 11:53:32.824129: step: 360/526, loss: 0.04290733486413956 2023-01-21 11:53:33.960683: step: 364/526, loss: 0.0001811981201171875 2023-01-21 11:53:35.063391: step: 368/526, loss: 4.758834620588459e-05 2023-01-21 11:53:36.196951: step: 372/526, loss: 0.0016165734268724918 2023-01-21 11:53:37.321991: step: 376/526, loss: 0.012668227776885033 2023-01-21 11:53:38.477555: step: 380/526, loss: 0.006336403079330921 2023-01-21 11:53:39.598422: step: 384/526, loss: 0.0005701064947061241 2023-01-21 11:53:40.717854: step: 388/526, loss: 0.06245927885174751 2023-01-21 11:53:41.848966: step: 392/526, loss: 0.037491992115974426 2023-01-21 11:53:42.964778: step: 396/526, loss: 0.00023846626572776586 2023-01-21 11:53:44.091084: step: 400/526, loss: 0.01639585569500923 2023-01-21 11:53:45.216678: step: 404/526, loss: 0.015825271606445312 2023-01-21 11:53:46.340987: step: 408/526, loss: 0.0014722824562340975 2023-01-21 11:53:47.442064: step: 412/526, loss: 0.4467828869819641 2023-01-21 11:53:48.578116: step: 416/526, loss: 0.005568218417465687 2023-01-21 11:53:49.733082: step: 420/526, loss: 0.03647371381521225 2023-01-21 11:53:50.858720: step: 424/526, loss: 0.011580754071474075 2023-01-21 11:53:51.979072: step: 428/526, loss: 0.00029234885005280375 2023-01-21 11:53:53.108709: step: 432/526, loss: 0.07661227881908417 2023-01-21 11:53:54.273995: step: 436/526, loss: 0.00045804979163222015 2023-01-21 11:53:55.424483: step: 440/526, loss: 0.008469962514936924 2023-01-21 11:53:56.564853: step: 444/526, loss: 0.248646542429924 2023-01-21 11:53:57.676442: step: 448/526, loss: 0.021618841215968132 2023-01-21 11:53:58.814807: step: 452/526, loss: 0.003064966294914484 2023-01-21 11:53:59.943862: step: 456/526, loss: 0.01999364048242569 2023-01-21 11:54:01.074777: step: 460/526, loss: 0.10535955429077148 2023-01-21 11:54:02.196935: step: 464/526, loss: 0.03682585060596466 2023-01-21 11:54:03.308827: step: 468/526, loss: -4.76837158203125e-06 2023-01-21 11:54:04.483656: step: 472/526, loss: 0.0159517303109169 2023-01-21 11:54:05.599426: step: 476/526, loss: 0.01844158209860325 2023-01-21 11:54:06.765452: step: 480/526, loss: 0.0036512850783765316 2023-01-21 11:54:07.876754: step: 484/526, loss: 0.0010453701252117753 2023-01-21 11:54:09.010798: step: 488/526, loss: 0.00029220583382993937 2023-01-21 11:54:10.139988: step: 492/526, loss: 0.0017512321937829256 2023-01-21 11:54:11.241104: step: 496/526, loss: 0.004153633024543524 2023-01-21 11:54:12.404934: step: 500/526, loss: 0.0034191133454442024 2023-01-21 11:54:13.544253: step: 504/526, loss: 0.007610511966049671 2023-01-21 11:54:14.653002: step: 508/526, loss: 0.0011955260997638106 2023-01-21 11:54:15.768536: step: 512/526, loss: 0.00015220644127111882 2023-01-21 11:54:16.894956: step: 516/526, loss: 0.010968017391860485 2023-01-21 11:54:18.037703: step: 520/526, loss: 0.004528808407485485 2023-01-21 11:54:19.164352: step: 524/526, loss: 0.02071418985724449 2023-01-21 11:54:20.260122: step: 528/526, loss: 0.0004985809791833162 2023-01-21 11:54:21.429443: step: 532/526, loss: 0.03955373913049698 2023-01-21 11:54:22.549487: step: 536/526, loss: 0.009877586737275124 2023-01-21 11:54:23.683678: step: 540/526, loss: 0.0025396349374204874 2023-01-21 11:54:24.799364: step: 544/526, loss: 0.001487827394157648 2023-01-21 11:54:25.955803: step: 548/526, loss: 0.0034872053656727076 2023-01-21 11:54:27.091276: step: 552/526, loss: 0.0031486512161791325 2023-01-21 11:54:28.217243: step: 556/526, loss: 0.00038394928560592234 2023-01-21 11:54:29.375933: step: 560/526, loss: 0.0001447677641408518 2023-01-21 11:54:30.531962: step: 564/526, loss: 4.043579247081652e-05 2023-01-21 11:54:31.694872: step: 568/526, loss: 0.003654670901596546 2023-01-21 11:54:32.813675: step: 572/526, loss: 0.0066085816361010075 2023-01-21 11:54:33.937507: step: 576/526, loss: 0.0010699272388592362 2023-01-21 11:54:35.065401: step: 580/526, loss: 0.006632233038544655 2023-01-21 11:54:36.208859: step: 584/526, loss: 0.009936332702636719 2023-01-21 11:54:37.350582: step: 588/526, loss: 0.0074797626584768295 2023-01-21 11:54:38.475598: step: 592/526, loss: 0.0006622314685955644 2023-01-21 11:54:39.606202: step: 596/526, loss: 0.025319386273622513 2023-01-21 11:54:40.717137: step: 600/526, loss: 0.007889366708695889 2023-01-21 11:54:41.859019: step: 604/526, loss: 0.011844825930893421 2023-01-21 11:54:42.988579: step: 608/526, loss: 7.534027099609375e-05 2023-01-21 11:54:44.108301: step: 612/526, loss: 0.0005170822259970009 2023-01-21 11:54:45.240571: step: 616/526, loss: 0.0006565094226971269 2023-01-21 11:54:46.384046: step: 620/526, loss: 0.0029603957664221525 2023-01-21 11:54:47.511866: step: 624/526, loss: 0.009745597839355469 2023-01-21 11:54:48.626921: step: 628/526, loss: 0.00015172958956100047 2023-01-21 11:54:49.768953: step: 632/526, loss: 0.08412046730518341 2023-01-21 11:54:50.915164: step: 636/526, loss: 0.02991456910967827 2023-01-21 11:54:52.043467: step: 640/526, loss: 0.029345132410526276 2023-01-21 11:54:53.189254: step: 644/526, loss: 0.047356415539979935 2023-01-21 11:54:54.320736: step: 648/526, loss: 0.00168523786123842 2023-01-21 11:54:55.469578: step: 652/526, loss: 0.06777477264404297 2023-01-21 11:54:56.598123: step: 656/526, loss: 0.001197004341520369 2023-01-21 11:54:57.706652: step: 660/526, loss: 0.012173175811767578 2023-01-21 11:54:58.848490: step: 664/526, loss: 0.02936248853802681 2023-01-21 11:54:59.991380: step: 668/526, loss: 0.13716450333595276 2023-01-21 11:55:01.142674: step: 672/526, loss: 0.0002035141078522429 2023-01-21 11:55:02.256746: step: 676/526, loss: 0.030060291290283203 2023-01-21 11:55:03.374851: step: 680/526, loss: 0.0007513046148233116 2023-01-21 11:55:04.480478: step: 684/526, loss: 0.019878387451171875 2023-01-21 11:55:05.645448: step: 688/526, loss: 0.00892949104309082 2023-01-21 11:55:06.760481: step: 692/526, loss: 0.006954574957489967 2023-01-21 11:55:07.886502: step: 696/526, loss: 0.00027439594850875437 2023-01-21 11:55:09.000960: step: 700/526, loss: 0.0015366554725915194 2023-01-21 11:55:10.128325: step: 704/526, loss: 0.017702102661132812 2023-01-21 11:55:11.271198: step: 708/526, loss: 0.11855965107679367 2023-01-21 11:55:12.392870: step: 712/526, loss: 0.0006343841669149697 2023-01-21 11:55:13.488865: step: 716/526, loss: 0.0032587528694421053 2023-01-21 11:55:14.608838: step: 720/526, loss: 0.00033979417639784515 2023-01-21 11:55:15.739596: step: 724/526, loss: 0.0005846023559570312 2023-01-21 11:55:16.855868: step: 728/526, loss: 0.004286289215087891 2023-01-21 11:55:17.980154: step: 732/526, loss: 0.025553608313202858 2023-01-21 11:55:19.118342: step: 736/526, loss: 0.0019258499378338456 2023-01-21 11:55:20.248312: step: 740/526, loss: 0.0750301331281662 2023-01-21 11:55:21.380116: step: 744/526, loss: 0.0254149679094553 2023-01-21 11:55:22.513244: step: 748/526, loss: 0.0009754180791787803 2023-01-21 11:55:23.625366: step: 752/526, loss: 0.2893388867378235 2023-01-21 11:55:24.791091: step: 756/526, loss: 0.014654017053544521 2023-01-21 11:55:25.883118: step: 760/526, loss: 0.003608894534409046 2023-01-21 11:55:27.003330: step: 764/526, loss: 0.016156386584043503 2023-01-21 11:55:28.109476: step: 768/526, loss: 0.0036128521896898746 2023-01-21 11:55:29.242486: step: 772/526, loss: 0.1001136377453804 2023-01-21 11:55:30.375535: step: 776/526, loss: 0.010996627621352673 2023-01-21 11:55:31.498989: step: 780/526, loss: 0.3185775876045227 2023-01-21 11:55:32.683828: step: 784/526, loss: 0.00040454865666106343 2023-01-21 11:55:33.827841: step: 788/526, loss: 0.033834267407655716 2023-01-21 11:55:34.952319: step: 792/526, loss: 0.04966411739587784 2023-01-21 11:55:36.083086: step: 796/526, loss: 0.011606693267822266 2023-01-21 11:55:37.207311: step: 800/526, loss: 0.0005444526905193925 2023-01-21 11:55:38.322073: step: 804/526, loss: 0.0015051126247271895 2023-01-21 11:55:39.445419: step: 808/526, loss: 0.04326934739947319 2023-01-21 11:55:40.591687: step: 812/526, loss: 0.0002534866507630795 2023-01-21 11:55:41.729146: step: 816/526, loss: 0.4787864685058594 2023-01-21 11:55:42.861080: step: 820/526, loss: 0.0009424686431884766 2023-01-21 11:55:43.974175: step: 824/526, loss: 0.010087060742080212 2023-01-21 11:55:45.099796: step: 828/526, loss: 0.007407856173813343 2023-01-21 11:55:46.230688: step: 832/526, loss: 0.0007472991710528731 2023-01-21 11:55:47.370467: step: 836/526, loss: 0.00671997107565403 2023-01-21 11:55:48.531876: step: 840/526, loss: 0.004812860395759344 2023-01-21 11:55:49.661738: step: 844/526, loss: 0.030461406335234642 2023-01-21 11:55:50.813660: step: 848/526, loss: 0.012452412396669388 2023-01-21 11:55:51.914216: step: 852/526, loss: 0.008803559467196465 2023-01-21 11:55:53.054577: step: 856/526, loss: 0.001974487444385886 2023-01-21 11:55:54.180676: step: 860/526, loss: 0.020607374608516693 2023-01-21 11:55:55.301791: step: 864/526, loss: 0.018451310694217682 2023-01-21 11:55:56.422404: step: 868/526, loss: 0.0005424499395303428 2023-01-21 11:55:57.564981: step: 872/526, loss: 0.00715980539098382 2023-01-21 11:55:58.708238: step: 876/526, loss: 0.00910329818725586 2023-01-21 11:55:59.840635: step: 880/526, loss: 0.013965129852294922 2023-01-21 11:56:00.964629: step: 884/526, loss: 0.0044203754514455795 2023-01-21 11:56:02.097453: step: 888/526, loss: 0.045957569032907486 2023-01-21 11:56:03.252441: step: 892/526, loss: 0.0010749816428869963 2023-01-21 11:56:04.410330: step: 896/526, loss: 0.011577701196074486 2023-01-21 11:56:05.516122: step: 900/526, loss: 0.00041518211946822703 2023-01-21 11:56:06.672029: step: 904/526, loss: 0.042954470962285995 2023-01-21 11:56:07.804609: step: 908/526, loss: 0.0006189346313476562 2023-01-21 11:56:08.958717: step: 912/526, loss: 0.026955222710967064 2023-01-21 11:56:10.112330: step: 916/526, loss: 0.00064001081045717 2023-01-21 11:56:11.227614: step: 920/526, loss: 0.001116847968660295 2023-01-21 11:56:12.370935: step: 924/526, loss: 7.21931501175277e-05 2023-01-21 11:56:13.480270: step: 928/526, loss: 0.23166485130786896 2023-01-21 11:56:14.574519: step: 932/526, loss: 0.0034762383438646793 2023-01-21 11:56:15.722063: step: 936/526, loss: 0.0320739783346653 2023-01-21 11:56:16.862343: step: 940/526, loss: 0.03420533984899521 2023-01-21 11:56:17.998046: step: 944/526, loss: 0.003828811924904585 2023-01-21 11:56:19.106450: step: 948/526, loss: 0.031687453389167786 2023-01-21 11:56:20.252533: step: 952/526, loss: 0.007382297422736883 2023-01-21 11:56:21.402861: step: 956/526, loss: 0.1173815205693245 2023-01-21 11:56:22.527053: step: 960/526, loss: 0.01780223846435547 2023-01-21 11:56:23.668797: step: 964/526, loss: 0.06351308524608612 2023-01-21 11:56:24.814563: step: 968/526, loss: 0.01303086243569851 2023-01-21 11:56:25.966133: step: 972/526, loss: 0.015380574390292168 2023-01-21 11:56:27.081563: step: 976/526, loss: 9.732246689964086e-05 2023-01-21 11:56:28.242896: step: 980/526, loss: 0.0003514289855957031 2023-01-21 11:56:29.380208: step: 984/526, loss: 0.005112933926284313 2023-01-21 11:56:30.507419: step: 988/526, loss: 0.0030620575416833162 2023-01-21 11:56:31.694542: step: 992/526, loss: 0.008502769283950329 2023-01-21 11:56:32.835171: step: 996/526, loss: 0.00021657944307662547 2023-01-21 11:56:33.965170: step: 1000/526, loss: 0.011288738809525967 2023-01-21 11:56:35.118532: step: 1004/526, loss: 0.026064587756991386 2023-01-21 11:56:36.294755: step: 1008/526, loss: 0.10760241001844406 2023-01-21 11:56:37.460291: step: 1012/526, loss: 0.0012613296275958419 2023-01-21 11:56:38.566750: step: 1016/526, loss: 0.0011058807140216231 2023-01-21 11:56:39.697699: step: 1020/526, loss: 0.00409202603623271 2023-01-21 11:56:40.842610: step: 1024/526, loss: 0.01920166052877903 2023-01-21 11:56:41.950899: step: 1028/526, loss: 0.00015954971604514867 2023-01-21 11:56:43.107506: step: 1032/526, loss: 4.5299530029296875e-05 2023-01-21 11:56:44.251719: step: 1036/526, loss: 0.0046485899947583675 2023-01-21 11:56:45.407249: step: 1040/526, loss: 8.96453821042087e-06 2023-01-21 11:56:46.560980: step: 1044/526, loss: 0.00014934540376998484 2023-01-21 11:56:47.749831: step: 1048/526, loss: 0.0034685134887695312 2023-01-21 11:56:48.892961: step: 1052/526, loss: 0.01864943467080593 2023-01-21 11:56:50.031360: step: 1056/526, loss: 0.019978713244199753 2023-01-21 11:56:51.194990: step: 1060/526, loss: 0.0019093513255938888 2023-01-21 11:56:52.318413: step: 1064/526, loss: 0.01450891513377428 2023-01-21 11:56:53.431571: step: 1068/526, loss: 0.08216586709022522 2023-01-21 11:56:54.582397: step: 1072/526, loss: 1.9073486328125e-05 2023-01-21 11:56:55.741636: step: 1076/526, loss: 0.0030183792114257812 2023-01-21 11:56:56.880469: step: 1080/526, loss: 0.0012455940013751388 2023-01-21 11:56:58.017439: step: 1084/526, loss: 0.0063187601044774055 2023-01-21 11:56:59.160074: step: 1088/526, loss: 0.0017118453979492188 2023-01-21 11:57:00.305824: step: 1092/526, loss: 0.04254655912518501 2023-01-21 11:57:01.425969: step: 1096/526, loss: 0.030933095142245293 2023-01-21 11:57:02.543824: step: 1100/526, loss: 0.0005690098041668534 2023-01-21 11:57:03.708584: step: 1104/526, loss: 0.035735514014959335 2023-01-21 11:57:04.849700: step: 1108/526, loss: 0.031011486425995827 2023-01-21 11:57:05.975342: step: 1112/526, loss: 0.022867776453495026 2023-01-21 11:57:07.142944: step: 1116/526, loss: 0.02181835100054741 2023-01-21 11:57:08.265879: step: 1120/526, loss: 0.002197647001594305 2023-01-21 11:57:09.396570: step: 1124/526, loss: 0.02172575145959854 2023-01-21 11:57:10.509762: step: 1128/526, loss: 0.024561645463109016 2023-01-21 11:57:11.633897: step: 1132/526, loss: 0.0028888701926916838 2023-01-21 11:57:12.738753: step: 1136/526, loss: 0.051204510033130646 2023-01-21 11:57:13.863031: step: 1140/526, loss: 0.003107261611148715 2023-01-21 11:57:15.016107: step: 1144/526, loss: 0.0001703739253571257 2023-01-21 11:57:16.141572: step: 1148/526, loss: 0.002641010330989957 2023-01-21 11:57:17.292857: step: 1152/526, loss: 0.028209306299686432 2023-01-21 11:57:18.402744: step: 1156/526, loss: 0.0008631706587038934 2023-01-21 11:57:19.554945: step: 1160/526, loss: 0.0034566400572657585 2023-01-21 11:57:20.708305: step: 1164/526, loss: 0.004465484991669655 2023-01-21 11:57:21.818420: step: 1168/526, loss: 0.005255413241684437 2023-01-21 11:57:22.978044: step: 1172/526, loss: 0.062155917286872864 2023-01-21 11:57:24.082988: step: 1176/526, loss: 0.02624053880572319 2023-01-21 11:57:25.223343: step: 1180/526, loss: 0.01065750140696764 2023-01-21 11:57:26.362631: step: 1184/526, loss: 0.02648334577679634 2023-01-21 11:57:27.479531: step: 1188/526, loss: 0.0012329102028161287 2023-01-21 11:57:28.596631: step: 1192/526, loss: 0.0015469550853595138 2023-01-21 11:57:29.707550: step: 1196/526, loss: 0.09007377922534943 2023-01-21 11:57:30.839098: step: 1200/526, loss: 0.012814425863325596 2023-01-21 11:57:32.001156: step: 1204/526, loss: 0.014897346496582031 2023-01-21 11:57:33.158796: step: 1208/526, loss: 0.001088905381038785 2023-01-21 11:57:34.296022: step: 1212/526, loss: 0.023442650213837624 2023-01-21 11:57:35.410772: step: 1216/526, loss: 0.04458579793572426 2023-01-21 11:57:36.541420: step: 1220/526, loss: 0.0002468109014444053 2023-01-21 11:57:37.669653: step: 1224/526, loss: 0.011535453610122204 2023-01-21 11:57:38.813549: step: 1228/526, loss: 0.022029779851436615 2023-01-21 11:57:39.950957: step: 1232/526, loss: 0.007533550262451172 2023-01-21 11:57:41.098579: step: 1236/526, loss: 0.015412712469696999 2023-01-21 11:57:42.221371: step: 1240/526, loss: 0.00017709731764625758 2023-01-21 11:57:43.352258: step: 1244/526, loss: 0.00043492316035553813 2023-01-21 11:57:44.472810: step: 1248/526, loss: 0.05779552459716797 2023-01-21 11:57:45.612585: step: 1252/526, loss: 0.02234034612774849 2023-01-21 11:57:46.733533: step: 1256/526, loss: 0.018366623669862747 2023-01-21 11:57:47.873686: step: 1260/526, loss: 0.0005928039900027215 2023-01-21 11:57:49.016562: step: 1264/526, loss: 0.01376113947480917 2023-01-21 11:57:50.171370: step: 1268/526, loss: 0.05060577392578125 2023-01-21 11:57:51.293049: step: 1272/526, loss: 0.0036373138427734375 2023-01-21 11:57:52.429971: step: 1276/526, loss: 0.029652167111635208 2023-01-21 11:57:53.605694: step: 1280/526, loss: 0.04644804075360298 2023-01-21 11:57:54.769058: step: 1284/526, loss: 0.0029702186584472656 2023-01-21 11:57:55.898453: step: 1288/526, loss: 0.003202104475349188 2023-01-21 11:57:57.047850: step: 1292/526, loss: 0.05603065341711044 2023-01-21 11:57:58.174450: step: 1296/526, loss: 0.016869639977812767 2023-01-21 11:57:59.338001: step: 1300/526, loss: 0.0038928985595703125 2023-01-21 11:58:00.473128: step: 1304/526, loss: 0.04448118433356285 2023-01-21 11:58:01.633382: step: 1308/526, loss: 0.0053289649076759815 2023-01-21 11:58:02.765339: step: 1312/526, loss: 0.44483301043510437 2023-01-21 11:58:03.901676: step: 1316/526, loss: 0.0007533073076047003 2023-01-21 11:58:05.034791: step: 1320/526, loss: 0.000522994960192591 2023-01-21 11:58:06.184866: step: 1324/526, loss: 0.25938504934310913 2023-01-21 11:58:07.330347: step: 1328/526, loss: 0.20688065886497498 2023-01-21 11:58:08.488578: step: 1332/526, loss: 0.02909240685403347 2023-01-21 11:58:09.606656: step: 1336/526, loss: 0.01870584487915039 2023-01-21 11:58:10.750111: step: 1340/526, loss: 0.025593310594558716 2023-01-21 11:58:11.853319: step: 1344/526, loss: 0.0005415916675701737 2023-01-21 11:58:12.964956: step: 1348/526, loss: 0.0066468240693211555 2023-01-21 11:58:14.082798: step: 1352/526, loss: 0.1011747419834137 2023-01-21 11:58:15.214323: step: 1356/526, loss: 0.030309487134218216 2023-01-21 11:58:16.337082: step: 1360/526, loss: 0.032256029546260834 2023-01-21 11:58:17.508568: step: 1364/526, loss: 0.021113110706210136 2023-01-21 11:58:18.663904: step: 1368/526, loss: 0.022073745727539062 2023-01-21 11:58:19.795892: step: 1372/526, loss: 0.017027663066983223 2023-01-21 11:58:20.951290: step: 1376/526, loss: 0.00087652204092592 2023-01-21 11:58:22.089844: step: 1380/526, loss: 0.0020315172150731087 2023-01-21 11:58:23.217426: step: 1384/526, loss: 0.005466843023896217 2023-01-21 11:58:24.323374: step: 1388/526, loss: 4.5108794438419864e-05 2023-01-21 11:58:25.452852: step: 1392/526, loss: 0.0009288787841796875 2023-01-21 11:58:26.644740: step: 1396/526, loss: 0.0031188009306788445 2023-01-21 11:58:27.749373: step: 1400/526, loss: 0.008556175045669079 2023-01-21 11:58:28.863487: step: 1404/526, loss: 0.02328958548605442 2023-01-21 11:58:30.016181: step: 1408/526, loss: 0.0011438369983807206 2023-01-21 11:58:31.172622: step: 1412/526, loss: 0.13003291189670563 2023-01-21 11:58:32.318403: step: 1416/526, loss: 0.005061817355453968 2023-01-21 11:58:33.470361: step: 1420/526, loss: 0.08237877488136292 2023-01-21 11:58:34.629404: step: 1424/526, loss: 0.006093216128647327 2023-01-21 11:58:35.773374: step: 1428/526, loss: 0.010690021328628063 2023-01-21 11:58:36.902473: step: 1432/526, loss: 0.005165672395378351 2023-01-21 11:58:38.063891: step: 1436/526, loss: 0.0027582168113440275 2023-01-21 11:58:39.186722: step: 1440/526, loss: 0.014203167520463467 2023-01-21 11:58:40.329030: step: 1444/526, loss: 0.0035703659523278475 2023-01-21 11:58:41.450499: step: 1448/526, loss: 0.0007405281066894531 2023-01-21 11:58:42.616029: step: 1452/526, loss: 0.048860300332307816 2023-01-21 11:58:43.765554: step: 1456/526, loss: 0.05387945473194122 2023-01-21 11:58:44.911626: step: 1460/526, loss: 0.0007026434177532792 2023-01-21 11:58:45.992926: step: 1464/526, loss: 0.0004672825161833316 2023-01-21 11:58:47.153496: step: 1468/526, loss: 0.334145724773407 2023-01-21 11:58:48.276340: step: 1472/526, loss: 0.00843515433371067 2023-01-21 11:58:49.430451: step: 1476/526, loss: 0.00026111601619049907 2023-01-21 11:58:50.582554: step: 1480/526, loss: 0.0380183607339859 2023-01-21 11:58:51.721205: step: 1484/526, loss: 0.00114526750985533 2023-01-21 11:58:52.899957: step: 1488/526, loss: 0.005035209935158491 2023-01-21 11:58:54.017213: step: 1492/526, loss: 0.0095077995210886 2023-01-21 11:58:55.148755: step: 1496/526, loss: 0.00024890899658203125 2023-01-21 11:58:56.271338: step: 1500/526, loss: 0.00029144284781068563 2023-01-21 11:58:57.395451: step: 1504/526, loss: 0.0007922649383544922 2023-01-21 11:58:58.557878: step: 1508/526, loss: 0.008802032098174095 2023-01-21 11:58:59.704049: step: 1512/526, loss: 0.006653976161032915 2023-01-21 11:59:00.856068: step: 1516/526, loss: 0.00032367705716751516 2023-01-21 11:59:02.007034: step: 1520/526, loss: 0.020620156079530716 2023-01-21 11:59:03.156477: step: 1524/526, loss: 0.0002572059747762978 2023-01-21 11:59:04.307682: step: 1528/526, loss: 0.00015363693819381297 2023-01-21 11:59:05.429755: step: 1532/526, loss: 0.013950919732451439 2023-01-21 11:59:06.579258: step: 1536/526, loss: 5.426407005870715e-05 2023-01-21 11:59:07.716239: step: 1540/526, loss: 0.026434611529111862 2023-01-21 11:59:08.852812: step: 1544/526, loss: 0.00018100740271620452 2023-01-21 11:59:09.981197: step: 1548/526, loss: 0.008823871612548828 2023-01-21 11:59:11.111544: step: 1552/526, loss: 3.1471254260395654e-06 2023-01-21 11:59:12.236200: step: 1556/526, loss: 0.07268200814723969 2023-01-21 11:59:13.396558: step: 1560/526, loss: 0.049806978553533554 2023-01-21 11:59:14.546186: step: 1564/526, loss: 0.012389754876494408 2023-01-21 11:59:15.680850: step: 1568/526, loss: 7.781983003951609e-05 2023-01-21 11:59:16.843789: step: 1572/526, loss: 0.014441299252212048 2023-01-21 11:59:18.053833: step: 1576/526, loss: 0.014146995730698109 2023-01-21 11:59:19.220713: step: 1580/526, loss: 0.0002803802490234375 2023-01-21 11:59:20.355462: step: 1584/526, loss: 0.034462932497262955 2023-01-21 11:59:21.501820: step: 1588/526, loss: 1.5753436088562012 2023-01-21 11:59:22.636983: step: 1592/526, loss: 0.001132965087890625 2023-01-21 11:59:23.773714: step: 1596/526, loss: 0.01793518103659153 2023-01-21 11:59:24.906440: step: 1600/526, loss: 0.030055763199925423 2023-01-21 11:59:26.055497: step: 1604/526, loss: 0.014671134762465954 2023-01-21 11:59:27.201380: step: 1608/526, loss: 0.0075120930559933186 2023-01-21 11:59:28.338505: step: 1612/526, loss: 0.009023762308061123 2023-01-21 11:59:29.478245: step: 1616/526, loss: 0.0018198012840002775 2023-01-21 11:59:30.604866: step: 1620/526, loss: 0.0001615524379303679 2023-01-21 11:59:31.723181: step: 1624/526, loss: 0.008576393127441406 2023-01-21 11:59:32.844474: step: 1628/526, loss: 0.0005743026849813759 2023-01-21 11:59:33.974550: step: 1632/526, loss: 0.0025877475272864103 2023-01-21 11:59:35.119470: step: 1636/526, loss: 0.04479098320007324 2023-01-21 11:59:36.258685: step: 1640/526, loss: 0.009057330898940563 2023-01-21 11:59:37.386775: step: 1644/526, loss: 0.00747337331995368 2023-01-21 11:59:38.540316: step: 1648/526, loss: 0.0029829025734215975 2023-01-21 11:59:39.652568: step: 1652/526, loss: 0.008117579855024815 2023-01-21 11:59:40.783411: step: 1656/526, loss: 0.010360908694565296 2023-01-21 11:59:41.938751: step: 1660/526, loss: 0.011524582281708717 2023-01-21 11:59:43.100956: step: 1664/526, loss: 0.03401918336749077 2023-01-21 11:59:44.227259: step: 1668/526, loss: 0.005031013395637274 2023-01-21 11:59:45.359871: step: 1672/526, loss: 0.004806423559784889 2023-01-21 11:59:46.498112: step: 1676/526, loss: 0.019953345879912376 2023-01-21 11:59:47.624246: step: 1680/526, loss: 0.01844792440533638 2023-01-21 11:59:48.761065: step: 1684/526, loss: 0.0008735657320357859 2023-01-21 11:59:49.924172: step: 1688/526, loss: 0.0035420418716967106 2023-01-21 11:59:51.076550: step: 1692/526, loss: 0.007732200901955366 2023-01-21 11:59:52.210906: step: 1696/526, loss: 0.0007713317754678428 2023-01-21 11:59:53.361134: step: 1700/526, loss: 0.03784923627972603 2023-01-21 11:59:54.493995: step: 1704/526, loss: 0.015230274759232998 2023-01-21 11:59:55.635509: step: 1708/526, loss: 0.0011760711204260588 2023-01-21 11:59:56.787918: step: 1712/526, loss: 0.010625649243593216 2023-01-21 11:59:57.919782: step: 1716/526, loss: 0.005139159969985485 2023-01-21 11:59:59.094560: step: 1720/526, loss: 0.03791771084070206 2023-01-21 12:00:00.233948: step: 1724/526, loss: 0.0262099988758564 2023-01-21 12:00:01.365595: step: 1728/526, loss: 0.0007447600364685059 2023-01-21 12:00:02.500067: step: 1732/526, loss: 0.00066547398455441 2023-01-21 12:00:03.628772: step: 1736/526, loss: 0.039075613021850586 2023-01-21 12:00:04.728213: step: 1740/526, loss: 0.006984138861298561 2023-01-21 12:00:05.865054: step: 1744/526, loss: 0.0003493308904580772 2023-01-21 12:00:07.031965: step: 1748/526, loss: 0.004170799162238836 2023-01-21 12:00:08.141483: step: 1752/526, loss: 0.001493167830631137 2023-01-21 12:00:09.329621: step: 1756/526, loss: 0.025445174425840378 2023-01-21 12:00:10.481948: step: 1760/526, loss: 0.012231064029037952 2023-01-21 12:00:11.637343: step: 1764/526, loss: 0.0006394386291503906 2023-01-21 12:00:12.766072: step: 1768/526, loss: 0.011926889419555664 2023-01-21 12:00:13.903747: step: 1772/526, loss: 0.04630432277917862 2023-01-21 12:00:15.029977: step: 1776/526, loss: 0.01020674780011177 2023-01-21 12:00:16.185206: step: 1780/526, loss: 0.03478426858782768 2023-01-21 12:00:17.342694: step: 1784/526, loss: 0.00114526750985533 2023-01-21 12:00:18.497225: step: 1788/526, loss: 0.0029139756225049496 2023-01-21 12:00:19.656691: step: 1792/526, loss: 0.03370952606201172 2023-01-21 12:00:20.777906: step: 1796/526, loss: 0.03203010559082031 2023-01-21 12:00:21.919524: step: 1800/526, loss: 0.0054104807786643505 2023-01-21 12:00:23.051664: step: 1804/526, loss: 0.013263989239931107 2023-01-21 12:00:24.177165: step: 1808/526, loss: 0.013837659731507301 2023-01-21 12:00:25.309138: step: 1812/526, loss: 0.03971576690673828 2023-01-21 12:00:26.435743: step: 1816/526, loss: 0.019826795905828476 2023-01-21 12:00:27.564021: step: 1820/526, loss: 0.00041570665780454874 2023-01-21 12:00:28.692763: step: 1824/526, loss: 0.010503023862838745 2023-01-21 12:00:29.840377: step: 1828/526, loss: 2.57968895311933e-05 2023-01-21 12:00:30.973085: step: 1832/526, loss: 0.014549542218446732 2023-01-21 12:00:32.147063: step: 1836/526, loss: 0.0004096984921488911 2023-01-21 12:00:33.278339: step: 1840/526, loss: 0.008773994632065296 2023-01-21 12:00:34.412133: step: 1844/526, loss: 0.06470699608325958 2023-01-21 12:00:35.536502: step: 1848/526, loss: 0.0035255432594567537 2023-01-21 12:00:36.648660: step: 1852/526, loss: 0.0013734817039221525 2023-01-21 12:00:37.803059: step: 1856/526, loss: 0.003048897022381425 2023-01-21 12:00:38.959425: step: 1860/526, loss: 0.027164697647094727 2023-01-21 12:00:40.157834: step: 1864/526, loss: 0.5487657189369202 2023-01-21 12:00:41.269755: step: 1868/526, loss: 0.01311645470559597 2023-01-21 12:00:42.397679: step: 1872/526, loss: 0.001676368759945035 2023-01-21 12:00:43.511007: step: 1876/526, loss: 0.0007895469898357987 2023-01-21 12:00:44.662330: step: 1880/526, loss: 0.0019960403442382812 2023-01-21 12:00:45.787749: step: 1884/526, loss: 0.08450088649988174 2023-01-21 12:00:46.921295: step: 1888/526, loss: 0.001560211181640625 2023-01-21 12:00:48.079049: step: 1892/526, loss: 4.310607982915826e-05 2023-01-21 12:00:49.240964: step: 1896/526, loss: 0.25707730650901794 2023-01-21 12:00:50.389624: step: 1900/526, loss: 0.009438895620405674 2023-01-21 12:00:51.521390: step: 1904/526, loss: 0.0024895668029785156 2023-01-21 12:00:52.675655: step: 1908/526, loss: 0.003894710447639227 2023-01-21 12:00:53.833451: step: 1912/526, loss: 0.00026617050752975047 2023-01-21 12:00:54.957741: step: 1916/526, loss: 9.689330909168348e-05 2023-01-21 12:00:56.088138: step: 1920/526, loss: 0.00023465156846214086 2023-01-21 12:00:57.201008: step: 1924/526, loss: 0.0017255783313885331 2023-01-21 12:00:58.324263: step: 1928/526, loss: 6.351470801746473e-05 2023-01-21 12:00:59.501767: step: 1932/526, loss: 0.0013942718505859375 2023-01-21 12:01:00.634189: step: 1936/526, loss: 0.008255386725068092 2023-01-21 12:01:01.761324: step: 1940/526, loss: 0.005553054623305798 2023-01-21 12:01:02.902195: step: 1944/526, loss: 0.0006347656017169356 2023-01-21 12:01:04.014669: step: 1948/526, loss: 0.008454657159745693 2023-01-21 12:01:05.129100: step: 1952/526, loss: 0.00027956964913755655 2023-01-21 12:01:06.251904: step: 1956/526, loss: 0.0069716451689600945 2023-01-21 12:01:07.377056: step: 1960/526, loss: 1.0779386758804321 2023-01-21 12:01:08.525756: step: 1964/526, loss: 0.018053723499178886 2023-01-21 12:01:09.659483: step: 1968/526, loss: 0.05226755142211914 2023-01-21 12:01:10.796964: step: 1972/526, loss: 0.024973010644316673 2023-01-21 12:01:11.930601: step: 1976/526, loss: 0.00859222374856472 2023-01-21 12:01:13.090982: step: 1980/526, loss: 0.007870674133300781 2023-01-21 12:01:14.251902: step: 1984/526, loss: 0.0009204865200445056 2023-01-21 12:01:15.380581: step: 1988/526, loss: 0.010213280096650124 2023-01-21 12:01:16.500603: step: 1992/526, loss: 0.0029568197205662727 2023-01-21 12:01:17.646800: step: 1996/526, loss: 0.00034322740975767374 2023-01-21 12:01:18.778411: step: 2000/526, loss: 0.0283355712890625 2023-01-21 12:01:19.904983: step: 2004/526, loss: 0.04610319435596466 2023-01-21 12:01:21.042313: step: 2008/526, loss: 0.035727739334106445 2023-01-21 12:01:22.193530: step: 2012/526, loss: 0.00164031982421875 2023-01-21 12:01:23.340276: step: 2016/526, loss: 0.013270759955048561 2023-01-21 12:01:24.471972: step: 2020/526, loss: 0.0016202926635742188 2023-01-21 12:01:25.603089: step: 2024/526, loss: 0.007674217224121094 2023-01-21 12:01:26.749164: step: 2028/526, loss: 0.0795714408159256 2023-01-21 12:01:27.874522: step: 2032/526, loss: 0.00148773193359375 2023-01-21 12:01:29.011143: step: 2036/526, loss: 0.000966835068538785 2023-01-21 12:01:30.167410: step: 2040/526, loss: 0.04186592251062393 2023-01-21 12:01:31.334544: step: 2044/526, loss: 0.006205367855727673 2023-01-21 12:01:32.473066: step: 2048/526, loss: 0.012000561691820621 2023-01-21 12:01:33.594968: step: 2052/526, loss: 0.03836984559893608 2023-01-21 12:01:34.710847: step: 2056/526, loss: 0.0773535743355751 2023-01-21 12:01:35.845022: step: 2060/526, loss: 0.004939270205795765 2023-01-21 12:01:36.963700: step: 2064/526, loss: 0.000782489834818989 2023-01-21 12:01:38.079650: step: 2068/526, loss: 0.018858909606933594 2023-01-21 12:01:39.207059: step: 2072/526, loss: 0.10593309253454208 2023-01-21 12:01:40.319356: step: 2076/526, loss: 0.03182802349328995 2023-01-21 12:01:41.474663: step: 2080/526, loss: 0.013780689798295498 2023-01-21 12:01:42.594749: step: 2084/526, loss: 0.010799598880112171 2023-01-21 12:01:43.707877: step: 2088/526, loss: 0.0016660690307617188 2023-01-21 12:01:44.827318: step: 2092/526, loss: 0.0021753788460046053 2023-01-21 12:01:45.975708: step: 2096/526, loss: 0.023459387943148613 2023-01-21 12:01:47.116063: step: 2100/526, loss: 0.007106876000761986 2023-01-21 12:01:48.262025: step: 2104/526, loss: 0.005678844638168812 ================================================== Loss: 0.029 -------------------- Dev: {'event': {'p': 0.5940695296523517, 'r': 0.7736351531291611, 'f1': 0.6720647773279351}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Test: {'event': {'p': 0.6000808734330773, 'r': 0.7745302713987474, 'f1': 0.6762360446570974}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Chinese: {'event': {'p': 0.6049382716049383, 'r': 0.9074074074074074, 'f1': 0.725925925925926}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Korean: {'event': {'p': 0.66, 'r': 0.5238095238095238, 'f1': 0.584070796460177}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} Russian: {'event': {'p': 0.46511627906976744, 'r': 0.5555555555555556, 'f1': 0.5063291139240507}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 18} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 19 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:02:31.857901: step: 4/526, loss: 0.0013688564067706466 2023-01-21 12:02:32.980091: step: 8/526, loss: 0.009413529187440872 2023-01-21 12:02:34.092071: step: 12/526, loss: 0.004992390051484108 2023-01-21 12:02:35.221774: step: 16/526, loss: 0.011704063974320889 2023-01-21 12:02:36.350182: step: 20/526, loss: 0.017786597833037376 2023-01-21 12:02:37.469347: step: 24/526, loss: 0.00257453927770257 2023-01-21 12:02:38.601704: step: 28/526, loss: 0.00282535539008677 2023-01-21 12:02:39.744366: step: 32/526, loss: 0.0008752822759561241 2023-01-21 12:02:40.895821: step: 36/526, loss: 0.0016429901588708162 2023-01-21 12:02:42.060522: step: 40/526, loss: 0.0029681206215173006 2023-01-21 12:02:43.198108: step: 44/526, loss: 0.002914714626967907 2023-01-21 12:02:44.334523: step: 48/526, loss: 0.00579118775203824 2023-01-21 12:02:45.473898: step: 52/526, loss: 0.00026912690373137593 2023-01-21 12:02:46.620948: step: 56/526, loss: 0.005245208740234375 2023-01-21 12:02:47.746745: step: 60/526, loss: 0.00550003070384264 2023-01-21 12:02:48.892777: step: 64/526, loss: 0.002261352725327015 2023-01-21 12:02:50.003247: step: 68/526, loss: 0.013428974896669388 2023-01-21 12:02:51.154906: step: 72/526, loss: 0.008398962207138538 2023-01-21 12:02:52.315940: step: 76/526, loss: 0.0016423225170001388 2023-01-21 12:02:53.453920: step: 80/526, loss: 0.017499733716249466 2023-01-21 12:02:54.580293: step: 84/526, loss: 0.017882632091641426 2023-01-21 12:02:55.731562: step: 88/526, loss: 0.013245927169919014 2023-01-21 12:02:56.882505: step: 92/526, loss: 0.0012380600674077868 2023-01-21 12:02:58.050243: step: 96/526, loss: 0.006217003334313631 2023-01-21 12:02:59.190355: step: 100/526, loss: 0.04063883051276207 2023-01-21 12:03:00.345402: step: 104/526, loss: 0.0019698143005371094 2023-01-21 12:03:01.469798: step: 108/526, loss: 0.0016646385192871094 2023-01-21 12:03:02.630160: step: 112/526, loss: 0.004708099644631147 2023-01-21 12:03:03.754477: step: 116/526, loss: 0.03622474893927574 2023-01-21 12:03:04.925856: step: 120/526, loss: 1.9168854123563506e-05 2023-01-21 12:03:06.054954: step: 124/526, loss: 0.0359533317387104 2023-01-21 12:03:07.249370: step: 128/526, loss: 0.044225696474313736 2023-01-21 12:03:08.384254: step: 132/526, loss: 0.0012380600674077868 2023-01-21 12:03:09.491241: step: 136/526, loss: 0.0010036468738690019 2023-01-21 12:03:10.627224: step: 140/526, loss: 0.002115965122357011 2023-01-21 12:03:11.754085: step: 144/526, loss: 0.010544968768954277 2023-01-21 12:03:12.910100: step: 148/526, loss: 0.0009315490606240928 2023-01-21 12:03:14.031278: step: 152/526, loss: 0.002503872150555253 2023-01-21 12:03:15.130535: step: 156/526, loss: 0.6769391298294067 2023-01-21 12:03:16.284799: step: 160/526, loss: 0.015960121527314186 2023-01-21 12:03:17.397080: step: 164/526, loss: 0.0006322860717773438 2023-01-21 12:03:18.515707: step: 168/526, loss: 0.00020385385141707957 2023-01-21 12:03:19.625797: step: 172/526, loss: 0.000362205522833392 2023-01-21 12:03:20.796861: step: 176/526, loss: 0.011219596490263939 2023-01-21 12:03:21.920896: step: 180/526, loss: 0.002458763075992465 2023-01-21 12:03:23.073197: step: 184/526, loss: 0.056160926818847656 2023-01-21 12:03:24.199532: step: 188/526, loss: 0.00288143171928823 2023-01-21 12:03:25.368553: step: 192/526, loss: 0.00013942719670012593 2023-01-21 12:03:26.495413: step: 196/526, loss: 0.001413250109180808 2023-01-21 12:03:27.625872: step: 200/526, loss: -5.817413693876006e-06 2023-01-21 12:03:28.767945: step: 204/526, loss: 0.0038799287285655737 2023-01-21 12:03:29.918718: step: 208/526, loss: 0.00246429443359375 2023-01-21 12:03:31.072793: step: 212/526, loss: 0.0011867821449413896 2023-01-21 12:03:32.226378: step: 216/526, loss: 0.005230140872299671 2023-01-21 12:03:33.335545: step: 220/526, loss: 0.000162029275088571 2023-01-21 12:03:34.472764: step: 224/526, loss: 0.08713188767433167 2023-01-21 12:03:35.605694: step: 228/526, loss: 0.0001279830903513357 2023-01-21 12:03:36.735840: step: 232/526, loss: 0.00016760826110839844 2023-01-21 12:03:37.884351: step: 236/526, loss: 0.007582283113151789 2023-01-21 12:03:39.021355: step: 240/526, loss: 0.006156730465590954 2023-01-21 12:03:40.156213: step: 244/526, loss: 0.3327018916606903 2023-01-21 12:03:41.313380: step: 248/526, loss: 0.026578903198242188 2023-01-21 12:03:42.466359: step: 252/526, loss: 0.031252577900886536 2023-01-21 12:03:43.642702: step: 256/526, loss: 0.04587908089160919 2023-01-21 12:03:44.769110: step: 260/526, loss: 0.006936264224350452 2023-01-21 12:03:45.875308: step: 264/526, loss: 0.034091949462890625 2023-01-21 12:03:47.009453: step: 268/526, loss: 0.014538002200424671 2023-01-21 12:03:48.171202: step: 272/526, loss: 0.0017744064098224044 2023-01-21 12:03:49.279610: step: 276/526, loss: 0.0016389847733080387 2023-01-21 12:03:50.427240: step: 280/526, loss: 0.00045623781625181437 2023-01-21 12:03:51.564342: step: 284/526, loss: 3.490447852527723e-05 2023-01-21 12:03:52.698412: step: 288/526, loss: 0.004144859500229359 2023-01-21 12:03:53.832323: step: 292/526, loss: 0.001394808292388916 2023-01-21 12:03:54.951645: step: 296/526, loss: 0.0065250396728515625 2023-01-21 12:03:56.062376: step: 300/526, loss: 0.00018525123596191406 2023-01-21 12:03:57.180630: step: 304/526, loss: 0.02166776731610298 2023-01-21 12:03:58.309045: step: 308/526, loss: 0.0007217407692223787 2023-01-21 12:03:59.449719: step: 312/526, loss: 0.020790722221136093 2023-01-21 12:04:00.621404: step: 316/526, loss: 0.0076885223388671875 2023-01-21 12:04:01.791297: step: 320/526, loss: 0.0006165504455566406 2023-01-21 12:04:02.937409: step: 324/526, loss: 0.00026769639225676656 2023-01-21 12:04:04.084918: step: 328/526, loss: 0.00012283996329642832 2023-01-21 12:04:05.202268: step: 332/526, loss: 0.00047454837476834655 2023-01-21 12:04:06.356128: step: 336/526, loss: 0.04615335538983345 2023-01-21 12:04:07.486849: step: 340/526, loss: 0.02505502849817276 2023-01-21 12:04:08.624068: step: 344/526, loss: 0.012150764465332031 2023-01-21 12:04:09.731199: step: 348/526, loss: 0.00016164779663085938 2023-01-21 12:04:10.859988: step: 352/526, loss: 2.555847095209174e-05 2023-01-21 12:04:11.980136: step: 356/526, loss: 0.0003909110964741558 2023-01-21 12:04:13.105738: step: 360/526, loss: 0.0009027481428347528 2023-01-21 12:04:14.225013: step: 364/526, loss: 0.0010707378387451172 2023-01-21 12:04:15.382054: step: 368/526, loss: 0.006009960081428289 2023-01-21 12:04:16.501505: step: 372/526, loss: 0.0016978741623461246 2023-01-21 12:04:17.623282: step: 376/526, loss: 2.593994213384576e-05 2023-01-21 12:04:18.778849: step: 380/526, loss: 0.41187816858291626 2023-01-21 12:04:19.931935: step: 384/526, loss: 0.01841878890991211 2023-01-21 12:04:21.064220: step: 388/526, loss: 0.0015591621631756425 2023-01-21 12:04:22.286525: step: 392/526, loss: 0.001306819962337613 2023-01-21 12:04:23.434619: step: 396/526, loss: 0.0013326645130291581 2023-01-21 12:04:24.585404: step: 400/526, loss: 0.005376148037612438 2023-01-21 12:04:25.691853: step: 404/526, loss: 0.0003391265927348286 2023-01-21 12:04:26.804188: step: 408/526, loss: 0.010370636358857155 2023-01-21 12:04:27.975406: step: 412/526, loss: 0.013152694329619408 2023-01-21 12:04:29.108115: step: 416/526, loss: 0.00027751922607421875 2023-01-21 12:04:30.239390: step: 420/526, loss: 0.003524303436279297 2023-01-21 12:04:31.387798: step: 424/526, loss: 0.0001924514799611643 2023-01-21 12:04:32.532359: step: 428/526, loss: 0.00030245783273130655 2023-01-21 12:04:33.674558: step: 432/526, loss: 0.008660316467285156 2023-01-21 12:04:34.812144: step: 436/526, loss: 0.0016996384365484118 2023-01-21 12:04:35.961794: step: 440/526, loss: 0.07145905494689941 2023-01-21 12:04:37.091314: step: 444/526, loss: 0.011229896917939186 2023-01-21 12:04:38.224411: step: 448/526, loss: 0.0004977226490154862 2023-01-21 12:04:39.403610: step: 452/526, loss: 0.0056528570130467415 2023-01-21 12:04:40.549275: step: 456/526, loss: 0.031906794756650925 2023-01-21 12:04:41.689861: step: 460/526, loss: 0.03872022405266762 2023-01-21 12:04:42.858626: step: 464/526, loss: 0.0749279037117958 2023-01-21 12:04:44.001234: step: 468/526, loss: 0.016022302210330963 2023-01-21 12:04:45.139417: step: 472/526, loss: 0.012515926733613014 2023-01-21 12:04:46.275565: step: 476/526, loss: 0.008679866790771484 2023-01-21 12:04:47.389699: step: 480/526, loss: 0.009264945983886719 2023-01-21 12:04:48.532584: step: 484/526, loss: 0.00012111663818359375 2023-01-21 12:04:49.668468: step: 488/526, loss: 0.0001070022553903982 2023-01-21 12:04:50.802525: step: 492/526, loss: 0.001993560930714011 2023-01-21 12:04:51.943697: step: 496/526, loss: 0.0893864706158638 2023-01-21 12:04:53.114342: step: 500/526, loss: 0.003783321473747492 2023-01-21 12:04:54.266410: step: 504/526, loss: 0.05378151312470436 2023-01-21 12:04:55.397808: step: 508/526, loss: 0.004395103082060814 2023-01-21 12:04:56.526443: step: 512/526, loss: 0.0027160646859556437 2023-01-21 12:04:57.654359: step: 516/526, loss: 0.0011004448169842362 2023-01-21 12:04:58.818406: step: 520/526, loss: 0.004351044073700905 2023-01-21 12:04:59.960861: step: 524/526, loss: 0.00025691985501907766 2023-01-21 12:05:01.100541: step: 528/526, loss: 0.07616004347801208 2023-01-21 12:05:02.248290: step: 532/526, loss: 0.019036103039979935 2023-01-21 12:05:03.417354: step: 536/526, loss: 0.03952808678150177 2023-01-21 12:05:04.595666: step: 540/526, loss: 0.04286599159240723 2023-01-21 12:05:05.717867: step: 544/526, loss: 0.004156684968620539 2023-01-21 12:05:06.833313: step: 548/526, loss: 0.004384326748549938 2023-01-21 12:05:07.991437: step: 552/526, loss: 0.053180109709501266 2023-01-21 12:05:09.111972: step: 556/526, loss: 0.19768953323364258 2023-01-21 12:05:10.224441: step: 560/526, loss: 0.051331423223018646 2023-01-21 12:05:11.345936: step: 564/526, loss: 0.009510518051683903 2023-01-21 12:05:12.493453: step: 568/526, loss: 0.008899974636733532 2023-01-21 12:05:13.645926: step: 572/526, loss: 0.7971934080123901 2023-01-21 12:05:14.774002: step: 576/526, loss: 5.550384594243951e-05 2023-01-21 12:05:15.950213: step: 580/526, loss: 0.08259773254394531 2023-01-21 12:05:17.063541: step: 584/526, loss: 0.002389383502304554 2023-01-21 12:05:18.212900: step: 588/526, loss: 0.07661718130111694 2023-01-21 12:05:19.323675: step: 592/526, loss: 0.01572561450302601 2023-01-21 12:05:20.464231: step: 596/526, loss: 0.002519035479053855 2023-01-21 12:05:21.597891: step: 600/526, loss: 0.06179647520184517 2023-01-21 12:05:22.727492: step: 604/526, loss: 5.702972339349799e-05 2023-01-21 12:05:23.874651: step: 608/526, loss: 0.0007839202880859375 2023-01-21 12:05:24.969134: step: 612/526, loss: 0.13024960458278656 2023-01-21 12:05:26.113295: step: 616/526, loss: 0.027873611077666283 2023-01-21 12:05:27.251724: step: 620/526, loss: 0.00025959016056731343 2023-01-21 12:05:28.389115: step: 624/526, loss: 0.005276489537209272 2023-01-21 12:05:29.517059: step: 628/526, loss: 0.0034673691261559725 2023-01-21 12:05:30.666191: step: 632/526, loss: 0.009805584326386452 2023-01-21 12:05:31.796560: step: 636/526, loss: 0.1329864263534546 2023-01-21 12:05:32.948139: step: 640/526, loss: 0.00028667450533248484 2023-01-21 12:05:34.112250: step: 644/526, loss: 0.00048513413639739156 2023-01-21 12:05:35.285347: step: 648/526, loss: 0.003796863602474332 2023-01-21 12:05:36.431711: step: 652/526, loss: 0.0006948470836505294 2023-01-21 12:05:37.556196: step: 656/526, loss: 0.0022129060234874487 2023-01-21 12:05:38.678611: step: 660/526, loss: 0.00074005126953125 2023-01-21 12:05:39.785315: step: 664/526, loss: 0.018998026847839355 2023-01-21 12:05:40.911689: step: 668/526, loss: 0.0002020835963776335 2023-01-21 12:05:42.032372: step: 672/526, loss: 0.0016401291359215975 2023-01-21 12:05:43.169610: step: 676/526, loss: 7.805824861861765e-05 2023-01-21 12:05:44.309199: step: 680/526, loss: 0.04405937343835831 2023-01-21 12:05:45.445779: step: 684/526, loss: 0.0008193969842977822 2023-01-21 12:05:46.618882: step: 688/526, loss: 0.011666107922792435 2023-01-21 12:05:47.777269: step: 692/526, loss: 0.0013137817149981856 2023-01-21 12:05:48.931799: step: 696/526, loss: 0.0564454086124897 2023-01-21 12:05:50.050081: step: 700/526, loss: 0.0020854950416833162 2023-01-21 12:05:51.182056: step: 704/526, loss: 0.0004592418554238975 2023-01-21 12:05:52.330657: step: 708/526, loss: 0.0007997512584552169 2023-01-21 12:05:53.454736: step: 712/526, loss: 0.0018640517955645919 2023-01-21 12:05:54.574287: step: 716/526, loss: 0.0017558097606524825 2023-01-21 12:05:55.689714: step: 720/526, loss: 0.005951118189841509 2023-01-21 12:05:56.807688: step: 724/526, loss: 0.011051272973418236 2023-01-21 12:05:57.964765: step: 728/526, loss: 3.461838059592992e-05 2023-01-21 12:05:59.070903: step: 732/526, loss: 0.0335872657597065 2023-01-21 12:06:00.205405: step: 736/526, loss: 0.0017366409301757812 2023-01-21 12:06:01.371247: step: 740/526, loss: 0.0009332179906778038 2023-01-21 12:06:02.493452: step: 744/526, loss: 0.0045379637740552425 2023-01-21 12:06:03.624051: step: 748/526, loss: 0.00244560232385993 2023-01-21 12:06:04.805868: step: 752/526, loss: 0.007321167271584272 2023-01-21 12:06:05.935863: step: 756/526, loss: 0.010950089432299137 2023-01-21 12:06:07.075983: step: 760/526, loss: 0.00978775043040514 2023-01-21 12:06:08.238878: step: 764/526, loss: 0.0007270813221111894 2023-01-21 12:06:09.351269: step: 768/526, loss: 0.0001143455519923009 2023-01-21 12:06:10.459694: step: 772/526, loss: 0.0007000922923907638 2023-01-21 12:06:11.619188: step: 776/526, loss: 0.02962322160601616 2023-01-21 12:06:12.753398: step: 780/526, loss: 0.009371757507324219 2023-01-21 12:06:13.881209: step: 784/526, loss: 0.03220539167523384 2023-01-21 12:06:14.988899: step: 788/526, loss: 0.0026785852387547493 2023-01-21 12:06:16.145745: step: 792/526, loss: 0.0012484551407396793 2023-01-21 12:06:17.317651: step: 796/526, loss: 0.002890300704166293 2023-01-21 12:06:18.435318: step: 800/526, loss: 0.0004921913496218622 2023-01-21 12:06:19.568718: step: 804/526, loss: 0.007667350582778454 2023-01-21 12:06:20.720613: step: 808/526, loss: 0.0024403573479503393 2023-01-21 12:06:21.844162: step: 812/526, loss: 0.03808155283331871 2023-01-21 12:06:22.991339: step: 816/526, loss: 0.001123619033023715 2023-01-21 12:06:24.151344: step: 820/526, loss: 0.0014673233963549137 2023-01-21 12:06:25.276387: step: 824/526, loss: 0.003963470458984375 2023-01-21 12:06:26.423458: step: 828/526, loss: 0.0033691406715661287 2023-01-21 12:06:27.595416: step: 832/526, loss: 0.00855865515768528 2023-01-21 12:06:28.722786: step: 836/526, loss: 0.06293001025915146 2023-01-21 12:06:29.859032: step: 840/526, loss: 0.0031220437958836555 2023-01-21 12:06:30.980640: step: 844/526, loss: 0.0012376784579828382 2023-01-21 12:06:32.144788: step: 848/526, loss: 0.0039009093306958675 2023-01-21 12:06:33.264277: step: 852/526, loss: 0.00121135707013309 2023-01-21 12:06:34.364076: step: 856/526, loss: 0.004396343603730202 2023-01-21 12:06:35.515008: step: 860/526, loss: 0.003361320588737726 2023-01-21 12:06:36.643386: step: 864/526, loss: 0.009370803833007812 2023-01-21 12:06:37.791577: step: 868/526, loss: 0.00013895034499000758 2023-01-21 12:06:38.943916: step: 872/526, loss: 0.0012997627491131425 2023-01-21 12:06:40.105140: step: 876/526, loss: 0.00011072159395553172 2023-01-21 12:06:41.234977: step: 880/526, loss: 0.0006776810041628778 2023-01-21 12:06:42.353654: step: 884/526, loss: 0.0007840633625164628 2023-01-21 12:06:43.499859: step: 888/526, loss: 0.002006149385124445 2023-01-21 12:06:44.650280: step: 892/526, loss: 0.07366943359375 2023-01-21 12:06:45.777776: step: 896/526, loss: 0.010111522860825062 2023-01-21 12:06:46.909557: step: 900/526, loss: 0.002279567765071988 2023-01-21 12:06:48.055147: step: 904/526, loss: 0.0200684554874897 2023-01-21 12:06:49.200292: step: 908/526, loss: 0.04132337495684624 2023-01-21 12:06:50.330745: step: 912/526, loss: 0.0006594657897949219 2023-01-21 12:06:51.485386: step: 916/526, loss: 0.012527895160019398 2023-01-21 12:06:52.644284: step: 920/526, loss: 0.02466306835412979 2023-01-21 12:06:53.793806: step: 924/526, loss: 0.0007373809348791838 2023-01-21 12:06:54.909583: step: 928/526, loss: 0.004340458195656538 2023-01-21 12:06:56.036397: step: 932/526, loss: 0.0008606493356637657 2023-01-21 12:06:57.160009: step: 936/526, loss: 0.0002482891140971333 2023-01-21 12:06:58.327722: step: 940/526, loss: 0.005392646882683039 2023-01-21 12:06:59.469736: step: 944/526, loss: 0.002089262008666992 2023-01-21 12:07:00.620785: step: 948/526, loss: 7.686614844715223e-05 2023-01-21 12:07:01.719869: step: 952/526, loss: 0.006218052003532648 2023-01-21 12:07:02.850148: step: 956/526, loss: 0.011279726400971413 2023-01-21 12:07:03.982840: step: 960/526, loss: 0.0017913817428052425 2023-01-21 12:07:05.123599: step: 964/526, loss: 0.025225544348359108 2023-01-21 12:07:06.288661: step: 968/526, loss: 0.0002473831409588456 2023-01-21 12:07:07.417615: step: 972/526, loss: 0.006342696957290173 2023-01-21 12:07:08.548211: step: 976/526, loss: 0.041419509798288345 2023-01-21 12:07:09.693628: step: 980/526, loss: 0.0053726197220385075 2023-01-21 12:07:10.857012: step: 984/526, loss: 0.012337112799286842 2023-01-21 12:07:11.976264: step: 988/526, loss: 0.0006395340315066278 2023-01-21 12:07:13.094253: step: 992/526, loss: 0.00015773772611282766 2023-01-21 12:07:14.225949: step: 996/526, loss: 0.0021259307395666838 2023-01-21 12:07:15.348000: step: 1000/526, loss: 0.2614225447177887 2023-01-21 12:07:16.486400: step: 1004/526, loss: 0.008846092969179153 2023-01-21 12:07:17.608932: step: 1008/526, loss: 0.0010235786903649569 2023-01-21 12:07:18.734551: step: 1012/526, loss: 0.0001354217529296875 2023-01-21 12:07:19.900817: step: 1016/526, loss: 0.007870960049331188 2023-01-21 12:07:21.036397: step: 1020/526, loss: 0.28966444730758667 2023-01-21 12:07:22.167866: step: 1024/526, loss: 0.03288641199469566 2023-01-21 12:07:23.273043: step: 1028/526, loss: 0.003364419797435403 2023-01-21 12:07:24.387042: step: 1032/526, loss: 0.00110626220703125 2023-01-21 12:07:25.519405: step: 1036/526, loss: 0.010199260897934437 2023-01-21 12:07:26.620556: step: 1040/526, loss: 0.0016743659507483244 2023-01-21 12:07:27.770328: step: 1044/526, loss: 0.0010484218364581466 2023-01-21 12:07:28.897529: step: 1048/526, loss: 0.06405496597290039 2023-01-21 12:07:30.005892: step: 1052/526, loss: 0.000982570694759488 2023-01-21 12:07:31.157032: step: 1056/526, loss: 9.89913969533518e-05 2023-01-21 12:07:32.288006: step: 1060/526, loss: 0.020412160083651543 2023-01-21 12:07:33.385878: step: 1064/526, loss: 0.005828285589814186 2023-01-21 12:07:34.524030: step: 1068/526, loss: 0.0015296937199309468 2023-01-21 12:07:35.668906: step: 1072/526, loss: 0.0021698474884033203 2023-01-21 12:07:36.831673: step: 1076/526, loss: 0.16330471634864807 2023-01-21 12:07:37.949915: step: 1080/526, loss: 0.011344050988554955 2023-01-21 12:07:39.102978: step: 1084/526, loss: 0.004983521066606045 2023-01-21 12:07:40.257354: step: 1088/526, loss: 0.03205425664782524 2023-01-21 12:07:41.385534: step: 1092/526, loss: 0.006979084108024836 2023-01-21 12:07:42.502311: step: 1096/526, loss: 7.934570021461695e-05 2023-01-21 12:07:43.661605: step: 1100/526, loss: 0.006379425525665283 2023-01-21 12:07:44.818098: step: 1104/526, loss: 0.06604786217212677 2023-01-21 12:07:45.928022: step: 1108/526, loss: 0.0006956100696697831 2023-01-21 12:07:47.056535: step: 1112/526, loss: 0.030487921088933945 2023-01-21 12:07:48.196540: step: 1116/526, loss: 0.007526207249611616 2023-01-21 12:07:49.362034: step: 1120/526, loss: 3.3572022914886475 2023-01-21 12:07:50.520559: step: 1124/526, loss: 0.010141563601791859 2023-01-21 12:07:51.660057: step: 1128/526, loss: 0.02692718431353569 2023-01-21 12:07:52.788162: step: 1132/526, loss: 0.017857838422060013 2023-01-21 12:07:53.907587: step: 1136/526, loss: 0.045717716217041016 2023-01-21 12:07:55.045239: step: 1140/526, loss: 0.004385566804558039 2023-01-21 12:07:56.193661: step: 1144/526, loss: 0.008656119927763939 2023-01-21 12:07:57.345465: step: 1148/526, loss: 0.0003086090146098286 2023-01-21 12:07:58.494074: step: 1152/526, loss: 0.00015363693819381297 2023-01-21 12:07:59.634177: step: 1156/526, loss: 0.00024380684772040695 2023-01-21 12:08:00.794812: step: 1160/526, loss: 0.1223064437508583 2023-01-21 12:08:01.947173: step: 1164/526, loss: 0.0002887725713662803 2023-01-21 12:08:03.072516: step: 1168/526, loss: 0.0001964569091796875 2023-01-21 12:08:04.222798: step: 1172/526, loss: 0.0037279129028320312 2023-01-21 12:08:05.339774: step: 1176/526, loss: 0.04041862487792969 2023-01-21 12:08:06.487667: step: 1180/526, loss: 0.003768491791561246 2023-01-21 12:08:07.614975: step: 1184/526, loss: 0.008487510494887829 2023-01-21 12:08:08.748769: step: 1188/526, loss: 0.004865836817771196 2023-01-21 12:08:09.887356: step: 1192/526, loss: 0.0011693955166265368 2023-01-21 12:08:11.022718: step: 1196/526, loss: 0.009299135766923428 2023-01-21 12:08:12.135044: step: 1200/526, loss: 0.004569625947624445 2023-01-21 12:08:13.249139: step: 1204/526, loss: 0.0004795074346475303 2023-01-21 12:08:14.422295: step: 1208/526, loss: 0.0012329102028161287 2023-01-21 12:08:15.613589: step: 1212/526, loss: 0.00013685226440429688 2023-01-21 12:08:16.738922: step: 1216/526, loss: 0.0016552448505535722 2023-01-21 12:08:17.867474: step: 1220/526, loss: 0.010214329697191715 2023-01-21 12:08:18.991951: step: 1224/526, loss: 0.020691586658358574 2023-01-21 12:08:20.137622: step: 1228/526, loss: 0.04051556438207626 2023-01-21 12:08:21.309380: step: 1232/526, loss: 0.0010293960804119706 2023-01-21 12:08:22.435449: step: 1236/526, loss: 0.004351616371423006 2023-01-21 12:08:23.591844: step: 1240/526, loss: 0.12619838118553162 2023-01-21 12:08:24.719935: step: 1244/526, loss: 0.04470691829919815 2023-01-21 12:08:25.853125: step: 1248/526, loss: 9.50813337112777e-05 2023-01-21 12:08:26.962294: step: 1252/526, loss: 0.005921554751694202 2023-01-21 12:08:28.153617: step: 1256/526, loss: 0.05605030059814453 2023-01-21 12:08:29.316263: step: 1260/526, loss: 0.03684492036700249 2023-01-21 12:08:30.447875: step: 1264/526, loss: 4.3201445805607364e-05 2023-01-21 12:08:31.589591: step: 1268/526, loss: 0.0005382538074627519 2023-01-21 12:08:32.736775: step: 1272/526, loss: 0.026052093133330345 2023-01-21 12:08:33.887379: step: 1276/526, loss: 0.023760223761200905 2023-01-21 12:08:35.023825: step: 1280/526, loss: 0.019644739106297493 2023-01-21 12:08:36.157837: step: 1284/526, loss: 0.07892999798059464 2023-01-21 12:08:37.284945: step: 1288/526, loss: 0.07939004898071289 2023-01-21 12:08:38.439147: step: 1292/526, loss: 0.007223606109619141 2023-01-21 12:08:39.592937: step: 1296/526, loss: 0.0036398889496922493 2023-01-21 12:08:40.715955: step: 1300/526, loss: 0.023408319801092148 2023-01-21 12:08:41.847320: step: 1304/526, loss: 0.0007158517837524414 2023-01-21 12:08:42.979501: step: 1308/526, loss: 0.0004914283636026084 2023-01-21 12:08:44.122028: step: 1312/526, loss: 0.015314054675400257 2023-01-21 12:08:45.254506: step: 1316/526, loss: 0.04575519636273384 2023-01-21 12:08:46.374137: step: 1320/526, loss: 0.00456581125035882 2023-01-21 12:08:47.497311: step: 1324/526, loss: 0.024317646399140358 2023-01-21 12:08:48.631052: step: 1328/526, loss: 0.0018795013893395662 2023-01-21 12:08:49.749545: step: 1332/526, loss: 0.0001222610444528982 2023-01-21 12:08:50.887960: step: 1336/526, loss: 6.780624244129285e-05 2023-01-21 12:08:52.020023: step: 1340/526, loss: 0.023294735699892044 2023-01-21 12:08:53.190134: step: 1344/526, loss: 0.005407619755715132 2023-01-21 12:08:54.342086: step: 1348/526, loss: 0.004045581910759211 2023-01-21 12:08:55.471550: step: 1352/526, loss: 0.002442646073177457 2023-01-21 12:08:56.603570: step: 1356/526, loss: 0.011979865841567516 2023-01-21 12:08:57.780018: step: 1360/526, loss: 0.0003926277277059853 2023-01-21 12:08:58.934719: step: 1364/526, loss: 0.056911028921604156 2023-01-21 12:09:00.094473: step: 1368/526, loss: 0.0006860733265057206 2023-01-21 12:09:01.276066: step: 1372/526, loss: 0.020235443487763405 2023-01-21 12:09:02.396254: step: 1376/526, loss: 0.0019110202556475997 2023-01-21 12:09:03.602759: step: 1380/526, loss: 0.0029584886506199837 2023-01-21 12:09:04.722684: step: 1384/526, loss: 0.056556131690740585 2023-01-21 12:09:05.849482: step: 1388/526, loss: 0.0019749163184314966 2023-01-21 12:09:06.985788: step: 1392/526, loss: 0.018282413482666016 2023-01-21 12:09:08.140712: step: 1396/526, loss: 0.049361322075128555 2023-01-21 12:09:09.296516: step: 1400/526, loss: 0.012869263999164104 2023-01-21 12:09:10.424818: step: 1404/526, loss: 0.013207531534135342 2023-01-21 12:09:11.584846: step: 1408/526, loss: 0.0507933609187603 2023-01-21 12:09:12.729174: step: 1412/526, loss: 0.001056671142578125 2023-01-21 12:09:13.885830: step: 1416/526, loss: 0.05953407287597656 2023-01-21 12:09:15.016965: step: 1420/526, loss: 0.00025424957857467234 2023-01-21 12:09:16.168979: step: 1424/526, loss: 0.008554554544389248 2023-01-21 12:09:17.308619: step: 1428/526, loss: 0.0001846313534770161 2023-01-21 12:09:18.496656: step: 1432/526, loss: 0.0018165588844567537 2023-01-21 12:09:19.601565: step: 1436/526, loss: 0.006369304843246937 2023-01-21 12:09:20.739054: step: 1440/526, loss: 1.2578611373901367 2023-01-21 12:09:21.855030: step: 1444/526, loss: 0.04019184038043022 2023-01-21 12:09:23.035645: step: 1448/526, loss: 0.0037845612969249487 2023-01-21 12:09:24.151271: step: 1452/526, loss: 0.011424875818192959 2023-01-21 12:09:25.290124: step: 1456/526, loss: 0.017191004008054733 2023-01-21 12:09:26.438795: step: 1460/526, loss: 0.00021476746769621968 2023-01-21 12:09:27.584096: step: 1464/526, loss: 0.02669076807796955 2023-01-21 12:09:28.700612: step: 1468/526, loss: 0.015716170892119408 2023-01-21 12:09:29.840609: step: 1472/526, loss: 0.3119939863681793 2023-01-21 12:09:30.978372: step: 1476/526, loss: 0.043900683522224426 2023-01-21 12:09:32.125661: step: 1480/526, loss: 0.013406753540039062 2023-01-21 12:09:33.247635: step: 1484/526, loss: 0.005054283421486616 2023-01-21 12:09:34.404554: step: 1488/526, loss: 9.54627976170741e-05 2023-01-21 12:09:35.513008: step: 1492/526, loss: 8.74519391800277e-05 2023-01-21 12:09:36.659684: step: 1496/526, loss: 0.0009054184192791581 2023-01-21 12:09:37.800232: step: 1500/526, loss: 0.02503795549273491 2023-01-21 12:09:38.939912: step: 1504/526, loss: 0.03236541897058487 2023-01-21 12:09:40.086570: step: 1508/526, loss: 0.011465073563158512 2023-01-21 12:09:41.212060: step: 1512/526, loss: 0.0006618499755859375 2023-01-21 12:09:42.363110: step: 1516/526, loss: 0.019887637346982956 2023-01-21 12:09:43.519891: step: 1520/526, loss: 0.024776078760623932 2023-01-21 12:09:44.665131: step: 1524/526, loss: 0.06181182712316513 2023-01-21 12:09:45.766705: step: 1528/526, loss: 0.012237167917191982 2023-01-21 12:09:46.900599: step: 1532/526, loss: 0.022452736273407936 2023-01-21 12:09:48.053792: step: 1536/526, loss: 0.004935645963996649 2023-01-21 12:09:49.201002: step: 1540/526, loss: 0.09390906989574432 2023-01-21 12:09:50.338502: step: 1544/526, loss: 0.06961756199598312 2023-01-21 12:09:51.471426: step: 1548/526, loss: 0.00020637513080146164 2023-01-21 12:09:52.626029: step: 1552/526, loss: 0.00033464431180618703 2023-01-21 12:09:53.744643: step: 1556/526, loss: 0.008463717065751553 2023-01-21 12:09:54.900632: step: 1560/526, loss: 0.006465005688369274 2023-01-21 12:09:56.034025: step: 1564/526, loss: 0.010308075696229935 2023-01-21 12:09:57.156902: step: 1568/526, loss: 0.00166234967764467 2023-01-21 12:09:58.290786: step: 1572/526, loss: 0.000884628330823034 2023-01-21 12:09:59.425317: step: 1576/526, loss: 0.02577190473675728 2023-01-21 12:10:00.588972: step: 1580/526, loss: 0.00014653205289505422 2023-01-21 12:10:01.741176: step: 1584/526, loss: 0.01144180353730917 2023-01-21 12:10:02.873483: step: 1588/526, loss: 0.06480150669813156 2023-01-21 12:10:04.021737: step: 1592/526, loss: 0.022624492645263672 2023-01-21 12:10:05.155059: step: 1596/526, loss: 0.007863807491958141 2023-01-21 12:10:06.296716: step: 1600/526, loss: 0.07386551052331924 2023-01-21 12:10:07.429034: step: 1604/526, loss: 0.00031976698664948344 2023-01-21 12:10:08.560540: step: 1608/526, loss: 0.0014957428211346269 2023-01-21 12:10:09.664543: step: 1612/526, loss: 0.0033864020369946957 2023-01-21 12:10:10.785576: step: 1616/526, loss: 0.00646553048864007 2023-01-21 12:10:11.936036: step: 1620/526, loss: 0.0007458686595782638 2023-01-21 12:10:13.080304: step: 1624/526, loss: 0.0003383636358194053 2023-01-21 12:10:14.204017: step: 1628/526, loss: 0.0008148193592205644 2023-01-21 12:10:15.349210: step: 1632/526, loss: 0.0019855499267578125 2023-01-21 12:10:16.468458: step: 1636/526, loss: 0.0309618953615427 2023-01-21 12:10:17.617814: step: 1640/526, loss: 0.11971358954906464 2023-01-21 12:10:18.761719: step: 1644/526, loss: 0.045342352241277695 2023-01-21 12:10:19.910853: step: 1648/526, loss: 0.012560510076582432 2023-01-21 12:10:21.079477: step: 1652/526, loss: 0.5659523606300354 2023-01-21 12:10:22.212547: step: 1656/526, loss: 8.678436279296875e-05 2023-01-21 12:10:23.349903: step: 1660/526, loss: 0.08686403930187225 2023-01-21 12:10:24.515033: step: 1664/526, loss: 0.01342916488647461 2023-01-21 12:10:25.632403: step: 1668/526, loss: 0.0015675545437261462 2023-01-21 12:10:26.766896: step: 1672/526, loss: 0.018215179443359375 2023-01-21 12:10:27.875407: step: 1676/526, loss: 0.0036699294578284025 2023-01-21 12:10:29.037833: step: 1680/526, loss: 0.002609443850815296 2023-01-21 12:10:30.238245: step: 1684/526, loss: 0.040708158165216446 2023-01-21 12:10:31.374497: step: 1688/526, loss: 0.00033969880314543843 2023-01-21 12:10:32.533919: step: 1692/526, loss: 0.007490825839340687 2023-01-21 12:10:33.668021: step: 1696/526, loss: 0.0001761198218446225 2023-01-21 12:10:34.800908: step: 1700/526, loss: 0.0016808509826660156 2023-01-21 12:10:35.925276: step: 1704/526, loss: 0.11907405406236649 2023-01-21 12:10:37.061250: step: 1708/526, loss: 0.000491237617097795 2023-01-21 12:10:38.198982: step: 1712/526, loss: 0.009769630618393421 2023-01-21 12:10:39.302886: step: 1716/526, loss: 0.004180431831628084 2023-01-21 12:10:40.441021: step: 1720/526, loss: 0.00036249158438295126 2023-01-21 12:10:41.567575: step: 1724/526, loss: 0.018470002338290215 2023-01-21 12:10:42.718031: step: 1728/526, loss: 0.0008161545265465975 2023-01-21 12:10:43.887684: step: 1732/526, loss: 0.0032661438453942537 2023-01-21 12:10:44.998782: step: 1736/526, loss: 0.003948402591049671 2023-01-21 12:10:46.150774: step: 1740/526, loss: 0.004609107971191406 2023-01-21 12:10:47.288048: step: 1744/526, loss: 1.087188684323337e-05 2023-01-21 12:10:48.432161: step: 1748/526, loss: 8.764266385696828e-05 2023-01-21 12:10:49.547399: step: 1752/526, loss: 0.0004631519550457597 2023-01-21 12:10:50.685911: step: 1756/526, loss: 0.04333467409014702 2023-01-21 12:10:51.837070: step: 1760/526, loss: 0.006010818760842085 2023-01-21 12:10:52.970171: step: 1764/526, loss: 0.012990808114409447 2023-01-21 12:10:54.089702: step: 1768/526, loss: 0.00030841826810501516 2023-01-21 12:10:55.229508: step: 1772/526, loss: 0.015382003039121628 2023-01-21 12:10:56.394449: step: 1776/526, loss: 0.0013078690972179174 2023-01-21 12:10:57.539523: step: 1780/526, loss: 0.01412954367697239 2023-01-21 12:10:58.666041: step: 1784/526, loss: 0.015067672356963158 2023-01-21 12:10:59.815866: step: 1788/526, loss: 0.0002401351957814768 2023-01-21 12:11:00.970589: step: 1792/526, loss: 0.010203742422163486 2023-01-21 12:11:02.083535: step: 1796/526, loss: 0.011493491940200329 2023-01-21 12:11:03.265294: step: 1800/526, loss: 0.00045299530029296875 2023-01-21 12:11:04.414772: step: 1804/526, loss: 0.00174713134765625 2023-01-21 12:11:05.578030: step: 1808/526, loss: 0.0053535462357103825 2023-01-21 12:11:06.748459: step: 1812/526, loss: 0.04392185062170029 2023-01-21 12:11:07.887449: step: 1816/526, loss: 0.03929167240858078 2023-01-21 12:11:09.016258: step: 1820/526, loss: 0.0029132843483239412 2023-01-21 12:11:10.158153: step: 1824/526, loss: 0.003966331481933594 2023-01-21 12:11:11.297760: step: 1828/526, loss: 4.119873119634576e-05 2023-01-21 12:11:12.468236: step: 1832/526, loss: 0.0009430170175619423 2023-01-21 12:11:13.603694: step: 1836/526, loss: 0.0008721351623535156 2023-01-21 12:11:14.746932: step: 1840/526, loss: 0.0004553794860839844 2023-01-21 12:11:15.893189: step: 1844/526, loss: 0.01766662672162056 2023-01-21 12:11:17.051613: step: 1848/526, loss: 0.06209850311279297 2023-01-21 12:11:18.184883: step: 1852/526, loss: 0.0024175643920898438 2023-01-21 12:11:19.339061: step: 1856/526, loss: 0.015548134222626686 2023-01-21 12:11:20.487889: step: 1860/526, loss: 0.05914344638586044 2023-01-21 12:11:21.606092: step: 1864/526, loss: 0.015493202954530716 2023-01-21 12:11:22.720487: step: 1868/526, loss: 0.0035613062791526318 2023-01-21 12:11:23.840016: step: 1872/526, loss: 0.03834743797779083 2023-01-21 12:11:24.996708: step: 1876/526, loss: 0.003623581025749445 2023-01-21 12:11:26.142852: step: 1880/526, loss: 0.10078783333301544 2023-01-21 12:11:27.319211: step: 1884/526, loss: 0.004490470979362726 2023-01-21 12:11:28.448472: step: 1888/526, loss: 0.029277898371219635 2023-01-21 12:11:29.621183: step: 1892/526, loss: 0.05825882405042648 2023-01-21 12:11:30.757790: step: 1896/526, loss: 0.0011039734818041325 2023-01-21 12:11:31.922754: step: 1900/526, loss: 0.01141433697193861 2023-01-21 12:11:33.077336: step: 1904/526, loss: 0.03253369405865669 2023-01-21 12:11:34.244405: step: 1908/526, loss: 0.04561123996973038 2023-01-21 12:11:35.392874: step: 1912/526, loss: 0.21059799194335938 2023-01-21 12:11:36.520716: step: 1916/526, loss: 0.0005908012972213328 2023-01-21 12:11:37.650754: step: 1920/526, loss: 0.03829069063067436 2023-01-21 12:11:38.795858: step: 1924/526, loss: 0.0013586044078692794 2023-01-21 12:11:39.975928: step: 1928/526, loss: 0.0010437011951580644 2023-01-21 12:11:41.106137: step: 1932/526, loss: 0.03102397918701172 2023-01-21 12:11:42.235625: step: 1936/526, loss: 0.04813690111041069 2023-01-21 12:11:43.429548: step: 1940/526, loss: 0.0011905670398846269 2023-01-21 12:11:44.571701: step: 1944/526, loss: 0.015375280752778053 2023-01-21 12:11:45.691114: step: 1948/526, loss: 0.0049228668212890625 2023-01-21 12:11:46.830874: step: 1952/526, loss: 0.015409660525619984 2023-01-21 12:11:47.982891: step: 1956/526, loss: 0.0035962106194347143 2023-01-21 12:11:49.098154: step: 1960/526, loss: 0.0007603645208291709 2023-01-21 12:11:50.202753: step: 1964/526, loss: 0.0023645402397960424 2023-01-21 12:11:51.360928: step: 1968/526, loss: 0.009099221788346767 2023-01-21 12:11:52.485752: step: 1972/526, loss: 1.1745659112930298 2023-01-21 12:11:53.610376: step: 1976/526, loss: 0.00012474060349632055 2023-01-21 12:11:54.742546: step: 1980/526, loss: 0.05372639000415802 2023-01-21 12:11:55.860311: step: 1984/526, loss: 0.0029954910278320312 2023-01-21 12:11:56.995933: step: 1988/526, loss: 0.02149381674826145 2023-01-21 12:11:58.126090: step: 1992/526, loss: 0.010166740976274014 2023-01-21 12:11:59.237538: step: 1996/526, loss: 0.006605816073715687 2023-01-21 12:12:00.369558: step: 2000/526, loss: 0.004673957824707031 2023-01-21 12:12:01.497079: step: 2004/526, loss: 0.05335378646850586 2023-01-21 12:12:02.611971: step: 2008/526, loss: 0.002769661135971546 2023-01-21 12:12:03.708087: step: 2012/526, loss: 0.012269878759980202 2023-01-21 12:12:04.837271: step: 2016/526, loss: 0.0002658843877725303 2023-01-21 12:12:05.957419: step: 2020/526, loss: 0.008564758114516735 2023-01-21 12:12:07.092052: step: 2024/526, loss: 0.020326614379882812 2023-01-21 12:12:08.216334: step: 2028/526, loss: 0.004642868414521217 2023-01-21 12:12:09.351081: step: 2032/526, loss: 0.05483055114746094 2023-01-21 12:12:10.505886: step: 2036/526, loss: 0.03961320221424103 2023-01-21 12:12:11.649665: step: 2040/526, loss: 0.0009190559503622353 2023-01-21 12:12:12.754647: step: 2044/526, loss: 0.00111560826189816 2023-01-21 12:12:13.897592: step: 2048/526, loss: 0.0001163482666015625 2023-01-21 12:12:15.022000: step: 2052/526, loss: 0.0025037764571607113 2023-01-21 12:12:16.204996: step: 2056/526, loss: 0.0027428148314356804 2023-01-21 12:12:17.341163: step: 2060/526, loss: 0.000681972480379045 2023-01-21 12:12:18.475424: step: 2064/526, loss: 0.04315071552991867 2023-01-21 12:12:19.585087: step: 2068/526, loss: 0.06378169357776642 2023-01-21 12:12:20.766781: step: 2072/526, loss: 0.05233726650476456 2023-01-21 12:12:21.884545: step: 2076/526, loss: 0.05550270155072212 2023-01-21 12:12:23.003184: step: 2080/526, loss: 0.000972366367932409 2023-01-21 12:12:24.112523: step: 2084/526, loss: 0.0005901813274249434 2023-01-21 12:12:25.257681: step: 2088/526, loss: 3.24249267578125e-05 2023-01-21 12:12:26.371332: step: 2092/526, loss: 0.002309608506038785 2023-01-21 12:12:27.505245: step: 2096/526, loss: 0.001995277591049671 2023-01-21 12:12:28.702826: step: 2100/526, loss: 0.010618305765092373 2023-01-21 12:12:29.866066: step: 2104/526, loss: 0.6264249682426453 ================================================== Loss: 0.035 -------------------- Dev: {'event': {'p': 0.6002076843198338, 'r': 0.7696404793608522, 'f1': 0.6744457409568262}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Test: {'event': {'p': 0.6195244055068836, 'r': 0.7750521920668059, 'f1': 0.6886158126594017}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Chinese: {'event': {'p': 0.6049382716049383, 'r': 0.9074074074074074, 'f1': 0.725925925925926}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Korean: {'event': {'p': 0.625, 'r': 0.47619047619047616, 'f1': 0.5405405405405405}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} Russian: {'event': {'p': 0.48717948717948717, 'r': 0.5277777777777778, 'f1': 0.5066666666666667}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 19} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 20 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:13:13.061055: step: 4/526, loss: 0.008662414737045765 2023-01-21 12:13:14.208185: step: 8/526, loss: 0.0022936819586902857 2023-01-21 12:13:15.327115: step: 12/526, loss: 0.026631593704223633 2023-01-21 12:13:16.448129: step: 16/526, loss: 0.004110813140869141 2023-01-21 12:13:17.612872: step: 20/526, loss: 0.03275737911462784 2023-01-21 12:13:18.769807: step: 24/526, loss: 0.008294296450912952 2023-01-21 12:13:19.894238: step: 28/526, loss: 0.0003543853817973286 2023-01-21 12:13:21.029667: step: 32/526, loss: 0.0267989169806242 2023-01-21 12:13:22.174416: step: 36/526, loss: 0.0016130447620525956 2023-01-21 12:13:23.345416: step: 40/526, loss: 0.0018735886551439762 2023-01-21 12:13:24.482765: step: 44/526, loss: 0.017012406140565872 2023-01-21 12:13:25.619816: step: 48/526, loss: 0.031696319580078125 2023-01-21 12:13:26.748690: step: 52/526, loss: 0.0003904342884197831 2023-01-21 12:13:27.897870: step: 56/526, loss: 0.0060860635712742805 2023-01-21 12:13:29.040740: step: 60/526, loss: 0.004594231024384499 2023-01-21 12:13:30.175146: step: 64/526, loss: 0.0503849983215332 2023-01-21 12:13:31.291672: step: 68/526, loss: 0.03251781314611435 2023-01-21 12:13:32.403821: step: 72/526, loss: 0.030513763427734375 2023-01-21 12:13:33.533766: step: 76/526, loss: 0.004304409027099609 2023-01-21 12:13:34.690272: step: 80/526, loss: 0.02356109768152237 2023-01-21 12:13:35.845595: step: 84/526, loss: 0.0033503533340990543 2023-01-21 12:13:36.977392: step: 88/526, loss: 0.10881023108959198 2023-01-21 12:13:38.116123: step: 92/526, loss: 0.005959033966064453 2023-01-21 12:13:39.236556: step: 96/526, loss: 0.038561441004276276 2023-01-21 12:13:40.372235: step: 100/526, loss: 0.08673276752233505 2023-01-21 12:13:41.563476: step: 104/526, loss: 0.009797859936952591 2023-01-21 12:13:42.686350: step: 108/526, loss: 0.0013260364066809416 2023-01-21 12:13:43.866210: step: 112/526, loss: 0.0035336497239768505 2023-01-21 12:13:44.982620: step: 116/526, loss: 0.0030756951309740543 2023-01-21 12:13:46.116661: step: 120/526, loss: 0.00119953160174191 2023-01-21 12:13:47.248491: step: 124/526, loss: 0.13245049118995667 2023-01-21 12:13:48.372168: step: 128/526, loss: 0.026577377691864967 2023-01-21 12:13:49.523276: step: 132/526, loss: 0.0005876541254110634 2023-01-21 12:13:50.684016: step: 136/526, loss: 4.6539309551008046e-05 2023-01-21 12:13:51.841048: step: 140/526, loss: 0.46959877014160156 2023-01-21 12:13:52.948975: step: 144/526, loss: 4.472732689464465e-05 2023-01-21 12:13:54.073179: step: 148/526, loss: 0.06671886146068573 2023-01-21 12:13:55.198176: step: 152/526, loss: 0.01216268539428711 2023-01-21 12:13:56.325661: step: 156/526, loss: 0.07796640694141388 2023-01-21 12:13:57.491930: step: 160/526, loss: 0.0018405914306640625 2023-01-21 12:13:58.611712: step: 164/526, loss: 0.013413524255156517 2023-01-21 12:13:59.760545: step: 168/526, loss: 0.01057729683816433 2023-01-21 12:14:00.915002: step: 172/526, loss: 0.003068685531616211 2023-01-21 12:14:02.059589: step: 176/526, loss: 0.011515045538544655 2023-01-21 12:14:03.204351: step: 180/526, loss: 0.0022478103637695312 2023-01-21 12:14:04.324870: step: 184/526, loss: 0.00024209023104049265 2023-01-21 12:14:05.476056: step: 188/526, loss: 0.015596580691635609 2023-01-21 12:14:06.593754: step: 192/526, loss: 0.001821327256038785 2023-01-21 12:14:07.731599: step: 196/526, loss: 0.002061271807178855 2023-01-21 12:14:08.875764: step: 200/526, loss: 0.0034610747825354338 2023-01-21 12:14:10.019947: step: 204/526, loss: 0.0010433197021484375 2023-01-21 12:14:11.145061: step: 208/526, loss: 0.066799096763134 2023-01-21 12:14:12.288040: step: 212/526, loss: 0.02425518073141575 2023-01-21 12:14:13.466425: step: 216/526, loss: 0.0002548217598814517 2023-01-21 12:14:14.615992: step: 220/526, loss: 8.0108642578125e-05 2023-01-21 12:14:15.727233: step: 224/526, loss: 0.0036037445534020662 2023-01-21 12:14:16.871684: step: 228/526, loss: 0.0024479867424815893 2023-01-21 12:14:18.050904: step: 232/526, loss: 0.0008649826049804688 2023-01-21 12:14:19.186543: step: 236/526, loss: 7.610320608364418e-05 2023-01-21 12:14:20.330890: step: 240/526, loss: 0.007566928863525391 2023-01-21 12:14:21.493818: step: 244/526, loss: 0.014372158795595169 2023-01-21 12:14:22.672592: step: 248/526, loss: 0.0122231962159276 2023-01-21 12:14:23.851778: step: 252/526, loss: 0.010809803381562233 2023-01-21 12:14:24.992791: step: 256/526, loss: 0.0030279161874204874 2023-01-21 12:14:26.125041: step: 260/526, loss: 0.017367172986268997 2023-01-21 12:14:27.257487: step: 264/526, loss: 0.003933716099709272 2023-01-21 12:14:28.371891: step: 268/526, loss: 0.0010600090026855469 2023-01-21 12:14:29.536884: step: 272/526, loss: 0.005539131350815296 2023-01-21 12:14:30.661069: step: 276/526, loss: 0.0018835067749023438 2023-01-21 12:14:31.809400: step: 280/526, loss: 0.04064541310071945 2023-01-21 12:14:32.944161: step: 284/526, loss: 0.018976211547851562 2023-01-21 12:14:34.086448: step: 288/526, loss: 0.000465583783807233 2023-01-21 12:14:35.198158: step: 292/526, loss: 0.016098307445645332 2023-01-21 12:14:36.318335: step: 296/526, loss: 0.03547964245080948 2023-01-21 12:14:37.424547: step: 300/526, loss: 3.4332275390625e-05 2023-01-21 12:14:38.554259: step: 304/526, loss: 0.0005561828729696572 2023-01-21 12:14:39.709077: step: 308/526, loss: 0.04275531694293022 2023-01-21 12:14:40.830815: step: 312/526, loss: 0.029764842242002487 2023-01-21 12:14:41.991175: step: 316/526, loss: 0.0011840820079669356 2023-01-21 12:14:43.140568: step: 320/526, loss: 0.00020751955162268132 2023-01-21 12:14:44.296178: step: 324/526, loss: 0.0038070681039243937 2023-01-21 12:14:45.416531: step: 328/526, loss: 0.0008877277723513544 2023-01-21 12:14:46.559314: step: 332/526, loss: 0.0111550809815526 2023-01-21 12:14:47.702067: step: 336/526, loss: 0.005178261082619429 2023-01-21 12:14:48.860429: step: 340/526, loss: 0.005114555358886719 2023-01-21 12:14:49.987947: step: 344/526, loss: 0.04802040755748749 2023-01-21 12:14:51.091753: step: 348/526, loss: 0.03889113664627075 2023-01-21 12:14:52.234226: step: 352/526, loss: 0.0025335310492664576 2023-01-21 12:14:53.397415: step: 356/526, loss: 0.0007738113636150956 2023-01-21 12:14:54.545311: step: 360/526, loss: 0.003078365232795477 2023-01-21 12:14:55.647649: step: 364/526, loss: 0.00016260147094726562 2023-01-21 12:14:56.757452: step: 368/526, loss: 0.00022807122149970382 2023-01-21 12:14:57.878533: step: 372/526, loss: 0.0009864807361736894 2023-01-21 12:14:59.023296: step: 376/526, loss: 0.013430309481918812 2023-01-21 12:15:00.153847: step: 380/526, loss: 0.039034079760313034 2023-01-21 12:15:01.272482: step: 384/526, loss: 0.022102834656834602 2023-01-21 12:15:02.461074: step: 388/526, loss: 0.04153728485107422 2023-01-21 12:15:03.631075: step: 392/526, loss: 0.019054889678955078 2023-01-21 12:15:04.760742: step: 396/526, loss: 0.002621936844661832 2023-01-21 12:15:05.921298: step: 400/526, loss: 0.0005519867409020662 2023-01-21 12:15:07.091173: step: 404/526, loss: 0.004413604736328125 2023-01-21 12:15:08.204580: step: 408/526, loss: 0.0004318237479310483 2023-01-21 12:15:09.339876: step: 412/526, loss: 0.0005514145013876259 2023-01-21 12:15:10.469705: step: 416/526, loss: 0.011383152566850185 2023-01-21 12:15:11.610471: step: 420/526, loss: 0.000705337559338659 2023-01-21 12:15:12.727367: step: 424/526, loss: 0.0006591796991415322 2023-01-21 12:15:13.878237: step: 428/526, loss: 0.016489028930664062 2023-01-21 12:15:15.028055: step: 432/526, loss: 0.0018028260674327612 2023-01-21 12:15:16.161280: step: 436/526, loss: 0.004520082380622625 2023-01-21 12:15:17.289855: step: 440/526, loss: 0.019710969179868698 2023-01-21 12:15:18.421223: step: 444/526, loss: 0.02992267720401287 2023-01-21 12:15:19.566870: step: 448/526, loss: 0.0019376755226403475 2023-01-21 12:15:20.736312: step: 452/526, loss: 0.010999870486557484 2023-01-21 12:15:21.853660: step: 456/526, loss: 0.001827335450798273 2023-01-21 12:15:22.983569: step: 460/526, loss: 0.030975341796875 2023-01-21 12:15:24.111607: step: 464/526, loss: 0.003909969702363014 2023-01-21 12:15:25.245362: step: 468/526, loss: 0.01716022565960884 2023-01-21 12:15:26.414490: step: 472/526, loss: 0.02118987962603569 2023-01-21 12:15:27.584708: step: 476/526, loss: 0.005733489990234375 2023-01-21 12:15:28.684790: step: 480/526, loss: 0.035212136805057526 2023-01-21 12:15:29.826085: step: 484/526, loss: 0.0033697127364575863 2023-01-21 12:15:30.939829: step: 488/526, loss: 0.00046815871610306203 2023-01-21 12:15:32.051371: step: 492/526, loss: 0.00011920929682673886 2023-01-21 12:15:33.176871: step: 496/526, loss: 0.00930414255708456 2023-01-21 12:15:34.351639: step: 500/526, loss: 0.03080577775835991 2023-01-21 12:15:35.481938: step: 504/526, loss: 0.0016635895008221269 2023-01-21 12:15:36.611019: step: 508/526, loss: 0.018720198422670364 2023-01-21 12:15:37.733370: step: 512/526, loss: 0.021756650879979134 2023-01-21 12:15:38.855547: step: 516/526, loss: 0.003030061721801758 2023-01-21 12:15:39.996319: step: 520/526, loss: 0.006154346279799938 2023-01-21 12:15:41.139853: step: 524/526, loss: 0.03164198622107506 2023-01-21 12:15:42.268757: step: 528/526, loss: 0.0017748832469806075 2023-01-21 12:15:43.431049: step: 532/526, loss: 0.9338005185127258 2023-01-21 12:15:44.583899: step: 536/526, loss: 0.016043568029999733 2023-01-21 12:15:45.715331: step: 540/526, loss: 0.0060829161666333675 2023-01-21 12:15:46.857648: step: 544/526, loss: 0.0071166991256177425 2023-01-21 12:15:47.969597: step: 548/526, loss: 0.022026633843779564 2023-01-21 12:15:49.113783: step: 552/526, loss: 0.012973785400390625 2023-01-21 12:15:50.241163: step: 556/526, loss: 0.04305992275476456 2023-01-21 12:15:51.413644: step: 560/526, loss: 0.1744460165500641 2023-01-21 12:15:52.527886: step: 564/526, loss: 0.005134296137839556 2023-01-21 12:15:53.662901: step: 568/526, loss: 0.01647663116455078 2023-01-21 12:15:54.776936: step: 572/526, loss: 0.017051076516509056 2023-01-21 12:15:55.913699: step: 576/526, loss: 0.00013923646474722773 2023-01-21 12:15:57.068866: step: 580/526, loss: 0.8529098629951477 2023-01-21 12:15:58.184581: step: 584/526, loss: 0.00033941271249204874 2023-01-21 12:15:59.321148: step: 588/526, loss: 0.04522562026977539 2023-01-21 12:16:00.456513: step: 592/526, loss: 0.013078402727842331 2023-01-21 12:16:01.586936: step: 596/526, loss: 7.896423630882055e-05 2023-01-21 12:16:02.719797: step: 600/526, loss: 0.00023504496493842453 2023-01-21 12:16:03.852459: step: 604/526, loss: 0.006578922271728516 2023-01-21 12:16:05.037943: step: 608/526, loss: 0.004994774237275124 2023-01-21 12:16:06.195854: step: 612/526, loss: 0.015529632568359375 2023-01-21 12:16:07.302114: step: 616/526, loss: 0.00028600695077329874 2023-01-21 12:16:08.466927: step: 620/526, loss: 0.045827675610780716 2023-01-21 12:16:09.608968: step: 624/526, loss: 0.029549982398748398 2023-01-21 12:16:10.734184: step: 628/526, loss: 0.020139122381806374 2023-01-21 12:16:11.870185: step: 632/526, loss: 0.08129234611988068 2023-01-21 12:16:12.990626: step: 636/526, loss: 0.025841902941465378 2023-01-21 12:16:14.133942: step: 640/526, loss: 0.00016098022751975805 2023-01-21 12:16:15.314745: step: 644/526, loss: 0.004547500517219305 2023-01-21 12:16:16.435391: step: 648/526, loss: 0.008652115240693092 2023-01-21 12:16:17.581893: step: 652/526, loss: 0.036055564880371094 2023-01-21 12:16:18.709429: step: 656/526, loss: 0.045095063745975494 2023-01-21 12:16:19.829779: step: 660/526, loss: 0.039093874394893646 2023-01-21 12:16:20.977510: step: 664/526, loss: 0.01661090739071369 2023-01-21 12:16:22.137010: step: 668/526, loss: 0.019431401044130325 2023-01-21 12:16:23.313761: step: 672/526, loss: 0.0045073507353663445 2023-01-21 12:16:24.445239: step: 676/526, loss: 0.0020751953125 2023-01-21 12:16:25.574518: step: 680/526, loss: 0.010302400216460228 2023-01-21 12:16:26.751092: step: 684/526, loss: 0.0017220497829839587 2023-01-21 12:16:27.912322: step: 688/526, loss: 0.010897255502641201 2023-01-21 12:16:29.053098: step: 692/526, loss: 0.0023220062721520662 2023-01-21 12:16:30.192848: step: 696/526, loss: 0.07908478379249573 2023-01-21 12:16:31.311294: step: 700/526, loss: 0.006512641906738281 2023-01-21 12:16:32.432854: step: 704/526, loss: 1.4495850336970761e-05 2023-01-21 12:16:33.585969: step: 708/526, loss: 9.069442603504285e-05 2023-01-21 12:16:34.763456: step: 712/526, loss: 3.070831735385582e-05 2023-01-21 12:16:35.881946: step: 716/526, loss: 0.005942916963249445 2023-01-21 12:16:37.026420: step: 720/526, loss: 0.027880191802978516 2023-01-21 12:16:38.133263: step: 724/526, loss: 0.014525700360536575 2023-01-21 12:16:39.265051: step: 728/526, loss: 0.020452117547392845 2023-01-21 12:16:40.461009: step: 732/526, loss: 0.0024368285667151213 2023-01-21 12:16:41.606975: step: 736/526, loss: 0.035151004791259766 2023-01-21 12:16:42.741544: step: 740/526, loss: 0.003737926483154297 2023-01-21 12:16:43.862051: step: 744/526, loss: 0.0006657600752077997 2023-01-21 12:16:44.995952: step: 748/526, loss: 0.0006807803874835372 2023-01-21 12:16:46.111943: step: 752/526, loss: 0.006023693364113569 2023-01-21 12:16:47.215169: step: 756/526, loss: 0.00024518967256881297 2023-01-21 12:16:48.339871: step: 760/526, loss: 0.00254230503924191 2023-01-21 12:16:49.466233: step: 764/526, loss: 0.009168815799057484 2023-01-21 12:16:50.579377: step: 768/526, loss: 0.009664822369813919 2023-01-21 12:16:51.707853: step: 772/526, loss: 0.011925315484404564 2023-01-21 12:16:52.840489: step: 776/526, loss: 0.03355713188648224 2023-01-21 12:16:53.979170: step: 780/526, loss: 0.0022844793274998665 2023-01-21 12:16:55.127229: step: 784/526, loss: 0.03353118896484375 2023-01-21 12:16:56.261864: step: 788/526, loss: 0.0007905960083007812 2023-01-21 12:16:57.376502: step: 792/526, loss: 0.0034228325821459293 2023-01-21 12:16:58.499747: step: 796/526, loss: 0.0060859681107103825 2023-01-21 12:16:59.645024: step: 800/526, loss: 0.02203807793557644 2023-01-21 12:17:00.794088: step: 804/526, loss: 0.0038338659796863794 2023-01-21 12:17:01.878638: step: 808/526, loss: 0.00023289918317459524 2023-01-21 12:17:02.993762: step: 812/526, loss: 8.01086389401462e-06 2023-01-21 12:17:04.134615: step: 816/526, loss: 0.009660053066909313 2023-01-21 12:17:05.276363: step: 820/526, loss: 0.006104851141571999 2023-01-21 12:17:06.410739: step: 824/526, loss: 0.00021295547776389867 2023-01-21 12:17:07.549838: step: 828/526, loss: 0.0024540661834180355 2023-01-21 12:17:08.658638: step: 832/526, loss: 0.0009092808468267322 2023-01-21 12:17:09.792357: step: 836/526, loss: 0.0017095566727221012 2023-01-21 12:17:10.938449: step: 840/526, loss: 0.002094268798828125 2023-01-21 12:17:12.095622: step: 844/526, loss: 0.014912033453583717 2023-01-21 12:17:13.212346: step: 848/526, loss: 0.0023727419320493937 2023-01-21 12:17:14.347909: step: 852/526, loss: 0.0012662888038903475 2023-01-21 12:17:15.491316: step: 856/526, loss: 0.0008008956792764366 2023-01-21 12:17:16.634503: step: 860/526, loss: 0.027820492163300514 2023-01-21 12:17:17.751435: step: 864/526, loss: 0.00093164446298033 2023-01-21 12:17:18.912915: step: 868/526, loss: 0.006908941548317671 2023-01-21 12:17:20.035558: step: 872/526, loss: 0.0052032470703125 2023-01-21 12:17:21.181190: step: 876/526, loss: 0.002834892366081476 2023-01-21 12:17:22.325607: step: 880/526, loss: 8.78334030858241e-05 2023-01-21 12:17:23.491323: step: 884/526, loss: 0.004516029264777899 2023-01-21 12:17:24.610985: step: 888/526, loss: 3.0231476557673886e-05 2023-01-21 12:17:25.773121: step: 892/526, loss: 0.17814789712429047 2023-01-21 12:17:26.920028: step: 896/526, loss: 0.0008721351623535156 2023-01-21 12:17:28.057422: step: 900/526, loss: 0.003397655440494418 2023-01-21 12:17:29.198898: step: 904/526, loss: 0.0005017280345782638 2023-01-21 12:17:30.324264: step: 908/526, loss: 0.0015083312755450606 2023-01-21 12:17:31.446378: step: 912/526, loss: 0.003467035247012973 2023-01-21 12:17:32.594656: step: 916/526, loss: 0.00012092590623069555 2023-01-21 12:17:33.761788: step: 920/526, loss: 0.0008860588422976434 2023-01-21 12:17:34.893136: step: 924/526, loss: 0.008481503464281559 2023-01-21 12:17:36.007795: step: 928/526, loss: 0.05268440395593643 2023-01-21 12:17:37.172186: step: 932/526, loss: 0.0008344650268554688 2023-01-21 12:17:38.297508: step: 936/526, loss: 0.08935165405273438 2023-01-21 12:17:39.424297: step: 940/526, loss: 0.00023298263840842992 2023-01-21 12:17:40.545445: step: 944/526, loss: 0.010291003622114658 2023-01-21 12:17:41.669180: step: 948/526, loss: 0.009380913339555264 2023-01-21 12:17:42.793230: step: 952/526, loss: 0.0025942802894860506 2023-01-21 12:17:43.930366: step: 956/526, loss: 3.6245583032723516e-05 2023-01-21 12:17:45.049913: step: 960/526, loss: 0.010713004507124424 2023-01-21 12:17:46.219325: step: 964/526, loss: 0.01556782703846693 2023-01-21 12:17:47.368105: step: 968/526, loss: 0.00559577951207757 2023-01-21 12:17:48.509883: step: 972/526, loss: 0.0298798568546772 2023-01-21 12:17:49.647986: step: 976/526, loss: 0.010061835870146751 2023-01-21 12:17:50.753186: step: 980/526, loss: 0.00012836456880904734 2023-01-21 12:17:51.885193: step: 984/526, loss: 0.061211489140987396 2023-01-21 12:17:53.028848: step: 988/526, loss: 0.00014433861360885203 2023-01-21 12:17:54.156406: step: 992/526, loss: 0.023717977106571198 2023-01-21 12:17:55.276091: step: 996/526, loss: 0.0044172764755785465 2023-01-21 12:17:56.401836: step: 1000/526, loss: 0.0002582550223451108 2023-01-21 12:17:57.513063: step: 1004/526, loss: 0.05900716781616211 2023-01-21 12:17:58.693557: step: 1008/526, loss: 0.0012770653702318668 2023-01-21 12:17:59.828775: step: 1012/526, loss: 0.00642442749813199 2023-01-21 12:18:00.970437: step: 1016/526, loss: 0.007118606474250555 2023-01-21 12:18:02.120461: step: 1020/526, loss: 0.0038975717034190893 2023-01-21 12:18:03.254211: step: 1024/526, loss: 0.01047973707318306 2023-01-21 12:18:04.404922: step: 1028/526, loss: -8.39233416627394e-06 2023-01-21 12:18:05.545095: step: 1032/526, loss: 0.026009369641542435 2023-01-21 12:18:06.680792: step: 1036/526, loss: 0.00013732911611441523 2023-01-21 12:18:07.837836: step: 1040/526, loss: 0.03740882873535156 2023-01-21 12:18:08.980660: step: 1044/526, loss: 0.026148606091737747 2023-01-21 12:18:10.124016: step: 1048/526, loss: 0.028035450726747513 2023-01-21 12:18:11.230826: step: 1052/526, loss: 0.0016407014336436987 2023-01-21 12:18:12.385516: step: 1056/526, loss: 0.0015090943779796362 2023-01-21 12:18:13.536013: step: 1060/526, loss: 0.007592296227812767 2023-01-21 12:18:14.648743: step: 1064/526, loss: 0.028738783672451973 2023-01-21 12:18:15.770515: step: 1068/526, loss: 0.00063323974609375 2023-01-21 12:18:16.906273: step: 1072/526, loss: 0.1336756944656372 2023-01-21 12:18:18.037924: step: 1076/526, loss: 0.006005287170410156 2023-01-21 12:18:19.193571: step: 1080/526, loss: 0.0259659793227911 2023-01-21 12:18:20.307653: step: 1084/526, loss: 0.038306236267089844 2023-01-21 12:18:21.440104: step: 1088/526, loss: 0.00036706926766783 2023-01-21 12:18:22.588981: step: 1092/526, loss: 0.0016206741565838456 2023-01-21 12:18:23.731614: step: 1096/526, loss: 0.021438980475068092 2023-01-21 12:18:24.891216: step: 1100/526, loss: 8.702278137207031e-05 2023-01-21 12:18:26.062472: step: 1104/526, loss: 0.012619209475815296 2023-01-21 12:18:27.194552: step: 1108/526, loss: 0.007122039794921875 2023-01-21 12:18:28.329128: step: 1112/526, loss: 0.001700592110864818 2023-01-21 12:18:29.476682: step: 1116/526, loss: 0.0017173767555505037 2023-01-21 12:18:30.626981: step: 1120/526, loss: 0.02399158664047718 2023-01-21 12:18:31.741989: step: 1124/526, loss: 0.009321880526840687 2023-01-21 12:18:32.877777: step: 1128/526, loss: 0.0030158995650708675 2023-01-21 12:18:34.041982: step: 1132/526, loss: 0.009538650512695312 2023-01-21 12:18:35.174041: step: 1136/526, loss: 0.003050136612728238 2023-01-21 12:18:36.333561: step: 1140/526, loss: 0.02655353583395481 2023-01-21 12:18:37.470984: step: 1144/526, loss: 0.007510185241699219 2023-01-21 12:18:38.615274: step: 1148/526, loss: 0.00955886859446764 2023-01-21 12:18:39.770481: step: 1152/526, loss: 0.0025625228881835938 2023-01-21 12:18:40.926891: step: 1156/526, loss: 0.008837985806167126 2023-01-21 12:18:42.099631: step: 1160/526, loss: 0.023253632709383965 2023-01-21 12:18:43.228941: step: 1164/526, loss: 4.77790817967616e-05 2023-01-21 12:18:44.393116: step: 1168/526, loss: 0.0711750015616417 2023-01-21 12:18:45.556618: step: 1172/526, loss: 0.0004561901150736958 2023-01-21 12:18:46.671385: step: 1176/526, loss: 0.007734775543212891 2023-01-21 12:18:47.799894: step: 1180/526, loss: 0.010203743353486061 2023-01-21 12:18:48.939382: step: 1184/526, loss: 0.007632446475327015 2023-01-21 12:18:50.088929: step: 1188/526, loss: 0.0016374588012695312 2023-01-21 12:18:51.228688: step: 1192/526, loss: 0.00011482239642646164 2023-01-21 12:18:52.340818: step: 1196/526, loss: 0.0007000923505984247 2023-01-21 12:18:53.507039: step: 1200/526, loss: 0.030180789530277252 2023-01-21 12:18:54.662318: step: 1204/526, loss: 0.004850482568144798 2023-01-21 12:18:55.811316: step: 1208/526, loss: 0.0007051468128338456 2023-01-21 12:18:56.928924: step: 1212/526, loss: 0.0001503944513387978 2023-01-21 12:18:58.077609: step: 1216/526, loss: 7.333754911087453e-05 2023-01-21 12:18:59.204442: step: 1220/526, loss: 8.983612497104332e-05 2023-01-21 12:19:00.359989: step: 1224/526, loss: 0.017517566680908203 2023-01-21 12:19:01.497753: step: 1228/526, loss: 0.0009068489307537675 2023-01-21 12:19:02.652473: step: 1232/526, loss: 0.0009227752452716231 2023-01-21 12:19:03.770722: step: 1236/526, loss: 0.018269730731844902 2023-01-21 12:19:04.914011: step: 1240/526, loss: 0.10956688970327377 2023-01-21 12:19:06.051551: step: 1244/526, loss: 0.012059974484145641 2023-01-21 12:19:07.187622: step: 1248/526, loss: 0.002231693360954523 2023-01-21 12:19:08.316568: step: 1252/526, loss: 0.006430435460060835 2023-01-21 12:19:09.448780: step: 1256/526, loss: 0.029808808118104935 2023-01-21 12:19:10.576605: step: 1260/526, loss: 0.09743861854076385 2023-01-21 12:19:11.726256: step: 1264/526, loss: 0.0007110595470294356 2023-01-21 12:19:12.908485: step: 1268/526, loss: 0.00442085275426507 2023-01-21 12:19:14.034534: step: 1272/526, loss: 5.769729978055693e-05 2023-01-21 12:19:15.178622: step: 1276/526, loss: 0.026811979711055756 2023-01-21 12:19:16.314748: step: 1280/526, loss: 0.009513664990663528 2023-01-21 12:19:17.489225: step: 1284/526, loss: 0.0004098892386537045 2023-01-21 12:19:18.622256: step: 1288/526, loss: 3.056526475120336e-05 2023-01-21 12:19:19.792709: step: 1292/526, loss: 0.0001312017411692068 2023-01-21 12:19:20.911124: step: 1296/526, loss: 0.0023645402397960424 2023-01-21 12:19:22.045834: step: 1300/526, loss: 0.0050830841064453125 2023-01-21 12:19:23.173140: step: 1304/526, loss: 0.011212158016860485 2023-01-21 12:19:24.317432: step: 1308/526, loss: 0.006687736138701439 2023-01-21 12:19:25.443745: step: 1312/526, loss: 0.022497082129120827 2023-01-21 12:19:26.578116: step: 1316/526, loss: 0.0004042625369038433 2023-01-21 12:19:27.709397: step: 1320/526, loss: 0.0005655288696289062 2023-01-21 12:19:28.866370: step: 1324/526, loss: 0.0006029129144735634 2023-01-21 12:19:29.993773: step: 1328/526, loss: 0.0004343033069744706 2023-01-21 12:19:31.108651: step: 1332/526, loss: 0.0008636474958620965 2023-01-21 12:19:32.246506: step: 1336/526, loss: 0.0014085769653320312 2023-01-21 12:19:33.400142: step: 1340/526, loss: 0.007404327392578125 2023-01-21 12:19:34.540106: step: 1344/526, loss: 1.2302399227337446e-05 2023-01-21 12:19:35.677646: step: 1348/526, loss: 0.0005500793340615928 2023-01-21 12:19:36.811437: step: 1352/526, loss: 0.013394355773925781 2023-01-21 12:19:37.945195: step: 1356/526, loss: 0.0038290023803710938 2023-01-21 12:19:39.087622: step: 1360/526, loss: 0.032061006873846054 2023-01-21 12:19:40.221509: step: 1364/526, loss: 0.0014435768825933337 2023-01-21 12:19:41.379210: step: 1368/526, loss: 0.0025581358931958675 2023-01-21 12:19:42.503353: step: 1372/526, loss: 0.0125579833984375 2023-01-21 12:19:43.651667: step: 1376/526, loss: 0.015253543853759766 2023-01-21 12:19:44.795888: step: 1380/526, loss: 0.00168437953107059 2023-01-21 12:19:45.933248: step: 1384/526, loss: 0.0017072678310796618 2023-01-21 12:19:47.105495: step: 1388/526, loss: 6.351470801746473e-05 2023-01-21 12:19:48.230908: step: 1392/526, loss: 0.0006200790521688759 2023-01-21 12:19:49.368408: step: 1396/526, loss: 0.011794853955507278 2023-01-21 12:19:50.530996: step: 1400/526, loss: 8.430481830146164e-05 2023-01-21 12:19:51.666324: step: 1404/526, loss: 0.03389101102948189 2023-01-21 12:19:52.796454: step: 1408/526, loss: 3.814697265625e-06 2023-01-21 12:19:53.929225: step: 1412/526, loss: 0.0016368867363780737 2023-01-21 12:19:55.051777: step: 1416/526, loss: 0.007273769471794367 2023-01-21 12:19:56.172558: step: 1420/526, loss: 0.002724885940551758 2023-01-21 12:19:57.339390: step: 1424/526, loss: 7.877350435592234e-05 2023-01-21 12:19:58.508809: step: 1428/526, loss: 0.000632858311291784 2023-01-21 12:19:59.624672: step: 1432/526, loss: 0.018408775329589844 2023-01-21 12:20:00.753850: step: 1436/526, loss: 0.050493624061346054 2023-01-21 12:20:01.908010: step: 1440/526, loss: 0.021533681079745293 2023-01-21 12:20:03.060513: step: 1444/526, loss: 0.05006542056798935 2023-01-21 12:20:04.228572: step: 1448/526, loss: 0.00021409988403320312 2023-01-21 12:20:05.388350: step: 1452/526, loss: 0.05549333244562149 2023-01-21 12:20:06.541409: step: 1456/526, loss: 0.43433818221092224 2023-01-21 12:20:07.679816: step: 1460/526, loss: 8.56399565236643e-05 2023-01-21 12:20:08.816203: step: 1464/526, loss: 0.007652473170310259 2023-01-21 12:20:09.978522: step: 1468/526, loss: 0.039629556238651276 2023-01-21 12:20:11.119775: step: 1472/526, loss: 0.000604820204898715 2023-01-21 12:20:12.247238: step: 1476/526, loss: 0.006085109896957874 2023-01-21 12:20:13.370192: step: 1480/526, loss: 0.0056047444231808186 2023-01-21 12:20:14.496637: step: 1484/526, loss: 0.057358741760253906 2023-01-21 12:20:15.640331: step: 1488/526, loss: 0.027588654309511185 2023-01-21 12:20:16.769853: step: 1492/526, loss: 0.40248459577560425 2023-01-21 12:20:17.900507: step: 1496/526, loss: 0.009130382910370827 2023-01-21 12:20:19.041388: step: 1500/526, loss: 0.004675865173339844 2023-01-21 12:20:20.184674: step: 1504/526, loss: 0.00014390944852493703 2023-01-21 12:20:21.328688: step: 1508/526, loss: 0.00026857852935791016 2023-01-21 12:20:22.457239: step: 1512/526, loss: 0.007316971197724342 2023-01-21 12:20:23.581248: step: 1516/526, loss: 0.0026476620696485043 2023-01-21 12:20:24.721893: step: 1520/526, loss: 0.019067764282226562 2023-01-21 12:20:25.893116: step: 1524/526, loss: 0.00047588348388671875 2023-01-21 12:20:27.047642: step: 1528/526, loss: 0.0011170387733727694 2023-01-21 12:20:28.216373: step: 1532/526, loss: 0.004993247799575329 2023-01-21 12:20:29.350008: step: 1536/526, loss: 0.00129871373064816 2023-01-21 12:20:30.466157: step: 1540/526, loss: 0.03675823286175728 2023-01-21 12:20:31.596194: step: 1544/526, loss: 1.2269428968429565 2023-01-21 12:20:32.752820: step: 1548/526, loss: 0.009882926940917969 2023-01-21 12:20:33.882503: step: 1552/526, loss: 0.0015347481239587069 2023-01-21 12:20:35.015665: step: 1556/526, loss: 0.007928013801574707 2023-01-21 12:20:36.151557: step: 1560/526, loss: 0.04843468591570854 2023-01-21 12:20:37.299441: step: 1564/526, loss: 0.018177462741732597 2023-01-21 12:20:38.460118: step: 1568/526, loss: 0.00067729945294559 2023-01-21 12:20:39.595768: step: 1572/526, loss: 0.0002335071621928364 2023-01-21 12:20:40.759574: step: 1576/526, loss: 0.00524482736364007 2023-01-21 12:20:41.908558: step: 1580/526, loss: 0.00015735626220703125 2023-01-21 12:20:43.089418: step: 1584/526, loss: 0.0009131431579589844 2023-01-21 12:20:44.221620: step: 1588/526, loss: 0.0005405425908975303 2023-01-21 12:20:45.326001: step: 1592/526, loss: 0.00019330979557707906 2023-01-21 12:20:46.499386: step: 1596/526, loss: 0.03341560438275337 2023-01-21 12:20:47.647187: step: 1600/526, loss: 0.0042653209529817104 2023-01-21 12:20:48.786462: step: 1604/526, loss: 0.0010721206199377775 2023-01-21 12:20:49.906067: step: 1608/526, loss: 0.054396819323301315 2023-01-21 12:20:51.054214: step: 1612/526, loss: 0.004126167390495539 2023-01-21 12:20:52.212632: step: 1616/526, loss: 0.002502918243408203 2023-01-21 12:20:53.345704: step: 1620/526, loss: 1.6307831174344756e-05 2023-01-21 12:20:54.478500: step: 1624/526, loss: 0.0012418747646734118 2023-01-21 12:20:55.603546: step: 1628/526, loss: 0.0021982192993164062 2023-01-21 12:20:56.707296: step: 1632/526, loss: 0.002097225282341242 2023-01-21 12:20:57.835983: step: 1636/526, loss: 0.0384831428527832 2023-01-21 12:20:58.993124: step: 1640/526, loss: 0.008099555969238281 2023-01-21 12:21:00.161211: step: 1644/526, loss: 0.007973861880600452 2023-01-21 12:21:01.285312: step: 1648/526, loss: 0.00038051605224609375 2023-01-21 12:21:02.426793: step: 1652/526, loss: 0.009347342886030674 2023-01-21 12:21:03.572274: step: 1656/526, loss: 0.0318608283996582 2023-01-21 12:21:04.704080: step: 1660/526, loss: 0.07543551921844482 2023-01-21 12:21:05.833550: step: 1664/526, loss: 0.001445007394067943 2023-01-21 12:21:06.966549: step: 1668/526, loss: 0.003109932178631425 2023-01-21 12:21:08.081205: step: 1672/526, loss: 0.02376575395464897 2023-01-21 12:21:09.216130: step: 1676/526, loss: 0.0015651227440685034 2023-01-21 12:21:10.327608: step: 1680/526, loss: 0.010297202505171299 2023-01-21 12:21:11.432394: step: 1684/526, loss: 0.00041303635225631297 2023-01-21 12:21:12.570156: step: 1688/526, loss: 0.002592372940853238 2023-01-21 12:21:13.679340: step: 1692/526, loss: 0.0707603469491005 2023-01-21 12:21:14.807900: step: 1696/526, loss: 0.009849262423813343 2023-01-21 12:21:15.947185: step: 1700/526, loss: 0.0008469581953249872 2023-01-21 12:21:17.093413: step: 1704/526, loss: 0.0008864402770996094 2023-01-21 12:21:18.236697: step: 1708/526, loss: 0.001918744994327426 2023-01-21 12:21:19.362554: step: 1712/526, loss: 0.026782894507050514 2023-01-21 12:21:20.521959: step: 1716/526, loss: 2.059936559817288e-05 2023-01-21 12:21:21.659306: step: 1720/526, loss: 0.022302914410829544 2023-01-21 12:21:22.815749: step: 1724/526, loss: 0.00028324127197265625 2023-01-21 12:21:23.944948: step: 1728/526, loss: 0.0019309998024255037 2023-01-21 12:21:25.088619: step: 1732/526, loss: 4.57763690064894e-06 2023-01-21 12:21:26.238955: step: 1736/526, loss: 0.004938888363540173 2023-01-21 12:21:27.386986: step: 1740/526, loss: 0.012362862005829811 2023-01-21 12:21:28.514346: step: 1744/526, loss: 0.00020093917555641383 2023-01-21 12:21:29.666733: step: 1748/526, loss: 0.0027158737648278475 2023-01-21 12:21:30.830591: step: 1752/526, loss: 0.01735553704202175 2023-01-21 12:21:31.955621: step: 1756/526, loss: 0.006782913580536842 2023-01-21 12:21:33.082961: step: 1760/526, loss: 0.001158905099146068 2023-01-21 12:21:34.215054: step: 1764/526, loss: 0.014737891964614391 2023-01-21 12:21:35.339394: step: 1768/526, loss: 0.001072788261808455 2023-01-21 12:21:36.493316: step: 1772/526, loss: 0.031169412657618523 2023-01-21 12:21:37.613067: step: 1776/526, loss: 0.03193054348230362 2023-01-21 12:21:38.761177: step: 1780/526, loss: 0.002986431121826172 2023-01-21 12:21:39.872740: step: 1784/526, loss: 0.00018558502779342234 2023-01-21 12:21:41.041941: step: 1788/526, loss: 0.0008165836334228516 2023-01-21 12:21:42.200569: step: 1792/526, loss: 0.0003337860107421875 2023-01-21 12:21:43.377897: step: 1796/526, loss: 0.0016701698768883944 2023-01-21 12:21:44.521516: step: 1800/526, loss: 0.42012128233909607 2023-01-21 12:21:45.636010: step: 1804/526, loss: 0.0033001780975610018 2023-01-21 12:21:46.769806: step: 1808/526, loss: 0.003208064939826727 2023-01-21 12:21:47.917646: step: 1812/526, loss: 0.0028989792335778475 2023-01-21 12:21:49.059528: step: 1816/526, loss: 0.03765994310379028 2023-01-21 12:21:50.227086: step: 1820/526, loss: 0.0039904117584228516 2023-01-21 12:21:51.379269: step: 1824/526, loss: 0.00241680140607059 2023-01-21 12:21:52.501556: step: 1828/526, loss: 0.0024551867973059416 2023-01-21 12:21:53.662867: step: 1832/526, loss: 0.0008853912586346269 2023-01-21 12:21:54.776685: step: 1836/526, loss: 0.010827924124896526 2023-01-21 12:21:55.896512: step: 1840/526, loss: 0.16075249016284943 2023-01-21 12:21:57.033344: step: 1844/526, loss: 0.09786505252122879 2023-01-21 12:21:58.149376: step: 1848/526, loss: 0.0006225586403161287 2023-01-21 12:21:59.289993: step: 1852/526, loss: 0.010860443115234375 2023-01-21 12:22:00.416898: step: 1856/526, loss: 0.0007551193702965975 2023-01-21 12:22:01.573816: step: 1860/526, loss: 0.02900247648358345 2023-01-21 12:22:02.726994: step: 1864/526, loss: 0.005084800999611616 2023-01-21 12:22:03.885485: step: 1868/526, loss: 0.0016214370261877775 2023-01-21 12:22:05.069715: step: 1872/526, loss: 0.05117174983024597 2023-01-21 12:22:06.308281: step: 1876/526, loss: 0.006907367613166571 2023-01-21 12:22:07.478461: step: 1880/526, loss: 0.0006016730912961066 2023-01-21 12:22:08.617277: step: 1884/526, loss: 0.0024353028275072575 2023-01-21 12:22:09.737754: step: 1888/526, loss: 0.024439144879579544 2023-01-21 12:22:10.862579: step: 1892/526, loss: 0.018228817731142044 2023-01-21 12:22:12.024155: step: 1896/526, loss: 0.04074735566973686 2023-01-21 12:22:13.193046: step: 1900/526, loss: 0.01277256105095148 2023-01-21 12:22:14.323912: step: 1904/526, loss: 0.05594091862440109 2023-01-21 12:22:15.444234: step: 1908/526, loss: 0.02311687543988228 2023-01-21 12:22:16.557462: step: 1912/526, loss: 0.06925544887781143 2023-01-21 12:22:17.697250: step: 1916/526, loss: 0.0009435653919354081 2023-01-21 12:22:18.822359: step: 1920/526, loss: 0.00385456089861691 2023-01-21 12:22:19.967810: step: 1924/526, loss: 0.031569674611091614 2023-01-21 12:22:21.085412: step: 1928/526, loss: 0.005666256416589022 2023-01-21 12:22:22.217194: step: 1932/526, loss: 0.0007661819690838456 2023-01-21 12:22:23.350515: step: 1936/526, loss: 0.02087726630270481 2023-01-21 12:22:24.512193: step: 1940/526, loss: 0.022549821063876152 2023-01-21 12:22:25.671241: step: 1944/526, loss: 0.0001398563472321257 2023-01-21 12:22:26.813259: step: 1948/526, loss: 0.037169814109802246 2023-01-21 12:22:27.952361: step: 1952/526, loss: 0.0021746635902673006 2023-01-21 12:22:29.084319: step: 1956/526, loss: 0.00014352798461914062 2023-01-21 12:22:30.213025: step: 1960/526, loss: 0.003190517658367753 2023-01-21 12:22:31.387865: step: 1964/526, loss: 0.0036753653548657894 2023-01-21 12:22:32.529571: step: 1968/526, loss: 2.555847095209174e-05 2023-01-21 12:22:33.643172: step: 1972/526, loss: 0.012353808619081974 2023-01-21 12:22:34.754667: step: 1976/526, loss: 0.0004582405090332031 2023-01-21 12:22:35.876967: step: 1980/526, loss: 2.212524486822076e-05 2023-01-21 12:22:37.018001: step: 1984/526, loss: 0.0007988453144207597 2023-01-21 12:22:38.160731: step: 1988/526, loss: 0.017854977399110794 2023-01-21 12:22:39.325499: step: 1992/526, loss: 0.0005401134840212762 2023-01-21 12:22:40.447159: step: 1996/526, loss: 0.0009823114378377795 2023-01-21 12:22:41.559912: step: 2000/526, loss: 0.0006556510925292969 2023-01-21 12:22:42.698941: step: 2004/526, loss: 0.05289506912231445 2023-01-21 12:22:43.870580: step: 2008/526, loss: 0.05361061170697212 2023-01-21 12:22:44.999168: step: 2012/526, loss: 8.687972876941785e-05 2023-01-21 12:22:46.108869: step: 2016/526, loss: 0.00011148452904308215 2023-01-21 12:22:47.292133: step: 2020/526, loss: 0.047094155102968216 2023-01-21 12:22:48.415022: step: 2024/526, loss: 0.00011892318434547633 2023-01-21 12:22:49.552959: step: 2028/526, loss: 0.022041702643036842 2023-01-21 12:22:50.716594: step: 2032/526, loss: 0.0018604277865961194 2023-01-21 12:22:51.871329: step: 2036/526, loss: 0.013186503201723099 2023-01-21 12:22:53.016536: step: 2040/526, loss: 0.0009047508356161416 2023-01-21 12:22:54.141184: step: 2044/526, loss: 0.005512333009392023 2023-01-21 12:22:55.270647: step: 2048/526, loss: 0.002853298094123602 2023-01-21 12:22:56.421787: step: 2052/526, loss: 0.014701366424560547 2023-01-21 12:22:57.560409: step: 2056/526, loss: 0.04941254109144211 2023-01-21 12:22:58.677458: step: 2060/526, loss: 0.0003387451288290322 2023-01-21 12:22:59.804383: step: 2064/526, loss: 0.00037937163142487407 2023-01-21 12:23:00.946579: step: 2068/526, loss: 0.0008299828041344881 2023-01-21 12:23:02.112527: step: 2072/526, loss: 0.020668793469667435 2023-01-21 12:23:03.273085: step: 2076/526, loss: 0.008508491329848766 2023-01-21 12:23:04.409665: step: 2080/526, loss: 0.0931495651602745 2023-01-21 12:23:05.546557: step: 2084/526, loss: 0.011777115054428577 2023-01-21 12:23:06.718403: step: 2088/526, loss: 0.012471389025449753 2023-01-21 12:23:07.855946: step: 2092/526, loss: 0.0012497901916503906 2023-01-21 12:23:08.985816: step: 2096/526, loss: 0.00043773651123046875 2023-01-21 12:23:10.124310: step: 2100/526, loss: 0.001194095704704523 2023-01-21 12:23:11.266665: step: 2104/526, loss: 9.72747802734375e-05 ================================================== Loss: 0.023 -------------------- Dev: {'event': {'p': 0.5604914933837429, 'r': 0.7896138482023968, 'f1': 0.6556108347153123}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Test: {'event': {'p': 0.5925349922239502, 'r': 0.7954070981210856, 'f1': 0.679144385026738}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Chinese: {'event': {'p': 0.49019607843137253, 'r': 0.9259259259259259, 'f1': 0.6410256410256411}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Korean: {'event': {'p': 0.5873015873015873, 'r': 0.5873015873015873, 'f1': 0.5873015873015873}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} Russian: {'event': {'p': 0.4583333333333333, 'r': 0.6111111111111112, 'f1': 0.5238095238095238}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 20} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 21 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:23:55.269595: step: 4/526, loss: 0.0009955406421795487 2023-01-21 12:23:56.405114: step: 8/526, loss: 0.032643888145685196 2023-01-21 12:23:57.527799: step: 12/526, loss: 6.0749058320652694e-05 2023-01-21 12:23:58.650099: step: 16/526, loss: 0.06737546622753143 2023-01-21 12:23:59.803595: step: 20/526, loss: 0.0030429840553551912 2023-01-21 12:24:00.951220: step: 24/526, loss: 0.03221569210290909 2023-01-21 12:24:02.070485: step: 28/526, loss: 0.006129169370979071 2023-01-21 12:24:03.227962: step: 32/526, loss: 0.0014341354835778475 2023-01-21 12:24:04.323738: step: 36/526, loss: 0.00033102036104537547 2023-01-21 12:24:05.456260: step: 40/526, loss: 0.007520007900893688 2023-01-21 12:24:06.607266: step: 44/526, loss: 0.004151535220444202 2023-01-21 12:24:07.739522: step: 48/526, loss: 0.027634240686893463 2023-01-21 12:24:08.888984: step: 52/526, loss: 0.014844131655991077 2023-01-21 12:24:10.026962: step: 56/526, loss: 0.012241172604262829 2023-01-21 12:24:11.166154: step: 60/526, loss: 0.0017638206481933594 2023-01-21 12:24:12.316449: step: 64/526, loss: 0.11033602058887482 2023-01-21 12:24:13.458206: step: 68/526, loss: 0.014161776751279831 2023-01-21 12:24:14.599907: step: 72/526, loss: 0.00036020277184434235 2023-01-21 12:24:15.732829: step: 76/526, loss: 0.002583503955975175 2023-01-21 12:24:16.883519: step: 80/526, loss: 0.00039119721623137593 2023-01-21 12:24:18.010615: step: 84/526, loss: 0.008385754190385342 2023-01-21 12:24:19.187037: step: 88/526, loss: 0.0198225025087595 2023-01-21 12:24:20.332544: step: 92/526, loss: 6.332397606456652e-05 2023-01-21 12:24:21.469778: step: 96/526, loss: 0.00029497145442292094 2023-01-21 12:24:22.608166: step: 100/526, loss: 0.009635734371840954 2023-01-21 12:24:23.739098: step: 104/526, loss: 0.0007391691324301064 2023-01-21 12:24:24.875933: step: 108/526, loss: 0.0006021499866619706 2023-01-21 12:24:26.021844: step: 112/526, loss: 0.00021438598923850805 2023-01-21 12:24:27.156212: step: 116/526, loss: 0.001714563462883234 2023-01-21 12:24:28.286873: step: 120/526, loss: 0.0015277386410161853 2023-01-21 12:24:29.414108: step: 124/526, loss: 0.009488677605986595 2023-01-21 12:24:30.546592: step: 128/526, loss: 0.00044264792813919485 2023-01-21 12:24:31.671862: step: 132/526, loss: 0.14051762223243713 2023-01-21 12:24:32.804378: step: 136/526, loss: 0.002780866576358676 2023-01-21 12:24:33.913974: step: 140/526, loss: 0.0004968166467733681 2023-01-21 12:24:35.075585: step: 144/526, loss: 0.048073720186948776 2023-01-21 12:24:36.228110: step: 148/526, loss: 0.005917549133300781 2023-01-21 12:24:37.336362: step: 152/526, loss: 0.004196167457848787 2023-01-21 12:24:38.464283: step: 156/526, loss: 0.011194133199751377 2023-01-21 12:24:39.614708: step: 160/526, loss: 0.007124233525246382 2023-01-21 12:24:40.749733: step: 164/526, loss: 0.0038658143021166325 2023-01-21 12:24:41.881376: step: 168/526, loss: 0.0017652511596679688 2023-01-21 12:24:43.022011: step: 172/526, loss: 0.0005502700805664062 2023-01-21 12:24:44.118144: step: 176/526, loss: 0.002340364269912243 2023-01-21 12:24:45.238748: step: 180/526, loss: 0.017264556139707565 2023-01-21 12:24:46.402896: step: 184/526, loss: 0.021281886845827103 2023-01-21 12:24:47.558457: step: 188/526, loss: 0.021313440054655075 2023-01-21 12:24:48.707506: step: 192/526, loss: 1.506805438111769e-05 2023-01-21 12:24:49.858015: step: 196/526, loss: 0.0015953064430505037 2023-01-21 12:24:51.006371: step: 200/526, loss: 0.019234657287597656 2023-01-21 12:24:52.123847: step: 204/526, loss: 0.00336456298828125 2023-01-21 12:24:53.287559: step: 208/526, loss: 0.020078469067811966 2023-01-21 12:24:54.420618: step: 212/526, loss: 0.028565121814608574 2023-01-21 12:24:55.553949: step: 216/526, loss: 0.0008527755853720009 2023-01-21 12:24:56.712766: step: 220/526, loss: 2.8753278456861153e-05 2023-01-21 12:24:57.867835: step: 224/526, loss: 0.006253433413803577 2023-01-21 12:24:58.981827: step: 228/526, loss: 7.905960228526965e-05 2023-01-21 12:25:00.113662: step: 232/526, loss: 0.04281196743249893 2023-01-21 12:25:01.237470: step: 236/526, loss: 0.00027484894962981343 2023-01-21 12:25:02.418459: step: 240/526, loss: 0.0028964995872229338 2023-01-21 12:25:03.560671: step: 244/526, loss: 0.019318198785185814 2023-01-21 12:25:04.693927: step: 248/526, loss: 5.044937279308215e-05 2023-01-21 12:25:05.827540: step: 252/526, loss: 0.006527137476950884 2023-01-21 12:25:07.013297: step: 256/526, loss: 0.006454754155129194 2023-01-21 12:25:08.126070: step: 260/526, loss: 0.03245840221643448 2023-01-21 12:25:09.286486: step: 264/526, loss: 0.006314659025520086 2023-01-21 12:25:10.407488: step: 268/526, loss: 0.01365203782916069 2023-01-21 12:25:11.543766: step: 272/526, loss: 0.003501224797219038 2023-01-21 12:25:12.660521: step: 276/526, loss: 0.0031845092307776213 2023-01-21 12:25:13.825618: step: 280/526, loss: 0.00025005341740325093 2023-01-21 12:25:14.943744: step: 284/526, loss: 0.0012862683506682515 2023-01-21 12:25:16.068568: step: 288/526, loss: 0.0013446807861328125 2023-01-21 12:25:17.218835: step: 292/526, loss: 0.04877147823572159 2023-01-21 12:25:18.363305: step: 296/526, loss: 0.012141609564423561 2023-01-21 12:25:19.503998: step: 300/526, loss: 0.006172561552375555 2023-01-21 12:25:20.626742: step: 304/526, loss: 0.03891048580408096 2023-01-21 12:25:21.729517: step: 308/526, loss: 0.0011526108719408512 2023-01-21 12:25:22.859204: step: 312/526, loss: 0.010182762518525124 2023-01-21 12:25:23.978506: step: 316/526, loss: 5.367398262023926e-05 2023-01-21 12:25:25.118697: step: 320/526, loss: 0.019713783636689186 2023-01-21 12:25:26.266265: step: 324/526, loss: 0.03429565578699112 2023-01-21 12:25:27.402704: step: 328/526, loss: 0.0009423256269656122 2023-01-21 12:25:28.539878: step: 332/526, loss: 9.117126319324598e-05 2023-01-21 12:25:29.680348: step: 336/526, loss: 2.5272369384765625e-05 2023-01-21 12:25:30.824264: step: 340/526, loss: 0.03172626718878746 2023-01-21 12:25:32.003017: step: 344/526, loss: 0.01759166829288006 2023-01-21 12:25:33.114905: step: 348/526, loss: 0.003347778460010886 2023-01-21 12:25:34.266414: step: 352/526, loss: 0.0017461776733398438 2023-01-21 12:25:35.402930: step: 356/526, loss: 0.03577089309692383 2023-01-21 12:25:36.525324: step: 360/526, loss: 0.016684724017977715 2023-01-21 12:25:37.669925: step: 364/526, loss: 0.015204716473817825 2023-01-21 12:25:38.800726: step: 368/526, loss: 0.000576877617277205 2023-01-21 12:25:39.930365: step: 372/526, loss: 7.643699791515246e-05 2023-01-21 12:25:41.055049: step: 376/526, loss: 0.18345527350902557 2023-01-21 12:25:42.191059: step: 380/526, loss: 0.0004954338073730469 2023-01-21 12:25:43.352051: step: 384/526, loss: 9.474754187976941e-05 2023-01-21 12:25:44.502541: step: 388/526, loss: 0.0012785911094397306 2023-01-21 12:25:45.638925: step: 392/526, loss: 0.00176830287091434 2023-01-21 12:25:46.780099: step: 396/526, loss: 0.04534664377570152 2023-01-21 12:25:47.906069: step: 400/526, loss: 0.000691795430611819 2023-01-21 12:25:49.033216: step: 404/526, loss: 0.007149696350097656 2023-01-21 12:25:50.170287: step: 408/526, loss: 0.00011005401756847277 2023-01-21 12:25:51.300148: step: 412/526, loss: 9.784698340808973e-05 2023-01-21 12:25:52.426603: step: 416/526, loss: 0.0015617371536791325 2023-01-21 12:25:53.560022: step: 420/526, loss: 0.002857112791389227 2023-01-21 12:25:54.699720: step: 424/526, loss: 0.00821542739868164 2023-01-21 12:25:55.841409: step: 428/526, loss: 0.018573950976133347 2023-01-21 12:25:57.013646: step: 432/526, loss: 0.0013235092628747225 2023-01-21 12:25:58.172840: step: 436/526, loss: 0.016424084082245827 2023-01-21 12:25:59.300019: step: 440/526, loss: 0.029625415802001953 2023-01-21 12:26:00.421687: step: 444/526, loss: 0.0004285812610760331 2023-01-21 12:26:01.563094: step: 448/526, loss: 0.005054282955825329 2023-01-21 12:26:02.701952: step: 452/526, loss: 5.7220458984375e-06 2023-01-21 12:26:03.824391: step: 456/526, loss: 0.017589950934052467 2023-01-21 12:26:04.957848: step: 460/526, loss: 0.0017057419754564762 2023-01-21 12:26:06.076057: step: 464/526, loss: 0.0012001991271972656 2023-01-21 12:26:07.250602: step: 468/526, loss: 0.010443019680678844 2023-01-21 12:26:08.395603: step: 472/526, loss: 0.00037508009700104594 2023-01-21 12:26:09.505391: step: 476/526, loss: 0.02084674872457981 2023-01-21 12:26:10.651134: step: 480/526, loss: 0.0100480318069458 2023-01-21 12:26:11.760887: step: 484/526, loss: 0.005688333418220282 2023-01-21 12:26:12.906600: step: 488/526, loss: 0.0008464812999591231 2023-01-21 12:26:14.037685: step: 492/526, loss: 0.0024870873894542456 2023-01-21 12:26:15.160571: step: 496/526, loss: 0.03071155585348606 2023-01-21 12:26:16.309216: step: 500/526, loss: 0.00010585784912109375 2023-01-21 12:26:17.457474: step: 504/526, loss: 0.03797359764575958 2023-01-21 12:26:18.611285: step: 508/526, loss: 0.00014562607975676656 2023-01-21 12:26:19.737322: step: 512/526, loss: 0.04983320087194443 2023-01-21 12:26:20.859423: step: 516/526, loss: 0.0011019706726074219 2023-01-21 12:26:21.970809: step: 520/526, loss: 1.0967256685034954e-06 2023-01-21 12:26:23.131877: step: 524/526, loss: 0.009551430121064186 2023-01-21 12:26:24.297822: step: 528/526, loss: 0.004874038510024548 2023-01-21 12:26:25.449230: step: 532/526, loss: 0.00013256072998046875 2023-01-21 12:26:26.583907: step: 536/526, loss: 0.008013343438506126 2023-01-21 12:26:27.725494: step: 540/526, loss: 0.0006774902576580644 2023-01-21 12:26:28.896221: step: 544/526, loss: 0.00036277773324400187 2023-01-21 12:26:30.044614: step: 548/526, loss: 0.014105415903031826 2023-01-21 12:26:31.211815: step: 552/526, loss: 0.001499176025390625 2023-01-21 12:26:32.343674: step: 556/526, loss: 0.01841278187930584 2023-01-21 12:26:33.491256: step: 560/526, loss: 0.012327957898378372 2023-01-21 12:26:34.636673: step: 564/526, loss: 0.0008825302356854081 2023-01-21 12:26:35.778956: step: 568/526, loss: 0.00016288759070448577 2023-01-21 12:26:36.937089: step: 572/526, loss: 0.0002874374622479081 2023-01-21 12:26:38.088478: step: 576/526, loss: 0.0012821197742596269 2023-01-21 12:26:39.204452: step: 580/526, loss: 0.00027785301790572703 2023-01-21 12:26:40.333755: step: 584/526, loss: 0.008230590261518955 2023-01-21 12:26:41.475814: step: 588/526, loss: 0.0018922805320471525 2023-01-21 12:26:42.626882: step: 592/526, loss: 0.0015671729343011975 2023-01-21 12:26:43.757573: step: 596/526, loss: 0.004811477847397327 2023-01-21 12:26:44.898038: step: 600/526, loss: 0.0003749847528524697 2023-01-21 12:26:46.053067: step: 604/526, loss: 0.017642594873905182 2023-01-21 12:26:47.166030: step: 608/526, loss: 0.0011973382206633687 2023-01-21 12:26:48.320266: step: 612/526, loss: 0.0010057450272142887 2023-01-21 12:26:49.471608: step: 616/526, loss: 0.019736479967832565 2023-01-21 12:26:50.596573: step: 620/526, loss: 0.00016841889009810984 2023-01-21 12:26:51.775301: step: 624/526, loss: 0.0011733531719073653 2023-01-21 12:26:52.914097: step: 628/526, loss: 0.01329965703189373 2023-01-21 12:26:54.046052: step: 632/526, loss: 0.020113373175263405 2023-01-21 12:26:55.152504: step: 636/526, loss: 0.0003427505725994706 2023-01-21 12:26:56.253732: step: 640/526, loss: 2.09808349609375e-05 2023-01-21 12:26:57.399370: step: 644/526, loss: 0.02185802534222603 2023-01-21 12:26:58.547187: step: 648/526, loss: 0.008469867520034313 2023-01-21 12:26:59.708480: step: 652/526, loss: 0.07594814151525497 2023-01-21 12:27:00.830910: step: 656/526, loss: 0.00014095305232331157 2023-01-21 12:27:01.974143: step: 660/526, loss: 0.02503986284136772 2023-01-21 12:27:03.086994: step: 664/526, loss: 0.0002735138114076108 2023-01-21 12:27:04.226564: step: 668/526, loss: 0.05170421674847603 2023-01-21 12:27:05.357509: step: 672/526, loss: 0.001561927841976285 2023-01-21 12:27:06.493305: step: 676/526, loss: 0.03464946523308754 2023-01-21 12:27:07.647093: step: 680/526, loss: 0.002483940217643976 2023-01-21 12:27:08.756026: step: 684/526, loss: 0.0005573272937908769 2023-01-21 12:27:09.898379: step: 688/526, loss: 0.011051654815673828 2023-01-21 12:27:11.083226: step: 692/526, loss: 0.024146461859345436 2023-01-21 12:27:12.217639: step: 696/526, loss: 6.67572021484375e-06 2023-01-21 12:27:13.381428: step: 700/526, loss: 0.0608489029109478 2023-01-21 12:27:14.525574: step: 704/526, loss: 0.0009290695306845009 2023-01-21 12:27:15.670114: step: 708/526, loss: 0.00016860962205100805 2023-01-21 12:27:16.796916: step: 712/526, loss: 2.765655517578125e-05 2023-01-21 12:27:17.919441: step: 716/526, loss: 0.008852767758071423 2023-01-21 12:27:19.038277: step: 720/526, loss: 0.0005374908214434981 2023-01-21 12:27:20.199698: step: 724/526, loss: 0.00016546249389648438 2023-01-21 12:27:21.322516: step: 728/526, loss: 0.003566169645637274 2023-01-21 12:27:22.454813: step: 732/526, loss: 0.049204062670469284 2023-01-21 12:27:23.617762: step: 736/526, loss: 0.007140922360122204 2023-01-21 12:27:24.732317: step: 740/526, loss: 0.00022611618624068797 2023-01-21 12:27:25.879636: step: 744/526, loss: 0.0006732940673828125 2023-01-21 12:27:27.058315: step: 748/526, loss: 4.138946678722277e-05 2023-01-21 12:27:28.194587: step: 752/526, loss: 0.005099296569824219 2023-01-21 12:27:29.363092: step: 756/526, loss: 0.006854534149169922 2023-01-21 12:27:30.489674: step: 760/526, loss: 2.0980833141948096e-06 2023-01-21 12:27:31.630633: step: 764/526, loss: 0.012611293233931065 2023-01-21 12:27:32.789908: step: 768/526, loss: 0.004787635989487171 2023-01-21 12:27:33.932404: step: 772/526, loss: 0.3584173321723938 2023-01-21 12:27:35.085325: step: 776/526, loss: 0.003460693173110485 2023-01-21 12:27:36.194747: step: 780/526, loss: 0.003250503446906805 2023-01-21 12:27:37.356013: step: 784/526, loss: 0.005113798193633556 2023-01-21 12:27:38.472599: step: 788/526, loss: 0.0018720626831054688 2023-01-21 12:27:39.609895: step: 792/526, loss: 0.0089095588773489 2023-01-21 12:27:40.739795: step: 796/526, loss: 0.05566377937793732 2023-01-21 12:27:41.932101: step: 800/526, loss: 4.38690176451928e-06 2023-01-21 12:27:43.097342: step: 804/526, loss: 0.02503671869635582 2023-01-21 12:27:44.221642: step: 808/526, loss: 8.301809430122375e-05 2023-01-21 12:27:45.372872: step: 812/526, loss: 0.03979034349322319 2023-01-21 12:27:46.498417: step: 816/526, loss: 0.00033016205998137593 2023-01-21 12:27:47.650488: step: 820/526, loss: 0.020508289337158203 2023-01-21 12:27:48.797847: step: 824/526, loss: 0.0008488655439577997 2023-01-21 12:27:49.925422: step: 828/526, loss: 0.0023378373589366674 2023-01-21 12:27:51.076032: step: 832/526, loss: 0.0005554199451580644 2023-01-21 12:27:52.215244: step: 836/526, loss: 0.04146156460046768 2023-01-21 12:27:53.363974: step: 840/526, loss: 0.008087158203125 2023-01-21 12:27:54.482358: step: 844/526, loss: 0.026282453909516335 2023-01-21 12:27:55.646651: step: 848/526, loss: 0.00015186071686912328 2023-01-21 12:27:56.800358: step: 852/526, loss: 0.012221241369843483 2023-01-21 12:27:57.939041: step: 856/526, loss: 0.016178417950868607 2023-01-21 12:27:59.115971: step: 860/526, loss: 0.18164634704589844 2023-01-21 12:28:00.230651: step: 864/526, loss: 0.03341054916381836 2023-01-21 12:28:01.362790: step: 868/526, loss: 0.01939220540225506 2023-01-21 12:28:02.506321: step: 872/526, loss: 0.018245983868837357 2023-01-21 12:28:03.633015: step: 876/526, loss: 0.014223290607333183 2023-01-21 12:28:04.806330: step: 880/526, loss: 0.010126161389052868 2023-01-21 12:28:05.945480: step: 884/526, loss: 0.0027590752579271793 2023-01-21 12:28:07.064590: step: 888/526, loss: 0.0005181313026696444 2023-01-21 12:28:08.229549: step: 892/526, loss: 0.0002513885556254536 2023-01-21 12:28:09.401705: step: 896/526, loss: 0.009302664548158646 2023-01-21 12:28:10.540102: step: 900/526, loss: 0.038389064371585846 2023-01-21 12:28:11.691625: step: 904/526, loss: 0.00014157296391204 2023-01-21 12:28:12.853074: step: 908/526, loss: 0.006889533717185259 2023-01-21 12:28:14.024835: step: 912/526, loss: 0.0006856918334960938 2023-01-21 12:28:15.149388: step: 916/526, loss: 2.346038854739163e-05 2023-01-21 12:28:16.294724: step: 920/526, loss: 0.06824145466089249 2023-01-21 12:28:17.416582: step: 924/526, loss: 0.00043606761028058827 2023-01-21 12:28:18.551224: step: 928/526, loss: 0.012229920364916325 2023-01-21 12:28:19.669771: step: 932/526, loss: 0.021875381469726562 2023-01-21 12:28:20.803820: step: 936/526, loss: 0.007502079010009766 2023-01-21 12:28:21.916484: step: 940/526, loss: -1.4019013178767636e-05 2023-01-21 12:28:23.075330: step: 944/526, loss: 0.0029477120842784643 2023-01-21 12:28:24.217330: step: 948/526, loss: 0.08689117431640625 2023-01-21 12:28:25.326788: step: 952/526, loss: 0.015105342492461205 2023-01-21 12:28:26.450720: step: 956/526, loss: 0.00018024444580078125 2023-01-21 12:28:27.558666: step: 960/526, loss: 0.07650360465049744 2023-01-21 12:28:28.685100: step: 964/526, loss: 0.0001720428408589214 2023-01-21 12:28:29.831915: step: 968/526, loss: 0.0004440307675395161 2023-01-21 12:28:30.966948: step: 972/526, loss: 0.02216949500143528 2023-01-21 12:28:32.108618: step: 976/526, loss: 0.004005050752311945 2023-01-21 12:28:33.240167: step: 980/526, loss: 0.0025488617829978466 2023-01-21 12:28:34.357943: step: 984/526, loss: 0.01591334491968155 2023-01-21 12:28:35.498358: step: 988/526, loss: 0.006067085079848766 2023-01-21 12:28:36.651321: step: 992/526, loss: 2.8991700673941523e-05 2023-01-21 12:28:37.777992: step: 996/526, loss: 0.00057392125017941 2023-01-21 12:28:38.906542: step: 1000/526, loss: 0.0010517120826989412 2023-01-21 12:28:40.050021: step: 1004/526, loss: 0.0011048555606976151 2023-01-21 12:28:41.175987: step: 1008/526, loss: 0.0008025169372558594 2023-01-21 12:28:42.316462: step: 1012/526, loss: 0.01602649874985218 2023-01-21 12:28:43.473093: step: 1016/526, loss: 0.010249900631606579 2023-01-21 12:28:44.588598: step: 1020/526, loss: 0.00043778421240858734 2023-01-21 12:28:45.745476: step: 1024/526, loss: 0.00016508101543877274 2023-01-21 12:28:46.896873: step: 1028/526, loss: 0.17733134329319 2023-01-21 12:28:48.044644: step: 1032/526, loss: 0.30967026948928833 2023-01-21 12:28:49.180093: step: 1036/526, loss: 0.01845226250588894 2023-01-21 12:28:50.332523: step: 1040/526, loss: 0.0006390571943484247 2023-01-21 12:28:51.487348: step: 1044/526, loss: 0.0012401581043377519 2023-01-21 12:28:52.659798: step: 1048/526, loss: 0.00530586251989007 2023-01-21 12:28:53.815119: step: 1052/526, loss: 0.00022907258244231343 2023-01-21 12:28:54.962099: step: 1056/526, loss: 7.658005051780492e-05 2023-01-21 12:28:56.103996: step: 1060/526, loss: 7.715225365245715e-05 2023-01-21 12:28:57.240232: step: 1064/526, loss: 0.00028133392333984375 2023-01-21 12:28:58.344498: step: 1068/526, loss: 0.0010894775623455644 2023-01-21 12:28:59.494695: step: 1072/526, loss: 0.4035331606864929 2023-01-21 12:29:00.642120: step: 1076/526, loss: 0.0030317308846861124 2023-01-21 12:29:01.747363: step: 1080/526, loss: 0.0018420219421386719 2023-01-21 12:29:02.910807: step: 1084/526, loss: 0.0005134582170285285 2023-01-21 12:29:04.052434: step: 1088/526, loss: 0.002246570773422718 2023-01-21 12:29:05.180084: step: 1092/526, loss: 0.0010133743053302169 2023-01-21 12:29:06.306638: step: 1096/526, loss: 0.004272079560905695 2023-01-21 12:29:07.465901: step: 1100/526, loss: 0.00017108916654251516 2023-01-21 12:29:08.609022: step: 1104/526, loss: 0.009885216131806374 2023-01-21 12:29:09.765961: step: 1108/526, loss: 0.050868988037109375 2023-01-21 12:29:10.893585: step: 1112/526, loss: 0.012102174572646618 2023-01-21 12:29:12.037396: step: 1116/526, loss: 0.00146827707067132 2023-01-21 12:29:13.173959: step: 1120/526, loss: 0.006963539402931929 2023-01-21 12:29:14.277549: step: 1124/526, loss: 0.00031266213045455515 2023-01-21 12:29:15.440569: step: 1128/526, loss: 0.00020513535127975047 2023-01-21 12:29:16.576001: step: 1132/526, loss: 0.0005605578771792352 2023-01-21 12:29:17.711944: step: 1136/526, loss: 0.010409070178866386 2023-01-21 12:29:18.838166: step: 1140/526, loss: 0.005105781834572554 2023-01-21 12:29:19.955211: step: 1144/526, loss: 0.00016026497178245336 2023-01-21 12:29:21.108592: step: 1148/526, loss: 1.7833708625403233e-05 2023-01-21 12:29:22.239285: step: 1152/526, loss: 0.0006553650018759072 2023-01-21 12:29:23.392425: step: 1156/526, loss: 0.013267803937196732 2023-01-21 12:29:24.526293: step: 1160/526, loss: 0.004382896237075329 2023-01-21 12:29:25.690591: step: 1164/526, loss: 0.08456211537122726 2023-01-21 12:29:26.798432: step: 1168/526, loss: 0.013112069107592106 2023-01-21 12:29:27.918837: step: 1172/526, loss: 0.019095564261078835 2023-01-21 12:29:29.041539: step: 1176/526, loss: 0.0008366585243493319 2023-01-21 12:29:30.159055: step: 1180/526, loss: 0.05835561826825142 2023-01-21 12:29:31.295319: step: 1184/526, loss: 0.017320729792118073 2023-01-21 12:29:32.437995: step: 1188/526, loss: 0.005169105716049671 2023-01-21 12:29:33.577855: step: 1192/526, loss: 0.0017440796364098787 2023-01-21 12:29:34.727074: step: 1196/526, loss: 0.2281607687473297 2023-01-21 12:29:35.859726: step: 1200/526, loss: 0.033351849764585495 2023-01-21 12:29:37.003757: step: 1204/526, loss: 0.007132434751838446 2023-01-21 12:29:38.148419: step: 1208/526, loss: 0.0007819175953045487 2023-01-21 12:29:39.262846: step: 1212/526, loss: 0.004265117924660444 2023-01-21 12:29:40.391234: step: 1216/526, loss: 0.08878135681152344 2023-01-21 12:29:41.590361: step: 1220/526, loss: 2.717971983656753e-05 2023-01-21 12:29:42.738439: step: 1224/526, loss: 0.00024590492830611765 2023-01-21 12:29:43.872213: step: 1228/526, loss: 0.036859702318906784 2023-01-21 12:29:44.978106: step: 1232/526, loss: 0.015033578500151634 2023-01-21 12:29:46.105535: step: 1236/526, loss: 0.02354106865823269 2023-01-21 12:29:47.264422: step: 1240/526, loss: 0.018633080646395683 2023-01-21 12:29:48.393212: step: 1244/526, loss: 0.0032699587754905224 2023-01-21 12:29:49.511826: step: 1248/526, loss: 0.014067268930375576 2023-01-21 12:29:50.642970: step: 1252/526, loss: 0.01876812055706978 2023-01-21 12:29:51.770088: step: 1256/526, loss: 0.0006593227153643966 2023-01-21 12:29:52.939102: step: 1260/526, loss: 0.005448865704238415 2023-01-21 12:29:54.064590: step: 1264/526, loss: 0.00014724732318427414 2023-01-21 12:29:55.199152: step: 1268/526, loss: 0.0012784005375579 2023-01-21 12:29:56.350666: step: 1272/526, loss: 0.12202782183885574 2023-01-21 12:29:57.474922: step: 1276/526, loss: 0.009560966864228249 2023-01-21 12:29:58.602455: step: 1280/526, loss: 0.010747050866484642 2023-01-21 12:29:59.733331: step: 1284/526, loss: 0.05109863355755806 2023-01-21 12:30:00.888309: step: 1288/526, loss: 0.01661095581948757 2023-01-21 12:30:02.011417: step: 1292/526, loss: 0.0007375717395916581 2023-01-21 12:30:03.172275: step: 1296/526, loss: 0.0022489549592137337 2023-01-21 12:30:04.289151: step: 1300/526, loss: 5.760193016612902e-05 2023-01-21 12:30:05.400110: step: 1304/526, loss: 0.018601704388856888 2023-01-21 12:30:06.547730: step: 1308/526, loss: 0.002262258669361472 2023-01-21 12:30:07.713415: step: 1312/526, loss: 0.006157303228974342 2023-01-21 12:30:08.846181: step: 1316/526, loss: 0.044877052307128906 2023-01-21 12:30:10.009264: step: 1320/526, loss: 0.020856190472841263 2023-01-21 12:30:11.134671: step: 1324/526, loss: 0.007565116975456476 2023-01-21 12:30:12.257341: step: 1328/526, loss: 0.00040340423583984375 2023-01-21 12:30:13.441711: step: 1332/526, loss: 0.011701202020049095 2023-01-21 12:30:14.570075: step: 1336/526, loss: 0.0007734298706054688 2023-01-21 12:30:15.712272: step: 1340/526, loss: 1.721382068353705e-05 2023-01-21 12:30:16.834339: step: 1344/526, loss: 0.05681648105382919 2023-01-21 12:30:17.997922: step: 1348/526, loss: 0.002610588213428855 2023-01-21 12:30:19.144087: step: 1352/526, loss: 0.002860641572624445 2023-01-21 12:30:20.275368: step: 1356/526, loss: 0.0002707481326069683 2023-01-21 12:30:21.384786: step: 1360/526, loss: 0.04086341708898544 2023-01-21 12:30:22.533065: step: 1364/526, loss: 0.12624701857566833 2023-01-21 12:30:23.674249: step: 1368/526, loss: 0.008828163146972656 2023-01-21 12:30:24.878479: step: 1372/526, loss: 0.0027866840828210115 2023-01-21 12:30:26.019621: step: 1376/526, loss: 0.0015918733552098274 2023-01-21 12:30:27.104942: step: 1380/526, loss: 0.003572464222088456 2023-01-21 12:30:28.252520: step: 1384/526, loss: 0.028323937207460403 2023-01-21 12:30:29.414183: step: 1388/526, loss: 0.005240440834313631 2023-01-21 12:30:30.557367: step: 1392/526, loss: 0.00039329531136900187 2023-01-21 12:30:31.696701: step: 1396/526, loss: 0.003461647080257535 2023-01-21 12:30:32.867454: step: 1400/526, loss: 0.01656322553753853 2023-01-21 12:30:33.998712: step: 1404/526, loss: 1.068115216185106e-05 2023-01-21 12:30:35.127680: step: 1408/526, loss: 0.0008235931163653731 2023-01-21 12:30:36.260621: step: 1412/526, loss: 0.019260406494140625 2023-01-21 12:30:37.397800: step: 1416/526, loss: 0.00432510394603014 2023-01-21 12:30:38.551370: step: 1420/526, loss: 0.016992948949337006 2023-01-21 12:30:39.693824: step: 1424/526, loss: 0.06199779361486435 2023-01-21 12:30:40.819550: step: 1428/526, loss: 0.0243072509765625 2023-01-21 12:30:41.974391: step: 1432/526, loss: 0.000606536865234375 2023-01-21 12:30:43.112078: step: 1436/526, loss: 0.0020130157936364412 2023-01-21 12:30:44.269371: step: 1440/526, loss: 0.011735916137695312 2023-01-21 12:30:45.437754: step: 1444/526, loss: 0.001981067704036832 2023-01-21 12:30:46.563757: step: 1448/526, loss: 0.0027264594100415707 2023-01-21 12:30:47.713802: step: 1452/526, loss: 0.018256474286317825 2023-01-21 12:30:48.843100: step: 1456/526, loss: 0.00015134812565520406 2023-01-21 12:30:49.970415: step: 1460/526, loss: 0.00010147095599677414 2023-01-21 12:30:51.099097: step: 1464/526, loss: 0.02340984344482422 2023-01-21 12:30:52.231075: step: 1468/526, loss: 0.0003719329833984375 2023-01-21 12:30:53.334191: step: 1472/526, loss: 3.414153979974799e-05 2023-01-21 12:30:54.531257: step: 1476/526, loss: 0.00016951560974121094 2023-01-21 12:30:55.654340: step: 1480/526, loss: 0.002627468202263117 2023-01-21 12:30:56.815917: step: 1484/526, loss: 0.0038399696350097656 2023-01-21 12:30:57.950190: step: 1488/526, loss: 0.005883598700165749 2023-01-21 12:30:59.087475: step: 1492/526, loss: 0.005117988679558039 2023-01-21 12:31:00.216539: step: 1496/526, loss: 0.0009925365447998047 2023-01-21 12:31:01.357410: step: 1500/526, loss: 0.0002210617094533518 2023-01-21 12:31:02.507648: step: 1504/526, loss: 0.0004280090215615928 2023-01-21 12:31:03.639495: step: 1508/526, loss: 0.0026557922828942537 2023-01-21 12:31:04.809471: step: 1512/526, loss: 1.4066697985981591e-05 2023-01-21 12:31:05.939433: step: 1516/526, loss: 0.004556751344352961 2023-01-21 12:31:07.109465: step: 1520/526, loss: 0.00031127932015806437 2023-01-21 12:31:08.229519: step: 1524/526, loss: 0.00597772654145956 2023-01-21 12:31:09.355068: step: 1528/526, loss: 0.003357029054313898 2023-01-21 12:31:10.506070: step: 1532/526, loss: 0.003941250033676624 2023-01-21 12:31:11.660259: step: 1536/526, loss: 0.03617596626281738 2023-01-21 12:31:12.779741: step: 1540/526, loss: 0.0002265930233988911 2023-01-21 12:31:13.887749: step: 1544/526, loss: 0.05983848497271538 2023-01-21 12:31:14.996602: step: 1548/526, loss: 9.183883958030492e-05 2023-01-21 12:31:16.122616: step: 1552/526, loss: 8.907318260753527e-05 2023-01-21 12:31:17.302550: step: 1556/526, loss: 0.021170424297451973 2023-01-21 12:31:18.441508: step: 1560/526, loss: 0.015160751529037952 2023-01-21 12:31:19.567644: step: 1564/526, loss: 0.0039014818612486124 2023-01-21 12:31:20.718992: step: 1568/526, loss: 1.7507259845733643 2023-01-21 12:31:21.857241: step: 1572/526, loss: 0.05242729187011719 2023-01-21 12:31:23.019272: step: 1576/526, loss: 0.04152173921465874 2023-01-21 12:31:24.136923: step: 1580/526, loss: 0.016037559136748314 2023-01-21 12:31:25.322392: step: 1584/526, loss: 0.03182535246014595 2023-01-21 12:31:26.448208: step: 1588/526, loss: 0.00027570725069381297 2023-01-21 12:31:27.579459: step: 1592/526, loss: 0.00564231863245368 2023-01-21 12:31:28.742480: step: 1596/526, loss: 0.0014621734153479338 2023-01-21 12:31:29.855711: step: 1600/526, loss: 0.025625038892030716 2023-01-21 12:31:31.042814: step: 1604/526, loss: 0.0006814957014285028 2023-01-21 12:31:32.196016: step: 1608/526, loss: 0.06305217742919922 2023-01-21 12:31:33.335298: step: 1612/526, loss: 0.01867065392434597 2023-01-21 12:31:34.471544: step: 1616/526, loss: 0.005424689967185259 2023-01-21 12:31:35.578494: step: 1620/526, loss: 0.05246276780962944 2023-01-21 12:31:36.726984: step: 1624/526, loss: 0.002684974577277899 2023-01-21 12:31:37.864858: step: 1628/526, loss: 0.02296895906329155 2023-01-21 12:31:39.001200: step: 1632/526, loss: 0.0018346786964684725 2023-01-21 12:31:40.140194: step: 1636/526, loss: 0.000812911952380091 2023-01-21 12:31:41.276236: step: 1640/526, loss: 0.014117050915956497 2023-01-21 12:31:42.424303: step: 1644/526, loss: 0.00014972686767578125 2023-01-21 12:31:43.550385: step: 1648/526, loss: 0.0015670761931687593 2023-01-21 12:31:44.688701: step: 1652/526, loss: 5.588531348621473e-05 2023-01-21 12:31:45.834192: step: 1656/526, loss: 0.0005754471058025956 2023-01-21 12:31:46.942006: step: 1660/526, loss: 0.005262947175651789 2023-01-21 12:31:48.076381: step: 1664/526, loss: 0.006017589475959539 2023-01-21 12:31:49.214813: step: 1668/526, loss: 0.00173015589825809 2023-01-21 12:31:50.346283: step: 1672/526, loss: 0.0006680488586425781 2023-01-21 12:31:51.482231: step: 1676/526, loss: 0.003529167268425226 2023-01-21 12:31:52.634901: step: 1680/526, loss: 0.0002471923944540322 2023-01-21 12:31:53.823332: step: 1684/526, loss: 0.008989477530121803 2023-01-21 12:31:54.945406: step: 1688/526, loss: 0.0008058547973632812 2023-01-21 12:31:56.085369: step: 1692/526, loss: 0.0038166046142578125 2023-01-21 12:31:57.247808: step: 1696/526, loss: 0.0017557144165039062 2023-01-21 12:31:58.369788: step: 1700/526, loss: 0.0081939697265625 2023-01-21 12:31:59.508412: step: 1704/526, loss: 0.021613502874970436 2023-01-21 12:32:00.644430: step: 1708/526, loss: 0.035605236887931824 2023-01-21 12:32:01.785317: step: 1712/526, loss: 0.0005512237548828125 2023-01-21 12:32:02.918074: step: 1716/526, loss: 0.01061544381082058 2023-01-21 12:32:04.038225: step: 1720/526, loss: 0.0032670022919774055 2023-01-21 12:32:05.179164: step: 1724/526, loss: 0.0008528709295205772 2023-01-21 12:32:06.302857: step: 1728/526, loss: 0.008709335699677467 2023-01-21 12:32:07.487359: step: 1732/526, loss: 0.029759027063846588 2023-01-21 12:32:08.633649: step: 1736/526, loss: 0.0010891915298998356 2023-01-21 12:32:09.740506: step: 1740/526, loss: 0.03299140930175781 2023-01-21 12:32:10.911106: step: 1744/526, loss: 0.006450462620705366 2023-01-21 12:32:12.028831: step: 1748/526, loss: 0.00022296907263807952 2023-01-21 12:32:13.177203: step: 1752/526, loss: 0.6999875903129578 2023-01-21 12:32:14.284598: step: 1756/526, loss: 0.013203429989516735 2023-01-21 12:32:15.405078: step: 1760/526, loss: 0.019240187481045723 2023-01-21 12:32:16.542395: step: 1764/526, loss: 0.014280319213867188 2023-01-21 12:32:17.672174: step: 1768/526, loss: 0.037392906844615936 2023-01-21 12:32:18.780521: step: 1772/526, loss: 0.003247309010475874 2023-01-21 12:32:19.933794: step: 1776/526, loss: 0.11433238536119461 2023-01-21 12:32:21.089593: step: 1780/526, loss: 0.0004230976337566972 2023-01-21 12:32:22.219507: step: 1784/526, loss: 0.04884061962366104 2023-01-21 12:32:23.338045: step: 1788/526, loss: 0.0024560929741710424 2023-01-21 12:32:24.467872: step: 1792/526, loss: 0.00013895034499000758 2023-01-21 12:32:25.602776: step: 1796/526, loss: 0.003559970995411277 2023-01-21 12:32:26.728180: step: 1800/526, loss: 0.19359055161476135 2023-01-21 12:32:27.839975: step: 1804/526, loss: 0.0281982421875 2023-01-21 12:32:28.971788: step: 1808/526, loss: 0.03411865234375 2023-01-21 12:32:30.092933: step: 1812/526, loss: 0.025551510974764824 2023-01-21 12:32:31.223887: step: 1816/526, loss: 0.054657746106386185 2023-01-21 12:32:32.372445: step: 1820/526, loss: 0.0024213790893554688 2023-01-21 12:32:33.545618: step: 1824/526, loss: 0.01178665179759264 2023-01-21 12:32:34.689330: step: 1828/526, loss: 4.5585635234601796e-05 2023-01-21 12:32:35.868765: step: 1832/526, loss: 0.01709447056055069 2023-01-21 12:32:36.993424: step: 1836/526, loss: 0.051303982734680176 2023-01-21 12:32:38.165192: step: 1840/526, loss: 0.015493965707719326 2023-01-21 12:32:39.288264: step: 1844/526, loss: 0.0005868911975994706 2023-01-21 12:32:40.433951: step: 1848/526, loss: 0.037518978118896484 2023-01-21 12:32:41.575591: step: 1852/526, loss: 0.0012385367881506681 2023-01-21 12:32:42.696714: step: 1856/526, loss: 0.005905342288315296 2023-01-21 12:32:43.877649: step: 1860/526, loss: 0.04580039903521538 2023-01-21 12:32:45.007858: step: 1864/526, loss: 0.00042896269587799907 2023-01-21 12:32:46.136346: step: 1868/526, loss: 0.024512099102139473 2023-01-21 12:32:47.274184: step: 1872/526, loss: 0.003866004990413785 2023-01-21 12:32:48.390249: step: 1876/526, loss: 5.073547436040826e-05 2023-01-21 12:32:49.536767: step: 1880/526, loss: 0.01756775751709938 2023-01-21 12:32:50.685112: step: 1884/526, loss: 0.07593861222267151 2023-01-21 12:32:51.832803: step: 1888/526, loss: 0.08946084976196289 2023-01-21 12:32:52.955232: step: 1892/526, loss: 0.07267895340919495 2023-01-21 12:32:54.057630: step: 1896/526, loss: 0.009779549203813076 2023-01-21 12:32:55.174947: step: 1900/526, loss: 0.0004269600030966103 2023-01-21 12:32:56.303282: step: 1904/526, loss: 0.0006159782642498612 2023-01-21 12:32:57.442857: step: 1908/526, loss: 0.0002957344113383442 2023-01-21 12:32:58.560069: step: 1912/526, loss: -2.956390289909905e-06 2023-01-21 12:32:59.748537: step: 1916/526, loss: 0.061963651329278946 2023-01-21 12:33:00.875401: step: 1920/526, loss: 5.4836273193359375e-06 2023-01-21 12:33:02.039311: step: 1924/526, loss: 0.007011794950813055 2023-01-21 12:33:03.182543: step: 1928/526, loss: 0.00021085739717818797 2023-01-21 12:33:04.304407: step: 1932/526, loss: 0.010194159112870693 2023-01-21 12:33:05.442907: step: 1936/526, loss: 0.009520625695586205 2023-01-21 12:33:06.563868: step: 1940/526, loss: 0.00036578180151991546 2023-01-21 12:33:07.712045: step: 1944/526, loss: 0.1800640970468521 2023-01-21 12:33:08.860065: step: 1948/526, loss: 5.559921555686742e-05 2023-01-21 12:33:10.031874: step: 1952/526, loss: 0.03282508999109268 2023-01-21 12:33:11.157612: step: 1956/526, loss: 0.0002084731968352571 2023-01-21 12:33:12.305446: step: 1960/526, loss: 0.006457519717514515 2023-01-21 12:33:13.472354: step: 1964/526, loss: 0.0986722931265831 2023-01-21 12:33:14.598558: step: 1968/526, loss: 0.03753481060266495 2023-01-21 12:33:15.723767: step: 1972/526, loss: 0.00035009384737350047 2023-01-21 12:33:16.885196: step: 1976/526, loss: 0.029037287458777428 2023-01-21 12:33:18.046297: step: 1980/526, loss: 0.00014438628568314016 2023-01-21 12:33:19.228092: step: 1984/526, loss: 0.005448341369628906 2023-01-21 12:33:20.360519: step: 1988/526, loss: 0.00855798739939928 2023-01-21 12:33:21.495346: step: 1992/526, loss: 0.007789993658661842 2023-01-21 12:33:22.670430: step: 1996/526, loss: 0.024349499493837357 2023-01-21 12:33:23.797365: step: 2000/526, loss: 0.0103591438382864 2023-01-21 12:33:24.947575: step: 2004/526, loss: 0.00107746128924191 2023-01-21 12:33:26.098466: step: 2008/526, loss: 0.007900619879364967 2023-01-21 12:33:27.245349: step: 2012/526, loss: 0.00010185241990257055 2023-01-21 12:33:28.402042: step: 2016/526, loss: 0.0004634857177734375 2023-01-21 12:33:29.532370: step: 2020/526, loss: 0.000782489834818989 2023-01-21 12:33:30.638692: step: 2024/526, loss: 0.014005064964294434 2023-01-21 12:33:31.749794: step: 2028/526, loss: 9.956360736396164e-05 2023-01-21 12:33:32.873176: step: 2032/526, loss: 0.006955218501389027 2023-01-21 12:33:33.999129: step: 2036/526, loss: 0.0004516601620707661 2023-01-21 12:33:35.143149: step: 2040/526, loss: 0.011570167727768421 2023-01-21 12:33:36.295124: step: 2044/526, loss: 0.02163248136639595 2023-01-21 12:33:37.449412: step: 2048/526, loss: 0.0007787704817019403 2023-01-21 12:33:38.618505: step: 2052/526, loss: 0.014159774407744408 2023-01-21 12:33:39.729422: step: 2056/526, loss: 0.00038356782170012593 2023-01-21 12:33:40.875648: step: 2060/526, loss: 0.0012542724143713713 2023-01-21 12:33:41.993651: step: 2064/526, loss: 0.007300854194909334 2023-01-21 12:33:43.131110: step: 2068/526, loss: 0.17027434706687927 2023-01-21 12:33:44.261298: step: 2072/526, loss: 0.0008005141862668097 2023-01-21 12:33:45.440600: step: 2076/526, loss: 0.0014862060779705644 2023-01-21 12:33:46.565113: step: 2080/526, loss: 0.03993682935833931 2023-01-21 12:33:47.685701: step: 2084/526, loss: 0.003623581025749445 2023-01-21 12:33:48.840347: step: 2088/526, loss: 0.003777885576710105 2023-01-21 12:33:50.004215: step: 2092/526, loss: 0.0019784928299486637 2023-01-21 12:33:51.168628: step: 2096/526, loss: 0.010195732116699219 2023-01-21 12:33:52.301752: step: 2100/526, loss: 0.00015802383131813258 2023-01-21 12:33:53.490358: step: 2104/526, loss: 0.0027029039338231087 ================================================== Loss: 0.022 -------------------- Dev: {'event': {'p': 0.5841784989858012, 'r': 0.7669773635153129, 'f1': 0.6632124352331606}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Test: {'event': {'p': 0.6179129005751849, 'r': 0.7849686847599165, 'f1': 0.6914942528735631}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Chinese: {'event': {'p': 0.5568181818181818, 'r': 0.9074074074074074, 'f1': 0.6901408450704225}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Korean: {'event': {'p': 0.6595744680851063, 'r': 0.49206349206349204, 'f1': 0.5636363636363635}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} Russian: {'event': {'p': 0.4634146341463415, 'r': 0.5277777777777778, 'f1': 0.4935064935064935}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 21} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 22 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:34:36.737698: step: 4/526, loss: 0.007927894592285156 2023-01-21 12:34:37.861730: step: 8/526, loss: 0.01830759085714817 2023-01-21 12:34:38.982278: step: 12/526, loss: 0.006164455320686102 2023-01-21 12:34:40.107320: step: 16/526, loss: 0.0023234367836266756 2023-01-21 12:34:41.261104: step: 20/526, loss: 0.0170153621584177 2023-01-21 12:34:42.415609: step: 24/526, loss: 0.009967708960175514 2023-01-21 12:34:43.536959: step: 28/526, loss: 0.0002826690615620464 2023-01-21 12:34:44.669653: step: 32/526, loss: 0.6450921297073364 2023-01-21 12:34:45.799900: step: 36/526, loss: 0.018439723178744316 2023-01-21 12:34:46.973008: step: 40/526, loss: 0.008635140024125576 2023-01-21 12:34:48.089668: step: 44/526, loss: 0.0040780543349683285 2023-01-21 12:34:49.239359: step: 48/526, loss: 0.00013108253187965602 2023-01-21 12:34:50.353777: step: 52/526, loss: 0.00545082101598382 2023-01-21 12:34:51.527461: step: 56/526, loss: 2.3269654775504023e-05 2023-01-21 12:34:52.655003: step: 60/526, loss: 0.0013195037608966231 2023-01-21 12:34:53.753714: step: 64/526, loss: 0.0001579284726176411 2023-01-21 12:34:54.860488: step: 68/526, loss: 0.00047860146150924265 2023-01-21 12:34:55.959942: step: 72/526, loss: 0.01788778230547905 2023-01-21 12:34:57.081159: step: 76/526, loss: 0.0006915092235431075 2023-01-21 12:34:58.212582: step: 80/526, loss: 0.003161525819450617 2023-01-21 12:34:59.380051: step: 84/526, loss: 0.015518855303525925 2023-01-21 12:35:00.510572: step: 88/526, loss: 0.05212097242474556 2023-01-21 12:35:01.669050: step: 92/526, loss: 0.007357597351074219 2023-01-21 12:35:02.829972: step: 96/526, loss: 0.007423591800034046 2023-01-21 12:35:03.966138: step: 100/526, loss: 0.008754349313676357 2023-01-21 12:35:05.090433: step: 104/526, loss: 0.00018110274686478078 2023-01-21 12:35:06.236594: step: 108/526, loss: 0.02275228500366211 2023-01-21 12:35:07.373857: step: 112/526, loss: 0.034697722643613815 2023-01-21 12:35:08.503839: step: 116/526, loss: 0.005748939700424671 2023-01-21 12:35:09.618986: step: 120/526, loss: 0.00022096632164902985 2023-01-21 12:35:10.743516: step: 124/526, loss: 0.00020313262939453125 2023-01-21 12:35:11.838164: step: 128/526, loss: 0.0030817033257335424 2023-01-21 12:35:12.991402: step: 132/526, loss: 0.014430142007768154 2023-01-21 12:35:14.134813: step: 136/526, loss: 0.0003222465456929058 2023-01-21 12:35:15.272820: step: 140/526, loss: 0.0027914047241210938 2023-01-21 12:35:16.393286: step: 144/526, loss: 2.403259350103326e-05 2023-01-21 12:35:17.505740: step: 148/526, loss: 4.482269287109375e-05 2023-01-21 12:35:18.617811: step: 152/526, loss: 0.0007287979242391884 2023-01-21 12:35:19.727789: step: 156/526, loss: 0.0012479304568842053 2023-01-21 12:35:20.862660: step: 160/526, loss: 9.35554489842616e-05 2023-01-21 12:35:21.997057: step: 164/526, loss: 0.0001068115234375 2023-01-21 12:35:23.133587: step: 168/526, loss: 0.005153274163603783 2023-01-21 12:35:24.269725: step: 172/526, loss: 0.0010914802551269531 2023-01-21 12:35:25.384587: step: 176/526, loss: 2.8371810913085938e-05 2023-01-21 12:35:26.507106: step: 180/526, loss: 0.0051104542799293995 2023-01-21 12:35:27.627814: step: 184/526, loss: 2.2315980459097773e-05 2023-01-21 12:35:28.754601: step: 188/526, loss: 0.00010414123244117945 2023-01-21 12:35:29.887619: step: 192/526, loss: 2.117157055181451e-05 2023-01-21 12:35:31.053689: step: 196/526, loss: 0.012742423452436924 2023-01-21 12:35:32.205898: step: 200/526, loss: 0.006153441034257412 2023-01-21 12:35:33.332242: step: 204/526, loss: 0.003979873843491077 2023-01-21 12:35:34.480488: step: 208/526, loss: 8.940697625803296e-06 2023-01-21 12:35:35.643512: step: 212/526, loss: 0.0009913444519042969 2023-01-21 12:35:36.785617: step: 216/526, loss: 0.010290145874023438 2023-01-21 12:35:37.897169: step: 220/526, loss: 0.0016716003883630037 2023-01-21 12:35:39.057576: step: 224/526, loss: 0.0007602691766805947 2023-01-21 12:35:40.165806: step: 228/526, loss: 0.010515929199755192 2023-01-21 12:35:41.271677: step: 232/526, loss: 0.0025112151633948088 2023-01-21 12:35:42.448412: step: 236/526, loss: 2.059936559817288e-05 2023-01-21 12:35:43.598359: step: 240/526, loss: 0.0011378765339031816 2023-01-21 12:35:44.723930: step: 244/526, loss: 0.004144859500229359 2023-01-21 12:35:45.912584: step: 248/526, loss: 0.0035165308509021997 2023-01-21 12:35:47.041963: step: 252/526, loss: 0.015324783511459827 2023-01-21 12:35:48.197384: step: 256/526, loss: 0.0038444518577307463 2023-01-21 12:35:49.353912: step: 260/526, loss: 0.006503296084702015 2023-01-21 12:35:50.484765: step: 264/526, loss: 0.00010218620445812121 2023-01-21 12:35:51.640314: step: 268/526, loss: 9.412765211891383e-05 2023-01-21 12:35:52.756442: step: 272/526, loss: 0.0021088600624352694 2023-01-21 12:35:53.875326: step: 276/526, loss: 0.007213306613266468 2023-01-21 12:35:55.003829: step: 280/526, loss: 0.007621001917868853 2023-01-21 12:35:56.151949: step: 284/526, loss: 0.007354068569839001 2023-01-21 12:35:57.261864: step: 288/526, loss: 0.00010778904106700793 2023-01-21 12:35:58.422683: step: 292/526, loss: 0.0006166935199871659 2023-01-21 12:35:59.548073: step: 296/526, loss: 0.0029600143898278475 2023-01-21 12:36:00.690430: step: 300/526, loss: 0.0003224372921977192 2023-01-21 12:36:01.825499: step: 304/526, loss: 0.057908058166503906 2023-01-21 12:36:02.947967: step: 308/526, loss: 0.00014481545076705515 2023-01-21 12:36:04.071937: step: 312/526, loss: 0.0019966126419603825 2023-01-21 12:36:05.195510: step: 316/526, loss: 0.03961486741900444 2023-01-21 12:36:06.365717: step: 320/526, loss: 0.004118156619369984 2023-01-21 12:36:07.503213: step: 324/526, loss: 0.008019733242690563 2023-01-21 12:36:08.659060: step: 328/526, loss: 0.0170777328312397 2023-01-21 12:36:09.771047: step: 332/526, loss: 0.009152126498520374 2023-01-21 12:36:10.926317: step: 336/526, loss: 3.366470627952367e-05 2023-01-21 12:36:12.066424: step: 340/526, loss: 0.00020987987227272242 2023-01-21 12:36:13.173525: step: 344/526, loss: 5.73158249608241e-05 2023-01-21 12:36:14.322017: step: 348/526, loss: 6.198883056640625e-05 2023-01-21 12:36:15.466960: step: 352/526, loss: 0.00716090202331543 2023-01-21 12:36:16.612105: step: 356/526, loss: 0.04049749672412872 2023-01-21 12:36:17.730997: step: 360/526, loss: 0.0026858330238610506 2023-01-21 12:36:18.874698: step: 364/526, loss: 0.07527142018079758 2023-01-21 12:36:20.014943: step: 368/526, loss: 0.04919242858886719 2023-01-21 12:36:21.126182: step: 372/526, loss: 0.00884170550853014 2023-01-21 12:36:22.273769: step: 376/526, loss: 0.0017216682899743319 2023-01-21 12:36:23.425843: step: 380/526, loss: 8.907318260753527e-05 2023-01-21 12:36:24.568857: step: 384/526, loss: 0.00685730017721653 2023-01-21 12:36:25.677163: step: 388/526, loss: 1.9168854123563506e-05 2023-01-21 12:36:26.811072: step: 392/526, loss: 3.976821608375758e-05 2023-01-21 12:36:27.947683: step: 396/526, loss: 0.018412113189697266 2023-01-21 12:36:29.076430: step: 400/526, loss: 0.00039801598177291453 2023-01-21 12:36:30.227318: step: 404/526, loss: 0.03472766652703285 2023-01-21 12:36:31.340264: step: 408/526, loss: 0.000697231269441545 2023-01-21 12:36:32.489247: step: 412/526, loss: 0.012841224670410156 2023-01-21 12:36:33.644923: step: 416/526, loss: 0.0001047134428517893 2023-01-21 12:36:34.777392: step: 420/526, loss: 0.0023822307121008635 2023-01-21 12:36:35.894306: step: 424/526, loss: 0.00858459435403347 2023-01-21 12:36:37.003419: step: 428/526, loss: 0.05080986022949219 2023-01-21 12:36:38.162301: step: 432/526, loss: 0.003403568407520652 2023-01-21 12:36:39.255048: step: 436/526, loss: 0.0002197265566792339 2023-01-21 12:36:40.374782: step: 440/526, loss: 0.001726341200992465 2023-01-21 12:36:41.551338: step: 444/526, loss: 0.00049762730486691 2023-01-21 12:36:42.671068: step: 448/526, loss: 0.0047966004349291325 2023-01-21 12:36:43.799299: step: 452/526, loss: 0.0840265303850174 2023-01-21 12:36:44.940026: step: 456/526, loss: 0.02573399432003498 2023-01-21 12:36:46.059507: step: 460/526, loss: 0.0239639300853014 2023-01-21 12:36:47.171335: step: 464/526, loss: 0.0007068633567541838 2023-01-21 12:36:48.296856: step: 468/526, loss: 0.0003845214960165322 2023-01-21 12:36:49.438623: step: 472/526, loss: 0.0009997368324548006 2023-01-21 12:36:50.595866: step: 476/526, loss: 0.0020549774635583162 2023-01-21 12:36:51.744320: step: 480/526, loss: 0.007887459360063076 2023-01-21 12:36:52.871706: step: 484/526, loss: 0.0002263069327455014 2023-01-21 12:36:54.040616: step: 488/526, loss: 0.0339902862906456 2023-01-21 12:36:55.210851: step: 492/526, loss: 0.15515127778053284 2023-01-21 12:36:56.331908: step: 496/526, loss: 0.001995182130485773 2023-01-21 12:36:57.480529: step: 500/526, loss: 0.0018994332058355212 2023-01-21 12:36:58.606023: step: 504/526, loss: 0.0006412506336346269 2023-01-21 12:36:59.734064: step: 508/526, loss: 0.09061269462108612 2023-01-21 12:37:00.850635: step: 512/526, loss: 0.032679032534360886 2023-01-21 12:37:01.979700: step: 516/526, loss: 0.03268270567059517 2023-01-21 12:37:03.104083: step: 520/526, loss: 0.00012302398681640625 2023-01-21 12:37:04.224670: step: 524/526, loss: 1.964569128176663e-05 2023-01-21 12:37:05.386246: step: 528/526, loss: 4.215240551275201e-05 2023-01-21 12:37:06.496755: step: 532/526, loss: 0.0028465748764574528 2023-01-21 12:37:07.643880: step: 536/526, loss: 0.04000244289636612 2023-01-21 12:37:08.781660: step: 540/526, loss: 0.016924476251006126 2023-01-21 12:37:09.928567: step: 544/526, loss: 0.0002304077206645161 2023-01-21 12:37:11.064922: step: 548/526, loss: 0.0010737419361248612 2023-01-21 12:37:12.215072: step: 552/526, loss: 0.011624526232481003 2023-01-21 12:37:13.346097: step: 556/526, loss: 2.3746490114717744e-05 2023-01-21 12:37:14.476028: step: 560/526, loss: 0.004565524868667126 2023-01-21 12:37:15.618977: step: 564/526, loss: 0.006567955017089844 2023-01-21 12:37:16.750993: step: 568/526, loss: 0.0016528129344806075 2023-01-21 12:37:17.892716: step: 572/526, loss: 0.0319090336561203 2023-01-21 12:37:19.040087: step: 576/526, loss: 0.0030702592339366674 2023-01-21 12:37:20.226144: step: 580/526, loss: 0.0033736228942871094 2023-01-21 12:37:21.358786: step: 584/526, loss: 0.008761786855757236 2023-01-21 12:37:22.511542: step: 588/526, loss: 0.0039994241669774055 2023-01-21 12:37:23.615823: step: 592/526, loss: 0.00029087066650390625 2023-01-21 12:37:24.721652: step: 596/526, loss: 0.00018396378436591476 2023-01-21 12:37:25.852822: step: 600/526, loss: 0.0013347625499591231 2023-01-21 12:37:26.991845: step: 604/526, loss: 0.0004343986511230469 2023-01-21 12:37:28.158226: step: 608/526, loss: 0.0002654552517924458 2023-01-21 12:37:29.277506: step: 612/526, loss: 5.836486889165826e-05 2023-01-21 12:37:30.417187: step: 616/526, loss: 0.00011577605619095266 2023-01-21 12:37:31.609534: step: 620/526, loss: 0.0035202025901526213 2023-01-21 12:37:32.766459: step: 624/526, loss: 0.0050741201266646385 2023-01-21 12:37:33.881337: step: 628/526, loss: 0.0012866974575445056 2023-01-21 12:37:35.023101: step: 632/526, loss: 0.0017437934875488281 2023-01-21 12:37:36.166147: step: 636/526, loss: 0.004202461335808039 2023-01-21 12:37:37.289044: step: 640/526, loss: 0.014503669925034046 2023-01-21 12:37:38.406698: step: 644/526, loss: 0.00015449525380972773 2023-01-21 12:37:39.543802: step: 648/526, loss: 0.006404972169548273 2023-01-21 12:37:40.671514: step: 652/526, loss: 0.03534698486328125 2023-01-21 12:37:41.761603: step: 656/526, loss: 0.01896214485168457 2023-01-21 12:37:42.912597: step: 660/526, loss: 0.06032419204711914 2023-01-21 12:37:44.046440: step: 664/526, loss: 0.001501321792602539 2023-01-21 12:37:45.181202: step: 668/526, loss: 0.00011882781836902723 2023-01-21 12:37:46.299325: step: 672/526, loss: 2.479553813827806e-06 2023-01-21 12:37:47.471689: step: 676/526, loss: 0.0029453278984874487 2023-01-21 12:37:48.631192: step: 680/526, loss: 4.2533876694506034e-05 2023-01-21 12:37:49.765791: step: 684/526, loss: 0.006904792971909046 2023-01-21 12:37:50.912691: step: 688/526, loss: 0.051347922533750534 2023-01-21 12:37:52.063676: step: 692/526, loss: 0.003003883408382535 2023-01-21 12:37:53.194871: step: 696/526, loss: 0.017427731305360794 2023-01-21 12:37:54.356348: step: 700/526, loss: 0.0020576477982103825 2023-01-21 12:37:55.473668: step: 704/526, loss: 0.02699604257941246 2023-01-21 12:37:56.617637: step: 708/526, loss: 0.018704796209931374 2023-01-21 12:37:57.768334: step: 712/526, loss: 0.0003643035888671875 2023-01-21 12:37:58.919588: step: 716/526, loss: 0.02192096970975399 2023-01-21 12:38:00.031940: step: 720/526, loss: 0.0002208709775004536 2023-01-21 12:38:01.185813: step: 724/526, loss: 0.013079643249511719 2023-01-21 12:38:02.323741: step: 728/526, loss: 0.01146697998046875 2023-01-21 12:38:03.465760: step: 732/526, loss: 0.008603858761489391 2023-01-21 12:38:04.617918: step: 736/526, loss: 0.0084381103515625 2023-01-21 12:38:05.791318: step: 740/526, loss: 0.00033397675724700093 2023-01-21 12:38:06.936197: step: 744/526, loss: 0.0003135681326966733 2023-01-21 12:38:08.100617: step: 748/526, loss: 3.967285010730848e-05 2023-01-21 12:38:09.210968: step: 752/526, loss: 0.00019092560978606343 2023-01-21 12:38:10.345054: step: 756/526, loss: 0.011512375436723232 2023-01-21 12:38:11.497286: step: 760/526, loss: 0.0016561507945880294 2023-01-21 12:38:12.635808: step: 764/526, loss: 0.007869553752243519 2023-01-21 12:38:13.764215: step: 768/526, loss: 6.12258882028982e-05 2023-01-21 12:38:14.928456: step: 772/526, loss: 0.006527757737785578 2023-01-21 12:38:16.080105: step: 776/526, loss: 0.029040098190307617 2023-01-21 12:38:17.233504: step: 780/526, loss: 0.3104965090751648 2023-01-21 12:38:18.374966: step: 784/526, loss: 0.0012784957652911544 2023-01-21 12:38:19.508340: step: 788/526, loss: 0.0036839963868260384 2023-01-21 12:38:20.672842: step: 792/526, loss: 9.15527380129788e-06 2023-01-21 12:38:21.807730: step: 796/526, loss: 0.000598907470703125 2023-01-21 12:38:22.926503: step: 800/526, loss: 0.0003949165402445942 2023-01-21 12:38:24.065766: step: 804/526, loss: 0.02554044872522354 2023-01-21 12:38:25.177319: step: 808/526, loss: 0.009838295169174671 2023-01-21 12:38:26.293867: step: 812/526, loss: 0.3151658773422241 2023-01-21 12:38:27.448289: step: 816/526, loss: 0.013678264804184437 2023-01-21 12:38:28.581032: step: 820/526, loss: 0.0452360138297081 2023-01-21 12:38:29.715965: step: 824/526, loss: 0.0013882637722417712 2023-01-21 12:38:30.832452: step: 828/526, loss: 0.003352546598762274 2023-01-21 12:38:31.963107: step: 832/526, loss: 0.0010985374683514237 2023-01-21 12:38:33.094554: step: 836/526, loss: 0.06973180919885635 2023-01-21 12:38:34.217337: step: 840/526, loss: 0.010907936841249466 2023-01-21 12:38:35.357693: step: 844/526, loss: 0.01543283462524414 2023-01-21 12:38:36.547502: step: 848/526, loss: 0.06298086792230606 2023-01-21 12:38:37.678213: step: 852/526, loss: 0.06486663967370987 2023-01-21 12:38:38.811018: step: 856/526, loss: 0.002442693803459406 2023-01-21 12:38:39.922430: step: 860/526, loss: 0.011987256817519665 2023-01-21 12:38:41.041323: step: 864/526, loss: 0.011977958492934704 2023-01-21 12:38:42.178794: step: 868/526, loss: 0.002745342208072543 2023-01-21 12:38:43.338652: step: 872/526, loss: 0.0009664535755291581 2023-01-21 12:38:44.480865: step: 876/526, loss: 0.00018625259690452367 2023-01-21 12:38:45.594967: step: 880/526, loss: 0.00108423235360533 2023-01-21 12:38:46.715399: step: 884/526, loss: 0.10615768283605576 2023-01-21 12:38:47.866756: step: 888/526, loss: 0.09314422309398651 2023-01-21 12:38:49.055860: step: 892/526, loss: 0.0025674819480627775 2023-01-21 12:38:50.174080: step: 896/526, loss: 9.021758887683973e-05 2023-01-21 12:38:51.308110: step: 900/526, loss: 0.0004505157412495464 2023-01-21 12:38:52.458303: step: 904/526, loss: 0.00024919508723542094 2023-01-21 12:38:53.599015: step: 908/526, loss: 0.0020849229767918587 2023-01-21 12:38:54.772523: step: 912/526, loss: 0.0002765655517578125 2023-01-21 12:38:55.895350: step: 916/526, loss: 0.000866842339746654 2023-01-21 12:38:57.038253: step: 920/526, loss: 0.03200606256723404 2023-01-21 12:38:58.166285: step: 924/526, loss: 0.0011563301086425781 2023-01-21 12:38:59.293981: step: 928/526, loss: 0.0001295089750783518 2023-01-21 12:39:00.414452: step: 932/526, loss: 0.0001903533993754536 2023-01-21 12:39:01.562315: step: 936/526, loss: 0.0025975226890295744 2023-01-21 12:39:02.733761: step: 940/526, loss: 0.0005887985462322831 2023-01-21 12:39:03.857616: step: 944/526, loss: 0.011120987124741077 2023-01-21 12:39:04.996479: step: 948/526, loss: 0.056415461003780365 2023-01-21 12:39:06.113531: step: 952/526, loss: 0.017654899507761 2023-01-21 12:39:07.258248: step: 956/526, loss: 0.0032493590842932463 2023-01-21 12:39:08.427022: step: 960/526, loss: 0.04450645670294762 2023-01-21 12:39:09.586866: step: 964/526, loss: 5.42163870704826e-05 2023-01-21 12:39:10.713065: step: 968/526, loss: 1.0967254638671875e-05 2023-01-21 12:39:11.872664: step: 972/526, loss: 0.007106209173798561 2023-01-21 12:39:13.004177: step: 976/526, loss: 0.0002753734588623047 2023-01-21 12:39:14.119747: step: 980/526, loss: 0.003381061600521207 2023-01-21 12:39:15.241584: step: 984/526, loss: 0.03594255447387695 2023-01-21 12:39:16.383107: step: 988/526, loss: 0.047146987169981 2023-01-21 12:39:17.534825: step: 992/526, loss: 0.013032246381044388 2023-01-21 12:39:18.678871: step: 996/526, loss: 0.0005305290105752647 2023-01-21 12:39:19.802513: step: 1000/526, loss: 0.03170814365148544 2023-01-21 12:39:20.923145: step: 1004/526, loss: 0.026973627507686615 2023-01-21 12:39:22.068736: step: 1008/526, loss: 0.029405688866972923 2023-01-21 12:39:23.224067: step: 1012/526, loss: 0.0001583099365234375 2023-01-21 12:39:24.359551: step: 1016/526, loss: 0.009625243954360485 2023-01-21 12:39:25.491959: step: 1020/526, loss: 0.00681152381002903 2023-01-21 12:39:26.636832: step: 1024/526, loss: 0.012583828531205654 2023-01-21 12:39:27.771204: step: 1028/526, loss: 0.00015993117995094508 2023-01-21 12:39:28.902598: step: 1032/526, loss: 3.61442580469884e-05 2023-01-21 12:39:30.043605: step: 1036/526, loss: 0.05180578678846359 2023-01-21 12:39:31.195913: step: 1040/526, loss: 0.0014400483341887593 2023-01-21 12:39:32.346703: step: 1044/526, loss: 0.01643834076821804 2023-01-21 12:39:33.483388: step: 1048/526, loss: 0.013430404476821423 2023-01-21 12:39:34.637915: step: 1052/526, loss: 0.00013236999802757055 2023-01-21 12:39:35.773403: step: 1056/526, loss: 0.025191688910126686 2023-01-21 12:39:36.915447: step: 1060/526, loss: 0.012255859561264515 2023-01-21 12:39:38.073037: step: 1064/526, loss: 0.039710428565740585 2023-01-21 12:39:39.198660: step: 1068/526, loss: 0.00537033099681139 2023-01-21 12:39:40.330528: step: 1072/526, loss: 4.491805884754285e-05 2023-01-21 12:39:41.446309: step: 1076/526, loss: 1.406669707648689e-05 2023-01-21 12:39:42.598141: step: 1080/526, loss: 0.000125885009765625 2023-01-21 12:39:43.703283: step: 1084/526, loss: 0.053790852427482605 2023-01-21 12:39:44.838141: step: 1088/526, loss: 1.773834264895413e-05 2023-01-21 12:39:45.973044: step: 1092/526, loss: 0.0011372566223144531 2023-01-21 12:39:47.114895: step: 1096/526, loss: 0.010393811389803886 2023-01-21 12:39:48.292656: step: 1100/526, loss: 0.0008558273548260331 2023-01-21 12:39:49.428326: step: 1104/526, loss: 0.00024337769718840718 2023-01-21 12:39:50.594208: step: 1108/526, loss: 0.001394939492456615 2023-01-21 12:39:51.722896: step: 1112/526, loss: 0.00951242446899414 2023-01-21 12:39:52.879078: step: 1116/526, loss: 0.002495241118595004 2023-01-21 12:39:54.008318: step: 1120/526, loss: 0.013024520128965378 2023-01-21 12:39:55.128129: step: 1124/526, loss: 0.0010013104183599353 2023-01-21 12:39:56.291689: step: 1128/526, loss: 0.016613198444247246 2023-01-21 12:39:57.417867: step: 1132/526, loss: 0.007111930754035711 2023-01-21 12:39:58.559954: step: 1136/526, loss: 1.296997106692288e-05 2023-01-21 12:39:59.699795: step: 1140/526, loss: 0.03424868360161781 2023-01-21 12:40:00.826391: step: 1144/526, loss: 0.0014286995865404606 2023-01-21 12:40:01.935159: step: 1148/526, loss: 0.009533500298857689 2023-01-21 12:40:03.070660: step: 1152/526, loss: 0.006971550174057484 2023-01-21 12:40:04.244410: step: 1156/526, loss: 0.03789863735437393 2023-01-21 12:40:05.373297: step: 1160/526, loss: 0.0006389617919921875 2023-01-21 12:40:06.493209: step: 1164/526, loss: 0.006878376007080078 2023-01-21 12:40:07.622127: step: 1168/526, loss: 0.002097606658935547 2023-01-21 12:40:08.756302: step: 1172/526, loss: 0.018703175708651543 2023-01-21 12:40:09.923626: step: 1176/526, loss: 0.005656242370605469 2023-01-21 12:40:11.052563: step: 1180/526, loss: 0.0009202957153320312 2023-01-21 12:40:12.190216: step: 1184/526, loss: 0.00011105537851108238 2023-01-21 12:40:13.350462: step: 1188/526, loss: 0.0008386612171307206 2023-01-21 12:40:14.464886: step: 1192/526, loss: 0.0035538673400878906 2023-01-21 12:40:15.629798: step: 1196/526, loss: 0.05066833272576332 2023-01-21 12:40:16.785219: step: 1200/526, loss: 0.002857208251953125 2023-01-21 12:40:17.924783: step: 1204/526, loss: 0.013005065731704235 2023-01-21 12:40:19.024749: step: 1208/526, loss: 0.00037097930908203125 2023-01-21 12:40:20.172286: step: 1212/526, loss: 0.0021614073775708675 2023-01-21 12:40:21.316593: step: 1216/526, loss: 0.0004459381161723286 2023-01-21 12:40:22.470017: step: 1220/526, loss: 0.016316134482622147 2023-01-21 12:40:23.609970: step: 1224/526, loss: 0.003688812255859375 2023-01-21 12:40:24.761855: step: 1228/526, loss: 0.000263214111328125 2023-01-21 12:40:25.918572: step: 1232/526, loss: 0.00021133421978447586 2023-01-21 12:40:27.039305: step: 1236/526, loss: 0.00019788742065429688 2023-01-21 12:40:28.159858: step: 1240/526, loss: 0.0004879951593466103 2023-01-21 12:40:29.291040: step: 1244/526, loss: 0.0003581047058105469 2023-01-21 12:40:30.478421: step: 1248/526, loss: 0.000263214111328125 2023-01-21 12:40:31.608114: step: 1252/526, loss: 0.0006858825800009072 2023-01-21 12:40:32.769311: step: 1256/526, loss: 0.03709583356976509 2023-01-21 12:40:33.892256: step: 1260/526, loss: 6.246566772460938e-05 2023-01-21 12:40:35.043993: step: 1264/526, loss: 2.059936559817288e-05 2023-01-21 12:40:36.201933: step: 1268/526, loss: 0.046073149889707565 2023-01-21 12:40:37.331971: step: 1272/526, loss: 2.098083541568485e-06 2023-01-21 12:40:38.479106: step: 1276/526, loss: 0.00023145676823332906 2023-01-21 12:40:39.608984: step: 1280/526, loss: 0.00011024475679732859 2023-01-21 12:40:40.729408: step: 1284/526, loss: 0.00864114798605442 2023-01-21 12:40:41.860931: step: 1288/526, loss: 0.013900947757065296 2023-01-21 12:40:43.051662: step: 1292/526, loss: 0.0022674561478197575 2023-01-21 12:40:44.204378: step: 1296/526, loss: 0.00211334228515625 2023-01-21 12:40:45.331685: step: 1300/526, loss: 9.460448927711695e-05 2023-01-21 12:40:46.476629: step: 1304/526, loss: 0.0007321357843466103 2023-01-21 12:40:47.655118: step: 1308/526, loss: 0.0005928039317950606 2023-01-21 12:40:48.823555: step: 1312/526, loss: 0.0014251709217205644 2023-01-21 12:40:49.934984: step: 1316/526, loss: 0.00107746128924191 2023-01-21 12:40:51.083080: step: 1320/526, loss: 4.787445141118951e-05 2023-01-21 12:40:52.209686: step: 1324/526, loss: 0.00011014938354492188 2023-01-21 12:40:53.341408: step: 1328/526, loss: 0.0008003234979696572 2023-01-21 12:40:54.511823: step: 1332/526, loss: 0.010824394412338734 2023-01-21 12:40:55.641386: step: 1336/526, loss: 4.4155120122013614e-05 2023-01-21 12:40:56.761495: step: 1340/526, loss: 0.0014827728737145662 2023-01-21 12:40:57.860245: step: 1344/526, loss: 0.00039892198401503265 2023-01-21 12:40:58.986675: step: 1348/526, loss: 0.0003302574041299522 2023-01-21 12:41:00.102055: step: 1352/526, loss: 0.0015388489700853825 2023-01-21 12:41:01.265421: step: 1356/526, loss: 0.07164879143238068 2023-01-21 12:41:02.379285: step: 1360/526, loss: 0.00306529994122684 2023-01-21 12:41:03.527550: step: 1364/526, loss: 0.007248306646943092 2023-01-21 12:41:04.656464: step: 1368/526, loss: 8.583068620282575e-07 2023-01-21 12:41:05.772309: step: 1372/526, loss: 0.0007380486349575222 2023-01-21 12:41:06.936359: step: 1376/526, loss: 8.478164818370715e-05 2023-01-21 12:41:08.067807: step: 1380/526, loss: 0.01052713394165039 2023-01-21 12:41:09.205057: step: 1384/526, loss: 0.015980910509824753 2023-01-21 12:41:10.348331: step: 1388/526, loss: 0.06886310875415802 2023-01-21 12:41:11.460926: step: 1392/526, loss: 0.0002882003609556705 2023-01-21 12:41:12.593861: step: 1396/526, loss: 0.0003448486386332661 2023-01-21 12:41:13.743414: step: 1400/526, loss: 0.011772585101425648 2023-01-21 12:41:14.883041: step: 1404/526, loss: 0.0002038955717580393 2023-01-21 12:41:16.009018: step: 1408/526, loss: 0.0018633842701092362 2023-01-21 12:41:17.143831: step: 1412/526, loss: 8.654594421386719e-05 2023-01-21 12:41:18.290142: step: 1416/526, loss: 0.004313564393669367 2023-01-21 12:41:19.411152: step: 1420/526, loss: 0.00011711120896507055 2023-01-21 12:41:20.538859: step: 1424/526, loss: 0.23872889578342438 2023-01-21 12:41:21.681381: step: 1428/526, loss: 0.6750664114952087 2023-01-21 12:41:22.823947: step: 1432/526, loss: 0.04905061796307564 2023-01-21 12:41:23.996360: step: 1436/526, loss: 0.007182503119111061 2023-01-21 12:41:25.238848: step: 1440/526, loss: 0.07318788021802902 2023-01-21 12:41:26.358423: step: 1444/526, loss: 0.06385822594165802 2023-01-21 12:41:27.492850: step: 1448/526, loss: 8.640289161121473e-05 2023-01-21 12:41:28.642479: step: 1452/526, loss: 0.015682749450206757 2023-01-21 12:41:29.770854: step: 1456/526, loss: 0.0009546280489303172 2023-01-21 12:41:30.926595: step: 1460/526, loss: 0.009378910064697266 2023-01-21 12:41:32.050710: step: 1464/526, loss: 0.056096456944942474 2023-01-21 12:41:33.191795: step: 1468/526, loss: 0.04520893469452858 2023-01-21 12:41:34.316764: step: 1472/526, loss: 0.10247516632080078 2023-01-21 12:41:35.482572: step: 1476/526, loss: 0.02235870435833931 2023-01-21 12:41:36.606464: step: 1480/526, loss: 0.006798172369599342 2023-01-21 12:41:37.751148: step: 1484/526, loss: 0.0004818916495423764 2023-01-21 12:41:38.854640: step: 1488/526, loss: 0.0001104354887502268 2023-01-21 12:41:40.001801: step: 1492/526, loss: 0.00032444001408293843 2023-01-21 12:41:41.135731: step: 1496/526, loss: 0.0039916993118822575 2023-01-21 12:41:42.234204: step: 1500/526, loss: 0.007906818762421608 2023-01-21 12:41:43.366833: step: 1504/526, loss: 0.0018359185196459293 2023-01-21 12:41:44.533295: step: 1508/526, loss: 0.017743302509188652 2023-01-21 12:41:45.671098: step: 1512/526, loss: 0.00039405826828442514 2023-01-21 12:41:46.789124: step: 1516/526, loss: 0.05438556522130966 2023-01-21 12:41:47.951465: step: 1520/526, loss: 0.04722042381763458 2023-01-21 12:41:49.100178: step: 1524/526, loss: 0.003061676165089011 2023-01-21 12:41:50.249158: step: 1528/526, loss: 0.0003373146173544228 2023-01-21 12:41:51.368730: step: 1532/526, loss: 5.4073334467830136e-05 2023-01-21 12:41:52.503509: step: 1536/526, loss: 3.790855407714844e-05 2023-01-21 12:41:53.662923: step: 1540/526, loss: 0.015591240487992764 2023-01-21 12:41:54.783086: step: 1544/526, loss: 0.023667145520448685 2023-01-21 12:41:55.920463: step: 1548/526, loss: 0.0005550384521484375 2023-01-21 12:41:57.051147: step: 1552/526, loss: 0.0005651474348269403 2023-01-21 12:41:58.204406: step: 1556/526, loss: 0.002125549130141735 2023-01-21 12:41:59.319184: step: 1560/526, loss: 0.0041055199690163136 2023-01-21 12:42:00.443513: step: 1564/526, loss: 0.46366769075393677 2023-01-21 12:42:01.614468: step: 1568/526, loss: 0.0003819942648988217 2023-01-21 12:42:02.744498: step: 1572/526, loss: 0.06436386704444885 2023-01-21 12:42:03.875006: step: 1576/526, loss: 0.0013162612449377775 2023-01-21 12:42:04.973402: step: 1580/526, loss: 0.0005640983581542969 2023-01-21 12:42:06.098955: step: 1584/526, loss: 0.0007989883306436241 2023-01-21 12:42:07.290456: step: 1588/526, loss: 0.02109508588910103 2023-01-21 12:42:08.414869: step: 1592/526, loss: 0.0007202147971838713 2023-01-21 12:42:09.568860: step: 1596/526, loss: 0.008037377148866653 2023-01-21 12:42:10.683861: step: 1600/526, loss: 0.0018079758156090975 2023-01-21 12:42:11.853667: step: 1604/526, loss: 0.02293567731976509 2023-01-21 12:42:12.993605: step: 1608/526, loss: 0.00011510848707985133 2023-01-21 12:42:14.137184: step: 1612/526, loss: 0.0007442474598065019 2023-01-21 12:42:15.283749: step: 1616/526, loss: 0.09318371117115021 2023-01-21 12:42:16.419766: step: 1620/526, loss: 0.1565997153520584 2023-01-21 12:42:17.527856: step: 1624/526, loss: 0.0023903846740722656 2023-01-21 12:42:18.674716: step: 1628/526, loss: 5.207061622058973e-05 2023-01-21 12:42:19.813527: step: 1632/526, loss: 0.36484020948410034 2023-01-21 12:42:20.952613: step: 1636/526, loss: 0.04972992092370987 2023-01-21 12:42:22.088488: step: 1640/526, loss: 0.01793498918414116 2023-01-21 12:42:23.222108: step: 1644/526, loss: 0.011297320947051048 2023-01-21 12:42:24.359635: step: 1648/526, loss: 9.5367431640625e-05 2023-01-21 12:42:25.482290: step: 1652/526, loss: 0.034090518951416016 2023-01-21 12:42:26.648677: step: 1656/526, loss: 0.012037324719130993 2023-01-21 12:42:27.766165: step: 1660/526, loss: 0.0037815095856785774 2023-01-21 12:42:28.905407: step: 1664/526, loss: 0.00022306441678665578 2023-01-21 12:42:30.024188: step: 1668/526, loss: 0.020898818969726562 2023-01-21 12:42:31.147787: step: 1672/526, loss: 0.0025756836403161287 2023-01-21 12:42:32.283889: step: 1676/526, loss: 0.0031532763969153166 2023-01-21 12:42:33.424717: step: 1680/526, loss: 0.0014539242256432772 2023-01-21 12:42:34.571895: step: 1684/526, loss: 0.0037601471412926912 2023-01-21 12:42:35.706881: step: 1688/526, loss: 0.03642597422003746 2023-01-21 12:42:36.832892: step: 1692/526, loss: 0.0008725167135708034 2023-01-21 12:42:37.946596: step: 1696/526, loss: 0.0001462936488678679 2023-01-21 12:42:39.069213: step: 1700/526, loss: 0.004838180728256702 2023-01-21 12:42:40.229400: step: 1704/526, loss: 0.008051491342484951 2023-01-21 12:42:41.365827: step: 1708/526, loss: 0.005165672395378351 2023-01-21 12:42:42.508316: step: 1712/526, loss: 0.28592395782470703 2023-01-21 12:42:43.643516: step: 1716/526, loss: 0.02497396431863308 2023-01-21 12:42:44.776966: step: 1720/526, loss: 0.0007458209875039756 2023-01-21 12:42:45.923798: step: 1724/526, loss: 0.0026048661675304174 2023-01-21 12:42:47.037916: step: 1728/526, loss: 0.0004734039248432964 2023-01-21 12:42:48.171843: step: 1732/526, loss: 0.042105820029973984 2023-01-21 12:42:49.321320: step: 1736/526, loss: 8.406638517044485e-05 2023-01-21 12:42:50.451089: step: 1740/526, loss: 0.0004470825369935483 2023-01-21 12:42:51.618544: step: 1744/526, loss: 0.022427940741181374 2023-01-21 12:42:52.747908: step: 1748/526, loss: 0.004199123475700617 2023-01-21 12:42:53.898065: step: 1752/526, loss: 0.0019474030705168843 2023-01-21 12:42:55.039415: step: 1756/526, loss: 0.00014629363431595266 2023-01-21 12:42:56.150744: step: 1760/526, loss: 7.390976406895788e-06 2023-01-21 12:42:57.271894: step: 1764/526, loss: 0.009794902987778187 2023-01-21 12:42:58.423991: step: 1768/526, loss: 0.01474075298756361 2023-01-21 12:42:59.575258: step: 1772/526, loss: 0.00010910034325206652 2023-01-21 12:43:00.707390: step: 1776/526, loss: 0.022666549310088158 2023-01-21 12:43:01.864325: step: 1780/526, loss: 0.00017051697068382055 2023-01-21 12:43:02.985396: step: 1784/526, loss: 5.187988426769152e-05 2023-01-21 12:43:04.147463: step: 1788/526, loss: 0.007553768344223499 2023-01-21 12:43:05.284456: step: 1792/526, loss: 0.07739105075597763 2023-01-21 12:43:06.448873: step: 1796/526, loss: 0.0001371383696096018 2023-01-21 12:43:07.624114: step: 1800/526, loss: 0.001453495118767023 2023-01-21 12:43:08.770689: step: 1804/526, loss: -1.411438006471144e-05 2023-01-21 12:43:09.894326: step: 1808/526, loss: 0.0005991935613565147 2023-01-21 12:43:11.055326: step: 1812/526, loss: 0.0014359474880620837 2023-01-21 12:43:12.203838: step: 1816/526, loss: 5.950928243692033e-05 2023-01-21 12:43:13.344363: step: 1820/526, loss: 0.0008203506586141884 2023-01-21 12:43:14.504212: step: 1824/526, loss: 0.010953378863632679 2023-01-21 12:43:15.630715: step: 1828/526, loss: 0.007882309146225452 2023-01-21 12:43:16.796883: step: 1832/526, loss: 0.0008634567493572831 2023-01-21 12:43:17.907718: step: 1836/526, loss: 0.0004085063992533833 2023-01-21 12:43:19.045773: step: 1840/526, loss: 0.0013549805153161287 2023-01-21 12:43:20.189019: step: 1844/526, loss: 0.011856365948915482 2023-01-21 12:43:21.346322: step: 1848/526, loss: 1.0117520093917847 2023-01-21 12:43:22.483157: step: 1852/526, loss: 0.023229314014315605 2023-01-21 12:43:23.614771: step: 1856/526, loss: 0.00020742416381835938 2023-01-21 12:43:24.770453: step: 1860/526, loss: 0.03857707977294922 2023-01-21 12:43:25.922393: step: 1864/526, loss: 0.0009004116291180253 2023-01-21 12:43:27.021116: step: 1868/526, loss: 0.00024518967256881297 2023-01-21 12:43:28.161204: step: 1872/526, loss: 0.0006789207691326737 2023-01-21 12:43:29.319848: step: 1876/526, loss: 0.003887748811393976 2023-01-21 12:43:30.470258: step: 1880/526, loss: 0.05139608308672905 2023-01-21 12:43:31.602002: step: 1884/526, loss: 0.006644821260124445 2023-01-21 12:43:32.784449: step: 1888/526, loss: 0.035083770751953125 2023-01-21 12:43:33.905028: step: 1892/526, loss: 0.0072428700514137745 2023-01-21 12:43:35.021403: step: 1896/526, loss: 0.004093742463737726 2023-01-21 12:43:36.163023: step: 1900/526, loss: 0.0022306444589048624 2023-01-21 12:43:37.325435: step: 1904/526, loss: 0.06239604949951172 2023-01-21 12:43:38.497248: step: 1908/526, loss: 0.00032558440580032766 2023-01-21 12:43:39.636800: step: 1912/526, loss: 0.07036600261926651 2023-01-21 12:43:40.745194: step: 1916/526, loss: 5.130767749506049e-05 2023-01-21 12:43:41.871229: step: 1920/526, loss: 0.022225569933652878 2023-01-21 12:43:43.064681: step: 1924/526, loss: 0.0007571220048703253 2023-01-21 12:43:44.189025: step: 1928/526, loss: 0.0005041122785769403 2023-01-21 12:43:45.325680: step: 1932/526, loss: 0.0003651142178568989 2023-01-21 12:43:46.467061: step: 1936/526, loss: 0.0773773193359375 2023-01-21 12:43:47.582797: step: 1940/526, loss: 0.016583634540438652 2023-01-21 12:43:48.708907: step: 1944/526, loss: 0.0005136490217410028 2023-01-21 12:43:49.832887: step: 1948/526, loss: 0.000676918076351285 2023-01-21 12:43:50.957268: step: 1952/526, loss: 0.0316162109375 2023-01-21 12:43:52.113395: step: 1956/526, loss: 0.007002830505371094 2023-01-21 12:43:53.272948: step: 1960/526, loss: 0.016753101721405983 2023-01-21 12:43:54.400376: step: 1964/526, loss: 0.03154182434082031 2023-01-21 12:43:55.536839: step: 1968/526, loss: 2.7751922971219756e-05 2023-01-21 12:43:56.694444: step: 1972/526, loss: 0.00022759436978958547 2023-01-21 12:43:57.840141: step: 1976/526, loss: 0.009221648797392845 2023-01-21 12:43:58.983374: step: 1980/526, loss: 0.003183126449584961 2023-01-21 12:44:00.106167: step: 1984/526, loss: 0.0009048461797647178 2023-01-21 12:44:01.243459: step: 1988/526, loss: 0.001155090401880443 2023-01-21 12:44:02.386217: step: 1992/526, loss: 0.0022377970162779093 2023-01-21 12:44:03.484724: step: 1996/526, loss: 0.0015698432689532638 2023-01-21 12:44:04.630604: step: 2000/526, loss: 0.004531955812126398 2023-01-21 12:44:05.792253: step: 2004/526, loss: 0.00027027129544876516 2023-01-21 12:44:06.924420: step: 2008/526, loss: 0.05013151094317436 2023-01-21 12:44:08.059249: step: 2012/526, loss: 0.023787595331668854 2023-01-21 12:44:09.238931: step: 2016/526, loss: 0.017963409423828125 2023-01-21 12:44:10.343807: step: 2020/526, loss: 0.01220550574362278 2023-01-21 12:44:11.471811: step: 2024/526, loss: 0.0020858377683907747 2023-01-21 12:44:12.582493: step: 2028/526, loss: 8.335113670909777e-05 2023-01-21 12:44:13.766541: step: 2032/526, loss: 0.0033470632042735815 2023-01-21 12:44:14.869364: step: 2036/526, loss: 0.0022212981712073088 2023-01-21 12:44:16.000469: step: 2040/526, loss: 0.000404548627557233 2023-01-21 12:44:17.112923: step: 2044/526, loss: 3.1948089599609375e-05 2023-01-21 12:44:18.245477: step: 2048/526, loss: 0.0012963295448571444 2023-01-21 12:44:19.415865: step: 2052/526, loss: 0.0009089470258913934 2023-01-21 12:44:20.573543: step: 2056/526, loss: 0.0011604309547692537 2023-01-21 12:44:21.696421: step: 2060/526, loss: 0.0005937576061114669 2023-01-21 12:44:22.826589: step: 2064/526, loss: 0.030721569433808327 2023-01-21 12:44:23.950434: step: 2068/526, loss: 0.0628567710518837 2023-01-21 12:44:25.092594: step: 2072/526, loss: 0.00010251998173771426 2023-01-21 12:44:26.230787: step: 2076/526, loss: 0.0002650260867085308 2023-01-21 12:44:27.356614: step: 2080/526, loss: 2.3508071535616182e-05 2023-01-21 12:44:28.493626: step: 2084/526, loss: 0.004207802005112171 2023-01-21 12:44:29.625559: step: 2088/526, loss: 0.027065325528383255 2023-01-21 12:44:30.788575: step: 2092/526, loss: 0.0002834320184774697 2023-01-21 12:44:31.932841: step: 2096/526, loss: 0.00039920807466842234 2023-01-21 12:44:33.063286: step: 2100/526, loss: 0.0002849579032044858 2023-01-21 12:44:34.200977: step: 2104/526, loss: 8.926392183639109e-05 ================================================== Loss: 0.020 -------------------- Dev: {'event': {'p': 0.5920245398773006, 'r': 0.7709720372836218, 'f1': 0.6697513013302487}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Test: {'event': {'p': 0.6204047914085089, 'r': 0.7839248434237995, 'f1': 0.6926446852663132}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Chinese: {'event': {'p': 0.573170731707317, 'r': 0.8703703703703703, 'f1': 0.6911764705882353}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Korean: {'event': {'p': 0.6170212765957447, 'r': 0.4603174603174603, 'f1': 0.5272727272727272}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} Russian: {'event': {'p': 0.45652173913043476, 'r': 0.5833333333333334, 'f1': 0.5121951219512195}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 22} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6389811738648948, 'r': 0.7683089214380826, 'f1': 0.6977025392986699}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Eng Test for Chinese: {'event': {'p': 0.6328626444159179, 'r': 0.7719206680584552, 'f1': 0.6955090524335763}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} Sample Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 13} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.5800575263662512, 'r': 0.8055925432756325, 'f1': 0.6744704570791527}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Eng Test for Russian: {'event': {'p': 0.5962534874451972, 'r': 0.7807933194154488, 'f1': 0.6761581920903954}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} Sample Russian: {'event': {'p': 0.48936170212765956, 'r': 0.6388888888888888, 'f1': 0.5542168674698795}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 16} ****************************** Epoch: 23 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:45:17.464927: step: 4/526, loss: 0.004995489027351141 2023-01-21 12:45:18.584956: step: 8/526, loss: 0.00027647020760923624 2023-01-21 12:45:19.712048: step: 12/526, loss: 0.0015367508167400956 2023-01-21 12:45:20.857347: step: 16/526, loss: 0.06378059834241867 2023-01-21 12:45:21.984205: step: 20/526, loss: 4.5490269258152694e-05 2023-01-21 12:45:23.143209: step: 24/526, loss: 0.02482783794403076 2023-01-21 12:45:24.285325: step: 28/526, loss: 0.00034933091956190765 2023-01-21 12:45:25.419684: step: 32/526, loss: 0.013775492087006569 2023-01-21 12:45:26.554982: step: 36/526, loss: 0.01672840118408203 2023-01-21 12:45:27.688813: step: 40/526, loss: 0.001422977540642023 2023-01-21 12:45:28.834333: step: 44/526, loss: 0.0017678738804534078 2023-01-21 12:45:29.980706: step: 48/526, loss: 0.0004125595442019403 2023-01-21 12:45:31.102502: step: 52/526, loss: 0.0027025220915675163 2023-01-21 12:45:32.270785: step: 56/526, loss: 0.000676059746183455 2023-01-21 12:45:33.424192: step: 60/526, loss: 0.0004665374872274697 2023-01-21 12:45:34.556467: step: 64/526, loss: 0.08500690758228302 2023-01-21 12:45:35.652295: step: 68/526, loss: 0.0008096694946289062 2023-01-21 12:45:36.778401: step: 72/526, loss: 0.03862740099430084 2023-01-21 12:45:37.927847: step: 76/526, loss: 0.004749011714011431 2023-01-21 12:45:39.098258: step: 80/526, loss: 0.024715041741728783 2023-01-21 12:45:40.226863: step: 84/526, loss: 0.00031595228938385844 2023-01-21 12:45:41.338925: step: 88/526, loss: 0.14136847853660583 2023-01-21 12:45:42.506312: step: 92/526, loss: 0.0005692958948202431 2023-01-21 12:45:43.635626: step: 96/526, loss: 0.00046720507089048624 2023-01-21 12:45:44.792758: step: 100/526, loss: 0.002639675047248602 2023-01-21 12:45:45.925023: step: 104/526, loss: 0.00017113685316871852 2023-01-21 12:45:47.074312: step: 108/526, loss: 0.00014209747314453125 2023-01-21 12:45:48.223619: step: 112/526, loss: 0.0005893707275390625 2023-01-21 12:45:49.351200: step: 116/526, loss: 0.0042701722122728825 2023-01-21 12:45:50.506633: step: 120/526, loss: 0.023633575066924095 2023-01-21 12:45:51.652049: step: 124/526, loss: 0.04454836994409561 2023-01-21 12:45:52.817699: step: 128/526, loss: 0.1259869933128357 2023-01-21 12:45:53.946014: step: 132/526, loss: 0.016383077949285507 2023-01-21 12:45:55.072719: step: 136/526, loss: 0.0038488388527184725 2023-01-21 12:45:56.223269: step: 140/526, loss: 0.00481073884293437 2023-01-21 12:45:57.350712: step: 144/526, loss: 0.0006084441556595266 2023-01-21 12:45:58.496664: step: 148/526, loss: 0.02895202673971653 2023-01-21 12:45:59.643427: step: 152/526, loss: 0.016843700781464577 2023-01-21 12:46:00.785344: step: 156/526, loss: 0.009022139944136143 2023-01-21 12:46:01.952850: step: 160/526, loss: 0.04529685899615288 2023-01-21 12:46:03.159367: step: 164/526, loss: 0.002326106885448098 2023-01-21 12:46:04.321524: step: 168/526, loss: 0.0005415916675701737 2023-01-21 12:46:05.449230: step: 172/526, loss: 0.009449243545532227 2023-01-21 12:46:06.598561: step: 176/526, loss: 0.01536102220416069 2023-01-21 12:46:07.730126: step: 180/526, loss: 0.0008292198181152344 2023-01-21 12:46:08.876744: step: 184/526, loss: 0.018462562933564186 2023-01-21 12:46:09.992168: step: 188/526, loss: 0.002765655517578125 2023-01-21 12:46:11.102381: step: 192/526, loss: 0.00808868370950222 2023-01-21 12:46:12.251801: step: 196/526, loss: 0.0014797210460528731 2023-01-21 12:46:13.375190: step: 200/526, loss: 0.0010051727294921875 2023-01-21 12:46:14.515187: step: 204/526, loss: 4.444122168933973e-05 2023-01-21 12:46:15.644431: step: 208/526, loss: 0.007484245114028454 2023-01-21 12:46:16.767195: step: 212/526, loss: 0.0037946701049804688 2023-01-21 12:46:17.907252: step: 216/526, loss: 0.013715839944779873 2023-01-21 12:46:19.040151: step: 220/526, loss: 0.007614707574248314 2023-01-21 12:46:20.188136: step: 224/526, loss: 0.023600006476044655 2023-01-21 12:46:21.313656: step: 228/526, loss: 0.0019225120777264237 2023-01-21 12:46:22.465233: step: 232/526, loss: 0.012050151824951172 2023-01-21 12:46:23.592121: step: 236/526, loss: 0.00013294219388626516 2023-01-21 12:46:24.728651: step: 240/526, loss: 0.0009471892844885588 2023-01-21 12:46:25.861821: step: 244/526, loss: 0.00407066335901618 2023-01-21 12:46:27.020354: step: 248/526, loss: 0.0004299163701944053 2023-01-21 12:46:28.149569: step: 252/526, loss: 0.012555313296616077 2023-01-21 12:46:29.271821: step: 256/526, loss: 0.04630317538976669 2023-01-21 12:46:30.384135: step: 260/526, loss: 0.002934694290161133 2023-01-21 12:46:31.505895: step: 264/526, loss: 0.07266712188720703 2023-01-21 12:46:32.663605: step: 268/526, loss: 0.0003631591680459678 2023-01-21 12:46:33.765039: step: 272/526, loss: 0.019245052710175514 2023-01-21 12:46:34.877989: step: 276/526, loss: 0.0011472702026367188 2023-01-21 12:46:36.024713: step: 280/526, loss: 0.0005469322204589844 2023-01-21 12:46:37.165663: step: 284/526, loss: 0.02994556538760662 2023-01-21 12:46:38.327770: step: 288/526, loss: 0.006078099831938744 2023-01-21 12:46:39.478619: step: 292/526, loss: 0.05078592523932457 2023-01-21 12:46:40.599924: step: 296/526, loss: 0.0004914283636026084 2023-01-21 12:46:41.727430: step: 300/526, loss: 0.00041999819222837687 2023-01-21 12:46:42.861659: step: 304/526, loss: 0.0035699845757335424 2023-01-21 12:46:44.018871: step: 308/526, loss: 0.006222724914550781 2023-01-21 12:46:45.156477: step: 312/526, loss: 0.009252166375517845 2023-01-21 12:46:46.308114: step: 316/526, loss: 0.00013742447481490672 2023-01-21 12:46:47.425996: step: 320/526, loss: 0.01678619347512722 2023-01-21 12:46:48.577811: step: 324/526, loss: 5.626678466796875e-05 2023-01-21 12:46:49.711122: step: 328/526, loss: 1.7166139514301904e-06 2023-01-21 12:46:50.816606: step: 332/526, loss: 9.822846186580136e-05 2023-01-21 12:46:51.938239: step: 336/526, loss: 0.00018215179443359375 2023-01-21 12:46:53.093130: step: 340/526, loss: 0.00020189284987282008 2023-01-21 12:46:54.234325: step: 344/526, loss: 0.005839729681611061 2023-01-21 12:46:55.409668: step: 348/526, loss: 0.00021705626568291336 2023-01-21 12:46:56.564379: step: 352/526, loss: 7.629394644936838e-07 2023-01-21 12:46:57.694903: step: 356/526, loss: 0.0013200759422034025 2023-01-21 12:46:58.860689: step: 360/526, loss: 0.010667610913515091 2023-01-21 12:46:59.982152: step: 364/526, loss: 0.007733345031738281 2023-01-21 12:47:01.095513: step: 368/526, loss: 0.024980831891298294 2023-01-21 12:47:02.251322: step: 372/526, loss: 0.0006782531854696572 2023-01-21 12:47:03.399477: step: 376/526, loss: 0.023363113403320312 2023-01-21 12:47:04.532335: step: 380/526, loss: 0.00031976698664948344 2023-01-21 12:47:05.685851: step: 384/526, loss: 0.00013618469529319555 2023-01-21 12:47:06.819681: step: 388/526, loss: 0.019942427054047585 2023-01-21 12:47:07.936952: step: 392/526, loss: 0.0007611275068484247 2023-01-21 12:47:09.077126: step: 396/526, loss: 0.004407405853271484 2023-01-21 12:47:10.199302: step: 400/526, loss: 0.0009386062738485634 2023-01-21 12:47:11.357987: step: 404/526, loss: 3.166198803228326e-05 2023-01-21 12:47:12.499302: step: 408/526, loss: 0.07454881072044373 2023-01-21 12:47:13.633914: step: 412/526, loss: 0.0027969360817223787 2023-01-21 12:47:14.739515: step: 416/526, loss: 0.003816795302554965 2023-01-21 12:47:15.865373: step: 420/526, loss: 0.00043487548828125 2023-01-21 12:47:16.978385: step: 424/526, loss: 0.002858400344848633 2023-01-21 12:47:18.162086: step: 428/526, loss: 3.776550147449598e-05 2023-01-21 12:47:19.293513: step: 432/526, loss: 0.004890060517936945 2023-01-21 12:47:20.431935: step: 436/526, loss: 0.0008033752674236894 2023-01-21 12:47:21.565343: step: 440/526, loss: 0.0015236377948895097 2023-01-21 12:47:22.704803: step: 444/526, loss: 0.023052215576171875 2023-01-21 12:47:23.845028: step: 448/526, loss: 0.026891136541962624 2023-01-21 12:47:24.962490: step: 452/526, loss: 0.0019168853759765625 2023-01-21 12:47:26.138031: step: 456/526, loss: 0.05372047424316406 2023-01-21 12:47:27.272402: step: 460/526, loss: 2.365112231927924e-05 2023-01-21 12:47:28.429931: step: 464/526, loss: 5.3691863286076114e-05 2023-01-21 12:47:29.545882: step: 468/526, loss: 0.0008518219692632556 2023-01-21 12:47:30.667829: step: 472/526, loss: 0.05989604443311691 2023-01-21 12:47:31.798611: step: 476/526, loss: 0.04166469722986221 2023-01-21 12:47:32.957076: step: 480/526, loss: 0.051915835589170456 2023-01-21 12:47:34.079155: step: 484/526, loss: 0.005378532223403454 2023-01-21 12:47:35.203741: step: 488/526, loss: 0.013560676947236061 2023-01-21 12:47:36.339899: step: 492/526, loss: 0.001990890596061945 2023-01-21 12:47:37.501741: step: 496/526, loss: 0.0232830997556448 2023-01-21 12:47:38.632421: step: 500/526, loss: 0.00031909943209029734 2023-01-21 12:47:39.779393: step: 504/526, loss: 0.011682510375976562 2023-01-21 12:47:40.876321: step: 508/526, loss: 0.02809171751141548 2023-01-21 12:47:42.041946: step: 512/526, loss: 0.04712257161736488 2023-01-21 12:47:43.200454: step: 516/526, loss: 0.07752871513366699 2023-01-21 12:47:44.308189: step: 520/526, loss: 0.0003688812430482358 2023-01-21 12:47:45.436817: step: 524/526, loss: 0.0041211131028831005 2023-01-21 12:47:46.561560: step: 528/526, loss: 0.0012222289806231856 2023-01-21 12:47:47.730962: step: 532/526, loss: 0.0576297789812088 2023-01-21 12:47:48.846413: step: 536/526, loss: 0.006798172369599342 2023-01-21 12:47:49.975956: step: 540/526, loss: 0.02004213258624077 2023-01-21 12:47:51.106828: step: 544/526, loss: 0.023140907287597656 2023-01-21 12:47:52.278824: step: 548/526, loss: 0.0010047436226159334 2023-01-21 12:47:53.425705: step: 552/526, loss: 1.068115216185106e-05 2023-01-21 12:47:54.532067: step: 556/526, loss: 0.00013217926607467234 2023-01-21 12:47:55.634518: step: 560/526, loss: 0.00028266909066587687 2023-01-21 12:47:56.786706: step: 564/526, loss: 0.0017215728294104338 2023-01-21 12:47:57.970239: step: 568/526, loss: 0.0007062911754474044 2023-01-21 12:47:59.114588: step: 572/526, loss: 0.00015115737915039062 2023-01-21 12:48:00.231176: step: 576/526, loss: 1.220703143189894e-05 2023-01-21 12:48:01.363902: step: 580/526, loss: -3.280639793956652e-05 2023-01-21 12:48:02.497722: step: 584/526, loss: 0.00160560617223382 2023-01-21 12:48:03.642311: step: 588/526, loss: 0.00043239592923782766 2023-01-21 12:48:04.783725: step: 592/526, loss: 0.017119886353611946 2023-01-21 12:48:05.927665: step: 596/526, loss: 0.0004044532834086567 2023-01-21 12:48:07.041047: step: 600/526, loss: 0.020192433148622513 2023-01-21 12:48:08.153035: step: 604/526, loss: 0.023328544571995735 2023-01-21 12:48:09.263603: step: 608/526, loss: 0.0003105163632426411 2023-01-21 12:48:10.410573: step: 612/526, loss: 0.003456688020378351 2023-01-21 12:48:11.537430: step: 616/526, loss: 0.0010906219249591231 2023-01-21 12:48:12.675339: step: 620/526, loss: 0.0527406707406044 2023-01-21 12:48:13.810094: step: 624/526, loss: 0.004522228147834539 2023-01-21 12:48:14.984600: step: 628/526, loss: 0.021945619955658913 2023-01-21 12:48:16.150784: step: 632/526, loss: 0.0007540702936239541 2023-01-21 12:48:17.251504: step: 636/526, loss: 6.608963303733617e-05 2023-01-21 12:48:18.405933: step: 640/526, loss: 0.014848614111542702 2023-01-21 12:48:19.548753: step: 644/526, loss: 0.0002599715953692794 2023-01-21 12:48:20.685716: step: 648/526, loss: 0.0009558200836181641 2023-01-21 12:48:21.833239: step: 652/526, loss: 0.0023431777954101562 2023-01-21 12:48:22.978755: step: 656/526, loss: 0.00024509429931640625 2023-01-21 12:48:24.136507: step: 660/526, loss: -1.0490416570974048e-06 2023-01-21 12:48:25.295073: step: 664/526, loss: 0.002943706465885043 2023-01-21 12:48:26.449071: step: 668/526, loss: 0.0008965492015704513 2023-01-21 12:48:27.608369: step: 672/526, loss: 0.0004547119315247983 2023-01-21 12:48:28.745391: step: 676/526, loss: 3.166198803228326e-05 2023-01-21 12:48:29.901750: step: 680/526, loss: 0.00245418562553823 2023-01-21 12:48:31.036045: step: 684/526, loss: 0.046572305262088776 2023-01-21 12:48:32.172839: step: 688/526, loss: 0.023625563830137253 2023-01-21 12:48:33.270465: step: 692/526, loss: 0.012465858832001686 2023-01-21 12:48:34.419284: step: 696/526, loss: 0.07640743255615234 2023-01-21 12:48:35.545820: step: 700/526, loss: 0.015327739529311657 2023-01-21 12:48:36.670555: step: 704/526, loss: 0.01274871826171875 2023-01-21 12:48:37.781736: step: 708/526, loss: 8.850097947288305e-05 2023-01-21 12:48:38.927357: step: 712/526, loss: 2.593994213384576e-05 2023-01-21 12:48:40.059349: step: 716/526, loss: 0.04666538164019585 2023-01-21 12:48:41.249164: step: 720/526, loss: 0.0005118369590491056 2023-01-21 12:48:42.396306: step: 724/526, loss: 0.0012772560585290194 2023-01-21 12:48:43.532629: step: 728/526, loss: 0.033873654901981354 2023-01-21 12:48:44.677546: step: 732/526, loss: 0.001488494803197682 2023-01-21 12:48:45.824575: step: 736/526, loss: 0.006680202670395374 2023-01-21 12:48:46.981205: step: 740/526, loss: 0.09278898686170578 2023-01-21 12:48:48.142443: step: 744/526, loss: 0.007106018252670765 2023-01-21 12:48:49.287114: step: 748/526, loss: 0.0015020847786217928 2023-01-21 12:48:50.418582: step: 752/526, loss: 0.008099175058305264 2023-01-21 12:48:51.536565: step: 756/526, loss: 0.02871551550924778 2023-01-21 12:48:52.691466: step: 760/526, loss: 0.05673703923821449 2023-01-21 12:48:53.838900: step: 764/526, loss: 0.0018929005600512028 2023-01-21 12:48:54.966446: step: 768/526, loss: 7.80105619924143e-05 2023-01-21 12:48:56.118123: step: 772/526, loss: 0.0005279541364870965 2023-01-21 12:48:57.272983: step: 776/526, loss: 0.00011672973778331652 2023-01-21 12:48:58.422034: step: 780/526, loss: 0.009870147332549095 2023-01-21 12:48:59.539088: step: 784/526, loss: 0.0007446289528161287 2023-01-21 12:49:00.673761: step: 788/526, loss: 0.0004868507385253906 2023-01-21 12:49:01.826741: step: 792/526, loss: 0.005836391821503639 2023-01-21 12:49:02.985504: step: 796/526, loss: 0.0003986358642578125 2023-01-21 12:49:04.153241: step: 800/526, loss: 0.0921657532453537 2023-01-21 12:49:05.289294: step: 804/526, loss: 3.604888843256049e-05 2023-01-21 12:49:06.444614: step: 808/526, loss: 0.010436106473207474 2023-01-21 12:49:07.587653: step: 812/526, loss: 0.07317447662353516 2023-01-21 12:49:08.724575: step: 816/526, loss: 2.4700166250113398e-05 2023-01-21 12:49:09.846565: step: 820/526, loss: 0.0016826629871502519 2023-01-21 12:49:11.006079: step: 824/526, loss: 0.0030189515091478825 2023-01-21 12:49:12.135776: step: 828/526, loss: 1.926422191900201e-05 2023-01-21 12:49:13.270182: step: 832/526, loss: 0.012021971866488457 2023-01-21 12:49:14.377305: step: 836/526, loss: 0.004403495695441961 2023-01-21 12:49:15.501823: step: 840/526, loss: 1.316070574830519e-05 2023-01-21 12:49:16.618450: step: 844/526, loss: 0.0027968406211584806 2023-01-21 12:49:17.737575: step: 848/526, loss: 0.00020885467529296875 2023-01-21 12:49:18.863110: step: 852/526, loss: 0.0009811401832848787 2023-01-21 12:49:20.009342: step: 856/526, loss: -1.182556115963962e-05 2023-01-21 12:49:21.134047: step: 860/526, loss: 0.007063866127282381 2023-01-21 12:49:22.263725: step: 864/526, loss: 2.231597864010837e-05 2023-01-21 12:49:23.397136: step: 868/526, loss: 0.008090972900390625 2023-01-21 12:49:24.539219: step: 872/526, loss: 0.05535387992858887 2023-01-21 12:49:25.675425: step: 876/526, loss: 0.000846481358166784 2023-01-21 12:49:26.815766: step: 880/526, loss: 0.004799842834472656 2023-01-21 12:49:27.957117: step: 884/526, loss: 0.0028884888160973787 2023-01-21 12:49:29.087804: step: 888/526, loss: 0.002346038818359375 2023-01-21 12:49:30.200631: step: 892/526, loss: 0.0571783073246479 2023-01-21 12:49:31.335095: step: 896/526, loss: 0.00010061264038085938 2023-01-21 12:49:32.456824: step: 900/526, loss: 0.005844783969223499 2023-01-21 12:49:33.572778: step: 904/526, loss: 2.5272368020523572e-06 2023-01-21 12:49:34.696090: step: 908/526, loss: 0.029066849499940872 2023-01-21 12:49:35.822601: step: 912/526, loss: 0.03276043012738228 2023-01-21 12:49:36.961264: step: 916/526, loss: 0.0020089149475097656 2023-01-21 12:49:38.120712: step: 920/526, loss: 0.00017299651517532766 2023-01-21 12:49:39.291374: step: 924/526, loss: 1.430511474609375e-06 2023-01-21 12:49:40.444371: step: 928/526, loss: 0.01763172261416912 2023-01-21 12:49:41.613873: step: 932/526, loss: 0.02422161214053631 2023-01-21 12:49:42.746715: step: 936/526, loss: 0.07827749848365784 2023-01-21 12:49:43.882468: step: 940/526, loss: 0.006887365132570267 2023-01-21 12:49:44.999658: step: 944/526, loss: 4.234314110362902e-05 2023-01-21 12:49:46.105935: step: 948/526, loss: 0.0005025863647460938 2023-01-21 12:49:47.230748: step: 952/526, loss: 0.022162437438964844 2023-01-21 12:49:48.399977: step: 956/526, loss: 0.011558962054550648 2023-01-21 12:49:49.529422: step: 960/526, loss: -5.34057608092553e-06 2023-01-21 12:49:50.645450: step: 964/526, loss: 0.006871223449707031 2023-01-21 12:49:51.779474: step: 968/526, loss: 0.31347334384918213 2023-01-21 12:49:52.898798: step: 972/526, loss: 0.002418327145278454 2023-01-21 12:49:53.989759: step: 976/526, loss: 0.0009500504238530993 2023-01-21 12:49:55.139081: step: 980/526, loss: 0.00171241769567132 2023-01-21 12:49:56.282169: step: 984/526, loss: 0.0006884574540890753 2023-01-21 12:49:57.456098: step: 988/526, loss: 0.0013051987625658512 2023-01-21 12:49:58.610717: step: 992/526, loss: 0.005212021060287952 2023-01-21 12:49:59.718379: step: 996/526, loss: 0.003074789186939597 2023-01-21 12:50:00.870774: step: 1000/526, loss: 0.02366046980023384 2023-01-21 12:50:02.016923: step: 1004/526, loss: 0.029489515349268913 2023-01-21 12:50:03.137788: step: 1008/526, loss: 8.535384949936997e-06 2023-01-21 12:50:04.255363: step: 1012/526, loss: 0.010569954290986061 2023-01-21 12:50:05.395859: step: 1016/526, loss: 0.009633922018110752 2023-01-21 12:50:06.523085: step: 1020/526, loss: 0.005270099733024836 2023-01-21 12:50:07.675329: step: 1024/526, loss: 0.03340597450733185 2023-01-21 12:50:08.803094: step: 1028/526, loss: 2.6226043701171875e-05 2023-01-21 12:50:09.943362: step: 1032/526, loss: 0.0009810447227209806 2023-01-21 12:50:11.101821: step: 1036/526, loss: 0.3842960298061371 2023-01-21 12:50:12.283672: step: 1040/526, loss: 0.006016159430146217 2023-01-21 12:50:13.435949: step: 1044/526, loss: 8.125305612338707e-05 2023-01-21 12:50:14.577566: step: 1048/526, loss: 0.0012985230423510075 2023-01-21 12:50:15.719454: step: 1052/526, loss: 0.0002464294375386089 2023-01-21 12:50:16.856154: step: 1056/526, loss: 0.004669857211410999 2023-01-21 12:50:17.979163: step: 1060/526, loss: 0.0007157326326705515 2023-01-21 12:50:19.122430: step: 1064/526, loss: 0.03144645690917969 2023-01-21 12:50:20.239846: step: 1068/526, loss: 0.011163617484271526 2023-01-21 12:50:21.395737: step: 1072/526, loss: 0.04239053651690483 2023-01-21 12:50:22.527864: step: 1076/526, loss: 0.03916988521814346 2023-01-21 12:50:23.674171: step: 1080/526, loss: 0.0012094498379155993 2023-01-21 12:50:24.799344: step: 1084/526, loss: 5.760193016612902e-05 2023-01-21 12:50:25.926229: step: 1088/526, loss: 0.00014019012451171875 2023-01-21 12:50:27.026661: step: 1092/526, loss: 3.8480757211800665e-05 2023-01-21 12:50:28.162100: step: 1096/526, loss: 0.0001485824614064768 2023-01-21 12:50:29.292078: step: 1100/526, loss: 9.202957153320312e-05 2023-01-21 12:50:30.405897: step: 1104/526, loss: 0.010206985287368298 2023-01-21 12:50:31.521853: step: 1108/526, loss: 0.001665306044742465 2023-01-21 12:50:32.665987: step: 1112/526, loss: 3.261566234868951e-05 2023-01-21 12:50:33.809432: step: 1116/526, loss: 9.369850886287168e-06 2023-01-21 12:50:34.947640: step: 1120/526, loss: 0.050013162195682526 2023-01-21 12:50:36.076095: step: 1124/526, loss: 0.05823250114917755 2023-01-21 12:50:37.213115: step: 1128/526, loss: 0.02266078069806099 2023-01-21 12:50:38.328598: step: 1132/526, loss: 0.0002397537318756804 2023-01-21 12:50:39.447528: step: 1136/526, loss: 0.0002952575741801411 2023-01-21 12:50:40.589082: step: 1140/526, loss: 0.00010452270362293348 2023-01-21 12:50:41.719801: step: 1144/526, loss: 1.0395049685030244e-05 2023-01-21 12:50:42.893706: step: 1148/526, loss: 0.00023469925508834422 2023-01-21 12:50:44.031531: step: 1152/526, loss: 0.00027751922607421875 2023-01-21 12:50:45.126386: step: 1156/526, loss: 6.67572021484375e-06 2023-01-21 12:50:46.261315: step: 1160/526, loss: 0.003107452532276511 2023-01-21 12:50:47.401090: step: 1164/526, loss: 0.0003467559872660786 2023-01-21 12:50:48.554926: step: 1168/526, loss: 0.0020801543723791838 2023-01-21 12:50:49.726481: step: 1172/526, loss: 0.00515360850840807 2023-01-21 12:50:50.879793: step: 1176/526, loss: 0.002410888671875 2023-01-21 12:50:51.990285: step: 1180/526, loss: 2.3555756342830136e-05 2023-01-21 12:50:53.145579: step: 1184/526, loss: 1.9073486328125e-05 2023-01-21 12:50:54.284660: step: 1188/526, loss: 0.00010356902930652723 2023-01-21 12:50:55.419042: step: 1192/526, loss: 0.0003186225949320942 2023-01-21 12:50:56.562512: step: 1196/526, loss: 0.005884647369384766 2023-01-21 12:50:57.701373: step: 1200/526, loss: 6.008148193359375e-05 2023-01-21 12:50:58.852926: step: 1204/526, loss: 4.00543212890625e-05 2023-01-21 12:51:00.014491: step: 1208/526, loss: 0.004184532444924116 2023-01-21 12:51:01.143827: step: 1212/526, loss: 0.004657554440200329 2023-01-21 12:51:02.276210: step: 1216/526, loss: 0.0033084871247410774 2023-01-21 12:51:03.425559: step: 1220/526, loss: 0.0008123398292809725 2023-01-21 12:51:04.592797: step: 1224/526, loss: 0.00013704299635719508 2023-01-21 12:51:05.729738: step: 1228/526, loss: 0.0007137298816815019 2023-01-21 12:51:06.858751: step: 1232/526, loss: 0.0443572998046875 2023-01-21 12:51:07.958252: step: 1236/526, loss: 0.00747184781357646 2023-01-21 12:51:09.128702: step: 1240/526, loss: 0.012370300479233265 2023-01-21 12:51:10.241163: step: 1244/526, loss: 0.07369355857372284 2023-01-21 12:51:11.363311: step: 1248/526, loss: 0.030431175604462624 2023-01-21 12:51:12.516492: step: 1252/526, loss: 5.035400317865424e-05 2023-01-21 12:51:13.737909: step: 1256/526, loss: 0.009019088931381702 2023-01-21 12:51:14.881699: step: 1260/526, loss: 0.0035072327591478825 2023-01-21 12:51:16.006749: step: 1264/526, loss: 0.0009922027820721269 2023-01-21 12:51:17.131650: step: 1268/526, loss: 0.01670055463910103 2023-01-21 12:51:18.268399: step: 1272/526, loss: 0.0004646301385946572 2023-01-21 12:51:19.418965: step: 1276/526, loss: 0.02537689357995987 2023-01-21 12:51:20.554012: step: 1280/526, loss: 0.0001780986785888672 2023-01-21 12:51:21.683282: step: 1284/526, loss: 0.0008717536693438888 2023-01-21 12:51:22.811152: step: 1288/526, loss: 0.010530662722885609 2023-01-21 12:51:23.991773: step: 1292/526, loss: 0.0009741783142089844 2023-01-21 12:51:25.121439: step: 1296/526, loss: 0.03679867088794708 2023-01-21 12:51:26.262281: step: 1300/526, loss: 0.003546428866684437 2023-01-21 12:51:27.382353: step: 1304/526, loss: 7.667541649425402e-05 2023-01-21 12:51:28.514032: step: 1308/526, loss: 0.039972878992557526 2023-01-21 12:51:29.672684: step: 1312/526, loss: 6.182193465065211e-05 2023-01-21 12:51:30.825566: step: 1316/526, loss: 0.0005231857649050653 2023-01-21 12:51:31.978719: step: 1320/526, loss: 0.11816863715648651 2023-01-21 12:51:33.113077: step: 1324/526, loss: 0.14015941321849823 2023-01-21 12:51:34.234357: step: 1328/526, loss: 0.006325960624963045 2023-01-21 12:51:35.400895: step: 1332/526, loss: 0.0002244949428131804 2023-01-21 12:51:36.521931: step: 1336/526, loss: 0.0004566192801576108 2023-01-21 12:51:37.662085: step: 1340/526, loss: 0.009648323059082031 2023-01-21 12:51:38.816130: step: 1344/526, loss: 0.00038394928560592234 2023-01-21 12:51:39.934966: step: 1348/526, loss: 0.0013529777061194181 2023-01-21 12:51:41.091703: step: 1352/526, loss: 0.07061338424682617 2023-01-21 12:51:42.249035: step: 1356/526, loss: 0.007945633493363857 2023-01-21 12:51:43.396333: step: 1360/526, loss: 2.5415420168428682e-05 2023-01-21 12:51:44.510712: step: 1364/526, loss: 0.0037233352195471525 2023-01-21 12:51:45.675253: step: 1368/526, loss: 0.033083345741033554 2023-01-21 12:51:46.810112: step: 1372/526, loss: 0.001010894775390625 2023-01-21 12:51:47.944441: step: 1376/526, loss: 0.00030393601628020406 2023-01-21 12:51:49.086028: step: 1380/526, loss: 4.806518700206652e-05 2023-01-21 12:51:50.240946: step: 1384/526, loss: 0.0007904052617959678 2023-01-21 12:51:51.376705: step: 1388/526, loss: 0.06448288261890411 2023-01-21 12:51:52.508483: step: 1392/526, loss: 0.008920097723603249 2023-01-21 12:51:53.623625: step: 1396/526, loss: 0.001951885293237865 2023-01-21 12:51:54.742184: step: 1400/526, loss: 0.0010952949523925781 2023-01-21 12:51:55.895881: step: 1404/526, loss: 7.45773286325857e-05 2023-01-21 12:51:56.991423: step: 1408/526, loss: 0.012462139129638672 2023-01-21 12:51:58.183530: step: 1412/526, loss: 0.08644609153270721 2023-01-21 12:51:59.332443: step: 1416/526, loss: 0.00010337829735362902 2023-01-21 12:52:00.477515: step: 1420/526, loss: 0.05374488607048988 2023-01-21 12:52:01.594056: step: 1424/526, loss: 0.0050987242721021175 2023-01-21 12:52:02.696880: step: 1428/526, loss: 0.043480873107910156 2023-01-21 12:52:03.843009: step: 1432/526, loss: 0.0016631126636639237 2023-01-21 12:52:04.988145: step: 1436/526, loss: 0.0013042449718341231 2023-01-21 12:52:06.131081: step: 1440/526, loss: 0.0020846365951001644 2023-01-21 12:52:07.261800: step: 1444/526, loss: 0.06253280490636826 2023-01-21 12:52:08.432737: step: 1448/526, loss: 0.5543445944786072 2023-01-21 12:52:09.577935: step: 1452/526, loss: 0.001891326974146068 2023-01-21 12:52:10.707720: step: 1456/526, loss: 0.0017368317348882556 2023-01-21 12:52:11.859721: step: 1460/526, loss: 8.077621168922633e-05 2023-01-21 12:52:12.963175: step: 1464/526, loss: 0.0007710456848144531 2023-01-21 12:52:14.085038: step: 1468/526, loss: 0.00141315464861691 2023-01-21 12:52:15.247470: step: 1472/526, loss: 0.017116069793701172 2023-01-21 12:52:16.364796: step: 1476/526, loss: 0.00010776519775390625 2023-01-21 12:52:17.487945: step: 1480/526, loss: 6.28471389063634e-05 2023-01-21 12:52:18.638693: step: 1484/526, loss: 0.014305973425507545 2023-01-21 12:52:19.795012: step: 1488/526, loss: 0.0008440017700195312 2023-01-21 12:52:20.938369: step: 1492/526, loss: 0.052460383623838425 2023-01-21 12:52:22.052999: step: 1496/526, loss: 0.017572879791259766 2023-01-21 12:52:23.187908: step: 1500/526, loss: 0.009979343973100185 2023-01-21 12:52:24.324635: step: 1504/526, loss: 3.051757857974735e-06 2023-01-21 12:52:25.429955: step: 1508/526, loss: 0.001967143965885043 2023-01-21 12:52:26.563557: step: 1512/526, loss: 0.02408142015337944 2023-01-21 12:52:27.689123: step: 1516/526, loss: 0.00844874419271946 2023-01-21 12:52:28.812522: step: 1520/526, loss: 0.0014387130504474044 2023-01-21 12:52:29.967672: step: 1524/526, loss: 0.009016323834657669 2023-01-21 12:52:31.100764: step: 1528/526, loss: 0.007923698052763939 2023-01-21 12:52:32.240373: step: 1532/526, loss: 0.0007848739624023438 2023-01-21 12:52:33.427620: step: 1536/526, loss: 0.014157581143081188 2023-01-21 12:52:34.567285: step: 1540/526, loss: 0.5811004042625427 2023-01-21 12:52:35.690606: step: 1544/526, loss: 0.01493148785084486 2023-01-21 12:52:36.814176: step: 1548/526, loss: 0.049453262239694595 2023-01-21 12:52:37.920139: step: 1552/526, loss: 0.00040273668128065765 2023-01-21 12:52:39.068521: step: 1556/526, loss: 0.0008309364202432334 2023-01-21 12:52:40.184183: step: 1560/526, loss: 0.00013446809316519648 2023-01-21 12:52:41.322291: step: 1564/526, loss: 0.06089053303003311 2023-01-21 12:52:42.465059: step: 1568/526, loss: 0.017856597900390625 2023-01-21 12:52:43.586490: step: 1572/526, loss: 1.8024446035269648e-05 2023-01-21 12:52:44.737986: step: 1576/526, loss: 0.006492328830063343 2023-01-21 12:52:45.874479: step: 1580/526, loss: 0.0007374763954430819 2023-01-21 12:52:47.017634: step: 1584/526, loss: 0.039803504943847656 2023-01-21 12:52:48.142266: step: 1588/526, loss: 7.133484177757055e-05 2023-01-21 12:52:49.265213: step: 1592/526, loss: 0.003092861268669367 2023-01-21 12:52:50.412539: step: 1596/526, loss: 0.0014581680297851562 2023-01-21 12:52:51.533903: step: 1600/526, loss: 7.820130122127011e-06 2023-01-21 12:52:52.645022: step: 1604/526, loss: 0.005928420927375555 2023-01-21 12:52:53.792213: step: 1608/526, loss: 0.010188293643295765 2023-01-21 12:52:54.921494: step: 1612/526, loss: 5.9700014389818534e-05 2023-01-21 12:52:56.066447: step: 1616/526, loss: 0.00165901193395257 2023-01-21 12:52:57.207237: step: 1620/526, loss: 0.048157691955566406 2023-01-21 12:52:58.347673: step: 1624/526, loss: 0.04464760050177574 2023-01-21 12:52:59.462262: step: 1628/526, loss: 0.004692935850471258 2023-01-21 12:53:00.571910: step: 1632/526, loss: 1.23977656585339e-06 2023-01-21 12:53:01.721289: step: 1636/526, loss: 0.009289741516113281 2023-01-21 12:53:02.860196: step: 1640/526, loss: 0.015642547979950905 2023-01-21 12:53:03.998092: step: 1644/526, loss: 0.0037658691871911287 2023-01-21 12:53:05.108870: step: 1648/526, loss: 0.01901531219482422 2023-01-21 12:53:06.217185: step: 1652/526, loss: 1.373290979245212e-05 2023-01-21 12:53:07.336628: step: 1656/526, loss: 0.00691752415150404 2023-01-21 12:53:08.448569: step: 1660/526, loss: 0.06677704304456711 2023-01-21 12:53:09.563629: step: 1664/526, loss: 0.00026683806208893657 2023-01-21 12:53:10.702294: step: 1668/526, loss: 0.00016689300537109375 2023-01-21 12:53:11.791369: step: 1672/526, loss: 0.00023756027803756297 2023-01-21 12:53:12.917866: step: 1676/526, loss: 0.00021667480177711695 2023-01-21 12:53:14.064494: step: 1680/526, loss: 0.004671812057495117 2023-01-21 12:53:15.214233: step: 1684/526, loss: 0.0045506954193115234 2023-01-21 12:53:16.380114: step: 1688/526, loss: 5.5599211918888614e-05 2023-01-21 12:53:17.542097: step: 1692/526, loss: 0.00588569650426507 2023-01-21 12:53:18.662510: step: 1696/526, loss: 0.01765899732708931 2023-01-21 12:53:19.804699: step: 1700/526, loss: 0.008499765768647194 2023-01-21 12:53:20.932132: step: 1704/526, loss: 0.0036149502266198397 2023-01-21 12:53:22.084261: step: 1708/526, loss: 0.00021781922259833664 2023-01-21 12:53:23.233798: step: 1712/526, loss: 2.4032591682043858e-05 2023-01-21 12:53:24.377414: step: 1716/526, loss: 0.013953781686723232 2023-01-21 12:53:25.512867: step: 1720/526, loss: 0.10758963227272034 2023-01-21 12:53:26.647345: step: 1724/526, loss: 0.000690460205078125 2023-01-21 12:53:27.813682: step: 1728/526, loss: 0.0019339561695232987 2023-01-21 12:53:28.965575: step: 1732/526, loss: 0.00045766832772642374 2023-01-21 12:53:30.094927: step: 1736/526, loss: 0.001689481781795621 2023-01-21 12:53:31.275337: step: 1740/526, loss: 2.8133392333984375e-05 2023-01-21 12:53:32.431111: step: 1744/526, loss: 0.04057254642248154 2023-01-21 12:53:33.601629: step: 1748/526, loss: 0.004868698306381702 2023-01-21 12:53:34.722190: step: 1752/526, loss: 0.0002922535059042275 2023-01-21 12:53:35.860028: step: 1756/526, loss: 0.0002534866507630795 2023-01-21 12:53:36.974880: step: 1760/526, loss: 0.016299057751893997 2023-01-21 12:53:38.113411: step: 1764/526, loss: 0.014888573437929153 2023-01-21 12:53:39.252233: step: 1768/526, loss: 0.01587958261370659 2023-01-21 12:53:40.404572: step: 1772/526, loss: 0.009452628903090954 2023-01-21 12:53:41.525933: step: 1776/526, loss: 0.0019157410133630037 2023-01-21 12:53:42.627549: step: 1780/526, loss: 0.03770046308636665 2023-01-21 12:53:43.761937: step: 1784/526, loss: 0.00021038056001998484 2023-01-21 12:53:44.868640: step: 1788/526, loss: 0.0023446083068847656 2023-01-21 12:53:45.982997: step: 1792/526, loss: 0.008762932382524014 2023-01-21 12:53:47.135806: step: 1796/526, loss: 0.00503196706995368 2023-01-21 12:53:48.322256: step: 1800/526, loss: 0.02203540876507759 2023-01-21 12:53:49.469549: step: 1804/526, loss: 0.002111339708790183 2023-01-21 12:53:50.584003: step: 1808/526, loss: 0.0003903389151673764 2023-01-21 12:53:51.740249: step: 1812/526, loss: 0.04990902170538902 2023-01-21 12:53:52.875482: step: 1816/526, loss: 0.24043616652488708 2023-01-21 12:53:54.043987: step: 1820/526, loss: 0.01688213460147381 2023-01-21 12:53:55.179692: step: 1824/526, loss: 0.022153044119477272 2023-01-21 12:53:56.329114: step: 1828/526, loss: 0.00613174494355917 2023-01-21 12:53:57.444943: step: 1832/526, loss: 0.0007495880126953125 2023-01-21 12:53:58.576058: step: 1836/526, loss: 0.001712036202661693 2023-01-21 12:53:59.749750: step: 1840/526, loss: 0.004852008540183306 2023-01-21 12:54:00.892093: step: 1844/526, loss: 4.520416405284777e-05 2023-01-21 12:54:02.027085: step: 1848/526, loss: 1.9955636162194423e-05 2023-01-21 12:54:03.155226: step: 1852/526, loss: 2.079009937006049e-05 2023-01-21 12:54:04.287615: step: 1856/526, loss: 7.591247413074598e-05 2023-01-21 12:54:05.472703: step: 1860/526, loss: 0.03272972255945206 2023-01-21 12:54:06.579768: step: 1864/526, loss: 6.8664553509734105e-06 2023-01-21 12:54:07.722401: step: 1868/526, loss: 0.00040216444176621735 2023-01-21 12:54:08.898157: step: 1872/526, loss: 0.14620837569236755 2023-01-21 12:54:10.048522: step: 1876/526, loss: 0.0017440796364098787 2023-01-21 12:54:11.187976: step: 1880/526, loss: 1.811981201171875e-05 2023-01-21 12:54:12.307405: step: 1884/526, loss: 0.0002244949428131804 2023-01-21 12:54:13.467861: step: 1888/526, loss: 0.006790351588279009 2023-01-21 12:54:14.572422: step: 1892/526, loss: 0.000678634678479284 2023-01-21 12:54:15.697875: step: 1896/526, loss: 0.0024310590233653784 2023-01-21 12:54:16.838126: step: 1900/526, loss: 0.03702693060040474 2023-01-21 12:54:17.961918: step: 1904/526, loss: 0.11375927925109863 2023-01-21 12:54:19.090623: step: 1908/526, loss: 0.009799766354262829 2023-01-21 12:54:20.259172: step: 1912/526, loss: 0.02885112725198269 2023-01-21 12:54:21.426211: step: 1916/526, loss: 0.0016406059730798006 2023-01-21 12:54:22.560933: step: 1920/526, loss: 5.474090721691027e-05 2023-01-21 12:54:23.688635: step: 1924/526, loss: 0.002544498536735773 2023-01-21 12:54:24.820294: step: 1928/526, loss: 0.003158187959343195 2023-01-21 12:54:25.961957: step: 1932/526, loss: 0.0005628585931845009 2023-01-21 12:54:27.098603: step: 1936/526, loss: 0.009442901238799095 2023-01-21 12:54:28.211577: step: 1940/526, loss: 0.00022220613027457148 2023-01-21 12:54:29.364509: step: 1944/526, loss: 0.00058155064471066 2023-01-21 12:54:30.499845: step: 1948/526, loss: 0.012954997830092907 2023-01-21 12:54:31.666407: step: 1952/526, loss: -2.155303991457913e-05 2023-01-21 12:54:32.817430: step: 1956/526, loss: 0.0025014879647642374 2023-01-21 12:54:33.951290: step: 1960/526, loss: 0.014387416653335094 2023-01-21 12:54:35.099713: step: 1964/526, loss: 0.00043506623478606343 2023-01-21 12:54:36.242542: step: 1968/526, loss: 0.0001312255917582661 2023-01-21 12:54:37.351285: step: 1972/526, loss: 0.0005878448719158769 2023-01-21 12:54:38.482786: step: 1976/526, loss: 0.0018257141346111894 2023-01-21 12:54:39.651194: step: 1980/526, loss: 7.114410254871473e-05 2023-01-21 12:54:40.800210: step: 1984/526, loss: 0.002582073211669922 2023-01-21 12:54:41.924603: step: 1988/526, loss: 0.0004734039248432964 2023-01-21 12:54:43.093714: step: 1992/526, loss: 0.019530678167939186 2023-01-21 12:54:44.202243: step: 1996/526, loss: 0.00031156541081145406 2023-01-21 12:54:45.355239: step: 2000/526, loss: 1.1444090887380298e-05 2023-01-21 12:54:46.477586: step: 2004/526, loss: 0.0009313584305346012 2023-01-21 12:54:47.637157: step: 2008/526, loss: 0.05644664913415909 2023-01-21 12:54:48.787820: step: 2012/526, loss: 0.004493999760597944 2023-01-21 12:54:49.922577: step: 2016/526, loss: 2.403259350103326e-05 2023-01-21 12:54:51.044181: step: 2020/526, loss: 0.0026079178787767887 2023-01-21 12:54:52.198734: step: 2024/526, loss: 0.025112152099609375 2023-01-21 12:54:53.323832: step: 2028/526, loss: 0.021279573440551758 2023-01-21 12:54:54.452429: step: 2032/526, loss: 0.00015411376080010086 2023-01-21 12:54:55.598260: step: 2036/526, loss: 0.0014522551791742444 2023-01-21 12:54:56.735941: step: 2040/526, loss: 0.00020856858463957906 2023-01-21 12:54:57.865020: step: 2044/526, loss: 0.000896072422619909 2023-01-21 12:54:58.988541: step: 2048/526, loss: 0.0009895324474200606 2023-01-21 12:55:00.112376: step: 2052/526, loss: 0.2350166290998459 2023-01-21 12:55:01.245488: step: 2056/526, loss: 0.007348204031586647 2023-01-21 12:55:02.381621: step: 2060/526, loss: 3.271103196311742e-05 2023-01-21 12:55:03.541596: step: 2064/526, loss: 0.014737558551132679 2023-01-21 12:55:04.733990: step: 2068/526, loss: 0.0003450393851380795 2023-01-21 12:55:05.902071: step: 2072/526, loss: 0.002077293349429965 2023-01-21 12:55:07.068057: step: 2076/526, loss: 0.00012512208195403218 2023-01-21 12:55:08.195062: step: 2080/526, loss: 0.8806449174880981 2023-01-21 12:55:09.321755: step: 2084/526, loss: 0.016659926623106003 2023-01-21 12:55:10.462284: step: 2088/526, loss: 0.008400822058320045 2023-01-21 12:55:11.605266: step: 2092/526, loss: 2.8896331059513614e-05 2023-01-21 12:55:12.752482: step: 2096/526, loss: 0.0043472289107739925 2023-01-21 12:55:13.872379: step: 2100/526, loss: 0.002697181887924671 2023-01-21 12:55:14.979890: step: 2104/526, loss: 0.0050490377470850945 ================================================== Loss: 0.018 -------------------- Dev: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Test: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Korean: {'event': {'p': 0.75, 'r': 0.47619047619047616, 'f1': 0.58252427184466}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} New best chinese model... New best russian model... ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 24 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 12:56:16.886597: step: 4/526, loss: 0.0003543853817973286 2023-01-21 12:56:18.013018: step: 8/526, loss: 0.0016130446456372738 2023-01-21 12:56:19.155483: step: 12/526, loss: 7.41958647267893e-05 2023-01-21 12:56:20.272327: step: 16/526, loss: 0.002418422605842352 2023-01-21 12:56:21.406701: step: 20/526, loss: 0.0007433891878463328 2023-01-21 12:56:22.515709: step: 24/526, loss: 0.006683063693344593 2023-01-21 12:56:23.658742: step: 28/526, loss: 1.087188684323337e-05 2023-01-21 12:56:24.811879: step: 32/526, loss: 0.00281944265589118 2023-01-21 12:56:25.928074: step: 36/526, loss: 1.7929078239831142e-05 2023-01-21 12:56:27.051947: step: 40/526, loss: 0.0001478672056691721 2023-01-21 12:56:28.195522: step: 44/526, loss: 0.06127338483929634 2023-01-21 12:56:29.347321: step: 48/526, loss: 0.04185633733868599 2023-01-21 12:56:30.476048: step: 52/526, loss: 8.39233416627394e-06 2023-01-21 12:56:31.592532: step: 56/526, loss: 0.03674716874957085 2023-01-21 12:56:32.722254: step: 60/526, loss: 0.002529716584831476 2023-01-21 12:56:33.882830: step: 64/526, loss: -1.0490416570974048e-06 2023-01-21 12:56:34.987755: step: 68/526, loss: 0.00017271040997002274 2023-01-21 12:56:36.111631: step: 72/526, loss: 0.0007569313165731728 2023-01-21 12:56:37.230384: step: 76/526, loss: 0.0017543792491778731 2023-01-21 12:56:38.381235: step: 80/526, loss: 0.0019897460006177425 2023-01-21 12:56:39.513633: step: 84/526, loss: 0.0011582374572753906 2023-01-21 12:56:40.664881: step: 88/526, loss: 0.013387167826294899 2023-01-21 12:56:41.779197: step: 92/526, loss: 0.028180886059999466 2023-01-21 12:56:42.911293: step: 96/526, loss: 0.005329036619514227 2023-01-21 12:56:44.036573: step: 100/526, loss: 0.009954262524843216 2023-01-21 12:56:45.177277: step: 104/526, loss: 0.002077818149700761 2023-01-21 12:56:46.320164: step: 108/526, loss: 0.005972862243652344 2023-01-21 12:56:47.460051: step: 112/526, loss: 0.011602973565459251 2023-01-21 12:56:48.589782: step: 116/526, loss: 0.04768562316894531 2023-01-21 12:56:49.762849: step: 120/526, loss: 0.0004989624139852822 2023-01-21 12:56:50.886305: step: 124/526, loss: 3.719329924933845e-06 2023-01-21 12:56:52.014768: step: 128/526, loss: 0.00041828156099654734 2023-01-21 12:56:53.118720: step: 132/526, loss: 0.0011971474159508944 2023-01-21 12:56:54.258521: step: 136/526, loss: 0.00016460419283248484 2023-01-21 12:56:55.403642: step: 140/526, loss: 1.8024446035269648e-05 2023-01-21 12:56:56.540279: step: 144/526, loss: 0.001442480250261724 2023-01-21 12:56:57.671711: step: 148/526, loss: 9.784699068404734e-05 2023-01-21 12:56:58.816095: step: 152/526, loss: -5.385280019254424e-06 2023-01-21 12:56:59.961943: step: 156/526, loss: 0.004082011990249157 2023-01-21 12:57:01.062858: step: 160/526, loss: 0.00688095111399889 2023-01-21 12:57:02.191940: step: 164/526, loss: 0.00023670197697356343 2023-01-21 12:57:03.316540: step: 168/526, loss: 0.011916732415556908 2023-01-21 12:57:04.440059: step: 172/526, loss: 5.745887756347656e-05 2023-01-21 12:57:05.579076: step: 176/526, loss: 1.1730194273695815e-05 2023-01-21 12:57:06.711592: step: 180/526, loss: 0.0007794380653649569 2023-01-21 12:57:07.848515: step: 184/526, loss: 0.0006580352783203125 2023-01-21 12:57:09.017186: step: 188/526, loss: 1.773834264895413e-05 2023-01-21 12:57:10.200231: step: 192/526, loss: 0.00013456345186568797 2023-01-21 12:57:11.344935: step: 196/526, loss: 0.018192196264863014 2023-01-21 12:57:12.468652: step: 200/526, loss: 0.00015716553025413305 2023-01-21 12:57:13.618006: step: 204/526, loss: 1.1110305422334932e-05 2023-01-21 12:57:14.779050: step: 208/526, loss: 0.03754596784710884 2023-01-21 12:57:15.907706: step: 212/526, loss: 0.056490231305360794 2023-01-21 12:57:17.031117: step: 216/526, loss: 0.016968728974461555 2023-01-21 12:57:18.171323: step: 220/526, loss: 8.578300912631676e-05 2023-01-21 12:57:19.308963: step: 224/526, loss: 0.000798130058683455 2023-01-21 12:57:20.442772: step: 228/526, loss: 1.4495850336970761e-05 2023-01-21 12:57:21.589739: step: 232/526, loss: 0.0005743026849813759 2023-01-21 12:57:22.746981: step: 236/526, loss: 0.00010976791963912547 2023-01-21 12:57:23.883351: step: 240/526, loss: 5.507469177246094e-05 2023-01-21 12:57:25.041199: step: 244/526, loss: 0.034583281725645065 2023-01-21 12:57:26.170470: step: 248/526, loss: 0.0036297799088060856 2023-01-21 12:57:27.324291: step: 252/526, loss: 0.05945110321044922 2023-01-21 12:57:28.441827: step: 256/526, loss: 2.8610236313397763e-07 2023-01-21 12:57:29.572128: step: 260/526, loss: 0.00013637542724609375 2023-01-21 12:57:30.711240: step: 264/526, loss: 0.024252891540527344 2023-01-21 12:57:31.841706: step: 268/526, loss: 1.926422191900201e-05 2023-01-21 12:57:32.986108: step: 272/526, loss: 0.009486389346420765 2023-01-21 12:57:34.104296: step: 276/526, loss: 0.007299995981156826 2023-01-21 12:57:35.221619: step: 280/526, loss: 0.0032514571212232113 2023-01-21 12:57:36.333600: step: 284/526, loss: 0.02510061301290989 2023-01-21 12:57:37.474198: step: 288/526, loss: 0.08238621056079865 2023-01-21 12:57:38.647420: step: 292/526, loss: 0.00023784636869095266 2023-01-21 12:57:39.801785: step: 296/526, loss: 0.0066242218017578125 2023-01-21 12:57:40.947272: step: 300/526, loss: 0.0002933502255473286 2023-01-21 12:57:42.092564: step: 304/526, loss: 0.0032695771660655737 2023-01-21 12:57:43.222985: step: 308/526, loss: 0.0004551887686830014 2023-01-21 12:57:44.372180: step: 312/526, loss: 0.0027387142181396484 2023-01-21 12:57:45.490962: step: 316/526, loss: 0.00040121079655364156 2023-01-21 12:57:46.628140: step: 320/526, loss: 0.026807021349668503 2023-01-21 12:57:47.732917: step: 324/526, loss: 0.0006674290052615106 2023-01-21 12:57:48.889660: step: 328/526, loss: 0.00032258033752441406 2023-01-21 12:57:50.003938: step: 332/526, loss: 0.0767589583992958 2023-01-21 12:57:51.148311: step: 336/526, loss: 0.0006695747142657638 2023-01-21 12:57:52.288840: step: 340/526, loss: 0.08317708969116211 2023-01-21 12:57:53.414809: step: 344/526, loss: 0.00021195411682128906 2023-01-21 12:57:54.550433: step: 348/526, loss: 0.0009900092845782638 2023-01-21 12:57:55.714722: step: 352/526, loss: 0.022437285631895065 2023-01-21 12:57:56.835955: step: 356/526, loss: -0.0 2023-01-21 12:57:57.984564: step: 360/526, loss: 0.0013381957542151213 2023-01-21 12:57:59.160629: step: 364/526, loss: 0.004854011815041304 2023-01-21 12:58:00.303455: step: 368/526, loss: 0.0022419930901378393 2023-01-21 12:58:01.436738: step: 372/526, loss: 0.010491943918168545 2023-01-21 12:58:02.575513: step: 376/526, loss: 0.002476310823112726 2023-01-21 12:58:03.690639: step: 380/526, loss: 0.00018367767916060984 2023-01-21 12:58:04.847579: step: 384/526, loss: 0.0026439668145030737 2023-01-21 12:58:05.962144: step: 388/526, loss: 0.00012774467177223414 2023-01-21 12:58:07.118138: step: 392/526, loss: 2.2697449821862392e-05 2023-01-21 12:58:08.227970: step: 396/526, loss: 0.0019767761696130037 2023-01-21 12:58:09.362211: step: 400/526, loss: 0.05824451521039009 2023-01-21 12:58:10.466632: step: 404/526, loss: 0.0007933616871014237 2023-01-21 12:58:11.592750: step: 408/526, loss: 0.005852317903190851 2023-01-21 12:58:12.727129: step: 412/526, loss: 0.023007966578006744 2023-01-21 12:58:13.892777: step: 416/526, loss: 0.00366554269567132 2023-01-21 12:58:15.002159: step: 420/526, loss: 0.04640350490808487 2023-01-21 12:58:16.104708: step: 424/526, loss: 0.0004723549063783139 2023-01-21 12:58:17.292965: step: 428/526, loss: 0.022696591913700104 2023-01-21 12:58:18.440453: step: 432/526, loss: 0.002528953365981579 2023-01-21 12:58:19.543918: step: 436/526, loss: 7.07626313669607e-05 2023-01-21 12:58:20.679453: step: 440/526, loss: 0.04513902962207794 2023-01-21 12:58:21.828722: step: 444/526, loss: 0.0002626419300213456 2023-01-21 12:58:22.956796: step: 448/526, loss: 0.0005931854248046875 2023-01-21 12:58:24.117457: step: 452/526, loss: 0.00011329651169944555 2023-01-21 12:58:25.239420: step: 456/526, loss: 0.0231507308781147 2023-01-21 12:58:26.377649: step: 460/526, loss: 0.0009976864093914628 2023-01-21 12:58:27.527102: step: 464/526, loss: 0.05659074708819389 2023-01-21 12:58:28.662093: step: 468/526, loss: 0.003830432891845703 2023-01-21 12:58:29.790736: step: 472/526, loss: 0.0009501457097940147 2023-01-21 12:58:30.905821: step: 476/526, loss: 0.006242990959435701 2023-01-21 12:58:32.071876: step: 480/526, loss: 0.01744966395199299 2023-01-21 12:58:33.197399: step: 484/526, loss: 1.678466833254788e-05 2023-01-21 12:58:34.321037: step: 488/526, loss: 7.247925623232732e-06 2023-01-21 12:58:35.441379: step: 492/526, loss: 0.04512348398566246 2023-01-21 12:58:36.573811: step: 496/526, loss: 0.004738807678222656 2023-01-21 12:58:37.712312: step: 500/526, loss: 0.0009008407942019403 2023-01-21 12:58:38.823505: step: 504/526, loss: 0.0022682668641209602 2023-01-21 12:58:39.964841: step: 508/526, loss: 0.005920696537941694 2023-01-21 12:58:41.114233: step: 512/526, loss: 0.04358692094683647 2023-01-21 12:58:42.240535: step: 516/526, loss: 0.00025053025456145406 2023-01-21 12:58:43.400158: step: 520/526, loss: 0.00191326136700809 2023-01-21 12:58:44.512488: step: 524/526, loss: 3.705024573719129e-05 2023-01-21 12:58:45.662773: step: 528/526, loss: 0.00021781922259833664 2023-01-21 12:58:46.838165: step: 532/526, loss: 0.005717658903449774 2023-01-21 12:58:47.986001: step: 536/526, loss: 0.00472602853551507 2023-01-21 12:58:49.163425: step: 540/526, loss: 0.0017313479911535978 2023-01-21 12:58:50.303847: step: 544/526, loss: 0.004014492034912109 2023-01-21 12:58:51.439583: step: 548/526, loss: 0.0023202896118164062 2023-01-21 12:58:52.608875: step: 552/526, loss: 7.972717139637098e-05 2023-01-21 12:58:53.762613: step: 556/526, loss: 0.0293413158506155 2023-01-21 12:58:54.879013: step: 560/526, loss: 0.008193492889404297 2023-01-21 12:58:56.012461: step: 564/526, loss: 0.006133651826530695 2023-01-21 12:58:57.176198: step: 568/526, loss: 0.0006778716924600303 2023-01-21 12:58:58.289879: step: 572/526, loss: 0.0003970146062783897 2023-01-21 12:58:59.433666: step: 576/526, loss: 0.00011472702317405492 2023-01-21 12:59:00.571685: step: 580/526, loss: 2.6321409677620977e-05 2023-01-21 12:59:01.739102: step: 584/526, loss: 3.681182715808973e-05 2023-01-21 12:59:02.861252: step: 588/526, loss: 1.0776519047794864e-05 2023-01-21 12:59:03.967912: step: 592/526, loss: 0.015050698071718216 2023-01-21 12:59:05.130402: step: 596/526, loss: 0.02680053934454918 2023-01-21 12:59:06.284472: step: 600/526, loss: 0.01822662353515625 2023-01-21 12:59:07.459307: step: 604/526, loss: 0.05724554508924484 2023-01-21 12:59:08.610843: step: 608/526, loss: 0.00027551650418899953 2023-01-21 12:59:09.751693: step: 612/526, loss: 0.0007148743025027215 2023-01-21 12:59:10.899060: step: 616/526, loss: 0.0012170792324468493 2023-01-21 12:59:12.034825: step: 620/526, loss: 3.2901763916015625e-05 2023-01-21 12:59:13.186312: step: 624/526, loss: 9.679794311523438e-05 2023-01-21 12:59:14.340480: step: 628/526, loss: 0.017675209790468216 2023-01-21 12:59:15.472658: step: 632/526, loss: 0.0008423805120401084 2023-01-21 12:59:16.612831: step: 636/526, loss: 0.004496097564697266 2023-01-21 12:59:17.817131: step: 640/526, loss: 7.61032133596018e-05 2023-01-21 12:59:18.971776: step: 644/526, loss: 0.02545623853802681 2023-01-21 12:59:20.122691: step: 648/526, loss: 0.016273116692900658 2023-01-21 12:59:21.270616: step: 652/526, loss: 0.035306550562381744 2023-01-21 12:59:22.387735: step: 656/526, loss: 0.0016454696888104081 2023-01-21 12:59:23.511237: step: 660/526, loss: 3.185272362316027e-05 2023-01-21 12:59:24.665920: step: 664/526, loss: 0.00437507638707757 2023-01-21 12:59:25.756450: step: 668/526, loss: 0.0035715103149414062 2023-01-21 12:59:26.871919: step: 672/526, loss: 0.0013638496166095138 2023-01-21 12:59:28.016428: step: 676/526, loss: 8.77380352903856e-06 2023-01-21 12:59:29.172903: step: 680/526, loss: 0.00018968581571243703 2023-01-21 12:59:30.313539: step: 684/526, loss: 0.007482767105102539 2023-01-21 12:59:31.458695: step: 688/526, loss: 0.0005333900917321444 2023-01-21 12:59:32.586979: step: 692/526, loss: 0.0004784584161825478 2023-01-21 12:59:33.774154: step: 696/526, loss: 0.0002773285086732358 2023-01-21 12:59:34.928113: step: 700/526, loss: 0.004928970243781805 2023-01-21 12:59:36.078590: step: 704/526, loss: 0.013131809420883656 2023-01-21 12:59:37.215472: step: 708/526, loss: 0.0003971099795307964 2023-01-21 12:59:38.358833: step: 712/526, loss: 0.02708374708890915 2023-01-21 12:59:39.476891: step: 716/526, loss: 0.05870051309466362 2023-01-21 12:59:40.640130: step: 720/526, loss: 0.00011129378981422633 2023-01-21 12:59:41.762639: step: 724/526, loss: 0.00032711029052734375 2023-01-21 12:59:42.916800: step: 728/526, loss: 0.47495585680007935 2023-01-21 12:59:44.052386: step: 732/526, loss: 0.005985069554299116 2023-01-21 12:59:45.148551: step: 736/526, loss: 0.0004085540713276714 2023-01-21 12:59:46.300098: step: 740/526, loss: 0.020210934802889824 2023-01-21 12:59:47.451749: step: 744/526, loss: 0.029415130615234375 2023-01-21 12:59:48.577537: step: 748/526, loss: 2.384185791015625e-05 2023-01-21 12:59:49.734317: step: 752/526, loss: 0.008944511413574219 2023-01-21 12:59:50.849599: step: 756/526, loss: 0.0005935192457400262 2023-01-21 12:59:51.988529: step: 760/526, loss: 0.0001256942778127268 2023-01-21 12:59:53.133626: step: 764/526, loss: 0.0003887176571879536 2023-01-21 12:59:54.268788: step: 768/526, loss: 0.0040260315872728825 2023-01-21 12:59:55.421671: step: 772/526, loss: 0.0012353897327557206 2023-01-21 12:59:56.528815: step: 776/526, loss: 1.850128137448337e-05 2023-01-21 12:59:57.665344: step: 780/526, loss: 0.02021818235516548 2023-01-21 12:59:58.803846: step: 784/526, loss: 0.028534317389130592 2023-01-21 12:59:59.927028: step: 788/526, loss: 0.0019321441650390625 2023-01-21 13:00:01.071133: step: 792/526, loss: 0.003497791476547718 2023-01-21 13:00:02.192544: step: 796/526, loss: 2.307891918462701e-05 2023-01-21 13:00:03.299175: step: 800/526, loss: 0.0014812469016760588 2023-01-21 13:00:04.438456: step: 804/526, loss: 0.0007188796880654991 2023-01-21 13:00:05.565253: step: 808/526, loss: 0.06550207734107971 2023-01-21 13:00:06.678618: step: 812/526, loss: 0.0005131721845827997 2023-01-21 13:00:07.825895: step: 816/526, loss: 0.0046897889114916325 2023-01-21 13:00:08.976449: step: 820/526, loss: 0.00021324156841728836 2023-01-21 13:00:10.132684: step: 824/526, loss: 0.0005862236139364541 2023-01-21 13:00:11.286815: step: 828/526, loss: 6.48498553346144e-06 2023-01-21 13:00:12.452847: step: 832/526, loss: 0.04538403078913689 2023-01-21 13:00:13.614355: step: 836/526, loss: 0.0003143310605082661 2023-01-21 13:00:14.741662: step: 840/526, loss: 0.01945781707763672 2023-01-21 13:00:15.869258: step: 844/526, loss: 0.001015520072542131 2023-01-21 13:00:17.035857: step: 848/526, loss: 0.0020095824729651213 2023-01-21 13:00:18.174853: step: 852/526, loss: 0.023306705057621002 2023-01-21 13:00:19.319928: step: 856/526, loss: 4.57763671875e-05 2023-01-21 13:00:20.435339: step: 860/526, loss: 0.014178657904267311 2023-01-21 13:00:21.560331: step: 864/526, loss: 0.0015453338855877519 2023-01-21 13:00:22.705547: step: 868/526, loss: 0.002031993819400668 2023-01-21 13:00:23.848464: step: 872/526, loss: 0.06039486080408096 2023-01-21 13:00:24.989807: step: 876/526, loss: 0.03189563751220703 2023-01-21 13:00:26.113789: step: 880/526, loss: 4.158019874012098e-05 2023-01-21 13:00:27.292669: step: 884/526, loss: 0.014531994238495827 2023-01-21 13:00:28.444548: step: 888/526, loss: 0.00010585784912109375 2023-01-21 13:00:29.628432: step: 892/526, loss: 0.0007903099758550525 2023-01-21 13:00:30.773637: step: 896/526, loss: 0.005035972688347101 2023-01-21 13:00:31.890356: step: 900/526, loss: 0.0354883186519146 2023-01-21 13:00:33.036832: step: 904/526, loss: 0.003201007843017578 2023-01-21 13:00:34.167196: step: 908/526, loss: 0.009349632076919079 2023-01-21 13:00:35.342116: step: 912/526, loss: 0.034232523292303085 2023-01-21 13:00:36.440116: step: 916/526, loss: 9.489059448242188e-05 2023-01-21 13:00:37.626607: step: 920/526, loss: 0.020899200811982155 2023-01-21 13:00:38.753870: step: 924/526, loss: 0.004879761021584272 2023-01-21 13:00:39.893479: step: 928/526, loss: 0.0004287242772988975 2023-01-21 13:00:41.036027: step: 932/526, loss: 0.0016637801891192794 2023-01-21 13:00:42.196641: step: 936/526, loss: 0.0013332366943359375 2023-01-21 13:00:43.377700: step: 940/526, loss: 0.0002853393380064517 2023-01-21 13:00:44.568022: step: 944/526, loss: 0.026480484753847122 2023-01-21 13:00:45.721463: step: 948/526, loss: 0.04964113235473633 2023-01-21 13:00:46.825168: step: 952/526, loss: 0.28571662306785583 2023-01-21 13:00:47.949318: step: 956/526, loss: 0.0009974478743970394 2023-01-21 13:00:49.107290: step: 960/526, loss: 0.0036619186867028475 2023-01-21 13:00:50.252918: step: 964/526, loss: 0.0006009578937664628 2023-01-21 13:00:51.388706: step: 968/526, loss: 0.0019067287212237716 2023-01-21 13:00:52.530504: step: 972/526, loss: 0.0072498321533203125 2023-01-21 13:00:53.642264: step: 976/526, loss: 0.01940612867474556 2023-01-21 13:00:54.825840: step: 980/526, loss: 0.08336753398180008 2023-01-21 13:00:55.987042: step: 984/526, loss: 0.0011601448059082031 2023-01-21 13:00:57.132180: step: 988/526, loss: 0.004457283299416304 2023-01-21 13:00:58.272992: step: 992/526, loss: 0.018612097948789597 2023-01-21 13:00:59.415772: step: 996/526, loss: 0.03325996547937393 2023-01-21 13:01:00.541889: step: 1000/526, loss: 0.005961274728178978 2023-01-21 13:01:01.665962: step: 1004/526, loss: 0.02926664426922798 2023-01-21 13:01:02.797057: step: 1008/526, loss: 0.06164827570319176 2023-01-21 13:01:03.923292: step: 1012/526, loss: 0.00033903124858625233 2023-01-21 13:01:05.050903: step: 1016/526, loss: 0.0006549835670739412 2023-01-21 13:01:06.228585: step: 1020/526, loss: 0.014649391174316406 2023-01-21 13:01:07.355137: step: 1024/526, loss: 0.0009042740566655993 2023-01-21 13:01:08.464976: step: 1028/526, loss: 0.002101993653923273 2023-01-21 13:01:09.593561: step: 1032/526, loss: 0.0008924484718590975 2023-01-21 13:01:10.759773: step: 1036/526, loss: 0.0008095264784060419 2023-01-21 13:01:11.935253: step: 1040/526, loss: 0.008431053720414639 2023-01-21 13:01:13.081657: step: 1044/526, loss: 3.767013822653098e-06 2023-01-21 13:01:14.276553: step: 1048/526, loss: 0.003498077392578125 2023-01-21 13:01:15.400686: step: 1052/526, loss: 0.0001203536958200857 2023-01-21 13:01:16.538456: step: 1056/526, loss: 0.2820812165737152 2023-01-21 13:01:17.636722: step: 1060/526, loss: 7.939338684082031e-05 2023-01-21 13:01:18.789038: step: 1064/526, loss: 0.00047206878662109375 2023-01-21 13:01:19.913464: step: 1068/526, loss: 7.772445314913057e-06 2023-01-21 13:01:21.046649: step: 1072/526, loss: 0.003941059578210115 2023-01-21 13:01:22.153986: step: 1076/526, loss: 0.014945030212402344 2023-01-21 13:01:23.336798: step: 1080/526, loss: 0.00442428607493639 2023-01-21 13:01:24.486541: step: 1084/526, loss: 0.0024404525756835938 2023-01-21 13:01:25.633640: step: 1088/526, loss: 1.5258792700478807e-06 2023-01-21 13:01:26.751987: step: 1092/526, loss: 0.00015316010103560984 2023-01-21 13:01:27.897906: step: 1096/526, loss: 0.00018272399029228836 2023-01-21 13:01:29.053070: step: 1100/526, loss: 1.4877318790240679e-05 2023-01-21 13:01:30.195299: step: 1104/526, loss: 0.040668487548828125 2023-01-21 13:01:31.311258: step: 1108/526, loss: 0.018201828002929688 2023-01-21 13:01:32.438057: step: 1112/526, loss: 0.0006000519497320056 2023-01-21 13:01:33.565004: step: 1116/526, loss: 0.03940659016370773 2023-01-21 13:01:34.690986: step: 1120/526, loss: 0.0017597199184820056 2023-01-21 13:01:35.826783: step: 1124/526, loss: 0.00037136077298782766 2023-01-21 13:01:36.963917: step: 1128/526, loss: 0.0001710891810944304 2023-01-21 13:01:38.108068: step: 1132/526, loss: 0.02481565624475479 2023-01-21 13:01:39.256256: step: 1136/526, loss: 2.1409989130916074e-05 2023-01-21 13:01:40.385187: step: 1140/526, loss: 0.00046899321023374796 2023-01-21 13:01:41.519415: step: 1144/526, loss: 0.0003192901785951108 2023-01-21 13:01:42.660118: step: 1148/526, loss: 0.02814960479736328 2023-01-21 13:01:43.786165: step: 1152/526, loss: 0.022788619622588158 2023-01-21 13:01:44.934534: step: 1156/526, loss: 0.002585887908935547 2023-01-21 13:01:46.106205: step: 1160/526, loss: 0.00947427749633789 2023-01-21 13:01:47.251443: step: 1164/526, loss: 0.0436311736702919 2023-01-21 13:01:48.405756: step: 1168/526, loss: 0.013833713717758656 2023-01-21 13:01:49.557378: step: 1172/526, loss: 0.013602066785097122 2023-01-21 13:01:50.684985: step: 1176/526, loss: 0.01815180853009224 2023-01-21 13:01:51.834659: step: 1180/526, loss: 0.0021659850608557463 2023-01-21 13:01:52.981285: step: 1184/526, loss: 5.931854320806451e-05 2023-01-21 13:01:54.122949: step: 1188/526, loss: 0.0044879917986691 2023-01-21 13:01:55.271958: step: 1192/526, loss: 0.0024885176680982113 2023-01-21 13:01:56.394104: step: 1196/526, loss: 0.08112716674804688 2023-01-21 13:01:57.518736: step: 1200/526, loss: 0.00940713845193386 2023-01-21 13:01:58.665204: step: 1204/526, loss: 0.004264450166374445 2023-01-21 13:01:59.806882: step: 1208/526, loss: 0.0004259109846316278 2023-01-21 13:02:00.935643: step: 1212/526, loss: 0.00015192032151389867 2023-01-21 13:02:02.099812: step: 1216/526, loss: 0.313162624835968 2023-01-21 13:02:03.226278: step: 1220/526, loss: 5.3882598876953125e-05 2023-01-21 13:02:04.353375: step: 1224/526, loss: 6.0033802583348006e-05 2023-01-21 13:02:05.460459: step: 1228/526, loss: 0.00025005341740325093 2023-01-21 13:02:06.595394: step: 1232/526, loss: 0.0013680459233000875 2023-01-21 13:02:07.738726: step: 1236/526, loss: 0.0001597404625499621 2023-01-21 13:02:08.865770: step: 1240/526, loss: 0.0002983093145303428 2023-01-21 13:02:09.978199: step: 1244/526, loss: 0.0068229674361646175 2023-01-21 13:02:11.083593: step: 1248/526, loss: 2.155303991457913e-05 2023-01-21 13:02:12.202824: step: 1252/526, loss: 0.2867697775363922 2023-01-21 13:02:13.343353: step: 1256/526, loss: 0.0001184463471872732 2023-01-21 13:02:14.498416: step: 1260/526, loss: 0.006624412722885609 2023-01-21 13:02:15.631020: step: 1264/526, loss: 0.005353641230612993 2023-01-21 13:02:16.778506: step: 1268/526, loss: 0.0012070179218426347 2023-01-21 13:02:17.891707: step: 1272/526, loss: 0.00045108795166015625 2023-01-21 13:02:19.005744: step: 1276/526, loss: 0.000192546853213571 2023-01-21 13:02:20.105601: step: 1280/526, loss: 0.002213859697803855 2023-01-21 13:02:21.248529: step: 1284/526, loss: 0.0009983063209801912 2023-01-21 13:02:22.369617: step: 1288/526, loss: 6.961822691664565e-06 2023-01-21 13:02:23.495965: step: 1292/526, loss: 0.005574035923928022 2023-01-21 13:02:24.606800: step: 1296/526, loss: 0.006134224124252796 2023-01-21 13:02:25.724343: step: 1300/526, loss: 0.00228710169903934 2023-01-21 13:02:26.855452: step: 1304/526, loss: 0.0001657485991017893 2023-01-21 13:02:28.011188: step: 1308/526, loss: 0.003320026444271207 2023-01-21 13:02:29.116127: step: 1312/526, loss: 0.00012602806964423507 2023-01-21 13:02:30.238888: step: 1316/526, loss: 0.006726741790771484 2023-01-21 13:02:31.357828: step: 1320/526, loss: 0.03466634452342987 2023-01-21 13:02:32.499652: step: 1324/526, loss: -1.506805438111769e-05 2023-01-21 13:02:33.606861: step: 1328/526, loss: 0.0002075195370707661 2023-01-21 13:02:34.747666: step: 1332/526, loss: 0.04439039155840874 2023-01-21 13:02:35.857678: step: 1336/526, loss: 0.0009511836105957627 2023-01-21 13:02:37.013258: step: 1340/526, loss: 0.0005867004510946572 2023-01-21 13:02:38.153374: step: 1344/526, loss: 6.895065598655492e-05 2023-01-21 13:02:39.279484: step: 1348/526, loss: 0.0010419845348224044 2023-01-21 13:02:40.396032: step: 1352/526, loss: 0.0008003235561773181 2023-01-21 13:02:41.493138: step: 1356/526, loss: 0.01979966089129448 2023-01-21 13:02:42.636177: step: 1360/526, loss: 8.37326078908518e-05 2023-01-21 13:02:43.745028: step: 1364/526, loss: 0.008275032043457031 2023-01-21 13:02:44.871941: step: 1368/526, loss: 0.0048618316650390625 2023-01-21 13:02:46.027095: step: 1372/526, loss: 0.00020580290583893657 2023-01-21 13:02:47.145670: step: 1376/526, loss: 0.003113794606178999 2023-01-21 13:02:48.292319: step: 1380/526, loss: 0.0343870185315609 2023-01-21 13:02:49.419813: step: 1384/526, loss: 0.0012341499095782638 2023-01-21 13:02:50.532400: step: 1388/526, loss: 4.9209596909349784e-05 2023-01-21 13:02:51.659293: step: 1392/526, loss: 0.0002454757923260331 2023-01-21 13:02:52.782730: step: 1396/526, loss: 0.0017900944221764803 2023-01-21 13:02:53.940289: step: 1400/526, loss: 0.0031234740745276213 2023-01-21 13:02:55.107349: step: 1404/526, loss: -1.33514404296875e-05 2023-01-21 13:02:56.236268: step: 1408/526, loss: 0.06752175837755203 2023-01-21 13:02:57.374674: step: 1412/526, loss: 0.019492531195282936 2023-01-21 13:02:58.508144: step: 1416/526, loss: 0.04481372609734535 2023-01-21 13:02:59.658163: step: 1420/526, loss: 0.00041065216646529734 2023-01-21 13:03:00.797517: step: 1424/526, loss: 0.0003224372921977192 2023-01-21 13:03:01.910489: step: 1428/526, loss: 0.00020689965458586812 2023-01-21 13:03:03.052174: step: 1432/526, loss: -2.6702882678364404e-06 2023-01-21 13:03:04.187380: step: 1436/526, loss: 0.04386467859148979 2023-01-21 13:03:05.353860: step: 1440/526, loss: 5.588532076217234e-05 2023-01-21 13:03:06.504927: step: 1444/526, loss: 0.003950232174247503 2023-01-21 13:03:07.647786: step: 1448/526, loss: 0.07215669751167297 2023-01-21 13:03:08.775511: step: 1452/526, loss: 1.182556115963962e-05 2023-01-21 13:03:09.904315: step: 1456/526, loss: 0.0023937225341796875 2023-01-21 13:03:11.012720: step: 1460/526, loss: 0.0002323150692973286 2023-01-21 13:03:12.150916: step: 1464/526, loss: 0.05893097072839737 2023-01-21 13:03:13.262220: step: 1468/526, loss: 0.6178361773490906 2023-01-21 13:03:14.398175: step: 1472/526, loss: 0.007234144490212202 2023-01-21 13:03:15.534087: step: 1476/526, loss: 0.0031185150146484375 2023-01-21 13:03:16.656145: step: 1480/526, loss: 0.0002645492786541581 2023-01-21 13:03:17.798292: step: 1484/526, loss: 0.00013699530973099172 2023-01-21 13:03:18.952149: step: 1488/526, loss: 0.007129478733986616 2023-01-21 13:03:20.089214: step: 1492/526, loss: 0.0006588220712728798 2023-01-21 13:03:21.241239: step: 1496/526, loss: 0.018398571759462357 2023-01-21 13:03:22.390457: step: 1500/526, loss: -9.250640687241685e-06 2023-01-21 13:03:23.511515: step: 1504/526, loss: 0.01704854890704155 2023-01-21 13:03:24.640030: step: 1508/526, loss: 7.758140418445691e-05 2023-01-21 13:03:25.788070: step: 1512/526, loss: 0.00525321951135993 2023-01-21 13:03:26.907479: step: 1516/526, loss: 0.00032634736271575093 2023-01-21 13:03:28.044130: step: 1520/526, loss: 0.015398693270981312 2023-01-21 13:03:29.172723: step: 1524/526, loss: 0.0006237507332116365 2023-01-21 13:03:30.303281: step: 1528/526, loss: 0.004484463017433882 2023-01-21 13:03:31.443396: step: 1532/526, loss: 0.05192699283361435 2023-01-21 13:03:32.560245: step: 1536/526, loss: 0.0070996759459376335 2023-01-21 13:03:33.681734: step: 1540/526, loss: 0.00010223388380836695 2023-01-21 13:03:34.814827: step: 1544/526, loss: 0.007494211196899414 2023-01-21 13:03:35.955007: step: 1548/526, loss: 0.03178148344159126 2023-01-21 13:03:37.111338: step: 1552/526, loss: 3.805160304182209e-05 2023-01-21 13:03:38.240059: step: 1556/526, loss: 0.047258712351322174 2023-01-21 13:03:39.361523: step: 1560/526, loss: 0.006660843268036842 2023-01-21 13:03:40.502262: step: 1564/526, loss: 0.08737526088953018 2023-01-21 13:03:41.606044: step: 1568/526, loss: 0.00011920928955078125 2023-01-21 13:03:42.713167: step: 1572/526, loss: 0.0018775940407067537 2023-01-21 13:03:43.888881: step: 1576/526, loss: 0.07419309765100479 2023-01-21 13:03:45.002504: step: 1580/526, loss: 0.0007379531743936241 2023-01-21 13:03:46.150505: step: 1584/526, loss: 0.05375843122601509 2023-01-21 13:03:47.298502: step: 1588/526, loss: 0.00013999939255882055 2023-01-21 13:03:48.452567: step: 1592/526, loss: 0.00026607513427734375 2023-01-21 13:03:49.569574: step: 1596/526, loss: 0.004611778538674116 2023-01-21 13:03:50.685557: step: 1600/526, loss: 0.007579517550766468 2023-01-21 13:03:51.820126: step: 1604/526, loss: 0.025613022968173027 2023-01-21 13:03:52.944447: step: 1608/526, loss: 0.00110626220703125 2023-01-21 13:03:54.083379: step: 1612/526, loss: 0.0025236131623387337 2023-01-21 13:03:55.219721: step: 1616/526, loss: 0.0002037048398051411 2023-01-21 13:03:56.370457: step: 1620/526, loss: 0.00012702941603492945 2023-01-21 13:03:57.514487: step: 1624/526, loss: 0.002845382783561945 2023-01-21 13:03:58.649210: step: 1628/526, loss: 8.049011375987902e-05 2023-01-21 13:03:59.782345: step: 1632/526, loss: 0.40420445799827576 2023-01-21 13:04:00.917347: step: 1636/526, loss: 5.0354006816633046e-05 2023-01-21 13:04:02.060843: step: 1640/526, loss: 2.956390380859375e-05 2023-01-21 13:04:03.188853: step: 1644/526, loss: 0.0019522667862474918 2023-01-21 13:04:04.320069: step: 1648/526, loss: 5.14984139954322e-06 2023-01-21 13:04:05.471882: step: 1652/526, loss: 0.00043964385986328125 2023-01-21 13:04:06.612532: step: 1656/526, loss: 0.020165156573057175 2023-01-21 13:04:07.737975: step: 1660/526, loss: 0.0002988815540447831 2023-01-21 13:04:08.835830: step: 1664/526, loss: 0.047823715955019 2023-01-21 13:04:09.947541: step: 1668/526, loss: -0.0 2023-01-21 13:04:11.083221: step: 1672/526, loss: 0.011056900024414062 2023-01-21 13:04:12.203860: step: 1676/526, loss: 0.003069305559620261 2023-01-21 13:04:13.360778: step: 1680/526, loss: 0.005118656437844038 2023-01-21 13:04:14.515222: step: 1684/526, loss: 0.003489971160888672 2023-01-21 13:04:15.634337: step: 1688/526, loss: 0.0018761157989501953 2023-01-21 13:04:16.760115: step: 1692/526, loss: 0.0032494545448571444 2023-01-21 13:04:17.919039: step: 1696/526, loss: 0.005176162347197533 2023-01-21 13:04:19.042064: step: 1700/526, loss: 0.0032299996819347143 2023-01-21 13:04:20.181965: step: 1704/526, loss: 4.673004150390625e-05 2023-01-21 13:04:21.344297: step: 1708/526, loss: 2.4890900022001006e-05 2023-01-21 13:04:22.488202: step: 1712/526, loss: 0.00055780413094908 2023-01-21 13:04:23.624757: step: 1716/526, loss: 0.01767864264547825 2023-01-21 13:04:24.763035: step: 1720/526, loss: 0.0003445625479798764 2023-01-21 13:04:25.906160: step: 1724/526, loss: 0.00011425017873989418 2023-01-21 13:04:27.028024: step: 1728/526, loss: 0.0015711784362792969 2023-01-21 13:04:28.177699: step: 1732/526, loss: 0.04193558543920517 2023-01-21 13:04:29.294926: step: 1736/526, loss: 5.91278057981981e-06 2023-01-21 13:04:30.416676: step: 1740/526, loss: 0.012828493490815163 2023-01-21 13:04:31.580837: step: 1744/526, loss: 0.00210914621129632 2023-01-21 13:04:32.703944: step: 1748/526, loss: 0.3500072658061981 2023-01-21 13:04:33.860014: step: 1752/526, loss: 0.005663109011948109 2023-01-21 13:04:34.997851: step: 1756/526, loss: 0.0003544807550497353 2023-01-21 13:04:36.129494: step: 1760/526, loss: 0.0001990795135498047 2023-01-21 13:04:37.235197: step: 1764/526, loss: 0.02201671525835991 2023-01-21 13:04:38.363500: step: 1768/526, loss: 0.0013246058952063322 2023-01-21 13:04:39.536039: step: 1772/526, loss: 0.00013217926607467234 2023-01-21 13:04:40.637377: step: 1776/526, loss: 0.014777851291000843 2023-01-21 13:04:41.788882: step: 1780/526, loss: 7.305145118152723e-05 2023-01-21 13:04:42.934078: step: 1784/526, loss: 0.011471175588667393 2023-01-21 13:04:44.081044: step: 1788/526, loss: 0.047315407544374466 2023-01-21 13:04:45.230977: step: 1792/526, loss: 0.007222938816994429 2023-01-21 13:04:46.351868: step: 1796/526, loss: 0.0006201743963174522 2023-01-21 13:04:47.542196: step: 1800/526, loss: 0.007343387696892023 2023-01-21 13:04:48.660130: step: 1804/526, loss: 0.0018410682678222656 2023-01-21 13:04:49.782316: step: 1808/526, loss: 0.0002980232238769531 2023-01-21 13:04:50.945392: step: 1812/526, loss: -5.4836273193359375e-06 2023-01-21 13:04:52.078806: step: 1816/526, loss: 0.03268451988697052 2023-01-21 13:04:53.214851: step: 1820/526, loss: 0.013419151306152344 2023-01-21 13:04:54.338735: step: 1824/526, loss: 0.02773771435022354 2023-01-21 13:04:55.476452: step: 1828/526, loss: 0.00028333664522506297 2023-01-21 13:04:56.623849: step: 1832/526, loss: 0.12107352912425995 2023-01-21 13:04:57.750615: step: 1836/526, loss: 0.024143792688846588 2023-01-21 13:04:58.853467: step: 1840/526, loss: 0.017164994031190872 2023-01-21 13:04:59.946694: step: 1844/526, loss: 8.449554297840223e-05 2023-01-21 13:05:01.061651: step: 1848/526, loss: 0.3749936819076538 2023-01-21 13:05:02.203887: step: 1852/526, loss: 0.00472183246165514 2023-01-21 13:05:03.329057: step: 1856/526, loss: 0.00114612584002316 2023-01-21 13:05:04.435307: step: 1860/526, loss: 2.6035308110294864e-05 2023-01-21 13:05:05.568726: step: 1864/526, loss: 0.0008787155384197831 2023-01-21 13:05:06.677397: step: 1868/526, loss: 0.015067482367157936 2023-01-21 13:05:07.814283: step: 1872/526, loss: 7.724762326688506e-06 2023-01-21 13:05:08.957869: step: 1876/526, loss: 0.0010809899540618062 2023-01-21 13:05:10.068716: step: 1880/526, loss: 0.0318942554295063 2023-01-21 13:05:11.211209: step: 1884/526, loss: 0.005975151434540749 2023-01-21 13:05:12.342344: step: 1888/526, loss: 0.008412457071244717 2023-01-21 13:05:13.496876: step: 1892/526, loss: 0.00031046869116835296 2023-01-21 13:05:14.639387: step: 1896/526, loss: 0.0007589340093545616 2023-01-21 13:05:15.760294: step: 1900/526, loss: 0.0001552104949951172 2023-01-21 13:05:16.897070: step: 1904/526, loss: 0.024302775040268898 2023-01-21 13:05:18.050394: step: 1908/526, loss: 0.0002381324884481728 2023-01-21 13:05:19.219276: step: 1912/526, loss: 0.018163396045565605 2023-01-21 13:05:20.368500: step: 1916/526, loss: 9.412765939487144e-05 2023-01-21 13:05:21.503905: step: 1920/526, loss: 1.3256073543743696e-05 2023-01-21 13:05:22.682557: step: 1924/526, loss: 0.06012916564941406 2023-01-21 13:05:23.807436: step: 1928/526, loss: 0.021240903064608574 2023-01-21 13:05:24.921771: step: 1932/526, loss: 0.00926980935037136 2023-01-21 13:05:26.045660: step: 1936/526, loss: 0.0023797035682946444 2023-01-21 13:05:27.215543: step: 1940/526, loss: 0.012895584106445312 2023-01-21 13:05:28.344874: step: 1944/526, loss: 0.0018672943115234375 2023-01-21 13:05:29.518459: step: 1948/526, loss: 0.014040709473192692 2023-01-21 13:05:30.687774: step: 1952/526, loss: 1.621246337890625e-05 2023-01-21 13:05:31.823508: step: 1956/526, loss: 0.00034794810926541686 2023-01-21 13:05:32.970760: step: 1960/526, loss: 0.0002079010009765625 2023-01-21 13:05:34.095184: step: 1964/526, loss: 0.0021209716796875 2023-01-21 13:05:35.248087: step: 1968/526, loss: 0.007848930545151234 2023-01-21 13:05:36.394767: step: 1972/526, loss: 0.0007929801940917969 2023-01-21 13:05:37.515966: step: 1976/526, loss: 0.00033969880314543843 2023-01-21 13:05:38.629621: step: 1980/526, loss: 0.0021348954178392887 2023-01-21 13:05:39.765110: step: 1984/526, loss: 0.0006766319274902344 2023-01-21 13:05:40.870267: step: 1988/526, loss: 0.13882236182689667 2023-01-21 13:05:42.013070: step: 1992/526, loss: 0.00038022996159270406 2023-01-21 13:05:43.140664: step: 1996/526, loss: 0.010604476556181908 2023-01-21 13:05:44.269916: step: 2000/526, loss: 0.0010848998790606856 2023-01-21 13:05:45.378839: step: 2004/526, loss: 0.00010786057100631297 2023-01-21 13:05:46.513019: step: 2008/526, loss: 0.01777353323996067 2023-01-21 13:05:47.639967: step: 2012/526, loss: 0.01016454678028822 2023-01-21 13:05:48.779155: step: 2016/526, loss: 0.011587905697524548 2023-01-21 13:05:49.902990: step: 2020/526, loss: 0.0012754440540447831 2023-01-21 13:05:51.049263: step: 2024/526, loss: 3.013610876223538e-05 2023-01-21 13:05:52.164654: step: 2028/526, loss: 0.00038013458834029734 2023-01-21 13:05:53.295197: step: 2032/526, loss: 0.0002818107604980469 2023-01-21 13:05:54.418556: step: 2036/526, loss: 3.4809113458322827e-06 2023-01-21 13:05:55.564042: step: 2040/526, loss: 0.009374809451401234 2023-01-21 13:05:56.677474: step: 2044/526, loss: 0.0018835067749023438 2023-01-21 13:05:57.810328: step: 2048/526, loss: 0.018840981647372246 2023-01-21 13:05:58.960150: step: 2052/526, loss: 0.0005063057178631425 2023-01-21 13:06:00.081337: step: 2056/526, loss: 8.964539119915571e-06 2023-01-21 13:06:01.223713: step: 2060/526, loss: 0.6696677207946777 2023-01-21 13:06:02.347621: step: 2064/526, loss: 4.634857032215223e-05 2023-01-21 13:06:03.476540: step: 2068/526, loss: 0.00021247864060569555 2023-01-21 13:06:04.656069: step: 2072/526, loss: 0.0015189171535894275 2023-01-21 13:06:05.768110: step: 2076/526, loss: 0.008635425940155983 2023-01-21 13:06:06.920236: step: 2080/526, loss: 0.0007112979656085372 2023-01-21 13:06:08.074540: step: 2084/526, loss: -1.087188684323337e-05 2023-01-21 13:06:09.212115: step: 2088/526, loss: -2.2506712411995977e-05 2023-01-21 13:06:10.328819: step: 2092/526, loss: 2.8991700673941523e-05 2023-01-21 13:06:11.455424: step: 2096/526, loss: 0.00027294160099700093 2023-01-21 13:06:12.608408: step: 2100/526, loss: 0.00018835067749023438 2023-01-21 13:06:13.759485: step: 2104/526, loss: 6.365776062011719e-05 ================================================== Loss: 0.018 -------------------- Dev: {'event': {'p': 0.5985552115583075, 'r': 0.7723035952063915, 'f1': 0.6744186046511628}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Test: {'event': {'p': 0.6182707993474714, 'r': 0.791231732776618, 'f1': 0.694139194139194}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Chinese: {'event': {'p': 0.5844155844155844, 'r': 0.8333333333333334, 'f1': 0.6870229007633588}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Korean: {'event': {'p': 0.6888888888888889, 'r': 0.49206349206349204, 'f1': 0.5740740740740741}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} Russian: {'event': {'p': 0.42857142857142855, 'r': 0.5, 'f1': 0.4615384615384615}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 24} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 25 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:06:56.919979: step: 4/526, loss: 0.02930307388305664 2023-01-21 13:06:58.065511: step: 8/526, loss: 0.04721498861908913 2023-01-21 13:06:59.187833: step: 12/526, loss: 1.7166138377433526e-06 2023-01-21 13:07:00.318681: step: 16/526, loss: 0.0007007598760537803 2023-01-21 13:07:01.446920: step: 20/526, loss: 0.025460338220000267 2023-01-21 13:07:02.567038: step: 24/526, loss: 4.954338146490045e-05 2023-01-21 13:07:03.712265: step: 28/526, loss: 0.00046215057955123484 2023-01-21 13:07:04.835004: step: 32/526, loss: 0.12230377644300461 2023-01-21 13:07:05.985947: step: 36/526, loss: 4.5299530029296875e-05 2023-01-21 13:07:07.125298: step: 40/526, loss: 0.040052127093076706 2023-01-21 13:07:08.230968: step: 44/526, loss: 0.0003772735653910786 2023-01-21 13:07:09.331790: step: 48/526, loss: 0.0046042026951909065 2023-01-21 13:07:10.471943: step: 52/526, loss: 0.00035533905611373484 2023-01-21 13:07:11.638579: step: 56/526, loss: 0.0014753341674804688 2023-01-21 13:07:12.778750: step: 60/526, loss: 0.0007164001581259072 2023-01-21 13:07:13.909631: step: 64/526, loss: 0.00021076203847769648 2023-01-21 13:07:15.025241: step: 68/526, loss: 0.0056937215849757195 2023-01-21 13:07:16.173261: step: 72/526, loss: 0.007486533839255571 2023-01-21 13:07:17.333825: step: 76/526, loss: 0.0001832008274504915 2023-01-21 13:07:18.452375: step: 80/526, loss: 5.245208740234375e-06 2023-01-21 13:07:19.580585: step: 84/526, loss: 3.910065061063506e-06 2023-01-21 13:07:20.742426: step: 88/526, loss: 0.005286216735839844 2023-01-21 13:07:21.883165: step: 92/526, loss: 2.8324126105871983e-05 2023-01-21 13:07:23.003235: step: 96/526, loss: 0.003423976944759488 2023-01-21 13:07:24.137657: step: 100/526, loss: 3.862381163344253e-06 2023-01-21 13:07:25.282598: step: 104/526, loss: 0.0007698058616369963 2023-01-21 13:07:26.410548: step: 108/526, loss: 0.0009496689308434725 2023-01-21 13:07:27.527993: step: 112/526, loss: 0.07140665501356125 2023-01-21 13:07:28.667108: step: 116/526, loss: 0.00467834435403347 2023-01-21 13:07:29.838902: step: 120/526, loss: 0.0016530990833416581 2023-01-21 13:07:30.982370: step: 124/526, loss: 0.0007596969371661544 2023-01-21 13:07:32.104964: step: 128/526, loss: 0.001174282981082797 2023-01-21 13:07:33.241389: step: 132/526, loss: 0.051836684346199036 2023-01-21 13:07:34.355799: step: 136/526, loss: 0.0017196654807776213 2023-01-21 13:07:35.488109: step: 140/526, loss: -1.2397763384797145e-06 2023-01-21 13:07:36.617442: step: 144/526, loss: 0.015848541632294655 2023-01-21 13:07:37.770627: step: 148/526, loss: 0.021215438842773438 2023-01-21 13:07:38.912834: step: 152/526, loss: 5.674362000718247e-06 2023-01-21 13:07:40.055076: step: 156/526, loss: 0.025526810437440872 2023-01-21 13:07:41.183668: step: 160/526, loss: 0.0024314881302416325 2023-01-21 13:07:42.332851: step: 164/526, loss: 0.0006305694114416838 2023-01-21 13:07:43.435006: step: 168/526, loss: 0.0001483917294535786 2023-01-21 13:07:44.574260: step: 172/526, loss: 0.0022073746658861637 2023-01-21 13:07:45.710845: step: 176/526, loss: 9.117127046920359e-05 2023-01-21 13:07:46.824962: step: 180/526, loss: 0.0001585006684763357 2023-01-21 13:07:47.945541: step: 184/526, loss: 0.04893188551068306 2023-01-21 13:07:49.065791: step: 188/526, loss: 8.03947405074723e-05 2023-01-21 13:07:50.239817: step: 192/526, loss: 0.001464462373405695 2023-01-21 13:07:51.350310: step: 196/526, loss: 3.4046173823298886e-05 2023-01-21 13:07:52.483128: step: 200/526, loss: 1.506805438111769e-05 2023-01-21 13:07:53.617210: step: 204/526, loss: 0.008358573541045189 2023-01-21 13:07:54.727902: step: 208/526, loss: 3.4618376957951114e-05 2023-01-21 13:07:55.847388: step: 212/526, loss: -6.675721124338452e-06 2023-01-21 13:07:56.978496: step: 216/526, loss: 0.032373812049627304 2023-01-21 13:07:58.132762: step: 220/526, loss: 3.2901763916015625e-05 2023-01-21 13:07:59.239921: step: 224/526, loss: 0.00879597757011652 2023-01-21 13:08:00.382425: step: 228/526, loss: 0.003954506013542414 2023-01-21 13:08:01.509168: step: 232/526, loss: 0.001010894775390625 2023-01-21 13:08:02.632906: step: 236/526, loss: 3.461838059592992e-05 2023-01-21 13:08:03.780570: step: 240/526, loss: -7.43865984986769e-06 2023-01-21 13:08:04.915071: step: 244/526, loss: 0.005058479495346546 2023-01-21 13:08:06.049601: step: 248/526, loss: 0.014174079522490501 2023-01-21 13:08:07.189522: step: 252/526, loss: 4.041195279569365e-05 2023-01-21 13:08:08.342115: step: 256/526, loss: 0.11412124335765839 2023-01-21 13:08:09.448544: step: 260/526, loss: 0.0003936767752747983 2023-01-21 13:08:10.594170: step: 264/526, loss: 0.007677269168198109 2023-01-21 13:08:11.695807: step: 268/526, loss: 0.0005231857649050653 2023-01-21 13:08:12.858702: step: 272/526, loss: 0.011204719543457031 2023-01-21 13:08:13.976438: step: 276/526, loss: 0.0002613067626953125 2023-01-21 13:08:15.096685: step: 280/526, loss: 0.081090547144413 2023-01-21 13:08:16.241940: step: 284/526, loss: 0.0013433456188067794 2023-01-21 13:08:17.362256: step: 288/526, loss: 0.00867304764688015 2023-01-21 13:08:18.472892: step: 292/526, loss: 8.258819434558973e-05 2023-01-21 13:08:19.634962: step: 296/526, loss: 0.0012937545543536544 2023-01-21 13:08:20.759907: step: 300/526, loss: 0.0004927635309286416 2023-01-21 13:08:21.877532: step: 304/526, loss: 0.023227596655488014 2023-01-21 13:08:23.020869: step: 308/526, loss: 0.026096869260072708 2023-01-21 13:08:24.150087: step: 312/526, loss: 0.019597148522734642 2023-01-21 13:08:25.286232: step: 316/526, loss: 0.0008947373135015368 2023-01-21 13:08:26.410612: step: 320/526, loss: -7.629391802765895e-07 2023-01-21 13:08:27.528558: step: 324/526, loss: 0.0008543014992028475 2023-01-21 13:08:28.622659: step: 328/526, loss: 9.441375732421875e-05 2023-01-21 13:08:29.739926: step: 332/526, loss: 0.0010314941173419356 2023-01-21 13:08:30.915599: step: 336/526, loss: 0.028462981805205345 2023-01-21 13:08:32.015043: step: 340/526, loss: -9.536742595628311e-08 2023-01-21 13:08:33.142626: step: 344/526, loss: 0.03591118007898331 2023-01-21 13:08:34.283214: step: 348/526, loss: 0.0002541542053222656 2023-01-21 13:08:35.443991: step: 352/526, loss: 0.07203960418701172 2023-01-21 13:08:36.576991: step: 356/526, loss: 0.002747821621596813 2023-01-21 13:08:37.746001: step: 360/526, loss: 0.00025215151254087687 2023-01-21 13:08:38.897833: step: 364/526, loss: 0.0003071785031352192 2023-01-21 13:08:40.010311: step: 368/526, loss: 1.964569128176663e-05 2023-01-21 13:08:41.170570: step: 372/526, loss: 0.0003894805849995464 2023-01-21 13:08:42.287811: step: 376/526, loss: 0.004029655363410711 2023-01-21 13:08:43.435691: step: 380/526, loss: 0.00010814666893566027 2023-01-21 13:08:44.563158: step: 384/526, loss: 0.011087036691606045 2023-01-21 13:08:45.684815: step: 388/526, loss: 0.0011505127185955644 2023-01-21 13:08:46.817775: step: 392/526, loss: 0.0016177654033526778 2023-01-21 13:08:47.973396: step: 396/526, loss: 0.000289726274786517 2023-01-21 13:08:49.133881: step: 400/526, loss: 8.125305612338707e-05 2023-01-21 13:08:50.275078: step: 404/526, loss: 0.00697903661057353 2023-01-21 13:08:51.370421: step: 408/526, loss: 1.0299683708581142e-05 2023-01-21 13:08:52.484183: step: 412/526, loss: -5.722046125811175e-07 2023-01-21 13:08:53.622208: step: 416/526, loss: 0.01753692701458931 2023-01-21 13:08:54.801242: step: 420/526, loss: 0.005646086297929287 2023-01-21 13:08:55.935153: step: 424/526, loss: 0.0003219604550395161 2023-01-21 13:08:57.064312: step: 428/526, loss: 0.004481506533920765 2023-01-21 13:08:58.214111: step: 432/526, loss: 0.013388622552156448 2023-01-21 13:08:59.347377: step: 436/526, loss: 0.00013456345186568797 2023-01-21 13:09:00.472931: step: 440/526, loss: 0.01545257493853569 2023-01-21 13:09:01.587234: step: 444/526, loss: 0.005069160833954811 2023-01-21 13:09:02.723455: step: 448/526, loss: 0.025766562670469284 2023-01-21 13:09:03.849573: step: 452/526, loss: 0.00119065644685179 2023-01-21 13:09:05.025100: step: 456/526, loss: 0.003124427981674671 2023-01-21 13:09:06.147301: step: 460/526, loss: -5.769729796156753e-06 2023-01-21 13:09:07.282537: step: 464/526, loss: 3.108978125965223e-05 2023-01-21 13:09:08.409914: step: 468/526, loss: 0.00037212372990325093 2023-01-21 13:09:09.534668: step: 472/526, loss: 3.2711028325138614e-05 2023-01-21 13:09:10.669269: step: 476/526, loss: 0.0009131432161666453 2023-01-21 13:09:11.817522: step: 480/526, loss: 0.00037479400634765625 2023-01-21 13:09:12.974713: step: 484/526, loss: 1.449584942747606e-05 2023-01-21 13:09:14.105030: step: 488/526, loss: 0.0009846687316894531 2023-01-21 13:09:15.246555: step: 492/526, loss: 0.0003299713134765625 2023-01-21 13:09:16.357917: step: 496/526, loss: 0.004880142398178577 2023-01-21 13:09:17.499830: step: 500/526, loss: 1.6593934560660273e-05 2023-01-21 13:09:18.615178: step: 504/526, loss: 0.0005884647252969444 2023-01-21 13:09:19.756801: step: 508/526, loss: 8.487701052217744e-06 2023-01-21 13:09:20.880106: step: 512/526, loss: 8.087157766567543e-05 2023-01-21 13:09:22.041457: step: 516/526, loss: 0.011248111724853516 2023-01-21 13:09:23.197852: step: 520/526, loss: 0.01270752027630806 2023-01-21 13:09:24.347158: step: 524/526, loss: 0.012514687143266201 2023-01-21 13:09:25.502372: step: 528/526, loss: 3.681182715808973e-05 2023-01-21 13:09:26.663459: step: 532/526, loss: 0.0004520416259765625 2023-01-21 13:09:27.812985: step: 536/526, loss: 0.003604126162827015 2023-01-21 13:09:28.941031: step: 540/526, loss: 6.12258882028982e-05 2023-01-21 13:09:30.067155: step: 544/526, loss: 0.0019002915360033512 2023-01-21 13:09:31.186102: step: 548/526, loss: 0.004593372344970703 2023-01-21 13:09:32.323882: step: 552/526, loss: 0.011429023928940296 2023-01-21 13:09:33.450293: step: 556/526, loss: 0.003665924072265625 2023-01-21 13:09:34.587471: step: 560/526, loss: 0.011133193969726562 2023-01-21 13:09:35.704881: step: 564/526, loss: 0.0021597864106297493 2023-01-21 13:09:36.828138: step: 568/526, loss: 0.006019401829689741 2023-01-21 13:09:37.970880: step: 572/526, loss: 0.10567913204431534 2023-01-21 13:09:39.134001: step: 576/526, loss: 0.0020154952071607113 2023-01-21 13:09:40.290484: step: 580/526, loss: 0.056601669639348984 2023-01-21 13:09:41.433568: step: 584/526, loss: 2.002715973503655e-06 2023-01-21 13:09:42.579572: step: 588/526, loss: 0.0002389907749602571 2023-01-21 13:09:43.714422: step: 592/526, loss: 0.006484031677246094 2023-01-21 13:09:44.885848: step: 596/526, loss: 0.014467430301010609 2023-01-21 13:09:46.001950: step: 600/526, loss: 0.00019912720017600805 2023-01-21 13:09:47.110179: step: 604/526, loss: 3.471374657237902e-05 2023-01-21 13:09:48.236301: step: 608/526, loss: 0.00933847390115261 2023-01-21 13:09:49.373537: step: 612/526, loss: 0.0008516312227584422 2023-01-21 13:09:50.495307: step: 616/526, loss: 0.0021648884285241365 2023-01-21 13:09:51.628241: step: 620/526, loss: 0.0027592184487730265 2023-01-21 13:09:52.756481: step: 624/526, loss: 0.0009501457680016756 2023-01-21 13:09:53.868966: step: 628/526, loss: 0.026536082848906517 2023-01-21 13:09:55.021974: step: 632/526, loss: 0.045363474637269974 2023-01-21 13:09:56.218008: step: 636/526, loss: 0.0002666473446879536 2023-01-21 13:09:57.350121: step: 640/526, loss: 0.037862878292798996 2023-01-21 13:09:58.486993: step: 644/526, loss: 0.00020294188288971782 2023-01-21 13:09:59.636899: step: 648/526, loss: 0.00828852690756321 2023-01-21 13:10:00.782282: step: 652/526, loss: 0.04934177175164223 2023-01-21 13:10:01.915003: step: 656/526, loss: 0.026671981438994408 2023-01-21 13:10:03.053605: step: 660/526, loss: 0.015010691247880459 2023-01-21 13:10:04.175143: step: 664/526, loss: 0.0007877349853515625 2023-01-21 13:10:05.335677: step: 668/526, loss: 0.005845355801284313 2023-01-21 13:10:06.427293: step: 672/526, loss: 9.536744016713783e-08 2023-01-21 13:10:07.592234: step: 676/526, loss: 0.0005153656238690019 2023-01-21 13:10:08.714747: step: 680/526, loss: 0.02548046037554741 2023-01-21 13:10:09.824756: step: 684/526, loss: 0.00014467239088844508 2023-01-21 13:10:10.977190: step: 688/526, loss: 0.0018161773914471269 2023-01-21 13:10:12.106171: step: 692/526, loss: 1.1634827387752011e-05 2023-01-21 13:10:13.284437: step: 696/526, loss: 0.0023976326920092106 2023-01-21 13:10:14.407828: step: 700/526, loss: 0.0037027359940111637 2023-01-21 13:10:15.554529: step: 704/526, loss: -1.1253356205997989e-05 2023-01-21 13:10:16.683554: step: 708/526, loss: 0.1141473799943924 2023-01-21 13:10:17.832128: step: 712/526, loss: 0.012693977914750576 2023-01-21 13:10:18.925975: step: 716/526, loss: 3.814697265625e-06 2023-01-21 13:10:20.081032: step: 720/526, loss: 0.36805686354637146 2023-01-21 13:10:21.205868: step: 724/526, loss: 0.015682794153690338 2023-01-21 13:10:22.344227: step: 728/526, loss: 0.017478562891483307 2023-01-21 13:10:23.478968: step: 732/526, loss: 0.03909454494714737 2023-01-21 13:10:24.659642: step: 736/526, loss: 7.82012939453125e-05 2023-01-21 13:10:25.789714: step: 740/526, loss: 0.0005135297542437911 2023-01-21 13:10:26.946124: step: 744/526, loss: 4.9591067181609105e-06 2023-01-21 13:10:28.131230: step: 748/526, loss: 0.00016045570373535156 2023-01-21 13:10:29.258065: step: 752/526, loss: 0.00846099853515625 2023-01-21 13:10:30.353786: step: 756/526, loss: 0.006530380342155695 2023-01-21 13:10:31.452625: step: 760/526, loss: 0.0071388245560228825 2023-01-21 13:10:32.581924: step: 764/526, loss: 0.0018610954284667969 2023-01-21 13:10:33.690128: step: 768/526, loss: 0.00010395050048828125 2023-01-21 13:10:34.823717: step: 772/526, loss: 0.0001646041782805696 2023-01-21 13:10:35.925668: step: 776/526, loss: 0.034526824951171875 2023-01-21 13:10:37.055434: step: 780/526, loss: 5.078315552964341e-06 2023-01-21 13:10:38.190599: step: 784/526, loss: 0.004657936282455921 2023-01-21 13:10:39.309230: step: 788/526, loss: 0.00014991759962867945 2023-01-21 13:10:40.426253: step: 792/526, loss: 0.0002193450927734375 2023-01-21 13:10:41.555321: step: 796/526, loss: 8.20159948489163e-06 2023-01-21 13:10:42.671036: step: 800/526, loss: 0.0005762100336141884 2023-01-21 13:10:43.786797: step: 804/526, loss: 9.126662916969508e-05 2023-01-21 13:10:44.908868: step: 808/526, loss: 0.06104621663689613 2023-01-21 13:10:46.054119: step: 812/526, loss: 0.0007038116455078125 2023-01-21 13:10:47.156954: step: 816/526, loss: 0.0010636806255206466 2023-01-21 13:10:48.268139: step: 820/526, loss: 0.0005322843790054321 2023-01-21 13:10:49.444685: step: 824/526, loss: 0.000579833984375 2023-01-21 13:10:50.546783: step: 828/526, loss: 8.316040475619957e-05 2023-01-21 13:10:51.660623: step: 832/526, loss: 0.0018756866920739412 2023-01-21 13:10:52.781330: step: 836/526, loss: 0.0017618179554119706 2023-01-21 13:10:53.925834: step: 840/526, loss: 0.03902454674243927 2023-01-21 13:10:55.061280: step: 844/526, loss: 0.04044094309210777 2023-01-21 13:10:56.191298: step: 848/526, loss: 0.04602847248315811 2023-01-21 13:10:57.332735: step: 852/526, loss: 0.012273024767637253 2023-01-21 13:10:58.468160: step: 856/526, loss: 0.0018209457630291581 2023-01-21 13:10:59.592144: step: 860/526, loss: 3.4332276754867053e-06 2023-01-21 13:11:00.728723: step: 864/526, loss: 0.015171051025390625 2023-01-21 13:11:01.831932: step: 868/526, loss: -3.623962356869015e-06 2023-01-21 13:11:02.965047: step: 872/526, loss: 0.0035392760764807463 2023-01-21 13:11:04.109794: step: 876/526, loss: 0.03569640964269638 2023-01-21 13:11:05.245066: step: 880/526, loss: 2.822876012942288e-05 2023-01-21 13:11:06.376501: step: 884/526, loss: 0.22572708129882812 2023-01-21 13:11:07.502264: step: 888/526, loss: 0.004665422718971968 2023-01-21 13:11:08.644282: step: 892/526, loss: 1.430511474609375e-05 2023-01-21 13:11:09.784648: step: 896/526, loss: 0.0027769089210778475 2023-01-21 13:11:10.912547: step: 900/526, loss: 0.0001483917294535786 2023-01-21 13:11:12.036353: step: 904/526, loss: 0.0030285834800451994 2023-01-21 13:11:13.190193: step: 908/526, loss: 0.00257282261736691 2023-01-21 13:11:14.364974: step: 912/526, loss: 0.017187023535370827 2023-01-21 13:11:15.497825: step: 916/526, loss: 0.011454201303422451 2023-01-21 13:11:16.638859: step: 920/526, loss: 0.3685436248779297 2023-01-21 13:11:17.774344: step: 924/526, loss: 0.019519424065947533 2023-01-21 13:11:18.918459: step: 928/526, loss: 0.04216348007321358 2023-01-21 13:11:20.085887: step: 932/526, loss: 0.027657128870487213 2023-01-21 13:11:21.227392: step: 936/526, loss: 0.007106495089828968 2023-01-21 13:11:22.382360: step: 940/526, loss: 0.0003859996795654297 2023-01-21 13:11:23.509099: step: 944/526, loss: 1.4591217222914565e-05 2023-01-21 13:11:24.644533: step: 948/526, loss: 0.0023360252380371094 2023-01-21 13:11:25.799671: step: 952/526, loss: 0.03049764782190323 2023-01-21 13:11:26.945518: step: 956/526, loss: 1.659393274167087e-05 2023-01-21 13:11:28.164399: step: 960/526, loss: -4.291534423828125e-06 2023-01-21 13:11:29.284668: step: 964/526, loss: 0.028342438861727715 2023-01-21 13:11:30.427459: step: 968/526, loss: 0.0014786720275878906 2023-01-21 13:11:31.562527: step: 972/526, loss: 0.007918739691376686 2023-01-21 13:11:32.703560: step: 976/526, loss: 0.0002422332763671875 2023-01-21 13:11:33.816694: step: 980/526, loss: 0.0006341934204101562 2023-01-21 13:11:34.941939: step: 984/526, loss: 0.0004841327609028667 2023-01-21 13:11:36.060790: step: 988/526, loss: 0.00021295547776389867 2023-01-21 13:11:37.203914: step: 992/526, loss: 0.00031604766263626516 2023-01-21 13:11:38.344989: step: 996/526, loss: 0.00686149625107646 2023-01-21 13:11:39.499122: step: 1000/526, loss: 0.04132575914263725 2023-01-21 13:11:40.658514: step: 1004/526, loss: 0.010339736938476562 2023-01-21 13:11:41.795763: step: 1008/526, loss: 0.00044403073843568563 2023-01-21 13:11:42.921627: step: 1012/526, loss: 0.0029230117797851562 2023-01-21 13:11:44.046262: step: 1016/526, loss: 0.0018386841984465718 2023-01-21 13:11:45.160489: step: 1020/526, loss: 0.00011067390732932836 2023-01-21 13:11:46.345288: step: 1024/526, loss: 0.01800861395895481 2023-01-21 13:11:47.470310: step: 1028/526, loss: 0.03227043151855469 2023-01-21 13:11:48.627234: step: 1032/526, loss: 0.024683762341737747 2023-01-21 13:11:49.759260: step: 1036/526, loss: 0.01766028441488743 2023-01-21 13:11:50.888456: step: 1040/526, loss: 2.0790101189049892e-05 2023-01-21 13:11:52.023280: step: 1044/526, loss: 0.0006039619329385459 2023-01-21 13:11:53.135329: step: 1048/526, loss: 0.0011785507667809725 2023-01-21 13:11:54.266801: step: 1052/526, loss: 0.019766902551054955 2023-01-21 13:11:55.434155: step: 1056/526, loss: 0.03691139444708824 2023-01-21 13:11:56.569316: step: 1060/526, loss: 0.0012372970813885331 2023-01-21 13:11:57.719323: step: 1064/526, loss: 1.506805438111769e-05 2023-01-21 13:11:58.855049: step: 1068/526, loss: 0.0003185272216796875 2023-01-21 13:12:00.025479: step: 1072/526, loss: 0.002008295152336359 2023-01-21 13:12:01.139099: step: 1076/526, loss: 0.00754814175888896 2023-01-21 13:12:02.284543: step: 1080/526, loss: 6.179809861350805e-05 2023-01-21 13:12:03.432812: step: 1084/526, loss: 6.732940528308973e-05 2023-01-21 13:12:04.557381: step: 1088/526, loss: 0.00010747909982455894 2023-01-21 13:12:05.707500: step: 1092/526, loss: 0.04722805321216583 2023-01-21 13:12:06.870847: step: 1096/526, loss: 0.006152534857392311 2023-01-21 13:12:08.033425: step: 1100/526, loss: 0.00010032653517555445 2023-01-21 13:12:09.209314: step: 1104/526, loss: 0.008452988229691982 2023-01-21 13:12:10.362189: step: 1108/526, loss: 0.0018753051990643144 2023-01-21 13:12:11.487036: step: 1112/526, loss: 7.152557373046875e-05 2023-01-21 13:12:12.629860: step: 1116/526, loss: 0.014368534088134766 2023-01-21 13:12:13.756002: step: 1120/526, loss: 0.00046539306640625 2023-01-21 13:12:14.887385: step: 1124/526, loss: 2.1743775505456142e-05 2023-01-21 13:12:16.053491: step: 1128/526, loss: 0.021529100835323334 2023-01-21 13:12:17.161649: step: 1132/526, loss: 0.004278755281120539 2023-01-21 13:12:18.299465: step: 1136/526, loss: 0.0002008915034821257 2023-01-21 13:12:19.428258: step: 1140/526, loss: 0.0007059097406454384 2023-01-21 13:12:20.524594: step: 1144/526, loss: 0.00031871796818450093 2023-01-21 13:12:21.644231: step: 1148/526, loss: 0.0003121376212220639 2023-01-21 13:12:22.773907: step: 1152/526, loss: 0.013891697861254215 2023-01-21 13:12:23.919890: step: 1156/526, loss: 0.024333763867616653 2023-01-21 13:12:25.039849: step: 1160/526, loss: 0.0025947571266442537 2023-01-21 13:12:26.150824: step: 1164/526, loss: 0.0001605987490620464 2023-01-21 13:12:27.277740: step: 1168/526, loss: 0.0017118455143645406 2023-01-21 13:12:28.430107: step: 1172/526, loss: 0.003590821987017989 2023-01-21 13:12:29.605497: step: 1176/526, loss: 0.024665450677275658 2023-01-21 13:12:30.754870: step: 1180/526, loss: 0.02268209494650364 2023-01-21 13:12:31.861579: step: 1184/526, loss: 0.0011107444297522306 2023-01-21 13:12:33.000753: step: 1188/526, loss: 2.9373170036706142e-05 2023-01-21 13:12:34.164779: step: 1192/526, loss: 0.02532787248492241 2023-01-21 13:12:35.330896: step: 1196/526, loss: 0.0009091377141885459 2023-01-21 13:12:36.463363: step: 1200/526, loss: 0.003118324326351285 2023-01-21 13:12:37.572637: step: 1204/526, loss: 0.0008320808410644531 2023-01-21 13:12:38.708895: step: 1208/526, loss: 0.00031528473482467234 2023-01-21 13:12:39.857123: step: 1212/526, loss: 0.031816866248846054 2023-01-21 13:12:40.988342: step: 1216/526, loss: 0.0009879112476482987 2023-01-21 13:12:42.134238: step: 1220/526, loss: 0.0006337165832519531 2023-01-21 13:12:43.284206: step: 1224/526, loss: 0.6496032476425171 2023-01-21 13:12:44.516627: step: 1228/526, loss: 1.2874604180979077e-05 2023-01-21 13:12:45.663420: step: 1232/526, loss: 0.0019760846626013517 2023-01-21 13:12:46.775113: step: 1236/526, loss: 0.0003295898495707661 2023-01-21 13:12:47.921239: step: 1240/526, loss: 0.003175163408741355 2023-01-21 13:12:49.057147: step: 1244/526, loss: 0.00020513535127975047 2023-01-21 13:12:50.187063: step: 1248/526, loss: 5.2165985835017636e-05 2023-01-21 13:12:51.320917: step: 1252/526, loss: 3.0517578125e-05 2023-01-21 13:12:52.455996: step: 1256/526, loss: 0.001052475068718195 2023-01-21 13:12:53.618041: step: 1260/526, loss: 0.014911365695297718 2023-01-21 13:12:54.759546: step: 1264/526, loss: 0.0041027069091796875 2023-01-21 13:12:55.902204: step: 1268/526, loss: 0.0008115768432617188 2023-01-21 13:12:57.089026: step: 1272/526, loss: 0.004632758907973766 2023-01-21 13:12:58.218089: step: 1276/526, loss: 0.0002115249662892893 2023-01-21 13:12:59.333168: step: 1280/526, loss: 0.0008586883777752519 2023-01-21 13:13:00.473304: step: 1284/526, loss: 0.0020294189453125 2023-01-21 13:13:01.620501: step: 1288/526, loss: 0.0013658524258062243 2023-01-21 13:13:02.751427: step: 1292/526, loss: 0.000529861485119909 2023-01-21 13:13:03.892537: step: 1296/526, loss: 0.003073024796321988 2023-01-21 13:13:05.014015: step: 1300/526, loss: 0.013638305477797985 2023-01-21 13:13:06.134121: step: 1304/526, loss: 0.0015863418811932206 2023-01-21 13:13:07.243512: step: 1308/526, loss: 0.00022411346435546875 2023-01-21 13:13:08.357109: step: 1312/526, loss: 6.809234764659777e-05 2023-01-21 13:13:09.478356: step: 1316/526, loss: 0.0008967399480752647 2023-01-21 13:13:10.587039: step: 1320/526, loss: 0.011912345886230469 2023-01-21 13:13:11.734628: step: 1324/526, loss: 3.337860107421875e-05 2023-01-21 13:13:12.883724: step: 1328/526, loss: 0.00169200892560184 2023-01-21 13:13:13.996386: step: 1332/526, loss: 0.00022325516329146922 2023-01-21 13:13:15.134093: step: 1336/526, loss: 0.011070584878325462 2023-01-21 13:13:16.296038: step: 1340/526, loss: 5.187988426769152e-05 2023-01-21 13:13:17.452290: step: 1344/526, loss: 1.163482647825731e-05 2023-01-21 13:13:18.575157: step: 1348/526, loss: 0.014441752806305885 2023-01-21 13:13:19.735063: step: 1352/526, loss: 0.00017395020404364914 2023-01-21 13:13:20.876717: step: 1356/526, loss: 0.0003949165402445942 2023-01-21 13:13:22.017190: step: 1360/526, loss: 0.00042104723979718983 2023-01-21 13:13:23.159372: step: 1364/526, loss: 0.000446128862677142 2023-01-21 13:13:24.305588: step: 1368/526, loss: 0.009949063882231712 2023-01-21 13:13:25.435939: step: 1372/526, loss: 0.004015636630356312 2023-01-21 13:13:26.551725: step: 1376/526, loss: 0.0001991748868022114 2023-01-21 13:13:27.703429: step: 1380/526, loss: 8.716584125068039e-05 2023-01-21 13:13:28.833610: step: 1384/526, loss: 0.0002954483206849545 2023-01-21 13:13:29.991061: step: 1388/526, loss: 0.06758461147546768 2023-01-21 13:13:31.116994: step: 1392/526, loss: 0.0004936218028888106 2023-01-21 13:13:32.243050: step: 1396/526, loss: 0.003154182340949774 2023-01-21 13:13:33.383485: step: 1400/526, loss: 0.044655606150627136 2023-01-21 13:13:34.529614: step: 1404/526, loss: 0.025870133191347122 2023-01-21 13:13:35.626106: step: 1408/526, loss: 0.004344177432358265 2023-01-21 13:13:36.752196: step: 1412/526, loss: 0.0005369663122110069 2023-01-21 13:13:37.892358: step: 1416/526, loss: 0.0012508392101153731 2023-01-21 13:13:39.024613: step: 1420/526, loss: 0.00033693312434479594 2023-01-21 13:13:40.181581: step: 1424/526, loss: 0.025281762704253197 2023-01-21 13:13:41.307627: step: 1428/526, loss: 0.0004226684395689517 2023-01-21 13:13:42.420888: step: 1432/526, loss: 2.841949390131049e-05 2023-01-21 13:13:43.610316: step: 1436/526, loss: 0.02558300644159317 2023-01-21 13:13:44.732739: step: 1440/526, loss: 0.0008280754555016756 2023-01-21 13:13:45.862898: step: 1444/526, loss: 0.00592694291844964 2023-01-21 13:13:46.958829: step: 1448/526, loss: 0.006555843632668257 2023-01-21 13:13:48.117471: step: 1452/526, loss: 6.065368506824598e-05 2023-01-21 13:13:49.252702: step: 1456/526, loss: 0.017021847888827324 2023-01-21 13:13:50.394501: step: 1460/526, loss: 0.00017132758512161672 2023-01-21 13:13:51.533603: step: 1464/526, loss: 0.00010938644118141383 2023-01-21 13:13:52.680671: step: 1468/526, loss: 0.030788231641054153 2023-01-21 13:13:53.793790: step: 1472/526, loss: 0.0008681774488650262 2023-01-21 13:13:54.917796: step: 1476/526, loss: 0.024807168170809746 2023-01-21 13:13:56.041550: step: 1480/526, loss: 0.5409547686576843 2023-01-21 13:13:57.174629: step: 1484/526, loss: 0.0005863190162926912 2023-01-21 13:13:58.320778: step: 1488/526, loss: 0.0024511339142918587 2023-01-21 13:13:59.466177: step: 1492/526, loss: 0.003298044204711914 2023-01-21 13:14:00.638173: step: 1496/526, loss: 0.0007040023920126259 2023-01-21 13:14:01.770613: step: 1500/526, loss: 0.015483285300433636 2023-01-21 13:14:02.909835: step: 1504/526, loss: 0.0019904612563550472 2023-01-21 13:14:04.079687: step: 1508/526, loss: 0.006766891572624445 2023-01-21 13:14:05.216629: step: 1512/526, loss: 0.014547348022460938 2023-01-21 13:14:06.333180: step: 1516/526, loss: 0.0007663726573809981 2023-01-21 13:14:07.515884: step: 1520/526, loss: 0.01921234279870987 2023-01-21 13:14:08.649718: step: 1524/526, loss: 0.0017212779494002461 2023-01-21 13:14:09.768130: step: 1528/526, loss: 3.337860107421875e-05 2023-01-21 13:14:10.887357: step: 1532/526, loss: 9.250641596736386e-06 2023-01-21 13:14:12.000976: step: 1536/526, loss: 0.0008694648859091103 2023-01-21 13:14:13.138578: step: 1540/526, loss: 0.007087897974997759 2023-01-21 13:14:14.284020: step: 1544/526, loss: 0.067865751683712 2023-01-21 13:14:15.436605: step: 1548/526, loss: 3.194809323758818e-05 2023-01-21 13:14:16.575403: step: 1552/526, loss: 0.005407905671745539 2023-01-21 13:14:17.715405: step: 1556/526, loss: 0.00016927719116210938 2023-01-21 13:14:18.846853: step: 1560/526, loss: 0.05326547846198082 2023-01-21 13:14:19.985304: step: 1564/526, loss: 0.044206809252500534 2023-01-21 13:14:21.103644: step: 1568/526, loss: 0.01596670225262642 2023-01-21 13:14:22.243747: step: 1572/526, loss: 0.00010375976853538305 2023-01-21 13:14:23.386767: step: 1576/526, loss: 0.06024513393640518 2023-01-21 13:14:24.511926: step: 1580/526, loss: 2.8944017685716972e-05 2023-01-21 13:14:25.668592: step: 1584/526, loss: 4.00543212890625e-05 2023-01-21 13:14:26.805736: step: 1588/526, loss: -4.38690176451928e-06 2023-01-21 13:14:27.927443: step: 1592/526, loss: 0.0002382278471486643 2023-01-21 13:14:29.047837: step: 1596/526, loss: 0.016375351697206497 2023-01-21 13:14:30.208492: step: 1600/526, loss: 0.0009118079906329513 2023-01-21 13:14:31.360322: step: 1604/526, loss: 0.013803672976791859 2023-01-21 13:14:32.497746: step: 1608/526, loss: 0.03435096517205238 2023-01-21 13:14:33.638415: step: 1612/526, loss: 0.025008726865053177 2023-01-21 13:14:34.771393: step: 1616/526, loss: 0.00011234283010708168 2023-01-21 13:14:35.911910: step: 1620/526, loss: 0.06903248280286789 2023-01-21 13:14:37.030574: step: 1624/526, loss: 0.00012454987154342234 2023-01-21 13:14:38.174014: step: 1628/526, loss: 0.006815052125602961 2023-01-21 13:14:39.313659: step: 1632/526, loss: 0.0050216675736010075 2023-01-21 13:14:40.482456: step: 1636/526, loss: 0.0023239136207848787 2023-01-21 13:14:41.645023: step: 1640/526, loss: 0.00264568324200809 2023-01-21 13:14:42.775289: step: 1644/526, loss: 0.004068136215209961 2023-01-21 13:14:43.906116: step: 1648/526, loss: 0.012709570117294788 2023-01-21 13:14:45.046155: step: 1652/526, loss: 0.0004522323433775455 2023-01-21 13:14:46.150721: step: 1656/526, loss: 0.09492845833301544 2023-01-21 13:14:47.300391: step: 1660/526, loss: 1.52587890625e-05 2023-01-21 13:14:48.463675: step: 1664/526, loss: 0.002302408218383789 2023-01-21 13:14:49.647422: step: 1668/526, loss: 0.005969524383544922 2023-01-21 13:14:50.803869: step: 1672/526, loss: 0.0042438507080078125 2023-01-21 13:14:51.951381: step: 1676/526, loss: 0.0013067246181890368 2023-01-21 13:14:53.086097: step: 1680/526, loss: 0.020853521302342415 2023-01-21 13:14:54.235207: step: 1684/526, loss: 0.00040068625821731985 2023-01-21 13:14:55.406523: step: 1688/526, loss: 8.735657320357859e-05 2023-01-21 13:14:56.579502: step: 1692/526, loss: 0.0009587288368493319 2023-01-21 13:14:57.718819: step: 1696/526, loss: 2.555847095209174e-05 2023-01-21 13:14:58.857527: step: 1700/526, loss: 0.00028171538724564016 2023-01-21 13:15:00.012969: step: 1704/526, loss: 0.039979077875614166 2023-01-21 13:15:01.124386: step: 1708/526, loss: 0.0009181976784020662 2023-01-21 13:15:02.269067: step: 1712/526, loss: 0.058440208435058594 2023-01-21 13:15:03.416882: step: 1716/526, loss: 0.0007172584882937372 2023-01-21 13:15:04.543661: step: 1720/526, loss: 6.341934204101562e-05 2023-01-21 13:15:05.679244: step: 1724/526, loss: 0.002150630811229348 2023-01-21 13:15:06.819400: step: 1728/526, loss: 0.0019634247291833162 2023-01-21 13:15:07.950947: step: 1732/526, loss: 0.0008495331276208162 2023-01-21 13:15:09.070868: step: 1736/526, loss: 0.00019321442232467234 2023-01-21 13:15:10.207277: step: 1740/526, loss: 2.9850005375919864e-05 2023-01-21 13:15:11.336670: step: 1744/526, loss: 0.0001634597865631804 2023-01-21 13:15:12.486012: step: 1748/526, loss: 0.21984100341796875 2023-01-21 13:15:13.626415: step: 1752/526, loss: 0.006243038456887007 2023-01-21 13:15:14.763959: step: 1756/526, loss: 0.00016789436631370336 2023-01-21 13:15:15.892132: step: 1760/526, loss: 0.008794593624770641 2023-01-21 13:15:17.020408: step: 1764/526, loss: 0.016346361488103867 2023-01-21 13:15:18.187768: step: 1768/526, loss: 0.00031871796818450093 2023-01-21 13:15:19.298752: step: 1772/526, loss: 0.01442489679902792 2023-01-21 13:15:20.430960: step: 1776/526, loss: 0.011054611764848232 2023-01-21 13:15:21.579132: step: 1780/526, loss: 0.021292496472597122 2023-01-21 13:15:22.726274: step: 1784/526, loss: 0.00636787386611104 2023-01-21 13:15:23.846659: step: 1788/526, loss: 2.5177003408316523e-05 2023-01-21 13:15:24.962022: step: 1792/526, loss: 0.008274412713944912 2023-01-21 13:15:26.118498: step: 1796/526, loss: 0.005169105716049671 2023-01-21 13:15:27.279157: step: 1800/526, loss: 0.046772003173828125 2023-01-21 13:15:28.420670: step: 1804/526, loss: 2.555847095209174e-05 2023-01-21 13:15:29.555500: step: 1808/526, loss: 0.00038657188997603953 2023-01-21 13:15:30.698001: step: 1812/526, loss: 0.00029544829158112407 2023-01-21 13:15:31.854966: step: 1816/526, loss: 0.01974639855325222 2023-01-21 13:15:33.009617: step: 1820/526, loss: 3.0040740966796875e-05 2023-01-21 13:15:34.128996: step: 1824/526, loss: 0.00049591064453125 2023-01-21 13:15:35.256260: step: 1828/526, loss: 0.012825775891542435 2023-01-21 13:15:36.454992: step: 1832/526, loss: 1.945495569088962e-05 2023-01-21 13:15:37.572277: step: 1836/526, loss: 0.02095937728881836 2023-01-21 13:15:38.738341: step: 1840/526, loss: 0.0006735802162438631 2023-01-21 13:15:39.884028: step: 1844/526, loss: 0.002120590303093195 2023-01-21 13:15:41.018006: step: 1848/526, loss: 0.03468761220574379 2023-01-21 13:15:42.150345: step: 1852/526, loss: 2.0980833141948096e-06 2023-01-21 13:15:43.273135: step: 1856/526, loss: 0.00227527623064816 2023-01-21 13:15:44.390603: step: 1860/526, loss: -3.051757857974735e-06 2023-01-21 13:15:45.565171: step: 1864/526, loss: 0.005071449093520641 2023-01-21 13:15:46.681606: step: 1868/526, loss: 0.010010338388383389 2023-01-21 13:15:47.826273: step: 1872/526, loss: 0.003715706057846546 2023-01-21 13:15:48.956365: step: 1876/526, loss: 7.190704491222277e-05 2023-01-21 13:15:50.109443: step: 1880/526, loss: 0.0013969421852380037 2023-01-21 13:15:51.254214: step: 1884/526, loss: 0.000670957553666085 2023-01-21 13:15:52.362913: step: 1888/526, loss: 0.0007257461547851562 2023-01-21 13:15:53.506595: step: 1892/526, loss: 7.02858014847152e-05 2023-01-21 13:15:54.638677: step: 1896/526, loss: 0.0012634277809411287 2023-01-21 13:15:55.794409: step: 1900/526, loss: 0.0018157958984375 2023-01-21 13:15:56.921646: step: 1904/526, loss: 0.06537322700023651 2023-01-21 13:15:58.042329: step: 1908/526, loss: 0.35081273317337036 2023-01-21 13:15:59.163471: step: 1912/526, loss: 0.005593585781753063 2023-01-21 13:16:00.311460: step: 1916/526, loss: 2.474784923833795e-05 2023-01-21 13:16:01.450034: step: 1920/526, loss: 0.03758401796221733 2023-01-21 13:16:02.576641: step: 1924/526, loss: 0.05770282819867134 2023-01-21 13:16:03.673809: step: 1928/526, loss: 0.027628041803836823 2023-01-21 13:16:04.816133: step: 1932/526, loss: 0.002439403673633933 2023-01-21 13:16:05.950769: step: 1936/526, loss: 1.773834264895413e-05 2023-01-21 13:16:07.083400: step: 1940/526, loss: 8.301735215354711e-05 2023-01-21 13:16:08.225070: step: 1944/526, loss: 0.000713348388671875 2023-01-21 13:16:09.340945: step: 1948/526, loss: 0.019584273919463158 2023-01-21 13:16:10.491584: step: 1952/526, loss: 0.0038118362426757812 2023-01-21 13:16:11.666429: step: 1956/526, loss: 4.4298172724666074e-05 2023-01-21 13:16:12.784056: step: 1960/526, loss: 4.5585635234601796e-05 2023-01-21 13:16:13.935687: step: 1964/526, loss: 0.014797830954194069 2023-01-21 13:16:15.048217: step: 1968/526, loss: 0.0013304711319506168 2023-01-21 13:16:16.171930: step: 1972/526, loss: 5.5122378398664296e-05 2023-01-21 13:16:17.315402: step: 1976/526, loss: 0.00906076468527317 2023-01-21 13:16:18.467584: step: 1980/526, loss: 9.074211266124621e-05 2023-01-21 13:16:19.587660: step: 1984/526, loss: 0.0001579284726176411 2023-01-21 13:16:20.754017: step: 1988/526, loss: 0.0009207725524902344 2023-01-21 13:16:21.910524: step: 1992/526, loss: 0.0011457443470135331 2023-01-21 13:16:23.069073: step: 1996/526, loss: 0.02708606794476509 2023-01-21 13:16:24.213749: step: 2000/526, loss: 4.75883498438634e-05 2023-01-21 13:16:25.371041: step: 2004/526, loss: 0.03226013109087944 2023-01-21 13:16:26.521361: step: 2008/526, loss: 0.0427764393389225 2023-01-21 13:16:27.710156: step: 2012/526, loss: 0.0005900383112020791 2023-01-21 13:16:28.849357: step: 2016/526, loss: 0.06280136853456497 2023-01-21 13:16:30.012492: step: 2020/526, loss: 8.411407907260582e-05 2023-01-21 13:16:31.149250: step: 2024/526, loss: 0.00017490386380814016 2023-01-21 13:16:32.275545: step: 2028/526, loss: 0.015264797024428844 2023-01-21 13:16:33.369265: step: 2032/526, loss: 9.250640687241685e-06 2023-01-21 13:16:34.501432: step: 2036/526, loss: 0.11339378356933594 2023-01-21 13:16:35.640722: step: 2040/526, loss: 0.0010231018532067537 2023-01-21 13:16:36.772438: step: 2044/526, loss: 0.0001900196075439453 2023-01-21 13:16:37.921102: step: 2048/526, loss: 1.888275073724799e-05 2023-01-21 13:16:39.030498: step: 2052/526, loss: 0.02208881452679634 2023-01-21 13:16:40.187119: step: 2056/526, loss: 0.00037832261295989156 2023-01-21 13:16:41.348963: step: 2060/526, loss: 0.0033551694359630346 2023-01-21 13:16:42.475944: step: 2064/526, loss: 4.751682354253717e-05 2023-01-21 13:16:43.627446: step: 2068/526, loss: 0.026750946417450905 2023-01-21 13:16:44.752187: step: 2072/526, loss: 0.0016255378723144531 2023-01-21 13:16:45.899769: step: 2076/526, loss: 0.02459249459207058 2023-01-21 13:16:47.050845: step: 2080/526, loss: 0.002175712725147605 2023-01-21 13:16:48.185740: step: 2084/526, loss: 0.0005808353307656944 2023-01-21 13:16:49.312833: step: 2088/526, loss: 2.1600722902803682e-05 2023-01-21 13:16:50.480536: step: 2092/526, loss: 0.023267697542905807 2023-01-21 13:16:51.602012: step: 2096/526, loss: 0.013891267590224743 2023-01-21 13:16:52.732649: step: 2100/526, loss: 0.0023319243919104338 2023-01-21 13:16:53.859427: step: 2104/526, loss: 0.0018161773914471269 ================================================== Loss: 0.015 -------------------- Dev: {'event': {'p': 0.6187904967602592, 'r': 0.762982689747004, 'f1': 0.6833631484794276}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Test: {'event': {'p': 0.6157007809288944, 'r': 0.7818371607515657, 'f1': 0.6888939986203725}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Chinese: {'event': {'p': 0.6125, 'r': 0.9074074074074074, 'f1': 0.7313432835820897}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Korean: {'event': {'p': 0.6904761904761905, 'r': 0.4603174603174603, 'f1': 0.5523809523809524}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} Russian: {'event': {'p': 0.475, 'r': 0.5277777777777778, 'f1': 0.5}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 25} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 26 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:17:37.339084: step: 4/526, loss: 0.007556724362075329 2023-01-21 13:17:38.495248: step: 8/526, loss: 0.0009531020768918097 2023-01-21 13:17:39.596388: step: 12/526, loss: 0.0008697509765625 2023-01-21 13:17:40.725109: step: 16/526, loss: 0.04036998748779297 2023-01-21 13:17:41.856037: step: 20/526, loss: 6.27517729299143e-05 2023-01-21 13:17:42.999473: step: 24/526, loss: 0.000240325927734375 2023-01-21 13:17:44.142290: step: 28/526, loss: 0.00011749267287086695 2023-01-21 13:17:45.276870: step: 32/526, loss: 0.00029087066650390625 2023-01-21 13:17:46.412923: step: 36/526, loss: 2.5081633793888614e-05 2023-01-21 13:17:47.555407: step: 40/526, loss: 0.016751719638705254 2023-01-21 13:17:48.694448: step: 44/526, loss: 0.00027332306490279734 2023-01-21 13:17:49.882710: step: 48/526, loss: 1.1730194273695815e-05 2023-01-21 13:17:51.023568: step: 52/526, loss: 0.0001827240048442036 2023-01-21 13:17:52.201249: step: 56/526, loss: 0.0011865615379065275 2023-01-21 13:17:53.349641: step: 60/526, loss: 0.015289688482880592 2023-01-21 13:17:54.487907: step: 64/526, loss: 0.0006167411920614541 2023-01-21 13:17:55.640449: step: 68/526, loss: 0.001567840576171875 2023-01-21 13:17:56.768613: step: 72/526, loss: 0.00014972686767578125 2023-01-21 13:17:57.894637: step: 76/526, loss: 0.09377880394458771 2023-01-21 13:17:59.042447: step: 80/526, loss: 0.0002579212305136025 2023-01-21 13:18:00.194486: step: 84/526, loss: 0.012417316436767578 2023-01-21 13:18:01.330797: step: 88/526, loss: 3.776550147449598e-05 2023-01-21 13:18:02.479394: step: 92/526, loss: 0.0004746437189169228 2023-01-21 13:18:03.609839: step: 96/526, loss: 0.0002685546933207661 2023-01-21 13:18:04.730132: step: 100/526, loss: 6.074905104469508e-05 2023-01-21 13:18:05.830758: step: 104/526, loss: 0.00011405945406295359 2023-01-21 13:18:06.967375: step: 108/526, loss: 6.408691115211695e-05 2023-01-21 13:18:08.100776: step: 112/526, loss: 0.0005860328674316406 2023-01-21 13:18:09.208508: step: 116/526, loss: 0.019515324383974075 2023-01-21 13:18:10.351444: step: 120/526, loss: 0.00012149811664130539 2023-01-21 13:18:11.505922: step: 124/526, loss: 0.00013446807861328125 2023-01-21 13:18:12.651426: step: 128/526, loss: 0.00024337769718840718 2023-01-21 13:18:13.797078: step: 132/526, loss: 4.558563159662299e-05 2023-01-21 13:18:14.913359: step: 136/526, loss: 3.185272362316027e-05 2023-01-21 13:18:16.051843: step: 140/526, loss: 4.119873119634576e-05 2023-01-21 13:18:17.184241: step: 144/526, loss: 0.004657173529267311 2023-01-21 13:18:18.300204: step: 148/526, loss: 0.0016197204822674394 2023-01-21 13:18:19.431520: step: 152/526, loss: 8.640289161121473e-05 2023-01-21 13:18:20.567601: step: 156/526, loss: 0.0006341457483358681 2023-01-21 13:18:21.678097: step: 160/526, loss: 0.00833044108003378 2023-01-21 13:18:22.805887: step: 164/526, loss: 0.07270584255456924 2023-01-21 13:18:23.936406: step: 168/526, loss: 0.001204681466333568 2023-01-21 13:18:25.059047: step: 172/526, loss: 5.283355858409777e-05 2023-01-21 13:18:26.218445: step: 176/526, loss: 0.0007898330804891884 2023-01-21 13:18:27.377306: step: 180/526, loss: 1.3351440202313825e-06 2023-01-21 13:18:28.523602: step: 184/526, loss: 0.025954438373446465 2023-01-21 13:18:29.653023: step: 188/526, loss: 0.03493804857134819 2023-01-21 13:18:30.786765: step: 192/526, loss: 0.015324211679399014 2023-01-21 13:18:31.915498: step: 196/526, loss: 1.926422191900201e-05 2023-01-21 13:18:33.059927: step: 200/526, loss: 0.0002452850458212197 2023-01-21 13:18:34.212827: step: 204/526, loss: 0.0545385368168354 2023-01-21 13:18:35.332582: step: 208/526, loss: 0.0003283977566752583 2023-01-21 13:18:36.448808: step: 212/526, loss: 0.009693384170532227 2023-01-21 13:18:37.603647: step: 216/526, loss: 0.02314901351928711 2023-01-21 13:18:38.744297: step: 220/526, loss: 0.0013080596690997481 2023-01-21 13:18:39.852363: step: 224/526, loss: 1.5354155038949102e-05 2023-01-21 13:18:40.987348: step: 228/526, loss: 0.001273346017114818 2023-01-21 13:18:42.110710: step: 232/526, loss: -6.198883056640625e-06 2023-01-21 13:18:43.233022: step: 236/526, loss: 0.04493580013513565 2023-01-21 13:18:44.366271: step: 240/526, loss: 0.0015001296997070312 2023-01-21 13:18:45.490969: step: 244/526, loss: 0.019270705059170723 2023-01-21 13:18:46.634439: step: 248/526, loss: 0.00045018194941803813 2023-01-21 13:18:47.746743: step: 252/526, loss: 0.008627128787338734 2023-01-21 13:18:48.869157: step: 256/526, loss: 0.004911995027214289 2023-01-21 13:18:49.994115: step: 260/526, loss: 0.007460975553840399 2023-01-21 13:18:51.115233: step: 264/526, loss: 0.00738677941262722 2023-01-21 13:18:52.265870: step: 268/526, loss: 0.037473488599061966 2023-01-21 13:18:53.371674: step: 272/526, loss: 0.0009285926935262978 2023-01-21 13:18:54.516665: step: 276/526, loss: 0.0006183624500408769 2023-01-21 13:18:55.696137: step: 280/526, loss: 2.613067590573337e-05 2023-01-21 13:18:56.846794: step: 284/526, loss: 0.00033843517303466797 2023-01-21 13:18:57.987802: step: 288/526, loss: 0.0003466606140136719 2023-01-21 13:18:59.139490: step: 292/526, loss: 0.019819164648652077 2023-01-21 13:19:00.265510: step: 296/526, loss: 3.681182715808973e-05 2023-01-21 13:19:01.403395: step: 300/526, loss: 0.00012569426326081157 2023-01-21 13:19:02.548782: step: 304/526, loss: 3.566742088878527e-05 2023-01-21 13:19:03.699825: step: 308/526, loss: 0.017421722412109375 2023-01-21 13:19:04.820423: step: 312/526, loss: 0.00769434031099081 2023-01-21 13:19:05.932477: step: 316/526, loss: 0.029249096289277077 2023-01-21 13:19:07.115074: step: 320/526, loss: 0.20623761415481567 2023-01-21 13:19:08.259250: step: 324/526, loss: 3.8719179428881034e-05 2023-01-21 13:19:09.376272: step: 328/526, loss: 0.006597327999770641 2023-01-21 13:19:10.511278: step: 332/526, loss: 0.017350196838378906 2023-01-21 13:19:11.642208: step: 336/526, loss: 4.234314110362902e-05 2023-01-21 13:19:12.781269: step: 340/526, loss: 0.00074939732439816 2023-01-21 13:19:13.913693: step: 344/526, loss: 0.00010004043724620715 2023-01-21 13:19:15.052791: step: 348/526, loss: 0.0003980636829510331 2023-01-21 13:19:16.184335: step: 352/526, loss: -4.2915337417070987e-07 2023-01-21 13:19:17.303030: step: 356/526, loss: 0.009605025872588158 2023-01-21 13:19:18.446777: step: 360/526, loss: 0.0008918762323446572 2023-01-21 13:19:19.557840: step: 364/526, loss: 0.01379938144236803 2023-01-21 13:19:20.668353: step: 368/526, loss: 0.004503726959228516 2023-01-21 13:19:21.839652: step: 372/526, loss: 0.008883095346391201 2023-01-21 13:19:22.980884: step: 376/526, loss: 0.004547023680061102 2023-01-21 13:19:24.127904: step: 380/526, loss: 0.0017359734047204256 2023-01-21 13:19:25.254580: step: 384/526, loss: 0.001637363457120955 2023-01-21 13:19:26.392058: step: 388/526, loss: 8.955002704169601e-05 2023-01-21 13:19:27.519222: step: 392/526, loss: 0.01388168428093195 2023-01-21 13:19:28.658705: step: 396/526, loss: 0.00014228820509742945 2023-01-21 13:19:29.790763: step: 400/526, loss: 0.0016307354671880603 2023-01-21 13:19:30.930472: step: 404/526, loss: 3.452301098150201e-05 2023-01-21 13:19:32.103885: step: 408/526, loss: 0.0004264831659384072 2023-01-21 13:19:33.248375: step: 412/526, loss: 0.00015449525380972773 2023-01-21 13:19:34.384346: step: 416/526, loss: 0.000974464463070035 2023-01-21 13:19:35.544660: step: 420/526, loss: 6.513595872092992e-05 2023-01-21 13:19:36.689844: step: 424/526, loss: 0.009265899658203125 2023-01-21 13:19:37.805926: step: 428/526, loss: 0.0015789985191076994 2023-01-21 13:19:38.961066: step: 432/526, loss: 0.00357818603515625 2023-01-21 13:19:40.087371: step: 436/526, loss: 8.935928053688258e-05 2023-01-21 13:19:41.258495: step: 440/526, loss: 0.005554104223847389 2023-01-21 13:19:42.388080: step: 444/526, loss: 6.198883056640625e-06 2023-01-21 13:19:43.518045: step: 448/526, loss: 3.404617018532008e-05 2023-01-21 13:19:44.667789: step: 452/526, loss: 0.0028277398087084293 2023-01-21 13:19:45.804953: step: 456/526, loss: 0.05595993995666504 2023-01-21 13:19:46.911614: step: 460/526, loss: 1.354217511106981e-05 2023-01-21 13:19:48.061099: step: 464/526, loss: 0.009227370843291283 2023-01-21 13:19:49.193110: step: 468/526, loss: 0.0004972458118572831 2023-01-21 13:19:50.350444: step: 472/526, loss: 0.00037469863309524953 2023-01-21 13:19:51.467505: step: 476/526, loss: 0.0009104252094402909 2023-01-21 13:19:52.600745: step: 480/526, loss: 0.00010395050048828125 2023-01-21 13:19:53.702104: step: 484/526, loss: 5.5122378398664296e-05 2023-01-21 13:19:54.830439: step: 488/526, loss: 0.006502723786979914 2023-01-21 13:19:55.970640: step: 492/526, loss: 0.00020742416381835938 2023-01-21 13:19:57.140473: step: 496/526, loss: 4.940032886224799e-05 2023-01-21 13:19:58.290562: step: 500/526, loss: 0.022144699469208717 2023-01-21 13:19:59.433032: step: 504/526, loss: 0.0005148887867107987 2023-01-21 13:20:00.543213: step: 508/526, loss: 0.0037399292923510075 2023-01-21 13:20:01.683656: step: 512/526, loss: 4.863738922722405e-06 2023-01-21 13:20:02.812983: step: 516/526, loss: 0.0009296417119912803 2023-01-21 13:20:03.927484: step: 520/526, loss: 0.007831573486328125 2023-01-21 13:20:05.056247: step: 524/526, loss: 0.016057778149843216 2023-01-21 13:20:06.192695: step: 528/526, loss: 0.009468460455536842 2023-01-21 13:20:07.303377: step: 532/526, loss: 1.8024444216280244e-05 2023-01-21 13:20:08.431739: step: 536/526, loss: 0.0016936302417889237 2023-01-21 13:20:09.542969: step: 540/526, loss: 0.0018484116299077868 2023-01-21 13:20:10.674506: step: 544/526, loss: 0.00066204066388309 2023-01-21 13:20:11.818146: step: 548/526, loss: 0.00150556571315974 2023-01-21 13:20:12.988593: step: 552/526, loss: 0.08858375251293182 2023-01-21 13:20:14.103007: step: 556/526, loss: 2.0694733393611386e-05 2023-01-21 13:20:15.225769: step: 560/526, loss: 0.02140979655086994 2023-01-21 13:20:16.362385: step: 564/526, loss: 6.132126145530492e-05 2023-01-21 13:20:17.502897: step: 568/526, loss: 4.606247239280492e-05 2023-01-21 13:20:18.638670: step: 572/526, loss: 0.02465667575597763 2023-01-21 13:20:19.801167: step: 576/526, loss: -2.2602082026423886e-05 2023-01-21 13:20:20.919719: step: 580/526, loss: 0.001093959785066545 2023-01-21 13:20:22.064546: step: 584/526, loss: 0.0027973176911473274 2023-01-21 13:20:23.208830: step: 588/526, loss: 0.0019176483619958162 2023-01-21 13:20:24.345772: step: 592/526, loss: 0.01902599260210991 2023-01-21 13:20:25.459595: step: 596/526, loss: 0.00029392243595793843 2023-01-21 13:20:26.596160: step: 600/526, loss: 0.09647312760353088 2023-01-21 13:20:27.737196: step: 604/526, loss: 0.0016601562965661287 2023-01-21 13:20:28.868733: step: 608/526, loss: 2.784729076665826e-05 2023-01-21 13:20:29.988084: step: 612/526, loss: 0.017209911718964577 2023-01-21 13:20:31.101451: step: 616/526, loss: 4.00543194700731e-06 2023-01-21 13:20:32.233483: step: 620/526, loss: 1.7642974853515625e-05 2023-01-21 13:20:33.389417: step: 624/526, loss: 0.0022843361366540194 2023-01-21 13:20:34.506795: step: 628/526, loss: 0.0005439758533611894 2023-01-21 13:20:35.663984: step: 632/526, loss: 0.01005172822624445 2023-01-21 13:20:36.788573: step: 636/526, loss: 2.3174286980065517e-05 2023-01-21 13:20:37.925260: step: 640/526, loss: -2.8610230629055877e-07 2023-01-21 13:20:39.039613: step: 644/526, loss: 0.008971786126494408 2023-01-21 13:20:40.206630: step: 648/526, loss: 3.0517578125e-05 2023-01-21 13:20:41.343346: step: 652/526, loss: 0.06203971058130264 2023-01-21 13:20:42.499961: step: 656/526, loss: 0.002213096711784601 2023-01-21 13:20:43.665140: step: 660/526, loss: 0.00011129378981422633 2023-01-21 13:20:44.800867: step: 664/526, loss: 0.021052028983831406 2023-01-21 13:20:45.938318: step: 668/526, loss: 0.002640247344970703 2023-01-21 13:20:47.097082: step: 672/526, loss: 0.012523031793534756 2023-01-21 13:20:48.280950: step: 676/526, loss: 0.001190781593322754 2023-01-21 13:20:49.436884: step: 680/526, loss: 0.0007537842029705644 2023-01-21 13:20:50.610509: step: 684/526, loss: 0.35065269470214844 2023-01-21 13:20:51.778230: step: 688/526, loss: 0.0002990722714457661 2023-01-21 13:20:52.905430: step: 692/526, loss: 0.009204578585922718 2023-01-21 13:20:54.017805: step: 696/526, loss: 0.009508133865892887 2023-01-21 13:20:55.167503: step: 700/526, loss: 0.00028228759765625 2023-01-21 13:20:56.325162: step: 704/526, loss: 0.0004445076046977192 2023-01-21 13:20:57.470464: step: 708/526, loss: 0.0019628526642918587 2023-01-21 13:20:58.611392: step: 712/526, loss: 0.002403402468189597 2023-01-21 13:20:59.727074: step: 716/526, loss: 0.018903542309999466 2023-01-21 13:21:00.863630: step: 720/526, loss: 0.0007581710815429688 2023-01-21 13:21:01.979498: step: 724/526, loss: 0.12390461564064026 2023-01-21 13:21:03.131333: step: 728/526, loss: 0.0007229328039102256 2023-01-21 13:21:04.245112: step: 732/526, loss: 0.005139350891113281 2023-01-21 13:21:05.363255: step: 736/526, loss: 2.6702880859375e-05 2023-01-21 13:21:06.499988: step: 740/526, loss: 0.012955952435731888 2023-01-21 13:21:07.602913: step: 744/526, loss: 3.957748504035408e-06 2023-01-21 13:21:08.720265: step: 748/526, loss: 6.0081479205109645e-06 2023-01-21 13:21:09.879667: step: 752/526, loss: 0.009667587466537952 2023-01-21 13:21:11.004131: step: 756/526, loss: 0.012731075286865234 2023-01-21 13:21:12.162488: step: 760/526, loss: 0.00022697450185660273 2023-01-21 13:21:13.362797: step: 764/526, loss: 0.03037109225988388 2023-01-21 13:21:14.488041: step: 768/526, loss: 0.09213504940271378 2023-01-21 13:21:15.619594: step: 772/526, loss: 7.62939453125e-06 2023-01-21 13:21:16.767476: step: 776/526, loss: 7.543564424850047e-05 2023-01-21 13:21:17.893250: step: 780/526, loss: 4.692077709478326e-05 2023-01-21 13:21:19.018528: step: 784/526, loss: 3.25202927342616e-05 2023-01-21 13:21:20.141082: step: 788/526, loss: 0.0003370285267010331 2023-01-21 13:21:21.269736: step: 792/526, loss: 0.005739402957260609 2023-01-21 13:21:22.418413: step: 796/526, loss: 0.0013688086764886975 2023-01-21 13:21:23.545506: step: 800/526, loss: 7.286071922862902e-05 2023-01-21 13:21:24.673855: step: 804/526, loss: 6.46591215627268e-05 2023-01-21 13:21:25.801837: step: 808/526, loss: 0.023021508008241653 2023-01-21 13:21:26.921673: step: 812/526, loss: 0.0003647804260253906 2023-01-21 13:21:28.072244: step: 816/526, loss: 0.02536792680621147 2023-01-21 13:21:29.214374: step: 820/526, loss: 0.00819616299122572 2023-01-21 13:21:30.325105: step: 824/526, loss: 0.04157199710607529 2023-01-21 13:21:31.426318: step: 828/526, loss: 0.02962922677397728 2023-01-21 13:21:32.640490: step: 832/526, loss: 0.0003867149353027344 2023-01-21 13:21:33.794080: step: 836/526, loss: 0.008469581604003906 2023-01-21 13:21:34.941322: step: 840/526, loss: 0.00365447998046875 2023-01-21 13:21:36.065773: step: 844/526, loss: 2.2792815798311494e-05 2023-01-21 13:21:37.214510: step: 848/526, loss: 0.0005125046009197831 2023-01-21 13:21:38.398808: step: 852/526, loss: 4.172772241872735e-05 2023-01-21 13:21:39.554662: step: 856/526, loss: 3.52859501617786e-06 2023-01-21 13:21:40.710796: step: 860/526, loss: 0.007661628536880016 2023-01-21 13:21:41.812227: step: 864/526, loss: 3.7765505112474784e-05 2023-01-21 13:21:42.928103: step: 868/526, loss: 0.03570890426635742 2023-01-21 13:21:44.065394: step: 872/526, loss: 0.04351940006017685 2023-01-21 13:21:45.187684: step: 876/526, loss: 0.0012695312034338713 2023-01-21 13:21:46.331557: step: 880/526, loss: 7.181167893577367e-05 2023-01-21 13:21:47.460296: step: 884/526, loss: 0.00034275054349564016 2023-01-21 13:21:48.625682: step: 888/526, loss: 4.19616708313697e-06 2023-01-21 13:21:49.766801: step: 892/526, loss: 0.0003177642938680947 2023-01-21 13:21:50.928356: step: 896/526, loss: 0.02492561563849449 2023-01-21 13:21:52.030595: step: 900/526, loss: -1.754760705807712e-05 2023-01-21 13:21:53.164968: step: 904/526, loss: 0.0002747535763774067 2023-01-21 13:21:54.315527: step: 908/526, loss: 0.0001509666326455772 2023-01-21 13:21:55.440599: step: 912/526, loss: 0.040938328951597214 2023-01-21 13:21:56.590917: step: 916/526, loss: 0.0013833999400958419 2023-01-21 13:21:57.718107: step: 920/526, loss: 0.00010170936729991809 2023-01-21 13:21:58.883316: step: 924/526, loss: 0.0002418518124613911 2023-01-21 13:22:00.022754: step: 928/526, loss: 0.00029201508732512593 2023-01-21 13:22:01.181331: step: 932/526, loss: 0.00101642613299191 2023-01-21 13:22:02.321971: step: 936/526, loss: 0.0035598755348473787 2023-01-21 13:22:03.435957: step: 940/526, loss: 4.053116208524443e-05 2023-01-21 13:22:04.604074: step: 944/526, loss: 4.901886131847277e-05 2023-01-21 13:22:05.748193: step: 948/526, loss: 0.010883331298828125 2023-01-21 13:22:06.919502: step: 952/526, loss: 0.015796469524502754 2023-01-21 13:22:08.052785: step: 956/526, loss: 0.001987934112548828 2023-01-21 13:22:09.188054: step: 960/526, loss: 0.08326959609985352 2023-01-21 13:22:10.298523: step: 964/526, loss: 0.00015087128849700093 2023-01-21 13:22:11.420937: step: 968/526, loss: 0.00042748451232910156 2023-01-21 13:22:12.537456: step: 972/526, loss: 2.5081633793888614e-05 2023-01-21 13:22:13.699133: step: 976/526, loss: 0.0030844209250062704 2023-01-21 13:22:14.836700: step: 980/526, loss: 0.02462296560406685 2023-01-21 13:22:15.966633: step: 984/526, loss: 0.0024687768891453743 2023-01-21 13:22:17.072149: step: 988/526, loss: 0.018194103613495827 2023-01-21 13:22:18.208368: step: 992/526, loss: 0.05408220365643501 2023-01-21 13:22:19.347611: step: 996/526, loss: 0.04926195368170738 2023-01-21 13:22:20.495131: step: 1000/526, loss: 0.0001615524379303679 2023-01-21 13:22:21.620860: step: 1004/526, loss: 1.3732910701946821e-05 2023-01-21 13:22:22.734808: step: 1008/526, loss: 0.0024916650727391243 2023-01-21 13:22:23.842962: step: 1012/526, loss: 0.00020990372286178172 2023-01-21 13:22:24.959495: step: 1016/526, loss: 0.002630996983498335 2023-01-21 13:22:26.068215: step: 1020/526, loss: 0.0008620262378826737 2023-01-21 13:22:27.218411: step: 1024/526, loss: 0.014086627401411533 2023-01-21 13:22:28.327576: step: 1028/526, loss: 8.611679368186742e-05 2023-01-21 13:22:29.493136: step: 1032/526, loss: 0.961580753326416 2023-01-21 13:22:30.599141: step: 1036/526, loss: 2.86102294921875e-06 2023-01-21 13:22:31.732012: step: 1040/526, loss: 0.00024867060710676014 2023-01-21 13:22:32.866968: step: 1044/526, loss: 0.0006837844848632812 2023-01-21 13:22:33.992352: step: 1048/526, loss: 0.0003898620489053428 2023-01-21 13:22:35.124399: step: 1052/526, loss: 4.901886131847277e-05 2023-01-21 13:22:36.252585: step: 1056/526, loss: 0.0005243301275186241 2023-01-21 13:22:37.478439: step: 1060/526, loss: 0.00020923613919876516 2023-01-21 13:22:38.630706: step: 1064/526, loss: 7.209777686512098e-05 2023-01-21 13:22:39.759488: step: 1068/526, loss: 0.0007066726684570312 2023-01-21 13:22:40.914154: step: 1072/526, loss: 0.0028179169166833162 2023-01-21 13:22:42.071537: step: 1076/526, loss: 0.018347740173339844 2023-01-21 13:22:43.211508: step: 1080/526, loss: 0.000834560371004045 2023-01-21 13:22:44.339471: step: 1084/526, loss: 0.0001028060942189768 2023-01-21 13:22:45.474176: step: 1088/526, loss: 0.0017168044578284025 2023-01-21 13:22:46.588031: step: 1092/526, loss: 4.100799742445815e-06 2023-01-21 13:22:47.735690: step: 1096/526, loss: 0.014684582129120827 2023-01-21 13:22:48.894098: step: 1100/526, loss: 4.4822695599577855e-06 2023-01-21 13:22:50.036844: step: 1104/526, loss: 0.014325332827866077 2023-01-21 13:22:51.154379: step: 1108/526, loss: 0.6066886186599731 2023-01-21 13:22:52.303227: step: 1112/526, loss: -2.174377368646674e-05 2023-01-21 13:22:53.455403: step: 1116/526, loss: 0.012491131201386452 2023-01-21 13:22:54.609145: step: 1120/526, loss: 0.05121555179357529 2023-01-21 13:22:55.737309: step: 1124/526, loss: 1.926422191900201e-05 2023-01-21 13:22:56.863726: step: 1128/526, loss: 0.0006259918445721269 2023-01-21 13:22:58.005379: step: 1132/526, loss: 8.773804438533261e-06 2023-01-21 13:22:59.134263: step: 1136/526, loss: 0.0011281968327239156 2023-01-21 13:23:00.261944: step: 1140/526, loss: 1.0204315003647935e-05 2023-01-21 13:23:01.429109: step: 1144/526, loss: -1.888275073724799e-05 2023-01-21 13:23:02.573372: step: 1148/526, loss: -2.47955313170678e-06 2023-01-21 13:23:03.679448: step: 1152/526, loss: 0.00010070801363326609 2023-01-21 13:23:04.826893: step: 1156/526, loss: 0.006709480658173561 2023-01-21 13:23:05.946591: step: 1160/526, loss: 3.9863589336164296e-05 2023-01-21 13:23:07.061689: step: 1164/526, loss: 0.012752914801239967 2023-01-21 13:23:08.215091: step: 1168/526, loss: 0.0033310416620224714 2023-01-21 13:23:09.318571: step: 1172/526, loss: 7.629394644936838e-07 2023-01-21 13:23:10.452877: step: 1176/526, loss: 0.031670376658439636 2023-01-21 13:23:11.585154: step: 1180/526, loss: 0.04897937923669815 2023-01-21 13:23:12.714222: step: 1184/526, loss: 5.617141869151965e-05 2023-01-21 13:23:13.918688: step: 1188/526, loss: 0.08169842511415482 2023-01-21 13:23:15.083568: step: 1192/526, loss: 9.441375732421875e-05 2023-01-21 13:23:16.205146: step: 1196/526, loss: 0.000514984130859375 2023-01-21 13:23:17.339265: step: 1200/526, loss: 0.0007412910345010459 2023-01-21 13:23:18.452626: step: 1204/526, loss: 7.581710633530747e-06 2023-01-21 13:23:19.599261: step: 1208/526, loss: 2.5844574338407256e-05 2023-01-21 13:23:20.729288: step: 1212/526, loss: 5.722046353184851e-06 2023-01-21 13:23:21.880602: step: 1216/526, loss: -5.91278057981981e-06 2023-01-21 13:23:23.012391: step: 1220/526, loss: 0.00439796457067132 2023-01-21 13:23:24.130029: step: 1224/526, loss: 0.00179376604501158 2023-01-21 13:23:25.286180: step: 1228/526, loss: 0.0030015946831554174 2023-01-21 13:23:26.469833: step: 1232/526, loss: 0.0011264800559729338 2023-01-21 13:23:27.587469: step: 1236/526, loss: 0.47378939390182495 2023-01-21 13:23:28.706493: step: 1240/526, loss: 2.6702882678364404e-05 2023-01-21 13:23:29.849693: step: 1244/526, loss: 0.014657068066298962 2023-01-21 13:23:30.985143: step: 1248/526, loss: 6.370544724632055e-05 2023-01-21 13:23:32.124458: step: 1252/526, loss: 0.000556993531063199 2023-01-21 13:23:33.298150: step: 1256/526, loss: 0.003033923916518688 2023-01-21 13:23:34.450095: step: 1260/526, loss: 0.01989002153277397 2023-01-21 13:23:35.578281: step: 1264/526, loss: 0.003425073577091098 2023-01-21 13:23:36.709113: step: 1268/526, loss: 0.00032329559326171875 2023-01-21 13:23:37.821371: step: 1272/526, loss: 8.39233416627394e-06 2023-01-21 13:23:38.962472: step: 1276/526, loss: 5.435943421616685e-06 2023-01-21 13:23:40.081668: step: 1280/526, loss: 0.007731151767075062 2023-01-21 13:23:41.228993: step: 1284/526, loss: 0.01676969602704048 2023-01-21 13:23:42.338933: step: 1288/526, loss: 2.002715973503655e-06 2023-01-21 13:23:43.479026: step: 1292/526, loss: 0.0001844406215241179 2023-01-21 13:23:44.616986: step: 1296/526, loss: 1.4114380974206142e-05 2023-01-21 13:23:45.785943: step: 1300/526, loss: 0.00155982980504632 2023-01-21 13:23:46.934749: step: 1304/526, loss: 0.028210550546646118 2023-01-21 13:23:48.050475: step: 1308/526, loss: -1.8978118532686494e-05 2023-01-21 13:23:49.173290: step: 1312/526, loss: 0.0002628326474223286 2023-01-21 13:23:50.304258: step: 1316/526, loss: 0.013109779916703701 2023-01-21 13:23:51.436174: step: 1320/526, loss: 0.004711723420768976 2023-01-21 13:23:52.574134: step: 1324/526, loss: 8.687973604537547e-05 2023-01-21 13:23:53.700721: step: 1328/526, loss: 1.1444091796875e-05 2023-01-21 13:23:54.822014: step: 1332/526, loss: 0.026537800207734108 2023-01-21 13:23:55.954589: step: 1336/526, loss: 5.264282299322076e-05 2023-01-21 13:23:57.078781: step: 1340/526, loss: 9.117127046920359e-05 2023-01-21 13:23:58.214567: step: 1344/526, loss: 0.000827407871838659 2023-01-21 13:23:59.336951: step: 1348/526, loss: 0.0022382736206054688 2023-01-21 13:24:00.499547: step: 1352/526, loss: 0.03650626912713051 2023-01-21 13:24:01.641484: step: 1356/526, loss: 0.004920482635498047 2023-01-21 13:24:02.780692: step: 1360/526, loss: 0.003426170442253351 2023-01-21 13:24:03.914731: step: 1364/526, loss: 0.056017305701971054 2023-01-21 13:24:05.076749: step: 1368/526, loss: 0.02686176262795925 2023-01-21 13:24:06.202835: step: 1372/526, loss: 0.008903312496840954 2023-01-21 13:24:07.359565: step: 1376/526, loss: 0.0030012130737304688 2023-01-21 13:24:08.528013: step: 1380/526, loss: 0.0007151603931561112 2023-01-21 13:24:09.677974: step: 1384/526, loss: 0.0038175582885742188 2023-01-21 13:24:10.786492: step: 1388/526, loss: 4.95910626341356e-06 2023-01-21 13:24:11.912158: step: 1392/526, loss: 0.0019253731006756425 2023-01-21 13:24:13.068400: step: 1396/526, loss: 7.486343747586943e-06 2023-01-21 13:24:14.213561: step: 1400/526, loss: 4.95910626341356e-06 2023-01-21 13:24:15.332588: step: 1404/526, loss: 1.5830992197152227e-05 2023-01-21 13:24:16.455823: step: 1408/526, loss: 0.002237510634586215 2023-01-21 13:24:17.590948: step: 1412/526, loss: 5.53131121705519e-06 2023-01-21 13:24:18.700350: step: 1416/526, loss: 0.00014820098294876516 2023-01-21 13:24:19.820673: step: 1420/526, loss: 0.0006631851429119706 2023-01-21 13:24:20.962928: step: 1424/526, loss: 0.03126220777630806 2023-01-21 13:24:22.144051: step: 1428/526, loss: -9.059906005859375e-06 2023-01-21 13:24:23.272446: step: 1432/526, loss: 0.00897140521556139 2023-01-21 13:24:24.414473: step: 1436/526, loss: 0.00128602993208915 2023-01-21 13:24:25.567780: step: 1440/526, loss: 0.000240325927734375 2023-01-21 13:24:26.706712: step: 1444/526, loss: 3.8909915019758046e-05 2023-01-21 13:24:27.860928: step: 1448/526, loss: 0.000702714896760881 2023-01-21 13:24:28.988936: step: 1452/526, loss: 0.020237160846590996 2023-01-21 13:24:30.109880: step: 1456/526, loss: 0.0599786750972271 2023-01-21 13:24:31.262051: step: 1460/526, loss: 0.00022621154494117945 2023-01-21 13:24:32.384373: step: 1464/526, loss: 2.994537317135837e-05 2023-01-21 13:24:33.521634: step: 1468/526, loss: 0.0005204200861044228 2023-01-21 13:24:34.623704: step: 1472/526, loss: 1.0967254638671875e-05 2023-01-21 13:24:35.755313: step: 1476/526, loss: 3.852844383800402e-05 2023-01-21 13:24:36.908028: step: 1480/526, loss: 0.0002305030939169228 2023-01-21 13:24:38.051539: step: 1484/526, loss: 0.0034009935334324837 2023-01-21 13:24:39.214159: step: 1488/526, loss: 2.765655608527595e-06 2023-01-21 13:24:40.382088: step: 1492/526, loss: 6.642342486884445e-05 2023-01-21 13:24:41.536677: step: 1496/526, loss: 2.6893614631262608e-05 2023-01-21 13:24:42.694643: step: 1500/526, loss: 0.003851318499073386 2023-01-21 13:24:43.855909: step: 1504/526, loss: 0.005389881320297718 2023-01-21 13:24:44.983134: step: 1508/526, loss: 0.0027903555892407894 2023-01-21 13:24:46.117006: step: 1512/526, loss: 2.899169885495212e-05 2023-01-21 13:24:47.256363: step: 1516/526, loss: 0.02359333075582981 2023-01-21 13:24:48.412793: step: 1520/526, loss: 5.8460234868107364e-05 2023-01-21 13:24:49.559435: step: 1524/526, loss: 0.012447738088667393 2023-01-21 13:24:50.707276: step: 1528/526, loss: 0.00042195318383164704 2023-01-21 13:24:51.819152: step: 1532/526, loss: 0.0006528854137286544 2023-01-21 13:24:52.996876: step: 1536/526, loss: -4.76837158203125e-06 2023-01-21 13:24:54.145764: step: 1540/526, loss: 0.015368843451142311 2023-01-21 13:24:55.259525: step: 1544/526, loss: 0.004864597227424383 2023-01-21 13:24:56.393447: step: 1548/526, loss: 0.0036489488556981087 2023-01-21 13:24:57.536253: step: 1552/526, loss: 0.0011518478859215975 2023-01-21 13:24:58.678947: step: 1556/526, loss: 0.0016561509110033512 2023-01-21 13:24:59.822101: step: 1560/526, loss: 0.0016248702304437757 2023-01-21 13:25:00.940007: step: 1564/526, loss: 0.012558747082948685 2023-01-21 13:25:02.046437: step: 1568/526, loss: 0.00028476715669967234 2023-01-21 13:25:03.204904: step: 1572/526, loss: 0.00011701584298862144 2023-01-21 13:25:04.354419: step: 1576/526, loss: 2.5081635612878017e-05 2023-01-21 13:25:05.477438: step: 1580/526, loss: 1.1253358024987392e-05 2023-01-21 13:25:06.621434: step: 1584/526, loss: 0.0037362100556492805 2023-01-21 13:25:07.835294: step: 1588/526, loss: 0.007292896509170532 2023-01-21 13:25:08.988012: step: 1592/526, loss: 0.014206696301698685 2023-01-21 13:25:10.146168: step: 1596/526, loss: -3.0517576306010596e-06 2023-01-21 13:25:11.274246: step: 1600/526, loss: 9.622573998058215e-05 2023-01-21 13:25:12.377173: step: 1604/526, loss: 0.00013065338134765625 2023-01-21 13:25:13.493617: step: 1608/526, loss: 0.00247707380913198 2023-01-21 13:25:14.653708: step: 1612/526, loss: 0.002977752825245261 2023-01-21 13:25:15.821604: step: 1616/526, loss: 5.455017162603326e-05 2023-01-21 13:25:16.954672: step: 1620/526, loss: 0.0002422332763671875 2023-01-21 13:25:18.126099: step: 1624/526, loss: 0.020637035369873047 2023-01-21 13:25:19.298829: step: 1628/526, loss: 0.0024837495293468237 2023-01-21 13:25:20.456635: step: 1632/526, loss: 2.002715973503655e-06 2023-01-21 13:25:21.600978: step: 1636/526, loss: 0.000663328159134835 2023-01-21 13:25:22.717168: step: 1640/526, loss: 3.604888843256049e-05 2023-01-21 13:25:23.864432: step: 1644/526, loss: 0.0050809383392333984 2023-01-21 13:25:25.013347: step: 1648/526, loss: 0.03238420560956001 2023-01-21 13:25:26.164099: step: 1652/526, loss: 0.01873464696109295 2023-01-21 13:25:27.288638: step: 1656/526, loss: 0.007911873050034046 2023-01-21 13:25:28.426994: step: 1660/526, loss: 0.0007429122924804688 2023-01-21 13:25:29.546488: step: 1664/526, loss: 0.030498409643769264 2023-01-21 13:25:30.659112: step: 1668/526, loss: 0.0016943931113928556 2023-01-21 13:25:31.812336: step: 1672/526, loss: 0.005672645755112171 2023-01-21 13:25:32.996276: step: 1676/526, loss: 0.007938671857118607 2023-01-21 13:25:34.137276: step: 1680/526, loss: 0.0017467498546466231 2023-01-21 13:25:35.250710: step: 1684/526, loss: 0.032073974609375 2023-01-21 13:25:36.379822: step: 1688/526, loss: 0.2091618925333023 2023-01-21 13:25:37.528327: step: 1692/526, loss: 0.022150231525301933 2023-01-21 13:25:38.666325: step: 1696/526, loss: 0.0018977165454998612 2023-01-21 13:25:39.809834: step: 1700/526, loss: 0.02743702009320259 2023-01-21 13:25:40.965081: step: 1704/526, loss: 0.036840301007032394 2023-01-21 13:25:42.078570: step: 1708/526, loss: 0.0004852294805459678 2023-01-21 13:25:43.205041: step: 1712/526, loss: 4.4536591303767636e-05 2023-01-21 13:25:44.385921: step: 1716/526, loss: 0.00040631293086335063 2023-01-21 13:25:45.541408: step: 1720/526, loss: 0.0007458687177859247 2023-01-21 13:25:46.690544: step: 1724/526, loss: 0.001009273575618863 2023-01-21 13:25:47.830243: step: 1728/526, loss: 0.00031156541081145406 2023-01-21 13:25:48.948157: step: 1732/526, loss: 1.2395102977752686 2023-01-21 13:25:50.081328: step: 1736/526, loss: 0.0013378143776208162 2023-01-21 13:25:51.210537: step: 1740/526, loss: 0.03308725729584694 2023-01-21 13:25:52.343962: step: 1744/526, loss: 0.10715971142053604 2023-01-21 13:25:53.500856: step: 1748/526, loss: 0.04743032529950142 2023-01-21 13:25:54.636416: step: 1752/526, loss: 5.626678557746345e-06 2023-01-21 13:25:55.795841: step: 1756/526, loss: 0.00030431748018600047 2023-01-21 13:25:56.943912: step: 1760/526, loss: 0.004824638366699219 2023-01-21 13:25:58.090582: step: 1764/526, loss: 0.023418808355927467 2023-01-21 13:25:59.271085: step: 1768/526, loss: 0.00148525252006948 2023-01-21 13:26:00.437602: step: 1772/526, loss: 0.03056774102151394 2023-01-21 13:26:01.619899: step: 1776/526, loss: 0.000388669956009835 2023-01-21 13:26:02.760894: step: 1780/526, loss: 0.0007776260026730597 2023-01-21 13:26:03.906760: step: 1784/526, loss: 0.017757128924131393 2023-01-21 13:26:05.062276: step: 1788/526, loss: 0.0242964755743742 2023-01-21 13:26:06.183712: step: 1792/526, loss: 0.016392040997743607 2023-01-21 13:26:07.304949: step: 1796/526, loss: 0.0017921447288244963 2023-01-21 13:26:08.413472: step: 1800/526, loss: 0.0018837453098967671 2023-01-21 13:26:09.531337: step: 1804/526, loss: 1.9168852304574102e-05 2023-01-21 13:26:10.653514: step: 1808/526, loss: 1.258850079466356e-05 2023-01-21 13:26:11.764406: step: 1812/526, loss: 8.592606172896922e-05 2023-01-21 13:26:12.910045: step: 1816/526, loss: 0.002609729766845703 2023-01-21 13:26:14.007448: step: 1820/526, loss: 0.0008083343273028731 2023-01-21 13:26:15.153069: step: 1824/526, loss: 0.00023536680964753032 2023-01-21 13:26:16.325298: step: 1828/526, loss: 0.0002002716064453125 2023-01-21 13:26:17.475030: step: 1832/526, loss: 0.005363273900002241 2023-01-21 13:26:18.611788: step: 1836/526, loss: 0.008018827065825462 2023-01-21 13:26:19.771049: step: 1840/526, loss: 0.007731246761977673 2023-01-21 13:26:20.912575: step: 1844/526, loss: 0.03594532236456871 2023-01-21 13:26:22.058902: step: 1848/526, loss: 0.3161148130893707 2023-01-21 13:26:23.174042: step: 1852/526, loss: 0.004541588015854359 2023-01-21 13:26:24.327046: step: 1856/526, loss: 0.00018424988957121968 2023-01-21 13:26:25.434434: step: 1860/526, loss: 0.0008784294477663934 2023-01-21 13:26:26.573204: step: 1864/526, loss: 0.00011901855759788305 2023-01-21 13:26:27.677529: step: 1868/526, loss: 0.0008966445457190275 2023-01-21 13:26:28.857079: step: 1872/526, loss: 0.0034414289984852076 2023-01-21 13:26:29.996073: step: 1876/526, loss: 0.00013656617375090718 2023-01-21 13:26:31.098759: step: 1880/526, loss: 6.999969627941027e-05 2023-01-21 13:26:32.232310: step: 1884/526, loss: 1.0580861568450928 2023-01-21 13:26:33.361351: step: 1888/526, loss: 0.008166313171386719 2023-01-21 13:26:34.490511: step: 1892/526, loss: 0.1165875494480133 2023-01-21 13:26:35.646934: step: 1896/526, loss: 4.920959327137098e-05 2023-01-21 13:26:36.810516: step: 1900/526, loss: 0.0001811981201171875 2023-01-21 13:26:37.966761: step: 1904/526, loss: 0.0002621650928631425 2023-01-21 13:26:39.119044: step: 1908/526, loss: -4.95910626341356e-06 2023-01-21 13:26:40.260776: step: 1912/526, loss: 0.004971933551132679 2023-01-21 13:26:41.420316: step: 1916/526, loss: 0.008744620718061924 2023-01-21 13:26:42.580167: step: 1920/526, loss: 7.343292236328125e-05 2023-01-21 13:26:43.723994: step: 1924/526, loss: 0.015537834726274014 2023-01-21 13:26:44.871081: step: 1928/526, loss: 0.00037384033203125 2023-01-21 13:26:45.987354: step: 1932/526, loss: 0.0004722595331259072 2023-01-21 13:26:47.119813: step: 1936/526, loss: 7.82012921263231e-06 2023-01-21 13:26:48.253571: step: 1940/526, loss: 0.0003864288446493447 2023-01-21 13:26:49.437252: step: 1944/526, loss: 0.0012104033958166838 2023-01-21 13:26:50.565261: step: 1948/526, loss: 0.010028649121522903 2023-01-21 13:26:51.711972: step: 1952/526, loss: 0.0003222465456929058 2023-01-21 13:26:52.833913: step: 1956/526, loss: 0.016926003620028496 2023-01-21 13:26:53.978412: step: 1960/526, loss: 0.008316612802445889 2023-01-21 13:26:55.103040: step: 1964/526, loss: 0.0005629539955407381 2023-01-21 13:26:56.260700: step: 1968/526, loss: 0.01044845674186945 2023-01-21 13:26:57.389785: step: 1972/526, loss: 0.0008127212640829384 2023-01-21 13:26:58.528511: step: 1976/526, loss: 0.0001337051362497732 2023-01-21 13:26:59.643853: step: 1980/526, loss: 1.4781951904296875e-05 2023-01-21 13:27:00.774625: step: 1984/526, loss: 0.00020828247943427414 2023-01-21 13:27:01.930106: step: 1988/526, loss: 0.0007927894475869834 2023-01-21 13:27:03.066271: step: 1992/526, loss: 0.0060024261474609375 2023-01-21 13:27:04.202678: step: 1996/526, loss: 0.0017284393543377519 2023-01-21 13:27:05.382159: step: 2000/526, loss: 0.0007156849023886025 2023-01-21 13:27:06.510124: step: 2004/526, loss: 0.0001312255917582661 2023-01-21 13:27:07.612976: step: 2008/526, loss: 1.2254714420123491e-05 2023-01-21 13:27:08.742480: step: 2012/526, loss: 7.934570749057457e-05 2023-01-21 13:27:09.906645: step: 2016/526, loss: 7.390976679744199e-05 2023-01-21 13:27:11.057583: step: 2020/526, loss: 0.013234615325927734 2023-01-21 13:27:12.181276: step: 2024/526, loss: 0.0004430771223269403 2023-01-21 13:27:13.365739: step: 2028/526, loss: 0.03345470875501633 2023-01-21 13:27:14.501822: step: 2032/526, loss: 0.02077198028564453 2023-01-21 13:27:15.662733: step: 2036/526, loss: 0.0002658843877725303 2023-01-21 13:27:16.803687: step: 2040/526, loss: 2.1362302504712716e-05 2023-01-21 13:27:17.927459: step: 2044/526, loss: 0.0018316269852221012 2023-01-21 13:27:19.081794: step: 2048/526, loss: 0.04275708273053169 2023-01-21 13:27:20.201993: step: 2052/526, loss: 2.0027162008773303e-06 2023-01-21 13:27:21.361086: step: 2056/526, loss: 0.12838879227638245 2023-01-21 13:27:22.452319: step: 2060/526, loss: 3.5762786865234375e-06 2023-01-21 13:27:23.591569: step: 2064/526, loss: -2.6702882678364404e-06 2023-01-21 13:27:24.738257: step: 2068/526, loss: 5.245209104032256e-05 2023-01-21 13:27:25.867752: step: 2072/526, loss: 0.0006593704456463456 2023-01-21 13:27:26.986202: step: 2076/526, loss: 0.0001570224849274382 2023-01-21 13:27:28.091722: step: 2080/526, loss: 0.0014127731556072831 2023-01-21 13:27:29.246005: step: 2084/526, loss: 0.00016708373732399195 2023-01-21 13:27:30.380797: step: 2088/526, loss: 4.158019874012098e-05 2023-01-21 13:27:31.508150: step: 2092/526, loss: 0.0018079758156090975 2023-01-21 13:27:32.639781: step: 2096/526, loss: 0.00243377685546875 2023-01-21 13:27:33.749120: step: 2100/526, loss: 1.9073486328125e-06 2023-01-21 13:27:34.872437: step: 2104/526, loss: 0.00010166168067371473 ================================================== Loss: 0.018 -------------------- Dev: {'event': {'p': 0.5684729064039409, 'r': 0.7683089214380826, 'f1': 0.653454133635334}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Test: {'event': {'p': 0.6110655737704918, 'r': 0.7781837160751566, 'f1': 0.6845730027548209}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Chinese: {'event': {'p': 0.5168539325842697, 'r': 0.8518518518518519, 'f1': 0.6433566433566433}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Korean: {'event': {'p': 0.62, 'r': 0.49206349206349204, 'f1': 0.5486725663716814}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} Russian: {'event': {'p': 0.4, 'r': 0.5, 'f1': 0.4444444444444445}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 26} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 27 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:28:18.226219: step: 4/526, loss: 0.0002765655517578125 2023-01-21 13:28:19.352100: step: 8/526, loss: 5.054473876953125e-05 2023-01-21 13:28:20.482430: step: 12/526, loss: 1.893043372547254e-05 2023-01-21 13:28:21.602080: step: 16/526, loss: 0.012279605492949486 2023-01-21 13:28:22.746758: step: 20/526, loss: 3.132819983875379e-05 2023-01-21 13:28:23.880463: step: 24/526, loss: 0.0015066623454913497 2023-01-21 13:28:25.002533: step: 28/526, loss: 6.961823237361386e-05 2023-01-21 13:28:26.119940: step: 32/526, loss: 0.0018840789562091231 2023-01-21 13:28:27.248231: step: 36/526, loss: 0.006211709696799517 2023-01-21 13:28:28.392684: step: 40/526, loss: 0.011351299472153187 2023-01-21 13:28:29.494455: step: 44/526, loss: 0.00311622628942132 2023-01-21 13:28:30.638511: step: 48/526, loss: 0.0006634712335653603 2023-01-21 13:28:31.782247: step: 52/526, loss: 4.043579247081652e-05 2023-01-21 13:28:32.926085: step: 56/526, loss: 0.005677509121596813 2023-01-21 13:28:34.057044: step: 60/526, loss: 0.0002114296075887978 2023-01-21 13:28:35.204934: step: 64/526, loss: 0.000727748847566545 2023-01-21 13:28:36.339540: step: 68/526, loss: 0.005809879396110773 2023-01-21 13:28:37.477616: step: 72/526, loss: 0.028379440307617188 2023-01-21 13:28:38.608379: step: 76/526, loss: 0.0004896164173260331 2023-01-21 13:28:39.735021: step: 80/526, loss: 0.0007036208990029991 2023-01-21 13:28:40.876336: step: 84/526, loss: -2.059936559817288e-05 2023-01-21 13:28:41.991621: step: 88/526, loss: 0.0008453846094198525 2023-01-21 13:28:43.125906: step: 92/526, loss: 1.659393274167087e-05 2023-01-21 13:28:44.225600: step: 96/526, loss: 0.0001484871027059853 2023-01-21 13:28:45.340771: step: 100/526, loss: 8.726119631319307e-06 2023-01-21 13:28:46.464193: step: 104/526, loss: 0.00019378661818336695 2023-01-21 13:28:47.599638: step: 108/526, loss: 9.098053124034777e-05 2023-01-21 13:28:48.759581: step: 112/526, loss: 0.00171070103533566 2023-01-21 13:28:49.905926: step: 116/526, loss: 0.0003231048467569053 2023-01-21 13:28:51.044443: step: 120/526, loss: 0.000392723100958392 2023-01-21 13:28:52.190568: step: 124/526, loss: 0.022927286103367805 2023-01-21 13:28:53.307408: step: 128/526, loss: 0.007363128941506147 2023-01-21 13:28:54.456625: step: 132/526, loss: 0.027758406475186348 2023-01-21 13:28:55.571817: step: 136/526, loss: 1.239776611328125e-05 2023-01-21 13:28:56.736837: step: 140/526, loss: -7.915497008070815e-06 2023-01-21 13:28:57.888331: step: 144/526, loss: 0.05431995540857315 2023-01-21 13:28:58.988174: step: 148/526, loss: 0.00033893584623001516 2023-01-21 13:29:00.104587: step: 152/526, loss: 2.765655608527595e-06 2023-01-21 13:29:01.233112: step: 156/526, loss: 0.00021991731773596257 2023-01-21 13:29:02.377436: step: 160/526, loss: 8.39233416627394e-06 2023-01-21 13:29:03.512797: step: 164/526, loss: 4.806518700206652e-05 2023-01-21 13:29:04.644877: step: 168/526, loss: 0.0008997917757369578 2023-01-21 13:29:05.770001: step: 172/526, loss: 7.534027190558845e-06 2023-01-21 13:29:06.898590: step: 176/526, loss: 5.168914867681451e-05 2023-01-21 13:29:08.021215: step: 180/526, loss: 0.0027696609031409025 2023-01-21 13:29:09.132085: step: 184/526, loss: 1.6880036127986386e-05 2023-01-21 13:29:10.276831: step: 188/526, loss: 0.0010470390552654862 2023-01-21 13:29:11.415493: step: 192/526, loss: 0.06860151886940002 2023-01-21 13:29:12.535388: step: 196/526, loss: 3.614425440900959e-05 2023-01-21 13:29:13.693560: step: 200/526, loss: 0.00404777517542243 2023-01-21 13:29:14.865942: step: 204/526, loss: 0.0008954048389568925 2023-01-21 13:29:16.042758: step: 208/526, loss: 3.051757857974735e-06 2023-01-21 13:29:17.184403: step: 212/526, loss: 0.0039577484130859375 2023-01-21 13:29:18.310120: step: 216/526, loss: 3.0040740966796875e-05 2023-01-21 13:29:19.446088: step: 220/526, loss: 0.001960658933967352 2023-01-21 13:29:20.572028: step: 224/526, loss: 0.00022811890812590718 2023-01-21 13:29:21.704227: step: 228/526, loss: 0.010171127505600452 2023-01-21 13:29:22.850295: step: 232/526, loss: 0.0003105163632426411 2023-01-21 13:29:23.986869: step: 236/526, loss: 0.002224540803581476 2023-01-21 13:29:25.131428: step: 240/526, loss: 0.0057853697799146175 2023-01-21 13:29:26.300381: step: 244/526, loss: 0.0003604888916015625 2023-01-21 13:29:27.428676: step: 248/526, loss: 0.0408940352499485 2023-01-21 13:29:28.595902: step: 252/526, loss: 7.057189577608369e-06 2023-01-21 13:29:29.740192: step: 256/526, loss: 9.760856482898816e-05 2023-01-21 13:29:30.880570: step: 260/526, loss: 0.0003845214960165322 2023-01-21 13:29:32.002891: step: 264/526, loss: 0.0005950927734375 2023-01-21 13:29:33.188184: step: 268/526, loss: 0.0004058837948832661 2023-01-21 13:29:34.344643: step: 272/526, loss: 0.006672763731330633 2023-01-21 13:29:35.498776: step: 276/526, loss: 0.0016333579551428556 2023-01-21 13:29:36.652479: step: 280/526, loss: 0.00050182337872684 2023-01-21 13:29:37.783104: step: 284/526, loss: 0.00019378661818336695 2023-01-21 13:29:38.950014: step: 288/526, loss: 0.0002842903195414692 2023-01-21 13:29:40.056007: step: 292/526, loss: 0.0001317977876169607 2023-01-21 13:29:41.171951: step: 296/526, loss: 2.646446409926284e-05 2023-01-21 13:29:42.354877: step: 300/526, loss: 0.029449844732880592 2023-01-21 13:29:43.504599: step: 304/526, loss: 0.009784317575395107 2023-01-21 13:29:44.627700: step: 308/526, loss: -9.72747784544481e-06 2023-01-21 13:29:45.748547: step: 312/526, loss: 0.0013284683227539062 2023-01-21 13:29:46.885592: step: 316/526, loss: 7.190704491222277e-05 2023-01-21 13:29:48.008150: step: 320/526, loss: 1.8787384760798886e-05 2023-01-21 13:29:49.149410: step: 324/526, loss: 9.088516526389867e-05 2023-01-21 13:29:50.262298: step: 328/526, loss: 0.0009706497075967491 2023-01-21 13:29:51.407243: step: 332/526, loss: 7.2479248046875e-05 2023-01-21 13:29:52.542570: step: 336/526, loss: -1.640319896978326e-05 2023-01-21 13:29:53.708762: step: 340/526, loss: 0.0007809638627804816 2023-01-21 13:29:54.839471: step: 344/526, loss: 0.00013914107694290578 2023-01-21 13:29:55.958181: step: 348/526, loss: -2.4795533590804553e-06 2023-01-21 13:29:57.076106: step: 352/526, loss: 3.4904482163256034e-05 2023-01-21 13:29:58.210723: step: 356/526, loss: -1.4019012269272935e-05 2023-01-21 13:29:59.366800: step: 360/526, loss: 0.00012416840763762593 2023-01-21 13:30:00.505999: step: 364/526, loss: 0.004153633024543524 2023-01-21 13:30:01.628955: step: 368/526, loss: 3.2711028325138614e-05 2023-01-21 13:30:02.782268: step: 372/526, loss: 0.00031347275944426656 2023-01-21 13:30:03.912539: step: 376/526, loss: 0.004353714175522327 2023-01-21 13:30:05.025003: step: 380/526, loss: 6.0987476899754256e-05 2023-01-21 13:30:06.185007: step: 384/526, loss: -3.433226083870977e-06 2023-01-21 13:30:07.321580: step: 388/526, loss: 0.00129871373064816 2023-01-21 13:30:08.454231: step: 392/526, loss: 0.002247190335765481 2023-01-21 13:30:09.573844: step: 396/526, loss: 0.011980055831372738 2023-01-21 13:30:10.734006: step: 400/526, loss: 9.269714792026207e-05 2023-01-21 13:30:11.876723: step: 404/526, loss: 2.1076202756376006e-05 2023-01-21 13:30:13.025239: step: 408/526, loss: 0.0003170013369526714 2023-01-21 13:30:14.182383: step: 412/526, loss: 0.0005190849187783897 2023-01-21 13:30:15.336589: step: 416/526, loss: 0.012317943386733532 2023-01-21 13:30:16.493976: step: 420/526, loss: 0.06505031138658524 2023-01-21 13:30:17.636147: step: 424/526, loss: 0.0004749298095703125 2023-01-21 13:30:18.772695: step: 428/526, loss: 0.0015095234848558903 2023-01-21 13:30:19.927297: step: 432/526, loss: 0.000209808349609375 2023-01-21 13:30:21.048098: step: 436/526, loss: 0.00159111013635993 2023-01-21 13:30:22.174202: step: 440/526, loss: 0.03256235271692276 2023-01-21 13:30:23.278464: step: 444/526, loss: 1.239776611328125e-05 2023-01-21 13:30:24.412476: step: 448/526, loss: 0.0009539604070596397 2023-01-21 13:30:25.546337: step: 452/526, loss: 6.389617919921875e-05 2023-01-21 13:30:26.667178: step: 456/526, loss: 8.773804438533261e-06 2023-01-21 13:30:27.796131: step: 460/526, loss: 0.003558540251106024 2023-01-21 13:30:28.958008: step: 464/526, loss: 0.0015316009521484375 2023-01-21 13:30:30.099583: step: 468/526, loss: 0.008338737301528454 2023-01-21 13:30:31.263666: step: 472/526, loss: 1.926422191900201e-05 2023-01-21 13:30:32.401288: step: 476/526, loss: 0.0008203506586141884 2023-01-21 13:30:33.517561: step: 480/526, loss: 0.0014940261607989669 2023-01-21 13:30:34.648573: step: 484/526, loss: 0.0001235961972270161 2023-01-21 13:30:35.798138: step: 488/526, loss: 0.0004393577401060611 2023-01-21 13:30:36.989165: step: 492/526, loss: 2.017021324718371e-05 2023-01-21 13:30:38.115711: step: 496/526, loss: 2.384185791015625e-05 2023-01-21 13:30:39.250562: step: 500/526, loss: 0.00401725759729743 2023-01-21 13:30:40.383849: step: 504/526, loss: 0.006469154264777899 2023-01-21 13:30:41.525556: step: 508/526, loss: 2.2220612663659267e-05 2023-01-21 13:30:42.667755: step: 512/526, loss: 2.593994213384576e-05 2023-01-21 13:30:43.789824: step: 516/526, loss: -8.01086389401462e-06 2023-01-21 13:30:44.924592: step: 520/526, loss: 0.020737363025546074 2023-01-21 13:30:46.052567: step: 524/526, loss: 1.888275073724799e-05 2023-01-21 13:30:47.234842: step: 528/526, loss: 0.0001293182431254536 2023-01-21 13:30:48.375677: step: 532/526, loss: 4.301071385270916e-05 2023-01-21 13:30:49.477333: step: 536/526, loss: 0.00024671555729582906 2023-01-21 13:30:50.602668: step: 540/526, loss: 1.3446808225126006e-05 2023-01-21 13:30:51.739309: step: 544/526, loss: 2.574920654296875e-05 2023-01-21 13:30:52.905841: step: 548/526, loss: 0.00031948089599609375 2023-01-21 13:30:54.007234: step: 552/526, loss: 3.337860107421875e-05 2023-01-21 13:30:55.161402: step: 556/526, loss: 0.009712982922792435 2023-01-21 13:30:56.272717: step: 560/526, loss: 8.735657320357859e-05 2023-01-21 13:30:57.433029: step: 564/526, loss: 0.0017586707836017013 2023-01-21 13:30:58.529017: step: 568/526, loss: 3.185272362316027e-05 2023-01-21 13:30:59.686115: step: 572/526, loss: 9.422302537132055e-05 2023-01-21 13:31:00.805858: step: 576/526, loss: 0.012394714169204235 2023-01-21 13:31:01.943821: step: 580/526, loss: 0.03634929656982422 2023-01-21 13:31:03.077514: step: 584/526, loss: 3.910065061063506e-05 2023-01-21 13:31:04.215953: step: 588/526, loss: 0.08449192345142365 2023-01-21 13:31:05.346539: step: 592/526, loss: 0.0006435394752770662 2023-01-21 13:31:06.463339: step: 596/526, loss: 0.014098548330366611 2023-01-21 13:31:07.604001: step: 600/526, loss: 0.0021797181107103825 2023-01-21 13:31:08.739289: step: 604/526, loss: 0.022112369537353516 2023-01-21 13:31:09.900056: step: 608/526, loss: 0.0021508217323571444 2023-01-21 13:31:11.030175: step: 612/526, loss: 9.975433204090223e-05 2023-01-21 13:31:12.166673: step: 616/526, loss: 0.016420556232333183 2023-01-21 13:31:13.332576: step: 620/526, loss: 0.0005218505975790322 2023-01-21 13:31:14.458347: step: 624/526, loss: 0.0005505562294274569 2023-01-21 13:31:15.621191: step: 628/526, loss: 0.005284309387207031 2023-01-21 13:31:16.763627: step: 632/526, loss: 0.0033848288003355265 2023-01-21 13:31:17.919065: step: 636/526, loss: 6.408691842807457e-05 2023-01-21 13:31:19.053204: step: 640/526, loss: 0.002997970674186945 2023-01-21 13:31:20.198233: step: 644/526, loss: 0.05346536636352539 2023-01-21 13:31:21.347079: step: 648/526, loss: 7.877349707996473e-05 2023-01-21 13:31:22.480230: step: 652/526, loss: 3.814697629422881e-05 2023-01-21 13:31:23.599007: step: 656/526, loss: 5.92231735936366e-05 2023-01-21 13:31:24.749934: step: 660/526, loss: 0.0013502121437340975 2023-01-21 13:31:25.873817: step: 664/526, loss: 0.00023527146549895406 2023-01-21 13:31:27.016705: step: 668/526, loss: 2.250671423098538e-05 2023-01-21 13:31:28.150773: step: 672/526, loss: 0.0001464843808207661 2023-01-21 13:31:29.287467: step: 676/526, loss: 0.07876253128051758 2023-01-21 13:31:30.411069: step: 680/526, loss: 0.0002496719534974545 2023-01-21 13:31:31.538876: step: 684/526, loss: 0.034255217760801315 2023-01-21 13:31:32.661247: step: 688/526, loss: 8.106231689453125e-06 2023-01-21 13:31:33.836498: step: 692/526, loss: 0.0017586946487426758 2023-01-21 13:31:34.966668: step: 696/526, loss: 0.006284141913056374 2023-01-21 13:31:36.131711: step: 700/526, loss: 0.00022354126849677414 2023-01-21 13:31:37.300972: step: 704/526, loss: 3.738403756869957e-05 2023-01-21 13:31:38.414357: step: 708/526, loss: 1.4591218132409267e-05 2023-01-21 13:31:39.541199: step: 712/526, loss: 1.6689300537109375e-05 2023-01-21 13:31:40.645388: step: 716/526, loss: 5.1307681133039296e-05 2023-01-21 13:31:41.838761: step: 720/526, loss: 0.009741402231156826 2023-01-21 13:31:42.972997: step: 724/526, loss: 0.005017471499741077 2023-01-21 13:31:44.124023: step: 728/526, loss: 0.0003294944763183594 2023-01-21 13:31:45.242865: step: 732/526, loss: 0.0008516788366250694 2023-01-21 13:31:46.361269: step: 736/526, loss: 0.0080420495942235 2023-01-21 13:31:47.510670: step: 740/526, loss: 8.125305612338707e-05 2023-01-21 13:31:48.659668: step: 744/526, loss: -3.4332275390625e-05 2023-01-21 13:31:49.812329: step: 748/526, loss: 0.0007475852617062628 2023-01-21 13:31:50.968142: step: 752/526, loss: 0.006243610754609108 2023-01-21 13:31:52.112742: step: 756/526, loss: 1.8596649169921875e-05 2023-01-21 13:31:53.234364: step: 760/526, loss: 6.160736666060984e-05 2023-01-21 13:31:54.388316: step: 764/526, loss: 0.0005108893383294344 2023-01-21 13:31:55.499193: step: 768/526, loss: 0.024231290444731712 2023-01-21 13:31:56.656638: step: 772/526, loss: 1.7166141788038658e-06 2023-01-21 13:31:57.790240: step: 776/526, loss: 3.328323509776965e-05 2023-01-21 13:31:58.936386: step: 780/526, loss: 0.009100914001464844 2023-01-21 13:32:00.100068: step: 784/526, loss: 0.00023050307936500758 2023-01-21 13:32:01.238379: step: 788/526, loss: 0.00033574103144928813 2023-01-21 13:32:02.377759: step: 792/526, loss: 1.9645693100756034e-05 2023-01-21 13:32:03.513842: step: 796/526, loss: 0.0018773078918457031 2023-01-21 13:32:04.621513: step: 800/526, loss: 0.00017633437528274953 2023-01-21 13:32:05.768803: step: 804/526, loss: 0.06858577579259872 2023-01-21 13:32:06.901740: step: 808/526, loss: 0.008789253421127796 2023-01-21 13:32:08.045796: step: 812/526, loss: 0.0004965782281942666 2023-01-21 13:32:09.177072: step: 816/526, loss: 0.0041870116256177425 2023-01-21 13:32:10.323528: step: 820/526, loss: 0.004493141081184149 2023-01-21 13:32:11.440413: step: 824/526, loss: 0.6073503494262695 2023-01-21 13:32:12.566615: step: 828/526, loss: 0.00023174285888671875 2023-01-21 13:32:13.719559: step: 832/526, loss: 0.032556917518377304 2023-01-21 13:32:14.847084: step: 836/526, loss: 0.002376651857048273 2023-01-21 13:32:15.968724: step: 840/526, loss: 0.007678401190787554 2023-01-21 13:32:17.124306: step: 844/526, loss: 0.004945182707160711 2023-01-21 13:32:18.243203: step: 848/526, loss: 0.00032529831514693797 2023-01-21 13:32:19.365572: step: 852/526, loss: 0.006610870361328125 2023-01-21 13:32:20.505737: step: 856/526, loss: 0.012574577704071999 2023-01-21 13:32:21.659778: step: 860/526, loss: 0.016125869005918503 2023-01-21 13:32:22.815448: step: 864/526, loss: 0.0003427505725994706 2023-01-21 13:32:23.968790: step: 868/526, loss: 0.006281661801040173 2023-01-21 13:32:25.112862: step: 872/526, loss: -1.9073486328125e-06 2023-01-21 13:32:26.286928: step: 876/526, loss: 7.43865984986769e-06 2023-01-21 13:32:27.393661: step: 880/526, loss: 0.0005115509266033769 2023-01-21 13:32:28.515331: step: 884/526, loss: 7.262230064952746e-05 2023-01-21 13:32:29.669585: step: 888/526, loss: 0.020418357104063034 2023-01-21 13:32:30.795719: step: 892/526, loss: 3.9768219721736386e-05 2023-01-21 13:32:31.959964: step: 896/526, loss: 0.016159821301698685 2023-01-21 13:32:33.092055: step: 900/526, loss: -8.106231689453125e-06 2023-01-21 13:32:34.220239: step: 904/526, loss: 0.007848739624023438 2023-01-21 13:32:35.383418: step: 908/526, loss: 0.02537841908633709 2023-01-21 13:32:36.512038: step: 912/526, loss: 4.4345855712890625e-05 2023-01-21 13:32:37.662035: step: 916/526, loss: 0.0004116058407817036 2023-01-21 13:32:38.793942: step: 920/526, loss: 0.1156230941414833 2023-01-21 13:32:39.927674: step: 924/526, loss: 0.0006291389581747353 2023-01-21 13:32:41.080119: step: 928/526, loss: 0.0011343955993652344 2023-01-21 13:32:42.225316: step: 932/526, loss: 1.773834264895413e-05 2023-01-21 13:32:43.383538: step: 936/526, loss: 0.001533126924186945 2023-01-21 13:32:44.528314: step: 940/526, loss: 0.00025386811466887593 2023-01-21 13:32:45.673566: step: 944/526, loss: 1.354217511106981e-05 2023-01-21 13:32:46.830840: step: 948/526, loss: 0.5998144149780273 2023-01-21 13:32:47.960355: step: 952/526, loss: 5.981922004139051e-05 2023-01-21 13:32:49.092309: step: 956/526, loss: 3.7002566386945546e-05 2023-01-21 13:32:50.240504: step: 960/526, loss: 0.00599795812740922 2023-01-21 13:32:51.398090: step: 964/526, loss: 0.0029155733063817024 2023-01-21 13:32:52.555730: step: 968/526, loss: 0.001951539539732039 2023-01-21 13:32:53.709215: step: 972/526, loss: 0.010029601864516735 2023-01-21 13:32:54.856519: step: 976/526, loss: 0.0002944946172647178 2023-01-21 13:32:55.990716: step: 980/526, loss: 7.3432925091765355e-06 2023-01-21 13:32:57.139243: step: 984/526, loss: 0.0006487369537353516 2023-01-21 13:32:58.283573: step: 988/526, loss: 0.04792454466223717 2023-01-21 13:32:59.419397: step: 992/526, loss: 8.049011375987902e-05 2023-01-21 13:33:00.572609: step: 996/526, loss: 0.004157734103500843 2023-01-21 13:33:01.734200: step: 1000/526, loss: 0.001708221505396068 2023-01-21 13:33:02.854583: step: 1004/526, loss: 0.0020089149475097656 2023-01-21 13:33:03.972104: step: 1008/526, loss: 4.911422820441658e-06 2023-01-21 13:33:05.116273: step: 1012/526, loss: 0.0021471979562193155 2023-01-21 13:33:06.253945: step: 1016/526, loss: 0.00010185241990257055 2023-01-21 13:33:07.376686: step: 1020/526, loss: 3.051757857974735e-06 2023-01-21 13:33:08.510871: step: 1024/526, loss: 3.0517578125e-05 2023-01-21 13:33:09.649427: step: 1028/526, loss: 0.0010009766556322575 2023-01-21 13:33:10.795055: step: 1032/526, loss: 0.00011339188495185226 2023-01-21 13:33:11.931313: step: 1036/526, loss: 0.2311851978302002 2023-01-21 13:33:13.103175: step: 1040/526, loss: 0.001352500868961215 2023-01-21 13:33:14.226823: step: 1044/526, loss: 0.0010618210071697831 2023-01-21 13:33:15.372583: step: 1048/526, loss: 3.852844747598283e-05 2023-01-21 13:33:16.517271: step: 1052/526, loss: 0.00031651853350922465 2023-01-21 13:33:17.670345: step: 1056/526, loss: 0.0011723518837243319 2023-01-21 13:33:18.831459: step: 1060/526, loss: 5.5313107623078395e-06 2023-01-21 13:33:19.966034: step: 1064/526, loss: 0.00017204285541083664 2023-01-21 13:33:21.085974: step: 1068/526, loss: 0.00351715087890625 2023-01-21 13:33:22.219547: step: 1072/526, loss: 3.9672853745287284e-05 2023-01-21 13:33:23.360663: step: 1076/526, loss: 0.0001314163237111643 2023-01-21 13:33:24.469793: step: 1080/526, loss: 5.741119457525201e-05 2023-01-21 13:33:25.600122: step: 1084/526, loss: 0.009083651937544346 2023-01-21 13:33:26.763969: step: 1088/526, loss: 0.06130790710449219 2023-01-21 13:33:27.916293: step: 1092/526, loss: 0.0018129348754882812 2023-01-21 13:33:29.086441: step: 1096/526, loss: 0.00509986886754632 2023-01-21 13:33:30.228905: step: 1100/526, loss: 0.013837147504091263 2023-01-21 13:33:31.354739: step: 1104/526, loss: 5.91278057981981e-06 2023-01-21 13:33:32.497394: step: 1108/526, loss: 0.0002700805780477822 2023-01-21 13:33:33.626336: step: 1112/526, loss: 0.00014400482177734375 2023-01-21 13:33:34.749397: step: 1116/526, loss: 0.023198891431093216 2023-01-21 13:33:35.883007: step: 1120/526, loss: 0.004236125852912664 2023-01-21 13:33:37.003735: step: 1124/526, loss: 0.1325935423374176 2023-01-21 13:33:38.143521: step: 1128/526, loss: 0.0166015625 2023-01-21 13:33:39.285226: step: 1132/526, loss: 0.0017906189896166325 2023-01-21 13:33:40.394608: step: 1136/526, loss: 0.00027728083659894764 2023-01-21 13:33:41.538690: step: 1140/526, loss: 0.004579353146255016 2023-01-21 13:33:42.682276: step: 1144/526, loss: 0.00022583008103538305 2023-01-21 13:33:43.859451: step: 1148/526, loss: 0.0007657051319256425 2023-01-21 13:33:45.011981: step: 1152/526, loss: 0.016341017559170723 2023-01-21 13:33:46.163542: step: 1156/526, loss: 5.7220458984375e-06 2023-01-21 13:33:47.298873: step: 1160/526, loss: 0.10946149379014969 2023-01-21 13:33:48.442443: step: 1164/526, loss: 0.009315108880400658 2023-01-21 13:33:49.589076: step: 1168/526, loss: 0.00040664675179868937 2023-01-21 13:33:50.762580: step: 1172/526, loss: 8.487701961712446e-06 2023-01-21 13:33:51.884471: step: 1176/526, loss: 1.3065338862361386e-05 2023-01-21 13:33:53.006870: step: 1180/526, loss: -7.152556946721234e-08 2023-01-21 13:33:54.136140: step: 1184/526, loss: 0.006881427951157093 2023-01-21 13:33:55.246218: step: 1188/526, loss: 2.86102294921875e-05 2023-01-21 13:33:56.424025: step: 1192/526, loss: 0.02059183083474636 2023-01-21 13:33:57.569534: step: 1196/526, loss: -8.583065209677443e-07 2023-01-21 13:33:58.719804: step: 1200/526, loss: 0.0018804550636559725 2023-01-21 13:33:59.850285: step: 1204/526, loss: 0.006845378782600164 2023-01-21 13:34:00.948658: step: 1208/526, loss: 8.535384949936997e-06 2023-01-21 13:34:02.083865: step: 1212/526, loss: 0.034250449389219284 2023-01-21 13:34:03.211894: step: 1216/526, loss: 7.677078428969253e-06 2023-01-21 13:34:04.357363: step: 1220/526, loss: 0.011291694827377796 2023-01-21 13:34:05.517903: step: 1224/526, loss: 0.0005156517145223916 2023-01-21 13:34:06.678927: step: 1228/526, loss: 0.0013139724032953382 2023-01-21 13:34:07.826255: step: 1232/526, loss: 0.0016376018757000566 2023-01-21 13:34:08.980966: step: 1236/526, loss: 0.0072765350341796875 2023-01-21 13:34:10.132521: step: 1240/526, loss: 3.2138825190486386e-05 2023-01-21 13:34:11.270596: step: 1244/526, loss: 1.1444091796875e-05 2023-01-21 13:34:12.413481: step: 1248/526, loss: 0.0003936767461709678 2023-01-21 13:34:13.594720: step: 1252/526, loss: 0.07505837082862854 2023-01-21 13:34:14.723570: step: 1256/526, loss: 0.0004282951704226434 2023-01-21 13:34:15.881062: step: 1260/526, loss: 0.006148433778434992 2023-01-21 13:34:17.028168: step: 1264/526, loss: 3.604888843256049e-05 2023-01-21 13:34:18.159619: step: 1268/526, loss: 0.00011596680269576609 2023-01-21 13:34:19.329772: step: 1272/526, loss: 0.00016794205293990672 2023-01-21 13:34:20.425702: step: 1276/526, loss: 0.0003505706845317036 2023-01-21 13:34:21.579479: step: 1280/526, loss: 6.48498553346144e-06 2023-01-21 13:34:22.705869: step: 1284/526, loss: 1.316070574830519e-05 2023-01-21 13:34:23.865566: step: 1288/526, loss: 0.10438423603773117 2023-01-21 13:34:24.992787: step: 1292/526, loss: 0.0013463973300531507 2023-01-21 13:34:26.135868: step: 1296/526, loss: 0.0001501083461334929 2023-01-21 13:34:27.320340: step: 1300/526, loss: 0.042860984802246094 2023-01-21 13:34:28.458861: step: 1304/526, loss: 0.00031223296537064016 2023-01-21 13:34:29.616224: step: 1308/526, loss: 0.0028631212189793587 2023-01-21 13:34:30.740344: step: 1312/526, loss: 0.0017815589671954513 2023-01-21 13:34:31.894980: step: 1316/526, loss: 0.023362064734101295 2023-01-21 13:34:33.039609: step: 1320/526, loss: 0.0005491256597451866 2023-01-21 13:34:34.138812: step: 1324/526, loss: 8.144378080032766e-05 2023-01-21 13:34:35.302176: step: 1328/526, loss: 0.0805664137005806 2023-01-21 13:34:36.475175: step: 1332/526, loss: 1.0871888662222773e-05 2023-01-21 13:34:37.598833: step: 1336/526, loss: 4.291534787626006e-05 2023-01-21 13:34:38.718771: step: 1340/526, loss: 0.0004345893976278603 2023-01-21 13:34:39.846676: step: 1344/526, loss: -1.716613724056515e-06 2023-01-21 13:34:40.991359: step: 1348/526, loss: 0.02762756496667862 2023-01-21 13:34:42.135816: step: 1352/526, loss: 0.042530059814453125 2023-01-21 13:34:43.272626: step: 1356/526, loss: 0.00038051605224609375 2023-01-21 13:34:44.407604: step: 1360/526, loss: 0.020390892401337624 2023-01-21 13:34:45.561298: step: 1364/526, loss: 0.0005153656238690019 2023-01-21 13:34:46.687657: step: 1368/526, loss: 0.000133514404296875 2023-01-21 13:34:47.818121: step: 1372/526, loss: 0.006777000613510609 2023-01-21 13:34:48.963683: step: 1376/526, loss: 0.0012608527904376388 2023-01-21 13:34:50.087159: step: 1380/526, loss: 0.008293438702821732 2023-01-21 13:34:51.224795: step: 1384/526, loss: 0.0027749063447117805 2023-01-21 13:34:52.349873: step: 1388/526, loss: 0.0037401202134788036 2023-01-21 13:34:53.501988: step: 1392/526, loss: 0.0009616852039471269 2023-01-21 13:34:54.627543: step: 1396/526, loss: 1.296997106692288e-05 2023-01-21 13:34:55.766554: step: 1400/526, loss: 0.008121108636260033 2023-01-21 13:34:56.917397: step: 1404/526, loss: 4.76837158203125e-06 2023-01-21 13:34:58.041684: step: 1408/526, loss: 0.00025072097196243703 2023-01-21 13:34:59.199872: step: 1412/526, loss: 0.0007710456848144531 2023-01-21 13:35:00.353041: step: 1416/526, loss: 0.026173783466219902 2023-01-21 13:35:01.511238: step: 1420/526, loss: 0.024512100964784622 2023-01-21 13:35:02.611123: step: 1424/526, loss: 0.07559618353843689 2023-01-21 13:35:03.714021: step: 1428/526, loss: 4.57763690064894e-06 2023-01-21 13:35:04.869959: step: 1432/526, loss: 0.7336997985839844 2023-01-21 13:35:06.000022: step: 1436/526, loss: 0.001161432359367609 2023-01-21 13:35:07.155137: step: 1440/526, loss: 0.012558316811919212 2023-01-21 13:35:08.283864: step: 1444/526, loss: 0.04235009849071503 2023-01-21 13:35:09.428457: step: 1448/526, loss: 1.087188684323337e-05 2023-01-21 13:35:10.534054: step: 1452/526, loss: 3.356933666509576e-05 2023-01-21 13:35:11.642185: step: 1456/526, loss: 0.0007003784412518144 2023-01-21 13:35:12.774452: step: 1460/526, loss: 0.0005109787452965975 2023-01-21 13:35:13.917672: step: 1464/526, loss: 4.9591067181609105e-06 2023-01-21 13:35:15.053685: step: 1468/526, loss: 5.722046353184851e-06 2023-01-21 13:35:16.216254: step: 1472/526, loss: 6.294249760685489e-05 2023-01-21 13:35:17.366461: step: 1476/526, loss: 0.0004680633428506553 2023-01-21 13:35:18.528480: step: 1480/526, loss: 0.0031726837623864412 2023-01-21 13:35:19.674186: step: 1484/526, loss: 1.76429750808893e-06 2023-01-21 13:35:20.800962: step: 1488/526, loss: 0.003940010443329811 2023-01-21 13:35:21.965205: step: 1492/526, loss: 0.007775497157126665 2023-01-21 13:35:23.086960: step: 1496/526, loss: 9.727478754939511e-06 2023-01-21 13:35:24.224279: step: 1500/526, loss: 8.64505855133757e-05 2023-01-21 13:35:25.389938: step: 1504/526, loss: 0.044715117663145065 2023-01-21 13:35:26.503854: step: 1508/526, loss: 0.002186393830925226 2023-01-21 13:35:27.655305: step: 1512/526, loss: 0.0029252052772790194 2023-01-21 13:35:28.799146: step: 1516/526, loss: 0.0035707473289221525 2023-01-21 13:35:29.956269: step: 1520/526, loss: 0.00016727446927689016 2023-01-21 13:35:31.075330: step: 1524/526, loss: 0.00037192701711319387 2023-01-21 13:35:32.209480: step: 1528/526, loss: -1.9073486328125e-06 2023-01-21 13:35:33.356538: step: 1532/526, loss: 0.06125602871179581 2023-01-21 13:35:34.481386: step: 1536/526, loss: 0.0005928039317950606 2023-01-21 13:35:35.641173: step: 1540/526, loss: 0.00128850934561342 2023-01-21 13:35:36.804623: step: 1544/526, loss: 0.0016193389892578125 2023-01-21 13:35:37.933916: step: 1548/526, loss: 0.08031503856182098 2023-01-21 13:35:39.091905: step: 1552/526, loss: 0.0033309936989098787 2023-01-21 13:35:40.244183: step: 1556/526, loss: 1.5926361811580136e-05 2023-01-21 13:35:41.374947: step: 1560/526, loss: 3.80516066798009e-05 2023-01-21 13:35:42.522398: step: 1564/526, loss: 0.0001712799130473286 2023-01-21 13:35:43.667489: step: 1568/526, loss: 0.009743690490722656 2023-01-21 13:35:44.830968: step: 1572/526, loss: 0.0004024505615234375 2023-01-21 13:35:45.947451: step: 1576/526, loss: -6.4373016357421875e-06 2023-01-21 13:35:47.072735: step: 1580/526, loss: 0.272746741771698 2023-01-21 13:35:48.206939: step: 1584/526, loss: 0.0018127441871911287 2023-01-21 13:35:49.373570: step: 1588/526, loss: 0.0005079269176349044 2023-01-21 13:35:50.540977: step: 1592/526, loss: 0.001050758408382535 2023-01-21 13:35:51.680641: step: 1596/526, loss: 0.0014330863486975431 2023-01-21 13:35:52.810103: step: 1600/526, loss: 0.015057658776640892 2023-01-21 13:35:53.935792: step: 1604/526, loss: 0.04213309288024902 2023-01-21 13:35:55.037856: step: 1608/526, loss: 0.0001644372969167307 2023-01-21 13:35:56.237335: step: 1612/526, loss: 0.0016588210128247738 2023-01-21 13:35:57.390713: step: 1616/526, loss: 3.898143768310547e-05 2023-01-21 13:35:58.534841: step: 1620/526, loss: 0.00263214111328125 2023-01-21 13:35:59.668150: step: 1624/526, loss: 6.771087646484375e-05 2023-01-21 13:36:00.787078: step: 1628/526, loss: 0.0016007423400878906 2023-01-21 13:36:01.963362: step: 1632/526, loss: 0.00038547517033293843 2023-01-21 13:36:03.092535: step: 1636/526, loss: 5.0735470722429454e-05 2023-01-21 13:36:04.221151: step: 1640/526, loss: 0.0005164146423339844 2023-01-21 13:36:05.324835: step: 1644/526, loss: 1.9073486328125e-05 2023-01-21 13:36:06.437272: step: 1648/526, loss: 0.020950699225068092 2023-01-21 13:36:07.572559: step: 1652/526, loss: 1.1634827387752011e-05 2023-01-21 13:36:08.699401: step: 1656/526, loss: 0.013421058654785156 2023-01-21 13:36:09.849797: step: 1660/526, loss: 0.01631312444806099 2023-01-21 13:36:10.977961: step: 1664/526, loss: -7.05719003235572e-06 2023-01-21 13:36:12.094438: step: 1668/526, loss: 0.0001385688956361264 2023-01-21 13:36:13.242775: step: 1672/526, loss: 0.01915550045669079 2023-01-21 13:36:14.392983: step: 1676/526, loss: 0.0016635895008221269 2023-01-21 13:36:15.531618: step: 1680/526, loss: 0.05268049240112305 2023-01-21 13:36:16.669570: step: 1684/526, loss: 5.831718590343371e-05 2023-01-21 13:36:17.808186: step: 1688/526, loss: 0.00021495818509720266 2023-01-21 13:36:18.972810: step: 1692/526, loss: 0.0003068924124818295 2023-01-21 13:36:20.116230: step: 1696/526, loss: 3.051757857974735e-06 2023-01-21 13:36:21.285857: step: 1700/526, loss: -1.52587890625e-05 2023-01-21 13:36:22.399929: step: 1704/526, loss: 1.4400482541532256e-05 2023-01-21 13:36:23.523555: step: 1708/526, loss: 0.019866324961185455 2023-01-21 13:36:24.650846: step: 1712/526, loss: 2.622604552016128e-05 2023-01-21 13:36:25.765451: step: 1716/526, loss: 0.0023622512817382812 2023-01-21 13:36:26.878971: step: 1720/526, loss: -7.62939453125e-06 2023-01-21 13:36:28.034614: step: 1724/526, loss: 0.00035152435884810984 2023-01-21 13:36:29.149283: step: 1728/526, loss: -1.4781952586417901e-06 2023-01-21 13:36:30.286655: step: 1732/526, loss: 0.000339508056640625 2023-01-21 13:36:31.444759: step: 1736/526, loss: 0.0001672744838288054 2023-01-21 13:36:32.611456: step: 1740/526, loss: 0.025621796026825905 2023-01-21 13:36:33.734403: step: 1744/526, loss: 0.0007631301996298134 2023-01-21 13:36:34.890766: step: 1748/526, loss: 0.002261257264763117 2023-01-21 13:36:36.032687: step: 1752/526, loss: 0.008528710342943668 2023-01-21 13:36:37.154886: step: 1756/526, loss: 0.013813590630888939 2023-01-21 13:36:38.262610: step: 1760/526, loss: 0.00030117033747956157 2023-01-21 13:36:39.404218: step: 1764/526, loss: 0.0006011963123455644 2023-01-21 13:36:40.531495: step: 1768/526, loss: 3.5285947888041846e-06 2023-01-21 13:36:41.656986: step: 1772/526, loss: 0.0001256942778127268 2023-01-21 13:36:42.791888: step: 1776/526, loss: 1.811981201171875e-05 2023-01-21 13:36:43.947567: step: 1780/526, loss: 1.7929078239831142e-05 2023-01-21 13:36:45.073863: step: 1784/526, loss: 0.00043010711669921875 2023-01-21 13:36:46.244320: step: 1788/526, loss: 0.0030249119736254215 2023-01-21 13:36:47.384236: step: 1792/526, loss: 0.0048306467942893505 2023-01-21 13:36:48.538089: step: 1796/526, loss: 0.001056671142578125 2023-01-21 13:36:49.689065: step: 1800/526, loss: 0.0009451866499148309 2023-01-21 13:36:50.808740: step: 1804/526, loss: 0.07289252430200577 2023-01-21 13:36:51.927860: step: 1808/526, loss: 0.08858004212379456 2023-01-21 13:36:53.064745: step: 1812/526, loss: 0.00022373200044967234 2023-01-21 13:36:54.211847: step: 1816/526, loss: 0.02032756805419922 2023-01-21 13:36:55.388763: step: 1820/526, loss: 0.017384672537446022 2023-01-21 13:36:56.508138: step: 1824/526, loss: 0.0001659393310546875 2023-01-21 13:36:57.638810: step: 1828/526, loss: 0.009712601080536842 2023-01-21 13:36:58.792660: step: 1832/526, loss: 3.967285010730848e-05 2023-01-21 13:36:59.954622: step: 1836/526, loss: 0.0002857208310160786 2023-01-21 13:37:01.081766: step: 1840/526, loss: 0.00012111663818359375 2023-01-21 13:37:02.212665: step: 1844/526, loss: 0.04125986248254776 2023-01-21 13:37:03.318623: step: 1848/526, loss: 0.001188087509945035 2023-01-21 13:37:04.450669: step: 1852/526, loss: 0.007689285557717085 2023-01-21 13:37:05.600549: step: 1856/526, loss: 0.00015559197345282882 2023-01-21 13:37:06.714507: step: 1860/526, loss: 0.0003532409609761089 2023-01-21 13:37:07.849399: step: 1864/526, loss: 0.0001885414239950478 2023-01-21 13:37:08.996302: step: 1868/526, loss: 0.0015705109108239412 2023-01-21 13:37:10.147666: step: 1872/526, loss: 0.000545501708984375 2023-01-21 13:37:11.264645: step: 1876/526, loss: 0.0002323150692973286 2023-01-21 13:37:12.400053: step: 1880/526, loss: 0.00027551650418899953 2023-01-21 13:37:13.545473: step: 1884/526, loss: 0.007864952087402344 2023-01-21 13:37:14.726533: step: 1888/526, loss: 0.27130481600761414 2023-01-21 13:37:15.846368: step: 1892/526, loss: 2.9277802241267636e-05 2023-01-21 13:37:16.986947: step: 1896/526, loss: 2.0027162008773303e-06 2023-01-21 13:37:18.094962: step: 1900/526, loss: 5.817413693876006e-06 2023-01-21 13:37:19.232982: step: 1904/526, loss: 0.0028642655815929174 2023-01-21 13:37:20.362010: step: 1908/526, loss: 6.10351571594947e-06 2023-01-21 13:37:21.491742: step: 1912/526, loss: 1.087188684323337e-05 2023-01-21 13:37:22.592761: step: 1916/526, loss: 0.0011740685440599918 2023-01-21 13:37:23.732909: step: 1920/526, loss: 0.09104882180690765 2023-01-21 13:37:24.850340: step: 1924/526, loss: 0.002863311907276511 2023-01-21 13:37:25.993324: step: 1928/526, loss: 1.678466833254788e-05 2023-01-21 13:37:27.118179: step: 1932/526, loss: 0.00023956299992278218 2023-01-21 13:37:28.301785: step: 1936/526, loss: 7.181167165981606e-05 2023-01-21 13:37:29.410919: step: 1940/526, loss: 0.0010404587956145406 2023-01-21 13:37:30.533469: step: 1944/526, loss: 0.02699403651058674 2023-01-21 13:37:31.678599: step: 1948/526, loss: 0.00089349749032408 2023-01-21 13:37:32.814277: step: 1952/526, loss: 0.010082244873046875 2023-01-21 13:37:33.933682: step: 1956/526, loss: 5.378723290050402e-05 2023-01-21 13:37:35.096400: step: 1960/526, loss: 0.002050781389698386 2023-01-21 13:37:36.230277: step: 1964/526, loss: 0.00013637542724609375 2023-01-21 13:37:37.350357: step: 1968/526, loss: 4.673004241340095e-06 2023-01-21 13:37:38.459944: step: 1972/526, loss: 0.0005458832019940019 2023-01-21 13:37:39.608955: step: 1976/526, loss: 0.00014085769362282008 2023-01-21 13:37:40.741501: step: 1980/526, loss: 0.0729871392250061 2023-01-21 13:37:41.893501: step: 1984/526, loss: 0.00127582554705441 2023-01-21 13:37:42.998633: step: 1988/526, loss: 0.01306066568940878 2023-01-21 13:37:44.104962: step: 1992/526, loss: 0.013928795233368874 2023-01-21 13:37:45.283284: step: 1996/526, loss: 0.010881423950195312 2023-01-21 13:37:46.412584: step: 2000/526, loss: 0.33638009428977966 2023-01-21 13:37:47.539823: step: 2004/526, loss: 3.528594970703125e-05 2023-01-21 13:37:48.672414: step: 2008/526, loss: 0.00032930372981354594 2023-01-21 13:37:49.805397: step: 2012/526, loss: 0.010987472720444202 2023-01-21 13:37:50.978115: step: 2016/526, loss: 0.012489271350204945 2023-01-21 13:37:52.124900: step: 2020/526, loss: 5.2165985835017636e-05 2023-01-21 13:37:53.247570: step: 2024/526, loss: 1.4829635802016128e-05 2023-01-21 13:37:54.380460: step: 2028/526, loss: 0.05006008222699165 2023-01-21 13:37:55.490236: step: 2032/526, loss: 5.760193016612902e-05 2023-01-21 13:37:56.639361: step: 2036/526, loss: 0.00137157435528934 2023-01-21 13:37:57.807697: step: 2040/526, loss: 0.026972580701112747 2023-01-21 13:37:58.934209: step: 2044/526, loss: 7.82012921263231e-06 2023-01-21 13:38:00.066533: step: 2048/526, loss: 1.239776611328125e-05 2023-01-21 13:38:01.198136: step: 2052/526, loss: 0.018094254657626152 2023-01-21 13:38:02.337117: step: 2056/526, loss: 0.0017058133380487561 2023-01-21 13:38:03.505097: step: 2060/526, loss: 0.0008918762323446572 2023-01-21 13:38:04.608823: step: 2064/526, loss: 0.00017433166794944555 2023-01-21 13:38:05.803124: step: 2068/526, loss: 9.889602370094508e-05 2023-01-21 13:38:06.953777: step: 2072/526, loss: 0.0003669738944154233 2023-01-21 13:38:08.115349: step: 2076/526, loss: 9.059906733455136e-05 2023-01-21 13:38:09.259834: step: 2080/526, loss: 0.011842727661132812 2023-01-21 13:38:10.394036: step: 2084/526, loss: 0.06062927097082138 2023-01-21 13:38:11.514626: step: 2088/526, loss: 0.016322897747159004 2023-01-21 13:38:12.657669: step: 2092/526, loss: -2.2029877072782256e-05 2023-01-21 13:38:13.799694: step: 2096/526, loss: 0.00043077467125840485 2023-01-21 13:38:14.923518: step: 2100/526, loss: 3.62396240234375e-05 2023-01-21 13:38:16.083195: step: 2104/526, loss: 0.02538776397705078 ================================================== Loss: 0.013 -------------------- Dev: {'event': {'p': 0.614240170031881, 'r': 0.7696404793608522, 'f1': 0.6832151300236406}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Test: {'event': {'p': 0.6294871794871795, 'r': 0.7687891440501043, 'f1': 0.6921992481203008}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Chinese: {'event': {'p': 0.631578947368421, 'r': 0.8888888888888888, 'f1': 0.7384615384615385}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Korean: {'event': {'p': 0.6818181818181818, 'r': 0.47619047619047616, 'f1': 0.5607476635514018}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} Russian: {'event': {'p': 0.4473684210526316, 'r': 0.4722222222222222, 'f1': 0.4594594594594595}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 27} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 28 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:38:59.099492: step: 4/526, loss: 0.012076378799974918 2023-01-21 13:39:00.281670: step: 8/526, loss: 0.012168694287538528 2023-01-21 13:39:01.409684: step: 12/526, loss: 0.0034654466435313225 2023-01-21 13:39:02.546179: step: 16/526, loss: 0.0021366120781749487 2023-01-21 13:39:03.669897: step: 20/526, loss: 0.00029096603975631297 2023-01-21 13:39:04.796276: step: 24/526, loss: 9.880065772449598e-05 2023-01-21 13:39:05.911019: step: 28/526, loss: 1.2207032341393642e-05 2023-01-21 13:39:07.021654: step: 32/526, loss: 0.00024100541486404836 2023-01-21 13:39:08.137804: step: 36/526, loss: 0.003618240589275956 2023-01-21 13:39:09.277876: step: 40/526, loss: 0.02555117756128311 2023-01-21 13:39:10.372275: step: 44/526, loss: 9.775161743164062e-06 2023-01-21 13:39:11.513829: step: 48/526, loss: 0.04503012076020241 2023-01-21 13:39:12.630558: step: 52/526, loss: 0.0017394066089764237 2023-01-21 13:39:13.790882: step: 56/526, loss: 0.06529799103736877 2023-01-21 13:39:14.948276: step: 60/526, loss: 3.566742088878527e-05 2023-01-21 13:39:16.065774: step: 64/526, loss: 2.098083541568485e-06 2023-01-21 13:39:17.222412: step: 68/526, loss: 0.02854161337018013 2023-01-21 13:39:18.348019: step: 72/526, loss: 0.06816143542528152 2023-01-21 13:39:19.465467: step: 76/526, loss: 0.018248939886689186 2023-01-21 13:39:20.587981: step: 80/526, loss: 8.58306884765625e-05 2023-01-21 13:39:21.718406: step: 84/526, loss: 0.0007057190523482859 2023-01-21 13:39:22.851367: step: 88/526, loss: 0.0052852630615234375 2023-01-21 13:39:23.980668: step: 92/526, loss: 1.125335711549269e-05 2023-01-21 13:39:25.100426: step: 96/526, loss: 0.00024313927860930562 2023-01-21 13:39:26.257650: step: 100/526, loss: 0.0002695083385333419 2023-01-21 13:39:27.390765: step: 104/526, loss: 0.0020261763129383326 2023-01-21 13:39:28.545723: step: 108/526, loss: 0.0001333236723439768 2023-01-21 13:39:29.705394: step: 112/526, loss: 0.0025209428276866674 2023-01-21 13:39:30.845300: step: 116/526, loss: 8.77380352903856e-06 2023-01-21 13:39:31.984784: step: 120/526, loss: 2.0980837689421605e-06 2023-01-21 13:39:33.161413: step: 124/526, loss: 0.036746978759765625 2023-01-21 13:39:34.283433: step: 128/526, loss: 0.0001458644837839529 2023-01-21 13:39:35.417481: step: 132/526, loss: 0.00012021065049339086 2023-01-21 13:39:36.569733: step: 136/526, loss: 0.0001428842660970986 2023-01-21 13:39:37.705117: step: 140/526, loss: 7.867813110351562e-06 2023-01-21 13:39:38.834156: step: 144/526, loss: 1.583099401614163e-05 2023-01-21 13:39:39.976627: step: 148/526, loss: 8.344650268554688e-06 2023-01-21 13:39:41.088914: step: 152/526, loss: 0.0009455680847167969 2023-01-21 13:39:42.233011: step: 156/526, loss: 0.0001331329403910786 2023-01-21 13:39:43.349209: step: 160/526, loss: 0.0005329131963662803 2023-01-21 13:39:44.492034: step: 164/526, loss: 0.001282692071981728 2023-01-21 13:39:45.615687: step: 168/526, loss: 0.0004152298206463456 2023-01-21 13:39:46.773354: step: 172/526, loss: 0.0011913300259038806 2023-01-21 13:39:47.883451: step: 176/526, loss: 0.03232569620013237 2023-01-21 13:39:48.985593: step: 180/526, loss: 0.0036855696234852076 2023-01-21 13:39:50.142438: step: 184/526, loss: 0.00052728655282408 2023-01-21 13:39:51.266695: step: 188/526, loss: 4.243850526108872e-06 2023-01-21 13:39:52.403455: step: 192/526, loss: 0.006767845246940851 2023-01-21 13:39:53.524825: step: 196/526, loss: 3.1089784897631034e-05 2023-01-21 13:39:54.699323: step: 200/526, loss: 0.0008615494007244706 2023-01-21 13:39:55.822086: step: 204/526, loss: 1.2493132999225054e-05 2023-01-21 13:39:56.978459: step: 208/526, loss: 0.006457710172981024 2023-01-21 13:39:58.127251: step: 212/526, loss: 0.00015373229689430445 2023-01-21 13:39:59.272046: step: 216/526, loss: 0.06100053712725639 2023-01-21 13:40:00.397846: step: 220/526, loss: 0.019083594903349876 2023-01-21 13:40:01.534776: step: 224/526, loss: 0.0010037422180175781 2023-01-21 13:40:02.662938: step: 228/526, loss: 0.02661914750933647 2023-01-21 13:40:03.773363: step: 232/526, loss: 9.193419828079641e-05 2023-01-21 13:40:04.931467: step: 236/526, loss: 0.08594484627246857 2023-01-21 13:40:06.078524: step: 240/526, loss: 2.8133392333984375e-05 2023-01-21 13:40:07.306274: step: 244/526, loss: 5.950927879894152e-05 2023-01-21 13:40:08.449909: step: 248/526, loss: 0.0008760452619753778 2023-01-21 13:40:09.616791: step: 252/526, loss: 0.003277254058048129 2023-01-21 13:40:10.782449: step: 256/526, loss: 0.0628608763217926 2023-01-21 13:40:11.951649: step: 260/526, loss: 0.02050018310546875 2023-01-21 13:40:13.098890: step: 264/526, loss: 0.00021457672119140625 2023-01-21 13:40:14.232132: step: 268/526, loss: 8.39233416627394e-06 2023-01-21 13:40:15.358284: step: 272/526, loss: 0.014435388147830963 2023-01-21 13:40:16.509919: step: 276/526, loss: 0.003130245255306363 2023-01-21 13:40:17.646778: step: 280/526, loss: 0.0004459381161723286 2023-01-21 13:40:18.798962: step: 284/526, loss: 1.3351441339182202e-06 2023-01-21 13:40:19.936663: step: 288/526, loss: 8.897781663108617e-05 2023-01-21 13:40:21.055318: step: 292/526, loss: 0.0006437301635742188 2023-01-21 13:40:22.171392: step: 296/526, loss: 0.004516029264777899 2023-01-21 13:40:23.295353: step: 300/526, loss: 0.009015274234116077 2023-01-21 13:40:24.434641: step: 304/526, loss: 0.02776656113564968 2023-01-21 13:40:25.542771: step: 308/526, loss: 6.10351571594947e-06 2023-01-21 13:40:26.672343: step: 312/526, loss: 0.00010604858834994957 2023-01-21 13:40:27.831812: step: 316/526, loss: 0.00014171600923873484 2023-01-21 13:40:28.982609: step: 320/526, loss: 0.0011230945819988847 2023-01-21 13:40:30.129750: step: 324/526, loss: 0.007213592529296875 2023-01-21 13:40:31.260762: step: 328/526, loss: 0.005086231045424938 2023-01-21 13:40:32.385839: step: 332/526, loss: 0.04716377332806587 2023-01-21 13:40:33.541668: step: 336/526, loss: 8.020401583053172e-05 2023-01-21 13:40:34.666631: step: 340/526, loss: 0.03883972018957138 2023-01-21 13:40:35.827583: step: 344/526, loss: 7.82012939453125e-05 2023-01-21 13:40:36.966157: step: 348/526, loss: 0.714342474937439 2023-01-21 13:40:38.100170: step: 352/526, loss: 0.00027503969613462687 2023-01-21 13:40:39.200839: step: 356/526, loss: 6.637573824264109e-05 2023-01-21 13:40:40.332291: step: 360/526, loss: 2.1791458493680693e-05 2023-01-21 13:40:41.472245: step: 364/526, loss: 0.0007914542802609503 2023-01-21 13:40:42.612946: step: 368/526, loss: 0.27310752868652344 2023-01-21 13:40:43.742266: step: 372/526, loss: 4.482269287109375e-05 2023-01-21 13:40:44.890770: step: 376/526, loss: 1.71661376953125e-05 2023-01-21 13:40:46.004763: step: 380/526, loss: 0.0014087200397625566 2023-01-21 13:40:47.122844: step: 384/526, loss: 0.00010261536226607859 2023-01-21 13:40:48.271430: step: 388/526, loss: 6.66618361719884e-05 2023-01-21 13:40:49.392020: step: 392/526, loss: 0.0027841569390147924 2023-01-21 13:40:50.569673: step: 396/526, loss: 0.0003030777152162045 2023-01-21 13:40:51.700711: step: 400/526, loss: 0.0002817154163494706 2023-01-21 13:40:52.845774: step: 404/526, loss: 7.686614844715223e-05 2023-01-21 13:40:54.016126: step: 408/526, loss: 0.0007841110345907509 2023-01-21 13:40:55.171959: step: 412/526, loss: 0.0003810882626567036 2023-01-21 13:40:56.329748: step: 416/526, loss: 1.1634827387752011e-05 2023-01-21 13:40:57.446395: step: 420/526, loss: 0.0004432678397279233 2023-01-21 13:40:58.600038: step: 424/526, loss: 0.006811332888901234 2023-01-21 13:40:59.755576: step: 428/526, loss: 0.0010438918834552169 2023-01-21 13:41:00.873646: step: 432/526, loss: 0.0017036438221111894 2023-01-21 13:41:01.989220: step: 436/526, loss: 0.0005007743602618575 2023-01-21 13:41:03.132641: step: 440/526, loss: 0.001735734986141324 2023-01-21 13:41:04.278741: step: 444/526, loss: 0.0012195587623864412 2023-01-21 13:41:05.401047: step: 448/526, loss: 0.00018138886662200093 2023-01-21 13:41:06.516146: step: 452/526, loss: 0.00047025681124068797 2023-01-21 13:41:07.655368: step: 456/526, loss: 0.004047823138535023 2023-01-21 13:41:08.831329: step: 460/526, loss: 0.014491081237792969 2023-01-21 13:41:09.951725: step: 464/526, loss: -3.4332276754867053e-06 2023-01-21 13:41:11.121094: step: 468/526, loss: 0.011139297857880592 2023-01-21 13:41:12.272936: step: 472/526, loss: 2.765655517578125e-05 2023-01-21 13:41:13.386770: step: 476/526, loss: 0.0001720428408589214 2023-01-21 13:41:14.504438: step: 480/526, loss: -8.96453821042087e-06 2023-01-21 13:41:15.677405: step: 484/526, loss: 0.0004913330194540322 2023-01-21 13:41:16.811123: step: 488/526, loss: 1.6975403923424892e-05 2023-01-21 13:41:17.953487: step: 492/526, loss: 0.017278004437685013 2023-01-21 13:41:19.064462: step: 496/526, loss: 4.4631960918195546e-05 2023-01-21 13:41:20.264696: step: 500/526, loss: 0.00047216415987350047 2023-01-21 13:41:21.385956: step: 504/526, loss: 0.07259654998779297 2023-01-21 13:41:22.508946: step: 508/526, loss: 1.087188684323337e-05 2023-01-21 13:41:23.632986: step: 512/526, loss: 0.00179119105450809 2023-01-21 13:41:24.805044: step: 516/526, loss: 0.00010814666893566027 2023-01-21 13:41:25.920221: step: 520/526, loss: 0.0005269527900964022 2023-01-21 13:41:27.038308: step: 524/526, loss: 0.0021200180053710938 2023-01-21 13:41:28.171328: step: 528/526, loss: 0.00012512206740211695 2023-01-21 13:41:29.301825: step: 532/526, loss: 0.004712152760475874 2023-01-21 13:41:30.402951: step: 536/526, loss: 0.00177001953125 2023-01-21 13:41:31.515909: step: 540/526, loss: 0.05389223247766495 2023-01-21 13:41:32.679537: step: 544/526, loss: 0.00669097900390625 2023-01-21 13:41:33.818432: step: 548/526, loss: 7.43865984986769e-06 2023-01-21 13:41:34.951486: step: 552/526, loss: 0.0019368171924725175 2023-01-21 13:41:36.111446: step: 556/526, loss: 0.5350887179374695 2023-01-21 13:41:37.255296: step: 560/526, loss: 2.384185791015625e-05 2023-01-21 13:41:38.379098: step: 564/526, loss: 0.004017638973891735 2023-01-21 13:41:39.464909: step: 568/526, loss: 1.4781952586417901e-06 2023-01-21 13:41:40.604637: step: 572/526, loss: 0.011543608270585537 2023-01-21 13:41:41.749899: step: 576/526, loss: 0.0011549950577318668 2023-01-21 13:41:42.903582: step: 580/526, loss: -2.86102294921875e-05 2023-01-21 13:41:44.032219: step: 584/526, loss: 0.003205585526302457 2023-01-21 13:41:45.162896: step: 588/526, loss: 0.0004784584161825478 2023-01-21 13:41:46.293560: step: 592/526, loss: 3.528594970703125e-05 2023-01-21 13:41:47.436921: step: 596/526, loss: 9.279252117266878e-05 2023-01-21 13:41:48.563207: step: 600/526, loss: 0.0007670402992516756 2023-01-21 13:41:49.697606: step: 604/526, loss: 0.0010917664039880037 2023-01-21 13:41:50.843033: step: 608/526, loss: 0.04972153156995773 2023-01-21 13:41:51.984546: step: 612/526, loss: 0.012523460201919079 2023-01-21 13:41:53.116797: step: 616/526, loss: 0.0001637458917684853 2023-01-21 13:41:54.222310: step: 620/526, loss: 0.0012747765285894275 2023-01-21 13:41:55.324039: step: 624/526, loss: -6.198883056640625e-06 2023-01-21 13:41:56.450766: step: 628/526, loss: 0.0009052276145666838 2023-01-21 13:41:57.609380: step: 632/526, loss: 0.05101451650261879 2023-01-21 13:41:58.753391: step: 636/526, loss: 0.004829693119972944 2023-01-21 13:41:59.887958: step: 640/526, loss: 0.0006771088228560984 2023-01-21 13:42:01.004301: step: 644/526, loss: 0.008210944943130016 2023-01-21 13:42:02.140698: step: 648/526, loss: 0.003708410309627652 2023-01-21 13:42:03.272155: step: 652/526, loss: -8.296966370835435e-06 2023-01-21 13:42:04.404718: step: 656/526, loss: 0.00738601665943861 2023-01-21 13:42:05.533284: step: 660/526, loss: 2.8705595468636602e-05 2023-01-21 13:42:06.687138: step: 664/526, loss: 0.003025054931640625 2023-01-21 13:42:07.846227: step: 668/526, loss: 0.010843753814697266 2023-01-21 13:42:08.960222: step: 672/526, loss: 1.7166139514301904e-06 2023-01-21 13:42:10.101554: step: 676/526, loss: 0.0007969856378622353 2023-01-21 13:42:11.215270: step: 680/526, loss: 0.028282547369599342 2023-01-21 13:42:12.345159: step: 684/526, loss: 0.00011711120896507055 2023-01-21 13:42:13.532791: step: 688/526, loss: 3.6048892070539296e-05 2023-01-21 13:42:14.683817: step: 692/526, loss: 4.863738922722405e-06 2023-01-21 13:42:15.836722: step: 696/526, loss: 0.0001727580965962261 2023-01-21 13:42:16.960331: step: 700/526, loss: 0.00011482238187454641 2023-01-21 13:42:18.097656: step: 704/526, loss: 0.008797836489975452 2023-01-21 13:42:19.236102: step: 708/526, loss: 0.0014982223510742188 2023-01-21 13:42:20.360677: step: 712/526, loss: 8.344650996150449e-05 2023-01-21 13:42:21.493504: step: 716/526, loss: 0.00037832261295989156 2023-01-21 13:42:22.642658: step: 720/526, loss: 0.055109117180109024 2023-01-21 13:42:23.790324: step: 724/526, loss: 0.00867166556417942 2023-01-21 13:42:24.919058: step: 728/526, loss: 0.002102804370224476 2023-01-21 13:42:26.019275: step: 732/526, loss: 0.002773284912109375 2023-01-21 13:42:27.158135: step: 736/526, loss: 0.0007508278358727694 2023-01-21 13:42:28.285944: step: 740/526, loss: 0.0006198883056640625 2023-01-21 13:42:29.425779: step: 744/526, loss: 0.0018260955112054944 2023-01-21 13:42:30.565587: step: 748/526, loss: 0.00012454987154342234 2023-01-21 13:42:31.687234: step: 752/526, loss: 0.009970664978027344 2023-01-21 13:42:32.827308: step: 756/526, loss: 0.021911144256591797 2023-01-21 13:42:33.938099: step: 760/526, loss: 1.0776519957289565e-05 2023-01-21 13:42:35.072058: step: 764/526, loss: 1.8692018784349784e-05 2023-01-21 13:42:36.211578: step: 768/526, loss: 2.536773718020413e-05 2023-01-21 13:42:37.359682: step: 772/526, loss: 6.866455805720761e-05 2023-01-21 13:42:38.480266: step: 776/526, loss: 0.0012810706393793225 2023-01-21 13:42:39.601960: step: 780/526, loss: 0.0014472008915618062 2023-01-21 13:42:40.738377: step: 784/526, loss: 0.02641601487994194 2023-01-21 13:42:41.852642: step: 788/526, loss: 0.002408027881756425 2023-01-21 13:42:42.990095: step: 792/526, loss: 2.021789623540826e-05 2023-01-21 13:42:44.094632: step: 796/526, loss: 0.024468136951327324 2023-01-21 13:42:45.234337: step: 800/526, loss: 0.0036302567459642887 2023-01-21 13:42:46.350592: step: 804/526, loss: 0.0004161834658589214 2023-01-21 13:42:47.549674: step: 808/526, loss: 0.007156944368034601 2023-01-21 13:42:48.671568: step: 812/526, loss: 0.0001233577640959993 2023-01-21 13:42:49.796000: step: 816/526, loss: 0.0041217803955078125 2023-01-21 13:42:50.945220: step: 820/526, loss: 0.01782817766070366 2023-01-21 13:42:52.074631: step: 824/526, loss: 8.869171142578125e-05 2023-01-21 13:42:53.201922: step: 828/526, loss: 0.0035326960496604443 2023-01-21 13:42:54.302589: step: 832/526, loss: 3.986358569818549e-05 2023-01-21 13:42:55.454071: step: 836/526, loss: 5.1116945542162284e-05 2023-01-21 13:42:56.574492: step: 840/526, loss: -2.86102294921875e-06 2023-01-21 13:42:57.738249: step: 844/526, loss: 0.0006128310924395919 2023-01-21 13:42:58.884422: step: 848/526, loss: 8.75473051564768e-05 2023-01-21 13:43:00.007168: step: 852/526, loss: -2.002715973503655e-06 2023-01-21 13:43:01.148213: step: 856/526, loss: 0.027760697528719902 2023-01-21 13:43:02.308430: step: 860/526, loss: 0.0004444122314453125 2023-01-21 13:43:03.440900: step: 864/526, loss: 2.3651124138268642e-05 2023-01-21 13:43:04.621560: step: 868/526, loss: 0.005697059445083141 2023-01-21 13:43:05.750228: step: 872/526, loss: 0.00011959076073253527 2023-01-21 13:43:06.898189: step: 876/526, loss: 0.009966659359633923 2023-01-21 13:43:08.044239: step: 880/526, loss: 0.06708285212516785 2023-01-21 13:43:09.185003: step: 884/526, loss: 8.354186866199598e-05 2023-01-21 13:43:10.327589: step: 888/526, loss: 0.004384040366858244 2023-01-21 13:43:11.476942: step: 892/526, loss: 0.007876587100327015 2023-01-21 13:43:12.644232: step: 896/526, loss: 0.01158838253468275 2023-01-21 13:43:13.762226: step: 900/526, loss: 0.0033540725708007812 2023-01-21 13:43:14.917560: step: 904/526, loss: 2.212524486822076e-05 2023-01-21 13:43:16.093650: step: 908/526, loss: 0.04241600260138512 2023-01-21 13:43:17.228947: step: 912/526, loss: 0.0035985945723950863 2023-01-21 13:43:18.394749: step: 916/526, loss: 0.000552749668713659 2023-01-21 13:43:19.555510: step: 920/526, loss: 0.00013828277587890625 2023-01-21 13:43:20.670435: step: 924/526, loss: 0.001910400460474193 2023-01-21 13:43:21.853801: step: 928/526, loss: 0.0011866569984704256 2023-01-21 13:43:22.998915: step: 932/526, loss: 0.0017074585193768144 2023-01-21 13:43:24.142656: step: 936/526, loss: 0.001060104463249445 2023-01-21 13:43:25.288945: step: 940/526, loss: 1.506805438111769e-05 2023-01-21 13:43:26.413268: step: 944/526, loss: 0.00022454261488746852 2023-01-21 13:43:27.559123: step: 948/526, loss: 0.0009961128234863281 2023-01-21 13:43:28.684788: step: 952/526, loss: 0.0019540786743164062 2023-01-21 13:43:29.824126: step: 956/526, loss: 1.792907642084174e-05 2023-01-21 13:43:30.961527: step: 960/526, loss: 0.0009687423589639366 2023-01-21 13:43:32.075092: step: 964/526, loss: 0.00042819976806640625 2023-01-21 13:43:33.212072: step: 968/526, loss: 0.0016031265258789062 2023-01-21 13:43:34.368304: step: 972/526, loss: 0.0008888245210982859 2023-01-21 13:43:35.486104: step: 976/526, loss: 0.0009263515239581466 2023-01-21 13:43:36.645679: step: 980/526, loss: 0.01993555948138237 2023-01-21 13:43:37.788756: step: 984/526, loss: 0.0012018203269690275 2023-01-21 13:43:38.969991: step: 988/526, loss: 0.00438580522313714 2023-01-21 13:43:40.120093: step: 992/526, loss: 0.029940033331513405 2023-01-21 13:43:41.230172: step: 996/526, loss: 0.009188842959702015 2023-01-21 13:43:42.363528: step: 1000/526, loss: 0.221476748585701 2023-01-21 13:43:43.486082: step: 1004/526, loss: 0.00042743684025481343 2023-01-21 13:43:44.625692: step: 1008/526, loss: 0.0022742270957678556 2023-01-21 13:43:45.770009: step: 1012/526, loss: 0.07159347832202911 2023-01-21 13:43:46.904985: step: 1016/526, loss: 0.0038324357010424137 2023-01-21 13:43:48.017057: step: 1020/526, loss: 0.0001354217529296875 2023-01-21 13:43:49.151048: step: 1024/526, loss: 4.6396256948355585e-05 2023-01-21 13:43:50.318778: step: 1028/526, loss: 0.0808926522731781 2023-01-21 13:43:51.449355: step: 1032/526, loss: 0.00011205673217773438 2023-01-21 13:43:52.598629: step: 1036/526, loss: 0.0029749395325779915 2023-01-21 13:43:53.718485: step: 1040/526, loss: 0.0007938385242596269 2023-01-21 13:43:54.859473: step: 1044/526, loss: 0.0007907628896646202 2023-01-21 13:43:55.976813: step: 1048/526, loss: 0.011900520883500576 2023-01-21 13:43:57.140094: step: 1052/526, loss: 0.000537872314453125 2023-01-21 13:43:58.254657: step: 1056/526, loss: 0.0026862144004553556 2023-01-21 13:43:59.379995: step: 1060/526, loss: 7.019042823230848e-05 2023-01-21 13:44:00.524238: step: 1064/526, loss: 6.67572021484375e-06 2023-01-21 13:44:01.676254: step: 1068/526, loss: 0.0005766868707723916 2023-01-21 13:44:02.797079: step: 1072/526, loss: 0.0001841068296926096 2023-01-21 13:44:03.903207: step: 1076/526, loss: 7.43865966796875e-05 2023-01-21 13:44:05.030174: step: 1080/526, loss: 0.012058544903993607 2023-01-21 13:44:06.179688: step: 1084/526, loss: 0.0003757476806640625 2023-01-21 13:44:07.300423: step: 1088/526, loss: 4.444122168933973e-05 2023-01-21 13:44:08.463055: step: 1092/526, loss: 0.0003086090146098286 2023-01-21 13:44:09.623406: step: 1096/526, loss: 0.0008723258506506681 2023-01-21 13:44:10.756382: step: 1100/526, loss: -9.96589551505167e-06 2023-01-21 13:44:11.876613: step: 1104/526, loss: 0.001121616456657648 2023-01-21 13:44:13.047792: step: 1108/526, loss: 0.0035851479042321444 2023-01-21 13:44:14.176649: step: 1112/526, loss: 0.00014991759962867945 2023-01-21 13:44:15.326810: step: 1116/526, loss: 0.009939765557646751 2023-01-21 13:44:16.433515: step: 1120/526, loss: 1.220703143189894e-05 2023-01-21 13:44:17.567898: step: 1124/526, loss: 2.8800963264075108e-05 2023-01-21 13:44:18.702072: step: 1128/526, loss: 0.050063323229551315 2023-01-21 13:44:19.830132: step: 1132/526, loss: 0.06774502247571945 2023-01-21 13:44:20.959816: step: 1136/526, loss: 0.013505172915756702 2023-01-21 13:44:22.081120: step: 1140/526, loss: 0.000812149082776159 2023-01-21 13:44:23.228114: step: 1144/526, loss: 0.005931663792580366 2023-01-21 13:44:24.375203: step: 1148/526, loss: 0.0008241653558798134 2023-01-21 13:44:25.502158: step: 1152/526, loss: 8.94546537892893e-05 2023-01-21 13:44:26.661005: step: 1156/526, loss: 0.0008180618169717491 2023-01-21 13:44:27.792655: step: 1160/526, loss: 7.839203317416832e-05 2023-01-21 13:44:28.939391: step: 1164/526, loss: 0.01895885355770588 2023-01-21 13:44:30.090593: step: 1168/526, loss: 0.0021076202392578125 2023-01-21 13:44:31.262620: step: 1172/526, loss: 0.006271648686379194 2023-01-21 13:44:32.382666: step: 1176/526, loss: 0.0004314422549214214 2023-01-21 13:44:33.518897: step: 1180/526, loss: 5.273819260764867e-05 2023-01-21 13:44:34.648753: step: 1184/526, loss: 0.0005692005506716669 2023-01-21 13:44:35.806513: step: 1188/526, loss: 0.009467697702348232 2023-01-21 13:44:36.923173: step: 1192/526, loss: 0.019235610961914062 2023-01-21 13:44:38.069084: step: 1196/526, loss: 0.00020694732666015625 2023-01-21 13:44:39.194788: step: 1200/526, loss: 9.34600848268019e-06 2023-01-21 13:44:40.312632: step: 1204/526, loss: 1.487731969973538e-05 2023-01-21 13:44:41.448189: step: 1208/526, loss: 0.017633533105254173 2023-01-21 13:44:42.592410: step: 1212/526, loss: 6.256104097701609e-05 2023-01-21 13:44:43.699370: step: 1216/526, loss: 0.015104102902114391 2023-01-21 13:44:44.850428: step: 1220/526, loss: 0.017835520207881927 2023-01-21 13:44:45.994352: step: 1224/526, loss: 4.463195728021674e-05 2023-01-21 13:44:47.115799: step: 1228/526, loss: 0.0007609367021359503 2023-01-21 13:44:48.261050: step: 1232/526, loss: 0.00041375160799361765 2023-01-21 13:44:49.392771: step: 1236/526, loss: 0.00022687911405228078 2023-01-21 13:44:50.554673: step: 1240/526, loss: 0.00036234856816008687 2023-01-21 13:44:51.699566: step: 1244/526, loss: 0.00010929107520496473 2023-01-21 13:44:52.825162: step: 1248/526, loss: 0.013018418103456497 2023-01-21 13:44:53.963192: step: 1252/526, loss: 0.024806689471006393 2023-01-21 13:44:55.102205: step: 1256/526, loss: 2.1123885744600557e-05 2023-01-21 13:44:56.252162: step: 1260/526, loss: 7.901191565906629e-05 2023-01-21 13:44:57.358711: step: 1264/526, loss: 0.0036570073571056128 2023-01-21 13:44:58.475707: step: 1268/526, loss: 0.0015617371536791325 2023-01-21 13:44:59.619025: step: 1272/526, loss: 0.00046758653479628265 2023-01-21 13:45:00.788083: step: 1276/526, loss: 0.008194160647690296 2023-01-21 13:45:01.904672: step: 1280/526, loss: 0.016090966761112213 2023-01-21 13:45:03.028831: step: 1284/526, loss: 1.1444091796875e-05 2023-01-21 13:45:04.182326: step: 1288/526, loss: 0.03010845184326172 2023-01-21 13:45:05.351433: step: 1292/526, loss: 0.00021958351135253906 2023-01-21 13:45:06.472903: step: 1296/526, loss: 4.081726001459174e-05 2023-01-21 13:45:07.604640: step: 1300/526, loss: 0.09383416175842285 2023-01-21 13:45:08.741660: step: 1304/526, loss: 0.0034228325821459293 2023-01-21 13:45:09.847935: step: 1308/526, loss: 3.108978125965223e-05 2023-01-21 13:45:10.983336: step: 1312/526, loss: 0.0033784867264330387 2023-01-21 13:45:12.119202: step: 1316/526, loss: 0.033480264246463776 2023-01-21 13:45:13.239080: step: 1320/526, loss: 0.0035385130904614925 2023-01-21 13:45:14.374357: step: 1324/526, loss: 0.008394241333007812 2023-01-21 13:45:15.490566: step: 1328/526, loss: 0.001901817275211215 2023-01-21 13:45:16.675448: step: 1332/526, loss: 0.00017967224994208664 2023-01-21 13:45:17.811794: step: 1336/526, loss: 0.0004960060468874872 2023-01-21 13:45:18.956416: step: 1340/526, loss: 0.015882110223174095 2023-01-21 13:45:20.104127: step: 1344/526, loss: 0.00036573412944562733 2023-01-21 13:45:21.220905: step: 1348/526, loss: 0.0001012802094919607 2023-01-21 13:45:22.350468: step: 1352/526, loss: 0.013698958791792393 2023-01-21 13:45:23.459015: step: 1356/526, loss: 0.001324558281339705 2023-01-21 13:45:24.598881: step: 1360/526, loss: 0.00024776457576081157 2023-01-21 13:45:25.718529: step: 1364/526, loss: 2.670288040462765e-06 2023-01-21 13:45:26.844026: step: 1368/526, loss: 1.4400482541532256e-05 2023-01-21 13:45:27.951402: step: 1372/526, loss: 0.0011069298489019275 2023-01-21 13:45:29.073468: step: 1376/526, loss: 0.007462263107299805 2023-01-21 13:45:30.190433: step: 1380/526, loss: 0.05401735380291939 2023-01-21 13:45:31.304467: step: 1384/526, loss: 4.3582916987361386e-05 2023-01-21 13:45:32.420043: step: 1388/526, loss: 2.250671423098538e-05 2023-01-21 13:45:33.573553: step: 1392/526, loss: 0.01971612125635147 2023-01-21 13:45:34.697636: step: 1396/526, loss: 0.0009097099537029862 2023-01-21 13:45:35.820743: step: 1400/526, loss: 2.28881845032447e-06 2023-01-21 13:45:37.010855: step: 1404/526, loss: 0.013045978732407093 2023-01-21 13:45:38.113661: step: 1408/526, loss: 0.0027154923882335424 2023-01-21 13:45:39.278966: step: 1412/526, loss: 1.3351454981602728e-06 2023-01-21 13:45:40.401803: step: 1416/526, loss: 0.005413246341049671 2023-01-21 13:45:41.535873: step: 1420/526, loss: 0.005953311920166016 2023-01-21 13:45:42.661996: step: 1424/526, loss: 0.013836859725415707 2023-01-21 13:45:43.821299: step: 1428/526, loss: 0.031794264912605286 2023-01-21 13:45:44.978646: step: 1432/526, loss: 1.0967254638671875e-05 2023-01-21 13:45:46.093843: step: 1436/526, loss: 0.000247764604864642 2023-01-21 13:45:47.239448: step: 1440/526, loss: 0.00011882782564498484 2023-01-21 13:45:48.351399: step: 1444/526, loss: 5.5313107623078395e-06 2023-01-21 13:45:49.505398: step: 1448/526, loss: 0.00026893618633039296 2023-01-21 13:45:50.639367: step: 1452/526, loss: 0.0003461838059592992 2023-01-21 13:45:51.776167: step: 1456/526, loss: 0.0025379182770848274 2023-01-21 13:45:52.880610: step: 1460/526, loss: 2.5367738999193534e-05 2023-01-21 13:45:54.034831: step: 1464/526, loss: 0.05057534947991371 2023-01-21 13:45:55.223013: step: 1468/526, loss: 0.00010166168794967234 2023-01-21 13:45:56.353962: step: 1472/526, loss: 5.8841706049861386e-05 2023-01-21 13:45:57.506237: step: 1476/526, loss: 0.025016499683260918 2023-01-21 13:45:58.630020: step: 1480/526, loss: 0.007387256249785423 2023-01-21 13:45:59.767800: step: 1484/526, loss: 0.0024724958930164576 2023-01-21 13:46:00.929999: step: 1488/526, loss: 0.0008812129963189363 2023-01-21 13:46:02.050846: step: 1492/526, loss: 0.0045608519576489925 2023-01-21 13:46:03.184465: step: 1496/526, loss: 0.003621578449383378 2023-01-21 13:46:04.334664: step: 1500/526, loss: 5.578995114774443e-06 2023-01-21 13:46:05.476514: step: 1504/526, loss: 0.012771034613251686 2023-01-21 13:46:06.578272: step: 1508/526, loss: 0.00037803652230650187 2023-01-21 13:46:07.704019: step: 1512/526, loss: 7.882118370616809e-05 2023-01-21 13:46:08.831447: step: 1516/526, loss: 0.0038187026511877775 2023-01-21 13:46:09.954941: step: 1520/526, loss: 0.008404826745390892 2023-01-21 13:46:11.083235: step: 1524/526, loss: 0.00011577606346691027 2023-01-21 13:46:12.247550: step: 1528/526, loss: 2.059936559817288e-05 2023-01-21 13:46:13.374328: step: 1532/526, loss: 0.0029659748543053865 2023-01-21 13:46:14.496751: step: 1536/526, loss: 0.019263554364442825 2023-01-21 13:46:15.595819: step: 1540/526, loss: 9.536744727256519e-08 2023-01-21 13:46:16.725577: step: 1544/526, loss: 3.1948088690114673e-06 2023-01-21 13:46:17.854188: step: 1548/526, loss: 0.003868675325065851 2023-01-21 13:46:19.019638: step: 1552/526, loss: 1.773834264895413e-05 2023-01-21 13:46:20.150135: step: 1556/526, loss: 0.024718379601836205 2023-01-21 13:46:21.252531: step: 1560/526, loss: 0.023740006610751152 2023-01-21 13:46:22.381664: step: 1564/526, loss: 0.00016088486881926656 2023-01-21 13:46:23.521082: step: 1568/526, loss: 0.012238502502441406 2023-01-21 13:46:24.659459: step: 1572/526, loss: 0.006568718235939741 2023-01-21 13:46:25.799580: step: 1576/526, loss: 0.026308823376893997 2023-01-21 13:46:26.944030: step: 1580/526, loss: 0.0013119220966473222 2023-01-21 13:46:28.072011: step: 1584/526, loss: 0.005526161286979914 2023-01-21 13:46:29.239097: step: 1588/526, loss: 0.006288337986916304 2023-01-21 13:46:30.385088: step: 1592/526, loss: 0.004463767632842064 2023-01-21 13:46:31.518920: step: 1596/526, loss: 3.5953522456111386e-05 2023-01-21 13:46:32.642913: step: 1600/526, loss: 5.3977968491381034e-05 2023-01-21 13:46:33.810752: step: 1604/526, loss: 0.0005950927734375 2023-01-21 13:46:34.954916: step: 1608/526, loss: 0.0020165443420410156 2023-01-21 13:46:36.104633: step: 1612/526, loss: 0.0012900352012366056 2023-01-21 13:46:37.235649: step: 1616/526, loss: 0.00015859604172874242 2023-01-21 13:46:38.369031: step: 1620/526, loss: 0.0005213737604208291 2023-01-21 13:46:39.509216: step: 1624/526, loss: 0.0107460031285882 2023-01-21 13:46:40.668707: step: 1628/526, loss: 0.000501442002132535 2023-01-21 13:46:41.843035: step: 1632/526, loss: 0.00048006774159148335 2023-01-21 13:46:42.989550: step: 1636/526, loss: 0.00040540695772506297 2023-01-21 13:46:44.125410: step: 1640/526, loss: 0.0008028030279092491 2023-01-21 13:46:45.241895: step: 1644/526, loss: 1.1157990229548886e-05 2023-01-21 13:46:46.383519: step: 1648/526, loss: 0.0015939592849463224 2023-01-21 13:46:47.553766: step: 1652/526, loss: 0.00019164086552336812 2023-01-21 13:46:48.674977: step: 1656/526, loss: 0.005586910527199507 2023-01-21 13:46:49.804361: step: 1660/526, loss: 2.021789623540826e-05 2023-01-21 13:46:50.964688: step: 1664/526, loss: 0.00043125153752043843 2023-01-21 13:46:52.103384: step: 1668/526, loss: 0.03317022696137428 2023-01-21 13:46:53.205865: step: 1672/526, loss: 0.0001363277406198904 2023-01-21 13:46:54.328616: step: 1676/526, loss: 0.0033693313598632812 2023-01-21 13:46:55.519824: step: 1680/526, loss: 0.000546312308870256 2023-01-21 13:46:56.647688: step: 1684/526, loss: 0.00048313141451217234 2023-01-21 13:46:57.791725: step: 1688/526, loss: 4.95910626341356e-06 2023-01-21 13:46:58.911262: step: 1692/526, loss: 0.0011812209850177169 2023-01-21 13:47:00.053897: step: 1696/526, loss: 2.6035309929284267e-05 2023-01-21 13:47:01.164186: step: 1700/526, loss: 3.604888843256049e-05 2023-01-21 13:47:02.310182: step: 1704/526, loss: 0.007018089760094881 2023-01-21 13:47:03.471857: step: 1708/526, loss: 0.0001440525084035471 2023-01-21 13:47:04.619186: step: 1712/526, loss: 0.003296089358627796 2023-01-21 13:47:05.747251: step: 1716/526, loss: 7.534026735811494e-06 2023-01-21 13:47:06.875199: step: 1720/526, loss: 1.6498564946232364e-05 2023-01-21 13:47:08.023801: step: 1724/526, loss: 0.0001220703125 2023-01-21 13:47:09.176357: step: 1728/526, loss: 4.634857032215223e-05 2023-01-21 13:47:10.330810: step: 1732/526, loss: 2.1934511096333154e-06 2023-01-21 13:47:11.478019: step: 1736/526, loss: 0.0006587982061319053 2023-01-21 13:47:12.599664: step: 1740/526, loss: 0.0007326126797124743 2023-01-21 13:47:13.752934: step: 1744/526, loss: 0.002924632979556918 2023-01-21 13:47:14.896455: step: 1748/526, loss: 1.2779236385540571e-05 2023-01-21 13:47:16.034052: step: 1752/526, loss: 1.0013580322265625e-05 2023-01-21 13:47:17.158578: step: 1756/526, loss: 4.4441225327318534e-05 2023-01-21 13:47:18.316220: step: 1760/526, loss: 0.0006335258949548006 2023-01-21 13:47:19.480623: step: 1764/526, loss: 2.079009937006049e-05 2023-01-21 13:47:20.621834: step: 1768/526, loss: 0.020188046619296074 2023-01-21 13:47:21.770113: step: 1772/526, loss: 6.923675391590223e-05 2023-01-21 13:47:22.898943: step: 1776/526, loss: 0.03169002756476402 2023-01-21 13:47:24.032700: step: 1780/526, loss: 3.280639793956652e-05 2023-01-21 13:47:25.162947: step: 1784/526, loss: 2.956390289909905e-06 2023-01-21 13:47:26.306753: step: 1788/526, loss: 0.0002522468566894531 2023-01-21 13:47:27.524814: step: 1792/526, loss: 0.00489730853587389 2023-01-21 13:47:28.720671: step: 1796/526, loss: 0.00023021697415970266 2023-01-21 13:47:29.868933: step: 1800/526, loss: 0.030421828851103783 2023-01-21 13:47:31.016657: step: 1804/526, loss: 0.00046758653479628265 2023-01-21 13:47:32.193443: step: 1808/526, loss: 1.3732910701946821e-05 2023-01-21 13:47:33.352847: step: 1812/526, loss: 0.15819358825683594 2023-01-21 13:47:34.524572: step: 1816/526, loss: 0.0006811618804931641 2023-01-21 13:47:35.665527: step: 1820/526, loss: 0.012120437808334827 2023-01-21 13:47:36.790760: step: 1824/526, loss: 0.0009313583723269403 2023-01-21 13:47:37.965631: step: 1828/526, loss: 0.0070618633180856705 2023-01-21 13:47:39.150123: step: 1832/526, loss: 0.00038537976797670126 2023-01-21 13:47:40.317437: step: 1836/526, loss: 0.0002094268857035786 2023-01-21 13:47:41.434126: step: 1840/526, loss: 0.015607452020049095 2023-01-21 13:47:42.578302: step: 1844/526, loss: 5.855560448253527e-05 2023-01-21 13:47:43.718550: step: 1848/526, loss: 2.8514863515738398e-05 2023-01-21 13:47:44.855312: step: 1852/526, loss: 0.06976642459630966 2023-01-21 13:47:45.974994: step: 1856/526, loss: 0.01127328909933567 2023-01-21 13:47:47.093192: step: 1860/526, loss: 7.43865984986769e-06 2023-01-21 13:47:48.274211: step: 1864/526, loss: 0.00010213851783191785 2023-01-21 13:47:49.426673: step: 1868/526, loss: 0.0011112213833257556 2023-01-21 13:47:50.568818: step: 1872/526, loss: 7.62939453125e-06 2023-01-21 13:47:51.703057: step: 1876/526, loss: -1.6689299400240998e-06 2023-01-21 13:47:52.839480: step: 1880/526, loss: 0.005048847291618586 2023-01-21 13:47:53.969095: step: 1884/526, loss: 0.0048728943802416325 2023-01-21 13:47:55.110693: step: 1888/526, loss: 0.01232070941478014 2023-01-21 13:47:56.284023: step: 1892/526, loss: 0.0005630493396893144 2023-01-21 13:47:57.415760: step: 1896/526, loss: 0.0005395889165811241 2023-01-21 13:47:58.586976: step: 1900/526, loss: 0.01865386962890625 2023-01-21 13:47:59.710547: step: 1904/526, loss: 0.006522941868752241 2023-01-21 13:48:00.854851: step: 1908/526, loss: 8.726119631319307e-06 2023-01-21 13:48:01.994600: step: 1912/526, loss: 6.27517729299143e-05 2023-01-21 13:48:03.177663: step: 1916/526, loss: 8.583069575252011e-05 2023-01-21 13:48:04.342785: step: 1920/526, loss: 1.9550323486328125e-05 2023-01-21 13:48:05.481516: step: 1924/526, loss: 0.0019906999077647924 2023-01-21 13:48:06.611660: step: 1928/526, loss: 0.0004419327015057206 2023-01-21 13:48:07.760599: step: 1932/526, loss: 0.11033286899328232 2023-01-21 13:48:08.881688: step: 1936/526, loss: 0.0006152152782306075 2023-01-21 13:48:10.016117: step: 1940/526, loss: 0.013238906860351562 2023-01-21 13:48:11.183042: step: 1944/526, loss: 0.022258948534727097 2023-01-21 13:48:12.330608: step: 1948/526, loss: 0.0025447846855968237 2023-01-21 13:48:13.484923: step: 1952/526, loss: 0.00025153160095214844 2023-01-21 13:48:14.646203: step: 1956/526, loss: 9.765625145519152e-05 2023-01-21 13:48:15.794178: step: 1960/526, loss: 0.02020130306482315 2023-01-21 13:48:16.927131: step: 1964/526, loss: 0.010028076358139515 2023-01-21 13:48:18.034199: step: 1968/526, loss: 0.016941452398896217 2023-01-21 13:48:19.144283: step: 1972/526, loss: 0.011068678461015224 2023-01-21 13:48:20.281478: step: 1976/526, loss: 0.00063071254407987 2023-01-21 13:48:21.446501: step: 1980/526, loss: 0.0012928008800372481 2023-01-21 13:48:22.569224: step: 1984/526, loss: 1.3637542906508315e-05 2023-01-21 13:48:23.703957: step: 1988/526, loss: 0.000211620339541696 2023-01-21 13:48:24.839996: step: 1992/526, loss: 0.00566787738353014 2023-01-21 13:48:25.989031: step: 1996/526, loss: 0.0005762100336141884 2023-01-21 13:48:27.149611: step: 2000/526, loss: 0.034627534449100494 2023-01-21 13:48:28.299340: step: 2004/526, loss: 0.007682514376938343 2023-01-21 13:48:29.417903: step: 2008/526, loss: 0.009882116690278053 2023-01-21 13:48:30.555499: step: 2012/526, loss: 9.441375368623994e-06 2023-01-21 13:48:31.680706: step: 2016/526, loss: 0.008030558004975319 2023-01-21 13:48:32.807230: step: 2020/526, loss: 0.0023218155838549137 2023-01-21 13:48:33.918573: step: 2024/526, loss: 0.0001678466796875 2023-01-21 13:48:35.070429: step: 2028/526, loss: 0.0009698867797851562 2023-01-21 13:48:36.208755: step: 2032/526, loss: 4.8065183364087716e-05 2023-01-21 13:48:37.337195: step: 2036/526, loss: 2.6416779292048886e-05 2023-01-21 13:48:38.469347: step: 2040/526, loss: 0.0009784698486328125 2023-01-21 13:48:39.646477: step: 2044/526, loss: 8.177757263183594e-05 2023-01-21 13:48:40.813962: step: 2048/526, loss: 0.002227401826530695 2023-01-21 13:48:41.946739: step: 2052/526, loss: 0.04100246727466583 2023-01-21 13:48:43.133454: step: 2056/526, loss: 0.011110449209809303 2023-01-21 13:48:44.317503: step: 2060/526, loss: 0.010821151547133923 2023-01-21 13:48:45.451462: step: 2064/526, loss: 0.0009624481317587197 2023-01-21 13:48:46.606875: step: 2068/526, loss: 0.0005439758533611894 2023-01-21 13:48:47.733477: step: 2072/526, loss: 1.029968279908644e-05 2023-01-21 13:48:48.914699: step: 2076/526, loss: 5.2642819355241954e-05 2023-01-21 13:48:50.087436: step: 2080/526, loss: 5.245209194981726e-06 2023-01-21 13:48:51.207466: step: 2084/526, loss: 1.6880036127986386e-05 2023-01-21 13:48:52.367240: step: 2088/526, loss: 0.00025663376436568797 2023-01-21 13:48:53.503399: step: 2092/526, loss: 0.0034324645530432463 2023-01-21 13:48:54.618081: step: 2096/526, loss: 1.163482647825731e-05 2023-01-21 13:48:55.774592: step: 2100/526, loss: 7.343292236328125e-05 2023-01-21 13:48:56.967894: step: 2104/526, loss: -4.95910626341356e-06 ================================================== Loss: 0.010 -------------------- Dev: {'event': {'p': 0.615795090715048, 'r': 0.7683089214380826, 'f1': 0.683649289099526}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Test: {'event': {'p': 0.6265366680796948, 'r': 0.7713987473903967, 'f1': 0.6914619883040936}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Chinese: {'event': {'p': 0.6282051282051282, 'r': 0.9074074074074074, 'f1': 0.7424242424242423}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Korean: {'event': {'p': 0.6666666666666666, 'r': 0.4444444444444444, 'f1': 0.5333333333333333}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} Russian: {'event': {'p': 0.4772727272727273, 'r': 0.5833333333333334, 'f1': 0.5250000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 28} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} ****************************** Epoch: 29 command: python train.py --model_name trg --xlmr_model_name xlm-roberta-large --batch_size 16 --xlmr_learning_rate 2e-5 --event_hidden_num 450 --accumulate_step 4 --max_epoch 30 --p1_data_weight 0.1 --learning_rate 6e-4 2023-01-21 13:49:40.818195: step: 4/526, loss: 1.811981201171875e-05 2023-01-21 13:49:41.954369: step: 8/526, loss: 0.0001941680966410786 2023-01-21 13:49:43.089509: step: 12/526, loss: 0.0012712478637695312 2023-01-21 13:49:44.206180: step: 16/526, loss: 0.0019008637173101306 2023-01-21 13:49:45.369807: step: 20/526, loss: 0.0001235961972270161 2023-01-21 13:49:46.519098: step: 24/526, loss: 0.002497005509212613 2023-01-21 13:49:47.631118: step: 28/526, loss: 0.0007276535616256297 2023-01-21 13:49:48.762065: step: 32/526, loss: -1.9073486328125e-06 2023-01-21 13:49:49.879458: step: 36/526, loss: 0.0001348495570709929 2023-01-21 13:49:51.018370: step: 40/526, loss: 0.00010986328561557457 2023-01-21 13:49:52.173571: step: 44/526, loss: 0.0010242462158203125 2023-01-21 13:49:53.287041: step: 48/526, loss: 0.00030574799166060984 2023-01-21 13:49:54.429062: step: 52/526, loss: 8.869171142578125e-05 2023-01-21 13:49:55.555057: step: 56/526, loss: 0.0015371321933344007 2023-01-21 13:49:56.702498: step: 60/526, loss: 0.00020799637422896922 2023-01-21 13:49:57.842970: step: 64/526, loss: 3.6239625842426904e-06 2023-01-21 13:49:58.956078: step: 68/526, loss: 1.2397766795402276e-06 2023-01-21 13:50:00.083350: step: 72/526, loss: 0.27656590938568115 2023-01-21 13:50:01.228048: step: 76/526, loss: 0.0013900757767260075 2023-01-21 13:50:02.368829: step: 80/526, loss: 0.002552032470703125 2023-01-21 13:50:03.457055: step: 84/526, loss: 0.0160719882696867 2023-01-21 13:50:04.576359: step: 88/526, loss: 0.00021638871112372726 2023-01-21 13:50:05.709149: step: 92/526, loss: 0.0005153656238690019 2023-01-21 13:50:06.874465: step: 96/526, loss: 9.469986252952367e-05 2023-01-21 13:50:08.042017: step: 100/526, loss: 3.814697265625e-06 2023-01-21 13:50:09.197369: step: 104/526, loss: 0.004469108767807484 2023-01-21 13:50:10.351662: step: 108/526, loss: 0.019799232482910156 2023-01-21 13:50:11.495386: step: 112/526, loss: 7.896423630882055e-05 2023-01-21 13:50:12.650542: step: 116/526, loss: 0.0029705048073083162 2023-01-21 13:50:13.780535: step: 120/526, loss: 0.005303001496940851 2023-01-21 13:50:14.951849: step: 124/526, loss: 0.055001452565193176 2023-01-21 13:50:16.090361: step: 128/526, loss: 0.0005161285516805947 2023-01-21 13:50:17.205166: step: 132/526, loss: 0.012682723812758923 2023-01-21 13:50:18.326911: step: 136/526, loss: 0.0180924404412508 2023-01-21 13:50:19.432142: step: 140/526, loss: 0.01165628433227539 2023-01-21 13:50:20.567042: step: 144/526, loss: 0.021371990442276 2023-01-21 13:50:21.699871: step: 148/526, loss: 0.00039253238355740905 2023-01-21 13:50:22.840062: step: 152/526, loss: 0.0012218474876135588 2023-01-21 13:50:23.962360: step: 156/526, loss: 0.007853222079575062 2023-01-21 13:50:25.153876: step: 160/526, loss: 0.0006039619911462069 2023-01-21 13:50:26.308397: step: 164/526, loss: 0.0007616042857989669 2023-01-21 13:50:27.423704: step: 168/526, loss: 0.006319236941635609 2023-01-21 13:50:28.567067: step: 172/526, loss: 0.00021772386389784515 2023-01-21 13:50:29.711644: step: 176/526, loss: 0.008054065518081188 2023-01-21 13:50:30.849201: step: 180/526, loss: 0.017675232142210007 2023-01-21 13:50:31.953699: step: 184/526, loss: 6.828307959949598e-05 2023-01-21 13:50:33.101967: step: 188/526, loss: 2.1791458493680693e-05 2023-01-21 13:50:34.228133: step: 192/526, loss: 0.0004935264587402344 2023-01-21 13:50:35.369557: step: 196/526, loss: -3.14712519866589e-06 2023-01-21 13:50:36.507476: step: 200/526, loss: 0.001275444170460105 2023-01-21 13:50:37.627981: step: 204/526, loss: 3.957748504035408e-06 2023-01-21 13:50:38.774367: step: 208/526, loss: 0.00024900436983443797 2023-01-21 13:50:39.925664: step: 212/526, loss: 0.005500125698745251 2023-01-21 13:50:41.062210: step: 216/526, loss: 0.011472702026367188 2023-01-21 13:50:42.187115: step: 220/526, loss: 0.00034508705721236765 2023-01-21 13:50:43.357969: step: 224/526, loss: 0.008842945098876953 2023-01-21 13:50:44.475125: step: 228/526, loss: 0.0005613326793536544 2023-01-21 13:50:45.583261: step: 232/526, loss: 0.002803516574203968 2023-01-21 13:50:46.717450: step: 236/526, loss: 0.0018253326416015625 2023-01-21 13:50:47.873028: step: 240/526, loss: 0.04392433166503906 2023-01-21 13:50:49.023968: step: 244/526, loss: 7.62939453125e-06 2023-01-21 13:50:50.140742: step: 248/526, loss: 1.0919571650447324e-05 2023-01-21 13:50:51.270447: step: 252/526, loss: 0.006730079650878906 2023-01-21 13:50:52.432535: step: 256/526, loss: 7.247925168485381e-06 2023-01-21 13:50:53.576579: step: 260/526, loss: 0.022701837122440338 2023-01-21 13:50:54.719666: step: 264/526, loss: 0.013193607330322266 2023-01-21 13:50:55.859428: step: 268/526, loss: 0.0010380744934082031 2023-01-21 13:50:56.999092: step: 272/526, loss: 0.00043334963265806437 2023-01-21 13:50:58.155219: step: 276/526, loss: 0.251460462808609 2023-01-21 13:50:59.286072: step: 280/526, loss: 0.03471508249640465 2023-01-21 13:51:00.420167: step: 284/526, loss: 0.00291862478479743 2023-01-21 13:51:01.538073: step: 288/526, loss: 2.965927342302166e-05 2023-01-21 13:51:02.677826: step: 292/526, loss: 0.0027446746826171875 2023-01-21 13:51:03.778104: step: 296/526, loss: -2.288818359375e-05 2023-01-21 13:51:04.918822: step: 300/526, loss: 0.0005229950184002519 2023-01-21 13:51:06.056907: step: 304/526, loss: 0.0008993148803710938 2023-01-21 13:51:07.201223: step: 308/526, loss: 0.37060850858688354 2023-01-21 13:51:08.356207: step: 312/526, loss: 0.008023500442504883 2023-01-21 13:51:09.469158: step: 316/526, loss: 0.001656055566854775 2023-01-21 13:51:10.615940: step: 320/526, loss: 5.2261355449445546e-05 2023-01-21 13:51:11.738411: step: 324/526, loss: 3.662109520519152e-05 2023-01-21 13:51:12.919079: step: 328/526, loss: 0.0005950451013632119 2023-01-21 13:51:14.065220: step: 332/526, loss: 0.5303598046302795 2023-01-21 13:51:15.207677: step: 336/526, loss: 7.61985793360509e-05 2023-01-21 13:51:16.326261: step: 340/526, loss: 0.012025165371596813 2023-01-21 13:51:17.445853: step: 344/526, loss: 0.00014505386934615672 2023-01-21 13:51:18.569563: step: 348/526, loss: 8.254051499534398e-05 2023-01-21 13:51:19.712994: step: 352/526, loss: 0.0009945393539965153 2023-01-21 13:51:20.846982: step: 356/526, loss: 2.8133392788731726e-06 2023-01-21 13:51:22.011628: step: 360/526, loss: 0.0002055168297374621 2023-01-21 13:51:23.155788: step: 364/526, loss: 1.26004224512144e-05 2023-01-21 13:51:24.287765: step: 368/526, loss: 0.0018686294788494706 2023-01-21 13:51:25.424783: step: 372/526, loss: 0.01757078245282173 2023-01-21 13:51:26.588486: step: 376/526, loss: 0.002975654788315296 2023-01-21 13:51:27.770327: step: 380/526, loss: 0.0005886078579351306 2023-01-21 13:51:28.917230: step: 384/526, loss: 4.234314110362902e-05 2023-01-21 13:51:30.064349: step: 388/526, loss: 0.00012245179095771164 2023-01-21 13:51:31.208377: step: 392/526, loss: 0.0032838822808116674 2023-01-21 13:51:32.342313: step: 396/526, loss: 0.0001922607480082661 2023-01-21 13:51:33.469481: step: 400/526, loss: 0.0005361557123251259 2023-01-21 13:51:34.598256: step: 404/526, loss: 4.4441225327318534e-05 2023-01-21 13:51:35.738143: step: 408/526, loss: 0.00023374558077193797 2023-01-21 13:51:36.852911: step: 412/526, loss: 0.0006167411920614541 2023-01-21 13:51:37.974244: step: 416/526, loss: 1.7881393432617188e-05 2023-01-21 13:51:39.095613: step: 420/526, loss: 2.2737367544323206e-13 2023-01-21 13:51:40.219903: step: 424/526, loss: 1.049041748046875e-05 2023-01-21 13:51:41.381580: step: 428/526, loss: -4.0054324017546605e-06 2023-01-21 13:51:42.517270: step: 432/526, loss: 0.0031707764137536287 2023-01-21 13:51:43.651241: step: 436/526, loss: 3.0136106943245977e-05 2023-01-21 13:51:44.808533: step: 440/526, loss: 8.630752745375503e-06 2023-01-21 13:51:45.956432: step: 444/526, loss: 0.00275421142578125 2023-01-21 13:51:47.126312: step: 448/526, loss: 0.031766414642333984 2023-01-21 13:51:48.245451: step: 452/526, loss: 6.065368506824598e-05 2023-01-21 13:51:49.389559: step: 456/526, loss: 2.2697449821862392e-05 2023-01-21 13:51:50.545136: step: 460/526, loss: 0.001425933907739818 2023-01-21 13:51:51.652477: step: 464/526, loss: 0.0022575377952307463 2023-01-21 13:51:52.830048: step: 468/526, loss: 6.4849853515625e-05 2023-01-21 13:51:53.971078: step: 472/526, loss: 0.0005176544073037803 2023-01-21 13:51:55.087363: step: 476/526, loss: 0.014089202508330345 2023-01-21 13:51:56.240013: step: 480/526, loss: 0.000640869140625 2023-01-21 13:51:57.409388: step: 484/526, loss: 0.00016174315533135086 2023-01-21 13:51:58.586182: step: 488/526, loss: 2.1648407710017636e-05 2023-01-21 13:51:59.725340: step: 492/526, loss: 3.0326844353112392e-05 2023-01-21 13:52:00.882328: step: 496/526, loss: 7.82012921263231e-06 2023-01-21 13:52:01.993840: step: 500/526, loss: 0.00164794921875 2023-01-21 13:52:03.127591: step: 504/526, loss: 4.138946678722277e-05 2023-01-21 13:52:04.230388: step: 508/526, loss: 0.003694486804306507 2023-01-21 13:52:05.347271: step: 512/526, loss: 0.03291931003332138 2023-01-21 13:52:06.498020: step: 516/526, loss: 3.089904930675402e-05 2023-01-21 13:52:07.636669: step: 520/526, loss: 0.00025177001953125 2023-01-21 13:52:08.771745: step: 524/526, loss: 0.0006361007690429688 2023-01-21 13:52:09.899483: step: 528/526, loss: 0.00387153634801507 2023-01-21 13:52:11.068046: step: 532/526, loss: 4.596710277837701e-05 2023-01-21 13:52:12.209612: step: 536/526, loss: 0.00012731552124023438 2023-01-21 13:52:13.324219: step: 540/526, loss: 0.005178546998649836 2023-01-21 13:52:14.463694: step: 544/526, loss: 4.730224463855848e-05 2023-01-21 13:52:15.582076: step: 548/526, loss: 0.04731016233563423 2023-01-21 13:52:16.714696: step: 552/526, loss: 6.12258882028982e-05 2023-01-21 13:52:17.828359: step: 556/526, loss: 8.821487426757812e-05 2023-01-21 13:52:18.991444: step: 560/526, loss: 0.008287430740892887 2023-01-21 13:52:20.139945: step: 564/526, loss: 6.580352874152595e-06 2023-01-21 13:52:21.256745: step: 568/526, loss: -6.008148375258315e-06 2023-01-21 13:52:22.382574: step: 572/526, loss: 0.0027103424072265625 2023-01-21 13:52:23.493304: step: 576/526, loss: 0.0021161078475415707 2023-01-21 13:52:24.639929: step: 580/526, loss: 0.009405947290360928 2023-01-21 13:52:25.791118: step: 584/526, loss: 0.0003705501731019467 2023-01-21 13:52:26.924121: step: 588/526, loss: 0.327279657125473 2023-01-21 13:52:28.040049: step: 592/526, loss: 0.001109218574129045 2023-01-21 13:52:29.207666: step: 596/526, loss: 0.01038427371531725 2023-01-21 13:52:30.354222: step: 600/526, loss: 0.0002617835998535156 2023-01-21 13:52:31.520437: step: 604/526, loss: 0.021316051483154297 2023-01-21 13:52:32.675365: step: 608/526, loss: 1.6117097402457148e-05 2023-01-21 13:52:33.774947: step: 612/526, loss: 1.1157990229548886e-05 2023-01-21 13:52:34.913197: step: 616/526, loss: 0.0001625060976948589 2023-01-21 13:52:36.064163: step: 620/526, loss: 0.004167747683823109 2023-01-21 13:52:37.188719: step: 624/526, loss: 5.6934357417048886e-05 2023-01-21 13:52:38.339630: step: 628/526, loss: 0.003048515412956476 2023-01-21 13:52:39.474229: step: 632/526, loss: 6.4849853515625e-05 2023-01-21 13:52:40.625346: step: 636/526, loss: 4.37736525782384e-05 2023-01-21 13:52:41.779546: step: 640/526, loss: 3.910064606316155e-06 2023-01-21 13:52:42.943523: step: 644/526, loss: 0.0007059097406454384 2023-01-21 13:52:44.079599: step: 648/526, loss: 1.4209747860149946e-05 2023-01-21 13:52:45.214321: step: 652/526, loss: 2.0790101189049892e-05 2023-01-21 13:52:46.357904: step: 656/526, loss: -1.029968279908644e-05 2023-01-21 13:52:47.479024: step: 660/526, loss: 6.222724914550781e-05 2023-01-21 13:52:48.630530: step: 664/526, loss: 0.0003966331423725933 2023-01-21 13:52:49.746345: step: 668/526, loss: 0.0001813888520700857 2023-01-21 13:52:50.874217: step: 672/526, loss: 0.013968468643724918 2023-01-21 13:52:51.989422: step: 676/526, loss: 7.05719003235572e-06 2023-01-21 13:52:53.111777: step: 680/526, loss: 0.08435797691345215 2023-01-21 13:52:54.316948: step: 684/526, loss: 0.009826088324189186 2023-01-21 13:52:55.455365: step: 688/526, loss: 0.0001464843808207661 2023-01-21 13:52:56.611209: step: 692/526, loss: 0.003316784044727683 2023-01-21 13:52:57.754555: step: 696/526, loss: 0.0219466220587492 2023-01-21 13:52:58.887995: step: 700/526, loss: 0.0021963119506835938 2023-01-21 13:53:00.011463: step: 704/526, loss: 0.0028383254539221525 2023-01-21 13:53:01.218658: step: 708/526, loss: 0.00020546912855934352 2023-01-21 13:53:02.350119: step: 712/526, loss: 1.182556115963962e-05 2023-01-21 13:53:03.470651: step: 716/526, loss: 1.7261505490751006e-05 2023-01-21 13:53:04.619637: step: 720/526, loss: 0.0004368782101664692 2023-01-21 13:53:05.760293: step: 724/526, loss: 2.098083541568485e-06 2023-01-21 13:53:06.881687: step: 728/526, loss: 2.5749208361958154e-05 2023-01-21 13:53:08.005291: step: 732/526, loss: 0.0009614944574423134 2023-01-21 13:53:09.131143: step: 736/526, loss: 7.309913780773059e-05 2023-01-21 13:53:10.231853: step: 740/526, loss: 0.09430694580078125 2023-01-21 13:53:11.330186: step: 744/526, loss: 1.5258788153005298e-06 2023-01-21 13:53:12.462384: step: 748/526, loss: 4.0626528061693534e-05 2023-01-21 13:53:13.620053: step: 752/526, loss: 3.4141543437726796e-05 2023-01-21 13:53:14.774604: step: 756/526, loss: 0.0001087188720703125 2023-01-21 13:53:15.926550: step: 760/526, loss: 0.006926632020622492 2023-01-21 13:53:17.049179: step: 764/526, loss: 1.2874604180979077e-05 2023-01-21 13:53:18.199921: step: 768/526, loss: 4.9400332500226796e-05 2023-01-21 13:53:19.292326: step: 772/526, loss: 1.0538101378188003e-05 2023-01-21 13:53:20.406218: step: 776/526, loss: 0.005696487613022327 2023-01-21 13:53:21.521639: step: 780/526, loss: 0.00010628700692905113 2023-01-21 13:53:22.661631: step: 784/526, loss: -1.773834264895413e-05 2023-01-21 13:53:23.798818: step: 788/526, loss: 1.201629675051663e-05 2023-01-21 13:53:24.914819: step: 792/526, loss: 0.00011272430856479332 2023-01-21 13:53:26.046458: step: 796/526, loss: 1.201629675051663e-05 2023-01-21 13:53:27.162644: step: 800/526, loss: 0.0012041092850267887 2023-01-21 13:53:28.300162: step: 804/526, loss: 0.00011693238775478676 2023-01-21 13:53:29.444962: step: 808/526, loss: 1.831054760259576e-05 2023-01-21 13:53:30.582959: step: 812/526, loss: 3.547668529790826e-05 2023-01-21 13:53:31.704246: step: 816/526, loss: 0.47016143798828125 2023-01-21 13:53:32.828699: step: 820/526, loss: 2.1934511096333154e-06 2023-01-21 13:53:33.955551: step: 824/526, loss: 0.0002132415829692036 2023-01-21 13:53:35.099567: step: 828/526, loss: 0.010844612494111061 2023-01-21 13:53:36.200278: step: 832/526, loss: -3.433226083870977e-06 2023-01-21 13:53:37.332421: step: 836/526, loss: 0.00046825408935546875 2023-01-21 13:53:38.446809: step: 840/526, loss: 0.024965476244688034 2023-01-21 13:53:39.578080: step: 844/526, loss: 0.004096412565559149 2023-01-21 13:53:40.714803: step: 848/526, loss: 0.001560211181640625 2023-01-21 13:53:41.861848: step: 852/526, loss: -9.489059266343247e-06 2023-01-21 13:53:43.007302: step: 856/526, loss: 0.04185066372156143 2023-01-21 13:53:44.165142: step: 860/526, loss: 5.14984139954322e-06 2023-01-21 13:53:45.303316: step: 864/526, loss: 2.288818359375e-05 2023-01-21 13:53:46.491844: step: 868/526, loss: 3.8528440200025216e-05 2023-01-21 13:53:47.620803: step: 872/526, loss: 0.02396259270608425 2023-01-21 13:53:48.742626: step: 876/526, loss: 0.011084240861237049 2023-01-21 13:53:49.858898: step: 880/526, loss: 0.011517142876982689 2023-01-21 13:53:50.979673: step: 884/526, loss: 0.0014192580711096525 2023-01-21 13:53:52.114540: step: 888/526, loss: 0.0008420944795943797 2023-01-21 13:53:53.294687: step: 892/526, loss: 0.01850261725485325 2023-01-21 13:53:54.401359: step: 896/526, loss: 0.000964736973401159 2023-01-21 13:53:55.554644: step: 900/526, loss: 0.004015636630356312 2023-01-21 13:53:56.707848: step: 904/526, loss: 0.0005403519026003778 2023-01-21 13:53:57.831887: step: 908/526, loss: 0.0002155303955078125 2023-01-21 13:53:58.973008: step: 912/526, loss: 0.0023027420975267887 2023-01-21 13:54:00.132252: step: 916/526, loss: 0.011924362741410732 2023-01-21 13:54:01.281380: step: 920/526, loss: 4.625320798368193e-06 2023-01-21 13:54:02.401935: step: 924/526, loss: 0.021325349807739258 2023-01-21 13:54:03.544531: step: 928/526, loss: 0.10978536307811737 2023-01-21 13:54:04.693991: step: 932/526, loss: 0.0001068115234375 2023-01-21 13:54:05.810040: step: 936/526, loss: 0.0017185688484460115 2023-01-21 13:54:06.924341: step: 940/526, loss: 7.43865984986769e-06 2023-01-21 13:54:08.070831: step: 944/526, loss: 0.0019600866362452507 2023-01-21 13:54:09.204447: step: 948/526, loss: 2.918243444582913e-05 2023-01-21 13:54:10.353716: step: 952/526, loss: 9.922981553245336e-05 2023-01-21 13:54:11.510526: step: 956/526, loss: 1.506805438111769e-05 2023-01-21 13:54:12.636180: step: 960/526, loss: 5.817413693876006e-06 2023-01-21 13:54:13.748635: step: 964/526, loss: 2.307891918462701e-05 2023-01-21 13:54:14.881352: step: 968/526, loss: 0.01246728841215372 2023-01-21 13:54:16.009194: step: 972/526, loss: 0.0005553245428018272 2023-01-21 13:54:17.146549: step: 976/526, loss: 0.0017802715301513672 2023-01-21 13:54:18.284166: step: 980/526, loss: 1.487731969973538e-05 2023-01-21 13:54:19.415270: step: 984/526, loss: 0.001964569091796875 2023-01-21 13:54:20.582712: step: 988/526, loss: -5.9127810345671605e-06 2023-01-21 13:54:21.699384: step: 992/526, loss: 0.023551084101200104 2023-01-21 13:54:22.815543: step: 996/526, loss: 0.00039691926212981343 2023-01-21 13:54:23.960706: step: 1000/526, loss: 0.0010014533763751388 2023-01-21 13:54:25.102233: step: 1004/526, loss: 0.00030012131901457906 2023-01-21 13:54:26.230689: step: 1008/526, loss: 0.04886522516608238 2023-01-21 13:54:27.359676: step: 1012/526, loss: 0.004866885952651501 2023-01-21 13:54:28.494717: step: 1016/526, loss: 0.013799190521240234 2023-01-21 13:54:29.612893: step: 1020/526, loss: 0.0004741668817587197 2023-01-21 13:54:30.713929: step: 1024/526, loss: 5.817413693876006e-06 2023-01-21 13:54:31.840262: step: 1028/526, loss: 9.822845640883315e-06 2023-01-21 13:54:32.965170: step: 1032/526, loss: 0.0004566192510537803 2023-01-21 13:54:34.079732: step: 1036/526, loss: 3.385544005141128e-06 2023-01-21 13:54:35.206100: step: 1040/526, loss: 0.00024147034855559468 2023-01-21 13:54:36.346175: step: 1044/526, loss: 0.0007737159612588584 2023-01-21 13:54:37.468083: step: 1048/526, loss: 0.006083107087761164 2023-01-21 13:54:38.610131: step: 1052/526, loss: 0.016718007624149323 2023-01-21 13:54:39.738988: step: 1056/526, loss: 0.023111343383789062 2023-01-21 13:54:40.881807: step: 1060/526, loss: 3.2043459214037284e-05 2023-01-21 13:54:41.990424: step: 1064/526, loss: 4.00543194700731e-06 2023-01-21 13:54:43.124049: step: 1068/526, loss: 0.04320569336414337 2023-01-21 13:54:44.226288: step: 1072/526, loss: 1.430511474609375e-06 2023-01-21 13:54:45.350882: step: 1076/526, loss: 0.0413476936519146 2023-01-21 13:54:46.508778: step: 1080/526, loss: 0.0049527171067893505 2023-01-21 13:54:47.619720: step: 1084/526, loss: 0.00012702943058684468 2023-01-21 13:54:48.740066: step: 1088/526, loss: 0.005206442438066006 2023-01-21 13:54:49.884076: step: 1092/526, loss: 0.0024322508834302425 2023-01-21 13:54:50.994414: step: 1096/526, loss: 0.00022602081298828125 2023-01-21 13:54:52.126938: step: 1100/526, loss: 7.82012921263231e-06 2023-01-21 13:54:53.255782: step: 1104/526, loss: 0.0010413170093670487 2023-01-21 13:54:54.368513: step: 1108/526, loss: 0.0013490677811205387 2023-01-21 13:54:55.506440: step: 1112/526, loss: 0.00021238326735328883 2023-01-21 13:54:56.627590: step: 1116/526, loss: 7.781982276355848e-05 2023-01-21 13:54:57.787750: step: 1120/526, loss: 0.0038629532791674137 2023-01-21 13:54:58.942444: step: 1124/526, loss: 5.1021575927734375e-05 2023-01-21 13:55:00.106772: step: 1128/526, loss: 3.395080420887098e-05 2023-01-21 13:55:01.250431: step: 1132/526, loss: 0.0017368317348882556 2023-01-21 13:55:02.381558: step: 1136/526, loss: 0.0020225525368005037 2023-01-21 13:55:03.482320: step: 1140/526, loss: 5.71250930079259e-05 2023-01-21 13:55:04.630782: step: 1144/526, loss: 0.00011968612670898438 2023-01-21 13:55:05.805427: step: 1148/526, loss: 0.050405073910951614 2023-01-21 13:55:06.954782: step: 1152/526, loss: 0.07323837280273438 2023-01-21 13:55:08.080332: step: 1156/526, loss: 0.0005089759943075478 2023-01-21 13:55:09.185975: step: 1160/526, loss: 0.11612377315759659 2023-01-21 13:55:10.296464: step: 1164/526, loss: 0.008359909057617188 2023-01-21 13:55:11.433403: step: 1168/526, loss: 0.002209282014518976 2023-01-21 13:55:12.551371: step: 1172/526, loss: 0.0002244949428131804 2023-01-21 13:55:13.720337: step: 1176/526, loss: 0.009678459726274014 2023-01-21 13:55:14.853933: step: 1180/526, loss: 0.0006263732793740928 2023-01-21 13:55:15.968693: step: 1184/526, loss: 0.0002487182791810483 2023-01-21 13:55:17.085377: step: 1188/526, loss: 0.03626585006713867 2023-01-21 13:55:18.246294: step: 1192/526, loss: 0.00010070801363326609 2023-01-21 13:55:19.380028: step: 1196/526, loss: 0.0008913993951864541 2023-01-21 13:55:20.532690: step: 1200/526, loss: 0.049355123192071915 2023-01-21 13:55:21.674893: step: 1204/526, loss: 1.354217511106981e-05 2023-01-21 13:55:22.807045: step: 1208/526, loss: 0.0001028060942189768 2023-01-21 13:55:23.944469: step: 1212/526, loss: 2.1457672119140625e-06 2023-01-21 13:55:25.044137: step: 1216/526, loss: 0.0002957344113383442 2023-01-21 13:55:26.149941: step: 1220/526, loss: 0.0002740860218182206 2023-01-21 13:55:27.251079: step: 1224/526, loss: 0.00034646986750885844 2023-01-21 13:55:28.348855: step: 1228/526, loss: 0.0009620666969567537 2023-01-21 13:55:29.501956: step: 1232/526, loss: 1.087188684323337e-05 2023-01-21 13:55:30.581405: step: 1236/526, loss: 3.1471254260395654e-06 2023-01-21 13:55:31.719997: step: 1240/526, loss: 1.8596649169921875e-05 2023-01-21 13:55:32.858084: step: 1244/526, loss: 0.00021181107149459422 2023-01-21 13:55:33.974805: step: 1248/526, loss: 0.004889774601906538 2023-01-21 13:55:35.123463: step: 1252/526, loss: 0.0003567695675883442 2023-01-21 13:55:36.276824: step: 1256/526, loss: 0.00042896269587799907 2023-01-21 13:55:37.404022: step: 1260/526, loss: 0.00012784004502464086 2023-01-21 13:55:38.517998: step: 1264/526, loss: 0.008442020043730736 2023-01-21 13:55:39.650556: step: 1268/526, loss: 0.0007042884826660156 2023-01-21 13:55:40.760441: step: 1272/526, loss: 0.00027899740962311625 2023-01-21 13:55:41.918347: step: 1276/526, loss: 0.00010652541823219508 2023-01-21 13:55:43.068836: step: 1280/526, loss: 0.00031480786856263876 2023-01-21 13:55:44.192763: step: 1284/526, loss: 5.722047262679553e-07 2023-01-21 13:55:45.366433: step: 1288/526, loss: 0.0017557144165039062 2023-01-21 13:55:46.499217: step: 1292/526, loss: 0.00169792165979743 2023-01-21 13:55:47.667902: step: 1296/526, loss: 1.869201696536038e-05 2023-01-21 13:55:48.823292: step: 1300/526, loss: 2.861022494471399e-06 2023-01-21 13:55:49.997768: step: 1304/526, loss: 9.555817086948082e-05 2023-01-21 13:55:51.127256: step: 1308/526, loss: 5.435943421616685e-06 2023-01-21 13:55:52.270882: step: 1312/526, loss: 0.009656143374741077 2023-01-21 13:55:53.432692: step: 1316/526, loss: 7.400513277389109e-05 2023-01-21 13:55:54.540525: step: 1320/526, loss: 1.5068055290612392e-05 2023-01-21 13:55:55.652051: step: 1324/526, loss: 0.0036869049072265625 2023-01-21 13:55:56.818687: step: 1328/526, loss: 4.1675568354548886e-05 2023-01-21 13:55:57.959464: step: 1332/526, loss: 0.007709693629294634 2023-01-21 13:55:59.121619: step: 1336/526, loss: 0.0002304077206645161 2023-01-21 13:56:00.261953: step: 1340/526, loss: 0.0011201858287677169 2023-01-21 13:56:01.398868: step: 1344/526, loss: 4.329681542003527e-05 2023-01-21 13:56:02.544696: step: 1348/526, loss: 4.00543194700731e-06 2023-01-21 13:56:03.660103: step: 1352/526, loss: 0.05244407430291176 2023-01-21 13:56:04.828861: step: 1356/526, loss: 0.0005008697044104338 2023-01-21 13:56:05.967681: step: 1360/526, loss: 0.042695045471191406 2023-01-21 13:56:07.134946: step: 1364/526, loss: 0.007861138321459293 2023-01-21 13:56:08.274271: step: 1368/526, loss: 0.00266685476526618 2023-01-21 13:56:09.418809: step: 1372/526, loss: 0.0014341354835778475 2023-01-21 13:56:10.548672: step: 1376/526, loss: 0.0016349791549146175 2023-01-21 13:56:11.652607: step: 1380/526, loss: 0.0015510559314861894 2023-01-21 13:56:12.784432: step: 1384/526, loss: 2.8038026357535273e-05 2023-01-21 13:56:13.890780: step: 1388/526, loss: 2.5129320420091972e-05 2023-01-21 13:56:15.016208: step: 1392/526, loss: 0.002948522800579667 2023-01-21 13:56:16.124267: step: 1396/526, loss: 0.00037593842716887593 2023-01-21 13:56:17.259671: step: 1400/526, loss: 0.0003245353582315147 2023-01-21 13:56:18.388236: step: 1404/526, loss: 5.817413330078125e-05 2023-01-21 13:56:19.515410: step: 1408/526, loss: 0.022128678858280182 2023-01-21 13:56:20.660994: step: 1412/526, loss: 5.455017162603326e-05 2023-01-21 13:56:21.773975: step: 1416/526, loss: 0.0045060161501169205 2023-01-21 13:56:22.896475: step: 1420/526, loss: -1.3351439065445447e-06 2023-01-21 13:56:24.027047: step: 1424/526, loss: 5.8937075664289296e-05 2023-01-21 13:56:25.154352: step: 1428/526, loss: 0.009972859174013138 2023-01-21 13:56:26.280438: step: 1432/526, loss: 0.018579674884676933 2023-01-21 13:56:27.438221: step: 1436/526, loss: 0.00012874603271484375 2023-01-21 13:56:28.571610: step: 1440/526, loss: 0.0007793426048010588 2023-01-21 13:56:29.711607: step: 1444/526, loss: 0.6036515831947327 2023-01-21 13:56:30.859822: step: 1448/526, loss: 0.13037262856960297 2023-01-21 13:56:31.993388: step: 1452/526, loss: 0.01944446563720703 2023-01-21 13:56:33.101461: step: 1456/526, loss: 0.031284235417842865 2023-01-21 13:56:34.252467: step: 1460/526, loss: 0.006911468226462603 2023-01-21 13:56:35.432491: step: 1464/526, loss: 1.0013580322265625e-05 2023-01-21 13:56:36.552002: step: 1468/526, loss: 2.8228761948412284e-05 2023-01-21 13:56:37.694378: step: 1472/526, loss: 0.06143951416015625 2023-01-21 13:56:38.825403: step: 1476/526, loss: 8.587837510276586e-05 2023-01-21 13:56:39.960268: step: 1480/526, loss: 0.0020687102805823088 2023-01-21 13:56:41.088517: step: 1484/526, loss: 0.007953548803925514 2023-01-21 13:56:42.250533: step: 1488/526, loss: 3.070831371587701e-05 2023-01-21 13:56:43.408765: step: 1492/526, loss: 6.542206392623484e-05 2023-01-21 13:56:44.554436: step: 1496/526, loss: 0.014448165893554688 2023-01-21 13:56:45.689191: step: 1500/526, loss: 0.0014653205871582031 2023-01-21 13:56:46.830187: step: 1504/526, loss: 1.945495569088962e-05 2023-01-21 13:56:47.960348: step: 1508/526, loss: 0.00021057129197288305 2023-01-21 13:56:49.111936: step: 1512/526, loss: 0.010966873727738857 2023-01-21 13:56:50.232761: step: 1516/526, loss: 4.19616708313697e-06 2023-01-21 13:56:51.395995: step: 1520/526, loss: 0.03119640424847603 2023-01-21 13:56:52.525498: step: 1524/526, loss: 3.24249276673072e-06 2023-01-21 13:56:53.642305: step: 1528/526, loss: 0.0028587342239916325 2023-01-21 13:56:54.801128: step: 1532/526, loss: 0.0004125595442019403 2023-01-21 13:56:55.956943: step: 1536/526, loss: 0.05422992631793022 2023-01-21 13:56:57.094006: step: 1540/526, loss: 2.174377368646674e-05 2023-01-21 13:56:58.236431: step: 1544/526, loss: 0.00015163421630859375 2023-01-21 13:56:59.351284: step: 1548/526, loss: 7.247925168485381e-06 2023-01-21 13:57:00.509237: step: 1552/526, loss: 0.002633476397022605 2023-01-21 13:57:01.660088: step: 1556/526, loss: 7.26699799997732e-05 2023-01-21 13:57:02.793891: step: 1560/526, loss: 0.0002200603485107422 2023-01-21 13:57:03.956843: step: 1564/526, loss: 0.0009292602771893144 2023-01-21 13:57:05.087503: step: 1568/526, loss: 0.009453201666474342 2023-01-21 13:57:06.230257: step: 1572/526, loss: 0.08656969666481018 2023-01-21 13:57:07.351733: step: 1576/526, loss: 1.33514404296875e-05 2023-01-21 13:57:08.476362: step: 1580/526, loss: 0.0017705917125567794 2023-01-21 13:57:09.582233: step: 1584/526, loss: 0.02142944373190403 2023-01-21 13:57:10.718911: step: 1588/526, loss: 0.007173466961830854 2023-01-21 13:57:11.852348: step: 1592/526, loss: 3.776550147449598e-05 2023-01-21 13:57:12.980010: step: 1596/526, loss: 0.0007255554082803428 2023-01-21 13:57:14.137669: step: 1600/526, loss: 0.001608562539331615 2023-01-21 13:57:15.292524: step: 1604/526, loss: 0.024193763732910156 2023-01-21 13:57:16.420819: step: 1608/526, loss: 0.0002079010009765625 2023-01-21 13:57:17.585780: step: 1612/526, loss: 6.732940528308973e-05 2023-01-21 13:57:18.699252: step: 1616/526, loss: 0.0003525734064169228 2023-01-21 13:57:19.837734: step: 1620/526, loss: 3.070831371587701e-05 2023-01-21 13:57:20.959190: step: 1624/526, loss: 1.7452241081628017e-05 2023-01-21 13:57:22.085774: step: 1628/526, loss: 2.002716064453125e-05 2023-01-21 13:57:23.212664: step: 1632/526, loss: 7.63893112889491e-05 2023-01-21 13:57:24.361006: step: 1636/526, loss: 2.6702882678364404e-06 2023-01-21 13:57:25.546712: step: 1640/526, loss: 0.017107296735048294 2023-01-21 13:57:26.685965: step: 1644/526, loss: 0.0029172897338867188 2023-01-21 13:57:27.803018: step: 1648/526, loss: 0.09816054999828339 2023-01-21 13:57:28.931134: step: 1652/526, loss: -2.6226043701171875e-06 2023-01-21 13:57:30.055774: step: 1656/526, loss: 0.0005792618030682206 2023-01-21 13:57:31.168992: step: 1660/526, loss: 5.53131121705519e-06 2023-01-21 13:57:32.316529: step: 1664/526, loss: 0.00031843187753111124 2023-01-21 13:57:33.445178: step: 1668/526, loss: 0.027492905035614967 2023-01-21 13:57:34.617059: step: 1672/526, loss: 1.2588501704158261e-05 2023-01-21 13:57:35.777331: step: 1676/526, loss: 1.277923547604587e-05 2023-01-21 13:57:36.918550: step: 1680/526, loss: 8.583069188716763e-07 2023-01-21 13:57:38.033428: step: 1684/526, loss: 9.860993304755539e-05 2023-01-21 13:57:39.171573: step: 1688/526, loss: 0.0006062507745809853 2023-01-21 13:57:40.307921: step: 1692/526, loss: 8.964539119915571e-06 2023-01-21 13:57:41.470768: step: 1696/526, loss: 0.004302597139030695 2023-01-21 13:57:42.604480: step: 1700/526, loss: 0.005199909675866365 2023-01-21 13:57:43.760511: step: 1704/526, loss: 0.007496881298720837 2023-01-21 13:57:44.901778: step: 1708/526, loss: 6.732941255904734e-05 2023-01-21 13:57:46.063345: step: 1712/526, loss: 0.47811076045036316 2023-01-21 13:57:47.203678: step: 1716/526, loss: 0.0004191398620605469 2023-01-21 13:57:48.340216: step: 1720/526, loss: 0.0011436461936682463 2023-01-21 13:57:49.461088: step: 1724/526, loss: 0.02382030338048935 2023-01-21 13:57:50.585965: step: 1728/526, loss: 2.441406286379788e-05 2023-01-21 13:57:51.721562: step: 1732/526, loss: 6.680488877464086e-05 2023-01-21 13:57:52.846831: step: 1736/526, loss: 6.561279587913305e-05 2023-01-21 13:57:54.000204: step: 1740/526, loss: 2.2220612663659267e-05 2023-01-21 13:57:55.156642: step: 1744/526, loss: 0.0026401046197861433 2023-01-21 13:57:56.293915: step: 1748/526, loss: 0.2119217813014984 2023-01-21 13:57:57.423079: step: 1752/526, loss: 0.009987641125917435 2023-01-21 13:57:58.547974: step: 1756/526, loss: 0.023770714178681374 2023-01-21 13:57:59.676660: step: 1760/526, loss: 0.0011073113419115543 2023-01-21 13:58:00.826224: step: 1764/526, loss: 0.00010299683344783261 2023-01-21 13:58:01.937678: step: 1768/526, loss: 4.2629242670955136e-05 2023-01-21 13:58:03.094667: step: 1772/526, loss: 0.00466079730540514 2023-01-21 13:58:04.261988: step: 1776/526, loss: 0.0020397186744958162 2023-01-21 13:58:05.413297: step: 1780/526, loss: 0.0014230727683752775 2023-01-21 13:58:06.557543: step: 1784/526, loss: 0.0004172325134277344 2023-01-21 13:58:07.681031: step: 1788/526, loss: 0.0006271362653933465 2023-01-21 13:58:08.795573: step: 1792/526, loss: 0.011373710818588734 2023-01-21 13:58:09.996861: step: 1796/526, loss: 0.005287742707878351 2023-01-21 13:58:11.115646: step: 1800/526, loss: -2.19345088225964e-06 2023-01-21 13:58:12.255896: step: 1804/526, loss: 2.6845931643038057e-05 2023-01-21 13:58:13.376396: step: 1808/526, loss: 0.0015560149913653731 2023-01-21 13:58:14.498516: step: 1812/526, loss: 0.06823697686195374 2023-01-21 13:58:15.623456: step: 1816/526, loss: 0.0019278526306152344 2023-01-21 13:58:16.724388: step: 1820/526, loss: 0.07867280393838882 2023-01-21 13:58:17.830690: step: 1824/526, loss: 0.00042743684025481343 2023-01-21 13:58:18.963198: step: 1828/526, loss: 0.0008561611175537109 2023-01-21 13:58:20.113847: step: 1832/526, loss: 0.00011920928955078125 2023-01-21 13:58:21.253542: step: 1836/526, loss: 0.0020898820366710424 2023-01-21 13:58:22.412209: step: 1840/526, loss: 0.019743727520108223 2023-01-21 13:58:23.531278: step: 1844/526, loss: 0.035468123853206635 2023-01-21 13:58:24.708476: step: 1848/526, loss: 0.015892982482910156 2023-01-21 13:58:25.831796: step: 1852/526, loss: 0.0007521629449911416 2023-01-21 13:58:26.991359: step: 1856/526, loss: 1.4495850336970761e-05 2023-01-21 13:58:28.166184: step: 1860/526, loss: 5.207061622058973e-05 2023-01-21 13:58:29.302605: step: 1864/526, loss: 2.6702882678364404e-06 2023-01-21 13:58:30.452867: step: 1868/526, loss: 0.0014774323208257556 2023-01-21 13:58:31.579167: step: 1872/526, loss: 0.011353207752108574 2023-01-21 13:58:32.732059: step: 1876/526, loss: 0.014170932583510876 2023-01-21 13:58:33.869356: step: 1880/526, loss: 4.487037949729711e-05 2023-01-21 13:58:34.986183: step: 1884/526, loss: 0.00039768218994140625 2023-01-21 13:58:36.140399: step: 1888/526, loss: 0.04080944135785103 2023-01-21 13:58:37.227771: step: 1892/526, loss: 0.02029399946331978 2023-01-21 13:58:38.388502: step: 1896/526, loss: 0.00759277306497097 2023-01-21 13:58:39.547722: step: 1900/526, loss: 0.002919864607974887 2023-01-21 13:58:40.651487: step: 1904/526, loss: 0.002685165498405695 2023-01-21 13:58:41.790982: step: 1908/526, loss: 1.3089180356473662e-05 2023-01-21 13:58:42.936089: step: 1912/526, loss: 0.00036067963810637593 2023-01-21 13:58:44.058373: step: 1916/526, loss: 0.00090112688485533 2023-01-21 13:58:45.176328: step: 1920/526, loss: 0.019569776952266693 2023-01-21 13:58:46.293772: step: 1924/526, loss: 0.00111732492223382 2023-01-21 13:58:47.447894: step: 1928/526, loss: 0.00013151168241165578 2023-01-21 13:58:48.604959: step: 1932/526, loss: 1.5258792700478807e-06 2023-01-21 13:58:49.765611: step: 1936/526, loss: 0.006577968131750822 2023-01-21 13:58:50.932859: step: 1940/526, loss: 2.269744800287299e-05 2023-01-21 13:58:52.049855: step: 1944/526, loss: 0.0001729965297272429 2023-01-21 13:58:53.229850: step: 1948/526, loss: 0.005018198397010565 2023-01-21 13:58:54.362550: step: 1952/526, loss: 0.0008781657088547945 2023-01-21 13:58:55.511158: step: 1956/526, loss: 1.4495850336970761e-05 2023-01-21 13:58:56.655618: step: 1960/526, loss: 0.007883788086473942 2023-01-21 13:58:57.804421: step: 1964/526, loss: 3.128051685052924e-05 2023-01-21 13:58:58.988350: step: 1968/526, loss: 0.06362781673669815 2023-01-21 13:59:00.133066: step: 1972/526, loss: 4.76837158203125e-06 2023-01-21 13:59:01.248255: step: 1976/526, loss: 0.005054474342614412 2023-01-21 13:59:02.397830: step: 1980/526, loss: 0.00013132095045875758 2023-01-21 13:59:03.518758: step: 1984/526, loss: 0.0284791961312294 2023-01-21 13:59:04.642730: step: 1988/526, loss: 7.295608520507812e-05 2023-01-21 13:59:05.786635: step: 1992/526, loss: 2.5558472771081142e-05 2023-01-21 13:59:06.916928: step: 1996/526, loss: 0.00021972657123114914 2023-01-21 13:59:08.071107: step: 2000/526, loss: 0.0012958527076989412 2023-01-21 13:59:09.209569: step: 2004/526, loss: 0.01424627285450697 2023-01-21 13:59:10.307513: step: 2008/526, loss: 6.175041198730469e-05 2023-01-21 13:59:11.464895: step: 2012/526, loss: 0.008519649505615234 2023-01-21 13:59:12.597514: step: 2016/526, loss: 0.019211864098906517 2023-01-21 13:59:13.726537: step: 2020/526, loss: 0.009011078625917435 2023-01-21 13:59:14.865033: step: 2024/526, loss: 0.002674675080925226 2023-01-21 13:59:15.997167: step: 2028/526, loss: 0.0013034821022301912 2023-01-21 13:59:17.132899: step: 2032/526, loss: 0.0012968461960554123 2023-01-21 13:59:18.263765: step: 2036/526, loss: 0.018309593200683594 2023-01-21 13:59:19.403767: step: 2040/526, loss: 0.011711501516401768 2023-01-21 13:59:20.532909: step: 2044/526, loss: 7.171630568336695e-05 2023-01-21 13:59:21.701504: step: 2048/526, loss: 6.408691842807457e-05 2023-01-21 13:59:22.835094: step: 2052/526, loss: 0.012289238162338734 2023-01-21 13:59:23.977559: step: 2056/526, loss: 0.08432511985301971 2023-01-21 13:59:25.130692: step: 2060/526, loss: 0.005443572998046875 2023-01-21 13:59:26.275556: step: 2064/526, loss: 0.04935331642627716 2023-01-21 13:59:27.395824: step: 2068/526, loss: 1.087188684323337e-05 2023-01-21 13:59:28.529955: step: 2072/526, loss: 0.0004212379571981728 2023-01-21 13:59:29.650274: step: 2076/526, loss: 1.296997106692288e-05 2023-01-21 13:59:30.777317: step: 2080/526, loss: 0.03268766403198242 2023-01-21 13:59:31.917994: step: 2084/526, loss: 0.022002030164003372 2023-01-21 13:59:33.058601: step: 2088/526, loss: 0.02983837202191353 2023-01-21 13:59:34.212619: step: 2092/526, loss: 0.003565883729606867 2023-01-21 13:59:35.329474: step: 2096/526, loss: 0.00304336566478014 2023-01-21 13:59:36.464081: step: 2100/526, loss: 0.013535975478589535 2023-01-21 13:59:37.658381: step: 2104/526, loss: 0.0045476434752345085 ================================================== Loss: 0.014 -------------------- Dev: {'event': {'p': 0.5856573705179283, 'r': 0.7829560585885486, 'f1': 0.67008547008547}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Test: {'event': {'p': 0.6119463632669646, 'r': 0.7860125260960334, 'f1': 0.6881425633995888}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Chinese: {'event': {'p': 0.5714285714285714, 'r': 0.8888888888888888, 'f1': 0.6956521739130435}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Korean: {'event': {'p': 0.6226415094339622, 'r': 0.5238095238095238, 'f1': 0.5689655172413793}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} Russian: {'event': {'p': 0.4772727272727273, 'r': 0.5833333333333334, 'f1': 0.5250000000000001}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 29} ================================================== Current best result: -------------------- Eng Dev for Chinese: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Chinese: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Chinese: {'event': {'p': 0.6621621621621622, 'r': 0.9074074074074074, 'f1': 0.765625}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} -------------------- Eng Dev for Korean: {'event': {'p': 0.6257995735607675, 'r': 0.7816245006657789, 'f1': 0.6950858496151568}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Eng Test for Korean: {'event': {'p': 0.623113410952997, 'r': 0.7541753653444676, 'f1': 0.6824085005903188}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} Sample Korean: {'event': {'p': 0.7446808510638298, 'r': 0.5555555555555556, 'f1': 0.6363636363636364}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 7} -------------------- Eng Dev for Russian: {'event': {'p': 0.6266519823788547, 'r': 0.7576564580559254, 'f1': 0.6859553948161543}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Eng Test for Russian: {'event': {'p': 0.6313522825150732, 'r': 0.7651356993736952, 'f1': 0.6918357715903728}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23} Sample Russian: {'event': {'p': 0.5384615384615384, 'r': 0.5833333333333334, 'f1': 0.5599999999999999}, 'argument': {'p': 0.0, 'r': 0.0, 'f1': 0.0}, 'combined': 0.0, 'epoch': 23}